From b64d814257b027e29a474bcd660f6372490138c7 Mon Sep 17 00:00:00 2001 From: Pali Rohár Date: Mon, 7 Sep 2020 13:27:17 +0200 Subject: arm64: dts: marvell: espressobin: Add ethernet switch aliases MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Espressobin boards have 3 ethernet ports and some of them got assigned more then one MAC address. MAC addresses are stored in U-Boot environment. Since commit a2c7023f7075c ("net: dsa: read mac address from DT for slave device") kernel can use MAC addresses from DT for particular DSA port. Currently Espressobin DTS file contains alias just for ethernet0. This patch defines additional ethernet aliases in Espressobin DTS files, so bootloader can fill correct MAC address for DSA switch ports if more MAC addresses were specified. DT alias ethernet1 is used for wan port, DT aliases ethernet2 and ethernet3 are used for lan ports for both Espressobin revisions (V5 and V7). Fixes: 5253cb8c00a6f ("arm64: dts: marvell: espressobin: add ethernet alias") Cc: # a2c7023f7075c: dsa: read mac address Signed-off-by: Pali Rohár Reviewed-by: Andrew Lunn Reviewed-by: Andre Heider Signed-off-by: Gregory CLEMENT --- .../boot/dts/marvell/armada-3720-espressobin-v7-emmc.dts | 10 ++++++++-- arch/arm64/boot/dts/marvell/armada-3720-espressobin-v7.dts | 10 ++++++++-- arch/arm64/boot/dts/marvell/armada-3720-espressobin.dtsi | 12 ++++++++---- 3 files changed, 24 insertions(+), 8 deletions(-) diff --git a/arch/arm64/boot/dts/marvell/armada-3720-espressobin-v7-emmc.dts b/arch/arm64/boot/dts/marvell/armada-3720-espressobin-v7-emmc.dts index 03733fd92732..215d2f702623 100644 --- a/arch/arm64/boot/dts/marvell/armada-3720-espressobin-v7-emmc.dts +++ b/arch/arm64/boot/dts/marvell/armada-3720-espressobin-v7-emmc.dts @@ -20,17 +20,23 @@ compatible = "globalscale,espressobin-v7-emmc", "globalscale,espressobin-v7", "globalscale,espressobin", "marvell,armada3720", "marvell,armada3710"; + + aliases { + /* ethernet1 is wan port */ + ethernet1 = &switch0port3; + ethernet3 = &switch0port1; + }; }; &switch0 { ports { - port@1 { + switch0port1: port@1 { reg = <1>; label = "lan1"; phy-handle = <&switch0phy0>; }; - port@3 { + switch0port3: port@3 { reg = <3>; label = "wan"; phy-handle = <&switch0phy2>; diff --git a/arch/arm64/boot/dts/marvell/armada-3720-espressobin-v7.dts b/arch/arm64/boot/dts/marvell/armada-3720-espressobin-v7.dts index 8570c5f47d7d..b6f4af8ebafb 100644 --- a/arch/arm64/boot/dts/marvell/armada-3720-espressobin-v7.dts +++ b/arch/arm64/boot/dts/marvell/armada-3720-espressobin-v7.dts @@ -19,17 +19,23 @@ model = "Globalscale Marvell ESPRESSOBin Board V7"; compatible = "globalscale,espressobin-v7", "globalscale,espressobin", "marvell,armada3720", "marvell,armada3710"; + + aliases { + /* ethernet1 is wan port */ + ethernet1 = &switch0port3; + ethernet3 = &switch0port1; + }; }; &switch0 { ports { - port@1 { + switch0port1: port@1 { reg = <1>; label = "lan1"; phy-handle = <&switch0phy0>; }; - port@3 { + switch0port3: port@3 { reg = <3>; label = "wan"; phy-handle = <&switch0phy2>; diff --git a/arch/arm64/boot/dts/marvell/armada-3720-espressobin.dtsi b/arch/arm64/boot/dts/marvell/armada-3720-espressobin.dtsi index b97218c72727..0775c16e0ec8 100644 --- a/arch/arm64/boot/dts/marvell/armada-3720-espressobin.dtsi +++ b/arch/arm64/boot/dts/marvell/armada-3720-espressobin.dtsi @@ -13,6 +13,10 @@ / { aliases { ethernet0 = ð0; + /* for dsa slave device */ + ethernet1 = &switch0port1; + ethernet2 = &switch0port2; + ethernet3 = &switch0port3; serial0 = &uart0; serial1 = &uart1; }; @@ -120,7 +124,7 @@ #address-cells = <1>; #size-cells = <0>; - port@0 { + switch0port0: port@0 { reg = <0>; label = "cpu"; ethernet = <ð0>; @@ -131,19 +135,19 @@ }; }; - port@1 { + switch0port1: port@1 { reg = <1>; label = "wan"; phy-handle = <&switch0phy0>; }; - port@2 { + switch0port2: port@2 { reg = <2>; label = "lan0"; phy-handle = <&switch0phy1>; }; - port@3 { + switch0port3: port@3 { reg = <3>; label = "lan1"; phy-handle = <&switch0phy2>; -- cgit v1.2.3 From 0b58725fb9a446890c1fd28fc6c9e393ce21acb7 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Wed, 16 Sep 2020 13:56:54 +0300 Subject: ARM: mvebu: drop pointless check for coherency_base The MMU off code path in ll_get_coherency_base() attempts to decide whether the coherency fabric is mapped by testing the value of coherency_base, which carries its virtual address if its mapped, and 0x0 otherwise. However, what the code actually does is take the virtual address of the coherency_base symbol, and compare it with 0x0, which are never equal, and so the branch is never taken. In fact, with the MMU off, dereferencing the VA of coherency_base is not possible to begin with, nor can its value be relied upon with the MMU off since it is not cleaned to the Dcache as is done with coherency_phys_base in armada_370_coherency_init(). Instead, the value of coherency_phys_base is returned, which results in the correct behavior since it will be 0x0 as well if the coherency fabric is not mapped, and it is accessible with the MMU off. So just drop the comparison and the branch. Fixes: 30cdef97107370a7 ("ARM: mvebu: make the coherency_ll.S functions work with no coherency fabric") Cc: Thomas Petazzoni Cc: Jason Cooper Cc: Andrew Lunn Cc: Gregory CLEMENT Cc: Sebastian Hesselbarth Signed-off-by: Ard Biesheuvel Signed-off-by: Gregory CLEMENT --- arch/arm/mach-mvebu/coherency_ll.S | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/arch/arm/mach-mvebu/coherency_ll.S b/arch/arm/mach-mvebu/coherency_ll.S index 2d962fe48821..a3a64bf97250 100644 --- a/arch/arm/mach-mvebu/coherency_ll.S +++ b/arch/arm/mach-mvebu/coherency_ll.S @@ -35,13 +35,8 @@ ENTRY(ll_get_coherency_base) /* * MMU is disabled, use the physical address of the coherency - * base address. However, if the coherency fabric isn't mapped - * (i.e its virtual address is zero), it means coherency is - * not enabled, so we return 0. + * base address, (or 0x0 if the coherency fabric is not mapped) */ - ldr r1, =coherency_base - cmp r1, #0 - beq 2f adr r1, 3f ldr r3, [r1] ldr r1, [r1, r3] -- cgit v1.2.3 From dea252fa41cd8ce332d148444e4799235a8a03ec Mon Sep 17 00:00:00 2001 From: Clément Péron Date: Sat, 3 Oct 2020 12:03:32 +0200 Subject: ARM: dts: sun4i-a10: fix cpu_alert temperature MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When running dtbs_check thermal_zone warn about the temperature declared. thermal-zones: cpu-thermal:trips:cpu-alert0:temperature:0:0: 850000 is greater than the maximum of 200000 It's indeed wrong the real value is 85°C and not 850°C. Signed-off-by: Clément Péron Signed-off-by: Maxime Ripard Link: https://lore.kernel.org/r/20201003100332.431178-1-peron.clem@gmail.com --- arch/arm/boot/dts/sun4i-a10.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/boot/dts/sun4i-a10.dtsi b/arch/arm/boot/dts/sun4i-a10.dtsi index 0f95a6ef8543..1c5a666c54b5 100644 --- a/arch/arm/boot/dts/sun4i-a10.dtsi +++ b/arch/arm/boot/dts/sun4i-a10.dtsi @@ -143,7 +143,7 @@ trips { cpu_alert0: cpu-alert0 { /* milliCelsius */ - temperature = <850000>; + temperature = <85000>; hysteresis = <2000>; type = "passive"; }; -- cgit v1.2.3 From 1b208bab34dc3f4ef8f408105017d4a7b72b2a2f Mon Sep 17 00:00:00 2001 From: Neil Armstrong Date: Thu, 17 Sep 2020 08:59:48 +0200 Subject: arm64: dts: meson-axg: add USB nodes This adds the USB Glue node, with the USB2 & USB3 controllers along the single USB2 PHY node. Signed-off-by: Neil Armstrong Reviewed-by: Kevin Hilman Reviewed-by: Martin Blumenstingl Signed-off-by: Kevin Hilman --- arch/arm64/boot/dts/amlogic/meson-axg.dtsi | 50 ++++++++++++++++++++++++++++++ 1 file changed, 50 insertions(+) diff --git a/arch/arm64/boot/dts/amlogic/meson-axg.dtsi b/arch/arm64/boot/dts/amlogic/meson-axg.dtsi index b9efc8469265..fae48efae83e 100644 --- a/arch/arm64/boot/dts/amlogic/meson-axg.dtsi +++ b/arch/arm64/boot/dts/amlogic/meson-axg.dtsi @@ -171,6 +171,46 @@ #size-cells = <2>; ranges; + usb: usb@ffe09080 { + compatible = "amlogic,meson-axg-usb-ctrl"; + reg = <0x0 0xffe09080 0x0 0x20>; + interrupts = ; + #address-cells = <2>; + #size-cells = <2>; + ranges; + + clocks = <&clkc CLKID_USB>, <&clkc CLKID_USB1_DDR_BRIDGE>; + clock-names = "usb_ctrl", "ddr"; + resets = <&reset RESET_USB_OTG>; + + dr_mode = "otg"; + + phys = <&usb2_phy1>; + phy-names = "usb2-phy1"; + + dwc2: usb@ff400000 { + compatible = "amlogic,meson-g12a-usb", "snps,dwc2"; + reg = <0x0 0xff400000 0x0 0x40000>; + interrupts = ; + clocks = <&clkc CLKID_USB1>; + clock-names = "otg"; + phys = <&usb2_phy1>; + dr_mode = "peripheral"; + g-rx-fifo-size = <192>; + g-np-tx-fifo-size = <128>; + g-tx-fifo-size = <128 128 16 16 16>; + }; + + dwc3: usb@ff500000 { + compatible = "snps,dwc3"; + reg = <0x0 0xff500000 0x0 0x100000>; + interrupts = ; + dr_mode = "host"; + maximum-speed = "high-speed"; + snps,dis_u2_susphy_quirk; + }; + }; + ethmac: ethernet@ff3f0000 { compatible = "amlogic,meson-axg-dwmac", "snps,dwmac-3.70a", @@ -1734,6 +1774,16 @@ clock-names = "core", "clkin0", "clkin1"; resets = <&reset RESET_SD_EMMC_C>; }; + + usb2_phy1: phy@9020 { + compatible = "amlogic,meson-gxl-usb2-phy"; + #phy-cells = <0>; + reg = <0x0 0x9020 0x0 0x20>; + clocks = <&clkc CLKID_USB>; + clock-names = "phy"; + resets = <&reset RESET_USB_OTG>; + reset-names = "phy"; + }; }; sram: sram@fffc0000 { -- cgit v1.2.3 From f450d2c219f6a6b79880c97bf910c3c72725eb70 Mon Sep 17 00:00:00 2001 From: Neil Armstrong Date: Thu, 17 Sep 2020 08:59:49 +0200 Subject: arm64: dts: meson-axg-s400: enable USB OTG This enables USB OTG on the S400 board. Signed-off-by: Neil Armstrong Reviewed-by: Kevin Hilman Signed-off-by: Kevin Hilman --- arch/arm64/boot/dts/amlogic/meson-axg-s400.dts | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts b/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts index cb1360ae1211..7740f97c240f 100644 --- a/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts +++ b/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts @@ -584,3 +584,9 @@ pinctrl-0 = <&uart_ao_a_pins>; pinctrl-names = "default"; }; + +&usb { + status = "okay"; + dr_mode = "otg"; + vbus-supply = <&usb_pwr>; +}; -- cgit v1.2.3 From a1afbbb0285797e01313779c71287d936d069245 Mon Sep 17 00:00:00 2001 From: Scott K Logan Date: Fri, 25 Sep 2020 01:43:53 -0700 Subject: arm64: dts: meson: add missing g12 rng clock This adds the missing perpheral clock for the RNG for Amlogic G12. As stated in amlogic,meson-rng.yaml, this isn't always necessary for the RNG to function, but is better to have in case the clock is disabled for some reason prior to loading. Signed-off-by: Scott K Logan Suggested-by: Neil Armstrong Reviewed-by: Neil Armstrong Signed-off-by: Kevin Hilman Link: https://lore.kernel.org/r/520a1a8ec7a958b3d918d89563ec7e93a4100a45.camel@cottsay.net --- arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi b/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi index 1e83ec5b8c91..81f490e404ca 100644 --- a/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi +++ b/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi @@ -282,6 +282,8 @@ hwrng: rng@218 { compatible = "amlogic,meson-rng"; reg = <0x0 0x218 0x0 0x4>; + clocks = <&clkc CLKID_RNG0>; + clock-names = "core"; }; }; -- cgit v1.2.3 From 1fdc97ae450ede2b4911d6737a57e6fca63b5f4a Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Fri, 25 Sep 2020 23:17:43 +0200 Subject: arm64: dts: amlogic: meson-g12: use the G12A specific dwmac compatible We have a dedicated "amlogic,meson-g12a-dwmac" compatible string for the Ethernet controller since commit 3efdb92426bf4 ("dt-bindings: net: dwmac-meson: Add a compatible string for G12A onwards"). Using the AXG compatible string worked fine so far because the dwmac-meson8b driver doesn't handle the newly introduced register bits for G12A. However, once that changes the driver must be probed with the correct compatible string to manage these new register bits. Signed-off-by: Martin Blumenstingl Reviewed-by: Neil Armstrong Signed-off-by: Kevin Hilman Link: https://lore.kernel.org/r/20200925211743.537496-1-martin.blumenstingl@googlemail.com --- arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi b/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi index 81f490e404ca..c95ebe615176 100644 --- a/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi +++ b/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi @@ -209,7 +209,7 @@ }; ethmac: ethernet@ff3f0000 { - compatible = "amlogic,meson-axg-dwmac", + compatible = "amlogic,meson-g12a-dwmac", "snps,dwmac-3.70a", "snps,dwmac"; reg = <0x0 0xff3f0000 0x0 0x10000>, -- cgit v1.2.3 From 45b9e04d5ba0b043783dfe2b19bb728e712cb32e Mon Sep 17 00:00:00 2001 From: Etienne Carriere Date: Thu, 8 Oct 2020 16:37:22 +0200 Subject: firmware: arm_scmi: Fix ARCH_COLD_RESET The defination for ARCH_COLD_RESET is wrong. Let us fix it according to the SCMI specification. Link: https://lore.kernel.org/r/20201008143722.21888-5-etienne.carriere@linaro.org Fixes: 95a15d80aa0d ("firmware: arm_scmi: Add RESET protocol in SCMI v2.0") Signed-off-by: Etienne Carriere Signed-off-by: Sudeep Holla --- drivers/firmware/arm_scmi/reset.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/firmware/arm_scmi/reset.c b/drivers/firmware/arm_scmi/reset.c index f063cfe17e02..a981a22cfe89 100644 --- a/drivers/firmware/arm_scmi/reset.c +++ b/drivers/firmware/arm_scmi/reset.c @@ -36,9 +36,7 @@ struct scmi_msg_reset_domain_reset { #define EXPLICIT_RESET_ASSERT BIT(1) #define ASYNCHRONOUS_RESET BIT(2) __le32 reset_state; -#define ARCH_RESET_TYPE BIT(31) -#define COLD_RESET_STATE BIT(0) -#define ARCH_COLD_RESET (ARCH_RESET_TYPE | COLD_RESET_STATE) +#define ARCH_COLD_RESET 0 }; struct scmi_msg_reset_notify { -- cgit v1.2.3 From 7adb2c8aaaa6a387af7140e57004beba2c04a4c6 Mon Sep 17 00:00:00 2001 From: Etienne Carriere Date: Thu, 8 Oct 2020 16:37:21 +0200 Subject: firmware: arm_scmi: Expand SMC/HVC message pool to more than one SMC/HVC can transmit only one message at the time as the shared memory needs to be protected and the calls are synchronous. However, in order to allow multiple threads to send SCMI messages simultaneously, we need a larger poll of memory. Let us just use value of 20 to keep it in sync mailbox transport implementation. Any other value must work perfectly. Link: https://lore.kernel.org/r/20201008143722.21888-4-etienne.carriere@linaro.org Fixes: 1dc6558062da ("firmware: arm_scmi: Add smc/hvc transport") Cc: Peng Fan Signed-off-by: Etienne Carriere [sudeep.holla: reworded the commit message to indicate the practicality] Signed-off-by: Sudeep Holla --- drivers/firmware/arm_scmi/smc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/firmware/arm_scmi/smc.c b/drivers/firmware/arm_scmi/smc.c index 1a03c3ec0230..82a82a5dc86a 100644 --- a/drivers/firmware/arm_scmi/smc.c +++ b/drivers/firmware/arm_scmi/smc.c @@ -149,6 +149,6 @@ static const struct scmi_transport_ops scmi_smc_ops = { const struct scmi_desc scmi_smc_desc = { .ops = &scmi_smc_ops, .max_rx_timeout_ms = 30, - .max_msg = 1, + .max_msg = 20, .max_msg_size = 128, }; -- cgit v1.2.3 From 722939528a37aa0cb22d441e2045c0cf53e78fb0 Mon Sep 17 00:00:00 2001 From: Sumit Garg Date: Thu, 17 Sep 2020 19:10:22 +0530 Subject: tee: client UUID: Skip REE kernel login method as well Since the addition of session's client UUID generation via commit [1], login via REE kernel method was disallowed. So fix that via passing nill UUID in case of TEE_IOCTL_LOGIN_REE_KERNEL method as well. Fixes: e33bcbab16d1 ("tee: add support for session's client UUID generation") [1] Signed-off-by: Sumit Garg Signed-off-by: Jens Wiklander --- drivers/tee/tee_core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/tee/tee_core.c b/drivers/tee/tee_core.c index 64637e09a095..2f6199ebf769 100644 --- a/drivers/tee/tee_core.c +++ b/drivers/tee/tee_core.c @@ -200,7 +200,8 @@ int tee_session_calc_client_uuid(uuid_t *uuid, u32 connection_method, int name_len; int rc; - if (connection_method == TEE_IOCTL_LOGIN_PUBLIC) { + if (connection_method == TEE_IOCTL_LOGIN_PUBLIC || + connection_method == TEE_IOCTL_LOGIN_REE_KERNEL) { /* Nil UUID to be passed to TEE environment */ uuid_copy(uuid, &uuid_null); return 0; -- cgit v1.2.3 From 9724722fde8f9bbd2b87340f00b9300c9284001e Mon Sep 17 00:00:00 2001 From: Sudeep Holla Date: Mon, 12 Oct 2020 14:26:24 +0100 Subject: firmware: arm_scmi: Add missing Rx size re-initialisation Few commands provide the list of description partially and require to be called consecutively until all the descriptors are fetched completely. In such cases, we don't release the buffers and reuse them for consecutive transmits. However, currently we don't reset the Rx size which will be set as per the response for the last transmit. This may result in incorrect response size being interpretted as the firmware may repond with size greater than the one set but we read only upto the size set by previous response. Let us reset the receive buffer size to max possible in such cases as we don't know the exact size of the response. Link: https://lore.kernel.org/r/20201012141746.32575-1-sudeep.holla@arm.com Fixes: b6f20ff8bd94 ("firmware: arm_scmi: add common infrastructure and support for base protocol") Reported-by: Etienne Carriere Signed-off-by: Sudeep Holla --- drivers/firmware/arm_scmi/base.c | 2 ++ drivers/firmware/arm_scmi/clock.c | 2 ++ drivers/firmware/arm_scmi/common.h | 2 ++ drivers/firmware/arm_scmi/driver.c | 8 ++++++++ drivers/firmware/arm_scmi/perf.c | 2 ++ drivers/firmware/arm_scmi/sensors.c | 2 ++ 6 files changed, 18 insertions(+) diff --git a/drivers/firmware/arm_scmi/base.c b/drivers/firmware/arm_scmi/base.c index 9853bd3c4d45..017e5d8bd869 100644 --- a/drivers/firmware/arm_scmi/base.c +++ b/drivers/firmware/arm_scmi/base.c @@ -197,6 +197,8 @@ static int scmi_base_implementation_list_get(const struct scmi_handle *handle, protocols_imp[tot_num_ret + loop] = *(list + loop); tot_num_ret += loop_num_ret; + + scmi_reset_rx_to_maxsz(handle, t); } while (loop_num_ret); scmi_xfer_put(handle, t); diff --git a/drivers/firmware/arm_scmi/clock.c b/drivers/firmware/arm_scmi/clock.c index c1cfe3ee3d55..4645677d86f1 100644 --- a/drivers/firmware/arm_scmi/clock.c +++ b/drivers/firmware/arm_scmi/clock.c @@ -192,6 +192,8 @@ scmi_clock_describe_rates_get(const struct scmi_handle *handle, u32 clk_id, } tot_rate_cnt += num_returned; + + scmi_reset_rx_to_maxsz(handle, t); /* * check for both returned and remaining to avoid infinite * loop due to buggy firmware diff --git a/drivers/firmware/arm_scmi/common.h b/drivers/firmware/arm_scmi/common.h index 37fb583f1bf5..65063fa948d4 100644 --- a/drivers/firmware/arm_scmi/common.h +++ b/drivers/firmware/arm_scmi/common.h @@ -147,6 +147,8 @@ int scmi_do_xfer_with_response(const struct scmi_handle *h, struct scmi_xfer *xfer); int scmi_xfer_get_init(const struct scmi_handle *h, u8 msg_id, u8 prot_id, size_t tx_size, size_t rx_size, struct scmi_xfer **p); +void scmi_reset_rx_to_maxsz(const struct scmi_handle *handle, + struct scmi_xfer *xfer); int scmi_handle_put(const struct scmi_handle *handle); struct scmi_handle *scmi_handle_get(struct device *dev); void scmi_set_handle(struct scmi_device *scmi_dev); diff --git a/drivers/firmware/arm_scmi/driver.c b/drivers/firmware/arm_scmi/driver.c index c5dea87edf8f..3dfd8b6a0ebf 100644 --- a/drivers/firmware/arm_scmi/driver.c +++ b/drivers/firmware/arm_scmi/driver.c @@ -402,6 +402,14 @@ int scmi_do_xfer(const struct scmi_handle *handle, struct scmi_xfer *xfer) return ret; } +void scmi_reset_rx_to_maxsz(const struct scmi_handle *handle, + struct scmi_xfer *xfer) +{ + struct scmi_info *info = handle_to_scmi_info(handle); + + xfer->rx.len = info->desc->max_msg_size; +} + #define SCMI_MAX_RESPONSE_TIMEOUT (2 * MSEC_PER_SEC) /** diff --git a/drivers/firmware/arm_scmi/perf.c b/drivers/firmware/arm_scmi/perf.c index ed475b40bd08..82fb3babff72 100644 --- a/drivers/firmware/arm_scmi/perf.c +++ b/drivers/firmware/arm_scmi/perf.c @@ -304,6 +304,8 @@ scmi_perf_describe_levels_get(const struct scmi_handle *handle, u32 domain, } tot_opp_cnt += num_returned; + + scmi_reset_rx_to_maxsz(handle, t); /* * check for both returned and remaining to avoid infinite * loop due to buggy firmware diff --git a/drivers/firmware/arm_scmi/sensors.c b/drivers/firmware/arm_scmi/sensors.c index 9703cf6356a0..b4232d611033 100644 --- a/drivers/firmware/arm_scmi/sensors.c +++ b/drivers/firmware/arm_scmi/sensors.c @@ -166,6 +166,8 @@ static int scmi_sensor_description_get(const struct scmi_handle *handle, } desc_index += num_returned; + + scmi_reset_rx_to_maxsz(handle, t); /* * check for both returned and remaining to avoid infinite * loop due to buggy firmware -- cgit v1.2.3 From afc18069a2cb7ead5f86623a5f3d4ad6e21f940d Mon Sep 17 00:00:00 2001 From: Kairui Song Date: Wed, 14 Oct 2020 17:24:28 +0800 Subject: x86/kexec: Use up-to-dated screen_info copy to fill boot params kexec_file_load() currently reuses the old boot_params.screen_info, but if drivers have change the hardware state, boot_param.screen_info could contain invalid info. For example, the video type might be no longer VGA, or the frame buffer address might be changed. If the kexec kernel keeps using the old screen_info, kexec'ed kernel may attempt to write to an invalid framebuffer memory region. There are two screen_info instances globally available, boot_params.screen_info and screen_info. Later one is a copy, and is updated by drivers. So let kexec_file_load use the updated copy. [ mingo: Tidied up the changelog. ] Signed-off-by: Kairui Song Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20201014092429.1415040-2-kasong@redhat.com --- arch/x86/kernel/kexec-bzimage64.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/x86/kernel/kexec-bzimage64.c b/arch/x86/kernel/kexec-bzimage64.c index 57c2ecf43134..ce831f9448e7 100644 --- a/arch/x86/kernel/kexec-bzimage64.c +++ b/arch/x86/kernel/kexec-bzimage64.c @@ -200,8 +200,7 @@ setup_boot_parameters(struct kimage *image, struct boot_params *params, params->hdr.hardware_subarch = boot_params.hdr.hardware_subarch; /* Copying screen_info will do? */ - memcpy(¶ms->screen_info, &boot_params.screen_info, - sizeof(struct screen_info)); + memcpy(¶ms->screen_info, &screen_info, sizeof(struct screen_info)); /* Fill in memsize later */ params->screen_info.ext_mem_k = 0; -- cgit v1.2.3 From 3cb73bc3fa2a3cb80b88aa63b48409939e0d996b Mon Sep 17 00:00:00 2001 From: Kairui Song Date: Wed, 14 Oct 2020 17:24:29 +0800 Subject: hyperv_fb: Update screen_info after removing old framebuffer On gen2 HyperV VM, hyperv_fb will remove the old framebuffer, and the new allocated framebuffer address could be at a differnt location, and it might be no longer a VGA framebuffer. Update screen_info so that after kexec the kernel won't try to reuse the old invalid/stale framebuffer address as VGA, corrupting memory. [ mingo: Tidied up the changelog. ] Signed-off-by: Kairui Song Signed-off-by: Ingo Molnar Cc: Dexuan Cui Cc: Jake Oshins Cc: Wei Hu Cc: "K. Y. Srinivasan" Cc: Haiyang Zhang Cc: Stephen Hemminger Link: https://lore.kernel.org/r/20201014092429.1415040-3-kasong@redhat.com --- drivers/video/fbdev/hyperv_fb.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/video/fbdev/hyperv_fb.c b/drivers/video/fbdev/hyperv_fb.c index 02411d89cb46..e36fb1a0ecdb 100644 --- a/drivers/video/fbdev/hyperv_fb.c +++ b/drivers/video/fbdev/hyperv_fb.c @@ -1114,8 +1114,15 @@ static int hvfb_getmem(struct hv_device *hdev, struct fb_info *info) getmem_done: remove_conflicting_framebuffers(info->apertures, KBUILD_MODNAME, false); - if (!gen2vm) + + if (gen2vm) { + /* framebuffer is reallocated, clear screen_info to avoid misuse from kexec */ + screen_info.lfb_size = 0; + screen_info.lfb_base = 0; + screen_info.orig_video_isVGA = 0; + } else { pci_dev_put(pdev); + } kfree(info->apertures); return 0; -- cgit v1.2.3 From f2ac57a4c49d40409c21c82d23b5706df9b438af Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Wed, 14 Oct 2020 07:30:51 +0200 Subject: x86/unwind/orc: Fix inactive tasks with stack pointer in %sp on GCC 10 compiled kernels GCC 10 optimizes the scheduler code differently than its predecessors. When CONFIG_DEBUG_SECTION_MISMATCH=y, the Makefile forces GCC not to inline some functions (-fno-inline-functions-called-once). Before GCC 10, "no-inlined" __schedule() starts with the usual prologue: push %bp mov %sp, %bp So the ORC unwinder simply picks stack pointer from %bp and unwinds from __schedule() just perfectly: $ cat /proc/1/stack [<0>] ep_poll+0x3e9/0x450 [<0>] do_epoll_wait+0xaa/0xc0 [<0>] __x64_sys_epoll_wait+0x1a/0x20 [<0>] do_syscall_64+0x33/0x40 [<0>] entry_SYSCALL_64_after_hwframe+0x44/0xa9 But now, with GCC 10, there is no %bp prologue in __schedule(): $ cat /proc/1/stack The ORC entry of the point in __schedule() is: sp:sp+88 bp:last_sp-48 type:call end:0 In this case, nobody subtracts sizeof "struct inactive_task_frame" in __unwind_start(). The struct is put on the stack by __switch_to_asm() and only then __switch_to_asm() stores %sp to task->thread.sp. But we start unwinding from a point in __schedule() (stored in frame->ret_addr by 'call') and not in __switch_to_asm(). So for these example values in __unwind_start(): sp=ffff94b50001fdc8 bp=ffff8e1f41d29340 ip=__schedule+0x1f0 The stack is: ffff94b50001fdc8: ffff8e1f41578000 # struct inactive_task_frame ffff94b50001fdd0: 0000000000000000 ffff94b50001fdd8: ffff8e1f41d29340 ffff94b50001fde0: ffff8e1f41611d40 # ... ffff94b50001fde8: ffffffff93c41920 # bx ffff94b50001fdf0: ffff8e1f41d29340 # bp ffff94b50001fdf8: ffffffff9376cad0 # ret_addr (and end of the struct) 0xffffffff9376cad0 is __schedule+0x1f0 (after the call to __switch_to_asm). Now follow those 88 bytes from the ORC entry (sp+88). The entry is correct, __schedule() really pushes 48 bytes (8*7) + 32 bytes via subq to store some local values (like 4U below). So to unwind, look at the offset 88-sizeof(long) = 0x50 from here: ffff94b50001fe00: ffff8e1f41578618 ffff94b50001fe08: 00000cc000000255 ffff94b50001fe10: 0000000500000004 ffff94b50001fe18: 7793fab6956b2d00 # NOTE (see below) ffff94b50001fe20: ffff8e1f41578000 ffff94b50001fe28: ffff8e1f41578000 ffff94b50001fe30: ffff8e1f41578000 ffff94b50001fe38: ffff8e1f41578000 ffff94b50001fe40: ffff94b50001fed8 ffff94b50001fe48: ffff8e1f41577ff0 ffff94b50001fe50: ffffffff9376cf12 Here ^^^^^^^^^^^^^^^^ is the correct ret addr from __schedule(). It translates to schedule+0x42 (insn after a call to __schedule()). BUT, unwind_next_frame() tries to take the address starting from 0xffff94b50001fdc8. That is exactly from thread.sp+88-sizeof(long) = 0xffff94b50001fdc8+88-8 = 0xffff94b50001fe18, which is garbage marked as NOTE above. So this quits the unwinding as 7793fab6956b2d00 is obviously not a kernel address. There was a fix to skip 'struct inactive_task_frame' in unwind_get_return_address_ptr in the following commit: 187b96db5ca7 ("x86/unwind/orc: Fix unwind_get_return_address_ptr() for inactive tasks") But we need to skip the struct already in the unwinder proper. So subtract the size (increase the stack pointer) of the structure in __unwind_start() directly. This allows for removal of the code added by commit 187b96db5ca7 completely, as the address is now at '(unsigned long *)state->sp - 1', the same as in the generic case. [ mingo: Cleaned up the changelog a bit, for better readability. ] Fixes: ee9f8fce9964 ("x86/unwind: Add the ORC unwinder") Bug: https://bugzilla.suse.com/show_bug.cgi?id=1176907 Signed-off-by: Jiri Slaby Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20201014053051.24199-1-jslaby@suse.cz --- arch/x86/kernel/unwind_orc.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) diff --git a/arch/x86/kernel/unwind_orc.c b/arch/x86/kernel/unwind_orc.c index ec88bbe08a32..4a96aa3de7d8 100644 --- a/arch/x86/kernel/unwind_orc.c +++ b/arch/x86/kernel/unwind_orc.c @@ -320,19 +320,12 @@ EXPORT_SYMBOL_GPL(unwind_get_return_address); unsigned long *unwind_get_return_address_ptr(struct unwind_state *state) { - struct task_struct *task = state->task; - if (unwind_done(state)) return NULL; if (state->regs) return &state->regs->ip; - if (task != current && state->sp == task->thread.sp) { - struct inactive_task_frame *frame = (void *)task->thread.sp; - return &frame->ret_addr; - } - if (state->sp) return (unsigned long *)state->sp - 1; @@ -662,7 +655,7 @@ void __unwind_start(struct unwind_state *state, struct task_struct *task, } else { struct inactive_task_frame *frame = (void *)task->thread.sp; - state->sp = task->thread.sp; + state->sp = task->thread.sp + sizeof(*frame); state->bp = READ_ONCE_NOCHECK(frame->bp); state->ip = READ_ONCE_NOCHECK(frame->ret_addr); state->signal = (void *)state->ip == ret_from_fork; -- cgit v1.2.3 From c7821c2d9c0dda0adf2bcf88e79b02a19a430be4 Mon Sep 17 00:00:00 2001 From: Cristian Marussi Date: Tue, 13 Oct 2020 14:31:09 +0100 Subject: firmware: arm_scmi: Fix locking in notifications When a protocol registers its events, the notification core takes care to rescan the hashtable of pending event handlers and activate all the possibly existent handlers referring to any of the events that are just registered by the new protocol. When a pending handler becomes active the core requests and enables the corresponding events in the SCMI firmware. If, for whatever reason, the enable fails, such invalid event handler must be finally removed and freed. Let us ensure to use the scmi_put_active_handler() helper which handles properly the needed additional locking. Failing to properly acquire all the needed mutexes exposes a race that leads to the following splat being observed: WARNING: CPU: 0 PID: 388 at lib/refcount.c:28 refcount_warn_saturate+0xf8/0x148 Hardware name: ARM LTD ARM Juno Development Platform/ARM Juno Development Platform, BIOS EDK II Jun 30 2020 pstate: 40000005 (nZcv daif -PAN -UAO BTYPE=--) pc : refcount_warn_saturate+0xf8/0x148 lr : refcount_warn_saturate+0xf8/0x148 Call trace: refcount_warn_saturate+0xf8/0x148 scmi_put_handler_unlocked.isra.10+0x204/0x208 scmi_put_handler+0x50/0xa0 scmi_unregister_notifier+0x1bc/0x240 scmi_notify_tester_remove+0x4c/0x68 [dummy_scmi_consumer] scmi_dev_remove+0x54/0x68 device_release_driver_internal+0x114/0x1e8 driver_detach+0x58/0xe8 bus_remove_driver+0x88/0xe0 driver_unregister+0x38/0x68 scmi_driver_unregister+0x1c/0x28 scmi_drv_exit+0x1c/0xae0 [dummy_scmi_consumer] __arm64_sys_delete_module+0x1a4/0x268 el0_svc_common.constprop.3+0x94/0x178 do_el0_svc+0x2c/0x98 el0_sync_handler+0x148/0x1a8 el0_sync+0x158/0x180 Link: https://lore.kernel.org/r/20201013133109.49821-1-cristian.marussi@arm.com Fixes: e7c215f358a35 ("firmware: arm_scmi: Add notification callbacks-registration") Signed-off-by: Cristian Marussi Signed-off-by: Sudeep Holla --- drivers/firmware/arm_scmi/notify.c | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-) diff --git a/drivers/firmware/arm_scmi/notify.c b/drivers/firmware/arm_scmi/notify.c index 2754f9d01636..c24e427dce0d 100644 --- a/drivers/firmware/arm_scmi/notify.c +++ b/drivers/firmware/arm_scmi/notify.c @@ -1403,15 +1403,21 @@ static void scmi_protocols_late_init(struct work_struct *work) "finalized PENDING handler - key:%X\n", hndl->key); ret = scmi_event_handler_enable_events(hndl); + if (ret) { + dev_dbg(ni->handle->dev, + "purging INVALID handler - key:%X\n", + hndl->key); + scmi_put_active_handler(ni, hndl); + } } else { ret = scmi_valid_pending_handler(ni, hndl); - } - if (ret) { - dev_dbg(ni->handle->dev, - "purging PENDING handler - key:%X\n", - hndl->key); - /* this hndl can be only a pending one */ - scmi_put_handler_unlocked(ni, hndl); + if (ret) { + dev_dbg(ni->handle->dev, + "purging PENDING handler - key:%X\n", + hndl->key); + /* this hndl can be only a pending one */ + scmi_put_handler_unlocked(ni, hndl); + } } } mutex_unlock(&ni->pending_mtx); -- cgit v1.2.3 From b9ceca6be43233845be70792be9b5ab315d2e010 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Tue, 13 Oct 2020 19:17:37 -0700 Subject: firmware: arm_scmi: Fix duplicate workqueue name When more than a single SCMI device are present in the system, the creation of the notification workqueue with the WQ_SYSFS flag will lead to the following sysfs duplicate node warning: sysfs: cannot create duplicate filename '/devices/virtual/workqueue/scmi_notify' CPU: 0 PID: 20 Comm: kworker/0:1 Not tainted 5.9.0-gdf4dd84a3f7d #29 Hardware name: Broadcom STB (Flattened Device Tree) Workqueue: events deferred_probe_work_func Backtrace: show_stack + 0x20/0x24 dump_stack + 0xbc/0xe0 sysfs_warn_dup + 0x70/0x80 sysfs_create_dir_ns + 0x15c/0x1a4 kobject_add_internal + 0x140/0x4d0 kobject_add + 0xc8/0x138 device_add + 0x1dc/0xc20 device_register + 0x24/0x28 workqueue_sysfs_register + 0xe4/0x1f0 alloc_workqueue + 0x448/0x6ac scmi_notification_init + 0x78/0x1dc scmi_probe + 0x268/0x4fc platform_drv_probe + 0x70/0xc8 really_probe + 0x184/0x728 driver_probe_device + 0xa4/0x278 __device_attach_driver + 0xe8/0x148 bus_for_each_drv + 0x108/0x158 __device_attach + 0x190/0x234 device_initial_probe + 0x1c/0x20 bus_probe_device + 0xdc/0xec deferred_probe_work_func + 0xd4/0x11c process_one_work + 0x420/0x8f0 worker_thread + 0x4fc/0x91c kthread + 0x21c/0x22c ret_from_fork + 0x14/0x20 kobject_add_internal failed for scmi_notify with -EEXIST, don't try to register things with the same name in the same directory. arm-scmi brcm_scmi@1: SCMI Notifications - Initialization Failed. arm-scmi brcm_scmi@1: SCMI Notifications NOT available. arm-scmi brcm_scmi@1: SCMI Protocol v1.0 'brcm-scmi:' Firmware version 0x1 Fix this by using dev_name(handle->dev) which guarantees that the name is unique and this also helps correlate which notification workqueue corresponds to which SCMI device instance. Link: https://lore.kernel.org/r/20201014021737.287340-1-f.fainelli@gmail.com Fixes: bd31b249692e ("firmware: arm_scmi: Add notification dispatch and delivery") Signed-off-by: Florian Fainelli [sudeep.holla: trimmed backtrace to remove all unwanted hexcodes and timestamps] Signed-off-by: Sudeep Holla --- drivers/firmware/arm_scmi/notify.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/firmware/arm_scmi/notify.c b/drivers/firmware/arm_scmi/notify.c index c24e427dce0d..ce336899d636 100644 --- a/drivers/firmware/arm_scmi/notify.c +++ b/drivers/firmware/arm_scmi/notify.c @@ -1474,7 +1474,7 @@ int scmi_notification_init(struct scmi_handle *handle) ni->gid = gid; ni->handle = handle; - ni->notify_wq = alloc_workqueue("scmi_notify", + ni->notify_wq = alloc_workqueue(dev_name(handle->dev), WQ_UNBOUND | WQ_FREEZABLE | WQ_SYSFS, 0); if (!ni->notify_wq) -- cgit v1.2.3 From c3b484c439b0bab7a698495f33ef16286a1000c4 Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Sun, 11 Oct 2020 19:51:21 -0700 Subject: x86/syscalls: Document the fact that syscalls 512-547 are a legacy mistake Since this commit: 6365b842aae4 ("x86/syscalls: Split the x32 syscalls into their own table") there is no need for special x32-specific syscall numbers. I forgot to update the comments in syscall_64.tbl. Add comments to make it clear to future contributors that this range is a legacy wart. Reported-by: Jessica Clarke Signed-off-by: Andy Lutomirski Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/6c56fb4ddd18fc60a238eb4d867e4b3d97c6351e.1602471055.git.luto@kernel.org --- arch/x86/entry/syscalls/syscall_64.tbl | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/arch/x86/entry/syscalls/syscall_64.tbl b/arch/x86/entry/syscalls/syscall_64.tbl index f30d6ae9a688..4adb5d2a3319 100644 --- a/arch/x86/entry/syscalls/syscall_64.tbl +++ b/arch/x86/entry/syscalls/syscall_64.tbl @@ -363,10 +363,10 @@ 439 common faccessat2 sys_faccessat2 # -# x32-specific system call numbers start at 512 to avoid cache impact -# for native 64-bit operation. The __x32_compat_sys stubs are created -# on-the-fly for compat_sys_*() compatibility system calls if X86_X32 -# is defined. +# Due to a historical design error, certain syscalls are numbered differently +# in x32 as compared to native x86_64. These syscalls have numbers 512-547. +# Do not add new syscalls to this range. Numbers 548 and above are available +# for non-x32 use. # 512 x32 rt_sigaction compat_sys_rt_sigaction 513 x32 rt_sigreturn compat_sys_x32_rt_sigreturn @@ -404,3 +404,5 @@ 545 x32 execveat compat_sys_execveat 546 x32 preadv2 compat_sys_preadv64v2 547 x32 pwritev2 compat_sys_pwritev64v2 +# This is the end of the legacy x32 range. Numbers 548 and above are +# not special and are not to be used for x32-specific syscalls. -- cgit v1.2.3 From 5be1805dc3961ce0465bcb0beab85fe8580af08d Mon Sep 17 00:00:00 2001 From: Georgi Djakov Date: Mon, 12 Oct 2020 22:40:34 +0300 Subject: interconnect: qcom: sdm845: Enable keepalive for the MM1 BCM After enabling interconnect scaling for display on the db845c board, in certain configurations the board hangs, while the following errors are observed on the console: Error sending AMC RPMH requests (-110) qcom_rpmh TCS Busy, retrying RPMH message send: addr=0x50000 qcom_rpmh TCS Busy, retrying RPMH message send: addr=0x50000 qcom_rpmh TCS Busy, retrying RPMH message send: addr=0x50000 ... In this specific case, the above is related to one of the sequencers being stuck, while client drivers are returning from probe and trying to disable the currently unused clock and interconnect resources. Generally we want to keep the multimedia NoC enabled like the rest of the NoCs, so let's set the keepalive flag on it too. Fixes: aae57773fbe0 ("interconnect: qcom: sdm845: Split qnodes into their respective NoCs") Reported-by: Amit Pundir Reviewed-by: Mike Tipton Tested-by: John Stultz Link: https://lore.kernel.org/r/20201012194034.26944-1-georgi.djakov@linaro.org Signed-off-by: Georgi Djakov --- drivers/interconnect/qcom/sdm845.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/interconnect/qcom/sdm845.c b/drivers/interconnect/qcom/sdm845.c index d79e3163e2c3..67e8fc3396f8 100644 --- a/drivers/interconnect/qcom/sdm845.c +++ b/drivers/interconnect/qcom/sdm845.c @@ -151,7 +151,7 @@ DEFINE_QBCM(bcm_mc0, "MC0", true, &ebi); DEFINE_QBCM(bcm_sh0, "SH0", true, &qns_llcc); DEFINE_QBCM(bcm_mm0, "MM0", false, &qns_mem_noc_hf); DEFINE_QBCM(bcm_sh1, "SH1", false, &qns_apps_io); -DEFINE_QBCM(bcm_mm1, "MM1", false, &qxm_camnoc_hf0_uncomp, &qxm_camnoc_hf1_uncomp, &qxm_camnoc_sf_uncomp, &qxm_camnoc_hf0, &qxm_camnoc_hf1, &qxm_mdp0, &qxm_mdp1); +DEFINE_QBCM(bcm_mm1, "MM1", true, &qxm_camnoc_hf0_uncomp, &qxm_camnoc_hf1_uncomp, &qxm_camnoc_sf_uncomp, &qxm_camnoc_hf0, &qxm_camnoc_hf1, &qxm_mdp0, &qxm_mdp1); DEFINE_QBCM(bcm_sh2, "SH2", false, &qns_memnoc_snoc); DEFINE_QBCM(bcm_mm2, "MM2", false, &qns2_mem_noc); DEFINE_QBCM(bcm_sh3, "SH3", false, &acm_tcu); -- cgit v1.2.3 From d3703b3e255f56d543aac183f8aafdbfd7096559 Mon Sep 17 00:00:00 2001 From: Georgi Djakov Date: Tue, 13 Oct 2020 16:59:11 +0300 Subject: interconnect: Aggregate before setting initial bandwidth When setting the initial bandwidth, make sure to call the aggregate() function (if such is implemented for the current provider), to handle cases when data needs to be aggregated first. Fixes: b1d681d8d324 ("interconnect: Add sync state support") Acked-by: Saravana Kannan Link: https://lore.kernel.org/r/20201013135913.29059-1-georgi.djakov@linaro.org Signed-off-by: Georgi Djakov --- drivers/interconnect/core.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/interconnect/core.c b/drivers/interconnect/core.c index eea47b4c84aa..974a66725d09 100644 --- a/drivers/interconnect/core.c +++ b/drivers/interconnect/core.c @@ -971,6 +971,9 @@ void icc_node_add(struct icc_node *node, struct icc_provider *provider) } node->avg_bw = node->init_avg; node->peak_bw = node->init_peak; + if (provider->aggregate) + provider->aggregate(node, 0, node->init_avg, node->init_peak, + &node->avg_bw, &node->peak_bw); provider->set(node, node); node->avg_bw = 0; node->peak_bw = 0; -- cgit v1.2.3 From 0f221a729049ab727c87b0fe47e309b952d879ce Mon Sep 17 00:00:00 2001 From: Georgi Djakov Date: Tue, 13 Oct 2020 16:59:12 +0300 Subject: interconnect: qcom: sdm845: Init BCMs before creating the nodes Currently if we use sync_state, by default the bandwidth is maxed out, but in order to set this in hardware, the BCMs (Bus Clock Managers) need to be initialized first. Move the BCM initialization before creating the nodes to fix this. Fixes: 7d3b0b0d8184 ("interconnect: qcom: Use icc_sync_state") Acked-by: Saravana Kannan Link: https://lore.kernel.org/r/20201013135913.29059-2-georgi.djakov@linaro.org Signed-off-by: Georgi Djakov --- drivers/interconnect/qcom/sdm845.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/interconnect/qcom/sdm845.c b/drivers/interconnect/qcom/sdm845.c index 67e8fc3396f8..5304aea3b058 100644 --- a/drivers/interconnect/qcom/sdm845.c +++ b/drivers/interconnect/qcom/sdm845.c @@ -489,6 +489,9 @@ static int qnoc_probe(struct platform_device *pdev) return ret; } + for (i = 0; i < qp->num_bcms; i++) + qcom_icc_bcm_init(qp->bcms[i], &pdev->dev); + for (i = 0; i < num_nodes; i++) { size_t j; @@ -512,9 +515,6 @@ static int qnoc_probe(struct platform_device *pdev) } data->num_nodes = num_nodes; - for (i = 0; i < qp->num_bcms; i++) - qcom_icc_bcm_init(qp->bcms[i], &pdev->dev); - platform_set_drvdata(pdev, qp); return 0; -- cgit v1.2.3 From 599809540f173f572c32a35d712accdc14d6357c Mon Sep 17 00:00:00 2001 From: Georgi Djakov Date: Tue, 13 Oct 2020 16:59:13 +0300 Subject: interconnect: qcom: sc7180: Init BCMs before creating the nodes Currently if we use sync_state, by default the bandwidth is maxed out, but in order to set this in hardware, the BCMs (Bus Clock Managers) need to be initialized first. Move the BCM initialization before creating the nodes to fix this. Fixes: 7d3b0b0d8184 ("interconnect: qcom: Use icc_sync_state") Acked-by: Saravana Kannan Link: https://lore.kernel.org/r/20201013135913.29059-3-georgi.djakov@linaro.org Signed-off-by: Georgi Djakov --- drivers/interconnect/qcom/sc7180.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/interconnect/qcom/sc7180.c b/drivers/interconnect/qcom/sc7180.c index bf11b82ed55c..8d9044ed18ab 100644 --- a/drivers/interconnect/qcom/sc7180.c +++ b/drivers/interconnect/qcom/sc7180.c @@ -553,6 +553,9 @@ static int qnoc_probe(struct platform_device *pdev) return ret; } + for (i = 0; i < qp->num_bcms; i++) + qcom_icc_bcm_init(qp->bcms[i], &pdev->dev); + for (i = 0; i < num_nodes; i++) { size_t j; @@ -576,9 +579,6 @@ static int qnoc_probe(struct platform_device *pdev) } data->num_nodes = num_nodes; - for (i = 0; i < qp->num_bcms; i++) - qcom_icc_bcm_init(qp->bcms[i], &pdev->dev); - platform_set_drvdata(pdev, qp); return 0; -- cgit v1.2.3 From b3e1ea16fb39fb6e1a1cf1dbdd6738531de3dc7d Mon Sep 17 00:00:00 2001 From: Jisheng Zhang Date: Thu, 15 Oct 2020 17:41:15 +0800 Subject: mmc: sdhci: Use Auto CMD Auto Select only when v4_mode is true sdhci-of-dwcmshc meets an eMMC read performance regression with below command after commit 427b6514d095 ("mmc: sdhci: Add Auto CMD Auto Select support"): dd if=/dev/mmcblk0 of=/dev/null bs=8192 count=100000 Before the commit, the above command gives 120MB/s After the commit, the above command gives 51.3 MB/s So it looks like sdhci-of-dwcmshc expects Version 4 Mode for Auto CMD Auto Select. Fix the performance degradation by ensuring v4_mode is true to use Auto CMD Auto Select. Fixes: 427b6514d095 ("mmc: sdhci: Add Auto CMD Auto Select support") Signed-off-by: Jisheng Zhang Acked-by: Adrian Hunter Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20201015174115.4cf2c19a@xhacker.debian Signed-off-by: Ulf Hansson --- drivers/mmc/host/sdhci.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c index 592a55a34b58..3561ae8a481a 100644 --- a/drivers/mmc/host/sdhci.c +++ b/drivers/mmc/host/sdhci.c @@ -1384,9 +1384,11 @@ static inline void sdhci_auto_cmd_select(struct sdhci_host *host, /* * In case of Version 4.10 or later, use of 'Auto CMD Auto * Select' is recommended rather than use of 'Auto CMD12 - * Enable' or 'Auto CMD23 Enable'. + * Enable' or 'Auto CMD23 Enable'. We require Version 4 Mode + * here because some controllers (e.g sdhci-of-dwmshc) expect it. */ - if (host->version >= SDHCI_SPEC_410 && (use_cmd12 || use_cmd23)) { + if (host->version >= SDHCI_SPEC_410 && host->v4_mode && + (use_cmd12 || use_cmd23)) { *mode |= SDHCI_TRNS_AUTO_SEL; ctrl2 = sdhci_readw(host, SDHCI_HOST_CONTROL2); -- cgit v1.2.3 From 1afc708dcae41bc538fda0f85d4f0ecf276fa505 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Wed, 14 Oct 2020 10:10:36 +0100 Subject: btrfs: fix relocation failure due to race with fallocate When doing a fallocate() we have a short time window, after reserving an extent and before starting a transaction, where if relocation for the block group containing the reserved extent happens, we can end up missing the extent in the data relocation inode causing relocation to fail later. This only happens when we don't pass a transaction to the internal fallocate function __btrfs_prealloc_file_range(), which is for all the cases where fallocate() is called from user space (the internal use cases include space cache extent allocation and relocation). When the race triggers the relocation failure, it produces a trace like the following: [200611.995995] ------------[ cut here ]------------ [200611.997084] BTRFS: Transaction aborted (error -2) [200611.998208] WARNING: CPU: 3 PID: 235845 at fs/btrfs/ctree.c:1074 __btrfs_cow_block+0x3a0/0x5b0 [btrfs] [200611.999042] Modules linked in: dm_thin_pool dm_persistent_data (...) [200612.003287] CPU: 3 PID: 235845 Comm: btrfs Not tainted 5.9.0-rc6-btrfs-next-69 #1 [200612.004442] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 [200612.006186] RIP: 0010:__btrfs_cow_block+0x3a0/0x5b0 [btrfs] [200612.007110] Code: 1b 00 00 02 72 2a 83 f8 fb 0f 84 b8 01 (...) [200612.007341] BTRFS warning (device sdb): Skipping commit of aborted transaction. [200612.008959] RSP: 0018:ffffaee38550f918 EFLAGS: 00010286 [200612.009672] BTRFS: error (device sdb) in cleanup_transaction:1901: errno=-30 Readonly filesystem [200612.010428] RAX: 0000000000000000 RBX: ffff9174d96f4000 RCX: 0000000000000000 [200612.011078] BTRFS info (device sdb): forced readonly [200612.011862] RDX: 0000000000000001 RSI: ffffffffa8161978 RDI: 00000000ffffffff [200612.013215] RBP: ffff9172569a0f80 R08: 0000000000000000 R09: 0000000000000000 [200612.014263] R10: 0000000000000000 R11: 0000000000000000 R12: ffff9174b8403b88 [200612.015203] R13: ffff9174b8400a88 R14: ffff9174c90f1000 R15: ffff9174a5a60e08 [200612.016182] FS: 00007fa55cf878c0(0000) GS:ffff9174ece00000(0000) knlGS:0000000000000000 [200612.017174] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [200612.018418] CR2: 00007f8fb8048148 CR3: 0000000428a46003 CR4: 00000000003706e0 [200612.019510] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [200612.020648] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [200612.021520] Call Trace: [200612.022434] btrfs_cow_block+0x10b/0x250 [btrfs] [200612.023407] do_relocation+0x54e/0x7b0 [btrfs] [200612.024343] ? do_raw_spin_unlock+0x4b/0xc0 [200612.025280] ? _raw_spin_unlock+0x29/0x40 [200612.026200] relocate_tree_blocks+0x3bc/0x6d0 [btrfs] [200612.027088] relocate_block_group+0x2f3/0x600 [btrfs] [200612.027961] btrfs_relocate_block_group+0x15e/0x340 [btrfs] [200612.028896] btrfs_relocate_chunk+0x38/0x110 [btrfs] [200612.029772] btrfs_balance+0xb22/0x1790 [btrfs] [200612.030601] ? btrfs_ioctl_balance+0x253/0x380 [btrfs] [200612.031414] btrfs_ioctl_balance+0x2cf/0x380 [btrfs] [200612.032279] btrfs_ioctl+0x620/0x36f0 [btrfs] [200612.033077] ? _raw_spin_unlock+0x29/0x40 [200612.033948] ? handle_mm_fault+0x116d/0x1ca0 [200612.034749] ? up_read+0x18/0x240 [200612.035542] ? __x64_sys_ioctl+0x83/0xb0 [200612.036244] __x64_sys_ioctl+0x83/0xb0 [200612.037269] do_syscall_64+0x33/0x80 [200612.038190] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [200612.038976] RIP: 0033:0x7fa55d07ed87 [200612.040127] Code: 00 00 00 48 8b 05 09 91 0c 00 64 c7 00 26 (...) [200612.041669] RSP: 002b:00007ffd5ebf03e8 EFLAGS: 00000206 ORIG_RAX: 0000000000000010 [200612.042437] RAX: ffffffffffffffda RBX: 0000000000000001 RCX: 00007fa55d07ed87 [200612.043511] RDX: 00007ffd5ebf0470 RSI: 00000000c4009420 RDI: 0000000000000003 [200612.044250] RBP: 0000000000000003 R08: 000055d8362642a0 R09: 00007fa55d148be0 [200612.044963] R10: fffffffffffff52e R11: 0000000000000206 R12: 00007ffd5ebf1614 [200612.045683] R13: 00007ffd5ebf0470 R14: 0000000000000002 R15: 00007ffd5ebf0470 [200612.046361] irq event stamp: 0 [200612.047040] hardirqs last enabled at (0): [<0000000000000000>] 0x0 [200612.047725] hardirqs last disabled at (0): [] copy_process+0x823/0x1bc0 [200612.048387] softirqs last enabled at (0): [] copy_process+0x823/0x1bc0 [200612.049024] softirqs last disabled at (0): [<0000000000000000>] 0x0 [200612.049722] ---[ end trace 49006c6876e65227 ]--- The race happens like this: 1) Task A starts an fallocate() (plain or zero range) and it calls __btrfs_prealloc_file_range() with the 'trans' parameter set to NULL; 2) Task A calls btrfs_reserve_extent() and gets an extent that belongs to block group X; 3) Before task A gets into btrfs_replace_file_extents(), through the call to insert_prealloc_file_extent(), task B starts relocation of block group X; 4) Task B enters btrfs_relocate_block_group() and it sets block group X to RO mode; 5) Task B enters relocate_block_group(), it calls prepare_to_relocate() whichs joins/starts a transaction and then commits the transaction; 6) Task B then starts scanning the extent tree looking for extents that belong to block group X - it does not find yet the extent reserved by task A, since that extent was not yet added to the extent tree, as its delayed reference was not even yet created at this point; 7) The data relocation inode ends up not having the extent reserved by task A associated to it; 8) Task A then starts a transaction through btrfs_replace_file_extents(), inserts a file extent item in the subvolume tree pointing to the reserved extent and creates a delayed reference for it; 9) Task A finishes and returns success to user space; 10) Later on, while relocation is still in progress, the leaf where task A inserted the new file extent item is COWed, so we end up at __btrfs_cow_block(), which calls btrfs_reloc_cow_block(), and that in turn calls relocation.c:replace_file_extents(); 11) At relocation.c:replace_file_extents() we iterate over all the items in the leaf and find the file extent item pointing to the extent that was allocated by task A, and then call relocation.c:get_new_location(), to find the new location for the extent; 12) However relocation.c:get_new_location() fails, returning -ENOENT, because it couldn't find a corresponding file extent item associated with the data relocation inode. This is because the extent was not seen in the extent tree at step 6). The -ENOENT error is propagated to __btrfs_cow_block(), which aborts the transaction. So fix this simply by decrementing the block group's number of reservations after calling insert_prealloc_file_extent(), as relocation waits for that counter to go down to zero before calling prepare_to_relocate() and start looking for extents in the extent tree. This issue only started to happen recently as of commit 8fccebfa534c79 ("btrfs: fix metadata reservation for fallocate that leads to transaction aborts"), because now we can reserve an extent before starting/joining a transaction, and previously we always did it after that, so relocation ended up waiting for a concurrent fallocate() to finish because before searching for the extents of the block group, it starts/joins a transaction and then commits it (at prepare_to_relocate()), which made it wait for the fallocate task to complete first. Fixes: 8fccebfa534c79 ("btrfs: fix metadata reservation for fallocate that leads to transaction aborts") Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/inode.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 936c3137c646..da58c58ef9aa 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -9672,10 +9672,16 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, * clear_offset by our extent size. */ clear_offset += ins.offset; - btrfs_dec_block_group_reservations(fs_info, ins.objectid); last_alloc = ins.offset; trans = insert_prealloc_file_extent(trans, inode, &ins, cur_offset); + /* + * Now that we inserted the prealloc extent we can finally + * decrement the number of reservations in the block group. + * If we did it before, we could race with relocation and have + * relocation miss the reserved extent, making it fail later. + */ + btrfs_dec_block_group_reservations(fs_info, ins.objectid); if (IS_ERR(trans)) { ret = PTR_ERR(trans); btrfs_free_reserved_extent(fs_info, ins.objectid, -- cgit v1.2.3 From f3362f0c18174a1f334a419ab7d567a36bd1b3f3 Mon Sep 17 00:00:00 2001 From: Anand Moon Date: Tue, 20 Oct 2020 14:01:41 +0200 Subject: arm64: dts: amlogic: add missing ethernet reset ID Add reset external reset of the ethernet mac controller Signed-off-by: Anand Moon Signed-off-by: Jerome Brunet Reviewed-by: Neil Armstrong Reviewed-by: Martin Blumenstingl Signed-off-by: Kevin Hilman Link: https://lore.kernel.org/r/20201020120141.298240-1-jbrunet@baylibre.com --- arch/arm64/boot/dts/amlogic/meson-axg.dtsi | 2 ++ arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi | 2 ++ arch/arm64/boot/dts/amlogic/meson-gx.dtsi | 3 +++ 3 files changed, 7 insertions(+) diff --git a/arch/arm64/boot/dts/amlogic/meson-axg.dtsi b/arch/arm64/boot/dts/amlogic/meson-axg.dtsi index fae48efae83e..724ee179b316 100644 --- a/arch/arm64/boot/dts/amlogic/meson-axg.dtsi +++ b/arch/arm64/boot/dts/amlogic/meson-axg.dtsi @@ -227,6 +227,8 @@ "timing-adjustment"; rx-fifo-depth = <4096>; tx-fifo-depth = <2048>; + resets = <&reset RESET_ETHERNET>; + reset-names = "stmmaceth"; status = "disabled"; }; diff --git a/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi b/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi index c95ebe615176..8514fe6a275a 100644 --- a/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi +++ b/arch/arm64/boot/dts/amlogic/meson-g12-common.dtsi @@ -224,6 +224,8 @@ "timing-adjustment"; rx-fifo-depth = <4096>; tx-fifo-depth = <2048>; + resets = <&reset RESET_ETHERNET>; + reset-names = "stmmaceth"; status = "disabled"; mdio0: mdio { diff --git a/arch/arm64/boot/dts/amlogic/meson-gx.dtsi b/arch/arm64/boot/dts/amlogic/meson-gx.dtsi index 0edd137151f8..726b91d3a905 100644 --- a/arch/arm64/boot/dts/amlogic/meson-gx.dtsi +++ b/arch/arm64/boot/dts/amlogic/meson-gx.dtsi @@ -13,6 +13,7 @@ #include #include #include +#include #include / { @@ -575,6 +576,8 @@ interrupt-names = "macirq"; rx-fifo-depth = <4096>; tx-fifo-depth = <2048>; + resets = <&reset RESET_ETHERNET>; + reset-names = "stmmaceth"; power-domains = <&pwrc PWRC_GXBB_ETHERNET_MEM_ID>; status = "disabled"; }; -- cgit v1.2.3 From 266cd33b59138501579c64648f54b93eab2e5adf Mon Sep 17 00:00:00 2001 From: Georgi Djakov Date: Thu, 22 Oct 2020 13:26:26 +0300 Subject: interconnect: qcom: Ensure that the floor bandwidth value is enforced Take into account the initial bandwidth from the framework and update the internal sum and max values before committing if needed. This will ensure that the floor bandwidth values are enforced until the providers get into sync state. Fixes: 7d3b0b0d8184 ("interconnect: qcom: Use icc_sync_state") Tested-by: Akash Asthana Link: https://lore.kernel.org/r/20201021155938.9223-1-georgi.djakov@linaro.org Signed-off-by: Georgi Djakov --- drivers/interconnect/qcom/icc-rpmh.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/interconnect/qcom/icc-rpmh.c b/drivers/interconnect/qcom/icc-rpmh.c index cf10a4b9611b..bf01d09dba6c 100644 --- a/drivers/interconnect/qcom/icc-rpmh.c +++ b/drivers/interconnect/qcom/icc-rpmh.c @@ -79,6 +79,7 @@ EXPORT_SYMBOL_GPL(qcom_icc_aggregate); int qcom_icc_set(struct icc_node *src, struct icc_node *dst) { struct qcom_icc_provider *qp; + struct qcom_icc_node *qn; struct icc_node *node; if (!src) @@ -87,6 +88,12 @@ int qcom_icc_set(struct icc_node *src, struct icc_node *dst) node = src; qp = to_qcom_provider(node->provider); + qn = node->data; + + qn->sum_avg[QCOM_ICC_BUCKET_AMC] = max_t(u64, qn->sum_avg[QCOM_ICC_BUCKET_AMC], + node->avg_bw); + qn->max_peak[QCOM_ICC_BUCKET_AMC] = max_t(u64, qn->max_peak[QCOM_ICC_BUCKET_AMC], + node->peak_bw); qcom_icc_bcm_voter_commit(qp->voter); -- cgit v1.2.3 From f8e48a3dca060e80f672d398d181db1298fbc86c Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 22 Oct 2020 12:23:02 +0200 Subject: lockdep: Fix preemption WARN for spurious IRQ-enable It is valid (albeit uncommon) to call local_irq_enable() without first having called local_irq_disable(). In this case we enter lockdep_hardirqs_on*() with IRQs enabled and trip a preemption warning for using __this_cpu_read(). Use this_cpu_read() instead to avoid the warning. Fixes: 4d004099a6 ("lockdep: Fix lockdep recursion") Reported-by: syzbot+53f8ce8bbc07924b6417@syzkaller.appspotmail.com Reported-by: kernel test robot Signed-off-by: Peter Zijlstra (Intel) --- kernel/locking/lockdep.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index 3e99dfef8408..fc206aefa970 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -4057,7 +4057,7 @@ void lockdep_hardirqs_on_prepare(unsigned long ip) if (unlikely(in_nmi())) return; - if (unlikely(__this_cpu_read(lockdep_recursion))) + if (unlikely(this_cpu_read(lockdep_recursion))) return; if (unlikely(lockdep_hardirqs_enabled())) { @@ -4126,7 +4126,7 @@ void noinstr lockdep_hardirqs_on(unsigned long ip) goto skip_checks; } - if (unlikely(__this_cpu_read(lockdep_recursion))) + if (unlikely(this_cpu_read(lockdep_recursion))) return; if (lockdep_hardirqs_enabled()) { -- cgit v1.2.3 From abee7c494d8c41bb388839bccc47e06247f0d7de Mon Sep 17 00:00:00 2001 From: Juergen Gross Date: Fri, 9 Oct 2020 16:42:25 +0200 Subject: x86/alternative: Don't call text_poke() in lazy TLB mode When running in lazy TLB mode the currently active page tables might be the ones of a previous process, e.g. when running a kernel thread. This can be problematic in case kernel code is being modified via text_poke() in a kernel thread, and on another processor exit_mmap() is active for the process which was running on the first cpu before the kernel thread. As text_poke() is using a temporary address space and the former address space (obtained via cpu_tlbstate.loaded_mm) is restored afterwards, there is a race possible in case the cpu on which exit_mmap() is running wants to make sure there are no stale references to that address space on any cpu active (this e.g. is required when running as a Xen PV guest, where this problem has been observed and analyzed). In order to avoid that, drop off TLB lazy mode before switching to the temporary address space. Fixes: cefa929c034eb5d ("x86/mm: Introduce temporary mm structs") Signed-off-by: Juergen Gross Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20201009144225.12019-1-jgross@suse.com --- arch/x86/kernel/alternative.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index cdaab30880b9..cd6be6f143e8 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -807,6 +807,15 @@ static inline temp_mm_state_t use_temporary_mm(struct mm_struct *mm) temp_mm_state_t temp_state; lockdep_assert_irqs_disabled(); + + /* + * Make sure not to be in TLB lazy mode, as otherwise we'll end up + * with a stale address space WITHOUT being in lazy mode after + * restoring the previous mm. + */ + if (this_cpu_read(cpu_tlbstate.is_lazy)) + leave_mm(smp_processor_id()); + temp_state.mm = this_cpu_read(cpu_tlbstate.loaded_mm); switch_mm_irqs_off(NULL, mm, current); -- cgit v1.2.3 From 938f324e7df25e89226c6fe137028af73cd6160b Mon Sep 17 00:00:00 2001 From: Guido Günther Date: Tue, 13 Oct 2020 12:32:46 +0200 Subject: drm/panel: mantix: Don't dereference NULL mode MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Don't dereference mode which was just NULL checked. Fixes: 72967d5616d3 ("drm/panel: Add panel driver for the Mantix MLAF057WE51-X DSI panel") Signed-off-by: Guido Günther Reported-by: Dan Carpenter Signed-off-by: Sam Ravnborg Reviewed-by: Sam Ravnborg Link: https://patchwork.freedesktop.org/patch/msgid/659158549f3c6cc1c71ceed0943e760e861c1206.1602584953.git.agx@sigxcpu.org --- drivers/gpu/drm/panel/panel-mantix-mlaf057we51.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/panel/panel-mantix-mlaf057we51.c b/drivers/gpu/drm/panel/panel-mantix-mlaf057we51.c index 3482e28e30fc..4a7fbf64bb7a 100644 --- a/drivers/gpu/drm/panel/panel-mantix-mlaf057we51.c +++ b/drivers/gpu/drm/panel/panel-mantix-mlaf057we51.c @@ -204,7 +204,7 @@ static int mantix_get_modes(struct drm_panel *panel, if (!mode) { dev_err(ctx->dev, "Failed to add mode %ux%u@%u\n", default_mode.hdisplay, default_mode.vdisplay, - drm_mode_vrefresh(mode)); + drm_mode_vrefresh(&default_mode)); return -ENOMEM; } -- cgit v1.2.3 From 6af672523fe4bd71f5e70c50258fd0fc09663fae Mon Sep 17 00:00:00 2001 From: Guido Günther Date: Tue, 13 Oct 2020 12:32:47 +0200 Subject: drm/panel: mantix: Fix panel reset MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The mantix panel needs two reset lines (RESX and TP_RSTN) deasserted to output an image. Only deasserting RESX is not enough and the display will stay blank. Deassert in prepare() and assert in unprepare() to keep device held in reset when off. Fixes: 72967d5616d3 ("drm/panel: Add panel driver for the Mantix MLAF057WE51-X DSI panel") Signed-off-by: Guido Günther Signed-off-by: Sam Ravnborg Reviewed-by: Sam Ravnborg Link: https://patchwork.freedesktop.org/patch/msgid/ba71a8ab010d263a8058dd4f711e3bcd95877bf2.1602584953.git.agx@sigxcpu.org --- drivers/gpu/drm/panel/panel-mantix-mlaf057we51.c | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/panel/panel-mantix-mlaf057we51.c b/drivers/gpu/drm/panel/panel-mantix-mlaf057we51.c index 4a7fbf64bb7a..0c5f22e95c2d 100644 --- a/drivers/gpu/drm/panel/panel-mantix-mlaf057we51.c +++ b/drivers/gpu/drm/panel/panel-mantix-mlaf057we51.c @@ -26,7 +26,9 @@ struct mantix { struct device *dev; struct drm_panel panel; + struct gpio_desc *reset_gpio; + struct gpio_desc *tp_rstn_gpio; struct regulator *avdd; struct regulator *avee; @@ -124,6 +126,10 @@ static int mantix_unprepare(struct drm_panel *panel) { struct mantix *ctx = panel_to_mantix(panel); + gpiod_set_value_cansleep(ctx->tp_rstn_gpio, 1); + usleep_range(5000, 6000); + gpiod_set_value_cansleep(ctx->reset_gpio, 1); + regulator_disable(ctx->avee); regulator_disable(ctx->avdd); /* T11 */ @@ -165,13 +171,10 @@ static int mantix_prepare(struct drm_panel *panel) return ret; } - /* T3+T5 */ - usleep_range(10000, 12000); - - gpiod_set_value_cansleep(ctx->reset_gpio, 1); - usleep_range(5150, 7000); - + /* T3 + T4 + time for voltage to become stable: */ + usleep_range(6000, 7000); gpiod_set_value_cansleep(ctx->reset_gpio, 0); + gpiod_set_value_cansleep(ctx->tp_rstn_gpio, 0); /* T6 */ msleep(50); @@ -236,12 +239,18 @@ static int mantix_probe(struct mipi_dsi_device *dsi) if (!ctx) return -ENOMEM; - ctx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_LOW); + ctx->reset_gpio = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH); if (IS_ERR(ctx->reset_gpio)) { dev_err(dev, "cannot get reset gpio\n"); return PTR_ERR(ctx->reset_gpio); } + ctx->tp_rstn_gpio = devm_gpiod_get(dev, "mantix,tp-rstn", GPIOD_OUT_HIGH); + if (IS_ERR(ctx->tp_rstn_gpio)) { + dev_err(dev, "cannot get tp-rstn gpio\n"); + return PTR_ERR(ctx->tp_rstn_gpio); + } + mipi_dsi_set_drvdata(dsi, ctx); ctx->dev = dev; -- cgit v1.2.3 From 2580a493a97da4a302cb66251b558bfc04c16e68 Mon Sep 17 00:00:00 2001 From: Guido Günther Date: Tue, 13 Oct 2020 12:32:48 +0200 Subject: dt-binding: display: Require two resets on mantix panel MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We need to reset both for the panel to show an image. Fixes: b9ab1248d801 ("dt-bindings: Add Mantix MLAF057WE51-X panel bindings") Signed-off-by: Guido Günther Signed-off-by: Sam Ravnborg Reviewed-by: Sam Ravnborg Link: https://patchwork.freedesktop.org/patch/msgid/15d3dc7eb4e031f380be1298ed3ac9085626f26b.1602584953.git.agx@sigxcpu.org --- .../devicetree/bindings/display/panel/mantix,mlaf057we51-x.yaml | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/Documentation/devicetree/bindings/display/panel/mantix,mlaf057we51-x.yaml b/Documentation/devicetree/bindings/display/panel/mantix,mlaf057we51-x.yaml index 937323cc9aaa..51f423297ec8 100644 --- a/Documentation/devicetree/bindings/display/panel/mantix,mlaf057we51-x.yaml +++ b/Documentation/devicetree/bindings/display/panel/mantix,mlaf057we51-x.yaml @@ -37,6 +37,9 @@ properties: reset-gpios: true + 'mantix,tp-rstn-gpios': + description: second reset line that triggers DSI config load + backlight: true required: @@ -63,6 +66,7 @@ examples: avee-supply = <®_avee>; vddi-supply = <®_1v8_p>; reset-gpios = <&gpio1 29 GPIO_ACTIVE_LOW>; + mantix,tp-rstn-gpios = <&gpio1 24 GPIO_ACTIVE_LOW>; backlight = <&backlight>; }; }; -- cgit v1.2.3 From 0add6e9b88d0632a25323aaf4987dbacb0e4ae64 Mon Sep 17 00:00:00 2001 From: Michael Walle Date: Fri, 23 Oct 2020 00:23:37 +0200 Subject: mmc: sdhci-of-esdhc: set timeout to max before tuning On rare occations there is the following error: mmc0: Tuning timeout, falling back to fixed sampling clock There are SD cards which takes a significant longer time to reply to the first CMD19 command. The eSDHC takes the data timeout value into account during the tuning period. The SDHCI core doesn't explicitly set this timeout for the tuning procedure. Thus on the slow cards, there might be a spurious "Buffer Read Ready" interrupt, which in turn triggers a wrong sequence of events. In the end this will lead to an unsuccessful tuning procedure and to the above error. To workaround this, set the timeout to the maximum value (which is the best we can do) and the SDHCI core will take care of the proper timeout handling. Fixes: ba49cbd0936e ("mmc: sdhci-of-esdhc: add tuning support") Signed-off-by: Michael Walle Acked-by: Adrian Hunter Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20201022222337.19857-1-michael@walle.cc Signed-off-by: Ulf Hansson --- drivers/mmc/host/sdhci-of-esdhc.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/drivers/mmc/host/sdhci-of-esdhc.c b/drivers/mmc/host/sdhci-of-esdhc.c index 0b45eff6fed4..baf7801a1804 100644 --- a/drivers/mmc/host/sdhci-of-esdhc.c +++ b/drivers/mmc/host/sdhci-of-esdhc.c @@ -1052,6 +1052,17 @@ static int esdhc_execute_tuning(struct mmc_host *mmc, u32 opcode) esdhc_tuning_block_enable(host, true); + /* + * The eSDHC controller takes the data timeout value into account + * during tuning. If the SD card is too slow sending the response, the + * timer will expire and a "Buffer Read Ready" interrupt without data + * is triggered. This leads to tuning errors. + * + * Just set the timeout to the maximum value because the core will + * already take care of it in sdhci_send_tuning(). + */ + sdhci_writeb(host, 0xe, SDHCI_TIMEOUT_CONTROL); + hs400_tuning = host->flags & SDHCI_HS400_TUNING; do { -- cgit v1.2.3 From 3f1b623a1be92103386bcab818e25885d6be9419 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Fri, 23 Oct 2020 17:00:41 +0800 Subject: vdpa: introduce config op to get valid iova range This patch introduce a config op to get valid iova range from the vDPA device. Signed-off-by: Jason Wang Link: https://lore.kernel.org/r/20201023090043.14430-2-jasowang@redhat.com Signed-off-by: Michael S. Tsirkin --- include/linux/vdpa.h | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/include/linux/vdpa.h b/include/linux/vdpa.h index eae0bfd87d91..30bc7a7223bb 100644 --- a/include/linux/vdpa.h +++ b/include/linux/vdpa.h @@ -52,6 +52,16 @@ struct vdpa_device { int nvqs; }; +/** + * vDPA IOVA range - the IOVA range support by the device + * @first: start of the IOVA range + * @last: end of the IOVA range + */ +struct vdpa_iova_range { + u64 first; + u64 last; +}; + /** * vDPA_config_ops - operations for configuring a vDPA device. * Note: vDPA device drivers are required to implement all of the @@ -151,6 +161,10 @@ struct vdpa_device { * @get_generation: Get device config generation (optional) * @vdev: vdpa device * Returns u32: device generation + * @get_iova_range: Get supported iova range (optional) + * @vdev: vdpa device + * Returns the iova range supported by + * the device. * @set_map: Set device memory mapping (optional) * Needed for device that using device * specific DMA translation (on-chip IOMMU) @@ -216,6 +230,7 @@ struct vdpa_config_ops { void (*set_config)(struct vdpa_device *vdev, unsigned int offset, const void *buf, unsigned int len); u32 (*get_generation)(struct vdpa_device *vdev); + struct vdpa_iova_range (*get_iova_range)(struct vdpa_device *vdev); /* DMA ops */ int (*set_map)(struct vdpa_device *vdev, struct vhost_iotlb *iotlb); -- cgit v1.2.3 From 1b48dc03e575a872404f33b04cd237953c5d7498 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Fri, 23 Oct 2020 17:00:42 +0800 Subject: vhost: vdpa: report iova range This patch introduces a new ioctl for vhost-vdpa device that can report the iova range by the device. For device that implements get_iova_range() method, we fetch it from the vDPA device. If device doesn't implement get_iova_range() but depends on platform IOMMU, we will query via DOMAIN_ATTR_GEOMETRY, otherwise [0, ULLONG_MAX] is assumed. For safety, this patch also rules out the map request which is not in the valid range. Signed-off-by: Jason Wang Link: https://lore.kernel.org/r/20201023090043.14430-3-jasowang@redhat.com Signed-off-by: Michael S. Tsirkin --- drivers/vhost/vdpa.c | 41 ++++++++++++++++++++++++++++++++++++++++ include/uapi/linux/vhost.h | 4 ++++ include/uapi/linux/vhost_types.h | 9 +++++++++ 3 files changed, 54 insertions(+) diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c index a2dbc85e0b0d..846de69d9c01 100644 --- a/drivers/vhost/vdpa.c +++ b/drivers/vhost/vdpa.c @@ -47,6 +47,7 @@ struct vhost_vdpa { int minor; struct eventfd_ctx *config_ctx; int in_batch; + struct vdpa_iova_range range; }; static DEFINE_IDA(vhost_vdpa_ida); @@ -337,6 +338,16 @@ static long vhost_vdpa_set_config_call(struct vhost_vdpa *v, u32 __user *argp) return 0; } +static long vhost_vdpa_get_iova_range(struct vhost_vdpa *v, u32 __user *argp) +{ + struct vhost_vdpa_iova_range range = { + .first = v->range.first, + .last = v->range.last, + }; + + return copy_to_user(argp, &range, sizeof(range)); +} + static long vhost_vdpa_vring_ioctl(struct vhost_vdpa *v, unsigned int cmd, void __user *argp) { @@ -471,6 +482,9 @@ static long vhost_vdpa_unlocked_ioctl(struct file *filep, features = VHOST_VDPA_BACKEND_FEATURES; r = copy_to_user(featurep, &features, sizeof(features)); break; + case VHOST_VDPA_GET_IOVA_RANGE: + r = vhost_vdpa_get_iova_range(v, argp); + break; default: r = vhost_dev_ioctl(&v->vdev, cmd, argp); if (r == -ENOIOCTLCMD) @@ -597,6 +611,10 @@ static int vhost_vdpa_process_iotlb_update(struct vhost_vdpa *v, long pinned; int ret = 0; + if (msg->iova < v->range.first || + msg->iova + msg->size - 1 > v->range.last) + return -EINVAL; + if (vhost_iotlb_itree_first(iotlb, msg->iova, msg->iova + msg->size - 1)) return -EEXIST; @@ -783,6 +801,27 @@ static void vhost_vdpa_free_domain(struct vhost_vdpa *v) v->domain = NULL; } +static void vhost_vdpa_set_iova_range(struct vhost_vdpa *v) +{ + struct vdpa_iova_range *range = &v->range; + struct iommu_domain_geometry geo; + struct vdpa_device *vdpa = v->vdpa; + const struct vdpa_config_ops *ops = vdpa->config; + + if (ops->get_iova_range) { + *range = ops->get_iova_range(vdpa); + } else if (v->domain && + !iommu_domain_get_attr(v->domain, + DOMAIN_ATTR_GEOMETRY, &geo) && + geo.force_aperture) { + range->first = geo.aperture_start; + range->last = geo.aperture_end; + } else { + range->first = 0; + range->last = ULLONG_MAX; + } +} + static int vhost_vdpa_open(struct inode *inode, struct file *filep) { struct vhost_vdpa *v; @@ -823,6 +862,8 @@ static int vhost_vdpa_open(struct inode *inode, struct file *filep) if (r) goto err_init_iotlb; + vhost_vdpa_set_iova_range(v); + filep->private_data = v; return 0; diff --git a/include/uapi/linux/vhost.h b/include/uapi/linux/vhost.h index 75232185324a..c998860d7bbc 100644 --- a/include/uapi/linux/vhost.h +++ b/include/uapi/linux/vhost.h @@ -146,4 +146,8 @@ /* Set event fd for config interrupt*/ #define VHOST_VDPA_SET_CONFIG_CALL _IOW(VHOST_VIRTIO, 0x77, int) + +/* Get the valid iova range */ +#define VHOST_VDPA_GET_IOVA_RANGE _IOR(VHOST_VIRTIO, 0x78, \ + struct vhost_vdpa_iova_range) #endif diff --git a/include/uapi/linux/vhost_types.h b/include/uapi/linux/vhost_types.h index 9a269a88a6ff..f7f6a3a28977 100644 --- a/include/uapi/linux/vhost_types.h +++ b/include/uapi/linux/vhost_types.h @@ -138,6 +138,15 @@ struct vhost_vdpa_config { __u8 buf[0]; }; +/* vhost vdpa IOVA range + * @first: First address that can be mapped by vhost-vDPA + * @last: Last address that can be mapped by vhost-vDPA + */ +struct vhost_vdpa_iova_range { + __u64 first; + __u64 last; +}; + /* Feature bits */ /* Log all write descriptors. Can be changed while device is active. */ #define VHOST_F_LOG_ALL 26 -- cgit v1.2.3 From 70a62fce262854c4aabaa47e46e82fce9f3f3938 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Fri, 23 Oct 2020 17:00:43 +0800 Subject: vdpa_sim: implement get_iova_range() This implements a sample get_iova_range() for the simulator which advertise [0, ULLONG_MAX] as the valid range. Signed-off-by: Jason Wang Link: https://lore.kernel.org/r/20201023090043.14430-4-jasowang@redhat.com Signed-off-by: Michael S. Tsirkin --- drivers/vdpa/vdpa_sim/vdpa_sim.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c index 62d640327145..ff6c9fd8d879 100644 --- a/drivers/vdpa/vdpa_sim/vdpa_sim.c +++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c @@ -574,6 +574,16 @@ static u32 vdpasim_get_generation(struct vdpa_device *vdpa) return vdpasim->generation; } +static struct vdpa_iova_range vdpasim_get_iova_range(struct vdpa_device *vdpa) +{ + struct vdpa_iova_range range = { + .first = 0ULL, + .last = ULLONG_MAX, + }; + + return range; +} + static int vdpasim_set_map(struct vdpa_device *vdpa, struct vhost_iotlb *iotlb) { @@ -657,6 +667,7 @@ static const struct vdpa_config_ops vdpasim_net_config_ops = { .get_config = vdpasim_get_config, .set_config = vdpasim_set_config, .get_generation = vdpasim_get_generation, + .get_iova_range = vdpasim_get_iova_range, .dma_map = vdpasim_dma_map, .dma_unmap = vdpasim_dma_unmap, .free = vdpasim_free, @@ -683,6 +694,7 @@ static const struct vdpa_config_ops vdpasim_net_batch_config_ops = { .get_config = vdpasim_get_config, .set_config = vdpasim_set_config, .get_generation = vdpasim_get_generation, + .get_iova_range = vdpasim_get_iova_range, .set_map = vdpasim_set_map, .free = vdpasim_free, }; -- cgit v1.2.3 From 930eae9d3dd474ca1d5de63ea05d73075405cf6a Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Sat, 15 Aug 2020 20:19:57 +0200 Subject: ARM: dts: meson8: remove two invalid interrupt lines from the GPU node The 3.10 vendor kernel defines the following GPU 20 interrupt lines: #define INT_MALI_GP AM_IRQ(160) #define INT_MALI_GP_MMU AM_IRQ(161) #define INT_MALI_PP AM_IRQ(162) #define INT_MALI_PMU AM_IRQ(163) #define INT_MALI_PP0 AM_IRQ(164) #define INT_MALI_PP0_MMU AM_IRQ(165) #define INT_MALI_PP1 AM_IRQ(166) #define INT_MALI_PP1_MMU AM_IRQ(167) #define INT_MALI_PP2 AM_IRQ(168) #define INT_MALI_PP2_MMU AM_IRQ(169) #define INT_MALI_PP3 AM_IRQ(170) #define INT_MALI_PP3_MMU AM_IRQ(171) #define INT_MALI_PP4 AM_IRQ(172) #define INT_MALI_PP4_MMU AM_IRQ(173) #define INT_MALI_PP5 AM_IRQ(174) #define INT_MALI_PP5_MMU AM_IRQ(175) #define INT_MALI_PP6 AM_IRQ(176) #define INT_MALI_PP6_MMU AM_IRQ(177) #define INT_MALI_PP7 AM_IRQ(178) #define INT_MALI_PP7_MMU AM_IRQ(179) However, the driver from the 3.10 vendor kernel does not use the following four interrupt lines: - INT_MALI_PP3 - INT_MALI_PP3_MMU - INT_MALI_PP7 - INT_MALI_PP7_MMU Drop the "pp3" and "ppmmu3" interrupt lines. This is also important because there is no matching entry in interrupt-names for it (meaning the "pp2" interrupt is actually assigned to the "pp3" interrupt line). Fixes: 7d3f6b536e72c9 ("ARM: dts: meson8: add the Mali-450 MP6 GPU") Reported-by: Thomas Graichen Signed-off-by: Martin Blumenstingl Signed-off-by: Kevin Hilman Tested-by: thomas graichen Reviewed-by: Neil Armstrong Link: https://lore.kernel.org/r/20200815181957.408649-1-martin.blumenstingl@googlemail.com --- arch/arm/boot/dts/meson8.dtsi | 2 -- 1 file changed, 2 deletions(-) diff --git a/arch/arm/boot/dts/meson8.dtsi b/arch/arm/boot/dts/meson8.dtsi index eedb92526968..a4ab8b96d0eb 100644 --- a/arch/arm/boot/dts/meson8.dtsi +++ b/arch/arm/boot/dts/meson8.dtsi @@ -239,8 +239,6 @@ , , , - , - , , , , -- cgit v1.2.3 From f7d933388f11f66e5af9e5663a17f26523fddd07 Mon Sep 17 00:00:00 2001 From: Jerome Brunet Date: Fri, 23 Oct 2020 11:41:39 +0200 Subject: arm64: dts: meson: odroid-n2 plus: fix vddcpu_a pwm On the odroid N2 plus, cpufreq is not available due to an error on the cpu regulators. vddcpu a and b get the same PWM. The one provided to vddcpu A is incorrect. Because vddcpu B PWM is busy the regulator cannot register: > pwm-regulator regulator-vddcpu-b: Failed to get PWM: -16 Like on the odroid n2, use PWM A out of GPIOE_2 for vddcpu A to fix the problem Fixes: 98d24896ee11 ("arm64: dts: meson: add support for the ODROID-N2+") Signed-off-by: Jerome Brunet Acked-by: Neil Armstrong Signed-off-by: Kevin Hilman Link: https://lore.kernel.org/r/20201023094139.809379-1-jbrunet@baylibre.com --- arch/arm64/boot/dts/amlogic/meson-g12b-odroid-n2-plus.dts | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/boot/dts/amlogic/meson-g12b-odroid-n2-plus.dts b/arch/arm64/boot/dts/amlogic/meson-g12b-odroid-n2-plus.dts index 5de2815ba99d..ce1198ad34e4 100644 --- a/arch/arm64/boot/dts/amlogic/meson-g12b-odroid-n2-plus.dts +++ b/arch/arm64/boot/dts/amlogic/meson-g12b-odroid-n2-plus.dts @@ -19,7 +19,7 @@ regulator-min-microvolt = <680000>; regulator-max-microvolt = <1040000>; - pwms = <&pwm_AO_cd 1 1500 0>; + pwms = <&pwm_ab 0 1500 0>; }; &vddcpu_b { -- cgit v1.2.3 From e08102d507f34e6591de521a4c2587c6f02c7996 Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Thu, 22 Oct 2020 16:43:08 +0100 Subject: io_uring: remove opcode check on ltimeout kill __io_kill_linked_timeout() already checks for REQ_F_LTIMEOUT_ACTIVE and it's set only for linked timeouts. No need to verify next request's opcode. Signed-off-by: Pavel Begunkov Signed-off-by: Jens Axboe --- fs/io_uring.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/fs/io_uring.c b/fs/io_uring.c index d40717f8647b..db7ad9e61146 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -1872,8 +1872,7 @@ static bool __io_kill_linked_timeout(struct io_kiocb *req) if (list_empty(&req->link_list)) return false; link = list_first_entry(&req->link_list, struct io_kiocb, link_list); - if (link->opcode != IORING_OP_LINK_TIMEOUT) - return false; + /* * Can happen if a linked timeout fired and link had been like * req -> link t-out -> link t-out [-> ...] -- cgit v1.2.3 From ac877d2edd094e161801d72b49cfb56c5fc860fb Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Thu, 22 Oct 2020 16:43:09 +0100 Subject: io_uring: don't adjust LINK_HEAD in cancel ltimeout An armed linked timeout can never be a head of a link, so we don't need to clear REQ_F_LINK_HEAD for it. Signed-off-by: Pavel Begunkov Signed-off-by: Jens Axboe --- fs/io_uring.c | 1 - 1 file changed, 1 deletion(-) diff --git a/fs/io_uring.c b/fs/io_uring.c index db7ad9e61146..043652929aa9 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -1856,7 +1856,6 @@ static bool io_link_cancel_timeout(struct io_kiocb *req) if (ret != -1) { io_cqring_fill_event(req, -ECANCELED); io_commit_cqring(ctx); - req->flags &= ~REQ_F_LINK_HEAD; io_put_req_deferred(req, 1); return true; } -- cgit v1.2.3 From cdfcc3ee04599ce51e5c84432c177163637dd0e0 Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Thu, 22 Oct 2020 16:43:10 +0100 Subject: io_uring: always clear LINK_TIMEOUT after cancel Move REQ_F_LINK_TIMEOUT clearing out of __io_kill_linked_timeout() because it might return early and leave the flag set. It's not a problem, but may be confusing. Signed-off-by: Pavel Begunkov Signed-off-by: Jens Axboe --- fs/io_uring.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/io_uring.c b/fs/io_uring.c index 043652929aa9..552c27850c36 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -1881,7 +1881,6 @@ static bool __io_kill_linked_timeout(struct io_kiocb *req) list_del_init(&link->link_list); wake_ev = io_link_cancel_timeout(link); - req->flags &= ~REQ_F_LINK_TIMEOUT; return wake_ev; } @@ -1893,6 +1892,7 @@ static void io_kill_linked_timeout(struct io_kiocb *req) spin_lock_irqsave(&ctx->completion_lock, flags); wake_ev = __io_kill_linked_timeout(req); + req->flags &= ~REQ_F_LINK_TIMEOUT; spin_unlock_irqrestore(&ctx->completion_lock, flags); if (wake_ev) -- cgit v1.2.3 From c9abd7ad832b9eef06d887f4971894af5de617fd Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Thu, 22 Oct 2020 16:43:11 +0100 Subject: io_uring: don't defer put of cancelled ltimeout Inline io_link_cancel_timeout() and __io_kill_linked_timeout() into io_kill_linked_timeout(). That allows to easily move a put of a cancelled linked timeout out of completion_lock and to not deferring it. It is also much more readable when not scattered across three different functions. Signed-off-by: Pavel Begunkov Signed-off-by: Jens Axboe --- fs/io_uring.c | 58 ++++++++++++++++++++-------------------------------------- 1 file changed, 20 insertions(+), 38 deletions(-) diff --git a/fs/io_uring.c b/fs/io_uring.c index 552c27850c36..f6cb2b62ce1a 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -1846,57 +1846,39 @@ static void __io_free_req(struct io_kiocb *req) percpu_ref_put(&ctx->refs); } -static bool io_link_cancel_timeout(struct io_kiocb *req) +static void io_kill_linked_timeout(struct io_kiocb *req) { - struct io_timeout_data *io = req->async_data; struct io_ring_ctx *ctx = req->ctx; - int ret; - - ret = hrtimer_try_to_cancel(&io->timer); - if (ret != -1) { - io_cqring_fill_event(req, -ECANCELED); - io_commit_cqring(ctx); - io_put_req_deferred(req, 1); - return true; - } - - return false; -} - -static bool __io_kill_linked_timeout(struct io_kiocb *req) -{ struct io_kiocb *link; - bool wake_ev; - - if (list_empty(&req->link_list)) - return false; - link = list_first_entry(&req->link_list, struct io_kiocb, link_list); + bool cancelled = false; + unsigned long flags; + spin_lock_irqsave(&ctx->completion_lock, flags); + link = list_first_entry_or_null(&req->link_list, struct io_kiocb, + link_list); /* * Can happen if a linked timeout fired and link had been like * req -> link t-out -> link t-out [-> ...] */ - if (!(link->flags & REQ_F_LTIMEOUT_ACTIVE)) - return false; - - list_del_init(&link->link_list); - wake_ev = io_link_cancel_timeout(link); - return wake_ev; -} - -static void io_kill_linked_timeout(struct io_kiocb *req) -{ - struct io_ring_ctx *ctx = req->ctx; - unsigned long flags; - bool wake_ev; + if (link && (link->flags & REQ_F_LTIMEOUT_ACTIVE)) { + struct io_timeout_data *io = link->async_data; + int ret; - spin_lock_irqsave(&ctx->completion_lock, flags); - wake_ev = __io_kill_linked_timeout(req); + list_del_init(&link->link_list); + ret = hrtimer_try_to_cancel(&io->timer); + if (ret != -1) { + io_cqring_fill_event(link, -ECANCELED); + io_commit_cqring(ctx); + cancelled = true; + } + } req->flags &= ~REQ_F_LINK_TIMEOUT; spin_unlock_irqrestore(&ctx->completion_lock, flags); - if (wake_ev) + if (cancelled) { io_cqring_ev_posted(ctx); + io_put_req(link); + } } static struct io_kiocb *io_req_link_next(struct io_kiocb *req) -- cgit v1.2.3 From feaadc4fc2ebdbd53ffed1735077725855a2af53 Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Thu, 22 Oct 2020 16:47:16 +0100 Subject: io_uring: don't miss setting IO_WQ_WORK_CONCURRENT Set IO_WQ_WORK_CONCURRENT for all REQ_F_FORCE_ASYNC requests, do that in that is also looks better. Signed-off-by: Pavel Begunkov Signed-off-by: Jens Axboe --- fs/io_uring.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/fs/io_uring.c b/fs/io_uring.c index f6cb2b62ce1a..3606ea572e61 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -1365,6 +1365,9 @@ static void io_prep_async_work(struct io_kiocb *req) io_req_init_async(req); id = req->work.identity; + if (req->flags & REQ_F_FORCE_ASYNC) + req->work.flags |= IO_WQ_WORK_CONCURRENT; + if (req->flags & REQ_F_ISREG) { if (def->hash_reg_file || (ctx->flags & IORING_SETUP_IOPOLL)) io_wq_hash_work(&req->work, file_inode(req->file)); @@ -6245,13 +6248,6 @@ fail_req: if (unlikely(ret)) goto fail_req; } - - /* - * Never try inline submit of IOSQE_ASYNC is set, go straight - * to async execution. - */ - io_req_init_async(req); - req->work.flags |= IO_WQ_WORK_CONCURRENT; io_queue_async_work(req); } else { if (sqe) { -- cgit v1.2.3 From 9aaf354352f1142831457492790d6bfa9c883021 Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Thu, 22 Oct 2020 16:47:17 +0100 Subject: io_uring: simplify nxt propagation in io_queue_sqe Don't overuse goto's, complex control flow doesn't make compilers happy and makes code harder to read. Signed-off-by: Pavel Begunkov Signed-off-by: Jens Axboe --- fs/io_uring.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/fs/io_uring.c b/fs/io_uring.c index 3606ea572e61..4d647d91dab2 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -6188,7 +6188,6 @@ again: */ if (ret == -EAGAIN && !(req->flags & REQ_F_NOWAIT)) { if (!io_arm_poll_handler(req)) { -punt: /* * Queued up for async execution, worker will release * submit reference when the iocb is actually submitted. @@ -6217,12 +6216,9 @@ punt: if (nxt) { req = nxt; - - if (req->flags & REQ_F_FORCE_ASYNC) { - linked_timeout = NULL; - goto punt; - } - goto again; + if (!(req->flags & REQ_F_FORCE_ASYNC)) + goto again; + io_queue_async_work(req); } exit: if (old_creds) -- cgit v1.2.3 From 0d63c148d6d9ac57c124b618f66269bb4558553b Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Thu, 22 Oct 2020 16:47:18 +0100 Subject: io_uring: simplify __io_queue_sqe() Restructure __io_queue_sqe() so it follows simple if/else if/else control flow. It's more readable and removes extra goto/labels. Signed-off-by: Pavel Begunkov Signed-off-by: Jens Axboe --- fs/io_uring.c | 28 +++++++++++----------------- 1 file changed, 11 insertions(+), 17 deletions(-) diff --git a/fs/io_uring.c b/fs/io_uring.c index 4d647d91dab2..74dcc4471e9b 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -6162,7 +6162,6 @@ static struct io_kiocb *io_prep_linked_timeout(struct io_kiocb *req) static void __io_queue_sqe(struct io_kiocb *req, struct io_comp_state *cs) { struct io_kiocb *linked_timeout; - struct io_kiocb *nxt; const struct cred *old_creds = NULL; int ret; @@ -6197,30 +6196,25 @@ again: if (linked_timeout) io_queue_linked_timeout(linked_timeout); - goto exit; - } + } else if (likely(!ret)) { + /* drop submission reference */ + req = io_put_req_find_next(req); + if (linked_timeout) + io_queue_linked_timeout(linked_timeout); - if (unlikely(ret)) { + if (req) { + if (!(req->flags & REQ_F_FORCE_ASYNC)) + goto again; + io_queue_async_work(req); + } + } else { /* un-prep timeout, so it'll be killed as any other linked */ req->flags &= ~REQ_F_LINK_TIMEOUT; req_set_fail_links(req); io_put_req(req); io_req_complete(req, ret); - goto exit; } - /* drop submission reference */ - nxt = io_put_req_find_next(req); - if (linked_timeout) - io_queue_linked_timeout(linked_timeout); - - if (nxt) { - req = nxt; - if (!(req->flags & REQ_F_FORCE_ASYNC)) - goto again; - io_queue_async_work(req); - } -exit: if (old_creds) revert_creds(old_creds); } -- cgit v1.2.3 From c3a98c3ad5c0dc60a1ac66bf91147a3f39cac96b Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Fri, 23 Oct 2020 15:27:48 -0700 Subject: crypto: x86/poly1305 - add back a needed assignment One of the assignments that was removed by commit 4a0c1de64bf9 ("crypto: x86/poly1305 - Remove assignments with no effect") is actually needed, since it affects the return value. This fixes the following crypto self-test failure: alg: shash: poly1305-simd test failed (wrong result) on test vector 2, cfg="init+update+final aligned buffer" Fixes: 4a0c1de64bf9 ("crypto: x86/poly1305 - Remove assignments with no effect") Signed-off-by: Eric Biggers Signed-off-by: Herbert Xu --- arch/x86/crypto/poly1305_glue.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/crypto/poly1305_glue.c b/arch/x86/crypto/poly1305_glue.c index e508dbd91813..c44aba290fbb 100644 --- a/arch/x86/crypto/poly1305_glue.c +++ b/arch/x86/crypto/poly1305_glue.c @@ -158,6 +158,7 @@ static unsigned int crypto_poly1305_setdctxkey(struct poly1305_desc_ctx *dctx, dctx->s[1] = get_unaligned_le32(&inp[4]); dctx->s[2] = get_unaligned_le32(&inp[8]); dctx->s[3] = get_unaligned_le32(&inp[12]); + acc += POLY1305_BLOCK_SIZE; dctx->sset = true; } } -- cgit v1.2.3 From df833050cced27e1b343cc8bc41f90191b289334 Mon Sep 17 00:00:00 2001 From: Alex Elder Date: Wed, 21 Oct 2020 20:00:29 -0500 Subject: net: ipa: command payloads already mapped IPA transactions describe actions to be performed by the IPA hardware. Three cases use IPA transactions: transmitting a socket buffer; providing a page to receive packet data; and issuing an IPA immediate command. An IPA transaction contains a scatter/gather list (SGL) to hold the set of actions to be performed. We map buffers in the SGL for DMA at the time they are added to the transaction. For skb TX transactions, we fill the SGL with a call to skb_to_sgvec(). Page RX transactions involve a single page pointer, and that is recorded in the SGL with sg_set_page(). In both of these cases we then map the SGL for DMA with a call to dma_map_sg(). Immediate commands are different. The payload for an immediate command comes from a region of coherent DMA memory, which must *not* be mapped for DMA. For that reason, gsi_trans_cmd_add() sort of hand-crafts each SGL entry added to a command transaction. This patch fixes a problem with the code that crafts the SGL entry for an immediate command. Previously a portion of the SGL entry was updated using sg_set_buf(). However this is not valid because it includes a call to virt_to_page() on the buffer, but the command buffer pointer is not a linear address. Since we never actually map the SGL for command transactions, there are very few fields in the SGL we need to fill. Specifically, we only need to record the DMA address and the length, so they can be used by __gsi_trans_commit() to fill a TRE. We additionally need to preserve the SGL flags so for_each_sg() still works. For that we can simply assign a null page pointer for command SGL entries. Fixes: 9dd441e4ed575 ("soc: qcom: ipa: GSI transactions") Reported-by: Stephen Boyd Tested-by: Stephen Boyd Signed-off-by: Alex Elder Link: https://lore.kernel.org/r/20201022010029.11877-1-elder@linaro.org Signed-off-by: Jakub Kicinski --- drivers/net/ipa/gsi_trans.c | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) diff --git a/drivers/net/ipa/gsi_trans.c b/drivers/net/ipa/gsi_trans.c index 43f5f5d93cb0..92642030e735 100644 --- a/drivers/net/ipa/gsi_trans.c +++ b/drivers/net/ipa/gsi_trans.c @@ -397,15 +397,24 @@ void gsi_trans_cmd_add(struct gsi_trans *trans, void *buf, u32 size, /* assert(which < trans->tre_count); */ - /* Set the page information for the buffer. We also need to fill in - * the DMA address and length for the buffer (something dma_map_sg() - * normally does). + /* Commands are quite different from data transfer requests. + * Their payloads come from a pool whose memory is allocated + * using dma_alloc_coherent(). We therefore do *not* map them + * for DMA (unlike what we do for pages and skbs). + * + * When a transaction completes, the SGL is normally unmapped. + * A command transaction has direction DMA_NONE, which tells + * gsi_trans_complete() to skip the unmapping step. + * + * The only things we use directly in a command scatter/gather + * entry are the DMA address and length. We still need the SG + * table flags to be maintained though, so assign a NULL page + * pointer for that purpose. */ sg = &trans->sgl[which]; - - sg_set_buf(sg, buf, size); + sg_assign_page(sg, NULL); sg_dma_address(sg) = addr; - sg_dma_len(sg) = sg->length; + sg_dma_len(sg) = size; info = &trans->info[which]; info->opcode = opcode; -- cgit v1.2.3 From 4f3391ce8f5a69e7e6d66d0a3fc654eb6dbdc919 Mon Sep 17 00:00:00 2001 From: Vinay Kumar Yadav Date: Fri, 23 Oct 2020 00:35:57 +0530 Subject: chelsio/chtls: fix tls record info to user chtls_pt_recvmsg() receives a skb with tls header and subsequent skb with data, need to finalize the data copy whenever next skb with tls header is available. but here current tls header is overwritten by next available tls header, ends up corrupting user buffer data. fixing it by finalizing current record whenever next skb contains tls header. v1->v2: - Improved commit message. Fixes: 17a7d24aa89d ("crypto: chtls - generic handling of data and hdr") Signed-off-by: Vinay Kumar Yadav Link: https://lore.kernel.org/r/20201022190556.21308-1-vinay.yadav@chelsio.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c index 9fb5ca6682ea..188d871f6b8c 100644 --- a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c +++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c @@ -1585,6 +1585,7 @@ skip_copy: tp->urg_data = 0; if ((avail + offset) >= skb->len) { + struct sk_buff *next_skb; if (ULP_SKB_CB(skb)->flags & ULPCB_FLAG_TLS_HDR) { tp->copied_seq += skb->len; hws->rcvpld = skb->hdr_len; @@ -1595,8 +1596,10 @@ skip_copy: chtls_free_skb(sk, skb); buffers_freed++; hws->copied_seq = 0; - if (copied >= target && - !skb_peek(&sk->sk_receive_queue)) + next_skb = skb_peek(&sk->sk_receive_queue); + if (copied >= target && !next_skb) + break; + if (ULP_SKB_CB(next_skb)->flags & ULPCB_FLAG_TLS_HDR) break; } } while (len > 0); -- cgit v1.2.3 From d701ec326a31945d1533b438a6feab753829b738 Mon Sep 17 00:00:00 2001 From: Shannon Nelson Date: Thu, 22 Oct 2020 16:55:29 -0700 Subject: ionic: clean up sparse complaints The sparse complaints around the static_asserts were obscuring more useful complaints. So, don't check the static_asserts, and fix the remaining sparse complaints. Signed-off-by: Shannon Nelson Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/pensando/ionic/ionic_dev.c | 4 ++-- drivers/net/ethernet/pensando/ionic/ionic_dev.h | 2 ++ drivers/net/ethernet/pensando/ionic/ionic_fw.c | 6 ++--- drivers/net/ethernet/pensando/ionic/ionic_lif.c | 28 +++++++++++------------ drivers/net/ethernet/pensando/ionic/ionic_main.c | 4 ++-- drivers/net/ethernet/pensando/ionic/ionic_stats.h | 2 +- drivers/net/ethernet/pensando/ionic/ionic_txrx.c | 10 ++++---- 7 files changed, 29 insertions(+), 27 deletions(-) diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.c b/drivers/net/ethernet/pensando/ionic/ionic_dev.c index 545c99b15df8..dc5fbc2704f3 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_dev.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.c @@ -333,7 +333,7 @@ int ionic_set_vf_config(struct ionic *ionic, int vf, u8 attr, u8 *data) union ionic_dev_cmd cmd = { .vf_setattr.opcode = IONIC_CMD_VF_SETATTR, .vf_setattr.attr = attr, - .vf_setattr.vf_index = vf, + .vf_setattr.vf_index = cpu_to_le16(vf), }; int err; @@ -391,7 +391,7 @@ void ionic_dev_cmd_queue_identify(struct ionic_dev *idev, { union ionic_dev_cmd cmd = { .q_identify.opcode = IONIC_CMD_Q_IDENTIFY, - .q_identify.lif_type = lif_type, + .q_identify.lif_type = cpu_to_le16(lif_type), .q_identify.type = qtype, .q_identify.ver = qver, }; diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.h b/drivers/net/ethernet/pensando/ionic/ionic_dev.h index c109cd5a0471..6c243b17312c 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_dev.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.h @@ -29,6 +29,7 @@ struct ionic_dev_bar { int res_index; }; +#ifndef __CHECKER__ /* Registers */ static_assert(sizeof(struct ionic_intr) == 32); @@ -119,6 +120,7 @@ static_assert(sizeof(struct ionic_vf_setattr_cmd) == 64); static_assert(sizeof(struct ionic_vf_setattr_comp) == 16); static_assert(sizeof(struct ionic_vf_getattr_cmd) == 64); static_assert(sizeof(struct ionic_vf_getattr_comp) == 16); +#endif /* __CHECKER__ */ struct ionic_devinfo { u8 asic_type; diff --git a/drivers/net/ethernet/pensando/ionic/ionic_fw.c b/drivers/net/ethernet/pensando/ionic/ionic_fw.c index f492ae406a60..d7bbf336c6f6 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_fw.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_fw.c @@ -27,9 +27,9 @@ static void ionic_dev_cmd_firmware_download(struct ionic_dev *idev, u64 addr, { union ionic_dev_cmd cmd = { .fw_download.opcode = IONIC_CMD_FW_DOWNLOAD, - .fw_download.offset = offset, - .fw_download.addr = addr, - .fw_download.length = length + .fw_download.offset = cpu_to_le32(offset), + .fw_download.addr = cpu_to_le64(addr), + .fw_download.length = cpu_to_le32(length), }; ionic_dev_cmd_go(idev, &cmd); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c index d655a7ae3058..591c644b8e69 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c @@ -1915,11 +1915,11 @@ static int ionic_get_vf_config(struct net_device *netdev, ret = -EINVAL; } else { ivf->vf = vf; - ivf->vlan = ionic->vfs[vf].vlanid; + ivf->vlan = le16_to_cpu(ionic->vfs[vf].vlanid); ivf->qos = 0; ivf->spoofchk = ionic->vfs[vf].spoofchk; ivf->linkstate = ionic->vfs[vf].linkstate; - ivf->max_tx_rate = ionic->vfs[vf].maxrate; + ivf->max_tx_rate = le32_to_cpu(ionic->vfs[vf].maxrate); ivf->trusted = ionic->vfs[vf].trusted; ether_addr_copy(ivf->mac, ionic->vfs[vf].macaddr); } @@ -2019,7 +2019,7 @@ static int ionic_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, ret = ionic_set_vf_config(ionic, vf, IONIC_VF_ATTR_VLAN, (u8 *)&vlan); if (!ret) - ionic->vfs[vf].vlanid = vlan; + ionic->vfs[vf].vlanid = cpu_to_le16(vlan); } up_write(&ionic->vf_op_lock); @@ -2048,7 +2048,7 @@ static int ionic_set_vf_rate(struct net_device *netdev, int vf, ret = ionic_set_vf_config(ionic, vf, IONIC_VF_ATTR_RATE, (u8 *)&tx_max); if (!ret) - lif->ionic->vfs[vf].maxrate = tx_max; + lif->ionic->vfs[vf].maxrate = cpu_to_le32(tx_max); } up_write(&ionic->vf_op_lock); @@ -2981,14 +2981,14 @@ void ionic_lif_unregister(struct ionic_lif *lif) static void ionic_lif_queue_identify(struct ionic_lif *lif) { + union ionic_q_identity __iomem *q_ident; struct ionic *ionic = lif->ionic; - union ionic_q_identity *q_ident; struct ionic_dev *idev; int qtype; int err; idev = &lif->ionic->idev; - q_ident = (union ionic_q_identity *)&idev->dev_cmd_regs->data; + q_ident = (union ionic_q_identity __iomem *)&idev->dev_cmd_regs->data; for (qtype = 0; qtype < ARRAY_SIZE(ionic_qtype_versions); qtype++) { struct ionic_qtype_info *qti = &lif->qtype_info[qtype]; @@ -3011,14 +3011,14 @@ static void ionic_lif_queue_identify(struct ionic_lif *lif) ionic_qtype_versions[qtype]); err = ionic_dev_cmd_wait(ionic, DEVCMD_TIMEOUT); if (!err) { - qti->version = q_ident->version; - qti->supported = q_ident->supported; - qti->features = le64_to_cpu(q_ident->features); - qti->desc_sz = le16_to_cpu(q_ident->desc_sz); - qti->comp_sz = le16_to_cpu(q_ident->comp_sz); - qti->sg_desc_sz = le16_to_cpu(q_ident->sg_desc_sz); - qti->max_sg_elems = le16_to_cpu(q_ident->max_sg_elems); - qti->sg_desc_stride = le16_to_cpu(q_ident->sg_desc_stride); + qti->version = readb(&q_ident->version); + qti->supported = readb(&q_ident->supported); + qti->features = readq(&q_ident->features); + qti->desc_sz = readw(&q_ident->desc_sz); + qti->comp_sz = readw(&q_ident->comp_sz); + qti->sg_desc_sz = readw(&q_ident->sg_desc_sz); + qti->max_sg_elems = readw(&q_ident->max_sg_elems); + qti->sg_desc_stride = readw(&q_ident->sg_desc_stride); } mutex_unlock(&ionic->dev_cmd_lock); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_main.c b/drivers/net/ethernet/pensando/ionic/ionic_main.c index ee0740881af3..d355676f6c16 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_main.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_main.c @@ -311,7 +311,7 @@ int ionic_adminq_post_wait(struct ionic_lif *lif, struct ionic_admin_ctx *ctx) static void ionic_dev_cmd_clean(struct ionic *ionic) { - union ionic_dev_cmd_regs *regs = ionic->idev.dev_cmd_regs; + union __iomem ionic_dev_cmd_regs *regs = ionic->idev.dev_cmd_regs; iowrite32(0, ®s->doorbell); memset_io(®s->cmd, 0, sizeof(regs->cmd)); @@ -333,7 +333,7 @@ int ionic_dev_cmd_wait(struct ionic *ionic, unsigned long max_seconds) */ max_wait = jiffies + (max_seconds * HZ); try_again: - opcode = idev->dev_cmd_regs->cmd.cmd.opcode; + opcode = readb(&idev->dev_cmd_regs->cmd.cmd.opcode); start_time = jiffies; do { done = ionic_dev_cmd_done(idev); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_stats.h b/drivers/net/ethernet/pensando/ionic/ionic_stats.h index 3f543512616e..2a725834f792 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_stats.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_stats.h @@ -49,7 +49,7 @@ extern const int ionic_num_stats_grps; (*((u64 *)(((u8 *)(base_ptr)) + (desc_ptr)->offset))) #define IONIC_READ_STAT_LE64(base_ptr, desc_ptr) \ - __le64_to_cpu(*((u64 *)(((u8 *)(base_ptr)) + (desc_ptr)->offset))) + __le64_to_cpu(*((__le64 *)(((u8 *)(base_ptr)) + (desc_ptr)->offset))) struct ionic_stat_desc { char name[ETH_GSTRING_LEN]; diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c index 169ac4f54640..8f6fc7142bc5 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c @@ -200,7 +200,7 @@ static void ionic_rx_clean(struct ionic_queue *q, if (likely(netdev->features & NETIF_F_RXCSUM)) { if (comp->csum_flags & IONIC_RXQ_COMP_CSUM_F_CALC) { skb->ip_summed = CHECKSUM_COMPLETE; - skb->csum = (__wsum)le16_to_cpu(comp->csum); + skb->csum = (__force __wsum)le16_to_cpu(comp->csum); stats->csum_complete++; } } else { @@ -812,6 +812,7 @@ static int ionic_tx_tso(struct ionic_queue *q, struct sk_buff *skb) skb_frag_t *frag; bool start, done; bool outer_csum; + dma_addr_t addr; bool has_vlan; u16 desc_len; u8 desc_nsge; @@ -893,11 +894,10 @@ static int ionic_tx_tso(struct ionic_queue *q, struct sk_buff *skb) if (frag_left > 0) { len = min(frag_left, left); frag_left -= len; - elem->addr = - cpu_to_le64(ionic_tx_map_frag(q, frag, - offset, len)); - if (dma_mapping_error(dev, elem->addr)) + addr = ionic_tx_map_frag(q, frag, offset, len); + if (dma_mapping_error(dev, addr)) goto err_out_abort; + elem->addr = cpu_to_le64(addr); elem->len = cpu_to_le16(len); elem++; desc_nsge++; -- cgit v1.2.3 From 43ecf7b46f2688fd37909801aee264f288b3917b Mon Sep 17 00:00:00 2001 From: Shannon Nelson Date: Thu, 22 Oct 2020 16:55:30 -0700 Subject: ionic: no rx flush in deinit Kmemleak pointed out to us that ionic_rx_flush() is sending skbs into napi_gro_XXX with a disabled napi context, and these end up getting lost and leaked. We can safely remove the flush. Fixes: 0f3154e6bcb3 ("ionic: Add Tx and Rx handling") Signed-off-by: Shannon Nelson Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/pensando/ionic/ionic_lif.c | 1 - drivers/net/ethernet/pensando/ionic/ionic_txrx.c | 13 ------------- drivers/net/ethernet/pensando/ionic/ionic_txrx.h | 1 - 3 files changed, 15 deletions(-) diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c index 591c644b8e69..a12df3946a07 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c @@ -1656,7 +1656,6 @@ static void ionic_txrx_deinit(struct ionic_lif *lif) if (lif->rxqcqs) { for (i = 0; i < lif->nxqs && lif->rxqcqs[i]; i++) { ionic_lif_qcq_deinit(lif, lif->rxqcqs[i]); - ionic_rx_flush(&lif->rxqcqs[i]->cq); ionic_rx_empty(&lif->rxqcqs[i]->q); } } diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c index 8f6fc7142bc5..35acb4d66e31 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c @@ -253,19 +253,6 @@ static bool ionic_rx_service(struct ionic_cq *cq, struct ionic_cq_info *cq_info) return true; } -void ionic_rx_flush(struct ionic_cq *cq) -{ - struct ionic_dev *idev = &cq->lif->ionic->idev; - u32 work_done; - - work_done = ionic_cq_service(cq, cq->num_descs, - ionic_rx_service, NULL, NULL); - - if (work_done) - ionic_intr_credits(idev->intr_ctrl, cq->bound_intr->index, - work_done, IONIC_INTR_CRED_RESET_COALESCE); -} - static int ionic_rx_page_alloc(struct ionic_queue *q, struct ionic_page_info *page_info) { diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.h b/drivers/net/ethernet/pensando/ionic/ionic_txrx.h index a5883be0413f..7667b72232b8 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.h @@ -4,7 +4,6 @@ #ifndef _IONIC_TXRX_H_ #define _IONIC_TXRX_H_ -void ionic_rx_flush(struct ionic_cq *cq); void ionic_tx_flush(struct ionic_cq *cq); void ionic_rx_fill(struct ionic_queue *q); -- cgit v1.2.3 From 0c32a28e247f51b0b67b5abb6e9368542e30c136 Mon Sep 17 00:00:00 2001 From: Shannon Nelson Date: Thu, 22 Oct 2020 16:55:31 -0700 Subject: ionic: fix mem leak in rx_empty The sentinel descriptor entry was getting missed in the traverse of the ring from head to tail, so change to a loop of 0 to the end. Fixes: f1d2e894f1b7 ("ionic: use index not pointer for queue tracking") Signed-off-by: Shannon Nelson Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/pensando/ionic/ionic_txrx.c | 24 +++++++++++------------- 1 file changed, 11 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c index 35acb4d66e31..b3d2250c77d0 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c @@ -400,22 +400,20 @@ static void ionic_rx_fill_cb(void *arg) void ionic_rx_empty(struct ionic_queue *q) { struct ionic_desc_info *desc_info; - struct ionic_rxq_desc *desc; - unsigned int i; - u16 idx; - - idx = q->tail_idx; - while (idx != q->head_idx) { - desc_info = &q->info[idx]; - desc = desc_info->desc; - desc->addr = 0; - desc->len = 0; + struct ionic_page_info *page_info; + unsigned int i, j; - for (i = 0; i < desc_info->npages; i++) - ionic_rx_page_free(q, &desc_info->pages[i]); + for (i = 0; i < q->num_descs; i++) { + desc_info = &q->info[i]; + for (j = 0; j < IONIC_RX_MAX_SG_ELEMS + 1; j++) { + page_info = &desc_info->pages[j]; + if (page_info->page) + ionic_rx_page_free(q, page_info); + } + desc_info->npages = 0; + desc_info->cb = NULL; desc_info->cb_arg = NULL; - idx = (idx + 1) & (q->num_descs - 1); } } -- cgit v1.2.3 From dab234227cbdc6a088c6f9bce38b5dcfefe90832 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Fri, 23 Oct 2020 14:32:36 +1100 Subject: net: ucc_geth: Drop extraneous parentheses in comparison Clang warns about the extra parentheses in this comparison: drivers/net/ethernet/freescale/ucc_geth.c:1361:28: warning: equality comparison with extraneous parentheses if ((ugeth->phy_interface == PHY_INTERFACE_MODE_SGMII)) ~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~ It seems clear the intent here is to do a comparison not an assignment, so drop the extra parentheses to avoid any confusion. Signed-off-by: Michael Ellerman Link: https://lore.kernel.org/r/20201023033236.3296988-1-mpe@ellerman.id.au Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/ucc_geth.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/freescale/ucc_geth.c b/drivers/net/ethernet/freescale/ucc_geth.c index 714b501be7d0..ba8869c3d891 100644 --- a/drivers/net/ethernet/freescale/ucc_geth.c +++ b/drivers/net/ethernet/freescale/ucc_geth.c @@ -1358,7 +1358,7 @@ static int adjust_enet_interface(struct ucc_geth_private *ugeth) (ugeth->phy_interface == PHY_INTERFACE_MODE_RTBI)) { upsmr |= UCC_GETH_UPSMR_TBIM; } - if ((ugeth->phy_interface == PHY_INTERFACE_MODE_SGMII)) + if (ugeth->phy_interface == PHY_INTERFACE_MODE_SGMII) upsmr |= UCC_GETH_UPSMR_SGMM; out_be32(&uf_regs->upsmr, upsmr); -- cgit v1.2.3 From 435ccfa894e35e3d4a1799e6ac030e48a7b69ef5 Mon Sep 17 00:00:00 2001 From: Arjun Roy Date: Fri, 23 Oct 2020 11:47:09 -0700 Subject: tcp: Prevent low rmem stalls with SO_RCVLOWAT. With SO_RCVLOWAT, under memory pressure, it is possible to enter a state where: 1. We have not received enough bytes to satisfy SO_RCVLOWAT. 2. We have not entered buffer pressure (see tcp_rmem_pressure()). 3. But, we do not have enough buffer space to accept more packets. In this case, we advertise 0 rwnd (due to #3) but the application does not drain the receive queue (no wakeup because of #1 and #2) so the flow stalls. Modify the heuristic for SO_RCVLOWAT so that, if we are advertising rwnd<=rcv_mss, force a wakeup to prevent a stall. Without this patch, setting tcp_rmem to 6143 and disabling TCP autotune causes a stalled flow. With this patch, no stall occurs. This is with RPC-style traffic with large messages. Fixes: 03f45c883c6f ("tcp: avoid extra wakeups for SO_RCVLOWAT users") Signed-off-by: Arjun Roy Acked-by: Soheil Hassas Yeganeh Acked-by: Neal Cardwell Signed-off-by: Eric Dumazet Link: https://lore.kernel.org/r/20201023184709.217614-1-arjunroy.kdev@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv4/tcp.c | 2 ++ net/ipv4/tcp_input.c | 3 ++- 2 files changed, 4 insertions(+), 1 deletion(-) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index bae4284bf542..b2bc3d7fe9e8 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -485,6 +485,8 @@ static inline bool tcp_stream_is_readable(const struct tcp_sock *tp, return true; if (tcp_rmem_pressure(sk)) return true; + if (tcp_receive_window(tp) <= inet_csk(sk)->icsk_ack.rcv_mss) + return true; } if (sk->sk_prot->stream_memory_read) return sk->sk_prot->stream_memory_read(sk); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index fc445833b5e5..389d1b340248 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -4908,7 +4908,8 @@ void tcp_data_ready(struct sock *sk) int avail = tp->rcv_nxt - tp->copied_seq; if (avail < sk->sk_rcvlowat && !tcp_rmem_pressure(sk) && - !sock_flag(sk, SOCK_DONE)) + !sock_flag(sk, SOCK_DONE) && + tcp_receive_window(tp) > inet_csk(sk)->icsk_ack.rcv_mss) return; sk->sk_data_ready(sk); -- cgit v1.2.3 From c8b5e2600a2cfa1cdfbecf151afd67aee227381d Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Sun, 25 Oct 2020 13:53:26 -0600 Subject: io_uring: use type appropriate io_kiocb handler for double poll io_poll_double_wake() is called for both request types - both pure poll requests, and internal polls. This means that we should be using the right handler based on the request type. Use the one that the original caller already assigned for the waitqueue handling, that will always match the correct type. Cc: stable@vger.kernel.org # v5.8+ Reported-by: Pavel Begunkov Signed-off-by: Jens Axboe --- fs/io_uring.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/fs/io_uring.c b/fs/io_uring.c index 74dcc4471e9b..2f6af230e86e 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -4959,8 +4959,10 @@ static int io_poll_double_wake(struct wait_queue_entry *wait, unsigned mode, /* make sure double remove sees this as being gone */ wait->private = NULL; spin_unlock(&poll->head->lock); - if (!done) - __io_async_wake(req, poll, mask, io_poll_task_func); + if (!done) { + /* use wait func handler, so it matches the rq type */ + poll->wait.func(&poll->wait, mode, sync, key); + } } refcount_dec(&req->refs); return 1; -- cgit v1.2.3 From 194810f78402128fe07676646cf9027fd3ed431c Mon Sep 17 00:00:00 2001 From: Dan Murphy Date: Fri, 16 Oct 2020 06:57:03 -0500 Subject: dt-bindings: leds: Update devicetree documents for ID_RGB Update the leds/common.yaml to indicate that the max color ID is 9. Reflect the same change in the leds-class-multicolor.yaml Reported-by: Zhen Lei Signed-off-by: Dan Murphy Link: https://lore.kernel.org/r/20201016115703.30184-1-dmurphy@ti.com Signed-off-by: Rob Herring --- Documentation/devicetree/bindings/leds/common.yaml | 2 +- .../devicetree/bindings/leds/leds-class-multicolor.yaml | 9 +++++---- 2 files changed, 6 insertions(+), 5 deletions(-) diff --git a/Documentation/devicetree/bindings/leds/common.yaml b/Documentation/devicetree/bindings/leds/common.yaml index 08b6700ca61e..f1211e7045f1 100644 --- a/Documentation/devicetree/bindings/leds/common.yaml +++ b/Documentation/devicetree/bindings/leds/common.yaml @@ -43,7 +43,7 @@ properties: LED_COLOR_ID available, add a new one. $ref: /schemas/types.yaml#definitions/uint32 minimum: 0 - maximum: 8 + maximum: 9 function-enumerator: description: diff --git a/Documentation/devicetree/bindings/leds/leds-class-multicolor.yaml b/Documentation/devicetree/bindings/leds/leds-class-multicolor.yaml index b1a53f054b89..37445c68cdef 100644 --- a/Documentation/devicetree/bindings/leds/leds-class-multicolor.yaml +++ b/Documentation/devicetree/bindings/leds/leds-class-multicolor.yaml @@ -16,7 +16,7 @@ description: | modules. This is achieved by adding multi-led nodes layer to the monochrome LED bindings. The nodes and properties defined in this document are unique to the multicolor - LED class. Common LED nodes and properties are inherited from the common.txt + LED class. Common LED nodes and properties are inherited from the common.yaml within this documentation directory. patternProperties: @@ -25,10 +25,11 @@ patternProperties: description: Represents the LEDs that are to be grouped. properties: color: - const: 8 # LED_COLOR_ID_MULTI description: | - For multicolor LED support this property should be defined as - LED_COLOR_ID_MULTI which can be found in include/linux/leds/common.h. + For multicolor LED support this property should be defined as either + LED_COLOR_ID_RGB or LED_COLOR_ID_MULTI which can be found in + include/linux/leds/common.h. + enum: [ 8, 9 ] $ref: "common.yaml#" -- cgit v1.2.3 From f83b03fc727ab56a77e68713d6e40299698f3c9f Mon Sep 17 00:00:00 2001 From: Fabien Parent Date: Sun, 18 Oct 2020 21:30:16 +0200 Subject: dt-bindings: mailbox: mtk-gce: fix incorrect mbox-cells value As the binding documentation says, #mbox-cells must have a value of 2, but the example use a value 3. The MT8173 device tree correctly use mbox-cells = <2>. This commit fixes the example. Fixes: 19d8e335d58a ("dt-binding: gce: remove atomic_exec in mboxes property") Reviewed-by: Matthias Brugger Signed-off-by: Fabien Parent Link: https://lore.kernel.org/r/20201018193016.3339045-1-fparent@baylibre.com Signed-off-by: Rob Herring --- Documentation/devicetree/bindings/mailbox/mtk-gce.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/mailbox/mtk-gce.txt b/Documentation/devicetree/bindings/mailbox/mtk-gce.txt index cf48cd806e00..7771ecaac586 100644 --- a/Documentation/devicetree/bindings/mailbox/mtk-gce.txt +++ b/Documentation/devicetree/bindings/mailbox/mtk-gce.txt @@ -47,7 +47,7 @@ Example: interrupts = ; clocks = <&infracfg CLK_INFRA_GCE>; clock-names = "gce"; - #mbox-cells = <3>; + #mbox-cells = <2>; }; Example for a client device: -- cgit v1.2.3 From ca05f33316559a04867295dd49f85aeedbfd6bfd Mon Sep 17 00:00:00 2001 From: Vincent Whitchurch Date: Wed, 21 Oct 2020 11:53:59 +0200 Subject: of: Fix reserved-memory overlap detection The reserved-memory overlap detection code fails to detect overlaps if either of the regions starts at address 0x0. The code explicitly checks for and ignores such regions, apparently in order to ignore dynamically allocated regions which have an address of 0x0 at this point. These dynamically allocated regions also have a size of 0x0 at this point, so fix this by removing the check and sorting the dynamically allocated regions ahead of any static regions at address 0x0. For example, there are two overlaps in this case but they are not currently reported: foo@0 { reg = <0x0 0x2000>; }; bar@0 { reg = <0x0 0x1000>; }; baz@1000 { reg = <0x1000 0x1000>; }; quux { size = <0x1000>; }; but they are after this patch: OF: reserved mem: OVERLAP DETECTED! bar@0 (0x00000000--0x00001000) overlaps with foo@0 (0x00000000--0x00002000) OF: reserved mem: OVERLAP DETECTED! foo@0 (0x00000000--0x00002000) overlaps with baz@1000 (0x00001000--0x00002000) Signed-off-by: Vincent Whitchurch Link: https://lore.kernel.org/r/ded6fd6b47b58741aabdcc6967f73eca6a3f311e.1603273666.git-series.vincent.whitchurch@axis.com Signed-off-by: Rob Herring --- drivers/of/of_reserved_mem.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/of/of_reserved_mem.c b/drivers/of/of_reserved_mem.c index bcd154485972..a7fbc5e37e19 100644 --- a/drivers/of/of_reserved_mem.c +++ b/drivers/of/of_reserved_mem.c @@ -200,6 +200,16 @@ static int __init __rmem_cmp(const void *a, const void *b) if (ra->base > rb->base) return 1; + /* + * Put the dynamic allocations (address == 0, size == 0) before static + * allocations at address 0x0 so that overlap detection works + * correctly. + */ + if (ra->size < rb->size) + return -1; + if (ra->size > rb->size) + return 1; + return 0; } @@ -217,8 +227,7 @@ static void __init __rmem_check_for_overlap(void) this = &reserved_mem[i]; next = &reserved_mem[i + 1]; - if (!(this->base && next->base)) - continue; + if (this->base + this->size > next->base) { phys_addr_t this_end, next_end; -- cgit v1.2.3 From c8fa942bd3fbd5b05351aff04649064f5d59f0d2 Mon Sep 17 00:00:00 2001 From: Anson Huang Date: Mon, 26 Oct 2020 15:45:00 +0800 Subject: arm64: defconfig: Build in CONFIG_GPIO_MXC by default i.MX SoC GPIO driver provides the basic functions of GPIO pin operations and IRQ operations, it is now changed from "def_bool y" to "tristate", so it should be explicitly enabled to make sure all consumers work normally. Signed-off-by: Anson Huang Signed-off-by: Shawn Guo --- arch/arm64/configs/defconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig index 17a2df6a263e..5cfe3cf6f2ac 100644 --- a/arch/arm64/configs/defconfig +++ b/arch/arm64/configs/defconfig @@ -500,6 +500,7 @@ CONFIG_GPIO_ALTERA=m CONFIG_GPIO_DWAPB=y CONFIG_GPIO_MB86S7X=y CONFIG_GPIO_MPC8XXX=y +CONFIG_GPIO_MXC=y CONFIG_GPIO_PL061=y CONFIG_GPIO_RCAR=y CONFIG_GPIO_UNIPHIER=y -- cgit v1.2.3 From f06dd1d41949018e050d18ec1058f16b388b96cf Mon Sep 17 00:00:00 2001 From: Anson Huang Date: Thu, 17 Sep 2020 13:33:48 +0800 Subject: ARM: imx_v6_v7_defconfig: Build in CONFIG_GPIO_MXC by default i.MX SoC GPIO driver provides the basic functions of GPIO pin operations and IRQ operations, it is now changed from "def_bool y" to "tristate", so it should be explicitly enabled to make sure all consumers work normally. Signed-off-by: Anson Huang Signed-off-by: Shawn Guo --- arch/arm/configs/imx_v6_v7_defconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/configs/imx_v6_v7_defconfig b/arch/arm/configs/imx_v6_v7_defconfig index 0fa79bd00219..221f5c340c86 100644 --- a/arch/arm/configs/imx_v6_v7_defconfig +++ b/arch/arm/configs/imx_v6_v7_defconfig @@ -217,6 +217,7 @@ CONFIG_GPIO_PCA953X=y CONFIG_GPIO_PCF857X=y CONFIG_GPIO_STMPE=y CONFIG_GPIO_74X164=y +CONFIG_GPIO_MXC=y CONFIG_POWER_RESET=y CONFIG_POWER_RESET_SYSCON=y CONFIG_POWER_RESET_SYSCON_POWEROFF=y -- cgit v1.2.3 From d6820bc6508c0cefd6e407cf5be50d9efda6c85c Mon Sep 17 00:00:00 2001 From: Anson Huang Date: Thu, 17 Sep 2020 13:33:49 +0800 Subject: ARM: multi_v7_defconfig: Build in CONFIG_GPIO_MXC by default i.MX SoC GPIO driver provides the basic functions of GPIO pin operations and IRQ operations, it is now changed from "def_bool y" to "tristate", so it should be explicitly enabled to make sure all consumers work normally. Signed-off-by: Anson Huang Signed-off-by: Shawn Guo --- arch/arm/configs/multi_v7_defconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/configs/multi_v7_defconfig b/arch/arm/configs/multi_v7_defconfig index e731cdf7c88c..a611b0c1e540 100644 --- a/arch/arm/configs/multi_v7_defconfig +++ b/arch/arm/configs/multi_v7_defconfig @@ -465,6 +465,7 @@ CONFIG_GPIO_PALMAS=y CONFIG_GPIO_TPS6586X=y CONFIG_GPIO_TPS65910=y CONFIG_GPIO_TWL4030=y +CONFIG_GPIO_MXC=y CONFIG_POWER_AVS=y CONFIG_ROCKCHIP_IODOMAIN=y CONFIG_POWER_RESET_AS3722=y -- cgit v1.2.3 From 6d7fe8aa45037776d5871c20f47a90901c841e4e Mon Sep 17 00:00:00 2001 From: Masahiro Yamada Date: Mon, 31 Aug 2020 12:58:03 +0900 Subject: MAINTAINERS: step down as maintainer of UniPhier SoCs and Denali driver I am leaving Socionext. Orphan the UniPhier platform and Denali NAND driver until somebody takes the role. Signed-off-by: Masahiro Yamada Signed-off-by: Arnd Bergmann --- MAINTAINERS | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index e73636b75f29..a238f954a5c9 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2642,10 +2642,8 @@ F: drivers/pinctrl/visconti/ N: visconti ARM/UNIPHIER ARCHITECTURE -M: Masahiro Yamada L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) -S: Maintained -T: git git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-uniphier.git +S: Orphan F: Documentation/devicetree/bindings/arm/socionext/uniphier.yaml F: Documentation/devicetree/bindings/gpio/socionext,uniphier-gpio.yaml F: Documentation/devicetree/bindings/pinctrl/socionext,uniphier-pinctrl.yaml @@ -5006,9 +5004,8 @@ T: git git://linuxtv.org/media_tree.git F: drivers/media/platform/sti/delta DENALI NAND DRIVER -M: Masahiro Yamada L: linux-mtd@lists.infradead.org -S: Supported +S: Orphan F: drivers/mtd/nand/raw/denali* DESIGNWARE EDMA CORE IP DRIVER -- cgit v1.2.3 From 421f2597bf4201ee70c23cb381e2ba683b3033e9 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 16 Oct 2020 08:18:47 +0200 Subject: MAINTAINERS: Move Kukjin Kim to credits Kukjin Kim has been maintaining the Samsung ARM architectures since 2010 up to 2016. He contributed many patches for the S3C, S5P and Exynos support. However since 2016 there is little activity from him on the LKML [1] so move his name to the CREDITS. Dear Kukjin, thank you for all the effort you put in to the upstream Samsung support. [1] https://lore.kernel.org/lkml/?q=f%3A%22Kukjin+Kim%22 Link: https://lore.kernel.org/r/20201016061848.6258-1-krzk@kernel.org Cc: Kukjin Kim Cc: Arnd Bergmann Cc: Olof Johansson Signed-off-by: Krzysztof Kozlowski Acked-by: Kukjin Kim Signed-off-by: Arnd Bergmann --- CREDITS | 4 ++++ MAINTAINERS | 2 -- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/CREDITS b/CREDITS index cb02b9923a52..5df027e12ff7 100644 --- a/CREDITS +++ b/CREDITS @@ -1910,6 +1910,10 @@ S: 660 Harvard Ave. #7 S: Santa Clara, CA 95051 S: USA +N: Kukjin Kim +E: kgene@kernel.org +D: Samsung S3C, S5P and Exynos ARM architectures + N: Russell King E: rmk@arm.linux.org.uk D: Linux/arm integrator, maintainer & hacker diff --git a/MAINTAINERS b/MAINTAINERS index a238f954a5c9..5664ee1f543f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2375,7 +2375,6 @@ F: sound/soc/rockchip/ N: rockchip ARM/SAMSUNG EXYNOS ARM ARCHITECTURES -M: Kukjin Kim M: Krzysztof Kozlowski L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) L: linux-samsung-soc@vger.kernel.org @@ -15486,7 +15485,6 @@ F: include/linux/clk/samsung.h F: include/linux/platform_data/clk-s3c2410.h SAMSUNG SPI DRIVERS -M: Kukjin Kim M: Krzysztof Kozlowski M: Andi Shyti L: linux-spi@vger.kernel.org -- cgit v1.2.3 From 666674cc1872abd917dc7a07a70ade3f3a1e91bf Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 16 Oct 2020 08:18:48 +0200 Subject: MAINTAINERS: Move Sangbeom Kim to credits Sangbeom Kim upstreamed the Samsung SoC Sound and PMIC (MFD, regulator, RTC) drivers. However his contributions and LKML activity ends in 2014: https://lore.kernel.org/lkml/?q=f%3A%22Sangbeom+Kim%22 Move Sangbeom Kim to the CREDITS file. Thank you for the effort you put in to the upstream Samsung support. Link: https://lore.kernel.org/r/20201016061848.6258-2-krzk@kernel.org Cc: Sangbeom Kim Cc: Sylwester Nawrocki Cc: Mark Brown Signed-off-by: Krzysztof Kozlowski Signed-off-by: Arnd Bergmann --- CREDITS | 5 +++++ MAINTAINERS | 2 -- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/CREDITS b/CREDITS index 5df027e12ff7..8592e45e3932 100644 --- a/CREDITS +++ b/CREDITS @@ -1914,6 +1914,11 @@ N: Kukjin Kim E: kgene@kernel.org D: Samsung S3C, S5P and Exynos ARM architectures +N: Sangbeom Kim +E: sbkim73@samsung.com +D: Samsung SoC Audio (ASoC) drivers +D: Samsung PMIC (RTC, regulators, MFD) drivers + N: Russell King E: rmk@arm.linux.org.uk D: Linux/arm integrator, maintainer & hacker diff --git a/MAINTAINERS b/MAINTAINERS index 5664ee1f543f..608fc8484c02 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -15368,7 +15368,6 @@ F: security/safesetid/ SAMSUNG AUDIO (ASoC) DRIVERS M: Krzysztof Kozlowski -M: Sangbeom Kim M: Sylwester Nawrocki L: alsa-devel@alsa-project.org (moderated for non-subscribers) S: Supported @@ -15403,7 +15402,6 @@ S: Maintained F: drivers/platform/x86/samsung-laptop.c SAMSUNG MULTIFUNCTION PMIC DEVICE DRIVERS -M: Sangbeom Kim M: Krzysztof Kozlowski M: Bartlomiej Zolnierkiewicz L: linux-kernel@vger.kernel.org -- cgit v1.2.3 From 7d54eb90cd44d809b6da56baa8b55299ca601473 Mon Sep 17 00:00:00 2001 From: Icenowy Zheng Date: Sun, 4 Oct 2020 07:50:18 +0800 Subject: dt-bindings: sram: sunxi-sram: add V3s compatible string Add compatible string for V3s, with H3 one as fallback. This is used in device tree now, but not standardized in DT binding. Signed-off-by: Icenowy Zheng Signed-off-by: Maxime Ripard Link: https://lore.kernel.org/r/20201003235018.1121618-2-icenowy@aosc.io Signed-off-by: Arnd Bergmann --- .../devicetree/bindings/sram/allwinner,sun4i-a10-system-control.yaml | 3 +++ 1 file changed, 3 insertions(+) diff --git a/Documentation/devicetree/bindings/sram/allwinner,sun4i-a10-system-control.yaml b/Documentation/devicetree/bindings/sram/allwinner,sun4i-a10-system-control.yaml index 6ebcbc153691..b66a07e21d1e 100644 --- a/Documentation/devicetree/bindings/sram/allwinner,sun4i-a10-system-control.yaml +++ b/Documentation/devicetree/bindings/sram/allwinner,sun4i-a10-system-control.yaml @@ -33,6 +33,9 @@ properties: - const: allwinner,sun4i-a10-system-control - const: allwinner,sun8i-a23-system-control - const: allwinner,sun8i-h3-system-control + - items: + - const: allwinner,sun8i-v3s-system-control + - const: allwinner,sun8i-h3-system-control - items: - const: allwinner,sun8i-r40-system-control - const: allwinner,sun4i-a10-system-control -- cgit v1.2.3 From b0fc70ce1f028e14a37c186d9f7a55e51439b83a Mon Sep 17 00:00:00 2001 From: Jisheng Zhang Date: Fri, 9 Oct 2020 15:08:31 +0800 Subject: arm64: berlin: Select DW_APB_TIMER_OF Berlin SoCs always contain some DW APB timers which can be used as an always-on broadcast timer. Link: https://lore.kernel.org/r/20201009150536.214181fb@xhacker.debian Cc: # v3.14+ Signed-off-by: Jisheng Zhang Signed-off-by: Arnd Bergmann --- arch/arm64/Kconfig.platforms | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm64/Kconfig.platforms b/arch/arm64/Kconfig.platforms index 6f2494dd6d60..5c4ac1c9f4e0 100644 --- a/arch/arm64/Kconfig.platforms +++ b/arch/arm64/Kconfig.platforms @@ -54,6 +54,7 @@ config ARCH_BCM_IPROC config ARCH_BERLIN bool "Marvell Berlin SoC Family" select DW_APB_ICTL + select DW_APB_TIMER_OF select GPIOLIB select PINCTRL help -- cgit v1.2.3 From 202f8e5c4975a95babf3bcdfb2c18952f06b030a Mon Sep 17 00:00:00 2001 From: Lubomir Rintel Date: Sat, 26 Sep 2020 01:48:04 +0200 Subject: ARM: dts: mmp3: Add power domain for the camera The camera interfaces on MMP3 are on a separate power island that needs to be turned on for them to operate and, ideally, turned off when the cameras are not in use. This hooks the power island with the camera interfaces in the device tree. Link: https://lore.kernel.org/r/20200925234805.228251-2-lkundrak@v3.sk Signed-off-by: Lubomir Rintel Signed-off-by: Arnd Bergmann --- arch/arm/boot/dts/mmp3.dtsi | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm/boot/dts/mmp3.dtsi b/arch/arm/boot/dts/mmp3.dtsi index cc4efd0efabd..4ae630d37d09 100644 --- a/arch/arm/boot/dts/mmp3.dtsi +++ b/arch/arm/boot/dts/mmp3.dtsi @@ -296,6 +296,7 @@ interrupts = ; clocks = <&soc_clocks MMP2_CLK_CCIC0>; clock-names = "axi"; + power-domains = <&soc_clocks MMP3_POWER_DOMAIN_CAMERA>; #clock-cells = <0>; clock-output-names = "mclk"; status = "disabled"; @@ -307,6 +308,7 @@ interrupts = ; clocks = <&soc_clocks MMP2_CLK_CCIC1>; clock-names = "axi"; + power-domains = <&soc_clocks MMP3_POWER_DOMAIN_CAMERA>; #clock-cells = <0>; clock-output-names = "mclk"; status = "disabled"; -- cgit v1.2.3 From 30bbbcda5be45dd2d8f060b4a39452061d72da38 Mon Sep 17 00:00:00 2001 From: Lubomir Rintel Date: Sat, 26 Sep 2020 01:48:05 +0200 Subject: ARM: dts: mmp2-olpc-xo-1-75: Use plural form of "-gpios" This makes validation happier. Link: https://lore.kernel.org/r/20200925234805.228251-3-lkundrak@v3.sk Signed-off-by: Lubomir Rintel Signed-off-by: Arnd Bergmann --- arch/arm/boot/dts/mmp2-olpc-xo-1-75.dts | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/boot/dts/mmp2-olpc-xo-1-75.dts b/arch/arm/boot/dts/mmp2-olpc-xo-1-75.dts index f1a41152e9dd..adde62d6fce7 100644 --- a/arch/arm/boot/dts/mmp2-olpc-xo-1-75.dts +++ b/arch/arm/boot/dts/mmp2-olpc-xo-1-75.dts @@ -227,12 +227,12 @@ /delete-property/ #size-cells; spi-slave; status = "okay"; - ready-gpio = <&gpio 125 GPIO_ACTIVE_HIGH>; + ready-gpios = <&gpio 125 GPIO_ACTIVE_HIGH>; slave { compatible = "olpc,xo1.75-ec"; spi-cpha; - cmd-gpio = <&gpio 155 GPIO_ACTIVE_HIGH>; + cmd-gpios = <&gpio 155 GPIO_ACTIVE_HIGH>; }; }; -- cgit v1.2.3 From 4cd2bb12981165f865d2b8ed92b446b52310ef74 Mon Sep 17 00:00:00 2001 From: Quanyang Wang Date: Tue, 29 Sep 2020 16:20:27 +0800 Subject: time/sched_clock: Mark sched_clock_read_begin/retry() as notrace Since sched_clock_read_begin() and sched_clock_read_retry() are called by notrace function sched_clock(), they shouldn't be traceable either, or else ftrace_graph_caller will run into a dead loop on the path as below (arm for instance): ftrace_graph_caller() prepare_ftrace_return() function_graph_enter() ftrace_push_return_trace() trace_clock_local() sched_clock() sched_clock_read_begin/retry() Fixes: 1b86abc1c645 ("sched_clock: Expose struct clock_read_data") Signed-off-by: Quanyang Wang Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra (Intel) Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20200929082027.16787-1-quanyang.wang@windriver.com --- kernel/time/sched_clock.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/time/sched_clock.c b/kernel/time/sched_clock.c index 0642013dace4..b1b9b12899f5 100644 --- a/kernel/time/sched_clock.c +++ b/kernel/time/sched_clock.c @@ -68,13 +68,13 @@ static inline u64 notrace cyc_to_ns(u64 cyc, u32 mult, u32 shift) return (cyc * mult) >> shift; } -struct clock_read_data *sched_clock_read_begin(unsigned int *seq) +notrace struct clock_read_data *sched_clock_read_begin(unsigned int *seq) { *seq = raw_read_seqcount_latch(&cd.seq); return cd.read_data + (*seq & 1); } -int sched_clock_read_retry(unsigned int seq) +notrace int sched_clock_read_retry(unsigned int seq) { return read_seqcount_latch_retry(&cd.seq, seq); } -- cgit v1.2.3 From 5254cb87c0423f73c8036235795788a132e8956e Mon Sep 17 00:00:00 2001 From: YueHaibing Date: Wed, 9 Sep 2020 21:48:50 +0800 Subject: hrtimer: Remove unused inline function debug_hrtimer_free() There is no caller in tree, remove it. Signed-off-by: YueHaibing Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/r/20200909134850.21940-1-yuehaibing@huawei.com --- kernel/time/hrtimer.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 3624b9b5835d..387b4bef7dd1 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -425,11 +425,6 @@ static inline void debug_hrtimer_deactivate(struct hrtimer *timer) debug_object_deactivate(timer, &hrtimer_debug_descr); } -static inline void debug_hrtimer_free(struct hrtimer *timer) -{ - debug_object_free(timer, &hrtimer_debug_descr); -} - static void __hrtimer_init(struct hrtimer *timer, clockid_t clock_id, enum hrtimer_mode mode); -- cgit v1.2.3 From 9010e3876e1c3f7b1c3769bee519d6a871589aca Mon Sep 17 00:00:00 2001 From: YueHaibing Date: Wed, 9 Sep 2020 21:47:49 +0800 Subject: timers: Remove unused inline funtion debug_timer_free() There is no caller in tree, remove it. Signed-off-by: YueHaibing Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/r/20200909134749.32300-1-yuehaibing@huawei.com --- kernel/time/timer.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/kernel/time/timer.c b/kernel/time/timer.c index de37e33a868d..c3ad64fb9d8b 100644 --- a/kernel/time/timer.c +++ b/kernel/time/timer.c @@ -732,11 +732,6 @@ static inline void debug_timer_deactivate(struct timer_list *timer) debug_object_deactivate(timer, &timer_debug_descr); } -static inline void debug_timer_free(struct timer_list *timer) -{ - debug_object_free(timer, &timer_debug_descr); -} - static inline void debug_timer_assert_init(struct timer_list *timer) { debug_object_assert_init(timer, &timer_debug_descr); -- cgit v1.2.3 From 84c971b356379c621df595bd00c3114579dfa59f Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Thu, 15 Oct 2020 11:36:40 +0200 Subject: drm/sun4i: frontend: Rework a bit the phase data The scaler filter phase setup in the allwinner kernel has two different cases for setting up the scaler filter, the first one using different phase parameters for the two channels, and the second one reusing the first channel parameters on the second channel. The allwinner kernel has a third option where the horizontal phase of the second channel will be set to a different value than the vertical one (and seems like it's the same value than one used on the first channel). However, that code path seems to never be taken, so we can ignore it for now, and it's essentially what we're doing so far as well. Since we will have always the same values across each components of the filter setup for a given channel, we can simplify a bit our frontend structure by only storing the phase value we want to apply to a given channel. Signed-off-by: Maxime Ripard Acked-by: Jernej Skrabec Link: https://patchwork.freedesktop.org/patch/msgid/20201015093642.261440-1-maxime@cerno.tech --- drivers/gpu/drm/sun4i/sun4i_frontend.c | 34 ++++++++-------------------------- drivers/gpu/drm/sun4i/sun4i_frontend.h | 6 +----- 2 files changed, 9 insertions(+), 31 deletions(-) diff --git a/drivers/gpu/drm/sun4i/sun4i_frontend.c b/drivers/gpu/drm/sun4i/sun4i_frontend.c index b51cc685c13a..b9eaaef41876 100644 --- a/drivers/gpu/drm/sun4i/sun4i_frontend.c +++ b/drivers/gpu/drm/sun4i/sun4i_frontend.c @@ -443,17 +443,17 @@ int sun4i_frontend_update_formats(struct sun4i_frontend *frontend, * related to the scaler FIR filter phase parameters. */ regmap_write(frontend->regs, SUN4I_FRONTEND_CH0_HORZPHASE_REG, - frontend->data->ch_phase[0].horzphase); + frontend->data->ch_phase[0]); regmap_write(frontend->regs, SUN4I_FRONTEND_CH1_HORZPHASE_REG, - frontend->data->ch_phase[1].horzphase); + frontend->data->ch_phase[1]); regmap_write(frontend->regs, SUN4I_FRONTEND_CH0_VERTPHASE0_REG, - frontend->data->ch_phase[0].vertphase[0]); + frontend->data->ch_phase[0]); regmap_write(frontend->regs, SUN4I_FRONTEND_CH1_VERTPHASE0_REG, - frontend->data->ch_phase[1].vertphase[0]); + frontend->data->ch_phase[1]); regmap_write(frontend->regs, SUN4I_FRONTEND_CH0_VERTPHASE1_REG, - frontend->data->ch_phase[0].vertphase[1]); + frontend->data->ch_phase[0]); regmap_write(frontend->regs, SUN4I_FRONTEND_CH1_VERTPHASE1_REG, - frontend->data->ch_phase[1].vertphase[1]); + frontend->data->ch_phase[1]); /* * Checking the input format is sufficient since we currently only @@ -687,30 +687,12 @@ static const struct dev_pm_ops sun4i_frontend_pm_ops = { }; static const struct sun4i_frontend_data sun4i_a10_frontend = { - .ch_phase = { - { - .horzphase = 0, - .vertphase = { 0, 0 }, - }, - { - .horzphase = 0xfc000, - .vertphase = { 0xfc000, 0xfc000 }, - }, - }, + .ch_phase = { 0x000, 0xfc000 }, .has_coef_rdy = true, }; static const struct sun4i_frontend_data sun8i_a33_frontend = { - .ch_phase = { - { - .horzphase = 0x400, - .vertphase = { 0x400, 0x400 }, - }, - { - .horzphase = 0x400, - .vertphase = { 0x400, 0x400 }, - }, - }, + .ch_phase = { 0x400, 0x400 }, .has_coef_access_ctrl = true, }; diff --git a/drivers/gpu/drm/sun4i/sun4i_frontend.h b/drivers/gpu/drm/sun4i/sun4i_frontend.h index 0c382c1ddb0f..2e7b76e50c2b 100644 --- a/drivers/gpu/drm/sun4i/sun4i_frontend.h +++ b/drivers/gpu/drm/sun4i/sun4i_frontend.h @@ -115,11 +115,7 @@ struct reset_control; struct sun4i_frontend_data { bool has_coef_access_ctrl; bool has_coef_rdy; - - struct { - u32 horzphase; - u32 vertphase[2]; - } ch_phase[2]; + u32 ch_phase[2]; }; struct sun4i_frontend { -- cgit v1.2.3 From 2db9ef9d9e6ea89a9feb5338f58d1f8f83875577 Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Thu, 15 Oct 2020 11:36:41 +0200 Subject: drm/sun4i: frontend: Reuse the ch0 phase for RGB formats When using the scaler on the A10-like frontend with single-planar formats, the current code will setup the channel 0 filter (used for the R or Y component) with a different phase parameter than the channel 1 filter (used for the G/B or U/V components). This creates a bleed out that keeps repeating on of the last line of the RGB plane across the rest of the display. The Allwinner BSP either applies the same phase parameter over both channels or use a separate one, the condition being whether the input format is YUV420 or not. Since YUV420 is both subsampled and multi-planar, and since YUYV is subsampled but single-planar, we can rule out the subsampling and assume that the condition is actually whether the format is single or multi-planar. And it looks like applying the same phase parameter over both channels for single-planar formats fixes our issue, while we keep the multi-planar formats working properly. Reported-by: Taras Galchenko Signed-off-by: Maxime Ripard Acked-by: Jernej Skrabec Link: https://patchwork.freedesktop.org/patch/msgid/20201015093642.261440-2-maxime@cerno.tech --- drivers/gpu/drm/sun4i/sun4i_frontend.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/sun4i/sun4i_frontend.c b/drivers/gpu/drm/sun4i/sun4i_frontend.c index b9eaaef41876..80f0c4392fef 100644 --- a/drivers/gpu/drm/sun4i/sun4i_frontend.c +++ b/drivers/gpu/drm/sun4i/sun4i_frontend.c @@ -407,6 +407,7 @@ int sun4i_frontend_update_formats(struct sun4i_frontend *frontend, struct drm_framebuffer *fb = state->fb; const struct drm_format_info *format = fb->format; uint64_t modifier = fb->modifier; + unsigned int ch1_phase_idx; u32 out_fmt_val; u32 in_fmt_val, in_mod_val, in_ps_val; unsigned int i; @@ -442,18 +443,19 @@ int sun4i_frontend_update_formats(struct sun4i_frontend *frontend, * I have no idea what this does exactly, but it seems to be * related to the scaler FIR filter phase parameters. */ + ch1_phase_idx = (format->num_planes > 1) ? 1 : 0; regmap_write(frontend->regs, SUN4I_FRONTEND_CH0_HORZPHASE_REG, frontend->data->ch_phase[0]); regmap_write(frontend->regs, SUN4I_FRONTEND_CH1_HORZPHASE_REG, - frontend->data->ch_phase[1]); + frontend->data->ch_phase[ch1_phase_idx]); regmap_write(frontend->regs, SUN4I_FRONTEND_CH0_VERTPHASE0_REG, frontend->data->ch_phase[0]); regmap_write(frontend->regs, SUN4I_FRONTEND_CH1_VERTPHASE0_REG, - frontend->data->ch_phase[1]); + frontend->data->ch_phase[ch1_phase_idx]); regmap_write(frontend->regs, SUN4I_FRONTEND_CH0_VERTPHASE1_REG, frontend->data->ch_phase[0]); regmap_write(frontend->regs, SUN4I_FRONTEND_CH1_VERTPHASE1_REG, - frontend->data->ch_phase[1]); + frontend->data->ch_phase[ch1_phase_idx]); /* * Checking the input format is sufficient since we currently only -- cgit v1.2.3 From e3190b5e9462067714d267c40d8c8c1d0463dda3 Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Thu, 15 Oct 2020 11:36:42 +0200 Subject: drm/sun4i: frontend: Fix the scaler phase on A33 The A33 has a different phase parameter in the Allwinner BSP on the channel1 than the one currently applied. Fix this. Signed-off-by: Maxime Ripard Acked-by: Jernej Skrabec Link: https://patchwork.freedesktop.org/patch/msgid/20201015093642.261440-3-maxime@cerno.tech --- drivers/gpu/drm/sun4i/sun4i_frontend.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/sun4i/sun4i_frontend.c b/drivers/gpu/drm/sun4i/sun4i_frontend.c index 80f0c4392fef..edb60ae0a9b7 100644 --- a/drivers/gpu/drm/sun4i/sun4i_frontend.c +++ b/drivers/gpu/drm/sun4i/sun4i_frontend.c @@ -694,7 +694,7 @@ static const struct sun4i_frontend_data sun4i_a10_frontend = { }; static const struct sun4i_frontend_data sun8i_a33_frontend = { - .ch_phase = { 0x400, 0x400 }, + .ch_phase = { 0x400, 0xfc400 }, .has_coef_access_ctrl = true, }; -- cgit v1.2.3 From 897dbea6b716c0f2c5bcd4ba1eb4d809caba290c Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 26 Oct 2020 12:49:05 +0300 Subject: drm/v3d: Fix double free in v3d_submit_cl_ioctl() Originally this error path used to leak "bin" but then we accidentally applied two separate commits to fix it and ended up with a double free. Signed-off-by: Dan Carpenter Signed-off-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20201026094905.GA1634423@mwanda --- drivers/gpu/drm/v3d/v3d_gem.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/v3d/v3d_gem.c b/drivers/gpu/drm/v3d/v3d_gem.c index 915f8bfdb58c..182c586525eb 100644 --- a/drivers/gpu/drm/v3d/v3d_gem.c +++ b/drivers/gpu/drm/v3d/v3d_gem.c @@ -568,7 +568,6 @@ v3d_submit_cl_ioctl(struct drm_device *dev, void *data, ret = v3d_job_init(v3d, file_priv, &bin->base, v3d_job_free, args->in_sync_bcl); if (ret) { - kfree(bin); v3d_job_put(&render->base); kfree(bin); return ret; -- cgit v1.2.3 From cb47755725da7b90fecbb2aa82ac3b24a7adb89b Mon Sep 17 00:00:00 2001 From: Zeng Tao Date: Tue, 1 Sep 2020 17:30:13 +0800 Subject: time: Prevent undefined behaviour in timespec64_to_ns() UBSAN reports: Undefined behaviour in ./include/linux/time64.h:127:27 signed integer overflow: 17179869187 * 1000000000 cannot be represented in type 'long long int' Call Trace: timespec64_to_ns include/linux/time64.h:127 [inline] set_cpu_itimer+0x65c/0x880 kernel/time/itimer.c:180 do_setitimer+0x8e/0x740 kernel/time/itimer.c:245 __x64_sys_setitimer+0x14c/0x2c0 kernel/time/itimer.c:336 do_syscall_64+0xa1/0x540 arch/x86/entry/common.c:295 Commit bd40a175769d ("y2038: itimer: change implementation to timespec64") replaced the original conversion which handled time clamping correctly with timespec64_to_ns() which has no overflow protection. Fix it in timespec64_to_ns() as this is not necessarily limited to the usage in itimers. [ tglx: Added comment and adjusted the fixes tag ] Fixes: 361a3bf00582 ("time64: Add time64.h header and define struct timespec64") Signed-off-by: Zeng Tao Signed-off-by: Thomas Gleixner Reviewed-by: Arnd Bergmann Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/1598952616-6416-1-git-send-email-prime.zeng@hisilicon.com --- include/linux/time64.h | 4 ++++ kernel/time/itimer.c | 4 ---- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/include/linux/time64.h b/include/linux/time64.h index c9dcb3e5781f..5117cb5b5656 100644 --- a/include/linux/time64.h +++ b/include/linux/time64.h @@ -124,6 +124,10 @@ static inline bool timespec64_valid_settod(const struct timespec64 *ts) */ static inline s64 timespec64_to_ns(const struct timespec64 *ts) { + /* Prevent multiplication overflow */ + if ((unsigned long long)ts->tv_sec >= KTIME_SEC_MAX) + return KTIME_MAX; + return ((s64) ts->tv_sec * NSEC_PER_SEC) + ts->tv_nsec; } diff --git a/kernel/time/itimer.c b/kernel/time/itimer.c index ca4e6d57d68b..00629e658ca1 100644 --- a/kernel/time/itimer.c +++ b/kernel/time/itimer.c @@ -172,10 +172,6 @@ static void set_cpu_itimer(struct task_struct *tsk, unsigned int clock_id, u64 oval, nval, ointerval, ninterval; struct cpu_itimer *it = &tsk->signal->it[clock_id]; - /* - * Use the to_ktime conversion because that clamps the maximum - * value to KTIME_MAX and avoid multiplication overflows. - */ nval = timespec64_to_ns(&value->it_value); ninterval = timespec64_to_ns(&value->it_interval); -- cgit v1.2.3 From 4230e2deaa484b385aa01d598b2aea8e7f2660a6 Mon Sep 17 00:00:00 2001 From: Zong Li Date: Wed, 21 Oct 2020 15:38:39 +0800 Subject: stop_machine, rcu: Mark functions as notrace Some architectures assume that the stopped CPUs don't make function calls to traceable functions when they are in the stopped state. See also commit cb9d7fd51d9f ("watchdog: Mark watchdog touch functions as notrace"). Violating this assumption causes kernel crashes when switching tracer on RISC-V. Mark rcu_momentary_dyntick_idle() and stop_machine_yield() notrace to prevent this. Fixes: 4ecf0a43e729 ("processor: get rid of cpu_relax_yield") Fixes: 366237e7b083 ("stop_machine: Provide RCU quiescent state in multi_cpu_stop()") Signed-off-by: Zong Li Signed-off-by: Thomas Gleixner Tested-by: Atish Patra Tested-by: Colin Ian King Acked-by: Steven Rostedt (VMware) Acked-by: Paul E. McKenney Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20201021073839.43935-1-zong.li@sifive.com --- kernel/rcu/tree.c | 2 +- kernel/stop_machine.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 06895ef85d69..2a52f42f64b6 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -409,7 +409,7 @@ bool rcu_eqs_special_set(int cpu) * * The caller must have disabled interrupts and must not be idle. */ -void rcu_momentary_dyntick_idle(void) +notrace void rcu_momentary_dyntick_idle(void) { int special; diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index 865bb0228ab6..890b79cf0e7c 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -178,7 +178,7 @@ static void ack_state(struct multi_stop_data *msdata) set_state(msdata, msdata->state + 1); } -void __weak stop_machine_yield(const struct cpumask *cpumask) +notrace void __weak stop_machine_yield(const struct cpumask *cpumask) { cpu_relax(); } -- cgit v1.2.3 From 0fa97e9403c767a4af63cfb5a3badbbe69426085 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Mon, 19 Oct 2020 18:14:07 -0700 Subject: arm64: vdso32: Allow ld.lld to properly link the VDSO As it stands now, the vdso32 Makefile hardcodes the linker to ld.bfd using -fuse-ld=bfd with $(CC). This was taken from the arm vDSO Makefile, as the comment notes, done in commit d2b30cd4b722 ("ARM: 8384/1: VDSO: force use of BFD linker"). Commit fe00e50b2db8 ("ARM: 8858/1: vdso: use $(LD) instead of $(CC) to link VDSO") changed that Makefile to use $(LD) directly instead of through $(CC), which matches how the rest of the kernel operates. Since then, LD=ld.lld means that the arm vDSO will be linked with ld.lld, which has shown no problems so far. Allow ld.lld to link this vDSO as we do the regular arm vDSO. To do this, we need to do a few things: * Add a LD_COMPAT variable, which defaults to $(CROSS_COMPILE_COMPAT)ld with gcc and $(LD) if LLVM is 1, which will be ld.lld, or $(CROSS_COMPILE_COMPAT)ld if not, which matches the logic of the main Makefile. It is overrideable for further customization and avoiding breakage. * Eliminate cc32-ldoption, which matches commit 055efab3120b ("kbuild: drop support for cc-ldoption"). With those, we can use $(LD_COMPAT) in cmd_ldvdso and change the flags from compiler linker flags to linker flags directly. We eliminate -mfloat-abi=soft because it is not handled by the linker. Reported-by: Nick Desaulniers Signed-off-by: Nathan Chancellor Tested-by: Nick Desaulniers Reviewed-by: Nick Desaulniers Reviewed-by: Vincenzo Frascino Link: https://github.com/ClangBuiltLinux/linux/issues/1033 Link: https://lore.kernel.org/r/20201020011406.1818918-1-natechancellor@gmail.com Signed-off-by: Will Deacon --- arch/arm64/kernel/vdso32/Makefile | 23 ++++++++++++----------- 1 file changed, 12 insertions(+), 11 deletions(-) diff --git a/arch/arm64/kernel/vdso32/Makefile b/arch/arm64/kernel/vdso32/Makefile index 7f96a1a9f68c..79280c53b9a6 100644 --- a/arch/arm64/kernel/vdso32/Makefile +++ b/arch/arm64/kernel/vdso32/Makefile @@ -22,16 +22,21 @@ endif CC_COMPAT ?= $(CC) CC_COMPAT += $(CC_COMPAT_CLANG_FLAGS) + +ifneq ($(LLVM),) +LD_COMPAT ?= $(LD) +else +LD_COMPAT ?= $(CROSS_COMPILE_COMPAT)ld +endif else CC_COMPAT ?= $(CROSS_COMPILE_COMPAT)gcc +LD_COMPAT ?= $(CROSS_COMPILE_COMPAT)ld endif cc32-option = $(call try-run,\ $(CC_COMPAT) $(1) -c -x c /dev/null -o "$$TMP",$(1),$(2)) cc32-disable-warning = $(call try-run,\ $(CC_COMPAT) -W$(strip $(1)) -c -x c /dev/null -o "$$TMP",-Wno-$(strip $(1))) -cc32-ldoption = $(call try-run,\ - $(CC_COMPAT) $(1) -nostdlib -x c /dev/null -o "$$TMP",$(1),$(2)) cc32-as-instr = $(call try-run,\ printf "%b\n" "$(1)" | $(CC_COMPAT) $(VDSO_AFLAGS) -c -x assembler -o "$$TMP" -,$(2),$(3)) @@ -122,14 +127,10 @@ dmbinstr := $(call cc32-as-instr,dmb ishld,-DCONFIG_AS_DMB_ISHLD=1) VDSO_CFLAGS += $(dmbinstr) VDSO_AFLAGS += $(dmbinstr) -VDSO_LDFLAGS := $(VDSO_CPPFLAGS) # From arm vDSO Makefile -VDSO_LDFLAGS += -Wl,-Bsymbolic -Wl,--no-undefined -Wl,-soname=linux-vdso.so.1 -VDSO_LDFLAGS += -Wl,-z,max-page-size=4096 -Wl,-z,common-page-size=4096 -VDSO_LDFLAGS += -nostdlib -shared -mfloat-abi=soft -VDSO_LDFLAGS += -Wl,--hash-style=sysv -VDSO_LDFLAGS += -Wl,--build-id=sha1 -VDSO_LDFLAGS += $(call cc32-ldoption,-fuse-ld=bfd) +VDSO_LDFLAGS += -Bsymbolic --no-undefined -soname=linux-vdso.so.1 +VDSO_LDFLAGS += -z max-page-size=4096 -z common-page-size=4096 +VDSO_LDFLAGS += -nostdlib -shared --hash-style=sysv --build-id=sha1 # Borrow vdsomunge.c from the arm vDSO @@ -189,8 +190,8 @@ quiet_cmd_vdsold_and_vdso_check = LD32 $@ cmd_vdsold_and_vdso_check = $(cmd_vdsold); $(cmd_vdso_check) quiet_cmd_vdsold = LD32 $@ - cmd_vdsold = $(CC_COMPAT) -Wp,-MD,$(depfile) $(VDSO_LDFLAGS) \ - -Wl,-T $(filter %.lds,$^) $(filter %.o,$^) -o $@ + cmd_vdsold = $(LD_COMPAT) $(VDSO_LDFLAGS) \ + -T $(filter %.lds,$^) $(filter %.o,$^) -o $@ quiet_cmd_vdsocc = CC32 $@ cmd_vdsocc = $(CC_COMPAT) -Wp,-MD,$(depfile) $(VDSO_CFLAGS) -c -o $@ $< quiet_cmd_vdsocc_gettimeofday = CC32 $@ -- cgit v1.2.3 From 8e90b4b1305a80b1d7712370a163eff269ac1ba2 Mon Sep 17 00:00:00 2001 From: Vasily Gorbik Date: Mon, 26 Oct 2020 10:52:26 +0100 Subject: s390: correct __bootdata / __bootdata_preserved macros Currently s390 build is broken. SECTCMP .boot.data error: section .boot.data differs between vmlinux and arch/s390/boot/compressed/vmlinux make[2]: *** [arch/s390/boot/section_cmp.boot.data] Error 1 SECTCMP .boot.preserved.data error: section .boot.preserved.data differs between vmlinux and arch/s390/boot/compressed/vmlinux make[2]: *** [arch/s390/boot/section_cmp.boot.preserved.data] Error 1 make[1]: *** [bzImage] Error 2 Commit 33def8498fdd ("treewide: Convert macro and uses of __section(foo) to __section("foo")") converted all __section(foo) to __section("foo"). This is wrong for __bootdata / __bootdata_preserved macros which want variable names to be a part of intermediate section names .boot.data. and .boot.preserved.data.. Those sections are later sorted by alignment + name and merged together into final .boot.data / .boot.preserved.data sections. Those sections must be identical in the decompressor and the decompressed kernel (that is checked during the build). Fixes: 33def8498fdd ("treewide: Convert macro and uses of __section(foo) to __section("foo")") Signed-off-by: Vasily Gorbik Signed-off-by: Heiko Carstens --- arch/s390/include/asm/sections.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/s390/include/asm/sections.h b/arch/s390/include/asm/sections.h index a996d3990a02..0c2151451ba5 100644 --- a/arch/s390/include/asm/sections.h +++ b/arch/s390/include/asm/sections.h @@ -26,14 +26,14 @@ static inline int arch_is_kernel_initmem_freed(unsigned long addr) * final .boot.data section, which should be identical in the decompressor and * the decompressed kernel (that is checked during the build). */ -#define __bootdata(var) __section(".boot.data.var") var +#define __bootdata(var) __section(".boot.data." #var) var /* * .boot.preserved.data is similar to .boot.data, but it is not part of the * .init section and thus will be preserved for later use in the decompressed * kernel. */ -#define __bootdata_preserved(var) __section(".boot.preserved.data.var") var +#define __bootdata_preserved(var) __section(".boot.preserved.data." #var) var extern unsigned long __sdma, __edma; extern unsigned long __stext_dma, __etext_dma; -- cgit v1.2.3 From 52abfcbd57eefdd54737fc8c2dc79d8f46d4a3e5 Mon Sep 17 00:00:00 2001 From: Gabriel Krisman Bertazi Date: Thu, 22 Oct 2020 16:58:41 -0400 Subject: blk-cgroup: Fix memleak on error path If new_blkg allocation raced with blk_policy change and blkg_lookup_check fails, new_blkg is leaked. Acked-by: Tejun Heo Signed-off-by: Gabriel Krisman Bertazi Signed-off-by: Jens Axboe --- block/blk-cgroup.c | 1 + 1 file changed, 1 insertion(+) diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index f9b55614d67d..f9389b7cf823 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -663,6 +663,7 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, blkg = blkg_lookup_check(pos, pol, q); if (IS_ERR(blkg)) { ret = PTR_ERR(blkg); + blkg_free(new_blkg); goto fail_unlock; } -- cgit v1.2.3 From f255c19b3ab46d3cad3b1b2e1036f4c926cb1d0c Mon Sep 17 00:00:00 2001 From: Gabriel Krisman Bertazi Date: Thu, 22 Oct 2020 16:58:42 -0400 Subject: blk-cgroup: Pre-allocate tree node on blkg_conf_prep Similarly to commit 457e490f2b741 ("blkcg: allocate struct blkcg_gq outside request queue spinlock"), blkg_create can also trigger occasional -ENOMEM failures at the radix insertion because any allocation inside blkg_create has to be non-blocking, making it more likely to fail. This causes trouble for userspace tools trying to configure io weights who need to deal with this condition. This patch reduces the occurrence of -ENOMEMs on this path by preloading the radix tree element on a GFP_KERNEL context, such that we guarantee the later non-blocking insertion won't fail. A similar solution exists in blkcg_init_queue for the same situation. Acked-by: Tejun Heo Signed-off-by: Gabriel Krisman Bertazi Signed-off-by: Jens Axboe --- block/blk-cgroup.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index f9389b7cf823..c68bdf58c9a6 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -657,6 +657,12 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, goto fail; } + if (radix_tree_preload(GFP_KERNEL)) { + blkg_free(new_blkg); + ret = -ENOMEM; + goto fail; + } + rcu_read_lock(); spin_lock_irq(&q->queue_lock); @@ -664,7 +670,7 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, if (IS_ERR(blkg)) { ret = PTR_ERR(blkg); blkg_free(new_blkg); - goto fail_unlock; + goto fail_preloaded; } if (blkg) { @@ -673,10 +679,12 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol, blkg = blkg_create(pos, q, new_blkg); if (IS_ERR(blkg)) { ret = PTR_ERR(blkg); - goto fail_unlock; + goto fail_preloaded; } } + radix_tree_preload_end(); + if (pos == blkcg) goto success; } @@ -686,6 +694,8 @@ success: ctx->body = input; return 0; +fail_preloaded: + radix_tree_preload_end(); fail_unlock: spin_unlock_irq(&q->queue_lock); rcu_read_unlock(); -- cgit v1.2.3 From 7837fa88704a66257404bb14144c9e4ab631a28a Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 14 Oct 2020 17:00:51 -0400 Subject: btrfs: drop the path before adding block group sysfs files Dave reported a problem with my rwsem conversion patch where we got the following lockdep splat: ====================================================== WARNING: possible circular locking dependency detected 5.9.0-default+ #1297 Not tainted ------------------------------------------------------ kswapd0/76 is trying to acquire lock: ffff9d5d25df2530 (&delayed_node->mutex){+.+.}-{3:3}, at: __btrfs_release_delayed_node.part.0+0x3f/0x320 [btrfs] but task is already holding lock: ffffffffa40cbba0 (fs_reclaim){+.+.}-{0:0}, at: __fs_reclaim_acquire+0x5/0x30 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #4 (fs_reclaim){+.+.}-{0:0}: __lock_acquire+0x582/0xac0 lock_acquire+0xca/0x430 fs_reclaim_acquire.part.0+0x25/0x30 kmem_cache_alloc+0x30/0x9c0 alloc_inode+0x81/0x90 iget_locked+0xcd/0x1a0 kernfs_get_inode+0x1b/0x130 kernfs_get_tree+0x136/0x210 sysfs_get_tree+0x1a/0x50 vfs_get_tree+0x1d/0xb0 path_mount+0x70f/0xa80 do_mount+0x75/0x90 __x64_sys_mount+0x8e/0xd0 do_syscall_64+0x2d/0x70 entry_SYSCALL_64_after_hwframe+0x44/0xa9 -> #3 (kernfs_mutex){+.+.}-{3:3}: __lock_acquire+0x582/0xac0 lock_acquire+0xca/0x430 __mutex_lock+0xa0/0xaf0 kernfs_add_one+0x23/0x150 kernfs_create_dir_ns+0x58/0x80 sysfs_create_dir_ns+0x70/0xd0 kobject_add_internal+0xbb/0x2d0 kobject_add+0x7a/0xd0 btrfs_sysfs_add_block_group_type+0x141/0x1d0 [btrfs] btrfs_read_block_groups+0x1f1/0x8c0 [btrfs] open_ctree+0x981/0x1108 [btrfs] btrfs_mount_root.cold+0xe/0xb0 [btrfs] legacy_get_tree+0x2d/0x60 vfs_get_tree+0x1d/0xb0 fc_mount+0xe/0x40 vfs_kern_mount.part.0+0x71/0x90 btrfs_mount+0x13b/0x3e0 [btrfs] legacy_get_tree+0x2d/0x60 vfs_get_tree+0x1d/0xb0 path_mount+0x70f/0xa80 do_mount+0x75/0x90 __x64_sys_mount+0x8e/0xd0 do_syscall_64+0x2d/0x70 entry_SYSCALL_64_after_hwframe+0x44/0xa9 -> #2 (btrfs-extent-00){++++}-{3:3}: __lock_acquire+0x582/0xac0 lock_acquire+0xca/0x430 down_read_nested+0x45/0x220 __btrfs_tree_read_lock+0x35/0x1c0 [btrfs] __btrfs_read_lock_root_node+0x3a/0x50 [btrfs] btrfs_search_slot+0x6d4/0xfd0 [btrfs] check_committed_ref+0x69/0x200 [btrfs] btrfs_cross_ref_exist+0x65/0xb0 [btrfs] run_delalloc_nocow+0x446/0x9b0 [btrfs] btrfs_run_delalloc_range+0x61/0x6a0 [btrfs] writepage_delalloc+0xae/0x160 [btrfs] __extent_writepage+0x262/0x420 [btrfs] extent_write_cache_pages+0x2b6/0x510 [btrfs] extent_writepages+0x43/0x90 [btrfs] do_writepages+0x40/0xe0 __writeback_single_inode+0x62/0x610 writeback_sb_inodes+0x20f/0x500 wb_writeback+0xef/0x4a0 wb_do_writeback+0x49/0x2e0 wb_workfn+0x81/0x340 process_one_work+0x233/0x5d0 worker_thread+0x50/0x3b0 kthread+0x137/0x150 ret_from_fork+0x1f/0x30 -> #1 (btrfs-fs-00){++++}-{3:3}: __lock_acquire+0x582/0xac0 lock_acquire+0xca/0x430 down_read_nested+0x45/0x220 __btrfs_tree_read_lock+0x35/0x1c0 [btrfs] __btrfs_read_lock_root_node+0x3a/0x50 [btrfs] btrfs_search_slot+0x6d4/0xfd0 [btrfs] btrfs_lookup_inode+0x3a/0xc0 [btrfs] __btrfs_update_delayed_inode+0x93/0x2c0 [btrfs] __btrfs_commit_inode_delayed_items+0x7de/0x850 [btrfs] __btrfs_run_delayed_items+0x8e/0x140 [btrfs] btrfs_commit_transaction+0x367/0xbc0 [btrfs] btrfs_mksubvol+0x2db/0x470 [btrfs] btrfs_mksnapshot+0x7b/0xb0 [btrfs] __btrfs_ioctl_snap_create+0x16f/0x1a0 [btrfs] btrfs_ioctl_snap_create_v2+0xb0/0xf0 [btrfs] btrfs_ioctl+0xd0b/0x2690 [btrfs] __x64_sys_ioctl+0x6f/0xa0 do_syscall_64+0x2d/0x70 entry_SYSCALL_64_after_hwframe+0x44/0xa9 -> #0 (&delayed_node->mutex){+.+.}-{3:3}: check_prev_add+0x91/0xc60 validate_chain+0xa6e/0x2a20 __lock_acquire+0x582/0xac0 lock_acquire+0xca/0x430 __mutex_lock+0xa0/0xaf0 __btrfs_release_delayed_node.part.0+0x3f/0x320 [btrfs] btrfs_evict_inode+0x3cc/0x560 [btrfs] evict+0xd6/0x1c0 dispose_list+0x48/0x70 prune_icache_sb+0x54/0x80 super_cache_scan+0x121/0x1a0 do_shrink_slab+0x16d/0x3b0 shrink_slab+0xb1/0x2e0 shrink_node+0x230/0x6a0 balance_pgdat+0x325/0x750 kswapd+0x206/0x4d0 kthread+0x137/0x150 ret_from_fork+0x1f/0x30 other info that might help us debug this: Chain exists of: &delayed_node->mutex --> kernfs_mutex --> fs_reclaim Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(fs_reclaim); lock(kernfs_mutex); lock(fs_reclaim); lock(&delayed_node->mutex); *** DEADLOCK *** 3 locks held by kswapd0/76: #0: ffffffffa40cbba0 (fs_reclaim){+.+.}-{0:0}, at: __fs_reclaim_acquire+0x5/0x30 #1: ffffffffa40b8b58 (shrinker_rwsem){++++}-{3:3}, at: shrink_slab+0x54/0x2e0 #2: ffff9d5d322390e8 (&type->s_umount_key#26){++++}-{3:3}, at: trylock_super+0x16/0x50 stack backtrace: CPU: 2 PID: 76 Comm: kswapd0 Not tainted 5.9.0-default+ #1297 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.0-59-gc9ba527-rebuilt.opensuse.org 04/01/2014 Call Trace: dump_stack+0x77/0x97 check_noncircular+0xff/0x110 ? save_trace+0x50/0x470 check_prev_add+0x91/0xc60 validate_chain+0xa6e/0x2a20 ? save_trace+0x50/0x470 __lock_acquire+0x582/0xac0 lock_acquire+0xca/0x430 ? __btrfs_release_delayed_node.part.0+0x3f/0x320 [btrfs] __mutex_lock+0xa0/0xaf0 ? __btrfs_release_delayed_node.part.0+0x3f/0x320 [btrfs] ? __lock_acquire+0x582/0xac0 ? __btrfs_release_delayed_node.part.0+0x3f/0x320 [btrfs] ? btrfs_evict_inode+0x30b/0x560 [btrfs] ? __btrfs_release_delayed_node.part.0+0x3f/0x320 [btrfs] __btrfs_release_delayed_node.part.0+0x3f/0x320 [btrfs] btrfs_evict_inode+0x3cc/0x560 [btrfs] evict+0xd6/0x1c0 dispose_list+0x48/0x70 prune_icache_sb+0x54/0x80 super_cache_scan+0x121/0x1a0 do_shrink_slab+0x16d/0x3b0 shrink_slab+0xb1/0x2e0 shrink_node+0x230/0x6a0 balance_pgdat+0x325/0x750 kswapd+0x206/0x4d0 ? finish_wait+0x90/0x90 ? balance_pgdat+0x750/0x750 kthread+0x137/0x150 ? kthread_mod_delayed_work+0xc0/0xc0 ret_from_fork+0x1f/0x30 This happens because we are still holding the path open when we start adding the sysfs files for the block groups, which creates a dependency on fs_reclaim via the tree lock. Fix this by dropping the path before we start doing anything with sysfs. Reported-by: David Sterba CC: stable@vger.kernel.org # 5.8+ Reviewed-by: Anand Jain Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/block-group.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c index c0f1d6818df7..3ba6f3839d39 100644 --- a/fs/btrfs/block-group.c +++ b/fs/btrfs/block-group.c @@ -2024,6 +2024,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info) key.offset = 0; btrfs_release_path(path); } + btrfs_release_path(path); list_for_each_entry(space_info, &info->space_info, list) { int i; -- cgit v1.2.3 From cad69d139651423b9cb8917f514740ed3f08b10a Mon Sep 17 00:00:00 2001 From: Pujin Shi Date: Wed, 14 Oct 2020 11:24:19 +0800 Subject: btrfs: tree-checker: fix incorrect printk format This patch addresses a compile warning: fs/btrfs/extent-tree.c: In function '__btrfs_free_extent': fs/btrfs/extent-tree.c:3187:4: warning: format '%lu' expects argument of type 'long unsigned int', but argument 8 has type 'unsigned int' [-Wformat=] Fixes: 1c2a07f598d5 ("btrfs: extent-tree: kill BUG_ON() in __btrfs_free_extent()") Reviewed-by: Filipe Manana Signed-off-by: Pujin Shi Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/extent-tree.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 3b21fee13e77..5fd60b13f4f8 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3185,7 +3185,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_tree_block_info *bi; if (item_size < sizeof(*ei) + sizeof(*bi)) { btrfs_crit(info, -"invalid extent item size for key (%llu, %u, %llu) owner %llu, has %u expect >= %lu", +"invalid extent item size for key (%llu, %u, %llu) owner %llu, has %u expect >= %zu", key.objectid, key.type, key.offset, owner_objectid, item_size, sizeof(*ei) + sizeof(*bi)); -- cgit v1.2.3 From 85d07fbe09efd1c529ff3e025e2f0d2c6c96a1b7 Mon Sep 17 00:00:00 2001 From: Daniel Xu Date: Thu, 8 Oct 2020 18:09:10 -0700 Subject: btrfs: tree-checker: validate number of chunk stripes and parity If there's no parity and num_stripes < ncopies, a crafted image can trigger a division by zero in calc_stripe_length(). The image was generated through fuzzing. CC: stable@vger.kernel.org # 5.4+ Reviewed-by: Qu Wenruo Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=209587 Signed-off-by: Daniel Xu Signed-off-by: David Sterba --- fs/btrfs/tree-checker.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c index f0ffd5ee77bd..8784b74f5232 100644 --- a/fs/btrfs/tree-checker.c +++ b/fs/btrfs/tree-checker.c @@ -760,18 +760,36 @@ int btrfs_check_chunk_valid(struct extent_buffer *leaf, u64 type; u64 features; bool mixed = false; + int raid_index; + int nparity; + int ncopies; length = btrfs_chunk_length(leaf, chunk); stripe_len = btrfs_chunk_stripe_len(leaf, chunk); num_stripes = btrfs_chunk_num_stripes(leaf, chunk); sub_stripes = btrfs_chunk_sub_stripes(leaf, chunk); type = btrfs_chunk_type(leaf, chunk); + raid_index = btrfs_bg_flags_to_raid_index(type); + ncopies = btrfs_raid_array[raid_index].ncopies; + nparity = btrfs_raid_array[raid_index].nparity; if (!num_stripes) { chunk_err(leaf, chunk, logical, "invalid chunk num_stripes, have %u", num_stripes); return -EUCLEAN; } + if (num_stripes < ncopies) { + chunk_err(leaf, chunk, logical, + "invalid chunk num_stripes < ncopies, have %u < %d", + num_stripes, ncopies); + return -EUCLEAN; + } + if (nparity && num_stripes == nparity) { + chunk_err(leaf, chunk, logical, + "invalid chunk num_stripes == nparity, have %u == %d", + num_stripes, nparity); + return -EUCLEAN; + } if (!IS_ALIGNED(logical, fs_info->sectorsize)) { chunk_err(leaf, chunk, logical, "invalid chunk logical, have %llu should aligned to %u", -- cgit v1.2.3 From 83bc1560e02e25c6439341352024ebe8488f4fbd Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 12 Oct 2020 11:55:23 +0100 Subject: btrfs: fix use-after-free on readahead extent after failure to create it If we fail to find suitable zones for a new readahead extent, we end up leaving a stale pointer in the global readahead extents radix tree (fs_info->reada_tree), which can trigger the following trace later on: [13367.696354] BUG: kernel NULL pointer dereference, address: 00000000000000b0 [13367.696802] #PF: supervisor read access in kernel mode [13367.697249] #PF: error_code(0x0000) - not-present page [13367.697721] PGD 0 P4D 0 [13367.698171] Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC PTI [13367.698632] CPU: 6 PID: 851214 Comm: btrfs Tainted: G W 5.9.0-rc6-btrfs-next-69 #1 [13367.699100] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 [13367.700069] RIP: 0010:__lock_acquire+0x20a/0x3970 [13367.700562] Code: ff 1f 0f b7 c0 48 0f (...) [13367.701609] RSP: 0018:ffffb14448f57790 EFLAGS: 00010046 [13367.702140] RAX: 0000000000000000 RBX: 29b935140c15e8cf RCX: 0000000000000000 [13367.702698] RDX: 0000000000000002 RSI: ffffffffb3d66bd0 RDI: 0000000000000046 [13367.703240] RBP: ffff8a52ba8ac040 R08: 00000c2866ad9288 R09: 0000000000000001 [13367.703783] R10: 0000000000000001 R11: 00000000b66d9b53 R12: ffff8a52ba8ac9b0 [13367.704330] R13: 0000000000000000 R14: ffff8a532b6333e8 R15: 0000000000000000 [13367.704880] FS: 00007fe1df6b5700(0000) GS:ffff8a5376600000(0000) knlGS:0000000000000000 [13367.705438] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [13367.705995] CR2: 00000000000000b0 CR3: 000000022cca8004 CR4: 00000000003706e0 [13367.706565] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [13367.707127] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [13367.707686] Call Trace: [13367.708246] ? ___slab_alloc+0x395/0x740 [13367.708820] ? reada_add_block+0xae/0xee0 [btrfs] [13367.709383] lock_acquire+0xb1/0x480 [13367.709955] ? reada_add_block+0xe0/0xee0 [btrfs] [13367.710537] ? reada_add_block+0xae/0xee0 [btrfs] [13367.711097] ? rcu_read_lock_sched_held+0x5d/0x90 [13367.711659] ? kmem_cache_alloc_trace+0x8d2/0x990 [13367.712221] ? lock_acquired+0x33b/0x470 [13367.712784] _raw_spin_lock+0x34/0x80 [13367.713356] ? reada_add_block+0xe0/0xee0 [btrfs] [13367.713966] reada_add_block+0xe0/0xee0 [btrfs] [13367.714529] ? btrfs_root_node+0x15/0x1f0 [btrfs] [13367.715077] btrfs_reada_add+0x117/0x170 [btrfs] [13367.715620] scrub_stripe+0x21e/0x10d0 [btrfs] [13367.716141] ? kvm_sched_clock_read+0x5/0x10 [13367.716657] ? __lock_acquire+0x41e/0x3970 [13367.717184] ? scrub_chunk+0x60/0x140 [btrfs] [13367.717697] ? find_held_lock+0x32/0x90 [13367.718254] ? scrub_chunk+0x60/0x140 [btrfs] [13367.718773] ? lock_acquired+0x33b/0x470 [13367.719278] ? scrub_chunk+0xcd/0x140 [btrfs] [13367.719786] scrub_chunk+0xcd/0x140 [btrfs] [13367.720291] scrub_enumerate_chunks+0x270/0x5c0 [btrfs] [13367.720787] ? finish_wait+0x90/0x90 [13367.721281] btrfs_scrub_dev+0x1ee/0x620 [btrfs] [13367.721762] ? rcu_read_lock_any_held+0x8e/0xb0 [13367.722235] ? preempt_count_add+0x49/0xa0 [13367.722710] ? __sb_start_write+0x19b/0x290 [13367.723192] btrfs_ioctl+0x7f5/0x36f0 [btrfs] [13367.723660] ? __fget_files+0x101/0x1d0 [13367.724118] ? find_held_lock+0x32/0x90 [13367.724559] ? __fget_files+0x101/0x1d0 [13367.724982] ? __x64_sys_ioctl+0x83/0xb0 [13367.725399] __x64_sys_ioctl+0x83/0xb0 [13367.725802] do_syscall_64+0x33/0x80 [13367.726188] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [13367.726574] RIP: 0033:0x7fe1df7add87 [13367.726948] Code: 00 00 00 48 8b 05 09 91 (...) [13367.727763] RSP: 002b:00007fe1df6b4d48 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [13367.728179] RAX: ffffffffffffffda RBX: 000055ce1fb596a0 RCX: 00007fe1df7add87 [13367.728604] RDX: 000055ce1fb596a0 RSI: 00000000c400941b RDI: 0000000000000003 [13367.729021] RBP: 0000000000000000 R08: 00007fe1df6b5700 R09: 0000000000000000 [13367.729431] R10: 00007fe1df6b5700 R11: 0000000000000246 R12: 00007ffd922b07de [13367.729842] R13: 00007ffd922b07df R14: 00007fe1df6b4e40 R15: 0000000000802000 [13367.730275] Modules linked in: btrfs blake2b_generic xor (...) [13367.732638] CR2: 00000000000000b0 [13367.733166] ---[ end trace d298b6805556acd9 ]--- What happens is the following: 1) At reada_find_extent() we don't find any existing readahead extent for the metadata extent starting at logical address X; 2) So we proceed to create a new one. We then call btrfs_map_block() to get information about which stripes contain extent X; 3) After that we iterate over the stripes and create only one zone for the readahead extent - only one because reada_find_zone() returned NULL for all iterations except for one, either because a memory allocation failed or it couldn't find the block group of the extent (it may have just been deleted); 4) We then add the new readahead extent to the readahead extents radix tree at fs_info->reada_tree; 5) Then we iterate over each zone of the new readahead extent, and find that the device used for that zone no longer exists, because it was removed or it was the source device of a device replace operation. Since this left 'have_zone' set to 0, after finishing the loop we jump to the 'error' label, call kfree() on the new readahead extent and return without removing it from the radix tree at fs_info->reada_tree; 6) Any future call to reada_find_extent() for the logical address X will find the stale pointer in the readahead extents radix tree, increment its reference counter, which can trigger the use-after-free right away or return it to the caller reada_add_block() that results in the use-after-free of the example trace above. So fix this by making sure we delete the readahead extent from the radix tree if we fail to setup zones for it (when 'have_zone = 0'). Fixes: 319450211842ba ("btrfs: reada: bypass adding extent when all zone failed") CC: stable@vger.kernel.org # 4.9+ Reviewed-by: Johannes Thumshirn Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/reada.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c index 9d4f5316a7e8..e261c3d0cec7 100644 --- a/fs/btrfs/reada.c +++ b/fs/btrfs/reada.c @@ -445,6 +445,8 @@ static struct reada_extent *reada_find_extent(struct btrfs_fs_info *fs_info, } have_zone = 1; } + if (!have_zone) + radix_tree_delete(&fs_info->reada_tree, index); spin_unlock(&fs_info->reada_lock); up_read(&fs_info->dev_replace.rwsem); -- cgit v1.2.3 From 66d204a16c94f24ad08290a7663ab67e7fc04e82 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Mon, 12 Oct 2020 11:55:24 +0100 Subject: btrfs: fix readahead hang and use-after-free after removing a device Very sporadically I had test case btrfs/069 from fstests hanging (for years, it is not a recent regression), with the following traces in dmesg/syslog: [162301.160628] BTRFS info (device sdc): dev_replace from /dev/sdd (devid 2) to /dev/sdg started [162301.181196] BTRFS info (device sdc): scrub: finished on devid 4 with status: 0 [162301.287162] BTRFS info (device sdc): dev_replace from /dev/sdd (devid 2) to /dev/sdg finished [162513.513792] INFO: task btrfs-transacti:1356167 blocked for more than 120 seconds. [162513.514318] Not tainted 5.9.0-rc6-btrfs-next-69 #1 [162513.514522] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [162513.514747] task:btrfs-transacti state:D stack: 0 pid:1356167 ppid: 2 flags:0x00004000 [162513.514751] Call Trace: [162513.514761] __schedule+0x5ce/0xd00 [162513.514765] ? _raw_spin_unlock_irqrestore+0x3c/0x60 [162513.514771] schedule+0x46/0xf0 [162513.514844] wait_current_trans+0xde/0x140 [btrfs] [162513.514850] ? finish_wait+0x90/0x90 [162513.514864] start_transaction+0x37c/0x5f0 [btrfs] [162513.514879] transaction_kthread+0xa4/0x170 [btrfs] [162513.514891] ? btrfs_cleanup_transaction+0x660/0x660 [btrfs] [162513.514894] kthread+0x153/0x170 [162513.514897] ? kthread_stop+0x2c0/0x2c0 [162513.514902] ret_from_fork+0x22/0x30 [162513.514916] INFO: task fsstress:1356184 blocked for more than 120 seconds. [162513.515192] Not tainted 5.9.0-rc6-btrfs-next-69 #1 [162513.515431] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [162513.515680] task:fsstress state:D stack: 0 pid:1356184 ppid:1356177 flags:0x00004000 [162513.515682] Call Trace: [162513.515688] __schedule+0x5ce/0xd00 [162513.515691] ? _raw_spin_unlock_irqrestore+0x3c/0x60 [162513.515697] schedule+0x46/0xf0 [162513.515712] wait_current_trans+0xde/0x140 [btrfs] [162513.515716] ? finish_wait+0x90/0x90 [162513.515729] start_transaction+0x37c/0x5f0 [btrfs] [162513.515743] btrfs_attach_transaction_barrier+0x1f/0x50 [btrfs] [162513.515753] btrfs_sync_fs+0x61/0x1c0 [btrfs] [162513.515758] ? __ia32_sys_fdatasync+0x20/0x20 [162513.515761] iterate_supers+0x87/0xf0 [162513.515765] ksys_sync+0x60/0xb0 [162513.515768] __do_sys_sync+0xa/0x10 [162513.515771] do_syscall_64+0x33/0x80 [162513.515774] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [162513.515781] RIP: 0033:0x7f5238f50bd7 [162513.515782] Code: Bad RIP value. [162513.515784] RSP: 002b:00007fff67b978e8 EFLAGS: 00000206 ORIG_RAX: 00000000000000a2 [162513.515786] RAX: ffffffffffffffda RBX: 000055b1fad2c560 RCX: 00007f5238f50bd7 [162513.515788] RDX: 00000000ffffffff RSI: 000000000daf0e74 RDI: 000000000000003a [162513.515789] RBP: 0000000000000032 R08: 000000000000000a R09: 00007f5239019be0 [162513.515791] R10: fffffffffffff24f R11: 0000000000000206 R12: 000000000000003a [162513.515792] R13: 00007fff67b97950 R14: 00007fff67b97906 R15: 000055b1fad1a340 [162513.515804] INFO: task fsstress:1356185 blocked for more than 120 seconds. [162513.516064] Not tainted 5.9.0-rc6-btrfs-next-69 #1 [162513.516329] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [162513.516617] task:fsstress state:D stack: 0 pid:1356185 ppid:1356177 flags:0x00000000 [162513.516620] Call Trace: [162513.516625] __schedule+0x5ce/0xd00 [162513.516628] ? _raw_spin_unlock_irqrestore+0x3c/0x60 [162513.516634] schedule+0x46/0xf0 [162513.516647] wait_current_trans+0xde/0x140 [btrfs] [162513.516650] ? finish_wait+0x90/0x90 [162513.516662] start_transaction+0x4d7/0x5f0 [btrfs] [162513.516679] btrfs_setxattr_trans+0x3c/0x100 [btrfs] [162513.516686] __vfs_setxattr+0x66/0x80 [162513.516691] __vfs_setxattr_noperm+0x70/0x200 [162513.516697] vfs_setxattr+0x6b/0x120 [162513.516703] setxattr+0x125/0x240 [162513.516709] ? lock_acquire+0xb1/0x480 [162513.516712] ? mnt_want_write+0x20/0x50 [162513.516721] ? rcu_read_lock_any_held+0x8e/0xb0 [162513.516723] ? preempt_count_add+0x49/0xa0 [162513.516725] ? __sb_start_write+0x19b/0x290 [162513.516727] ? preempt_count_add+0x49/0xa0 [162513.516732] path_setxattr+0xba/0xd0 [162513.516739] __x64_sys_setxattr+0x27/0x30 [162513.516741] do_syscall_64+0x33/0x80 [162513.516743] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [162513.516745] RIP: 0033:0x7f5238f56d5a [162513.516746] Code: Bad RIP value. [162513.516748] RSP: 002b:00007fff67b97868 EFLAGS: 00000202 ORIG_RAX: 00000000000000bc [162513.516750] RAX: ffffffffffffffda RBX: 0000000000000001 RCX: 00007f5238f56d5a [162513.516751] RDX: 000055b1fbb0d5a0 RSI: 00007fff67b978a0 RDI: 000055b1fbb0d470 [162513.516753] RBP: 000055b1fbb0d5a0 R08: 0000000000000001 R09: 00007fff67b97700 [162513.516754] R10: 0000000000000004 R11: 0000000000000202 R12: 0000000000000004 [162513.516756] R13: 0000000000000024 R14: 0000000000000001 R15: 00007fff67b978a0 [162513.516767] INFO: task fsstress:1356196 blocked for more than 120 seconds. [162513.517064] Not tainted 5.9.0-rc6-btrfs-next-69 #1 [162513.517365] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [162513.517763] task:fsstress state:D stack: 0 pid:1356196 ppid:1356177 flags:0x00004000 [162513.517780] Call Trace: [162513.517786] __schedule+0x5ce/0xd00 [162513.517789] ? _raw_spin_unlock_irqrestore+0x3c/0x60 [162513.517796] schedule+0x46/0xf0 [162513.517810] wait_current_trans+0xde/0x140 [btrfs] [162513.517814] ? finish_wait+0x90/0x90 [162513.517829] start_transaction+0x37c/0x5f0 [btrfs] [162513.517845] btrfs_attach_transaction_barrier+0x1f/0x50 [btrfs] [162513.517857] btrfs_sync_fs+0x61/0x1c0 [btrfs] [162513.517862] ? __ia32_sys_fdatasync+0x20/0x20 [162513.517865] iterate_supers+0x87/0xf0 [162513.517869] ksys_sync+0x60/0xb0 [162513.517872] __do_sys_sync+0xa/0x10 [162513.517875] do_syscall_64+0x33/0x80 [162513.517878] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [162513.517881] RIP: 0033:0x7f5238f50bd7 [162513.517883] Code: Bad RIP value. [162513.517885] RSP: 002b:00007fff67b978e8 EFLAGS: 00000206 ORIG_RAX: 00000000000000a2 [162513.517887] RAX: ffffffffffffffda RBX: 000055b1fad2c560 RCX: 00007f5238f50bd7 [162513.517889] RDX: 0000000000000000 RSI: 000000007660add2 RDI: 0000000000000053 [162513.517891] RBP: 0000000000000032 R08: 0000000000000067 R09: 00007f5239019be0 [162513.517893] R10: fffffffffffff24f R11: 0000000000000206 R12: 0000000000000053 [162513.517895] R13: 00007fff67b97950 R14: 00007fff67b97906 R15: 000055b1fad1a340 [162513.517908] INFO: task fsstress:1356197 blocked for more than 120 seconds. [162513.518298] Not tainted 5.9.0-rc6-btrfs-next-69 #1 [162513.518672] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [162513.519157] task:fsstress state:D stack: 0 pid:1356197 ppid:1356177 flags:0x00000000 [162513.519160] Call Trace: [162513.519165] __schedule+0x5ce/0xd00 [162513.519168] ? _raw_spin_unlock_irqrestore+0x3c/0x60 [162513.519174] schedule+0x46/0xf0 [162513.519190] wait_current_trans+0xde/0x140 [btrfs] [162513.519193] ? finish_wait+0x90/0x90 [162513.519206] start_transaction+0x4d7/0x5f0 [btrfs] [162513.519222] btrfs_create+0x57/0x200 [btrfs] [162513.519230] lookup_open+0x522/0x650 [162513.519246] path_openat+0x2b8/0xa50 [162513.519270] do_filp_open+0x91/0x100 [162513.519275] ? find_held_lock+0x32/0x90 [162513.519280] ? lock_acquired+0x33b/0x470 [162513.519285] ? do_raw_spin_unlock+0x4b/0xc0 [162513.519287] ? _raw_spin_unlock+0x29/0x40 [162513.519295] do_sys_openat2+0x20d/0x2d0 [162513.519300] do_sys_open+0x44/0x80 [162513.519304] do_syscall_64+0x33/0x80 [162513.519307] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [162513.519309] RIP: 0033:0x7f5238f4a903 [162513.519310] Code: Bad RIP value. [162513.519312] RSP: 002b:00007fff67b97758 EFLAGS: 00000246 ORIG_RAX: 0000000000000055 [162513.519314] RAX: ffffffffffffffda RBX: 00000000ffffffff RCX: 00007f5238f4a903 [162513.519316] RDX: 0000000000000000 RSI: 00000000000001b6 RDI: 000055b1fbb0d470 [162513.519317] RBP: 00007fff67b978c0 R08: 0000000000000001 R09: 0000000000000002 [162513.519319] R10: 00007fff67b974f7 R11: 0000000000000246 R12: 0000000000000013 [162513.519320] R13: 00000000000001b6 R14: 00007fff67b97906 R15: 000055b1fad1c620 [162513.519332] INFO: task btrfs:1356211 blocked for more than 120 seconds. [162513.519727] Not tainted 5.9.0-rc6-btrfs-next-69 #1 [162513.520115] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [162513.520508] task:btrfs state:D stack: 0 pid:1356211 ppid:1356178 flags:0x00004002 [162513.520511] Call Trace: [162513.520516] __schedule+0x5ce/0xd00 [162513.520519] ? _raw_spin_unlock_irqrestore+0x3c/0x60 [162513.520525] schedule+0x46/0xf0 [162513.520544] btrfs_scrub_pause+0x11f/0x180 [btrfs] [162513.520548] ? finish_wait+0x90/0x90 [162513.520562] btrfs_commit_transaction+0x45a/0xc30 [btrfs] [162513.520574] ? start_transaction+0xe0/0x5f0 [btrfs] [162513.520596] btrfs_dev_replace_finishing+0x6d8/0x711 [btrfs] [162513.520619] btrfs_dev_replace_by_ioctl.cold+0x1cc/0x1fd [btrfs] [162513.520639] btrfs_ioctl+0x2a25/0x36f0 [btrfs] [162513.520643] ? do_sigaction+0xf3/0x240 [162513.520645] ? find_held_lock+0x32/0x90 [162513.520648] ? do_sigaction+0xf3/0x240 [162513.520651] ? lock_acquired+0x33b/0x470 [162513.520655] ? _raw_spin_unlock_irq+0x24/0x50 [162513.520657] ? lockdep_hardirqs_on+0x7d/0x100 [162513.520660] ? _raw_spin_unlock_irq+0x35/0x50 [162513.520662] ? do_sigaction+0xf3/0x240 [162513.520671] ? __x64_sys_ioctl+0x83/0xb0 [162513.520672] __x64_sys_ioctl+0x83/0xb0 [162513.520677] do_syscall_64+0x33/0x80 [162513.520679] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [162513.520681] RIP: 0033:0x7fc3cd307d87 [162513.520682] Code: Bad RIP value. [162513.520684] RSP: 002b:00007ffe30a56bb8 EFLAGS: 00000202 ORIG_RAX: 0000000000000010 [162513.520686] RAX: ffffffffffffffda RBX: 0000000000000004 RCX: 00007fc3cd307d87 [162513.520687] RDX: 00007ffe30a57a30 RSI: 00000000ca289435 RDI: 0000000000000003 [162513.520689] RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000 [162513.520690] R10: 0000000000000008 R11: 0000000000000202 R12: 0000000000000003 [162513.520692] R13: 0000557323a212e0 R14: 00007ffe30a5a520 R15: 0000000000000001 [162513.520703] Showing all locks held in the system: [162513.520712] 1 lock held by khungtaskd/54: [162513.520713] #0: ffffffffb40a91a0 (rcu_read_lock){....}-{1:2}, at: debug_show_all_locks+0x15/0x197 [162513.520728] 1 lock held by in:imklog/596: [162513.520729] #0: ffff8f3f0d781400 (&f->f_pos_lock){+.+.}-{3:3}, at: __fdget_pos+0x4d/0x60 [162513.520782] 1 lock held by btrfs-transacti/1356167: [162513.520784] #0: ffff8f3d810cc848 (&fs_info->transaction_kthread_mutex){+.+.}-{3:3}, at: transaction_kthread+0x4a/0x170 [btrfs] [162513.520798] 1 lock held by btrfs/1356190: [162513.520800] #0: ffff8f3d57644470 (sb_writers#15){.+.+}-{0:0}, at: mnt_want_write_file+0x22/0x60 [162513.520805] 1 lock held by fsstress/1356184: [162513.520806] #0: ffff8f3d576440e8 (&type->s_umount_key#62){++++}-{3:3}, at: iterate_supers+0x6f/0xf0 [162513.520811] 3 locks held by fsstress/1356185: [162513.520812] #0: ffff8f3d57644470 (sb_writers#15){.+.+}-{0:0}, at: mnt_want_write+0x20/0x50 [162513.520815] #1: ffff8f3d80a650b8 (&type->i_mutex_dir_key#10){++++}-{3:3}, at: vfs_setxattr+0x50/0x120 [162513.520820] #2: ffff8f3d57644690 (sb_internal#2){.+.+}-{0:0}, at: start_transaction+0x40e/0x5f0 [btrfs] [162513.520833] 1 lock held by fsstress/1356196: [162513.520834] #0: ffff8f3d576440e8 (&type->s_umount_key#62){++++}-{3:3}, at: iterate_supers+0x6f/0xf0 [162513.520838] 3 locks held by fsstress/1356197: [162513.520839] #0: ffff8f3d57644470 (sb_writers#15){.+.+}-{0:0}, at: mnt_want_write+0x20/0x50 [162513.520843] #1: ffff8f3d506465e8 (&type->i_mutex_dir_key#10){++++}-{3:3}, at: path_openat+0x2a7/0xa50 [162513.520846] #2: ffff8f3d57644690 (sb_internal#2){.+.+}-{0:0}, at: start_transaction+0x40e/0x5f0 [btrfs] [162513.520858] 2 locks held by btrfs/1356211: [162513.520859] #0: ffff8f3d810cde30 (&fs_info->dev_replace.lock_finishing_cancel_unmount){+.+.}-{3:3}, at: btrfs_dev_replace_finishing+0x52/0x711 [btrfs] [162513.520877] #1: ffff8f3d57644690 (sb_internal#2){.+.+}-{0:0}, at: start_transaction+0x40e/0x5f0 [btrfs] This was weird because the stack traces show that a transaction commit, triggered by a device replace operation, is blocking trying to pause any running scrubs but there are no stack traces of blocked tasks doing a scrub. After poking around with drgn, I noticed there was a scrub task that was constantly running and blocking for shorts periods of time: >>> t = find_task(prog, 1356190) >>> prog.stack_trace(t) #0 __schedule+0x5ce/0xcfc #1 schedule+0x46/0xe4 #2 schedule_timeout+0x1df/0x475 #3 btrfs_reada_wait+0xda/0x132 #4 scrub_stripe+0x2a8/0x112f #5 scrub_chunk+0xcd/0x134 #6 scrub_enumerate_chunks+0x29e/0x5ee #7 btrfs_scrub_dev+0x2d5/0x91b #8 btrfs_ioctl+0x7f5/0x36e7 #9 __x64_sys_ioctl+0x83/0xb0 #10 do_syscall_64+0x33/0x77 #11 entry_SYSCALL_64+0x7c/0x156 Which corresponds to: int btrfs_reada_wait(void *handle) { struct reada_control *rc = handle; struct btrfs_fs_info *fs_info = rc->fs_info; while (atomic_read(&rc->elems)) { if (!atomic_read(&fs_info->reada_works_cnt)) reada_start_machine(fs_info); wait_event_timeout(rc->wait, atomic_read(&rc->elems) == 0, (HZ + 9) / 10); } (...) So the counter "rc->elems" was set to 1 and never decreased to 0, causing the scrub task to loop forever in that function. Then I used the following script for drgn to check the readahead requests: $ cat dump_reada.py import sys import drgn from drgn import NULL, Object, cast, container_of, execscript, \ reinterpret, sizeof from drgn.helpers.linux import * mnt_path = b"/home/fdmanana/btrfs-tests/scratch_1" mnt = None for mnt in for_each_mount(prog, dst = mnt_path): pass if mnt is None: sys.stderr.write(f'Error: mount point {mnt_path} not found\n') sys.exit(1) fs_info = cast('struct btrfs_fs_info *', mnt.mnt.mnt_sb.s_fs_info) def dump_re(re): nzones = re.nzones.value_() print(f're at {hex(re.value_())}') print(f'\t logical {re.logical.value_()}') print(f'\t refcnt {re.refcnt.value_()}') print(f'\t nzones {nzones}') for i in range(nzones): dev = re.zones[i].device name = dev.name.str.string_() print(f'\t\t dev id {dev.devid.value_()} name {name}') print() for _, e in radix_tree_for_each(fs_info.reada_tree): re = cast('struct reada_extent *', e) dump_re(re) $ drgn dump_reada.py re at 0xffff8f3da9d25ad8 logical 38928384 refcnt 1 nzones 1 dev id 0 name b'/dev/sdd' $ So there was one readahead extent with a single zone corresponding to the source device of that last device replace operation logged in dmesg/syslog. Also the ID of that zone's device was 0 which is a special value set in the source device of a device replace operation when the operation finishes (constant BTRFS_DEV_REPLACE_DEVID set at btrfs_dev_replace_finishing()), confirming again that device /dev/sdd was the source of a device replace operation. Normally there should be as many zones in the readahead extent as there are devices, and I wasn't expecting the extent to be in a block group with a 'single' profile, so I went and confirmed with the following drgn script that there weren't any single profile block groups: $ cat dump_block_groups.py import sys import drgn from drgn import NULL, Object, cast, container_of, execscript, \ reinterpret, sizeof from drgn.helpers.linux import * mnt_path = b"/home/fdmanana/btrfs-tests/scratch_1" mnt = None for mnt in for_each_mount(prog, dst = mnt_path): pass if mnt is None: sys.stderr.write(f'Error: mount point {mnt_path} not found\n') sys.exit(1) fs_info = cast('struct btrfs_fs_info *', mnt.mnt.mnt_sb.s_fs_info) BTRFS_BLOCK_GROUP_DATA = (1 << 0) BTRFS_BLOCK_GROUP_SYSTEM = (1 << 1) BTRFS_BLOCK_GROUP_METADATA = (1 << 2) BTRFS_BLOCK_GROUP_RAID0 = (1 << 3) BTRFS_BLOCK_GROUP_RAID1 = (1 << 4) BTRFS_BLOCK_GROUP_DUP = (1 << 5) BTRFS_BLOCK_GROUP_RAID10 = (1 << 6) BTRFS_BLOCK_GROUP_RAID5 = (1 << 7) BTRFS_BLOCK_GROUP_RAID6 = (1 << 8) BTRFS_BLOCK_GROUP_RAID1C3 = (1 << 9) BTRFS_BLOCK_GROUP_RAID1C4 = (1 << 10) def bg_flags_string(bg): flags = bg.flags.value_() ret = '' if flags & BTRFS_BLOCK_GROUP_DATA: ret = 'data' if flags & BTRFS_BLOCK_GROUP_METADATA: if len(ret) > 0: ret += '|' ret += 'meta' if flags & BTRFS_BLOCK_GROUP_SYSTEM: if len(ret) > 0: ret += '|' ret += 'system' if flags & BTRFS_BLOCK_GROUP_RAID0: ret += ' raid0' elif flags & BTRFS_BLOCK_GROUP_RAID1: ret += ' raid1' elif flags & BTRFS_BLOCK_GROUP_DUP: ret += ' dup' elif flags & BTRFS_BLOCK_GROUP_RAID10: ret += ' raid10' elif flags & BTRFS_BLOCK_GROUP_RAID5: ret += ' raid5' elif flags & BTRFS_BLOCK_GROUP_RAID6: ret += ' raid6' elif flags & BTRFS_BLOCK_GROUP_RAID1C3: ret += ' raid1c3' elif flags & BTRFS_BLOCK_GROUP_RAID1C4: ret += ' raid1c4' else: ret += ' single' return ret def dump_bg(bg): print() print(f'block group at {hex(bg.value_())}') print(f'\t start {bg.start.value_()} length {bg.length.value_()}') print(f'\t flags {bg.flags.value_()} - {bg_flags_string(bg)}') bg_root = fs_info.block_group_cache_tree.address_of_() for bg in rbtree_inorder_for_each_entry('struct btrfs_block_group', bg_root, 'cache_node'): dump_bg(bg) $ drgn dump_block_groups.py block group at 0xffff8f3d673b0400 start 22020096 length 16777216 flags 258 - system raid6 block group at 0xffff8f3d53ddb400 start 38797312 length 536870912 flags 260 - meta raid6 block group at 0xffff8f3d5f4d9c00 start 575668224 length 2147483648 flags 257 - data raid6 block group at 0xffff8f3d08189000 start 2723151872 length 67108864 flags 258 - system raid6 block group at 0xffff8f3db70ff000 start 2790260736 length 1073741824 flags 260 - meta raid6 block group at 0xffff8f3d5f4dd800 start 3864002560 length 67108864 flags 258 - system raid6 block group at 0xffff8f3d67037000 start 3931111424 length 2147483648 flags 257 - data raid6 $ So there were only 2 reasons left for having a readahead extent with a single zone: reada_find_zone(), called when creating a readahead extent, returned NULL either because we failed to find the corresponding block group or because a memory allocation failed. With some additional and custom tracing I figured out that on every further ocurrence of the problem the block group had just been deleted when we were looping to create the zones for the readahead extent (at reada_find_extent()), so we ended up with only one zone in the readahead extent, corresponding to a device that ends up getting replaced. So after figuring that out it became obvious why the hang happens: 1) Task A starts a scrub on any device of the filesystem, except for device /dev/sdd; 2) Task B starts a device replace with /dev/sdd as the source device; 3) Task A calls btrfs_reada_add() from scrub_stripe() and it is currently starting to scrub a stripe from block group X. This call to btrfs_reada_add() is the one for the extent tree. When btrfs_reada_add() calls reada_add_block(), it passes the logical address of the extent tree's root node as its 'logical' argument - a value of 38928384; 4) Task A then enters reada_find_extent(), called from reada_add_block(). It finds there isn't any existing readahead extent for the logical address 38928384, so it proceeds to the path of creating a new one. It calls btrfs_map_block() to find out which stripes exist for the block group X. On the first iteration of the for loop that iterates over the stripes, it finds the stripe for device /dev/sdd, so it creates one zone for that device and adds it to the readahead extent. Before getting into the second iteration of the loop, the cleanup kthread deletes block group X because it was empty. So in the iterations for the remaining stripes it does not add more zones to the readahead extent, because the calls to reada_find_zone() returned NULL because they couldn't find block group X anymore. As a result the new readahead extent has a single zone, corresponding to the device /dev/sdd; 4) Before task A returns to btrfs_reada_add() and queues the readahead job for the readahead work queue, task B finishes the device replace and at btrfs_dev_replace_finishing() swaps the device /dev/sdd with the new device /dev/sdg; 5) Task A returns to reada_add_block(), which increments the counter "->elems" of the reada_control structure allocated at btrfs_reada_add(). Then it returns back to btrfs_reada_add() and calls reada_start_machine(). This queues a job in the readahead work queue to run the function reada_start_machine_worker(), which calls __reada_start_machine(). At __reada_start_machine() we take the device list mutex and for each device found in the current device list, we call reada_start_machine_dev() to start the readahead work. However at this point the device /dev/sdd was already freed and is not in the device list anymore. This means the corresponding readahead for the extent at 38928384 is never started, and therefore the "->elems" counter of the reada_control structure allocated at btrfs_reada_add() never goes down to 0, causing the call to btrfs_reada_wait(), done by the scrub task, to wait forever. Note that the readahead request can be made either after the device replace started or before it started, however in pratice it is very unlikely that a device replace is able to start after a readahead request is made and is able to complete before the readahead request completes - maybe only on a very small and nearly empty filesystem. This hang however is not the only problem we can have with readahead and device removals. When the readahead extent has other zones other than the one corresponding to the device that is being removed (either by a device replace or a device remove operation), we risk having a use-after-free on the device when dropping the last reference of the readahead extent. For example if we create a readahead extent with two zones, one for the device /dev/sdd and one for the device /dev/sde: 1) Before the readahead worker starts, the device /dev/sdd is removed, and the corresponding btrfs_device structure is freed. However the readahead extent still has the zone pointing to the device structure; 2) When the readahead worker starts, it only finds device /dev/sde in the current device list of the filesystem; 3) It starts the readahead work, at reada_start_machine_dev(), using the device /dev/sde; 4) Then when it finishes reading the extent from device /dev/sde, it calls __readahead_hook() which ends up dropping the last reference on the readahead extent through the last call to reada_extent_put(); 5) At reada_extent_put() it iterates over each zone of the readahead extent and attempts to delete an element from the device's 'reada_extents' radix tree, resulting in a use-after-free, as the device pointer of the zone for /dev/sdd is now stale. We can also access the device after dropping the last reference of a zone, through reada_zone_release(), also called by reada_extent_put(). And a device remove suffers the same problem, however since it shrinks the device size down to zero before removing the device, it is very unlikely to still have readahead requests not completed by the time we free the device, the only possibility is if the device has a very little space allocated. While the hang problem is exclusive to scrub, since it is currently the only user of btrfs_reada_add() and btrfs_reada_wait(), the use-after-free problem affects any path that triggers readhead, which includes btree_readahead_hook() and __readahead_hook() (a readahead worker can trigger readahed for the children of a node) for example - any path that ends up calling reada_add_block() can trigger the use-after-free after a device is removed. So fix this by waiting for any readahead requests for a device to complete before removing a device, ensuring that while waiting for existing ones no new ones can be made. This problem has been around for a very long time - the readahead code was added in 2011, device remove exists since 2008 and device replace was introduced in 2013, hard to pick a specific commit for a git Fixes tag. CC: stable@vger.kernel.org # 4.4+ Reviewed-by: Josef Bacik Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/ctree.h | 2 ++ fs/btrfs/dev-replace.c | 5 +++++ fs/btrfs/reada.c | 45 +++++++++++++++++++++++++++++++++++++++++++++ fs/btrfs/volumes.c | 3 +++ fs/btrfs/volumes.h | 1 + 5 files changed, 56 insertions(+) diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index aac3d6f4e35b..0378933d163c 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -3564,6 +3564,8 @@ struct reada_control *btrfs_reada_add(struct btrfs_root *root, int btrfs_reada_wait(void *handle); void btrfs_reada_detach(void *handle); int btree_readahead_hook(struct extent_buffer *eb, int err); +void btrfs_reada_remove_dev(struct btrfs_device *dev); +void btrfs_reada_undo_remove_dev(struct btrfs_device *dev); static inline int is_fstree(u64 rootid) { diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c index 4a0243cb9d97..5b9e3f3ace22 100644 --- a/fs/btrfs/dev-replace.c +++ b/fs/btrfs/dev-replace.c @@ -688,6 +688,9 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, } btrfs_wait_ordered_roots(fs_info, U64_MAX, 0, (u64)-1); + if (!scrub_ret) + btrfs_reada_remove_dev(src_device); + /* * We have to use this loop approach because at this point src_device * has to be available for transaction commit to complete, yet new @@ -696,6 +699,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, while (1) { trans = btrfs_start_transaction(root, 0); if (IS_ERR(trans)) { + btrfs_reada_undo_remove_dev(src_device); mutex_unlock(&dev_replace->lock_finishing_cancel_unmount); return PTR_ERR(trans); } @@ -746,6 +750,7 @@ error: up_write(&dev_replace->rwsem); mutex_unlock(&fs_info->chunk_mutex); mutex_unlock(&fs_info->fs_devices->device_list_mutex); + btrfs_reada_undo_remove_dev(src_device); btrfs_rm_dev_replace_blocked(fs_info); if (tgt_device) btrfs_destroy_dev_replace_tgtdev(tgt_device); diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c index e261c3d0cec7..d9a166eb344e 100644 --- a/fs/btrfs/reada.c +++ b/fs/btrfs/reada.c @@ -421,6 +421,9 @@ static struct reada_extent *reada_find_extent(struct btrfs_fs_info *fs_info, if (!dev->bdev) continue; + if (test_bit(BTRFS_DEV_STATE_NO_READA, &dev->dev_state)) + continue; + if (dev_replace_is_ongoing && dev == fs_info->dev_replace.tgtdev) { /* @@ -1022,3 +1025,45 @@ void btrfs_reada_detach(void *handle) kref_put(&rc->refcnt, reada_control_release); } + +/* + * Before removing a device (device replace or device remove ioctls), call this + * function to wait for all existing readahead requests on the device and to + * make sure no one queues more readahead requests for the device. + * + * Must be called without holding neither the device list mutex nor the device + * replace semaphore, otherwise it will deadlock. + */ +void btrfs_reada_remove_dev(struct btrfs_device *dev) +{ + struct btrfs_fs_info *fs_info = dev->fs_info; + + /* Serialize with readahead extent creation at reada_find_extent(). */ + spin_lock(&fs_info->reada_lock); + set_bit(BTRFS_DEV_STATE_NO_READA, &dev->dev_state); + spin_unlock(&fs_info->reada_lock); + + /* + * There might be readahead requests added to the radix trees which + * were not yet added to the readahead work queue. We need to start + * them and wait for their completion, otherwise we can end up with + * use-after-free problems when dropping the last reference on the + * readahead extents and their zones, as they need to access the + * device structure. + */ + reada_start_machine(fs_info); + btrfs_flush_workqueue(fs_info->readahead_workers); +} + +/* + * If when removing a device (device replace or device remove ioctls) an error + * happens after calling btrfs_reada_remove_dev(), call this to undo what that + * function did. This is safe to call even if btrfs_reada_remove_dev() was not + * called before. + */ +void btrfs_reada_undo_remove_dev(struct btrfs_device *dev) +{ + spin_lock(&dev->fs_info->reada_lock); + clear_bit(BTRFS_DEV_STATE_NO_READA, &dev->dev_state); + spin_unlock(&dev->fs_info->reada_lock); +} diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 58b9c419a2b6..1991bc5a6f59 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -2099,6 +2099,8 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path, mutex_unlock(&uuid_mutex); ret = btrfs_shrink_device(device, 0); + if (!ret) + btrfs_reada_remove_dev(device); mutex_lock(&uuid_mutex); if (ret) goto error_undo; @@ -2179,6 +2181,7 @@ out: return ret; error_undo: + btrfs_reada_undo_remove_dev(device); if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) { mutex_lock(&fs_info->chunk_mutex); list_add(&device->dev_alloc_list, diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index bf27ac07d315..f2177263748e 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -50,6 +50,7 @@ struct btrfs_io_geometry { #define BTRFS_DEV_STATE_MISSING (2) #define BTRFS_DEV_STATE_REPLACE_TGT (3) #define BTRFS_DEV_STATE_FLUSH_SENT (4) +#define BTRFS_DEV_STATE_NO_READA (5) struct btrfs_device { struct list_head dev_list; /* device_list_mutex */ -- cgit v1.2.3 From 5223cc60b40ae525ae6c94e98824129f1a5b4ae5 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 19 Oct 2020 16:02:29 -0400 Subject: btrfs: drop the path before adding qgroup items when enabling qgroups When enabling qgroups we walk the tree_root and then add a qgroup item for every root that we have. This creates a lock dependency on the tree_root and qgroup_root, which results in the following lockdep splat (with tree locks using rwsem), eg. in tests btrfs/017 or btrfs/022: ====================================================== WARNING: possible circular locking dependency detected 5.9.0-default+ #1299 Not tainted ------------------------------------------------------ btrfs/24552 is trying to acquire lock: ffff9142dfc5f630 (btrfs-quota-00){++++}-{3:3}, at: __btrfs_tree_read_lock+0x35/0x1c0 [btrfs] but task is already holding lock: ffff9142dfc5d0b0 (btrfs-root-00){++++}-{3:3}, at: __btrfs_tree_read_lock+0x35/0x1c0 [btrfs] which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (btrfs-root-00){++++}-{3:3}: __lock_acquire+0x3fb/0x730 lock_acquire.part.0+0x6a/0x130 down_read_nested+0x46/0x130 __btrfs_tree_read_lock+0x35/0x1c0 [btrfs] __btrfs_read_lock_root_node+0x3a/0x50 [btrfs] btrfs_search_slot_get_root+0x11d/0x290 [btrfs] btrfs_search_slot+0xc3/0x9f0 [btrfs] btrfs_insert_item+0x6e/0x140 [btrfs] btrfs_create_tree+0x1cb/0x240 [btrfs] btrfs_quota_enable+0xcd/0x790 [btrfs] btrfs_ioctl_quota_ctl+0xc9/0xe0 [btrfs] __x64_sys_ioctl+0x83/0xa0 do_syscall_64+0x2d/0x70 entry_SYSCALL_64_after_hwframe+0x44/0xa9 -> #0 (btrfs-quota-00){++++}-{3:3}: check_prev_add+0x91/0xc30 validate_chain+0x491/0x750 __lock_acquire+0x3fb/0x730 lock_acquire.part.0+0x6a/0x130 down_read_nested+0x46/0x130 __btrfs_tree_read_lock+0x35/0x1c0 [btrfs] __btrfs_read_lock_root_node+0x3a/0x50 [btrfs] btrfs_search_slot_get_root+0x11d/0x290 [btrfs] btrfs_search_slot+0xc3/0x9f0 [btrfs] btrfs_insert_empty_items+0x58/0xa0 [btrfs] add_qgroup_item.part.0+0x72/0x210 [btrfs] btrfs_quota_enable+0x3bb/0x790 [btrfs] btrfs_ioctl_quota_ctl+0xc9/0xe0 [btrfs] __x64_sys_ioctl+0x83/0xa0 do_syscall_64+0x2d/0x70 entry_SYSCALL_64_after_hwframe+0x44/0xa9 other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(btrfs-root-00); lock(btrfs-quota-00); lock(btrfs-root-00); lock(btrfs-quota-00); *** DEADLOCK *** 5 locks held by btrfs/24552: #0: ffff9142df431478 (sb_writers#10){.+.+}-{0:0}, at: mnt_want_write_file+0x22/0xa0 #1: ffff9142f9b10cc0 (&fs_info->subvol_sem){++++}-{3:3}, at: btrfs_ioctl_quota_ctl+0x7b/0xe0 [btrfs] #2: ffff9142f9b11a08 (&fs_info->qgroup_ioctl_lock){+.+.}-{3:3}, at: btrfs_quota_enable+0x3b/0x790 [btrfs] #3: ffff9142df431698 (sb_internal#2){.+.+}-{0:0}, at: start_transaction+0x406/0x510 [btrfs] #4: ffff9142dfc5d0b0 (btrfs-root-00){++++}-{3:3}, at: __btrfs_tree_read_lock+0x35/0x1c0 [btrfs] stack backtrace: CPU: 1 PID: 24552 Comm: btrfs Not tainted 5.9.0-default+ #1299 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.0-59-gc9ba527-rebuilt.opensuse.org 04/01/2014 Call Trace: dump_stack+0x77/0x97 check_noncircular+0xf3/0x110 check_prev_add+0x91/0xc30 validate_chain+0x491/0x750 __lock_acquire+0x3fb/0x730 lock_acquire.part.0+0x6a/0x130 ? __btrfs_tree_read_lock+0x35/0x1c0 [btrfs] ? lock_acquire+0xc4/0x140 ? __btrfs_tree_read_lock+0x35/0x1c0 [btrfs] down_read_nested+0x46/0x130 ? __btrfs_tree_read_lock+0x35/0x1c0 [btrfs] __btrfs_tree_read_lock+0x35/0x1c0 [btrfs] ? btrfs_root_node+0xd9/0x200 [btrfs] __btrfs_read_lock_root_node+0x3a/0x50 [btrfs] btrfs_search_slot_get_root+0x11d/0x290 [btrfs] btrfs_search_slot+0xc3/0x9f0 [btrfs] btrfs_insert_empty_items+0x58/0xa0 [btrfs] add_qgroup_item.part.0+0x72/0x210 [btrfs] btrfs_quota_enable+0x3bb/0x790 [btrfs] btrfs_ioctl_quota_ctl+0xc9/0xe0 [btrfs] __x64_sys_ioctl+0x83/0xa0 do_syscall_64+0x2d/0x70 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Fix this by dropping the path whenever we find a root item, add the qgroup item, and then re-lookup the root item we found and continue processing roots. Reported-by: David Sterba Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/qgroup.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 580899bdb991..c54ea6586632 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -1026,6 +1026,10 @@ int btrfs_quota_enable(struct btrfs_fs_info *fs_info) btrfs_item_key_to_cpu(leaf, &found_key, slot); if (found_key.type == BTRFS_ROOT_REF_KEY) { + + /* Release locks on tree_root before we access quota_root */ + btrfs_release_path(path); + ret = add_qgroup_item(trans, quota_root, found_key.offset); if (ret) { @@ -1044,6 +1048,20 @@ int btrfs_quota_enable(struct btrfs_fs_info *fs_info) btrfs_abort_transaction(trans, ret); goto out_free_path; } + ret = btrfs_search_slot_for_read(tree_root, &found_key, + path, 1, 0); + if (ret < 0) { + btrfs_abort_transaction(trans, ret); + goto out_free_path; + } + if (ret > 0) { + /* + * Shouldn't happen, but in case it does we + * don't need to do the btrfs_next_item, just + * continue. + */ + continue; + } } ret = btrfs_next_item(tree_root, path); if (ret < 0) { -- cgit v1.2.3 From 49d11bead7d596e031fbd34051d8765587cd645b Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 19 Oct 2020 16:02:31 -0400 Subject: btrfs: add a helper to read the tree_root commit root for backref lookup I got the following lockdep splat with tree locks converted to rwsem patches on btrfs/104: ====================================================== WARNING: possible circular locking dependency detected 5.9.0+ #102 Not tainted ------------------------------------------------------ btrfs-cleaner/903 is trying to acquire lock: ffff8e7fab6ffe30 (btrfs-root-00){++++}-{3:3}, at: __btrfs_tree_read_lock+0x32/0x170 but task is already holding lock: ffff8e7fab628a88 (&fs_info->commit_root_sem){++++}-{3:3}, at: btrfs_find_all_roots+0x41/0x80 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #3 (&fs_info->commit_root_sem){++++}-{3:3}: down_read+0x40/0x130 caching_thread+0x53/0x5a0 btrfs_work_helper+0xfa/0x520 process_one_work+0x238/0x540 worker_thread+0x55/0x3c0 kthread+0x13a/0x150 ret_from_fork+0x1f/0x30 -> #2 (&caching_ctl->mutex){+.+.}-{3:3}: __mutex_lock+0x7e/0x7b0 btrfs_cache_block_group+0x1e0/0x510 find_free_extent+0xb6e/0x12f0 btrfs_reserve_extent+0xb3/0x1b0 btrfs_alloc_tree_block+0xb1/0x330 alloc_tree_block_no_bg_flush+0x4f/0x60 __btrfs_cow_block+0x11d/0x580 btrfs_cow_block+0x10c/0x220 commit_cowonly_roots+0x47/0x2e0 btrfs_commit_transaction+0x595/0xbd0 sync_filesystem+0x74/0x90 generic_shutdown_super+0x22/0x100 kill_anon_super+0x14/0x30 btrfs_kill_super+0x12/0x20 deactivate_locked_super+0x36/0xa0 cleanup_mnt+0x12d/0x190 task_work_run+0x5c/0xa0 exit_to_user_mode_prepare+0x1df/0x200 syscall_exit_to_user_mode+0x54/0x280 entry_SYSCALL_64_after_hwframe+0x44/0xa9 -> #1 (&space_info->groups_sem){++++}-{3:3}: down_read+0x40/0x130 find_free_extent+0x2ed/0x12f0 btrfs_reserve_extent+0xb3/0x1b0 btrfs_alloc_tree_block+0xb1/0x330 alloc_tree_block_no_bg_flush+0x4f/0x60 __btrfs_cow_block+0x11d/0x580 btrfs_cow_block+0x10c/0x220 commit_cowonly_roots+0x47/0x2e0 btrfs_commit_transaction+0x595/0xbd0 sync_filesystem+0x74/0x90 generic_shutdown_super+0x22/0x100 kill_anon_super+0x14/0x30 btrfs_kill_super+0x12/0x20 deactivate_locked_super+0x36/0xa0 cleanup_mnt+0x12d/0x190 task_work_run+0x5c/0xa0 exit_to_user_mode_prepare+0x1df/0x200 syscall_exit_to_user_mode+0x54/0x280 entry_SYSCALL_64_after_hwframe+0x44/0xa9 -> #0 (btrfs-root-00){++++}-{3:3}: __lock_acquire+0x1167/0x2150 lock_acquire+0xb9/0x3d0 down_read_nested+0x43/0x130 __btrfs_tree_read_lock+0x32/0x170 __btrfs_read_lock_root_node+0x3a/0x50 btrfs_search_slot+0x614/0x9d0 btrfs_find_root+0x35/0x1b0 btrfs_read_tree_root+0x61/0x120 btrfs_get_root_ref+0x14b/0x600 find_parent_nodes+0x3e6/0x1b30 btrfs_find_all_roots_safe+0xb4/0x130 btrfs_find_all_roots+0x60/0x80 btrfs_qgroup_trace_extent_post+0x27/0x40 btrfs_add_delayed_data_ref+0x3fd/0x460 btrfs_free_extent+0x42/0x100 __btrfs_mod_ref+0x1d7/0x2f0 walk_up_proc+0x11c/0x400 walk_up_tree+0xf0/0x180 btrfs_drop_snapshot+0x1c7/0x780 btrfs_clean_one_deleted_snapshot+0xfb/0x110 cleaner_kthread+0xd4/0x140 kthread+0x13a/0x150 ret_from_fork+0x1f/0x30 other info that might help us debug this: Chain exists of: btrfs-root-00 --> &caching_ctl->mutex --> &fs_info->commit_root_sem Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&fs_info->commit_root_sem); lock(&caching_ctl->mutex); lock(&fs_info->commit_root_sem); lock(btrfs-root-00); *** DEADLOCK *** 3 locks held by btrfs-cleaner/903: #0: ffff8e7fab628838 (&fs_info->cleaner_mutex){+.+.}-{3:3}, at: cleaner_kthread+0x6e/0x140 #1: ffff8e7faadac640 (sb_internal){.+.+}-{0:0}, at: start_transaction+0x40b/0x5c0 #2: ffff8e7fab628a88 (&fs_info->commit_root_sem){++++}-{3:3}, at: btrfs_find_all_roots+0x41/0x80 stack backtrace: CPU: 0 PID: 903 Comm: btrfs-cleaner Not tainted 5.9.0+ #102 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-2.fc32 04/01/2014 Call Trace: dump_stack+0x8b/0xb0 check_noncircular+0xcf/0xf0 __lock_acquire+0x1167/0x2150 ? __bfs+0x42/0x210 lock_acquire+0xb9/0x3d0 ? __btrfs_tree_read_lock+0x32/0x170 down_read_nested+0x43/0x130 ? __btrfs_tree_read_lock+0x32/0x170 __btrfs_tree_read_lock+0x32/0x170 __btrfs_read_lock_root_node+0x3a/0x50 btrfs_search_slot+0x614/0x9d0 ? find_held_lock+0x2b/0x80 btrfs_find_root+0x35/0x1b0 ? do_raw_spin_unlock+0x4b/0xa0 btrfs_read_tree_root+0x61/0x120 btrfs_get_root_ref+0x14b/0x600 find_parent_nodes+0x3e6/0x1b30 btrfs_find_all_roots_safe+0xb4/0x130 btrfs_find_all_roots+0x60/0x80 btrfs_qgroup_trace_extent_post+0x27/0x40 btrfs_add_delayed_data_ref+0x3fd/0x460 btrfs_free_extent+0x42/0x100 __btrfs_mod_ref+0x1d7/0x2f0 walk_up_proc+0x11c/0x400 walk_up_tree+0xf0/0x180 btrfs_drop_snapshot+0x1c7/0x780 ? btrfs_clean_one_deleted_snapshot+0x73/0x110 btrfs_clean_one_deleted_snapshot+0xfb/0x110 cleaner_kthread+0xd4/0x140 ? btrfs_alloc_root+0x50/0x50 kthread+0x13a/0x150 ? kthread_create_worker_on_cpu+0x40/0x40 ret_from_fork+0x1f/0x30 BTRFS info (device sdb): disk space caching is enabled BTRFS info (device sdb): has skinny extents This happens because qgroups does a backref lookup when we create a delayed ref. From here it may have to look up a root from an indirect ref, which does a normal lookup on the tree_root, which takes the read lock on the tree_root nodes. To fix this we need to add a variant for looking up roots that searches the commit root of the tree_root. Then when we do the backref search using the commit root we are sure to not take any locks on the tree_root nodes. This gets rid of the lockdep splat when running btrfs/104. Reviewed-by: Filipe Manana Signed-off-by: Josef Bacik Signed-off-by: David Sterba --- fs/btrfs/backref.c | 13 ++++- fs/btrfs/disk-io.c | 139 ++++++++++++++++++++++++++++++++++++++--------------- fs/btrfs/disk-io.h | 3 ++ 3 files changed, 114 insertions(+), 41 deletions(-) diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index b3268f4ea5f3..771a036867dc 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -544,7 +544,18 @@ static int resolve_indirect_ref(struct btrfs_fs_info *fs_info, int level = ref->level; struct btrfs_key search_key = ref->key_for_search; - root = btrfs_get_fs_root(fs_info, ref->root_id, false); + /* + * If we're search_commit_root we could possibly be holding locks on + * other tree nodes. This happens when qgroups does backref walks when + * adding new delayed refs. To deal with this we need to look in cache + * for the root, and if we don't find it then we need to search the + * tree_root's commit root, thus the btrfs_get_fs_root_commit_root usage + * here. + */ + if (path->search_commit_root) + root = btrfs_get_fs_root_commit_root(fs_info, path, ref->root_id); + else + root = btrfs_get_fs_root(fs_info, ref->root_id, false); if (IS_ERR(root)) { ret = PTR_ERR(root); goto out_free; diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 764001609a15..278001a49d2e 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1281,32 +1281,26 @@ int btrfs_add_log_tree(struct btrfs_trans_handle *trans, return 0; } -struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root, - struct btrfs_key *key) +static struct btrfs_root *read_tree_root_path(struct btrfs_root *tree_root, + struct btrfs_path *path, + struct btrfs_key *key) { struct btrfs_root *root; struct btrfs_fs_info *fs_info = tree_root->fs_info; - struct btrfs_path *path; u64 generation; int ret; int level; - path = btrfs_alloc_path(); - if (!path) - return ERR_PTR(-ENOMEM); - root = btrfs_alloc_root(fs_info, key->objectid, GFP_NOFS); - if (!root) { - ret = -ENOMEM; - goto alloc_fail; - } + if (!root) + return ERR_PTR(-ENOMEM); ret = btrfs_find_root(tree_root, key, path, &root->root_item, &root->root_key); if (ret) { if (ret > 0) ret = -ENOENT; - goto find_fail; + goto fail; } generation = btrfs_root_generation(&root->root_item); @@ -1317,21 +1311,31 @@ struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root, if (IS_ERR(root->node)) { ret = PTR_ERR(root->node); root->node = NULL; - goto find_fail; + goto fail; } else if (!btrfs_buffer_uptodate(root->node, generation, 0)) { ret = -EIO; - goto find_fail; + goto fail; } root->commit_root = btrfs_root_node(root); -out: - btrfs_free_path(path); return root; - -find_fail: +fail: btrfs_put_root(root); -alloc_fail: - root = ERR_PTR(ret); - goto out; + return ERR_PTR(ret); +} + +struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root, + struct btrfs_key *key) +{ + struct btrfs_root *root; + struct btrfs_path *path; + + path = btrfs_alloc_path(); + if (!path) + return ERR_PTR(-ENOMEM); + root = read_tree_root_path(tree_root, path, key); + btrfs_free_path(path); + + return root; } /* @@ -1419,6 +1423,31 @@ static struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info, return root; } +static struct btrfs_root *btrfs_get_global_root(struct btrfs_fs_info *fs_info, + u64 objectid) +{ + if (objectid == BTRFS_ROOT_TREE_OBJECTID) + return btrfs_grab_root(fs_info->tree_root); + if (objectid == BTRFS_EXTENT_TREE_OBJECTID) + return btrfs_grab_root(fs_info->extent_root); + if (objectid == BTRFS_CHUNK_TREE_OBJECTID) + return btrfs_grab_root(fs_info->chunk_root); + if (objectid == BTRFS_DEV_TREE_OBJECTID) + return btrfs_grab_root(fs_info->dev_root); + if (objectid == BTRFS_CSUM_TREE_OBJECTID) + return btrfs_grab_root(fs_info->csum_root); + if (objectid == BTRFS_QUOTA_TREE_OBJECTID) + return btrfs_grab_root(fs_info->quota_root) ? + fs_info->quota_root : ERR_PTR(-ENOENT); + if (objectid == BTRFS_UUID_TREE_OBJECTID) + return btrfs_grab_root(fs_info->uuid_root) ? + fs_info->uuid_root : ERR_PTR(-ENOENT); + if (objectid == BTRFS_FREE_SPACE_TREE_OBJECTID) + return btrfs_grab_root(fs_info->free_space_root) ? + fs_info->free_space_root : ERR_PTR(-ENOENT); + return NULL; +} + int btrfs_insert_fs_root(struct btrfs_fs_info *fs_info, struct btrfs_root *root) { @@ -1518,25 +1547,9 @@ static struct btrfs_root *btrfs_get_root_ref(struct btrfs_fs_info *fs_info, struct btrfs_key key; int ret; - if (objectid == BTRFS_ROOT_TREE_OBJECTID) - return btrfs_grab_root(fs_info->tree_root); - if (objectid == BTRFS_EXTENT_TREE_OBJECTID) - return btrfs_grab_root(fs_info->extent_root); - if (objectid == BTRFS_CHUNK_TREE_OBJECTID) - return btrfs_grab_root(fs_info->chunk_root); - if (objectid == BTRFS_DEV_TREE_OBJECTID) - return btrfs_grab_root(fs_info->dev_root); - if (objectid == BTRFS_CSUM_TREE_OBJECTID) - return btrfs_grab_root(fs_info->csum_root); - if (objectid == BTRFS_QUOTA_TREE_OBJECTID) - return btrfs_grab_root(fs_info->quota_root) ? - fs_info->quota_root : ERR_PTR(-ENOENT); - if (objectid == BTRFS_UUID_TREE_OBJECTID) - return btrfs_grab_root(fs_info->uuid_root) ? - fs_info->uuid_root : ERR_PTR(-ENOENT); - if (objectid == BTRFS_FREE_SPACE_TREE_OBJECTID) - return btrfs_grab_root(fs_info->free_space_root) ? - fs_info->free_space_root : ERR_PTR(-ENOENT); + root = btrfs_get_global_root(fs_info, objectid); + if (root) + return root; again: root = btrfs_lookup_fs_root(fs_info, objectid); if (root) { @@ -1621,6 +1634,52 @@ struct btrfs_root *btrfs_get_new_fs_root(struct btrfs_fs_info *fs_info, return btrfs_get_root_ref(fs_info, objectid, anon_dev, true); } +/* + * btrfs_get_fs_root_commit_root - return a root for the given objectid + * @fs_info: the fs_info + * @objectid: the objectid we need to lookup + * + * This is exclusively used for backref walking, and exists specifically because + * of how qgroups does lookups. Qgroups will do a backref lookup at delayed ref + * creation time, which means we may have to read the tree_root in order to look + * up a fs root that is not in memory. If the root is not in memory we will + * read the tree root commit root and look up the fs root from there. This is a + * temporary root, it will not be inserted into the radix tree as it doesn't + * have the most uptodate information, it'll simply be discarded once the + * backref code is finished using the root. + */ +struct btrfs_root *btrfs_get_fs_root_commit_root(struct btrfs_fs_info *fs_info, + struct btrfs_path *path, + u64 objectid) +{ + struct btrfs_root *root; + struct btrfs_key key; + + ASSERT(path->search_commit_root && path->skip_locking); + + /* + * This can return -ENOENT if we ask for a root that doesn't exist, but + * since this is called via the backref walking code we won't be looking + * up a root that doesn't exist, unless there's corruption. So if root + * != NULL just return it. + */ + root = btrfs_get_global_root(fs_info, objectid); + if (root) + return root; + + root = btrfs_lookup_fs_root(fs_info, objectid); + if (root) + return root; + + key.objectid = objectid; + key.type = BTRFS_ROOT_ITEM_KEY; + key.offset = (u64)-1; + root = read_tree_root_path(fs_info->tree_root, path, &key); + btrfs_release_path(path); + + return root; +} + /* * called by the kthread helper functions to finally call the bio end_io * functions. This is where read checksum verification actually happens diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h index fee69ced58b4..182540bdcea0 100644 --- a/fs/btrfs/disk-io.h +++ b/fs/btrfs/disk-io.h @@ -69,6 +69,9 @@ struct btrfs_root *btrfs_get_fs_root(struct btrfs_fs_info *fs_info, u64 objectid, bool check_ref); struct btrfs_root *btrfs_get_new_fs_root(struct btrfs_fs_info *fs_info, u64 objectid, dev_t anon_dev); +struct btrfs_root *btrfs_get_fs_root_commit_root(struct btrfs_fs_info *fs_info, + struct btrfs_path *path, + u64 objectid); void btrfs_free_fs_info(struct btrfs_fs_info *fs_info); int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info); -- cgit v1.2.3 From 9480b4e75b7108ee68ecf5bc6b4bd68e8031c521 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Mon, 26 Oct 2020 09:12:10 +0000 Subject: cachefiles: Handle readpage error correctly If ->readpage returns an error, it has already unlocked the page. Fixes: 5e929b33c393 ("CacheFiles: Handle truncate unlocking the page we're reading") Cc: stable@vger.kernel.org Signed-off-by: Matthew Wilcox (Oracle) Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- fs/cachefiles/rdwr.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/cachefiles/rdwr.c b/fs/cachefiles/rdwr.c index 3080cda9e824..8bda092e60c5 100644 --- a/fs/cachefiles/rdwr.c +++ b/fs/cachefiles/rdwr.c @@ -121,7 +121,7 @@ static int cachefiles_read_reissue(struct cachefiles_object *object, _debug("reissue read"); ret = bmapping->a_ops->readpage(NULL, backpage); if (ret < 0) - goto unlock_discard; + goto discard; } /* but the page may have been read before the monitor was installed, so @@ -138,6 +138,7 @@ static int cachefiles_read_reissue(struct cachefiles_object *object, unlock_discard: unlock_page(backpage); +discard: spin_lock_irq(&object->work_lock); list_del(&monitor->op_link); spin_unlock_irq(&object->work_lock); -- cgit v1.2.3 From 831e3405c2a344018a18fcc2665acc5a38c3a707 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Sat, 10 Oct 2020 11:25:39 +0800 Subject: scsi: core: Don't start concurrent async scan on same host The current scanning mechanism is supposed to fall back to a synchronous host scan if an asynchronous scan is in progress. However, this rule isn't strictly respected, scsi_prep_async_scan() doesn't hold scan_mutex when checking shost->async_scan. When scsi_scan_host() is called concurrently, two async scans on same host can be started and a hang in do_scan_async() is observed. Fixes this issue by checking & setting shost->async_scan atomically with shost->scan_mutex. Link: https://lore.kernel.org/r/20201010032539.426615-1-ming.lei@redhat.com Cc: Christoph Hellwig Cc: Ewan D. Milne Cc: Hannes Reinecke Cc: Bart Van Assche Reviewed-by: Lee Duncan Reviewed-by: Bart Van Assche Signed-off-by: Ming Lei Signed-off-by: Martin K. Petersen --- drivers/scsi/scsi_scan.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index f2437a7570ce..9af50e6f94c4 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -1714,15 +1714,16 @@ static void scsi_sysfs_add_devices(struct Scsi_Host *shost) */ static struct async_scan_data *scsi_prep_async_scan(struct Scsi_Host *shost) { - struct async_scan_data *data; + struct async_scan_data *data = NULL; unsigned long flags; if (strncmp(scsi_scan_type, "sync", 4) == 0) return NULL; + mutex_lock(&shost->scan_mutex); if (shost->async_scan) { shost_printk(KERN_DEBUG, shost, "%s called twice\n", __func__); - return NULL; + goto err; } data = kmalloc(sizeof(*data), GFP_KERNEL); @@ -1733,7 +1734,6 @@ static struct async_scan_data *scsi_prep_async_scan(struct Scsi_Host *shost) goto err; init_completion(&data->prev_finished); - mutex_lock(&shost->scan_mutex); spin_lock_irqsave(shost->host_lock, flags); shost->async_scan = 1; spin_unlock_irqrestore(shost->host_lock, flags); @@ -1748,6 +1748,7 @@ static struct async_scan_data *scsi_prep_async_scan(struct Scsi_Host *shost) return data; err: + mutex_unlock(&shost->scan_mutex); kfree(data); return NULL; } -- cgit v1.2.3 From 98f9d33c5d0d779e6c96a4a8ac995085b11924f0 Mon Sep 17 00:00:00 2001 From: Daniel Wagner Date: Wed, 14 Oct 2020 09:30:48 +0200 Subject: scsi: qla2xxx: Return EBUSY on fcport deletion When the fcport is about to be deleted we should return EBUSY instead of ENODEV. Only for EBUSY will the request be requeued in a multipath setup. Also return EBUSY when the firmware has not yet started to avoid dropping the request. Link: https://lore.kernel.org/r/20201014073048.36219-1-dwagner@suse.de Reviewed-by: Arun Easi Signed-off-by: Daniel Wagner Signed-off-by: Martin K. Petersen --- drivers/scsi/qla2xxx/qla_nvme.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/qla2xxx/qla_nvme.c b/drivers/scsi/qla2xxx/qla_nvme.c index 1f9005125313..b7a1dc24db38 100644 --- a/drivers/scsi/qla2xxx/qla_nvme.c +++ b/drivers/scsi/qla2xxx/qla_nvme.c @@ -554,10 +554,12 @@ static int qla_nvme_post_cmd(struct nvme_fc_local_port *lport, fcport = qla_rport->fcport; - if (!qpair || !fcport || (qpair && !qpair->fw_started) || - (fcport && fcport->deleted)) + if (!qpair || !fcport) return -ENODEV; + if (!qpair->fw_started || fcport->deleted) + return -EBUSY; + vha = fcport->vha; if (!(fcport->nvme_flag & NVME_FLAG_REGISTERED)) -- cgit v1.2.3 From 2f4843b172c2c0360ee7792ad98025fae7baefde Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Thu, 22 Oct 2020 11:00:05 +0200 Subject: scsi: mptfusion: Fix null pointer dereferences in mptscsih_remove() The mptscsih_remove() function triggers a kernel oops if the Scsi_Host pointer (ioc->sh) is NULL, as can be seen in this syslog: ioc0: LSI53C1030 B2: Capabilities={Initiator,Target} Begin: Waiting for root file system ... scsi host2: error handler thread failed to spawn, error = -4 mptspi: ioc0: WARNING - Unable to register controller with SCSI subsystem Backtrace: [<000000001045b7cc>] mptspi_probe+0x248/0x3d0 [mptspi] [<0000000040946470>] pci_device_probe+0x1ac/0x2d8 [<0000000040add668>] really_probe+0x1bc/0x988 [<0000000040ade704>] driver_probe_device+0x160/0x218 [<0000000040adee24>] device_driver_attach+0x160/0x188 [<0000000040adef90>] __driver_attach+0x144/0x320 [<0000000040ad7c78>] bus_for_each_dev+0xd4/0x158 [<0000000040adc138>] driver_attach+0x4c/0x80 [<0000000040adb3ec>] bus_add_driver+0x3e0/0x498 [<0000000040ae0130>] driver_register+0xf4/0x298 [<00000000409450c4>] __pci_register_driver+0x78/0xa8 [<000000000007d248>] mptspi_init+0x18c/0x1c4 [mptspi] This patch adds the necessary NULL-pointer checks. Successfully tested on a HP C8000 parisc workstation with buggy SCSI drives. Link: https://lore.kernel.org/r/20201022090005.GA9000@ls3530.fritz.box Cc: Signed-off-by: Helge Deller Signed-off-by: Martin K. Petersen --- drivers/message/fusion/mptscsih.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/message/fusion/mptscsih.c b/drivers/message/fusion/mptscsih.c index a5ef9faf71c7..e7f0d4ae0f96 100644 --- a/drivers/message/fusion/mptscsih.c +++ b/drivers/message/fusion/mptscsih.c @@ -1176,8 +1176,10 @@ mptscsih_remove(struct pci_dev *pdev) MPT_SCSI_HOST *hd; int sz1; - if((hd = shost_priv(host)) == NULL) - return; + if (host == NULL) + hd = NULL; + else + hd = shost_priv(host); mptscsih_shutdown(pdev); @@ -1193,14 +1195,15 @@ mptscsih_remove(struct pci_dev *pdev) "Free'd ScsiLookup (%d) memory\n", ioc->name, sz1)); - kfree(hd->info_kbuf); + if (hd) + kfree(hd->info_kbuf); /* NULL the Scsi_Host pointer */ ioc->sh = NULL; - scsi_host_put(host); - + if (host) + scsi_host_put(host); mpt_detach(pdev); } -- cgit v1.2.3 From a305e7dc5fa86ff9cf6cd2da30215a92d43c9285 Mon Sep 17 00:00:00 2001 From: "Tianci.Yin" Date: Thu, 22 Oct 2020 11:40:26 +0800 Subject: drm/amdgpu: disable DCN and VCN for navi10 blockchain SKU(v3) The blockchain SKU has no display and video support, remove them. Reviewed-by: Alex Deucher Signed-off-by: Tianci.Yin Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/nv.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c index 1ce741a0c6a7..03462c857498 100644 --- a/drivers/gpu/drm/amd/amdgpu/nv.c +++ b/drivers/gpu/drm/amd/amdgpu/nv.c @@ -455,6 +455,14 @@ void nv_set_virt_ops(struct amdgpu_device *adev) adev->virt.ops = &xgpu_nv_virt_ops; } +static bool nv_is_blockchain_sku(struct pci_dev *pdev) +{ + if (pdev->device == 0x731E && + (pdev->revision == 0xC6 || pdev->revision == 0xC7)) + return true; + return false; +} + int nv_set_ip_blocks(struct amdgpu_device *adev) { int r; @@ -483,7 +491,8 @@ int nv_set_ip_blocks(struct amdgpu_device *adev) if (adev->enable_virtual_display || amdgpu_sriov_vf(adev)) amdgpu_device_ip_block_add(adev, &dce_virtual_ip_block); #if defined(CONFIG_DRM_AMD_DC) - else if (amdgpu_device_has_dc_support(adev)) + else if (amdgpu_device_has_dc_support(adev) && + !nv_is_blockchain_sku(adev->pdev)) amdgpu_device_ip_block_add(adev, &dm_ip_block); #endif amdgpu_device_ip_block_add(adev, &gfx_v10_0_ip_block); @@ -491,7 +500,8 @@ int nv_set_ip_blocks(struct amdgpu_device *adev) if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT && !amdgpu_sriov_vf(adev)) amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block); - amdgpu_device_ip_block_add(adev, &vcn_v2_0_ip_block); + if (!nv_is_blockchain_sku(adev->pdev)) + amdgpu_device_ip_block_add(adev, &vcn_v2_0_ip_block); amdgpu_device_ip_block_add(adev, &jpeg_v2_0_ip_block); if (adev->enable_mes) amdgpu_device_ip_block_add(adev, &mes_v10_1_ip_block); -- cgit v1.2.3 From 8942881144a7365143f196f5eafed24783a424a3 Mon Sep 17 00:00:00 2001 From: "Tianci.Yin" Date: Wed, 14 Oct 2020 17:05:50 +0800 Subject: drm/amdgpu: add DID for navi10 blockchain SKU Reviewed-by: Alex Deucher Reviewed-by: Guchun Chen Signed-off-by: Tianci.Yin Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index c241317edee7..42d9748921f5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -1066,6 +1066,7 @@ static const struct pci_device_id pciidlist[] = { {0x1002, 0x7319, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_NAVI10}, {0x1002, 0x731A, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_NAVI10}, {0x1002, 0x731B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_NAVI10}, + {0x1002, 0x731E, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_NAVI10}, {0x1002, 0x731F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_NAVI10}, /* Navi14 */ {0x1002, 0x7340, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_NAVI14}, -- cgit v1.2.3 From c4aa8dff6091cc9536aeb255e544b0b4ba29faf4 Mon Sep 17 00:00:00 2001 From: Madhav Chauhan Date: Fri, 16 Oct 2020 18:03:07 +0530 Subject: drm/amdgpu: don't map BO in reserved region MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 2MB area is reserved at top inside VM. Suggested-by: Christian König Signed-off-by: Madhav Chauhan Reviewed-by: Christian König Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index aa7f230c71bf..7e8265da9f25 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -596,6 +596,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, struct ww_acquire_ctx ticket; struct list_head list, duplicates; uint64_t va_flags; + uint64_t vm_size; int r = 0; if (args->va_address < AMDGPU_VA_RESERVED_SIZE) { @@ -616,6 +617,15 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, args->va_address &= AMDGPU_GMC_HOLE_MASK; + vm_size = adev->vm_manager.max_pfn * AMDGPU_GPU_PAGE_SIZE; + vm_size -= AMDGPU_VA_RESERVED_SIZE; + if (args->va_address + args->map_size > vm_size) { + dev_dbg(&dev->pdev->dev, + "va_address 0x%llx is in top reserved area 0x%llx\n", + args->va_address + args->map_size, vm_size); + return -EINVAL; + } + if ((args->flags & ~valid_flags) && (args->flags & ~prt_flags)) { dev_dbg(&dev->pdev->dev, "invalid flags combination 0x%08X\n", args->flags); -- cgit v1.2.3 From 866e09f0110c6e86071954033e3067975946592a Mon Sep 17 00:00:00 2001 From: David Galiffi Date: Wed, 29 Apr 2020 13:31:12 -0400 Subject: drm/amd/display: Fixed panic during seamless boot. [why] get_pixel_clk_frequency_100hz is undefined in clock_source_funcs. [how] set function pointer: ".get_pixel_clk_frequency_100hz = get_pixel_clk_frequency_100hz" Signed-off-by: David Galiffi Reviewed-by: Bhawanpreet Lakha Acked-by: Aurabindo Pillai Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c index 9cc65dc1970f..49ae5ff12da6 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clock_source.c @@ -1149,7 +1149,8 @@ static uint32_t dcn3_get_pix_clk_dividers( static const struct clock_source_funcs dcn3_clk_src_funcs = { .cs_power_down = dce110_clock_source_power_down, .program_pix_clk = dcn3_program_pix_clk, - .get_pix_clk_dividers = dcn3_get_pix_clk_dividers + .get_pix_clk_dividers = dcn3_get_pix_clk_dividers, + .get_pixel_clk_frequency_100hz = get_pixel_clk_frequency_100hz }; #endif /*****************************************/ -- cgit v1.2.3 From 55bb919be4e4973cd037a04f527ecc6686800437 Mon Sep 17 00:00:00 2001 From: Christian König Date: Mon, 12 Oct 2020 13:09:36 +0200 Subject: drm/amdgpu: increase the reserved VM size to 2MB MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Ideally this should be a multiple of the VM block size. 2MB should at least fit for Vega/Navi. Signed-off-by: Christian König Reviewed-by: Madhav Chauhan Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index c6abb16e8018..58c83a7ad0fd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -112,8 +112,8 @@ struct amdgpu_bo_list_entry; #define AMDGPU_MMHUB_0 1 #define AMDGPU_MMHUB_1 2 -/* hardcode that limit for now */ -#define AMDGPU_VA_RESERVED_SIZE (1ULL << 20) +/* Reserve 2MB at top/bottom of address space for kernel use */ +#define AMDGPU_VA_RESERVED_SIZE (2ULL << 20) /* max vmids dedicated for process */ #define AMDGPU_VM_MAX_RESERVED_VMID 1 -- cgit v1.2.3 From 140b2ef1c28d3f5a5fc767368eaa8b45afc5bf1f Mon Sep 17 00:00:00 2001 From: Dmytro Laktyushkin Date: Thu, 15 Oct 2020 14:49:56 -0400 Subject: drm/amd/display: prevent null pointer access Prevent null pointer access when checking odm tree. Signed-off-by: Dmytro Laktyushkin Acked-by: Aurabindo Pillai Signed-off-by: Alex Deucher Cc: --- drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c index 24fb39a11e5d..2455d210ccf6 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c @@ -2105,12 +2105,12 @@ static bool dcn30_internal_validate_bw( if (split[i]) { if (odm) { - if (split[i] == 4 && old_pipe->next_odm_pipe->next_odm_pipe) + if (split[i] == 4 && old_pipe->next_odm_pipe && old_pipe->next_odm_pipe->next_odm_pipe) old_index = old_pipe->next_odm_pipe->next_odm_pipe->pipe_idx; else if (old_pipe->next_odm_pipe) old_index = old_pipe->next_odm_pipe->pipe_idx; } else { - if (split[i] == 4 && old_pipe->bottom_pipe->bottom_pipe && + if (split[i] == 4 && old_pipe->bottom_pipe && old_pipe->bottom_pipe->bottom_pipe && old_pipe->bottom_pipe->bottom_pipe->plane_state == old_pipe->plane_state) old_index = old_pipe->bottom_pipe->bottom_pipe->pipe_idx; else if (old_pipe->bottom_pipe && @@ -2150,10 +2150,12 @@ static bool dcn30_internal_validate_bw( goto validate_fail; newly_split[pipe_4to1->pipe_idx] = true; - if (odm && old_pipe->next_odm_pipe->next_odm_pipe->next_odm_pipe) + if (odm && old_pipe->next_odm_pipe && old_pipe->next_odm_pipe->next_odm_pipe + && old_pipe->next_odm_pipe->next_odm_pipe->next_odm_pipe) old_index = old_pipe->next_odm_pipe->next_odm_pipe->next_odm_pipe->pipe_idx; - else if (!odm && old_pipe->bottom_pipe->bottom_pipe->bottom_pipe && - old_pipe->bottom_pipe->bottom_pipe->bottom_pipe->plane_state == old_pipe->plane_state) + else if (!odm && old_pipe->bottom_pipe && old_pipe->bottom_pipe->bottom_pipe && + old_pipe->bottom_pipe->bottom_pipe->bottom_pipe && + old_pipe->bottom_pipe->bottom_pipe->bottom_pipe->plane_state == old_pipe->plane_state) old_index = old_pipe->bottom_pipe->bottom_pipe->bottom_pipe->pipe_idx; else old_index = -1; -- cgit v1.2.3 From 1acd4577a66f5125ede038ee16c33d05ddcaf153 Mon Sep 17 00:00:00 2001 From: Ricardo Cañuelo Date: Wed, 21 Oct 2020 13:43:06 +0200 Subject: dt-bindings: i2c: convert i2c-cros-ec-tunnel to json-schema MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Convert the google,cros-ec-i2c-tunnel binding to YAML and add it as a property of google,cros-ec.yaml. Reviewed-by: Enric Balletbo i Serra Signed-off-by: Ricardo Cañuelo Link: https://lore.kernel.org/r/20201021114308.25485-2-ricardo.canuelo@collabora.com [robh: add ref to i2c-controller.yaml] Signed-off-by: Rob Herring --- .../bindings/i2c/google,cros-ec-i2c-tunnel.yaml | 66 ++++++++++++++++++++++ .../devicetree/bindings/i2c/i2c-cros-ec-tunnel.txt | 39 ------------- .../devicetree/bindings/mfd/google,cros-ec.yaml | 5 ++ 3 files changed, 71 insertions(+), 39 deletions(-) create mode 100644 Documentation/devicetree/bindings/i2c/google,cros-ec-i2c-tunnel.yaml delete mode 100644 Documentation/devicetree/bindings/i2c/i2c-cros-ec-tunnel.txt diff --git a/Documentation/devicetree/bindings/i2c/google,cros-ec-i2c-tunnel.yaml b/Documentation/devicetree/bindings/i2c/google,cros-ec-i2c-tunnel.yaml new file mode 100644 index 000000000000..b386e4128a79 --- /dev/null +++ b/Documentation/devicetree/bindings/i2c/google,cros-ec-i2c-tunnel.yaml @@ -0,0 +1,66 @@ +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause) +%YAML 1.2 +--- + +$id: http://devicetree.org/schemas/i2c/google,cros-ec-i2c-tunnel.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: I2C bus that tunnels through the ChromeOS EC (cros-ec) + +maintainers: + - Doug Anderson + - Benson Leung + - Enric Balletbo i Serra + +description: | + On some ChromeOS board designs we've got a connection to the EC + (embedded controller) but no direct connection to some devices on the + other side of the EC (like a battery and PMIC). To get access to + those devices we need to tunnel our i2c commands through the EC. + + The node for this device should be under a cros-ec node like + google,cros-ec-spi or google,cros-ec-i2c. + +allOf: + - $ref: i2c-controller.yaml# + +properties: + compatible: + const: google,cros-ec-i2c-tunnel + + google,remote-bus: + description: The EC bus we'd like to talk to. + $ref: /schemas/types.yaml#/definitions/uint32 + +required: + - compatible + - google,remote-bus + +unevaluatedProperties: false + +examples: + - | + spi0 { + #address-cells = <1>; + #size-cells = <0>; + + cros-ec@0 { + compatible = "google,cros-ec-spi"; + reg = <0>; + spi-max-frequency = <5000000>; + + i2c-tunnel { + compatible = "google,cros-ec-i2c-tunnel"; + #address-cells = <1>; + #size-cells = <0>; + + google,remote-bus = <0>; + + battery: sbs-battery@b { + compatible = "sbs,sbs-battery"; + reg = <0xb>; + sbs,poll-retry-count = <1>; + }; + }; + }; + }; diff --git a/Documentation/devicetree/bindings/i2c/i2c-cros-ec-tunnel.txt b/Documentation/devicetree/bindings/i2c/i2c-cros-ec-tunnel.txt deleted file mode 100644 index 898f030eba62..000000000000 --- a/Documentation/devicetree/bindings/i2c/i2c-cros-ec-tunnel.txt +++ /dev/null @@ -1,39 +0,0 @@ -I2C bus that tunnels through the ChromeOS EC (cros-ec) -====================================================== -On some ChromeOS board designs we've got a connection to the EC (embedded -controller) but no direct connection to some devices on the other side of -the EC (like a battery and PMIC). To get access to those devices we need -to tunnel our i2c commands through the EC. - -The node for this device should be under a cros-ec node like google,cros-ec-spi -or google,cros-ec-i2c. - - -Required properties: -- compatible: google,cros-ec-i2c-tunnel -- google,remote-bus: The EC bus we'd like to talk to. - -Optional child nodes: -- One node per I2C device connected to the tunnelled I2C bus. - - -Example: - cros-ec@0 { - compatible = "google,cros-ec-spi"; - - ... - - i2c-tunnel { - compatible = "google,cros-ec-i2c-tunnel"; - #address-cells = <1>; - #size-cells = <0>; - - google,remote-bus = <0>; - - battery: sbs-battery@b { - compatible = "sbs,sbs-battery"; - reg = <0xb>; - sbs,poll-retry-count = <1>; - }; - }; - } diff --git a/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml b/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml index f49c0d5d31ad..c45cf30ea3aa 100644 --- a/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml +++ b/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml @@ -71,6 +71,11 @@ properties: wakeup-source: description: Button can wake-up the system. +patternProperties: + "^i2c-tunnel[0-9]*$": + type: object + $ref: "/schemas/i2c/google,cros-ec-i2c-tunnel.yaml#" + required: - compatible -- cgit v1.2.3 From 50d68feee01039e74e296c3d8453c0d72b546307 Mon Sep 17 00:00:00 2001 From: Ricardo Cañuelo Date: Wed, 21 Oct 2020 13:43:07 +0200 Subject: dt-bindings: input: convert cros-ec-keyb to json-schema MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Convert the google,cros-ec-keyb binding to YAML and add it as a property of google,cros-ec.yaml Reviewed-by: Enric Balletbo i Serra Signed-off-by: Ricardo Cañuelo Link: https://lore.kernel.org/r/20201021114308.25485-3-ricardo.canuelo@collabora.com Signed-off-by: Rob Herring --- .../devicetree/bindings/input/cros-ec-keyb.txt | 72 ----------------- .../bindings/input/google,cros-ec-keyb.yaml | 92 ++++++++++++++++++++++ .../devicetree/bindings/mfd/google,cros-ec.yaml | 3 + 3 files changed, 95 insertions(+), 72 deletions(-) delete mode 100644 Documentation/devicetree/bindings/input/cros-ec-keyb.txt create mode 100644 Documentation/devicetree/bindings/input/google,cros-ec-keyb.yaml diff --git a/Documentation/devicetree/bindings/input/cros-ec-keyb.txt b/Documentation/devicetree/bindings/input/cros-ec-keyb.txt deleted file mode 100644 index 0f6355ce39b5..000000000000 --- a/Documentation/devicetree/bindings/input/cros-ec-keyb.txt +++ /dev/null @@ -1,72 +0,0 @@ -ChromeOS EC Keyboard - -Google's ChromeOS EC Keyboard is a simple matrix keyboard implemented on -a separate EC (Embedded Controller) device. It provides a message for reading -key scans from the EC. These are then converted into keycodes for processing -by the kernel. - -This binding is based on matrix-keymap.txt and extends/modifies it as follows: - -Required properties: -- compatible: "google,cros-ec-keyb" - -Optional properties: -- google,needs-ghost-filter: True to enable a ghost filter for the matrix -keyboard. This is recommended if the EC does not have its own logic or -hardware for this. - - -Example: - -cros-ec-keyb { - compatible = "google,cros-ec-keyb"; - keypad,num-rows = <8>; - keypad,num-columns = <13>; - google,needs-ghost-filter; - /* - * Keymap entries take the form of 0xRRCCKKKK where - * RR=Row CC=Column KKKK=Key Code - * The values below are for a US keyboard layout and - * are taken from the Linux driver. Note that the - * 102ND key is not used for US keyboards. - */ - linux,keymap = < - /* CAPSLCK F1 B F10 */ - 0x0001003a 0x0002003b 0x00030030 0x00040044 - /* N = R_ALT ESC */ - 0x00060031 0x0008000d 0x000a0064 0x01010001 - /* F4 G F7 H */ - 0x0102003e 0x01030022 0x01040041 0x01060023 - /* ' F9 BKSPACE L_CTRL */ - 0x01080028 0x01090043 0x010b000e 0x0200001d - /* TAB F3 T F6 */ - 0x0201000f 0x0202003d 0x02030014 0x02040040 - /* ] Y 102ND [ */ - 0x0205001b 0x02060015 0x02070056 0x0208001a - /* F8 GRAVE F2 5 */ - 0x02090042 0x03010029 0x0302003c 0x03030006 - /* F5 6 - \ */ - 0x0304003f 0x03060007 0x0308000c 0x030b002b - /* R_CTRL A D F */ - 0x04000061 0x0401001e 0x04020020 0x04030021 - /* S K J ; */ - 0x0404001f 0x04050025 0x04060024 0x04080027 - /* L ENTER Z C */ - 0x04090026 0x040b001c 0x0501002c 0x0502002e - /* V X , M */ - 0x0503002f 0x0504002d 0x05050033 0x05060032 - /* L_SHIFT / . SPACE */ - 0x0507002a 0x05080035 0x05090034 0x050B0039 - /* 1 3 4 2 */ - 0x06010002 0x06020004 0x06030005 0x06040003 - /* 8 7 0 9 */ - 0x06050009 0x06060008 0x0608000b 0x0609000a - /* L_ALT DOWN RIGHT Q */ - 0x060a0038 0x060b006c 0x060c006a 0x07010010 - /* E R W I */ - 0x07020012 0x07030013 0x07040011 0x07050017 - /* U R_SHIFT P O */ - 0x07060016 0x07070036 0x07080019 0x07090018 - /* UP LEFT */ - 0x070b0067 0x070c0069>; -}; diff --git a/Documentation/devicetree/bindings/input/google,cros-ec-keyb.yaml b/Documentation/devicetree/bindings/input/google,cros-ec-keyb.yaml new file mode 100644 index 000000000000..8e50c14a9d77 --- /dev/null +++ b/Documentation/devicetree/bindings/input/google,cros-ec-keyb.yaml @@ -0,0 +1,92 @@ +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause) +%YAML 1.2 +--- + +$id: http://devicetree.org/schemas/input/google,cros-ec-keyb.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: ChromeOS EC Keyboard + +maintainers: + - Simon Glass + - Benson Leung + - Enric Balletbo i Serra + +description: | + Google's ChromeOS EC Keyboard is a simple matrix keyboard + implemented on a separate EC (Embedded Controller) device. It provides + a message for reading key scans from the EC. These are then converted + into keycodes for processing by the kernel. + +allOf: + - $ref: "/schemas/input/matrix-keymap.yaml#" + +properties: + compatible: + const: google,cros-ec-keyb + + google,needs-ghost-filter: + description: + Enable a ghost filter for the matrix keyboard. This is recommended + if the EC does not have its own logic or hardware for this. + type: boolean + +required: + - compatible + +unevaluatedProperties: false + +examples: + - | + cros-ec-keyb { + compatible = "google,cros-ec-keyb"; + keypad,num-rows = <8>; + keypad,num-columns = <13>; + google,needs-ghost-filter; + /* + * Keymap entries take the form of 0xRRCCKKKK where + * RR=Row CC=Column KKKK=Key Code + * The values below are for a US keyboard layout and + * are taken from the Linux driver. Note that the + * 102ND key is not used for US keyboards. + */ + linux,keymap = < + /* CAPSLCK F1 B F10 */ + 0x0001003a 0x0002003b 0x00030030 0x00040044 + /* N = R_ALT ESC */ + 0x00060031 0x0008000d 0x000a0064 0x01010001 + /* F4 G F7 H */ + 0x0102003e 0x01030022 0x01040041 0x01060023 + /* ' F9 BKSPACE L_CTRL */ + 0x01080028 0x01090043 0x010b000e 0x0200001d + /* TAB F3 T F6 */ + 0x0201000f 0x0202003d 0x02030014 0x02040040 + /* ] Y 102ND [ */ + 0x0205001b 0x02060015 0x02070056 0x0208001a + /* F8 GRAVE F2 5 */ + 0x02090042 0x03010029 0x0302003c 0x03030006 + /* F5 6 - \ */ + 0x0304003f 0x03060007 0x0308000c 0x030b002b + /* R_CTRL A D F */ + 0x04000061 0x0401001e 0x04020020 0x04030021 + /* S K J ; */ + 0x0404001f 0x04050025 0x04060024 0x04080027 + /* L ENTER Z C */ + 0x04090026 0x040b001c 0x0501002c 0x0502002e + /* V X , M */ + 0x0503002f 0x0504002d 0x05050033 0x05060032 + /* L_SHIFT / . SPACE */ + 0x0507002a 0x05080035 0x05090034 0x050B0039 + /* 1 3 4 2 */ + 0x06010002 0x06020004 0x06030005 0x06040003 + /* 8 7 0 9 */ + 0x06050009 0x06060008 0x0608000b 0x0609000a + /* L_ALT DOWN RIGHT Q */ + 0x060a0038 0x060b006c 0x060c006a 0x07010010 + /* E R W I */ + 0x07020012 0x07030013 0x07040011 0x07050017 + /* U R_SHIFT P O */ + 0x07060016 0x07070036 0x07080019 0x07090018 + /* UP LEFT */ + 0x070b0067 0x070c0069>; + }; diff --git a/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml b/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml index c45cf30ea3aa..351bfb6d37ba 100644 --- a/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml +++ b/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml @@ -71,6 +71,9 @@ properties: wakeup-source: description: Button can wake-up the system. + keyboard-controller: + $ref: "/schemas/input/google,cros-ec-keyb.yaml#" + patternProperties: "^i2c-tunnel[0-9]*$": type: object -- cgit v1.2.3 From 5c024e68d79b1ed8029afd2cae87c67865426cbe Mon Sep 17 00:00:00 2001 From: Ricardo Cañuelo Date: Wed, 21 Oct 2020 13:43:08 +0200 Subject: mfd: google,cros-ec: add missing properties MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add missing properties that are currently used in the examples of subnode bindings and in many DTs. Also updates the example in sound/google,cros-ec-codec.yaml to comply with the google,cros-ec binding. Reviewed-by: Enric Balletbo i Serra Signed-off-by: Ricardo Cañuelo Link: https://lore.kernel.org/r/20201021114308.25485-4-ricardo.canuelo@collabora.com [robh: Add missing '#address-cells' and '#size-cells'] Signed-off-by: Rob Herring --- .../devicetree/bindings/mfd/google,cros-ec.yaml | 48 ++++++++++++++++++++++ .../bindings/sound/google,cros-ec-codec.yaml | 26 +++++++----- 2 files changed, 64 insertions(+), 10 deletions(-) diff --git a/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml b/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml index 351bfb6d37ba..76bf16ee27ec 100644 --- a/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml +++ b/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml @@ -59,6 +59,14 @@ properties: whether this nvram is present or not. type: boolean + mtk,rpmsg-name: + description: + Must be defined if the cros-ec is a rpmsg device for a Mediatek + ARM Cortex M4 Co-processor. Contains the name pf the rpmsg + device. Used to match the subnode to the rpmsg device announced by + the SCP. + $ref: "/schemas/types.yaml#/definitions/string" + spi-max-frequency: description: Maximum SPI frequency of the device in Hz. @@ -71,14 +79,54 @@ properties: wakeup-source: description: Button can wake-up the system. + '#address-cells': + const: 1 + + '#size-cells': + const: 0 + + typec: + $ref: "/schemas/chrome/google,cros-ec-typec.yaml#" + + ec-pwm: + $ref: "/schemas/pwm/google,cros-ec-pwm.yaml#" + keyboard-controller: $ref: "/schemas/input/google,cros-ec-keyb.yaml#" + codecs: + type: object + additionalProperties: false + + properties: + '#address-cells': + const: 2 + + '#size-cells': + const: 1 + + patternProperties: + "^ec-codec@[a-f0-9]+$": + type: object + $ref: "/schemas/sound/google,cros-ec-codec.yaml#" + + required: + - "#address-cells" + - "#size-cells" + patternProperties: "^i2c-tunnel[0-9]*$": type: object $ref: "/schemas/i2c/google,cros-ec-i2c-tunnel.yaml#" + "^regulator@[0-9]+$": + type: object + $ref: "/schemas/regulator/google,cros-ec-regulator.yaml#" + + "^extcon[0-9]*$": + type: object + $ref: "/schemas/extcon/extcon-usbc-cros-ec.yaml#" + required: - compatible diff --git a/Documentation/devicetree/bindings/sound/google,cros-ec-codec.yaml b/Documentation/devicetree/bindings/sound/google,cros-ec-codec.yaml index 3b9143af2c7c..acfb9db021dc 100644 --- a/Documentation/devicetree/bindings/sound/google,cros-ec-codec.yaml +++ b/Documentation/devicetree/bindings/sound/google,cros-ec-codec.yaml @@ -11,9 +11,10 @@ maintainers: description: | Google's ChromeOS EC codec is a digital mic codec provided by the - Embedded Controller (EC) and is controlled via a host-command interface. - An EC codec node should only be found as a sub-node of the EC node (see - Documentation/devicetree/bindings/mfd/google,cros-ec.yaml). + Embedded Controller (EC) and is controlled via a host-command + interface. An EC codec node should only be found inside the "codecs" + subnode of a cros-ec node. + (see Documentation/devicetree/bindings/mfd/google,cros-ec.yaml). properties: compatible: @@ -54,14 +55,19 @@ examples: #size-cells = <0>; cros-ec@0 { compatible = "google,cros-ec-spi"; - #address-cells = <2>; - #size-cells = <1>; reg = <0>; - cros_ec_codec: ec-codec@10500000 { - compatible = "google,cros-ec-codec"; - #sound-dai-cells = <1>; - reg = <0x0 0x10500000 0x80000>; - memory-region = <&reserved_mem>; + + codecs { + #address-cells = <2>; + #size-cells = <1>; + + cros_ec_codec: ec-codec@10500000 { + compatible = "google,cros-ec-codec"; + #sound-dai-cells = <1>; + reg = <0x0 0x10500000 0x80000>; + memory-region = <&reserved_mem>; + }; + }; }; }; -- cgit v1.2.3 From 6ad8838de4e9ce7ccb19abeec169d224ddb07dde Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 20 Apr 2020 21:24:47 -0500 Subject: dt-bindings: More whitespace clean-ups in schema files Clean-up incorrect indentation, extra spaces, and missing EOF newline in schema files. Most of the clean-ups are for list indentation which should always be 2 spaces more than the preceding keyword. Found with yamllint (now integrated into the checks). Cc: linux-arm-kernel@lists.infradead.org Cc: dri-devel@lists.freedesktop.org Cc: linux-gpio@vger.kernel.org Cc: linux-i2c@vger.kernel.org Cc: linux-iio@vger.kernel.org Cc: linux-pm@vger.kernel.org Cc: alsa-devel@alsa-project.org Cc: linux-mmc@vger.kernel.org Cc: linux-mtd@lists.infradead.org Cc: linux-serial@vger.kernel.org Cc: linux-usb@vger.kernel.org Acked-by: Wolfram Sang # for I2C Acked-by: Sam Ravnborg # for display Acked-by: Jonathan Cameron #for-iio Signed-off-by: Rob Herring --- .../devicetree/bindings/arm/stm32/stm32.yaml | 8 +- .../bindings/display/bridge/toshiba,tc358775.yaml | 36 ++++---- .../bindings/display/panel/ilitek,ili9881c.yaml | 5 +- Documentation/devicetree/bindings/eeprom/at25.yaml | 6 +- .../bindings/gpio/kontron,sl28cpld-gpio.yaml | 4 +- .../devicetree/bindings/i2c/ingenic,i2c.yaml | 8 +- .../devicetree/bindings/iio/adc/adi,ad7291.yaml | 3 +- .../devicetree/bindings/iio/adc/adi,ad7768-1.yaml | 3 +- .../bindings/iio/adc/cosmic,10001-adc.yaml | 4 +- .../devicetree/bindings/iio/adc/holt,hi8435.yaml | 2 +- .../interrupt-controller/ti,pruss-intc.yaml | 12 +-- .../devicetree/bindings/mfd/ene-kb3930.yaml | 2 +- .../devicetree/bindings/mmc/arasan,sdhci.yaml | 8 +- .../devicetree/bindings/mmc/sdhci-am654.yaml | 15 ++- .../bindings/pci/socionext,uniphier-pcie-ep.yaml | 18 ++-- .../bindings/phy/socionext,uniphier-ahci-phy.yaml | 6 +- .../devicetree/bindings/phy/ti,omap-usb2.yaml | 20 ++-- .../bindings/pinctrl/actions,s500-pinctrl.yaml | 102 ++++++++++----------- .../bindings/pinctrl/pinctrl-mt8192.yaml | 2 +- .../bindings/pinctrl/qcom,msm8226-pinctrl.yaml | 6 +- .../bindings/pinctrl/toshiba,visconti-pinctrl.yaml | 24 ++--- .../bindings/power/reset/reboot-mode.yaml | 6 +- .../bindings/power/supply/ingenic,battery.yaml | 8 +- .../power/supply/summit,smb347-charger.yaml | 16 ++-- .../devicetree/bindings/riscv/sifive-l2-cache.yaml | 4 +- Documentation/devicetree/bindings/rng/imx-rng.yaml | 6 +- .../devicetree/bindings/serial/fsl-imx-uart.yaml | 34 +++---- .../devicetree/bindings/sound/mchp,spdifrx.yaml | 4 +- .../devicetree/bindings/sound/mchp,spdiftx.yaml | 4 +- .../devicetree/bindings/sound/qcom,lpass-cpu.yaml | 40 ++++---- .../devicetree/bindings/timer/arm,sp804.yaml | 12 +-- .../devicetree/bindings/usb/cdns,usb3.yaml | 4 +- .../devicetree/bindings/usb/ti,hd3ss3220.yaml | 2 +- .../devicetree/bindings/w1/fsl-imx-owire.yaml | 8 +- 34 files changed, 223 insertions(+), 219 deletions(-) diff --git a/Documentation/devicetree/bindings/arm/stm32/stm32.yaml b/Documentation/devicetree/bindings/arm/stm32/stm32.yaml index deacb4e686e8..6b87314eb9ed 100644 --- a/Documentation/devicetree/bindings/arm/stm32/stm32.yaml +++ b/Documentation/devicetree/bindings/arm/stm32/stm32.yaml @@ -54,8 +54,8 @@ properties: - const: st,stm32mp157 - description: Odyssey STM32MP1 SoM based Boards items: - - enum: - - seeed,stm32mp157c-odyssey - - const: seeed,stm32mp157c-odyssey-som - - const: st,stm32mp157 + - enum: + - seeed,stm32mp157c-odyssey + - const: seeed,stm32mp157c-odyssey-som + - const: st,stm32mp157 ... diff --git a/Documentation/devicetree/bindings/display/bridge/toshiba,tc358775.yaml b/Documentation/devicetree/bindings/display/bridge/toshiba,tc358775.yaml index 31f085d8ab13..e51d5cccc0e9 100644 --- a/Documentation/devicetree/bindings/display/bridge/toshiba,tc358775.yaml +++ b/Documentation/devicetree/bindings/display/bridge/toshiba,tc358775.yaml @@ -7,17 +7,17 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: Toshiba TC358775 DSI to LVDS bridge bindings maintainers: - - Vinay Simha BN + - Vinay Simha BN description: | - This binding supports DSI to LVDS bridge TC358775 + This binding supports DSI to LVDS bridge TC358775 - MIPI DSI-RX Data 4-lane, CLK 1-lane with data rates up to 800 Mbps/lane. - Video frame size: - Up to 1600x1200 24-bit/pixel resolution for single-link LVDS display panel - limited by 135 MHz LVDS speed - Up to WUXGA (1920x1200 24-bit pixels) resolution for dual-link LVDS display - panel, limited by 270 MHz LVDS speed. + MIPI DSI-RX Data 4-lane, CLK 1-lane with data rates up to 800 Mbps/lane. + Video frame size: + Up to 1600x1200 24-bit/pixel resolution for single-link LVDS display panel + limited by 135 MHz LVDS speed + Up to WUXGA (1920x1200 24-bit pixels) resolution for dual-link LVDS display + panel, limited by 270 MHz LVDS speed. properties: compatible: @@ -29,7 +29,7 @@ properties: vdd-supply: maxItems: 1 - description: 1.2V LVDS Power Supply + description: 1.2V LVDS Power Supply vddio-supply: maxItems: 1 @@ -77,16 +77,16 @@ properties: - port@1 required: - - compatible - - reg - - vdd-supply - - vddio-supply - - stby-gpios - - reset-gpios - - ports + - compatible + - reg + - vdd-supply + - vddio-supply + - stby-gpios + - reset-gpios + - ports examples: - - | + - | #include /* For single-link LVDS display panel */ @@ -147,7 +147,7 @@ examples: }; }; - - | + - | /* For dual-link LVDS display panel */ i2c@78b8000 { diff --git a/Documentation/devicetree/bindings/display/panel/ilitek,ili9881c.yaml b/Documentation/devicetree/bindings/display/panel/ilitek,ili9881c.yaml index c60b3bd74337..b2fcec4f22fd 100644 --- a/Documentation/devicetree/bindings/display/panel/ilitek,ili9881c.yaml +++ b/Documentation/devicetree/bindings/display/panel/ilitek,ili9881c.yaml @@ -13,9 +13,8 @@ properties: compatible: items: - enum: - - bananapi,lhr050h41 - - feixin,k101-im2byl02 - + - bananapi,lhr050h41 + - feixin,k101-im2byl02 - const: ilitek,ili9881c backlight: true diff --git a/Documentation/devicetree/bindings/eeprom/at25.yaml b/Documentation/devicetree/bindings/eeprom/at25.yaml index 9810619a2b5c..744973637678 100644 --- a/Documentation/devicetree/bindings/eeprom/at25.yaml +++ b/Documentation/devicetree/bindings/eeprom/at25.yaml @@ -81,14 +81,14 @@ properties: at25,byte-len: $ref: /schemas/types.yaml#/definitions/uint32 description: - Total eeprom size in bytes. Deprecated, use "size" property instead. + Total eeprom size in bytes. Deprecated, use "size" property instead. deprecated: true at25,addr-mode: $ref: /schemas/types.yaml#/definitions/uint32 description: - Addr-mode flags, as defined in include/linux/spi/eeprom.h. - Deprecated, use "address-width" property instead. + Addr-mode flags, as defined in include/linux/spi/eeprom.h. + Deprecated, use "address-width" property instead. deprecated: true at25,page-size: diff --git a/Documentation/devicetree/bindings/gpio/kontron,sl28cpld-gpio.yaml b/Documentation/devicetree/bindings/gpio/kontron,sl28cpld-gpio.yaml index e2d2c10e536a..b032471831e7 100644 --- a/Documentation/devicetree/bindings/gpio/kontron,sl28cpld-gpio.yaml +++ b/Documentation/devicetree/bindings/gpio/kontron,sl28cpld-gpio.yaml @@ -43,8 +43,8 @@ properties: gpio-controller: true gpio-line-names: - minItems: 1 - maxItems: 8 + minItems: 1 + maxItems: 8 required: - compatible diff --git a/Documentation/devicetree/bindings/i2c/ingenic,i2c.yaml b/Documentation/devicetree/bindings/i2c/ingenic,i2c.yaml index 0e7b4b8a7e48..e1e65eb4f795 100644 --- a/Documentation/devicetree/bindings/i2c/ingenic,i2c.yaml +++ b/Documentation/devicetree/bindings/i2c/ingenic,i2c.yaml @@ -19,11 +19,11 @@ properties: compatible: oneOf: - enum: - - ingenic,jz4770-i2c - - ingenic,x1000-i2c + - ingenic,jz4770-i2c + - ingenic,x1000-i2c - items: - - const: ingenic,jz4780-i2c - - const: ingenic,jz4770-i2c + - const: ingenic,jz4780-i2c + - const: ingenic,jz4770-i2c reg: maxItems: 1 diff --git a/Documentation/devicetree/bindings/iio/adc/adi,ad7291.yaml b/Documentation/devicetree/bindings/iio/adc/adi,ad7291.yaml index 6feafb7e531e..930f9e3904d7 100644 --- a/Documentation/devicetree/bindings/iio/adc/adi,ad7291.yaml +++ b/Documentation/devicetree/bindings/iio/adc/adi,ad7291.yaml @@ -43,4 +43,5 @@ examples: vref-supply = <&adc_vref>; }; }; -... \ No newline at end of file +... + diff --git a/Documentation/devicetree/bindings/iio/adc/adi,ad7768-1.yaml b/Documentation/devicetree/bindings/iio/adc/adi,ad7768-1.yaml index d3733ad8785a..8f32800fe5b7 100644 --- a/Documentation/devicetree/bindings/iio/adc/adi,ad7768-1.yaml +++ b/Documentation/devicetree/bindings/iio/adc/adi,ad7768-1.yaml @@ -46,7 +46,8 @@ properties: spi-max-frequency: true spi-cpol: true - spi-cpha : true + + spi-cpha: true "#io-channel-cells": const: 1 diff --git a/Documentation/devicetree/bindings/iio/adc/cosmic,10001-adc.yaml b/Documentation/devicetree/bindings/iio/adc/cosmic,10001-adc.yaml index 5d92b477e23f..4e695b97d015 100644 --- a/Documentation/devicetree/bindings/iio/adc/cosmic,10001-adc.yaml +++ b/Documentation/devicetree/bindings/iio/adc/cosmic,10001-adc.yaml @@ -22,8 +22,8 @@ properties: adc-reserved-channels: $ref: /schemas/types.yaml#/definitions/uint32 description: - Bitmask of reserved channels, i.e. channels that cannot be - used by the OS. + Bitmask of reserved channels, i.e. channels that cannot be + used by the OS. clocks: maxItems: 1 diff --git a/Documentation/devicetree/bindings/iio/adc/holt,hi8435.yaml b/Documentation/devicetree/bindings/iio/adc/holt,hi8435.yaml index 9514c3381c42..52490cbb0af0 100644 --- a/Documentation/devicetree/bindings/iio/adc/holt,hi8435.yaml +++ b/Documentation/devicetree/bindings/iio/adc/holt,hi8435.yaml @@ -21,7 +21,7 @@ properties: gpios: description: - GPIO used for controlling the reset pin + GPIO used for controlling the reset pin maxItems: 1 spi-max-frequency: true diff --git a/Documentation/devicetree/bindings/interrupt-controller/ti,pruss-intc.yaml b/Documentation/devicetree/bindings/interrupt-controller/ti,pruss-intc.yaml index bbf79d125675..1c4c009dedd0 100644 --- a/Documentation/devicetree/bindings/interrupt-controller/ti,pruss-intc.yaml +++ b/Documentation/devicetree/bindings/interrupt-controller/ti,pruss-intc.yaml @@ -94,12 +94,12 @@ properties: instances. required: - - compatible - - reg - - interrupts - - interrupt-names - - interrupt-controller - - "#interrupt-cells" + - compatible + - reg + - interrupts + - interrupt-names + - interrupt-controller + - "#interrupt-cells" additionalProperties: false diff --git a/Documentation/devicetree/bindings/mfd/ene-kb3930.yaml b/Documentation/devicetree/bindings/mfd/ene-kb3930.yaml index 074243c40891..08af356f5d27 100644 --- a/Documentation/devicetree/bindings/mfd/ene-kb3930.yaml +++ b/Documentation/devicetree/bindings/mfd/ene-kb3930.yaml @@ -17,7 +17,7 @@ properties: compatible: items: - enum: - - dell,wyse-ariel-ec # Dell Wyse Ariel board (3020) + - dell,wyse-ariel-ec # Dell Wyse Ariel board (3020) - const: ene,kb3930 reg: maxItems: 1 diff --git a/Documentation/devicetree/bindings/mmc/arasan,sdhci.yaml b/Documentation/devicetree/bindings/mmc/arasan,sdhci.yaml index 58fe9d02a781..0753289fba84 100644 --- a/Documentation/devicetree/bindings/mmc/arasan,sdhci.yaml +++ b/Documentation/devicetree/bindings/mmc/arasan,sdhci.yaml @@ -32,11 +32,11 @@ allOf: clock-output-names: oneOf: - items: - - const: clk_out_sd0 - - const: clk_in_sd0 + - const: clk_out_sd0 + - const: clk_in_sd0 - items: - - const: clk_out_sd1 - - const: clk_in_sd1 + - const: clk_out_sd1 + - const: clk_in_sd1 properties: compatible: diff --git a/Documentation/devicetree/bindings/mmc/sdhci-am654.yaml b/Documentation/devicetree/bindings/mmc/sdhci-am654.yaml index ac79f3adf20b..2cd859066462 100644 --- a/Documentation/devicetree/bindings/mmc/sdhci-am654.yaml +++ b/Documentation/devicetree/bindings/mmc/sdhci-am654.yaml @@ -3,7 +3,7 @@ %YAML 1.2 --- $id: "http://devicetree.org/schemas/mmc/sdhci-am654.yaml#" -$schema : "http://devicetree.org/meta-schemas/core.yaml#" +$schema: "http://devicetree.org/meta-schemas/core.yaml#" title: TI AM654 MMC Controller @@ -163,13 +163,12 @@ properties: ti,driver-strength-ohm: description: DLL drive strength in ohms $ref: "/schemas/types.yaml#/definitions/uint32" - oneOf: - - enum: - - 33 - - 40 - - 50 - - 66 - - 100 + enum: + - 33 + - 40 + - 50 + - 66 + - 100 ti,strobe-sel: description: strobe select delay for HS400 speed mode. diff --git a/Documentation/devicetree/bindings/pci/socionext,uniphier-pcie-ep.yaml b/Documentation/devicetree/bindings/pci/socionext,uniphier-pcie-ep.yaml index f4292d2c54e3..d6cf8a560ef0 100644 --- a/Documentation/devicetree/bindings/pci/socionext,uniphier-pcie-ep.yaml +++ b/Documentation/devicetree/bindings/pci/socionext,uniphier-pcie-ep.yaml @@ -29,16 +29,16 @@ properties: reg-names: oneOf: - items: - - const: dbi - - const: dbi2 - - const: link - - const: addr_space + - const: dbi + - const: dbi2 + - const: link + - const: addr_space - items: - - const: dbi - - const: dbi2 - - const: link - - const: addr_space - - const: atu + - const: dbi + - const: dbi2 + - const: link + - const: addr_space + - const: atu clocks: maxItems: 2 diff --git a/Documentation/devicetree/bindings/phy/socionext,uniphier-ahci-phy.yaml b/Documentation/devicetree/bindings/phy/socionext,uniphier-ahci-phy.yaml index bab2ff4d9dc9..34756347a14e 100644 --- a/Documentation/devicetree/bindings/phy/socionext,uniphier-ahci-phy.yaml +++ b/Documentation/devicetree/bindings/phy/socionext,uniphier-ahci-phy.yaml @@ -31,10 +31,10 @@ properties: clock-names: oneOf: - items: # for PXs2 - - const: link + - const: link - items: # for others - - const: link - - const: phy + - const: link + - const: phy resets: maxItems: 2 diff --git a/Documentation/devicetree/bindings/phy/ti,omap-usb2.yaml b/Documentation/devicetree/bindings/phy/ti,omap-usb2.yaml index 15207ca9548f..ab77a65c5c55 100644 --- a/Documentation/devicetree/bindings/phy/ti,omap-usb2.yaml +++ b/Documentation/devicetree/bindings/phy/ti,omap-usb2.yaml @@ -7,23 +7,23 @@ $schema: http://devicetree.org/meta-schemas/core.yaml# title: OMAP USB2 PHY maintainers: - - Kishon Vijay Abraham I - - Roger Quadros + - Kishon Vijay Abraham I + - Roger Quadros properties: compatible: oneOf: - items: - - enum: - - ti,dra7x-usb2 - - ti,dra7x-usb2-phy2 - - ti,am654-usb2 - - enum: - - ti,omap-usb2 + - enum: + - ti,dra7x-usb2 + - ti,dra7x-usb2-phy2 + - ti,am654-usb2 + - enum: + - ti,omap-usb2 - items: - - const: ti,am437x-usb2 + - const: ti,am437x-usb2 - items: - - const: ti,omap-usb2 + - const: ti,omap-usb2 reg: maxItems: 1 diff --git a/Documentation/devicetree/bindings/pinctrl/actions,s500-pinctrl.yaml b/Documentation/devicetree/bindings/pinctrl/actions,s500-pinctrl.yaml index 33391d30c00c..ccdd9e3820d7 100644 --- a/Documentation/devicetree/bindings/pinctrl/actions,s500-pinctrl.yaml +++ b/Documentation/devicetree/bindings/pinctrl/actions,s500-pinctrl.yaml @@ -76,22 +76,22 @@ patternProperties: items: oneOf: - enum: [lcd0_d18_mfp, rmii_crs_dv_mfp, rmii_txd0_mfp, - rmii_txd1_mfp, rmii_txen_mfp, rmii_rxen_mfp, rmii_rxd1_mfp, - rmii_rxd0_mfp, rmii_ref_clk_mfp, i2s_d0_mfp, i2s_pcm1_mfp, - i2s0_pcm0_mfp, i2s1_pcm0_mfp, i2s_d1_mfp, ks_in2_mfp, - ks_in1_mfp, ks_in0_mfp, ks_in3_mfp, ks_out0_mfp, - ks_out1_mfp, ks_out2_mfp, lvds_o_pn_mfp, dsi_dn0_mfp, - dsi_dp2_mfp, lcd0_d17_mfp, dsi_dp3_mfp, dsi_dn3_mfp, - dsi_dp0_mfp, lvds_ee_pn_mfp, spi0_i2c_pcm_mfp, - spi0_i2s_pcm_mfp, dsi_dnp1_cp_mfp, lvds_e_pn_mfp, - dsi_dn2_mfp, uart2_rtsb_mfp, uart2_ctsb_mfp, uart3_rtsb_mfp, - uart3_ctsb_mfp, sd0_d0_mfp, sd0_d1_mfp, sd0_d2_d3_mfp, - sd1_d0_d3_mfp, sd0_cmd_mfp, sd0_clk_mfp, sd1_cmd_mfp, - uart0_rx_mfp, clko_25m_mfp, csi_cn_cp_mfp, sens0_ckout_mfp, - uart0_tx_mfp, i2c0_mfp, csi_dn_dp_mfp, sen0_pclk_mfp, - pcm1_in_mfp, pcm1_clk_mfp, pcm1_sync_mfp, pcm1_out_mfp, - dnand_data_wr_mfp, dnand_acle_ce0_mfp, nand_ceb2_mfp, - nand_ceb3_mfp] + rmii_txd1_mfp, rmii_txen_mfp, rmii_rxen_mfp, rmii_rxd1_mfp, + rmii_rxd0_mfp, rmii_ref_clk_mfp, i2s_d0_mfp, i2s_pcm1_mfp, + i2s0_pcm0_mfp, i2s1_pcm0_mfp, i2s_d1_mfp, ks_in2_mfp, + ks_in1_mfp, ks_in0_mfp, ks_in3_mfp, ks_out0_mfp, + ks_out1_mfp, ks_out2_mfp, lvds_o_pn_mfp, dsi_dn0_mfp, + dsi_dp2_mfp, lcd0_d17_mfp, dsi_dp3_mfp, dsi_dn3_mfp, + dsi_dp0_mfp, lvds_ee_pn_mfp, spi0_i2c_pcm_mfp, + spi0_i2s_pcm_mfp, dsi_dnp1_cp_mfp, lvds_e_pn_mfp, + dsi_dn2_mfp, uart2_rtsb_mfp, uart2_ctsb_mfp, uart3_rtsb_mfp, + uart3_ctsb_mfp, sd0_d0_mfp, sd0_d1_mfp, sd0_d2_d3_mfp, + sd1_d0_d3_mfp, sd0_cmd_mfp, sd0_clk_mfp, sd1_cmd_mfp, + uart0_rx_mfp, clko_25m_mfp, csi_cn_cp_mfp, sens0_ckout_mfp, + uart0_tx_mfp, i2c0_mfp, csi_dn_dp_mfp, sen0_pclk_mfp, + pcm1_in_mfp, pcm1_clk_mfp, pcm1_sync_mfp, pcm1_out_mfp, + dnand_data_wr_mfp, dnand_acle_ce0_mfp, nand_ceb2_mfp, + nand_ceb3_mfp] minItems: 1 maxItems: 32 @@ -100,10 +100,10 @@ patternProperties: Specify the alternative function to be configured for the given gpio pin groups. enum: [nor, eth_rmii, eth_smii, spi0, spi1, spi2, spi3, sens0, - sens1, uart0, uart1, uart2, uart3, uart4, uart5, uart6, i2s0, - i2s1, pcm1, pcm0, ks, jtag, pwm0, pwm1, pwm2, pwm3, pwm4, pwm5, - p0, sd0, sd1, sd2, i2c0, i2c1, i2c3, dsi, lvds, usb30, clko_25m, - mipi_csi, nand, spdif, ts, lcd0] + sens1, uart0, uart1, uart2, uart3, uart4, uart5, uart6, i2s0, + i2s1, pcm1, pcm0, ks, jtag, pwm0, pwm1, pwm2, pwm3, pwm4, pwm5, + p0, sd0, sd1, sd2, i2c0, i2c1, i2c3, dsi, lvds, usb30, clko_25m, + mipi_csi, nand, spdif, ts, lcd0] required: - groups @@ -126,14 +126,14 @@ patternProperties: items: oneOf: - enum: [sirq_drv, rmii_txd01_txen_drv, rmii_rxer_drv, - rmii_crs_drv, rmii_rxd10_drv, rmii_ref_clk_drv, - smi_mdc_mdio_drv, i2s_d0_drv, i2s_bclk0_drv, i2s3_drv, - i2s13_drv, pcm1_drv, ks_in_drv, ks_out_drv, lvds_all_drv, - lcd_dsi_drv, dsi_drv, sd0_d0_d3_drv, sd1_d0_d3_drv, - sd0_cmd_drv, sd0_clk_drv, sd1_cmd_drv, sd1_clk_drv, - spi0_all_drv, uart0_rx_drv, uart0_tx_drv, uart2_all_drv, - i2c0_all_drv, i2c12_all_drv, sens0_pclk_drv, - sens0_ckout_drv, uart3_all_drv] + rmii_crs_drv, rmii_rxd10_drv, rmii_ref_clk_drv, + smi_mdc_mdio_drv, i2s_d0_drv, i2s_bclk0_drv, i2s3_drv, + i2s13_drv, pcm1_drv, ks_in_drv, ks_out_drv, lvds_all_drv, + lcd_dsi_drv, dsi_drv, sd0_d0_d3_drv, sd1_d0_d3_drv, + sd0_cmd_drv, sd0_clk_drv, sd1_cmd_drv, sd1_clk_drv, + spi0_all_drv, uart0_rx_drv, uart0_tx_drv, uart2_all_drv, + i2c0_all_drv, i2c12_all_drv, sens0_pclk_drv, + sens0_ckout_drv, uart3_all_drv] minItems: 1 maxItems: 32 @@ -144,29 +144,29 @@ patternProperties: items: oneOf: - enum: [dnand_dqs, dnand_dqsn, eth_txd0, eth_txd1, eth_txen, - eth_rxer, eth_crs_dv, eth_rxd1, eth_rxd0, eth_ref_clk, - eth_mdc, eth_mdio, sirq0, sirq1, sirq2, i2s_d0, i2s_bclk0, - i2s_lrclk0, i2s_mclk0, i2s_d1, i2s_bclk1, i2s_lrclk1, - i2s_mclk1, ks_in0, ks_in1, ks_in2, ks_in3, ks_out0, ks_out1, - ks_out2, lvds_oep, lvds_oen, lvds_odp, lvds_odn, lvds_ocp, - lvds_ocn, lvds_obp, lvds_obn, lvds_oap, lvds_oan, lvds_eep, - lvds_een, lvds_edp, lvds_edn, lvds_ecp, lvds_ecn, lvds_ebp, - lvds_ebn, lvds_eap, lvds_ean, lcd0_d18, lcd0_d17, dsi_dp3, - dsi_dn3, dsi_dp1, dsi_dn1, dsi_cp, dsi_cn, dsi_dp0, dsi_dn0, - dsi_dp2, dsi_dn2, sd0_d0, sd0_d1, sd0_d2, sd0_d3, sd1_d0, - sd1_d1, sd1_d2, sd1_d3, sd0_cmd, sd0_clk, sd1_cmd, sd1_clk, - spi0_sclk, spi0_ss, spi0_miso, spi0_mosi, uart0_rx, - uart0_tx, i2c0_sclk, i2c0_sdata, sensor0_pclk, - sensor0_ckout, dnand_ale, dnand_cle, dnand_ceb0, dnand_ceb1, - dnand_ceb2, dnand_ceb3, uart2_rx, uart2_tx, uart2_rtsb, - uart2_ctsb, uart3_rx, uart3_tx, uart3_rtsb, uart3_ctsb, - pcm1_in, pcm1_clk, pcm1_sync, pcm1_out, i2c1_sclk, - i2c1_sdata, i2c2_sclk, i2c2_sdata, csi_dn0, csi_dp0, - csi_dn1, csi_dp1, csi_dn2, csi_dp2, csi_dn3, csi_dp3, - csi_cn, csi_cp, dnand_d0, dnand_d1, dnand_d2, dnand_d3, - dnand_d4, dnand_d5, dnand_d6, dnand_d7, dnand_rb, dnand_rdb, - dnand_rdbn, dnand_wrb, porb, clko_25m, bsel, pkg0, pkg1, - pkg2, pkg3] + eth_rxer, eth_crs_dv, eth_rxd1, eth_rxd0, eth_ref_clk, + eth_mdc, eth_mdio, sirq0, sirq1, sirq2, i2s_d0, i2s_bclk0, + i2s_lrclk0, i2s_mclk0, i2s_d1, i2s_bclk1, i2s_lrclk1, + i2s_mclk1, ks_in0, ks_in1, ks_in2, ks_in3, ks_out0, ks_out1, + ks_out2, lvds_oep, lvds_oen, lvds_odp, lvds_odn, lvds_ocp, + lvds_ocn, lvds_obp, lvds_obn, lvds_oap, lvds_oan, lvds_eep, + lvds_een, lvds_edp, lvds_edn, lvds_ecp, lvds_ecn, lvds_ebp, + lvds_ebn, lvds_eap, lvds_ean, lcd0_d18, lcd0_d17, dsi_dp3, + dsi_dn3, dsi_dp1, dsi_dn1, dsi_cp, dsi_cn, dsi_dp0, dsi_dn0, + dsi_dp2, dsi_dn2, sd0_d0, sd0_d1, sd0_d2, sd0_d3, sd1_d0, + sd1_d1, sd1_d2, sd1_d3, sd0_cmd, sd0_clk, sd1_cmd, sd1_clk, + spi0_sclk, spi0_ss, spi0_miso, spi0_mosi, uart0_rx, + uart0_tx, i2c0_sclk, i2c0_sdata, sensor0_pclk, + sensor0_ckout, dnand_ale, dnand_cle, dnand_ceb0, dnand_ceb1, + dnand_ceb2, dnand_ceb3, uart2_rx, uart2_tx, uart2_rtsb, + uart2_ctsb, uart3_rx, uart3_tx, uart3_rtsb, uart3_ctsb, + pcm1_in, pcm1_clk, pcm1_sync, pcm1_out, i2c1_sclk, + i2c1_sdata, i2c2_sclk, i2c2_sdata, csi_dn0, csi_dp0, + csi_dn1, csi_dp1, csi_dn2, csi_dp2, csi_dn3, csi_dp3, + csi_cn, csi_cp, dnand_d0, dnand_d1, dnand_d2, dnand_d3, + dnand_d4, dnand_d5, dnand_d6, dnand_d7, dnand_rb, dnand_rdb, + dnand_rdbn, dnand_wrb, porb, clko_25m, bsel, pkg0, pkg1, + pkg2, pkg3] minItems: 1 maxItems: 64 diff --git a/Documentation/devicetree/bindings/pinctrl/pinctrl-mt8192.yaml b/Documentation/devicetree/bindings/pinctrl/pinctrl-mt8192.yaml index 5556def6b99b..c4c071211611 100644 --- a/Documentation/devicetree/bindings/pinctrl/pinctrl-mt8192.yaml +++ b/Documentation/devicetree/bindings/pinctrl/pinctrl-mt8192.yaml @@ -106,7 +106,7 @@ patternProperties: required: - pinmux - additionalProperties: false + additionalProperties: false required: - compatible diff --git a/Documentation/devicetree/bindings/pinctrl/qcom,msm8226-pinctrl.yaml b/Documentation/devicetree/bindings/pinctrl/qcom,msm8226-pinctrl.yaml index 1f0f5757f9e1..040d2ada3669 100644 --- a/Documentation/devicetree/bindings/pinctrl/qcom,msm8226-pinctrl.yaml +++ b/Documentation/devicetree/bindings/pinctrl/qcom,msm8226-pinctrl.yaml @@ -71,9 +71,9 @@ patternProperties: Specify the alternative function to be configured for the specified pins. Functions are only valid for gpio pins. enum: [ gpio, cci_i2c0, blsp_uim1, blsp_uim2, blsp_uim3, blsp_uim5, - blsp_i2c1, blsp_i2c2, blsp_i2c3, blsp_i2c5, blsp_spi1, - blsp_spi2, blsp_spi3, blsp_spi5, blsp_uart1, blsp_uart2, - blsp_uart3, blsp_uart5, cam_mclk0, cam_mclk1, wlan ] + blsp_i2c1, blsp_i2c2, blsp_i2c3, blsp_i2c5, blsp_spi1, + blsp_spi2, blsp_spi3, blsp_spi5, blsp_uart1, blsp_uart2, + blsp_uart3, blsp_uart5, cam_mclk0, cam_mclk1, wlan ] drive-strength: enum: [2, 4, 6, 8, 10, 12, 14, 16] diff --git a/Documentation/devicetree/bindings/pinctrl/toshiba,visconti-pinctrl.yaml b/Documentation/devicetree/bindings/pinctrl/toshiba,visconti-pinctrl.yaml index d0d1a01140ea..9f1dab0c2430 100644 --- a/Documentation/devicetree/bindings/pinctrl/toshiba,visconti-pinctrl.yaml +++ b/Documentation/devicetree/bindings/pinctrl/toshiba,visconti-pinctrl.yaml @@ -40,24 +40,24 @@ patternProperties: Function to mux. $ref: "/schemas/types.yaml#/definitions/string" enum: [i2c0, i2c1, i2c2, i2c3, i2c4, i2c5, i2c6, i2c7, i2c8, - spi0, spi1, spi2, spi3, spi4, spi5, spi6, - uart0, uart1, uart2, uart3, pwm, pcmif_out, pcmif_in] + spi0, spi1, spi2, spi3, spi4, spi5, spi6, + uart0, uart1, uart2, uart3, pwm, pcmif_out, pcmif_in] groups: description: Name of the pin group to use for the functions. $ref: "/schemas/types.yaml#/definitions/string" enum: [i2c0_grp, i2c1_grp, i2c2_grp, i2c3_grp, i2c4_grp, - i2c5_grp, i2c6_grp, i2c7_grp, i2c8_grp, - spi0_grp, spi0_cs0_grp, spi0_cs1_grp, spi0_cs2_grp, - spi1_grp, spi2_grp, spi3_grp, spi4_grp, spi5_grp, spi6_grp, - uart0_grp, uart1_grp, uart2_grp, uart3_grp, - pwm0_gpio4_grp, pwm0_gpio8_grp, pwm0_gpio12_grp, - pwm0_gpio16_grp, pwm1_gpio5_grp, pwm1_gpio9_grp, - pwm1_gpio13_grp, pwm1_gpio17_grp, pwm2_gpio6_grp, - pwm2_gpio10_grp, pwm2_gpio14_grp, pwm2_gpio18_grp, - pwm3_gpio7_grp, pwm3_gpio11_grp, pwm3_gpio15_grp, - pwm3_gpio19_grp, pcmif_out_grp, pcmif_in_grp] + i2c5_grp, i2c6_grp, i2c7_grp, i2c8_grp, + spi0_grp, spi0_cs0_grp, spi0_cs1_grp, spi0_cs2_grp, + spi1_grp, spi2_grp, spi3_grp, spi4_grp, spi5_grp, spi6_grp, + uart0_grp, uart1_grp, uart2_grp, uart3_grp, + pwm0_gpio4_grp, pwm0_gpio8_grp, pwm0_gpio12_grp, + pwm0_gpio16_grp, pwm1_gpio5_grp, pwm1_gpio9_grp, + pwm1_gpio13_grp, pwm1_gpio17_grp, pwm2_gpio6_grp, + pwm2_gpio10_grp, pwm2_gpio14_grp, pwm2_gpio18_grp, + pwm3_gpio7_grp, pwm3_gpio11_grp, pwm3_gpio15_grp, + pwm3_gpio19_grp, pcmif_out_grp, pcmif_in_grp] drive-strength: enum: [2, 4, 6, 8, 16, 24, 32] diff --git a/Documentation/devicetree/bindings/power/reset/reboot-mode.yaml b/Documentation/devicetree/bindings/power/reset/reboot-mode.yaml index a6c91026d4cc..a2dca38577f7 100644 --- a/Documentation/devicetree/bindings/power/reset/reboot-mode.yaml +++ b/Documentation/devicetree/bindings/power/reset/reboot-mode.yaml @@ -28,9 +28,9 @@ description: | properties: mode-normal: - $ref: /schemas/types.yaml#/definitions/uint32 - description: | - Default value to set on a reboot if no command was provided. + $ref: /schemas/types.yaml#/definitions/uint32 + description: + Default value to set on a reboot if no command was provided. patternProperties: "^mode-.*$": diff --git a/Documentation/devicetree/bindings/power/supply/ingenic,battery.yaml b/Documentation/devicetree/bindings/power/supply/ingenic,battery.yaml index 867e3e6b7e80..76c227a7cd5c 100644 --- a/Documentation/devicetree/bindings/power/supply/ingenic,battery.yaml +++ b/Documentation/devicetree/bindings/power/supply/ingenic,battery.yaml @@ -15,10 +15,10 @@ properties: oneOf: - const: ingenic,jz4740-battery - items: - - enum: - - ingenic,jz4725b-battery - - ingenic,jz4770-battery - - const: ingenic,jz4740-battery + - enum: + - ingenic,jz4725b-battery + - ingenic,jz4770-battery + - const: ingenic,jz4740-battery io-channels: maxItems: 1 diff --git a/Documentation/devicetree/bindings/power/supply/summit,smb347-charger.yaml b/Documentation/devicetree/bindings/power/supply/summit,smb347-charger.yaml index 193a23af2007..983fc215c1e5 100644 --- a/Documentation/devicetree/bindings/power/supply/summit,smb347-charger.yaml +++ b/Documentation/devicetree/bindings/power/supply/summit,smb347-charger.yaml @@ -84,12 +84,12 @@ allOf: then: properties: summit,mains-current-limit-microamp: - enum: [ 300000, 500000, 700000, 1000000, - 1500000, 1800000, 2000000] + enum: [ 300000, 500000, 700000, 1000000, + 1500000, 1800000, 2000000] summit,usb-current-limit-microamp: - enum: [ 300000, 500000, 700000, 1000000, - 1500000, 1800000, 2000000] + enum: [ 300000, 500000, 700000, 1000000, + 1500000, 1800000, 2000000] summit,charge-current-compensation-microamp: enum: [200000, 450000, 600000, 900000] @@ -97,12 +97,12 @@ allOf: else: properties: summit,mains-current-limit-microamp: - enum: [ 300000, 500000, 700000, 900000, 1200000, - 1500000, 1800000, 2000000, 2200000, 2500000] + enum: [ 300000, 500000, 700000, 900000, 1200000, + 1500000, 1800000, 2000000, 2200000, 2500000] summit,usb-current-limit-microamp: - enum: [ 300000, 500000, 700000, 900000, 1200000, - 1500000, 1800000, 2000000, 2200000, 2500000] + enum: [ 300000, 500000, 700000, 900000, 1200000, + 1500000, 1800000, 2000000, 2200000, 2500000] summit,charge-current-compensation-microamp: enum: [250000, 700000, 900000, 1200000] diff --git a/Documentation/devicetree/bindings/riscv/sifive-l2-cache.yaml b/Documentation/devicetree/bindings/riscv/sifive-l2-cache.yaml index 3f4a1939554d..efc0198eeb74 100644 --- a/Documentation/devicetree/bindings/riscv/sifive-l2-cache.yaml +++ b/Documentation/devicetree/bindings/riscv/sifive-l2-cache.yaml @@ -25,8 +25,8 @@ select: properties: compatible: items: - - enum: - - sifive,fu540-c000-ccache + - enum: + - sifive,fu540-c000-ccache required: - compatible diff --git a/Documentation/devicetree/bindings/rng/imx-rng.yaml b/Documentation/devicetree/bindings/rng/imx-rng.yaml index 4ad1e456a801..07f6ff89bcc1 100644 --- a/Documentation/devicetree/bindings/rng/imx-rng.yaml +++ b/Documentation/devicetree/bindings/rng/imx-rng.yaml @@ -19,9 +19,9 @@ properties: - const: fsl,imx21-rnga - items: - enum: - - fsl,imx6sl-rngb - - fsl,imx6sll-rngb - - fsl,imx6ull-rngb + - fsl,imx6sl-rngb + - fsl,imx6sll-rngb + - fsl,imx6ull-rngb - const: fsl,imx25-rngb - const: fsl,imx35-rngc diff --git a/Documentation/devicetree/bindings/serial/fsl-imx-uart.yaml b/Documentation/devicetree/bindings/serial/fsl-imx-uart.yaml index 9ff85bc6859c..9702c07a6b6c 100644 --- a/Documentation/devicetree/bindings/serial/fsl-imx-uart.yaml +++ b/Documentation/devicetree/bindings/serial/fsl-imx-uart.yaml @@ -20,30 +20,30 @@ properties: - const: fsl,imx21-uart - items: - enum: - - fsl,imx25-uart - - fsl,imx27-uart - - fsl,imx31-uart - - fsl,imx35-uart - - fsl,imx50-uart - - fsl,imx51-uart - - fsl,imx53-uart - - fsl,imx6q-uart + - fsl,imx25-uart + - fsl,imx27-uart + - fsl,imx31-uart + - fsl,imx35-uart + - fsl,imx50-uart + - fsl,imx51-uart + - fsl,imx53-uart + - fsl,imx6q-uart - const: fsl,imx21-uart - items: - enum: - - fsl,imx6sl-uart - - fsl,imx6sll-uart - - fsl,imx6sx-uart + - fsl,imx6sl-uart + - fsl,imx6sll-uart + - fsl,imx6sx-uart - const: fsl,imx6q-uart - const: fsl,imx21-uart - items: - enum: - - fsl,imx6ul-uart - - fsl,imx7d-uart - - fsl,imx8mm-uart - - fsl,imx8mn-uart - - fsl,imx8mp-uart - - fsl,imx8mq-uart + - fsl,imx6ul-uart + - fsl,imx7d-uart + - fsl,imx8mm-uart + - fsl,imx8mn-uart + - fsl,imx8mp-uart + - fsl,imx8mq-uart - const: fsl,imx6q-uart reg: diff --git a/Documentation/devicetree/bindings/sound/mchp,spdifrx.yaml b/Documentation/devicetree/bindings/sound/mchp,spdifrx.yaml index 7d8bd4e14434..4a2129005c0f 100644 --- a/Documentation/devicetree/bindings/sound/mchp,spdifrx.yaml +++ b/Documentation/devicetree/bindings/sound/mchp,spdifrx.yaml @@ -10,8 +10,8 @@ maintainers: - Codrin Ciubotariu description: - The Microchip Sony/Philips Digital Interface Receiver is a - serial port compliant with the IEC-60958 standard. + The Microchip Sony/Philips Digital Interface Receiver is a serial port + compliant with the IEC-60958 standard. properties: "#sound-dai-cells": diff --git a/Documentation/devicetree/bindings/sound/mchp,spdiftx.yaml b/Documentation/devicetree/bindings/sound/mchp,spdiftx.yaml index a03b0b871fc9..bdfb63387c53 100644 --- a/Documentation/devicetree/bindings/sound/mchp,spdiftx.yaml +++ b/Documentation/devicetree/bindings/sound/mchp,spdiftx.yaml @@ -10,8 +10,8 @@ maintainers: - Codrin Ciubotariu description: - The Microchip Sony/Philips Digital Interface Transmitter is a - serial port compliant with the IEC-60958 standard. + The Microchip Sony/Philips Digital Interface Transmitter is a serial port + compliant with the IEC-60958 standard. properties: "#sound-dai-cells": diff --git a/Documentation/devicetree/bindings/sound/qcom,lpass-cpu.yaml b/Documentation/devicetree/bindings/sound/qcom,lpass-cpu.yaml index f6f9fb49f385..1e23c0e20bc1 100644 --- a/Documentation/devicetree/bindings/sound/qcom,lpass-cpu.yaml +++ b/Documentation/devicetree/bindings/sound/qcom,lpass-cpu.yaml @@ -26,8 +26,10 @@ properties: reg: maxItems: 2 description: LPAIF core registers + reg-names: - maxItems: 2 + maxItems: 2 + clocks: minItems: 3 maxItems: 6 @@ -39,8 +41,10 @@ properties: interrupts: maxItems: 2 description: LPAIF DMA buffer interrupt + interrupt-names: maxItems: 2 + qcom,adsp: $ref: /schemas/types.yaml#/definitions/phandle description: Phandle for the audio DSP node @@ -141,31 +145,31 @@ allOf: properties: clock-names: oneOf: - - items: #for I2S - - const: pcnoc-sway-clk - - const: audio-core - - const: mclk0 - - const: pcnoc-mport-clk - - const: mi2s-bit-clk0 - - const: mi2s-bit-clk1 - - items: #for HDMI - - const: pcnoc-sway-clk - - const: audio-core - - const: pcnoc-mport-clk + - items: #for I2S + - const: pcnoc-sway-clk + - const: audio-core + - const: mclk0 + - const: pcnoc-mport-clk + - const: mi2s-bit-clk0 + - const: mi2s-bit-clk1 + - items: #for HDMI + - const: pcnoc-sway-clk + - const: audio-core + - const: pcnoc-mport-clk reg-names: anyOf: - items: #for I2S - - const: lpass-lpaif + - const: lpass-lpaif - items: #for I2S and HDMI - - const: lpass-hdmiif - - const: lpass-lpaif + - const: lpass-hdmiif + - const: lpass-lpaif interrupt-names: anyOf: - items: #for I2S - - const: lpass-irq-lpaif + - const: lpass-irq-lpaif - items: #for I2S and HDMI - - const: lpass-irq-lpaif - - const: lpass-irq-hdmi + - const: lpass-irq-lpaif + - const: lpass-irq-hdmi required: - iommus - power-domains diff --git a/Documentation/devicetree/bindings/timer/arm,sp804.yaml b/Documentation/devicetree/bindings/timer/arm,sp804.yaml index e35d3053250a..960e2bd66a97 100644 --- a/Documentation/devicetree/bindings/timer/arm,sp804.yaml +++ b/Documentation/devicetree/bindings/timer/arm,sp804.yaml @@ -33,8 +33,8 @@ properties: compatible: items: - enum: - - arm,sp804 - - hisilicon,sp804 + - arm,sp804 + - hisilicon,sp804 - const: arm,primecell interrupts: @@ -58,11 +58,11 @@ properties: clock is used for all clock inputs. oneOf: - items: - - description: clock for timer 1 - - description: clock for timer 2 - - description: bus clock + - description: clock for timer 1 + - description: clock for timer 2 + - description: bus clock - items: - - description: unified clock for both timers and the bus + - description: unified clock for both timers and the bus clock-names: true # The original binding did not specify any clock names, and there is no diff --git a/Documentation/devicetree/bindings/usb/cdns,usb3.yaml b/Documentation/devicetree/bindings/usb/cdns,usb3.yaml index ac20b98e9910..d6af2794d444 100644 --- a/Documentation/devicetree/bindings/usb/cdns,usb3.yaml +++ b/Documentation/devicetree/bindings/usb/cdns,usb3.yaml @@ -44,8 +44,8 @@ properties: enum: [super-speed, high-speed, full-speed] phys: - minItems: 1 - maxItems: 2 + minItems: 1 + maxItems: 2 phy-names: minItems: 1 diff --git a/Documentation/devicetree/bindings/usb/ti,hd3ss3220.yaml b/Documentation/devicetree/bindings/usb/ti,hd3ss3220.yaml index 5fe9e6211ba2..52ceb07294a3 100644 --- a/Documentation/devicetree/bindings/usb/ti,hd3ss3220.yaml +++ b/Documentation/devicetree/bindings/usb/ti,hd3ss3220.yaml @@ -17,7 +17,7 @@ description: |- properties: compatible: - const: ti,hd3ss3220 + const: ti,hd3ss3220 reg: maxItems: 1 diff --git a/Documentation/devicetree/bindings/w1/fsl-imx-owire.yaml b/Documentation/devicetree/bindings/w1/fsl-imx-owire.yaml index 1aaf3e768c81..55adea827c34 100644 --- a/Documentation/devicetree/bindings/w1/fsl-imx-owire.yaml +++ b/Documentation/devicetree/bindings/w1/fsl-imx-owire.yaml @@ -15,10 +15,10 @@ properties: - const: fsl,imx21-owire - items: - enum: - - fsl,imx27-owire - - fsl,imx50-owire - - fsl,imx51-owire - - fsl,imx53-owire + - fsl,imx27-owire + - fsl,imx50-owire + - fsl,imx51-owire + - fsl,imx53-owire - const: fsl,imx21-owire reg: -- cgit v1.2.3 From 62298364bd489b06d16370fd258c7be6a906729c Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 5 Oct 2020 13:38:29 -0500 Subject: dt-bindings: Explicitly allow additional properties in board/SoC schemas In order to add meta-schema checks for additional/unevaluatedProperties being present, all schema need to make this explicit. As the top-level board/SoC schemas always have additional properties, add 'additionalProperties: true'. Acked-by: Krzysztof Kozlowski Acked-by: Viresh Kumar Acked-by: Geert Uytterhoeven Signed-off-by: Rob Herring Link: https://lore.kernel.org/r/20201005183830.486085-4-robh@kernel.org Signed-off-by: Rob Herring --- Documentation/devicetree/bindings/arm/actions.yaml | 2 ++ Documentation/devicetree/bindings/arm/altera.yaml | 3 +++ Documentation/devicetree/bindings/arm/amazon,al.yaml | 2 ++ Documentation/devicetree/bindings/arm/amlogic.yaml | 3 +++ Documentation/devicetree/bindings/arm/arm,integrator.yaml | 2 ++ Documentation/devicetree/bindings/arm/arm,realview.yaml | 2 ++ Documentation/devicetree/bindings/arm/arm,versatile.yaml | 2 ++ Documentation/devicetree/bindings/arm/arm,vexpress-juno.yaml | 2 ++ Documentation/devicetree/bindings/arm/atmel-at91.yaml | 2 ++ Documentation/devicetree/bindings/arm/axxia.yaml | 2 ++ Documentation/devicetree/bindings/arm/bcm/bcm2835.yaml | 2 ++ Documentation/devicetree/bindings/arm/bcm/brcm,bcm11351.yaml | 2 ++ Documentation/devicetree/bindings/arm/bcm/brcm,bcm21664.yaml | 2 ++ Documentation/devicetree/bindings/arm/bcm/brcm,bcm23550.yaml | 2 ++ Documentation/devicetree/bindings/arm/bcm/brcm,bcm4708.yaml | 3 +++ Documentation/devicetree/bindings/arm/bcm/brcm,cygnus.yaml | 2 ++ Documentation/devicetree/bindings/arm/bcm/brcm,hr2.yaml | 2 ++ Documentation/devicetree/bindings/arm/bcm/brcm,ns2.yaml | 2 ++ Documentation/devicetree/bindings/arm/bcm/brcm,nsp.yaml | 2 ++ Documentation/devicetree/bindings/arm/bcm/brcm,stingray.yaml | 2 ++ Documentation/devicetree/bindings/arm/bcm/brcm,vulcan-soc.yaml | 2 ++ Documentation/devicetree/bindings/arm/bitmain.yaml | 3 +++ Documentation/devicetree/bindings/arm/calxeda.yaml | 2 ++ Documentation/devicetree/bindings/arm/digicolor.yaml | 2 ++ Documentation/devicetree/bindings/arm/fsl.yaml | 2 ++ Documentation/devicetree/bindings/arm/hisilicon/hisilicon.yaml | 3 +++ Documentation/devicetree/bindings/arm/intel,keembay.yaml | 3 +++ Documentation/devicetree/bindings/arm/intel-ixp4xx.yaml | 2 ++ Documentation/devicetree/bindings/arm/keystone/ti,k3-sci-common.yaml | 2 ++ Documentation/devicetree/bindings/arm/marvell/armada-7k-8k.yaml | 2 ++ Documentation/devicetree/bindings/arm/mediatek.yaml | 3 +++ Documentation/devicetree/bindings/arm/microchip,sparx5.yaml | 2 ++ Documentation/devicetree/bindings/arm/moxart.yaml | 1 + Documentation/devicetree/bindings/arm/mrvl/mrvl.yaml | 3 +++ Documentation/devicetree/bindings/arm/mstar/mstar.yaml | 2 ++ Documentation/devicetree/bindings/arm/nxp/lpc32xx.yaml | 2 ++ Documentation/devicetree/bindings/arm/qcom.yaml | 2 ++ Documentation/devicetree/bindings/arm/rda.yaml | 2 ++ Documentation/devicetree/bindings/arm/realtek.yaml | 3 +++ Documentation/devicetree/bindings/arm/renesas.yaml | 2 ++ Documentation/devicetree/bindings/arm/rockchip.yaml | 3 +++ Documentation/devicetree/bindings/arm/samsung/samsung-boards.yaml | 2 ++ Documentation/devicetree/bindings/arm/sirf.yaml | 3 +++ Documentation/devicetree/bindings/arm/socionext/milbeaut.yaml | 3 +++ Documentation/devicetree/bindings/arm/socionext/uniphier.yaml | 2 ++ Documentation/devicetree/bindings/arm/spear.yaml | 3 +++ Documentation/devicetree/bindings/arm/sprd/sprd.yaml | 2 ++ Documentation/devicetree/bindings/arm/sti.yaml | 3 +++ Documentation/devicetree/bindings/arm/stm32/stm32.yaml | 3 +++ Documentation/devicetree/bindings/arm/sunxi.yaml | 2 ++ Documentation/devicetree/bindings/arm/tegra.yaml | 2 ++ Documentation/devicetree/bindings/arm/ti/k3.yaml | 3 +++ Documentation/devicetree/bindings/arm/ti/nspire.yaml | 3 +++ Documentation/devicetree/bindings/arm/ti/ti,davinci.yaml | 3 +++ Documentation/devicetree/bindings/arm/toshiba.yaml | 3 +++ Documentation/devicetree/bindings/arm/ux500.yaml | 2 ++ Documentation/devicetree/bindings/arm/vt8500.yaml | 3 +++ Documentation/devicetree/bindings/arm/xilinx.yaml | 2 ++ Documentation/devicetree/bindings/arm/zte.yaml | 2 ++ Documentation/devicetree/bindings/mips/ingenic/devices.yaml | 3 +++ Documentation/devicetree/bindings/mips/loongson/devices.yaml | 3 +++ Documentation/devicetree/bindings/riscv/sifive.yaml | 3 +++ 62 files changed, 146 insertions(+) diff --git a/Documentation/devicetree/bindings/arm/actions.yaml b/Documentation/devicetree/bindings/arm/actions.yaml index fe22c66e9c15..02dc72c97645 100644 --- a/Documentation/devicetree/bindings/arm/actions.yaml +++ b/Documentation/devicetree/bindings/arm/actions.yaml @@ -49,3 +49,5 @@ properties: - enum: - ucrobotics,bubblegum-96 # uCRobotics Bubblegum-96 - const: actions,s900 + +additionalProperties: true diff --git a/Documentation/devicetree/bindings/arm/altera.yaml b/Documentation/devicetree/bindings/arm/altera.yaml index 0bc5020b7539..c15c92fdf2ed 100644 --- a/Documentation/devicetree/bindings/arm/altera.yaml +++ b/Documentation/devicetree/bindings/arm/altera.yaml @@ -19,4 +19,7 @@ properties: - altr,socfpga-arria5 - altr,socfpga-arria10 - const: altr,socfpga + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/amazon,al.yaml b/Documentation/devicetree/bindings/arm/amazon,al.yaml index a3a4d710bd02..0f03135d91b6 100644 --- a/Documentation/devicetree/bindings/arm/amazon,al.yaml +++ b/Documentation/devicetree/bindings/arm/amazon,al.yaml @@ -30,4 +30,6 @@ properties: - amazon,al-alpine-v3-evp - const: amazon,al-alpine-v3 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/amlogic.yaml b/Documentation/devicetree/bindings/arm/amlogic.yaml index 0ee7c5b7b3f6..3341788d1096 100644 --- a/Documentation/devicetree/bindings/arm/amlogic.yaml +++ b/Documentation/devicetree/bindings/arm/amlogic.yaml @@ -173,4 +173,7 @@ properties: - enum: - amlogic,ad401 - const: amlogic,a1 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/arm,integrator.yaml b/Documentation/devicetree/bindings/arm/arm,integrator.yaml index f0daf990e077..528eee64290a 100644 --- a/Documentation/devicetree/bindings/arm/arm,integrator.yaml +++ b/Documentation/devicetree/bindings/arm/arm,integrator.yaml @@ -83,4 +83,6 @@ required: - compatible - core-module@10000000 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/arm,realview.yaml b/Documentation/devicetree/bindings/arm/arm,realview.yaml index 1d0b4e2bc7d2..4f9b21f49e84 100644 --- a/Documentation/devicetree/bindings/arm/arm,realview.yaml +++ b/Documentation/devicetree/bindings/arm/arm,realview.yaml @@ -120,4 +120,6 @@ required: - compatible - soc +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/arm,versatile.yaml b/Documentation/devicetree/bindings/arm/arm,versatile.yaml index 06efd2a075c9..34b437c72751 100644 --- a/Documentation/devicetree/bindings/arm/arm,versatile.yaml +++ b/Documentation/devicetree/bindings/arm/arm,versatile.yaml @@ -68,4 +68,6 @@ required: - compatible - core-module@10000000 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/arm,vexpress-juno.yaml b/Documentation/devicetree/bindings/arm/arm,vexpress-juno.yaml index 26829a803fda..55ef656d1192 100644 --- a/Documentation/devicetree/bindings/arm/arm,vexpress-juno.yaml +++ b/Documentation/devicetree/bindings/arm/arm,vexpress-juno.yaml @@ -216,4 +216,6 @@ allOf: required: - arm,hbi +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/atmel-at91.yaml b/Documentation/devicetree/bindings/arm/atmel-at91.yaml index 614c91956798..6fc5a22ad962 100644 --- a/Documentation/devicetree/bindings/arm/atmel-at91.yaml +++ b/Documentation/devicetree/bindings/arm/atmel-at91.yaml @@ -184,4 +184,6 @@ properties: - const: atmel,samv71 - const: atmel,samv7 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/axxia.yaml b/Documentation/devicetree/bindings/arm/axxia.yaml index 3ea5f2fdcd96..e0d2bb71cf50 100644 --- a/Documentation/devicetree/bindings/arm/axxia.yaml +++ b/Documentation/devicetree/bindings/arm/axxia.yaml @@ -18,4 +18,6 @@ properties: - const: lsi,axm5516-amarillo - const: lsi,axm5516 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/bcm/bcm2835.yaml b/Documentation/devicetree/bindings/arm/bcm/bcm2835.yaml index dd52e29b0642..812ae8cc5959 100644 --- a/Documentation/devicetree/bindings/arm/bcm/bcm2835.yaml +++ b/Documentation/devicetree/bindings/arm/bcm/bcm2835.yaml @@ -51,4 +51,6 @@ properties: - raspberrypi,3-compute-module-lite - const: brcm,bcm2837 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/bcm/brcm,bcm11351.yaml b/Documentation/devicetree/bindings/arm/bcm/brcm,bcm11351.yaml index 497600a2ffb9..c60324357435 100644 --- a/Documentation/devicetree/bindings/arm/bcm/brcm,bcm11351.yaml +++ b/Documentation/devicetree/bindings/arm/bcm/brcm,bcm11351.yaml @@ -18,4 +18,6 @@ properties: - brcm,bcm28155-ap - const: brcm,bcm11351 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/bcm/brcm,bcm21664.yaml b/Documentation/devicetree/bindings/arm/bcm/brcm,bcm21664.yaml index e0ee931723dc..b3020757380f 100644 --- a/Documentation/devicetree/bindings/arm/bcm/brcm,bcm21664.yaml +++ b/Documentation/devicetree/bindings/arm/bcm/brcm,bcm21664.yaml @@ -18,4 +18,6 @@ properties: - brcm,bcm21664-garnet - const: brcm,bcm21664 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/bcm/brcm,bcm23550.yaml b/Documentation/devicetree/bindings/arm/bcm/brcm,bcm23550.yaml index 40d12ea56e54..37f3a6fcde76 100644 --- a/Documentation/devicetree/bindings/arm/bcm/brcm,bcm23550.yaml +++ b/Documentation/devicetree/bindings/arm/bcm/brcm,bcm23550.yaml @@ -18,4 +18,6 @@ properties: - brcm,bcm23550-sparrow - const: brcm,bcm23550 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/bcm/brcm,bcm4708.yaml b/Documentation/devicetree/bindings/arm/bcm/brcm,bcm4708.yaml index 988e0bbb2a62..434d3c6db61e 100644 --- a/Documentation/devicetree/bindings/arm/bcm/brcm,bcm4708.yaml +++ b/Documentation/devicetree/bindings/arm/bcm/brcm,bcm4708.yaml @@ -87,4 +87,7 @@ properties: - const: brcm,brcm53012 - const: brcm,brcm53016 - const: brcm,bcm4708 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/bcm/brcm,cygnus.yaml b/Documentation/devicetree/bindings/arm/bcm/brcm,cygnus.yaml index 9ba7b16e1fc4..432ccf990f9e 100644 --- a/Documentation/devicetree/bindings/arm/bcm/brcm,cygnus.yaml +++ b/Documentation/devicetree/bindings/arm/bcm/brcm,cygnus.yaml @@ -26,4 +26,6 @@ properties: - brcm,bcm58305 - const: brcm,cygnus +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/bcm/brcm,hr2.yaml b/Documentation/devicetree/bindings/arm/bcm/brcm,hr2.yaml index ae614b6722c2..294948399f82 100644 --- a/Documentation/devicetree/bindings/arm/bcm/brcm,hr2.yaml +++ b/Documentation/devicetree/bindings/arm/bcm/brcm,hr2.yaml @@ -25,4 +25,6 @@ properties: - const: brcm,bcm53342 - const: brcm,hr2 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/bcm/brcm,ns2.yaml b/Documentation/devicetree/bindings/arm/bcm/brcm,ns2.yaml index 0749adf94c28..c4847abbecd8 100644 --- a/Documentation/devicetree/bindings/arm/bcm/brcm,ns2.yaml +++ b/Documentation/devicetree/bindings/arm/bcm/brcm,ns2.yaml @@ -20,4 +20,6 @@ properties: - brcm,ns2-xmc - const: brcm,ns2 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/bcm/brcm,nsp.yaml b/Documentation/devicetree/bindings/arm/bcm/brcm,nsp.yaml index 8c2cacb2bb4f..476bc23a7f75 100644 --- a/Documentation/devicetree/bindings/arm/bcm/brcm,nsp.yaml +++ b/Documentation/devicetree/bindings/arm/bcm/brcm,nsp.yaml @@ -33,4 +33,6 @@ properties: - brcm,bcm88312 - const: brcm,nsp +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/bcm/brcm,stingray.yaml b/Documentation/devicetree/bindings/arm/bcm/brcm,stingray.yaml index c13cb96545a2..c638e04ebae0 100644 --- a/Documentation/devicetree/bindings/arm/bcm/brcm,stingray.yaml +++ b/Documentation/devicetree/bindings/arm/bcm/brcm,stingray.yaml @@ -21,4 +21,6 @@ properties: - brcm,bcm958802a802x - const: brcm,stingray +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/bcm/brcm,vulcan-soc.yaml b/Documentation/devicetree/bindings/arm/bcm/brcm,vulcan-soc.yaml index ccdf9f99cb2b..4eba182abd53 100644 --- a/Documentation/devicetree/bindings/arm/bcm/brcm,vulcan-soc.yaml +++ b/Documentation/devicetree/bindings/arm/bcm/brcm,vulcan-soc.yaml @@ -19,4 +19,6 @@ properties: - cavium,thunderx2-cn9900 - const: brcm,vulcan-soc +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/bitmain.yaml b/Documentation/devicetree/bindings/arm/bitmain.yaml index 5880083ab8d0..90ba02be48ce 100644 --- a/Documentation/devicetree/bindings/arm/bitmain.yaml +++ b/Documentation/devicetree/bindings/arm/bitmain.yaml @@ -17,4 +17,7 @@ properties: - enum: - bitmain,sophon-edge - const: bitmain,bm1880 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/calxeda.yaml b/Documentation/devicetree/bindings/arm/calxeda.yaml index aa5571d23c39..46f78addebb0 100644 --- a/Documentation/devicetree/bindings/arm/calxeda.yaml +++ b/Documentation/devicetree/bindings/arm/calxeda.yaml @@ -20,3 +20,5 @@ properties: - enum: - calxeda,highbank - calxeda,ecx-2000 + +additionalProperties: true diff --git a/Documentation/devicetree/bindings/arm/digicolor.yaml b/Documentation/devicetree/bindings/arm/digicolor.yaml index 849e20518339..a35de3c9e284 100644 --- a/Documentation/devicetree/bindings/arm/digicolor.yaml +++ b/Documentation/devicetree/bindings/arm/digicolor.yaml @@ -15,4 +15,6 @@ properties: compatible: const: cnxt,cx92755 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/fsl.yaml b/Documentation/devicetree/bindings/arm/fsl.yaml index 1ca9dfa8ce9a..934289446abb 100644 --- a/Documentation/devicetree/bindings/arm/fsl.yaml +++ b/Documentation/devicetree/bindings/arm/fsl.yaml @@ -621,4 +621,6 @@ properties: - fsl,s32v234-evb # S32V234-EVB2 Customer Evaluation Board - const: fsl,s32v234 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/hisilicon/hisilicon.yaml b/Documentation/devicetree/bindings/arm/hisilicon/hisilicon.yaml index 43b8ce2227aa..b38458022946 100644 --- a/Documentation/devicetree/bindings/arm/hisilicon/hisilicon.yaml +++ b/Documentation/devicetree/bindings/arm/hisilicon/hisilicon.yaml @@ -64,4 +64,7 @@ properties: items: - const: H836ASDJ - const: hisilicon,sd5203 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/intel,keembay.yaml b/Documentation/devicetree/bindings/arm/intel,keembay.yaml index 06a7b05f435f..69cd30872928 100644 --- a/Documentation/devicetree/bindings/arm/intel,keembay.yaml +++ b/Documentation/devicetree/bindings/arm/intel,keembay.yaml @@ -16,4 +16,7 @@ properties: - enum: - intel,keembay-evm - const: intel,keembay + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/intel-ixp4xx.yaml b/Documentation/devicetree/bindings/arm/intel-ixp4xx.yaml index f18302efb90e..d72e92bdf7c1 100644 --- a/Documentation/devicetree/bindings/arm/intel-ixp4xx.yaml +++ b/Documentation/devicetree/bindings/arm/intel-ixp4xx.yaml @@ -22,3 +22,5 @@ properties: - enum: - gateworks,gw2358 - const: intel,ixp43x + +additionalProperties: true diff --git a/Documentation/devicetree/bindings/arm/keystone/ti,k3-sci-common.yaml b/Documentation/devicetree/bindings/arm/keystone/ti,k3-sci-common.yaml index 7597bc93a55f..5cbcacaeb441 100644 --- a/Documentation/devicetree/bindings/arm/keystone/ti,k3-sci-common.yaml +++ b/Documentation/devicetree/bindings/arm/keystone/ti,k3-sci-common.yaml @@ -42,3 +42,5 @@ properties: - description: TI-SCI processor id for the remote processor device - description: TI-SCI host id to which processor control ownership should be transferred to + +additionalProperties: true diff --git a/Documentation/devicetree/bindings/arm/marvell/armada-7k-8k.yaml b/Documentation/devicetree/bindings/arm/marvell/armada-7k-8k.yaml index a9828c50c0fb..e9bf3054529f 100644 --- a/Documentation/devicetree/bindings/arm/marvell/armada-7k-8k.yaml +++ b/Documentation/devicetree/bindings/arm/marvell/armada-7k-8k.yaml @@ -59,3 +59,5 @@ properties: - const: marvell,cn9130 - const: marvell,armada-ap807-quad - const: marvell,armada-ap807 + +additionalProperties: true diff --git a/Documentation/devicetree/bindings/arm/mediatek.yaml b/Documentation/devicetree/bindings/arm/mediatek.yaml index 30908963ae26..f736e8c859fa 100644 --- a/Documentation/devicetree/bindings/arm/mediatek.yaml +++ b/Documentation/devicetree/bindings/arm/mediatek.yaml @@ -119,4 +119,7 @@ properties: - const: google,krane-sku176 - const: google,krane - const: mediatek,mt8183 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/microchip,sparx5.yaml b/Documentation/devicetree/bindings/arm/microchip,sparx5.yaml index ecf6fa12e6ad..6193388c6318 100644 --- a/Documentation/devicetree/bindings/arm/microchip,sparx5.yaml +++ b/Documentation/devicetree/bindings/arm/microchip,sparx5.yaml @@ -62,4 +62,6 @@ required: - compatible - axi@600000000 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/moxart.yaml b/Documentation/devicetree/bindings/arm/moxart.yaml index c068df59fad2..670d24ce8ec5 100644 --- a/Documentation/devicetree/bindings/arm/moxart.yaml +++ b/Documentation/devicetree/bindings/arm/moxart.yaml @@ -16,4 +16,5 @@ properties: - const: moxa,moxart-uc-7112-lx - const: moxa,moxart +additionalProperties: true ... diff --git a/Documentation/devicetree/bindings/arm/mrvl/mrvl.yaml b/Documentation/devicetree/bindings/arm/mrvl/mrvl.yaml index 3235ec9e9bad..d58116136154 100644 --- a/Documentation/devicetree/bindings/arm/mrvl/mrvl.yaml +++ b/Documentation/devicetree/bindings/arm/mrvl/mrvl.yaml @@ -35,4 +35,7 @@ properties: - enum: - dell,wyse-ariel - const: marvell,mmp3 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/mstar/mstar.yaml b/Documentation/devicetree/bindings/arm/mstar/mstar.yaml index c2f980b00b06..7c787405bb2f 100644 --- a/Documentation/devicetree/bindings/arm/mstar/mstar.yaml +++ b/Documentation/devicetree/bindings/arm/mstar/mstar.yaml @@ -31,3 +31,5 @@ properties: - enum: - 70mai,midrived08 # 70mai midrive d08 - const: mstar,mercury5 + +additionalProperties: true diff --git a/Documentation/devicetree/bindings/arm/nxp/lpc32xx.yaml b/Documentation/devicetree/bindings/arm/nxp/lpc32xx.yaml index f7f024910e71..214c97bc3063 100644 --- a/Documentation/devicetree/bindings/arm/nxp/lpc32xx.yaml +++ b/Documentation/devicetree/bindings/arm/nxp/lpc32xx.yaml @@ -21,4 +21,6 @@ properties: - ea,ea3250 - phytec,phy3250 - const: nxp,lpc3250 + +additionalProperties: true ... diff --git a/Documentation/devicetree/bindings/arm/qcom.yaml b/Documentation/devicetree/bindings/arm/qcom.yaml index ad25deba4d86..c97d4a580f47 100644 --- a/Documentation/devicetree/bindings/arm/qcom.yaml +++ b/Documentation/devicetree/bindings/arm/qcom.yaml @@ -178,4 +178,6 @@ properties: - qcom,sm8250-mtp - const: qcom,sm8250 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/rda.yaml b/Documentation/devicetree/bindings/arm/rda.yaml index 9672aa0c760d..a5c0444aa2b4 100644 --- a/Documentation/devicetree/bindings/arm/rda.yaml +++ b/Documentation/devicetree/bindings/arm/rda.yaml @@ -19,4 +19,6 @@ properties: - xunlong,orangepi-i96 # Orange Pi i96 - const: rda,8810pl +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/realtek.yaml b/Documentation/devicetree/bindings/arm/realtek.yaml index 845f9c76d6f7..9fb0297fe1ce 100644 --- a/Documentation/devicetree/bindings/arm/realtek.yaml +++ b/Documentation/devicetree/bindings/arm/realtek.yaml @@ -54,4 +54,7 @@ properties: - enum: - realtek,mjolnir # Realtek Mjolnir EVB - const: realtek,rtd1619 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/renesas.yaml b/Documentation/devicetree/bindings/arm/renesas.yaml index 01a6d0c571ad..ff94c45eefb0 100644 --- a/Documentation/devicetree/bindings/arm/renesas.yaml +++ b/Documentation/devicetree/bindings/arm/renesas.yaml @@ -299,4 +299,6 @@ properties: - renesas,rzn1d400-db # RZN1D-DB (RZ/N1D Demo Board for the RZ/N1D 400 pins package) - const: renesas,r9a06g032 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/rockchip.yaml b/Documentation/devicetree/bindings/arm/rockchip.yaml index 65b4cc2c63f7..b621752aaa65 100644 --- a/Documentation/devicetree/bindings/arm/rockchip.yaml +++ b/Documentation/devicetree/bindings/arm/rockchip.yaml @@ -569,4 +569,7 @@ properties: items: - const: zkmagic,a95x-z2 - const: rockchip,rk3318 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/samsung/samsung-boards.yaml b/Documentation/devicetree/bindings/arm/samsung/samsung-boards.yaml index eb92f9eefaba..272508010b02 100644 --- a/Documentation/devicetree/bindings/arm/samsung/samsung-boards.yaml +++ b/Documentation/devicetree/bindings/arm/samsung/samsung-boards.yaml @@ -180,3 +180,5 @@ properties: required: - compatible + +additionalProperties: true diff --git a/Documentation/devicetree/bindings/arm/sirf.yaml b/Documentation/devicetree/bindings/arm/sirf.yaml index 0b597032c923..b25eb35d1b66 100644 --- a/Documentation/devicetree/bindings/arm/sirf.yaml +++ b/Documentation/devicetree/bindings/arm/sirf.yaml @@ -24,4 +24,7 @@ properties: - items: - const: sirf,prima2-cb - const: sirf,prima2 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/socionext/milbeaut.yaml b/Documentation/devicetree/bindings/arm/socionext/milbeaut.yaml index 2bd519d2e855..aa1d4afbc510 100644 --- a/Documentation/devicetree/bindings/arm/socionext/milbeaut.yaml +++ b/Documentation/devicetree/bindings/arm/socionext/milbeaut.yaml @@ -19,4 +19,7 @@ properties: - enum: - socionext,milbeaut-m10v-evb - const: socionext,sc2000a + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/socionext/uniphier.yaml b/Documentation/devicetree/bindings/arm/socionext/uniphier.yaml index 6caf1f9be390..8c0e91658474 100644 --- a/Documentation/devicetree/bindings/arm/socionext/uniphier.yaml +++ b/Documentation/devicetree/bindings/arm/socionext/uniphier.yaml @@ -60,3 +60,5 @@ properties: - enum: - socionext,uniphier-pxs3-ref - const: socionext,uniphier-pxs3 + +additionalProperties: true diff --git a/Documentation/devicetree/bindings/arm/spear.yaml b/Documentation/devicetree/bindings/arm/spear.yaml index f6ec731c9531..605ad3f882ef 100644 --- a/Documentation/devicetree/bindings/arm/spear.yaml +++ b/Documentation/devicetree/bindings/arm/spear.yaml @@ -22,4 +22,7 @@ properties: - st,spear320 - st,spear1310 - st,spear1340 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/sprd/sprd.yaml b/Documentation/devicetree/bindings/arm/sprd/sprd.yaml index 0258a96bfbde..7b6ae3070396 100644 --- a/Documentation/devicetree/bindings/arm/sprd/sprd.yaml +++ b/Documentation/devicetree/bindings/arm/sprd/sprd.yaml @@ -30,4 +30,6 @@ properties: - sprd,sp9863a-1h10 - const: sprd,sc9863a +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/sti.yaml b/Documentation/devicetree/bindings/arm/sti.yaml index 47f9b8eebaa0..b1f28d16d3fb 100644 --- a/Documentation/devicetree/bindings/arm/sti.yaml +++ b/Documentation/devicetree/bindings/arm/sti.yaml @@ -20,4 +20,7 @@ properties: - st,stih407 - st,stih410 - st,stih418 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/stm32/stm32.yaml b/Documentation/devicetree/bindings/arm/stm32/stm32.yaml index 6b87314eb9ed..009b424e456e 100644 --- a/Documentation/devicetree/bindings/arm/stm32/stm32.yaml +++ b/Documentation/devicetree/bindings/arm/stm32/stm32.yaml @@ -58,4 +58,7 @@ properties: - seeed,stm32mp157c-odyssey - const: seeed,stm32mp157c-odyssey-som - const: st,stm32mp157 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/sunxi.yaml b/Documentation/devicetree/bindings/arm/sunxi.yaml index afa00268c7db..cab8e1b6417b 100644 --- a/Documentation/devicetree/bindings/arm/sunxi.yaml +++ b/Documentation/devicetree/bindings/arm/sunxi.yaml @@ -893,3 +893,5 @@ properties: items: - const: xunlong,orangepi-zero-plus2-h3 - const: allwinner,sun8i-h3 + +additionalProperties: true diff --git a/Documentation/devicetree/bindings/arm/tegra.yaml b/Documentation/devicetree/bindings/arm/tegra.yaml index 8ae44948e873..767e86354c8e 100644 --- a/Documentation/devicetree/bindings/arm/tegra.yaml +++ b/Documentation/devicetree/bindings/arm/tegra.yaml @@ -125,3 +125,5 @@ properties: - enum: - nvidia,tegra234-vdk - const: nvidia,tegra234 + +additionalProperties: true diff --git a/Documentation/devicetree/bindings/arm/ti/k3.yaml b/Documentation/devicetree/bindings/arm/ti/k3.yaml index 829751209543..c6e1c1e63e43 100644 --- a/Documentation/devicetree/bindings/arm/ti/k3.yaml +++ b/Documentation/devicetree/bindings/arm/ti/k3.yaml @@ -32,4 +32,7 @@ properties: - description: K3 J7200 SoC items: - const: ti,j7200 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/ti/nspire.yaml b/Documentation/devicetree/bindings/arm/ti/nspire.yaml index e372b43da62f..cc2023bb7fa6 100644 --- a/Documentation/devicetree/bindings/arm/ti/nspire.yaml +++ b/Documentation/devicetree/bindings/arm/ti/nspire.yaml @@ -21,4 +21,7 @@ properties: - ti,nspire-tp # Clickpad models - ti,nspire-clp + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/ti/ti,davinci.yaml b/Documentation/devicetree/bindings/arm/ti/ti,davinci.yaml index a8765ba29476..c022d325fc08 100644 --- a/Documentation/devicetree/bindings/arm/ti/ti,davinci.yaml +++ b/Documentation/devicetree/bindings/arm/ti/ti,davinci.yaml @@ -23,4 +23,7 @@ properties: - enbw,cmc # EnBW AM1808 based CMC board - lego,ev3 # LEGO MINDSTORMS EV3 (AM1808 based) - const: ti,da850 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/toshiba.yaml b/Documentation/devicetree/bindings/arm/toshiba.yaml index 0e066290238e..001bbbcd1432 100644 --- a/Documentation/devicetree/bindings/arm/toshiba.yaml +++ b/Documentation/devicetree/bindings/arm/toshiba.yaml @@ -19,4 +19,7 @@ properties: - enum: - toshiba,tmpv7708-rm-mbrc # TMPV7708 RM main board - const: toshiba,tmpv7708 + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/ux500.yaml b/Documentation/devicetree/bindings/arm/ux500.yaml index accaee906050..5db7cfba81a4 100644 --- a/Documentation/devicetree/bindings/arm/ux500.yaml +++ b/Documentation/devicetree/bindings/arm/ux500.yaml @@ -34,3 +34,5 @@ properties: items: - const: samsung,golden - const: st-ericsson,u8500 + +additionalProperties: true diff --git a/Documentation/devicetree/bindings/arm/vt8500.yaml b/Documentation/devicetree/bindings/arm/vt8500.yaml index 7b25b6fa34e9..29ff399551ca 100644 --- a/Documentation/devicetree/bindings/arm/vt8500.yaml +++ b/Documentation/devicetree/bindings/arm/vt8500.yaml @@ -21,3 +21,6 @@ properties: - wm,wm8650 - wm,wm8750 - wm,wm8850 + +additionalProperties: true + diff --git a/Documentation/devicetree/bindings/arm/xilinx.yaml b/Documentation/devicetree/bindings/arm/xilinx.yaml index c73b1f5c7f49..e0c6787f6e94 100644 --- a/Documentation/devicetree/bindings/arm/xilinx.yaml +++ b/Documentation/devicetree/bindings/arm/xilinx.yaml @@ -111,4 +111,6 @@ properties: - const: xlnx,zynqmp-zcu111 - const: xlnx,zynqmp +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/arm/zte.yaml b/Documentation/devicetree/bindings/arm/zte.yaml index 2d3fefdccdff..672f8129cd31 100644 --- a/Documentation/devicetree/bindings/arm/zte.yaml +++ b/Documentation/devicetree/bindings/arm/zte.yaml @@ -23,4 +23,6 @@ properties: - zte,zx296718-evb - const: zte,zx296718 +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/mips/ingenic/devices.yaml b/Documentation/devicetree/bindings/mips/ingenic/devices.yaml index dc21b4630c25..ee00d414df10 100644 --- a/Documentation/devicetree/bindings/mips/ingenic/devices.yaml +++ b/Documentation/devicetree/bindings/mips/ingenic/devices.yaml @@ -52,4 +52,7 @@ properties: items: - const: yna,cu2000-neo - const: ingenic,x2000e + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/mips/loongson/devices.yaml b/Documentation/devicetree/bindings/mips/loongson/devices.yaml index d25e80aa8b2a..9fee6708e6f5 100644 --- a/Documentation/devicetree/bindings/mips/loongson/devices.yaml +++ b/Documentation/devicetree/bindings/mips/loongson/devices.yaml @@ -36,4 +36,7 @@ properties: - description: Virtual Loongson64 Quad Core + VirtIO items: - const: loongson,loongson64v-4core-virtio + +additionalProperties: true + ... diff --git a/Documentation/devicetree/bindings/riscv/sifive.yaml b/Documentation/devicetree/bindings/riscv/sifive.yaml index 3ab532713dc1..3a8647d1da4c 100644 --- a/Documentation/devicetree/bindings/riscv/sifive.yaml +++ b/Documentation/devicetree/bindings/riscv/sifive.yaml @@ -22,4 +22,7 @@ properties: - sifive,hifive-unleashed-a00 - const: sifive,fu540-c000 - const: sifive,fu540 + +additionalProperties: true + ... -- cgit v1.2.3 From f84e2c5c528d937564e4a9e3411418e4c914b1fb Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Tue, 6 Oct 2020 12:19:08 -0500 Subject: dt-bindings: Another round of adding missing 'additionalProperties/unevalutatedProperties' Another round of wack-a-mole. The json-schema default is additional unknown properties are allowed, but for DT all properties should be defined. Signed-off-by: Rob Herring --- .../devicetree/bindings/display/bridge/toshiba,tc358775.yaml | 2 ++ Documentation/devicetree/bindings/edac/amazon,al-mc-edac.yaml | 1 + Documentation/devicetree/bindings/iio/adc/lltc,ltc2497.yaml | 2 ++ Documentation/devicetree/bindings/iio/humidity/ti,hdc2010.yaml | 2 ++ .../devicetree/bindings/interrupt-controller/ti,sci-inta.yaml | 2 ++ .../devicetree/bindings/interrupt-controller/ti,sci-intr.yaml | 2 ++ Documentation/devicetree/bindings/leds/backlight/common.yaml | 2 ++ Documentation/devicetree/bindings/leds/leds-lp50xx.yaml | 8 ++++++++ .../devicetree/bindings/mmc/microchip,dw-sparx5-sdhci.yaml | 2 ++ Documentation/devicetree/bindings/mmc/sdhci-am654.yaml | 2 ++ Documentation/devicetree/bindings/net/intel,dwmac-plat.yaml | 2 ++ Documentation/devicetree/bindings/net/ti,dp83822.yaml | 2 ++ Documentation/devicetree/bindings/phy/ti,omap-usb2.yaml | 2 ++ Documentation/devicetree/bindings/power/reset/reboot-mode.yaml | 2 ++ Documentation/devicetree/bindings/regulator/mps,mp886x.yaml | 2 ++ Documentation/devicetree/bindings/regulator/pfuze100.yaml | 2 ++ Documentation/devicetree/bindings/sound/realtek,rt1015p.yaml | 2 ++ 17 files changed, 39 insertions(+) diff --git a/Documentation/devicetree/bindings/display/bridge/toshiba,tc358775.yaml b/Documentation/devicetree/bindings/display/bridge/toshiba,tc358775.yaml index e51d5cccc0e9..fd3113aa9ccd 100644 --- a/Documentation/devicetree/bindings/display/bridge/toshiba,tc358775.yaml +++ b/Documentation/devicetree/bindings/display/bridge/toshiba,tc358775.yaml @@ -85,6 +85,8 @@ required: - reset-gpios - ports +additionalProperties: false + examples: - | #include diff --git a/Documentation/devicetree/bindings/edac/amazon,al-mc-edac.yaml b/Documentation/devicetree/bindings/edac/amazon,al-mc-edac.yaml index a25387df0865..57e5270a0741 100644 --- a/Documentation/devicetree/bindings/edac/amazon,al-mc-edac.yaml +++ b/Documentation/devicetree/bindings/edac/amazon,al-mc-edac.yaml @@ -48,6 +48,7 @@ required: - "#address-cells" - "#size-cells" +additionalProperties: false examples: - | diff --git a/Documentation/devicetree/bindings/iio/adc/lltc,ltc2497.yaml b/Documentation/devicetree/bindings/iio/adc/lltc,ltc2497.yaml index 6a176f551d75..c1772b568cd1 100644 --- a/Documentation/devicetree/bindings/iio/adc/lltc,ltc2497.yaml +++ b/Documentation/devicetree/bindings/iio/adc/lltc,ltc2497.yaml @@ -28,6 +28,8 @@ required: - reg - vref-supply +additionalProperties: false + examples: - | i2c { diff --git a/Documentation/devicetree/bindings/iio/humidity/ti,hdc2010.yaml b/Documentation/devicetree/bindings/iio/humidity/ti,hdc2010.yaml index dc870eb2875f..7037f82ec753 100644 --- a/Documentation/devicetree/bindings/iio/humidity/ti,hdc2010.yaml +++ b/Documentation/devicetree/bindings/iio/humidity/ti,hdc2010.yaml @@ -32,6 +32,8 @@ required: - compatible - reg +additionalProperties: false + examples: - | i2c0 { diff --git a/Documentation/devicetree/bindings/interrupt-controller/ti,sci-inta.yaml b/Documentation/devicetree/bindings/interrupt-controller/ti,sci-inta.yaml index c7cd05656a3e..f6c3fcc4bdfd 100644 --- a/Documentation/devicetree/bindings/interrupt-controller/ti,sci-inta.yaml +++ b/Documentation/devicetree/bindings/interrupt-controller/ti,sci-inta.yaml @@ -79,6 +79,8 @@ required: - ti,sci-dev-id - ti,interrupt-ranges +unevaluatedProperties: false + examples: - | bus { diff --git a/Documentation/devicetree/bindings/interrupt-controller/ti,sci-intr.yaml b/Documentation/devicetree/bindings/interrupt-controller/ti,sci-intr.yaml index cff6a956afb4..e12aee42b126 100644 --- a/Documentation/devicetree/bindings/interrupt-controller/ti,sci-intr.yaml +++ b/Documentation/devicetree/bindings/interrupt-controller/ti,sci-intr.yaml @@ -88,6 +88,8 @@ required: - ti,sci-dev-id - ti,interrupt-ranges +unevaluatedProperties: false + examples: - | main_gpio_intr: interrupt-controller0 { diff --git a/Documentation/devicetree/bindings/leds/backlight/common.yaml b/Documentation/devicetree/bindings/leds/backlight/common.yaml index 4e7e95e331a5..bc817f77d2b1 100644 --- a/Documentation/devicetree/bindings/leds/backlight/common.yaml +++ b/Documentation/devicetree/bindings/leds/backlight/common.yaml @@ -32,3 +32,5 @@ properties: that a LED can be made so bright that it gets damaged or causes damage due to restrictions in a specific system, such as mounting conditions. $ref: /schemas/types.yaml#definitions/uint32 + +additionalProperties: true diff --git a/Documentation/devicetree/bindings/leds/leds-lp50xx.yaml b/Documentation/devicetree/bindings/leds/leds-lp50xx.yaml index 947542a253ec..c192b5feadc7 100644 --- a/Documentation/devicetree/bindings/leds/leds-lp50xx.yaml +++ b/Documentation/devicetree/bindings/leds/leds-lp50xx.yaml @@ -46,6 +46,12 @@ properties: vled-supply: description: LED supply. + '#address-cells': + const: 1 + + '#size-cells': + const: 0 + patternProperties: '^multi-led@[0-9a-f]$': type: object @@ -69,6 +75,8 @@ required: - compatible - reg +additionalProperties: false + examples: - | #include diff --git a/Documentation/devicetree/bindings/mmc/microchip,dw-sparx5-sdhci.yaml b/Documentation/devicetree/bindings/mmc/microchip,dw-sparx5-sdhci.yaml index 55883290543b..69ff065c9a39 100644 --- a/Documentation/devicetree/bindings/mmc/microchip,dw-sparx5-sdhci.yaml +++ b/Documentation/devicetree/bindings/mmc/microchip,dw-sparx5-sdhci.yaml @@ -46,6 +46,8 @@ required: - clocks - clock-names +unevaluatedProperties: false + examples: - | #include diff --git a/Documentation/devicetree/bindings/mmc/sdhci-am654.yaml b/Documentation/devicetree/bindings/mmc/sdhci-am654.yaml index 2cd859066462..1ae945434c53 100644 --- a/Documentation/devicetree/bindings/mmc/sdhci-am654.yaml +++ b/Documentation/devicetree/bindings/mmc/sdhci-am654.yaml @@ -186,6 +186,8 @@ required: - clock-names - ti,otap-del-sel-legacy +unevaluatedProperties: false + examples: - | #include diff --git a/Documentation/devicetree/bindings/net/intel,dwmac-plat.yaml b/Documentation/devicetree/bindings/net/intel,dwmac-plat.yaml index fa3ebba4e635..c1948ce00081 100644 --- a/Documentation/devicetree/bindings/net/intel,dwmac-plat.yaml +++ b/Documentation/devicetree/bindings/net/intel,dwmac-plat.yaml @@ -46,6 +46,8 @@ required: - clocks - clock-names +unevaluatedProperties: false + examples: # FIXME: Remove defines and include the correct header file # once it is available in mainline. diff --git a/Documentation/devicetree/bindings/net/ti,dp83822.yaml b/Documentation/devicetree/bindings/net/ti,dp83822.yaml index 55913534cbc2..75e8712e903a 100644 --- a/Documentation/devicetree/bindings/net/ti,dp83822.yaml +++ b/Documentation/devicetree/bindings/net/ti,dp83822.yaml @@ -65,6 +65,8 @@ properties: required: - reg +unevaluatedProperties: false + examples: - | mdio0 { diff --git a/Documentation/devicetree/bindings/phy/ti,omap-usb2.yaml b/Documentation/devicetree/bindings/phy/ti,omap-usb2.yaml index ab77a65c5c55..83d5d0aceb04 100644 --- a/Documentation/devicetree/bindings/phy/ti,omap-usb2.yaml +++ b/Documentation/devicetree/bindings/phy/ti,omap-usb2.yaml @@ -62,6 +62,8 @@ required: - clocks - clock-names +additionalProperties: false + examples: - | usb0_phy: phy@4100000 { diff --git a/Documentation/devicetree/bindings/power/reset/reboot-mode.yaml b/Documentation/devicetree/bindings/power/reset/reboot-mode.yaml index a2dca38577f7..9c6fda6b1dd9 100644 --- a/Documentation/devicetree/bindings/power/reset/reboot-mode.yaml +++ b/Documentation/devicetree/bindings/power/reset/reboot-mode.yaml @@ -36,6 +36,8 @@ patternProperties: "^mode-.*$": $ref: /schemas/types.yaml#/definitions/uint32 +additionalProperties: false + examples: - | reboot-mode { diff --git a/Documentation/devicetree/bindings/regulator/mps,mp886x.yaml b/Documentation/devicetree/bindings/regulator/mps,mp886x.yaml index ba175b30f468..9245b7199439 100644 --- a/Documentation/devicetree/bindings/regulator/mps,mp886x.yaml +++ b/Documentation/devicetree/bindings/regulator/mps,mp886x.yaml @@ -41,6 +41,8 @@ required: - enable-gpios - mps,fb-voltage-divider +unevaluatedProperties: false + examples: - | #include diff --git a/Documentation/devicetree/bindings/regulator/pfuze100.yaml b/Documentation/devicetree/bindings/regulator/pfuze100.yaml index c6de49685db7..f578e72778a7 100644 --- a/Documentation/devicetree/bindings/regulator/pfuze100.yaml +++ b/Documentation/devicetree/bindings/regulator/pfuze100.yaml @@ -80,6 +80,8 @@ required: - compatible - reg +additionalProperties: false + examples: - | i2c { diff --git a/Documentation/devicetree/bindings/sound/realtek,rt1015p.yaml b/Documentation/devicetree/bindings/sound/realtek,rt1015p.yaml index def1db298eac..644b68edf3e1 100644 --- a/Documentation/devicetree/bindings/sound/realtek,rt1015p.yaml +++ b/Documentation/devicetree/bindings/sound/realtek,rt1015p.yaml @@ -26,6 +26,8 @@ properties: required: - compatible +additionalProperties: false + examples: - | #include -- cgit v1.2.3 From 665e0224a3d76f36da40bd9012270fa629aa42ed Mon Sep 17 00:00:00 2001 From: Tyrel Datwyler Date: Sat, 24 Oct 2020 19:13:55 -0500 Subject: scsi: ibmvscsi: Fix potential race after loss of transport After a loss of transport due to an adapter migration or crash/disconnect from the host partner there is a tiny window where we can race adjusting the request_limit of the adapter. The request limit is atomically increased/decreased to track the number of inflight requests against the allowed limit of our VIOS partner. After a transport loss we set the request_limit to zero to reflect this state. However, there is a window where the adapter may attempt to queue a command because the transport loss event hasn't been fully processed yet and request_limit is still greater than zero. The hypercall to send the event will fail and the error path will increment the request_limit as a result. If the adapter processes the transport event prior to this increment the request_limit becomes out of sync with the adapter state and can result in SCSI commands being submitted on the now reset connection prior to an SRP Login resulting in a protocol violation. Fix this race by protecting request_limit with the host lock when changing the value via atomic_set() to indicate no transport. Link: https://lore.kernel.org/r/20201025001355.4527-1-tyreld@linux.ibm.com Signed-off-by: Tyrel Datwyler Signed-off-by: Martin K. Petersen --- drivers/scsi/ibmvscsi/ibmvscsi.c | 36 ++++++++++++++++++++++++++---------- 1 file changed, 26 insertions(+), 10 deletions(-) diff --git a/drivers/scsi/ibmvscsi/ibmvscsi.c b/drivers/scsi/ibmvscsi/ibmvscsi.c index b1f3017b6547..29fcc44be2d5 100644 --- a/drivers/scsi/ibmvscsi/ibmvscsi.c +++ b/drivers/scsi/ibmvscsi/ibmvscsi.c @@ -806,6 +806,22 @@ static void purge_requests(struct ibmvscsi_host_data *hostdata, int error_code) spin_unlock_irqrestore(hostdata->host->host_lock, flags); } +/** + * ibmvscsi_set_request_limit - Set the adapter request_limit in response to + * an adapter failure, reset, or SRP Login. Done under host lock to prevent + * race with SCSI command submission. + * @hostdata: adapter to adjust + * @limit: new request limit + */ +static void ibmvscsi_set_request_limit(struct ibmvscsi_host_data *hostdata, int limit) +{ + unsigned long flags; + + spin_lock_irqsave(hostdata->host->host_lock, flags); + atomic_set(&hostdata->request_limit, limit); + spin_unlock_irqrestore(hostdata->host->host_lock, flags); +} + /** * ibmvscsi_reset_host - Reset the connection to the server * @hostdata: struct ibmvscsi_host_data to reset @@ -813,7 +829,7 @@ static void purge_requests(struct ibmvscsi_host_data *hostdata, int error_code) static void ibmvscsi_reset_host(struct ibmvscsi_host_data *hostdata) { scsi_block_requests(hostdata->host); - atomic_set(&hostdata->request_limit, 0); + ibmvscsi_set_request_limit(hostdata, 0); purge_requests(hostdata, DID_ERROR); hostdata->action = IBMVSCSI_HOST_ACTION_RESET; @@ -1146,13 +1162,13 @@ static void login_rsp(struct srp_event_struct *evt_struct) dev_info(hostdata->dev, "SRP_LOGIN_REJ reason %u\n", evt_struct->xfer_iu->srp.login_rej.reason); /* Login failed. */ - atomic_set(&hostdata->request_limit, -1); + ibmvscsi_set_request_limit(hostdata, -1); return; default: dev_err(hostdata->dev, "Invalid login response typecode 0x%02x!\n", evt_struct->xfer_iu->srp.login_rsp.opcode); /* Login failed. */ - atomic_set(&hostdata->request_limit, -1); + ibmvscsi_set_request_limit(hostdata, -1); return; } @@ -1163,7 +1179,7 @@ static void login_rsp(struct srp_event_struct *evt_struct) * This value is set rather than added to request_limit because * request_limit could have been set to -1 by this client. */ - atomic_set(&hostdata->request_limit, + ibmvscsi_set_request_limit(hostdata, be32_to_cpu(evt_struct->xfer_iu->srp.login_rsp.req_lim_delta)); /* If we had any pending I/Os, kick them */ @@ -1195,13 +1211,13 @@ static int send_srp_login(struct ibmvscsi_host_data *hostdata) login->req_buf_fmt = cpu_to_be16(SRP_BUF_FORMAT_DIRECT | SRP_BUF_FORMAT_INDIRECT); - spin_lock_irqsave(hostdata->host->host_lock, flags); /* Start out with a request limit of 0, since this is negotiated in * the login request we are just sending and login requests always * get sent by the driver regardless of request_limit. */ - atomic_set(&hostdata->request_limit, 0); + ibmvscsi_set_request_limit(hostdata, 0); + spin_lock_irqsave(hostdata->host->host_lock, flags); rc = ibmvscsi_send_srp_event(evt_struct, hostdata, login_timeout * 2); spin_unlock_irqrestore(hostdata->host->host_lock, flags); dev_info(hostdata->dev, "sent SRP login\n"); @@ -1781,7 +1797,7 @@ static void ibmvscsi_handle_crq(struct viosrp_crq *crq, return; case VIOSRP_CRQ_XPORT_EVENT: /* Hypervisor telling us the connection is closed */ scsi_block_requests(hostdata->host); - atomic_set(&hostdata->request_limit, 0); + ibmvscsi_set_request_limit(hostdata, 0); if (crq->format == 0x06) { /* We need to re-setup the interpartition connection */ dev_info(hostdata->dev, "Re-enabling adapter!\n"); @@ -2137,12 +2153,12 @@ static void ibmvscsi_do_work(struct ibmvscsi_host_data *hostdata) } hostdata->action = IBMVSCSI_HOST_ACTION_NONE; + spin_unlock_irqrestore(hostdata->host->host_lock, flags); if (rc) { - atomic_set(&hostdata->request_limit, -1); + ibmvscsi_set_request_limit(hostdata, -1); dev_err(hostdata->dev, "error after %s\n", action); } - spin_unlock_irqrestore(hostdata->host->host_lock, flags); scsi_unblock_requests(hostdata->host); } @@ -2226,7 +2242,7 @@ static int ibmvscsi_probe(struct vio_dev *vdev, const struct vio_device_id *id) init_waitqueue_head(&hostdata->work_wait_q); hostdata->host = host; hostdata->dev = dev; - atomic_set(&hostdata->request_limit, -1); + ibmvscsi_set_request_limit(hostdata, -1); hostdata->host->max_sectors = IBMVSCSI_MAX_SECTORS_DEFAULT; if (map_persist_bufs(hostdata)) { -- cgit v1.2.3 From a1d2afc5dde29a943d32bf92eb0408c9f19541fc Mon Sep 17 00:00:00 2001 From: Martin Leung Date: Wed, 7 Oct 2020 12:17:22 -0400 Subject: drm/amd/display: adding ddc_gpio_vga_reg_list to ddc reg def'ns why: oem-related ddc read/write fails without these regs how: copy from hw_factory_dcn20.c Signed-off-by: Martin Leung Acked-by: Aurabindo Pillai Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_factory_dcn30.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_factory_dcn30.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_factory_dcn30.c index 3be2c90b0c61..21583699f992 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_factory_dcn30.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_factory_dcn30.c @@ -117,6 +117,12 @@ static const struct ddc_registers ddc_data_regs_dcn[] = { ddc_data_regs_dcn2(4), ddc_data_regs_dcn2(5), ddc_data_regs_dcn2(6), + { + DDC_GPIO_VGA_REG_LIST(DATA), + .ddc_setup = 0, + .phy_aux_cntl = 0, + .dc_gpio_aux_ctrl_5 = 0 + } }; static const struct ddc_registers ddc_clk_regs_dcn[] = { @@ -126,6 +132,12 @@ static const struct ddc_registers ddc_clk_regs_dcn[] = { ddc_clk_regs_dcn2(4), ddc_clk_regs_dcn2(5), ddc_clk_regs_dcn2(6), + { + DDC_GPIO_VGA_REG_LIST(CLK), + .ddc_setup = 0, + .phy_aux_cntl = 0, + .dc_gpio_aux_ctrl_5 = 0 + } }; static const struct ddc_sh_mask ddc_shift[] = { -- cgit v1.2.3 From 0689dcf3e4d6b89cc2087139561dc12b60461dca Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 26 Oct 2020 10:25:36 -0400 Subject: drm/amdgpu/display: use kvzalloc again in dc_create_state It looks this was accidently lost in a follow up patch. dc context is large and we don't need contiguous pages. Fixes: e4863f118a7d ("drm/amd/display: Multi display cause system lag on mode change") Reviewed-by: Nicholas Kazlauskas Signed-off-by: Alex Deucher Cc: Aric Cyr Cc: Alex Xu Reported-by: Alex Xu (Hello71) Tested-by: Alex Xu (Hello71) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/dc/core/dc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 1eb29c362122..45ad05f6e03b 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -1571,8 +1571,8 @@ static void init_state(struct dc *dc, struct dc_state *context) struct dc_state *dc_create_state(struct dc *dc) { - struct dc_state *context = kzalloc(sizeof(struct dc_state), - GFP_KERNEL); + struct dc_state *context = kvzalloc(sizeof(struct dc_state), + GFP_KERNEL); if (!context) return NULL; -- cgit v1.2.3 From 920bb38c518408fa2600eaefa0af9e82cf48f166 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 23 Oct 2020 09:46:54 +0200 Subject: drm/amd/display: Fix kernel panic by dal_gpio_open() error Currently both error code paths handled in dal_gpio_open_ex() issues ASSERT_CRITICAL(), and this leads to a kernel panic unnecessarily if CONFIG_KGDB is enabled. Since basically both are non-critical errors and can be recovered, drop those assert calls and use a safer one, BREAK_TO_DEBUGGER(), for allowing the debugging, instead. BugLink: https://bugzilla.opensuse.org/show_bug.cgi?id=1177973 Cc: Acked-by: Alex Deucher Reviewed-by: Nicholas Kazlauskas Signed-off-by: Takashi Iwai Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c b/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c index f67c18375bfd..dac427b68fd7 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_base.c @@ -63,13 +63,13 @@ enum gpio_result dal_gpio_open_ex( enum gpio_mode mode) { if (gpio->pin) { - ASSERT_CRITICAL(false); + BREAK_TO_DEBUGGER(); return GPIO_RESULT_ALREADY_OPENED; } // No action if allocation failed during gpio construct if (!gpio->hw_container.ddc) { - ASSERT_CRITICAL(false); + BREAK_TO_DEBUGGER(); return GPIO_RESULT_NON_SPECIFIC_ERROR; } gpio->mode = mode; -- cgit v1.2.3 From 8b7dc1fe1a5c1093551f6cd7dfbb941bd9081c2e Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 23 Oct 2020 09:46:55 +0200 Subject: drm/amd/display: Don't invoke kgdb_breakpoint() unconditionally ASSERT_CRITICAL() invokes kgdb_breakpoint() whenever either CONFIG_KGDB or CONFIG_HAVE_KGDB is set. This, however, may lead to a kernel panic when no kdb stuff is attached, since the kgdb_breakpoint() call issues INT3. It's nothing but a surprise for normal end-users. For avoiding the pitfall, make the kgdb_breakpoint() call only when CONFIG_DEBUG_KERNEL_DC is set. https://bugzilla.opensuse.org/show_bug.cgi?id=1177973 Cc: Acked-by: Alex Deucher Reviewed-by: Nicholas Kazlauskas Signed-off-by: Takashi Iwai Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/dc/os_types.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/dc/os_types.h b/drivers/gpu/drm/amd/display/dc/os_types.h index 330acaaed79a..32758b245754 100644 --- a/drivers/gpu/drm/amd/display/dc/os_types.h +++ b/drivers/gpu/drm/amd/display/dc/os_types.h @@ -94,7 +94,7 @@ * general debug capabilities * */ -#if defined(CONFIG_HAVE_KGDB) || defined(CONFIG_KGDB) +#if defined(CONFIG_DEBUG_KERNEL_DC) && (defined(CONFIG_HAVE_KGDB) || defined(CONFIG_KGDB)) #define ASSERT_CRITICAL(expr) do { \ if (WARN_ON(!(expr))) { \ kgdb_breakpoint(); \ -- cgit v1.2.3 From bcc3775dcf920b25d7123c141ed61f7f9f9bb6c3 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 23 Oct 2020 09:46:56 +0200 Subject: drm/amd/display: Clean up debug macros This patch simplifies the ASSERT*() and BREAK_TO_DEBUGGER() macros: - Move the dependency check of CONFIG_KGDB into Kconfig - Unify the kgdb_breakpoint() call - Drop the non-existing CONFIG_HAVE_KGDB Also align the behavior of ASSERT() macro in both cases with and without CONFIG_DEBUG_KERNEL_DC. Acked-by: Alex Deucher Reviewed-by: Nicholas Kazlauskas Signed-off-by: Takashi Iwai Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/display/Kconfig | 1 + drivers/gpu/drm/amd/display/dc/os_types.h | 33 +++++++++++-------------------- 2 files changed, 13 insertions(+), 21 deletions(-) diff --git a/drivers/gpu/drm/amd/display/Kconfig b/drivers/gpu/drm/amd/display/Kconfig index f24abf428534..60dfdd432aba 100644 --- a/drivers/gpu/drm/amd/display/Kconfig +++ b/drivers/gpu/drm/amd/display/Kconfig @@ -42,6 +42,7 @@ config DRM_AMD_DC_SI config DEBUG_KERNEL_DC bool "Enable kgdb break in DC" depends on DRM_AMD_DC + depends on KGDB help Choose this option if you want to hit kdgb_break in assert. diff --git a/drivers/gpu/drm/amd/display/dc/os_types.h b/drivers/gpu/drm/amd/display/dc/os_types.h index 32758b245754..95cb56929e79 100644 --- a/drivers/gpu/drm/amd/display/dc/os_types.h +++ b/drivers/gpu/drm/amd/display/dc/os_types.h @@ -94,36 +94,27 @@ * general debug capabilities * */ -#if defined(CONFIG_DEBUG_KERNEL_DC) && (defined(CONFIG_HAVE_KGDB) || defined(CONFIG_KGDB)) -#define ASSERT_CRITICAL(expr) do { \ - if (WARN_ON(!(expr))) { \ - kgdb_breakpoint(); \ - } \ -} while (0) +#ifdef CONFIG_DEBUG_KERNEL_DC +#define dc_breakpoint() kgdb_breakpoint() #else -#define ASSERT_CRITICAL(expr) do { \ - if (WARN_ON(!(expr))) { \ - ; \ - } \ -} while (0) +#define dc_breakpoint() do {} while (0) #endif -#if defined(CONFIG_DEBUG_KERNEL_DC) -#define ASSERT(expr) ASSERT_CRITICAL(expr) +#define ASSERT_CRITICAL(expr) do { \ + if (WARN_ON(!(expr))) \ + dc_breakpoint(); \ + } while (0) -#else -#define ASSERT(expr) WARN_ON_ONCE(!(expr)) -#endif +#define ASSERT(expr) do { \ + if (WARN_ON_ONCE(!(expr))) \ + dc_breakpoint(); \ + } while (0) -#if defined(CONFIG_DEBUG_KERNEL_DC) && (defined(CONFIG_HAVE_KGDB) || defined(CONFIG_KGDB)) #define BREAK_TO_DEBUGGER() \ do { \ DRM_DEBUG_DRIVER("%s():%d\n", __func__, __LINE__); \ - kgdb_breakpoint(); \ + dc_breakpoint(); \ } while (0) -#else -#define BREAK_TO_DEBUGGER() DRM_DEBUG_DRIVER("%s():%d\n", __func__, __LINE__) -#endif #define DC_ERR(...) do { \ dm_error(__VA_ARGS__); \ -- cgit v1.2.3 From fab09aaee80389a37d8ab49396afbb77fa86583a Mon Sep 17 00:00:00 2001 From: John Garry Date: Thu, 15 Oct 2020 16:27:05 +0800 Subject: scsi: hisi_sas: Stop using queue #0 always for v2 hw In commit 8d98416a55eb ("scsi: hisi_sas: Switch v3 hw to MQ"), the dispatch function was changed to choose the delivery queue based on the request tag HW queue index. This heavily degrades performance for v2 hw, since the HW queues are not exposed there, and, as such, HW queue #0 is used for every command. Revert to previous behaviour for when nr_hw_queues is not set, that being to choose the HW queue based on target device index. Link: https://lore.kernel.org/r/1602750425-240341-1-git-send-email-john.garry@huawei.com Fixes: 8d98416a55eb ("scsi: hisi_sas: Switch v3 hw to MQ") Signed-off-by: John Garry Signed-off-by: Martin K. Petersen --- drivers/scsi/hisi_sas/hisi_sas_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c index 128583dfccf2..c8dd8588f800 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_main.c +++ b/drivers/scsi/hisi_sas/hisi_sas_main.c @@ -445,7 +445,7 @@ static int hisi_sas_task_prep(struct sas_task *task, } } - if (scmd) { + if (scmd && hisi_hba->shost->nr_hw_queues) { unsigned int dq_index; u32 blk_tag; -- cgit v1.2.3 From edebc8407b8891ec0ea9ca4089f3d3343a5e50dd Mon Sep 17 00:00:00 2001 From: Bob Pearson Date: Fri, 16 Oct 2020 16:13:44 -0500 Subject: RDMA/rxe: Fix small problem in network_type patch The patch referenced below has a typo that results in using the wrong L2 header size for outbound traffic. (V4 <-> V6). It also breaks kernel-side RC traffic because they use AVs that use RDMA_NETWORK_XXX enums instead of RXE_NETWORK_TYPE_XXX enums. Fix this by transcoding between these enum types. Fixes: e0d696d201dd ("RDMA/rxe: Move the definitions for rxe_av.network_type to uAPI") Link: https://lore.kernel.org/r/20201016211343.22906-1-rpearson@hpe.com Signed-off-by: Bob Pearson Signed-off-by: Jason Gunthorpe --- drivers/infiniband/sw/rxe/rxe_av.c | 35 +++++++++++++++++++++++++++++++---- drivers/infiniband/sw/rxe/rxe_net.c | 2 +- 2 files changed, 32 insertions(+), 5 deletions(-) diff --git a/drivers/infiniband/sw/rxe/rxe_av.c b/drivers/infiniband/sw/rxe/rxe_av.c index 38021e2c8688..df0d173d6acb 100644 --- a/drivers/infiniband/sw/rxe/rxe_av.c +++ b/drivers/infiniband/sw/rxe/rxe_av.c @@ -16,15 +16,24 @@ void rxe_init_av(struct rdma_ah_attr *attr, struct rxe_av *av) int rxe_av_chk_attr(struct rxe_dev *rxe, struct rdma_ah_attr *attr) { + const struct ib_global_route *grh = rdma_ah_read_grh(attr); struct rxe_port *port; + int type; port = &rxe->port; if (rdma_ah_get_ah_flags(attr) & IB_AH_GRH) { - u8 sgid_index = rdma_ah_read_grh(attr)->sgid_index; + if (grh->sgid_index > port->attr.gid_tbl_len) { + pr_warn("invalid sgid index = %d\n", + grh->sgid_index); + return -EINVAL; + } - if (sgid_index > port->attr.gid_tbl_len) { - pr_warn("invalid sgid index = %d\n", sgid_index); + type = rdma_gid_attr_network_type(grh->sgid_attr); + if (type < RDMA_NETWORK_IPV4 || + type > RDMA_NETWORK_IPV6) { + pr_warn("invalid network type for rdma_rxe = %d\n", + type); return -EINVAL; } } @@ -65,11 +74,29 @@ void rxe_av_to_attr(struct rxe_av *av, struct rdma_ah_attr *attr) void rxe_av_fill_ip_info(struct rxe_av *av, struct rdma_ah_attr *attr) { const struct ib_gid_attr *sgid_attr = attr->grh.sgid_attr; + int ibtype; + int type; rdma_gid2ip((struct sockaddr *)&av->sgid_addr, &sgid_attr->gid); rdma_gid2ip((struct sockaddr *)&av->dgid_addr, &rdma_ah_read_grh(attr)->dgid); - av->network_type = rdma_gid_attr_network_type(sgid_attr); + + ibtype = rdma_gid_attr_network_type(sgid_attr); + + switch (ibtype) { + case RDMA_NETWORK_IPV4: + type = RXE_NETWORK_TYPE_IPV4; + break; + case RDMA_NETWORK_IPV6: + type = RXE_NETWORK_TYPE_IPV4; + break; + default: + /* not reached - checked in rxe_av_chk_attr */ + type = 0; + break; + } + + av->network_type = type; } struct rxe_av *rxe_get_av(struct rxe_pkt_info *pkt) diff --git a/drivers/infiniband/sw/rxe/rxe_net.c b/drivers/infiniband/sw/rxe/rxe_net.c index 575e1a4ec821..34bef7d8e6b4 100644 --- a/drivers/infiniband/sw/rxe/rxe_net.c +++ b/drivers/infiniband/sw/rxe/rxe_net.c @@ -442,7 +442,7 @@ struct sk_buff *rxe_init_packet(struct rxe_dev *rxe, struct rxe_av *av, if (IS_ERR(attr)) return NULL; - if (av->network_type == RXE_NETWORK_TYPE_IPV6) + if (av->network_type == RXE_NETWORK_TYPE_IPV4) hdr_len = ETH_HLEN + sizeof(struct udphdr) + sizeof(struct iphdr); else -- cgit v1.2.3 From fbdd0049d98d44914fc57d4b91f867f4996c787b Mon Sep 17 00:00:00 2001 From: Parav Pandit Date: Mon, 26 Oct 2020 15:43:59 +0200 Subject: RDMA/mlx5: Fix devlink deadlock on net namespace deletion When a mlx5 core devlink instance is reloaded in different net namespace, its associated IB device is deleted and recreated. Example sequence is: $ ip netns add foo $ devlink dev reload pci/0000:00:08.0 netns foo $ ip netns del foo mlx5 IB device needs to attach and detach the netdevice to it through the netdev notifier chain during load and unload sequence. A below call graph of the unload flow. cleanup_net() down_read(&pernet_ops_rwsem); <- first sem acquired ops_pre_exit_list() pre_exit() devlink_pernet_pre_exit() devlink_reload() mlx5_devlink_reload_down() mlx5_unload_one() [...] mlx5_ib_remove() mlx5_ib_unbind_slave_port() mlx5_remove_netdev_notifier() unregister_netdevice_notifier() down_write(&pernet_ops_rwsem);<- recurrsive lock Hence, when net namespace is deleted, mlx5 reload results in deadlock. When deadlock occurs, devlink mutex is also held. This not only deadlocks the mlx5 device under reload, but all the processes which attempt to access unrelated devlink devices are deadlocked. Hence, fix this by mlx5 ib driver to register for per net netdev notifier instead of global one, which operats on the net namespace without holding the pernet_ops_rwsem. Fixes: 4383cfcc65e7 ("net/mlx5: Add devlink reload") Link: https://lore.kernel.org/r/20201026134359.23150-1-parav@nvidia.com Signed-off-by: Parav Pandit Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe --- drivers/infiniband/hw/mlx5/main.c | 6 ++++-- drivers/net/ethernet/mellanox/mlx5/core/lib/mlx5.h | 5 ----- include/linux/mlx5/driver.h | 18 ++++++++++++++++++ 3 files changed, 22 insertions(+), 7 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 89e04ca62ae0..246e3cbe0b2c 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -3305,7 +3305,8 @@ static int mlx5_add_netdev_notifier(struct mlx5_ib_dev *dev, u8 port_num) int err; dev->port[port_num].roce.nb.notifier_call = mlx5_netdev_event; - err = register_netdevice_notifier(&dev->port[port_num].roce.nb); + err = register_netdevice_notifier_net(mlx5_core_net(dev->mdev), + &dev->port[port_num].roce.nb); if (err) { dev->port[port_num].roce.nb.notifier_call = NULL; return err; @@ -3317,7 +3318,8 @@ static int mlx5_add_netdev_notifier(struct mlx5_ib_dev *dev, u8 port_num) static void mlx5_remove_netdev_notifier(struct mlx5_ib_dev *dev, u8 port_num) { if (dev->port[port_num].roce.nb.notifier_call) { - unregister_netdevice_notifier(&dev->port[port_num].roce.nb); + unregister_netdevice_notifier_net(mlx5_core_net(dev->mdev), + &dev->port[port_num].roce.nb); dev->port[port_num].roce.nb.notifier_call = NULL; } } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/mlx5.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/mlx5.h index d046db7bb047..3a9fa629503f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/mlx5.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/mlx5.h @@ -90,9 +90,4 @@ int mlx5_create_encryption_key(struct mlx5_core_dev *mdev, u32 key_type, u32 *p_key_id); void mlx5_destroy_encryption_key(struct mlx5_core_dev *mdev, u32 key_id); -static inline struct net *mlx5_core_net(struct mlx5_core_dev *dev) -{ - return devlink_net(priv_to_devlink(dev)); -} - #endif diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index add85094f9a5..0f23e1ed5e71 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -1213,4 +1213,22 @@ static inline bool mlx5_is_roce_enabled(struct mlx5_core_dev *dev) return val.vbool; } +/** + * mlx5_core_net - Provide net namespace of the mlx5_core_dev + * @dev: mlx5 core device + * + * mlx5_core_net() returns the net namespace of mlx5 core device. + * This can be called only in below described limited context. + * (a) When a devlink instance for mlx5_core is registered and + * when devlink reload operation is disabled. + * or + * (b) during devlink reload reload_down() and reload_up callbacks + * where it is ensured that devlink instance's net namespace is + * stable. + */ +static inline struct net *mlx5_core_net(struct mlx5_core_dev *dev) +{ + return devlink_net(priv_to_devlink(dev)); +} + #endif /* MLX5_DRIVER_H */ -- cgit v1.2.3 From 7d66a71488d7c14506ab81d6455c095992efca04 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Mon, 26 Oct 2020 10:26:21 +0200 Subject: RDMA/uverbs: Fix false error in query gid IOCTL Some drivers (such as EFA) have a GID table, but aren't IB/RoCE devices. Remove the unnecessary rdma_ib_or_roce() check. This fixes rdma-core failures for EFA when it uses the new ioctl interface for querying the GID table. Fixes: 9f85cbe50aa0 ("RDMA/uverbs: Expose the new GID query API to user space") Link: https://lore.kernel.org/r/20201026082621.32463-1-galpress@amazon.com Signed-off-by: Gal Pressman Signed-off-by: Jason Gunthorpe --- drivers/infiniband/core/uverbs_std_types_device.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/infiniband/core/uverbs_std_types_device.c b/drivers/infiniband/core/uverbs_std_types_device.c index f367d523a46b..302f898c5833 100644 --- a/drivers/infiniband/core/uverbs_std_types_device.c +++ b/drivers/infiniband/core/uverbs_std_types_device.c @@ -401,9 +401,6 @@ static int UVERBS_HANDLER(UVERBS_METHOD_QUERY_GID_ENTRY)( if (!rdma_is_port_valid(ib_dev, port_num)) return -EINVAL; - if (!rdma_ib_or_roce(ib_dev, port_num)) - return -EOPNOTSUPP; - gid_attr = rdma_get_gid_attr(ib_dev, port_num, gid_index); if (IS_ERR(gid_attr)) return PTR_ERR(gid_attr); -- cgit v1.2.3 From bf9a76a470d83355200adaa5d5b55d118f229ecb Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 26 Oct 2020 15:39:37 -0700 Subject: arch/um: partially revert the conversion to __section() macro A couple of um files ended up not including the header file that defines the __section() macro, and the simplest fix is to just revert the change for those files. Fixes: 33def8498fdd treewide: Convert macro and uses of __section(foo) to __section("foo") Reported-and-tested-by: Guenter Roeck Cc: Joe Perches Signed-off-by: Linus Torvalds --- arch/um/kernel/skas/clone.c | 2 +- arch/x86/um/stub_segv.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/um/kernel/skas/clone.c b/arch/um/kernel/skas/clone.c index 95c355181dcd..bfb70c456b30 100644 --- a/arch/um/kernel/skas/clone.c +++ b/arch/um/kernel/skas/clone.c @@ -21,7 +21,7 @@ * on some systems. */ -void __section(".__syscall_stub") +void __attribute__ ((__section__ (".__syscall_stub"))) stub_clone_handler(void) { struct stub_data *data = (struct stub_data *) STUB_DATA; diff --git a/arch/x86/um/stub_segv.c b/arch/x86/um/stub_segv.c index fdcd58af707a..27361cbb7ca9 100644 --- a/arch/x86/um/stub_segv.c +++ b/arch/x86/um/stub_segv.c @@ -8,7 +8,7 @@ #include #include -void __section(".__syscall_stub") +void __attribute__ ((__section__ (".__syscall_stub"))) stub_segv_handler(int sig, siginfo_t *info, void *p) { ucontext_t *uc = p; -- cgit v1.2.3 From 4525c8781ec0701ce824e8bd379ae1b129e26568 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 26 Oct 2020 15:45:22 -0700 Subject: scsi: qla2xxx: remove incorrect sparse #ifdef The code to try to shut up sparse warnings about questionable locking didn't shut up sparse: it made the result not parse as valid C at all, since the end result now has a label with no statement. The proper fix is to just always lock the hardware, the same way Bart did in commit 8ae178760b23 ("scsi: qla2xxx: Simplify the functions for dumping firmware"). That avoids the whole problem with having locking that is not statically obvious. But in the meantime, just remove the incorrect attempt at trying to avoid a sparse warning that just made things worse. This was exposed by commit 3e6efab865ac ("scsi: qla2xxx: Fix reset of MPI firmware"), very similarly to how commit cbb01c2f2f63 ("scsi: qla2xxx: Fix MPI failure AEN (8200) handling") exposed the same problem in another place, and caused that commit 8ae178760b23. Please don't add code to just shut up sparse without actually fixing what sparse complains about. Reported-by: Luc Van Oostenryck Cc: Bart Van Assche Cc: Arun Easi Signed-off-by: Linus Torvalds --- drivers/scsi/qla2xxx/qla_tmpl.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/scsi/qla2xxx/qla_tmpl.c b/drivers/scsi/qla2xxx/qla_tmpl.c index 84f4416d366f..bd8623ee156a 100644 --- a/drivers/scsi/qla2xxx/qla_tmpl.c +++ b/drivers/scsi/qla2xxx/qla_tmpl.c @@ -1001,10 +1001,8 @@ qla27xx_mpi_fwdump(scsi_qla_host_t *vha, int hardware_locked) { ulong flags = 0; -#ifndef __CHECKER__ if (!hardware_locked) spin_lock_irqsave(&vha->hw->hardware_lock, flags); -#endif if (!vha->hw->mpi_fw_dump) { ql_log(ql_log_warn, vha, 0x02f3, "-> mpi_fwdump no buffer\n"); } else { @@ -1050,10 +1048,8 @@ qla27xx_mpi_fwdump(scsi_qla_host_t *vha, int hardware_locked) } bailout: -#ifndef __CHECKER__ if (!hardware_locked) spin_unlock_irqrestore(&vha->hw->hardware_lock, flags); -#endif } void -- cgit v1.2.3 From ee7a376421dd7bc65b610d42d42c084a0d16d6fa Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 23 Oct 2020 14:22:12 +0300 Subject: net: hns3: clean up a return in hclge_tm_bp_setup() Smatch complains that "ret" might be uninitialized if we don't enter the loop. We do always enter the loop so it's a false positive, but it's cleaner to just return a literal zero and that silences the warning as well. Signed-off-by: Dan Carpenter Link: https://lore.kernel.org/r/20201023112212.GA282278@mwanda Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c index 15f69fa86323..e8495f58a1a8 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c @@ -1373,7 +1373,7 @@ static int hclge_tm_bp_setup(struct hclge_dev *hdev) return ret; } - return ret; + return 0; } int hclge_pause_setup_hw(struct hclge_dev *hdev, bool init) -- cgit v1.2.3 From 937d8420588421eaa5c7aa5c79b26b42abb288ef Mon Sep 17 00:00:00 2001 From: Raju Rangoju Date: Fri, 23 Oct 2020 17:28:52 +0530 Subject: cxgb4: set up filter action after rewrites The current code sets up the filter action field before rewrites are set up. When the action 'switch' is used with rewrites, this may result in initial few packets that get switched out don't have rewrites applied on them. So, make sure filter action is set up along with rewrites or only after everything else is set up for rewrites. Fixes: 12b276fbf6e0 ("cxgb4: add support to create hash filters") Signed-off-by: Raju Rangoju Link: https://lore.kernel.org/r/20201023115852.18262-1-rajur@chelsio.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c | 56 +++++++++++------------ drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h | 4 ++ 2 files changed, 31 insertions(+), 29 deletions(-) diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c index 6ec5f2f26f05..4e55f7081644 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_filter.c @@ -145,13 +145,13 @@ static int configure_filter_smac(struct adapter *adap, struct filter_entry *f) int err; /* do a set-tcb for smac-sel and CWR bit.. */ - err = set_tcb_tflag(adap, f, f->tid, TF_CCTRL_CWR_S, 1, 1); - if (err) - goto smac_err; - err = set_tcb_field(adap, f, f->tid, TCB_SMAC_SEL_W, TCB_SMAC_SEL_V(TCB_SMAC_SEL_M), TCB_SMAC_SEL_V(f->smt->idx), 1); + if (err) + goto smac_err; + + err = set_tcb_tflag(adap, f, f->tid, TF_CCTRL_CWR_S, 1, 1); if (!err) return 0; @@ -862,6 +862,7 @@ int set_filter_wr(struct adapter *adapter, int fidx) FW_FILTER_WR_DIRSTEERHASH_V(f->fs.dirsteerhash) | FW_FILTER_WR_LPBK_V(f->fs.action == FILTER_SWITCH) | FW_FILTER_WR_DMAC_V(f->fs.newdmac) | + FW_FILTER_WR_SMAC_V(f->fs.newsmac) | FW_FILTER_WR_INSVLAN_V(f->fs.newvlan == VLAN_INSERT || f->fs.newvlan == VLAN_REWRITE) | FW_FILTER_WR_RMVLAN_V(f->fs.newvlan == VLAN_REMOVE || @@ -879,7 +880,7 @@ int set_filter_wr(struct adapter *adapter, int fidx) FW_FILTER_WR_OVLAN_VLD_V(f->fs.val.ovlan_vld) | FW_FILTER_WR_IVLAN_VLDM_V(f->fs.mask.ivlan_vld) | FW_FILTER_WR_OVLAN_VLDM_V(f->fs.mask.ovlan_vld)); - fwr->smac_sel = 0; + fwr->smac_sel = f->smt->idx; fwr->rx_chan_rx_rpl_iq = htons(FW_FILTER_WR_RX_CHAN_V(0) | FW_FILTER_WR_RX_RPL_IQ_V(adapter->sge.fw_evtq.abs_id)); @@ -1323,11 +1324,8 @@ static void mk_act_open_req6(struct filter_entry *f, struct sk_buff *skb, TX_QUEUE_V(f->fs.nat_mode) | T5_OPT_2_VALID_F | RX_CHANNEL_V(cxgb4_port_e2cchan(f->dev)) | - CONG_CNTRL_V((f->fs.action == FILTER_DROP) | - (f->fs.dirsteer << 1)) | PACE_V((f->fs.maskhash) | - ((f->fs.dirsteerhash) << 1)) | - CCTRL_ECN_V(f->fs.action == FILTER_SWITCH)); + ((f->fs.dirsteerhash) << 1))); } static void mk_act_open_req(struct filter_entry *f, struct sk_buff *skb, @@ -1363,11 +1361,8 @@ static void mk_act_open_req(struct filter_entry *f, struct sk_buff *skb, TX_QUEUE_V(f->fs.nat_mode) | T5_OPT_2_VALID_F | RX_CHANNEL_V(cxgb4_port_e2cchan(f->dev)) | - CONG_CNTRL_V((f->fs.action == FILTER_DROP) | - (f->fs.dirsteer << 1)) | PACE_V((f->fs.maskhash) | - ((f->fs.dirsteerhash) << 1)) | - CCTRL_ECN_V(f->fs.action == FILTER_SWITCH)); + ((f->fs.dirsteerhash) << 1))); } static int cxgb4_set_hash_filter(struct net_device *dev, @@ -2039,6 +2034,20 @@ void hash_filter_rpl(struct adapter *adap, const struct cpl_act_open_rpl *rpl) } return; } + switch (f->fs.action) { + case FILTER_PASS: + if (f->fs.dirsteer) + set_tcb_tflag(adap, f, tid, + TF_DIRECT_STEER_S, 1, 1); + break; + case FILTER_DROP: + set_tcb_tflag(adap, f, tid, TF_DROP_S, 1, 1); + break; + case FILTER_SWITCH: + set_tcb_tflag(adap, f, tid, TF_LPBK_S, 1, 1); + break; + } + break; default: @@ -2106,22 +2115,11 @@ void filter_rpl(struct adapter *adap, const struct cpl_set_tcb_rpl *rpl) if (ctx) ctx->result = 0; } else if (ret == FW_FILTER_WR_FLT_ADDED) { - int err = 0; - - if (f->fs.newsmac) - err = configure_filter_smac(adap, f); - - if (!err) { - f->pending = 0; /* async setup completed */ - f->valid = 1; - if (ctx) { - ctx->result = 0; - ctx->tid = idx; - } - } else { - clear_filter(adap, f); - if (ctx) - ctx->result = err; + f->pending = 0; /* async setup completed */ + f->valid = 1; + if (ctx) { + ctx->result = 0; + ctx->tid = idx; } } else { /* Something went wrong. Issue a warning about the diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h b/drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h index 50232e063f49..92473dda55d9 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h @@ -50,6 +50,10 @@ #define TCB_T_FLAGS_M 0xffffffffffffffffULL #define TCB_T_FLAGS_V(x) ((__u64)(x) << TCB_T_FLAGS_S) +#define TF_DROP_S 22 +#define TF_DIRECT_STEER_S 23 +#define TF_LPBK_S 59 + #define TF_CCTRL_ECE_S 60 #define TF_CCTRL_CWR_S 61 #define TF_CCTRL_RFR_S 62 -- cgit v1.2.3 From af545bb5ee53f5261db631db2ac4cde54038bdaf Mon Sep 17 00:00:00 2001 From: Jeff Vander Stoep Date: Fri, 23 Oct 2020 16:37:57 +0200 Subject: vsock: use ns_capable_noaudit() on socket create During __vsock_create() CAP_NET_ADMIN is used to determine if the vsock_sock->trusted should be set to true. This value is used later for determing if a remote connection should be allowed to connect to a restricted VM. Unfortunately, if the caller doesn't have CAP_NET_ADMIN, an audit message such as an selinux denial is generated even if the caller does not want a trusted socket. Logging errors on success is confusing. To avoid this, switch the capable(CAP_NET_ADMIN) check to the noaudit version. Reported-by: Roman Kiryanov https://android-review.googlesource.com/c/device/generic/goldfish/+/1468545/ Signed-off-by: Jeff Vander Stoep Reviewed-by: James Morris Link: https://lore.kernel.org/r/20201023143757.377574-1-jeffv@google.com Signed-off-by: Jakub Kicinski --- net/vmw_vsock/af_vsock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 9e93bc201cc0..b4d7b8aba003 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -739,7 +739,7 @@ static struct sock *__vsock_create(struct net *net, vsk->buffer_min_size = psk->buffer_min_size; vsk->buffer_max_size = psk->buffer_max_size; } else { - vsk->trusted = capable(CAP_NET_ADMIN); + vsk->trusted = ns_capable_noaudit(&init_user_ns, CAP_NET_ADMIN); vsk->owner = get_current_cred(); vsk->connect_timeout = VSOCK_DEFAULT_CONNECT_TIMEOUT; vsk->buffer_size = VSOCK_DEFAULT_BUFFER_SIZE; -- cgit v1.2.3 From 4a9baf45fd72615a804947a8495b73c4a0a4cb54 Mon Sep 17 00:00:00 2001 From: Karsten Graul Date: Fri, 23 Oct 2020 20:48:28 +0200 Subject: net/smc: fix null pointer dereference in smc_listen_decline() smc_listen_work() calls smc_listen_decline() on label out_decl, providing the ini pointer variable. But this pointer can still be null when the label out_decl is reached. Fix this by checking the ini variable in smc_listen_work() and call smc_listen_decline() with the result directly. Fixes: a7c9c5f4af7f ("net/smc: CLC accept / confirm V2") Signed-off-by: Karsten Graul Signed-off-by: Jakub Kicinski --- net/smc/af_smc.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 82be0bd0f6e8..e9f487c8c6d5 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -1317,10 +1317,10 @@ static void smc_listen_out_err(struct smc_sock *new_smc) /* listen worker: decline and fall back if possible */ static void smc_listen_decline(struct smc_sock *new_smc, int reason_code, - struct smc_init_info *ini, u8 version) + int local_first, u8 version) { /* RDMA setup failed, switch back to TCP */ - if (ini->first_contact_local) + if (local_first) smc_lgr_cleanup_early(&new_smc->conn); else smc_conn_free(&new_smc->conn); @@ -1768,7 +1768,8 @@ static void smc_listen_work(struct work_struct *work) out_unlock: mutex_unlock(&smc_server_lgr_pending); out_decl: - smc_listen_decline(new_smc, rc, ini, version); + smc_listen_decline(new_smc, rc, ini ? ini->first_contact_local : 0, + version); out_free: kfree(ini); kfree(buf); -- cgit v1.2.3 From 96d6fded958d971a3695009e0ed43aca6c598283 Mon Sep 17 00:00:00 2001 From: Karsten Graul Date: Fri, 23 Oct 2020 20:48:29 +0200 Subject: net/smc: fix suppressed return code The patch that repaired the invalid return code in smcd_new_buf_create() missed to take care of errno ENOSPC which has a special meaning that no more DMBEs can be registered on the device. Fix that by keeping this errno value during the translation of the return code. Fixes: 6b1bbf94ab36 ("net/smc: fix invalid return code in smcd_new_buf_create()") Signed-off-by: Karsten Graul Signed-off-by: Jakub Kicinski --- net/smc/smc_core.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c index d790c43c473f..2b19863f7171 100644 --- a/net/smc/smc_core.c +++ b/net/smc/smc_core.c @@ -1615,8 +1615,11 @@ static struct smc_buf_desc *smcd_new_buf_create(struct smc_link_group *lgr, rc = smc_ism_register_dmb(lgr, bufsize, buf_desc); if (rc) { kfree(buf_desc); - return (rc == -ENOMEM) ? ERR_PTR(-EAGAIN) : - ERR_PTR(-EIO); + if (rc == -ENOMEM) + return ERR_PTR(-EAGAIN); + if (rc == -ENOSPC) + return ERR_PTR(-ENOSPC); + return ERR_PTR(-EIO); } buf_desc->pages = virt_to_page(buf_desc->cpu_addr); /* CDC header stored in buf. So, pretend it was smaller */ -- cgit v1.2.3 From 1dc0d1cf6f3d910ce3fffa83c5ae40c564e12373 Mon Sep 17 00:00:00 2001 From: Karsten Graul Date: Fri, 23 Oct 2020 20:48:30 +0200 Subject: s390/ism: fix incorrect system EID The system EID that is defined by the ISM driver is not correct. Using an incorrect system EID allows to communicate with remote Linux systems that use the same incorrect system EID, but when it comes to interoperability with other operating systems then the system EIDs do never match which prevents SMC-Dv2 communication. Using the correct system EID fixes this problem. Fixes: 201091ebb2a1 ("net/smc: introduce System Enterprise ID (SEID)") Signed-off-by: Karsten Graul Signed-off-by: Jakub Kicinski --- drivers/s390/net/ism_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/s390/net/ism_drv.c b/drivers/s390/net/ism_drv.c index fe96ca3c88a5..26cc943d2034 100644 --- a/drivers/s390/net/ism_drv.c +++ b/drivers/s390/net/ism_drv.c @@ -390,7 +390,7 @@ static int ism_move(struct smcd_dev *smcd, u64 dmb_tok, unsigned int idx, } static struct ism_systemeid SYSTEM_EID = { - .seid_string = "IBM-SYSZ-IBMSEID00000000", + .seid_string = "IBM-SYSZ-ISMSEID00000000", .serial_number = "0000", .type = "0000", }; -- cgit v1.2.3 From 24cb90964665c1e7d322bce754623b7155528252 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Mon, 26 Oct 2020 11:20:39 -0300 Subject: ARM: imx_v4_v5_defconfig: Select CONFIG_GPIO_MXC Since commit 12d16b397ce0 ("gpio: mxc: Support module build") the CONFIG_GPIO_MXC option needs to be explicitly selected. Select it to avoid boot issues on imx25/imx27 due to the lack of the GPIO driver. Signed-off-by: Fabio Estevam Signed-off-by: Shawn Guo --- arch/arm/configs/imx_v4_v5_defconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/configs/imx_v4_v5_defconfig b/arch/arm/configs/imx_v4_v5_defconfig index aeb1209e0804..bb70acc6b526 100644 --- a/arch/arm/configs/imx_v4_v5_defconfig +++ b/arch/arm/configs/imx_v4_v5_defconfig @@ -93,6 +93,7 @@ CONFIG_SPI=y CONFIG_SPI_IMX=y CONFIG_SPI_SPIDEV=y CONFIG_GPIO_SYSFS=y +CONFIG_GPIO_MXC=y CONFIG_W1=y CONFIG_W1_MASTER_MXC=y CONFIG_W1_SLAVE_THERM=y -- cgit v1.2.3 From ccee91b568e35e87d61ae72b51fd07ddf5a3d999 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Mon, 26 Oct 2020 11:20:40 -0300 Subject: ARM: multi_v5_defconfig: Select CONFIG_GPIO_MXC Since commit 12d16b397ce0 ("gpio: mxc: Support module build") the CONFIG_GPIO_MXC option needs to be explicitly selected. Select it to avoid boot issues on imx25/imx27 due to the lack of the GPIO driver. Signed-off-by: Fabio Estevam Signed-off-by: Shawn Guo --- arch/arm/configs/multi_v5_defconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/configs/multi_v5_defconfig b/arch/arm/configs/multi_v5_defconfig index 70b709a669d2..e00be9faa23b 100644 --- a/arch/arm/configs/multi_v5_defconfig +++ b/arch/arm/configs/multi_v5_defconfig @@ -166,6 +166,7 @@ CONFIG_SPI_IMX=y CONFIG_SPI_ORION=y CONFIG_GPIO_ASPEED=m CONFIG_GPIO_ASPEED_SGPIO=y +CONFIG_GPIO_MXC=y CONFIG_POWER_RESET=y CONFIG_POWER_RESET_GPIO=y CONFIG_POWER_RESET_QNAP=y -- cgit v1.2.3 From 1601559be3e4213148b4cb4a1abe672b00bf4f67 Mon Sep 17 00:00:00 2001 From: Amit Cohen Date: Sat, 24 Oct 2020 16:37:31 +0300 Subject: mlxsw: Only advertise link modes supported by both driver and device During port creation the driver instructs the device to advertise all the supported link modes queried from the device. Since cited commit not all the link modes supported by the device are supported by the driver. This can result in the device negotiating a link mode that is not recognized by the driver causing ethtool to show an unsupported speed: $ ethtool swp1 ... Speed: Unknown! This is especially problematic when the netdev is enslaved to a bond, as the bond driver uses unknown speed as an indication that the link is down: [13048.900895] net_ratelimit: 86 callbacks suppressed [13048.900902] t_bond0: (slave swp52): failed to get link speed/duplex [13048.912160] t_bond0: (slave swp49): failed to get link speed/duplex Fix this by making sure that only link modes that are supported by both the device and the driver are advertised. Fixes: b97cd891268d ("mlxsw: Remove 56G speed support") Signed-off-by: Amit Cohen Signed-off-by: Ido Schimmel Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 9 +++++-- drivers/net/ethernet/mellanox/mlxsw/spectrum.h | 1 + .../net/ethernet/mellanox/mlxsw/spectrum_ethtool.c | 30 ++++++++++++++++++++++ 3 files changed, 38 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 16b47fce540b..b08853f71b2b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -1174,11 +1174,14 @@ mlxsw_sp_port_speed_by_width_set(struct mlxsw_sp_port *mlxsw_sp_port) u32 eth_proto_cap, eth_proto_admin, eth_proto_oper; const struct mlxsw_sp_port_type_speed_ops *ops; char ptys_pl[MLXSW_REG_PTYS_LEN]; + u32 eth_proto_cap_masked; int err; ops = mlxsw_sp->port_type_speed_ops; - /* Set advertised speeds to supported speeds. */ + /* Set advertised speeds to speeds supported by both the driver + * and the device. + */ ops->reg_ptys_eth_pack(mlxsw_sp, ptys_pl, mlxsw_sp_port->local_port, 0, false); err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl); @@ -1187,8 +1190,10 @@ mlxsw_sp_port_speed_by_width_set(struct mlxsw_sp_port *mlxsw_sp_port) ops->reg_ptys_eth_unpack(mlxsw_sp, ptys_pl, ð_proto_cap, ð_proto_admin, ð_proto_oper); + eth_proto_cap_masked = ops->ptys_proto_cap_masked_get(eth_proto_cap); ops->reg_ptys_eth_pack(mlxsw_sp, ptys_pl, mlxsw_sp_port->local_port, - eth_proto_cap, mlxsw_sp_port->link.autoneg); + eth_proto_cap_masked, + mlxsw_sp_port->link.autoneg); return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl); } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 3e26eb6cb140..74b3959b36d4 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -342,6 +342,7 @@ struct mlxsw_sp_port_type_speed_ops { u32 *p_eth_proto_cap, u32 *p_eth_proto_admin, u32 *p_eth_proto_oper); + u32 (*ptys_proto_cap_masked_get)(u32 eth_proto_cap); }; static inline struct net_device * diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c index 2096b6478958..540616469e28 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c @@ -1303,6 +1303,20 @@ mlxsw_sp1_reg_ptys_eth_unpack(struct mlxsw_sp *mlxsw_sp, char *payload, p_eth_proto_oper); } +static u32 mlxsw_sp1_ptys_proto_cap_masked_get(u32 eth_proto_cap) +{ + u32 ptys_proto_cap_masked = 0; + int i; + + for (i = 0; i < MLXSW_SP1_PORT_LINK_MODE_LEN; i++) { + if (mlxsw_sp1_port_link_mode[i].mask & eth_proto_cap) + ptys_proto_cap_masked |= + mlxsw_sp1_port_link_mode[i].mask; + } + + return ptys_proto_cap_masked; +} + const struct mlxsw_sp_port_type_speed_ops mlxsw_sp1_port_type_speed_ops = { .from_ptys_supported_port = mlxsw_sp1_from_ptys_supported_port, .from_ptys_link = mlxsw_sp1_from_ptys_link, @@ -1313,6 +1327,7 @@ const struct mlxsw_sp_port_type_speed_ops mlxsw_sp1_port_type_speed_ops = { .to_ptys_speed = mlxsw_sp1_to_ptys_speed, .reg_ptys_eth_pack = mlxsw_sp1_reg_ptys_eth_pack, .reg_ptys_eth_unpack = mlxsw_sp1_reg_ptys_eth_unpack, + .ptys_proto_cap_masked_get = mlxsw_sp1_ptys_proto_cap_masked_get, }; static const enum ethtool_link_mode_bit_indices @@ -1731,6 +1746,20 @@ mlxsw_sp2_reg_ptys_eth_unpack(struct mlxsw_sp *mlxsw_sp, char *payload, p_eth_proto_admin, p_eth_proto_oper); } +static u32 mlxsw_sp2_ptys_proto_cap_masked_get(u32 eth_proto_cap) +{ + u32 ptys_proto_cap_masked = 0; + int i; + + for (i = 0; i < MLXSW_SP2_PORT_LINK_MODE_LEN; i++) { + if (mlxsw_sp2_port_link_mode[i].mask & eth_proto_cap) + ptys_proto_cap_masked |= + mlxsw_sp2_port_link_mode[i].mask; + } + + return ptys_proto_cap_masked; +} + const struct mlxsw_sp_port_type_speed_ops mlxsw_sp2_port_type_speed_ops = { .from_ptys_supported_port = mlxsw_sp2_from_ptys_supported_port, .from_ptys_link = mlxsw_sp2_from_ptys_link, @@ -1741,4 +1770,5 @@ const struct mlxsw_sp_port_type_speed_ops mlxsw_sp2_port_type_speed_ops = { .to_ptys_speed = mlxsw_sp2_to_ptys_speed, .reg_ptys_eth_pack = mlxsw_sp2_reg_ptys_eth_pack, .reg_ptys_eth_unpack = mlxsw_sp2_reg_ptys_eth_unpack, + .ptys_proto_cap_masked_get = mlxsw_sp2_ptys_proto_cap_masked_get, }; -- cgit v1.2.3 From adc80b6cfedff6dad8b93d46a5ea2775fd5af9ec Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Sat, 24 Oct 2020 16:37:32 +0300 Subject: mlxsw: core: Fix memory leak on module removal Free the devlink instance during the teardown sequence in the non-reload case to avoid the following memory leak. unreferenced object 0xffff888232895000 (size 2048): comm "modprobe", pid 1073, jiffies 4295568857 (age 164.871s) hex dump (first 32 bytes): 00 01 00 00 00 00 ad de 22 01 00 00 00 00 ad de ........"....... 10 50 89 32 82 88 ff ff 10 50 89 32 82 88 ff ff .P.2.....P.2.... backtrace: [<00000000c704e9a6>] __kmalloc+0x13a/0x2a0 [<00000000ee30129d>] devlink_alloc+0xff/0x760 [<0000000092ab3e5d>] 0xffffffffa042e5b0 [<000000004f3f8a31>] 0xffffffffa042f6ad [<0000000092800b4b>] 0xffffffffa0491df3 [<00000000c4843903>] local_pci_probe+0xcb/0x170 [<000000006993ded7>] pci_device_probe+0x2c2/0x4e0 [<00000000a8e0de75>] really_probe+0x2c5/0xf90 [<00000000d42ba75d>] driver_probe_device+0x1eb/0x340 [<00000000bcc95e05>] device_driver_attach+0x294/0x300 [<000000000e2bc177>] __driver_attach+0x167/0x2f0 [<000000007d44cd6e>] bus_for_each_dev+0x148/0x1f0 [<000000003cd5a91e>] driver_attach+0x45/0x60 [<000000000041ce51>] bus_add_driver+0x3b8/0x720 [<00000000f5215476>] driver_register+0x230/0x4e0 [<00000000d79356f5>] __pci_register_driver+0x190/0x200 Fixes: a22712a96291 ("mlxsw: core: Fix devlink unregister flow") Signed-off-by: Ido Schimmel Reported-by: Vadim Pasternak Tested-by: Oleksandr Shamray Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlxsw/core.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c index 7f77c2a71d1c..a2efbb1e3cf4 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core.c @@ -2064,6 +2064,8 @@ void mlxsw_core_bus_device_unregister(struct mlxsw_core *mlxsw_core, if (!reload) devlink_resources_unregister(devlink, NULL); mlxsw_core->bus->fini(mlxsw_core->bus_priv); + if (!reload) + devlink_free(devlink); return; -- cgit v1.2.3 From 0daf2bf5a2dcf33d446b76360908f109816e2e21 Mon Sep 17 00:00:00 2001 From: Amit Cohen Date: Sat, 24 Oct 2020 16:37:33 +0300 Subject: mlxsw: core: Fix use-after-free in mlxsw_emad_trans_finish() Each EMAD transaction stores the skb used to issue the EMAD request ('trans->tx_skb') so that the request could be retried in case of a timeout. The skb can be freed when a corresponding response is received or as part of the retry logic (e.g., failed retransmit, exceeded maximum number of retries). The two tasks (i.e., response processing and retransmits) are synchronized by the atomic 'trans->active' field which ensures that responses to inactive transactions are ignored. In case of a failed retransmit the transaction is finished and all of its resources are freed. However, the current code does not mark it as inactive. Syzkaller was able to hit a race condition in which a concurrent response is processed while the transaction's resources are being freed, resulting in a use-after-free [1]. Fix the issue by making sure to mark the transaction as inactive after a failed retransmit and free its resources only if a concurrent task did not already do that. [1] BUG: KASAN: use-after-free in consume_skb+0x30/0x370 net/core/skbuff.c:833 Read of size 4 at addr ffff88804f570494 by task syz-executor.0/1004 CPU: 0 PID: 1004 Comm: syz-executor.0 Not tainted 5.8.0-rc7+ #68 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0xf6/0x16e lib/dump_stack.c:118 print_address_description.constprop.0+0x1c/0x250 mm/kasan/report.c:383 __kasan_report mm/kasan/report.c:513 [inline] kasan_report.cold+0x1f/0x37 mm/kasan/report.c:530 check_memory_region_inline mm/kasan/generic.c:186 [inline] check_memory_region+0x14e/0x1b0 mm/kasan/generic.c:192 instrument_atomic_read include/linux/instrumented.h:56 [inline] atomic_read include/asm-generic/atomic-instrumented.h:27 [inline] refcount_read include/linux/refcount.h:147 [inline] skb_unref include/linux/skbuff.h:1044 [inline] consume_skb+0x30/0x370 net/core/skbuff.c:833 mlxsw_emad_trans_finish+0x64/0x1c0 drivers/net/ethernet/mellanox/mlxsw/core.c:592 mlxsw_emad_process_response drivers/net/ethernet/mellanox/mlxsw/core.c:651 [inline] mlxsw_emad_rx_listener_func+0x5c9/0xac0 drivers/net/ethernet/mellanox/mlxsw/core.c:672 mlxsw_core_skb_receive+0x4df/0x770 drivers/net/ethernet/mellanox/mlxsw/core.c:2063 mlxsw_pci_cqe_rdq_handle drivers/net/ethernet/mellanox/mlxsw/pci.c:595 [inline] mlxsw_pci_cq_tasklet+0x12a6/0x2520 drivers/net/ethernet/mellanox/mlxsw/pci.c:651 tasklet_action_common.isra.0+0x13f/0x3e0 kernel/softirq.c:550 __do_softirq+0x223/0x964 kernel/softirq.c:292 asm_call_on_stack+0x12/0x20 arch/x86/entry/entry_64.S:711 Allocated by task 1006: save_stack+0x1b/0x40 mm/kasan/common.c:48 set_track mm/kasan/common.c:56 [inline] __kasan_kmalloc mm/kasan/common.c:494 [inline] __kasan_kmalloc.constprop.0+0xc2/0xd0 mm/kasan/common.c:467 slab_post_alloc_hook mm/slab.h:586 [inline] slab_alloc_node mm/slub.c:2824 [inline] slab_alloc mm/slub.c:2832 [inline] kmem_cache_alloc+0xcd/0x2e0 mm/slub.c:2837 __build_skb+0x21/0x60 net/core/skbuff.c:311 __netdev_alloc_skb+0x1e2/0x360 net/core/skbuff.c:464 netdev_alloc_skb include/linux/skbuff.h:2810 [inline] mlxsw_emad_alloc drivers/net/ethernet/mellanox/mlxsw/core.c:756 [inline] mlxsw_emad_reg_access drivers/net/ethernet/mellanox/mlxsw/core.c:787 [inline] mlxsw_core_reg_access_emad+0x1ab/0x1420 drivers/net/ethernet/mellanox/mlxsw/core.c:1817 mlxsw_reg_trans_query+0x39/0x50 drivers/net/ethernet/mellanox/mlxsw/core.c:1831 mlxsw_sp_sb_pm_occ_clear drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c:260 [inline] mlxsw_sp_sb_occ_max_clear+0xbff/0x10a0 drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c:1365 mlxsw_devlink_sb_occ_max_clear+0x76/0xb0 drivers/net/ethernet/mellanox/mlxsw/core.c:1037 devlink_nl_cmd_sb_occ_max_clear_doit+0x1ec/0x280 net/core/devlink.c:1765 genl_family_rcv_msg_doit net/netlink/genetlink.c:669 [inline] genl_family_rcv_msg net/netlink/genetlink.c:714 [inline] genl_rcv_msg+0x617/0x980 net/netlink/genetlink.c:731 netlink_rcv_skb+0x152/0x440 net/netlink/af_netlink.c:2470 genl_rcv+0x24/0x40 net/netlink/genetlink.c:742 netlink_unicast_kernel net/netlink/af_netlink.c:1304 [inline] netlink_unicast+0x53a/0x750 net/netlink/af_netlink.c:1330 netlink_sendmsg+0x850/0xd90 net/netlink/af_netlink.c:1919 sock_sendmsg_nosec net/socket.c:651 [inline] sock_sendmsg+0x150/0x190 net/socket.c:671 ____sys_sendmsg+0x6d8/0x840 net/socket.c:2359 ___sys_sendmsg+0xff/0x170 net/socket.c:2413 __sys_sendmsg+0xe5/0x1b0 net/socket.c:2446 do_syscall_64+0x56/0xa0 arch/x86/entry/common.c:384 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Freed by task 73: save_stack+0x1b/0x40 mm/kasan/common.c:48 set_track mm/kasan/common.c:56 [inline] kasan_set_free_info mm/kasan/common.c:316 [inline] __kasan_slab_free+0x12c/0x170 mm/kasan/common.c:455 slab_free_hook mm/slub.c:1474 [inline] slab_free_freelist_hook mm/slub.c:1507 [inline] slab_free mm/slub.c:3072 [inline] kmem_cache_free+0xbe/0x380 mm/slub.c:3088 kfree_skbmem net/core/skbuff.c:622 [inline] kfree_skbmem+0xef/0x1b0 net/core/skbuff.c:616 __kfree_skb net/core/skbuff.c:679 [inline] consume_skb net/core/skbuff.c:837 [inline] consume_skb+0xe1/0x370 net/core/skbuff.c:831 mlxsw_emad_trans_finish+0x64/0x1c0 drivers/net/ethernet/mellanox/mlxsw/core.c:592 mlxsw_emad_transmit_retry.isra.0+0x9d/0xc0 drivers/net/ethernet/mellanox/mlxsw/core.c:613 mlxsw_emad_trans_timeout_work+0x43/0x50 drivers/net/ethernet/mellanox/mlxsw/core.c:625 process_one_work+0xa3e/0x17a0 kernel/workqueue.c:2269 worker_thread+0x9e/0x1050 kernel/workqueue.c:2415 kthread+0x355/0x470 kernel/kthread.c:291 ret_from_fork+0x22/0x30 arch/x86/entry/entry_64.S:293 The buggy address belongs to the object at ffff88804f5703c0 which belongs to the cache skbuff_head_cache of size 224 The buggy address is located 212 bytes inside of 224-byte region [ffff88804f5703c0, ffff88804f5704a0) The buggy address belongs to the page: page:ffffea00013d5c00 refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 flags: 0x100000000000200(slab) raw: 0100000000000200 dead000000000100 dead000000000122 ffff88806c625400 raw: 0000000000000000 00000000000c000c 00000001ffffffff 0000000000000000 page dumped because: kasan: bad access detected Memory state around the buggy address: ffff88804f570380: fc fc fc fc fc fc fc fc fb fb fb fb fb fb fb fb ffff88804f570400: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb >ffff88804f570480: fb fb fb fb fc fc fc fc fc fc fc fc fc fc fc fc ^ ffff88804f570500: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ffff88804f570580: 00 00 00 00 00 00 00 00 00 00 00 00 fc fc fc fc Fixes: caf7297e7ab5f ("mlxsw: core: Introduce support for asynchronous EMAD register access") Signed-off-by: Amit Cohen Reviewed-by: Jiri Pirko Signed-off-by: Ido Schimmel Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlxsw/core.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c index a2efbb1e3cf4..937b8e46f8c7 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core.c @@ -620,6 +620,9 @@ static void mlxsw_emad_transmit_retry(struct mlxsw_core *mlxsw_core, err = mlxsw_emad_transmit(trans->core, trans); if (err == 0) return; + + if (!atomic_dec_and_test(&trans->active)) + return; } else { err = -EIO; } -- cgit v1.2.3 From 21d6a11e2cadfb8446265a3efff0e2aad206e15e Mon Sep 17 00:00:00 2001 From: Vasundhara Volam Date: Mon, 26 Oct 2020 00:18:17 -0400 Subject: bnxt_en: Fix regression in workqueue cleanup logic in bnxt_remove_one(). A recent patch has moved the workqueue cleanup logic before calling unregister_netdev() in bnxt_remove_one(). This caused a regression because the workqueue can be restarted if the device is still open. Workqueue cleanup must be done after unregister_netdev(). The workqueue will not restart itself after the device is closed. Call bnxt_cancel_sp_work() after unregister_netdev() and call bnxt_dl_fw_reporters_destroy() after that. This fixes the regession and the original NULL ptr dereference issue. Fixes: b16939b59cc0 ("bnxt_en: Fix NULL ptr dereference crash in bnxt_fw_reset_task()") Signed-off-by: Vasundhara Volam Signed-off-by: Michael Chan Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index fa147865e33f..e4e5ea080391 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -12108,15 +12108,16 @@ static void bnxt_remove_one(struct pci_dev *pdev) if (BNXT_PF(bp)) bnxt_sriov_disable(bp); + if (BNXT_PF(bp)) + devlink_port_type_clear(&bp->dl_port); + pci_disable_pcie_error_reporting(pdev); + unregister_netdev(dev); clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state); + /* Flush any pending tasks */ bnxt_cancel_sp_work(bp); bp->sp_event = 0; bnxt_dl_fw_reporters_destroy(bp, true); - if (BNXT_PF(bp)) - devlink_port_type_clear(&bp->dl_port); - pci_disable_pcie_error_reporting(pdev); - unregister_netdev(dev); bnxt_dl_unregister(bp); bnxt_shutdown_tc(bp); -- cgit v1.2.3 From 631ce27a3006fc0b732bfd589c6df505f62eadd9 Mon Sep 17 00:00:00 2001 From: Vasundhara Volam Date: Mon, 26 Oct 2020 00:18:18 -0400 Subject: bnxt_en: Invoke cancel_delayed_work_sync() for PFs also. As part of the commit b148bb238c02 ("bnxt_en: Fix possible crash in bnxt_fw_reset_task()."), cancel_delayed_work_sync() is called only for VFs to fix a possible crash by cancelling any pending delayed work items. It was assumed by mistake that the flush_workqueue() call on the PF would flush delayed work items as well. As flush_workqueue() does not cancel the delayed workqueue, extend the fix for PFs. This fix will avoid the system crash, if there are any pending delayed work items in fw_reset_task() during driver's .remove() call. Unify the workqueue cleanup logic for both PF and VF by calling cancel_work_sync() and cancel_delayed_work_sync() directly in bnxt_remove_one(). Fixes: b148bb238c02 ("bnxt_en: Fix possible crash in bnxt_fw_reset_task().") Reviewed-by: Pavan Chebbi Reviewed-by: Andy Gospodarek Signed-off-by: Vasundhara Volam Signed-off-by: Michael Chan Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 13 ++----------- 1 file changed, 2 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index e4e5ea080391..7be232018015 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -1160,16 +1160,6 @@ static void bnxt_queue_sp_work(struct bnxt *bp) schedule_work(&bp->sp_task); } -static void bnxt_cancel_sp_work(struct bnxt *bp) -{ - if (BNXT_PF(bp)) { - flush_workqueue(bnxt_pf_wq); - } else { - cancel_work_sync(&bp->sp_task); - cancel_delayed_work_sync(&bp->fw_reset_task); - } -} - static void bnxt_sched_reset(struct bnxt *bp, struct bnxt_rx_ring_info *rxr) { if (!rxr->bnapi->in_reset) { @@ -12114,7 +12104,8 @@ static void bnxt_remove_one(struct pci_dev *pdev) unregister_netdev(dev); clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state); /* Flush any pending tasks */ - bnxt_cancel_sp_work(bp); + cancel_work_sync(&bp->sp_task); + cancel_delayed_work_sync(&bp->fw_reset_task); bp->sp_event = 0; bnxt_dl_fw_reporters_destroy(bp, true); -- cgit v1.2.3 From f75d9a0aa96721d20011cd5f8c7a24eb32728589 Mon Sep 17 00:00:00 2001 From: Vasundhara Volam Date: Mon, 26 Oct 2020 00:18:19 -0400 Subject: bnxt_en: Re-write PCI BARs after PCI fatal error. When a PCIe fatal error occurs, the internal latched BAR addresses in the chip get reset even though the BAR register values in config space are retained. pci_restore_state() will not rewrite the BAR addresses if the BAR address values are valid, causing the chip's internal BAR addresses to stay invalid. So we need to zero the BAR registers during PCIe fatal error to force pci_restore_state() to restore the BAR addresses. These write cycles to the BAR registers will cause the proper BAR addresses to latch internally. Fixes: 6316ea6db93d ("bnxt_en: Enable AER support.") Signed-off-by: Vasundhara Volam Signed-off-by: Michael Chan Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 19 ++++++++++++++++++- drivers/net/ethernet/broadcom/bnxt/bnxt.h | 1 + 2 files changed, 19 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 7be232018015..8012386b4a0f 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -12852,6 +12852,9 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev, return PCI_ERS_RESULT_DISCONNECT; } + if (state == pci_channel_io_frozen) + set_bit(BNXT_STATE_PCI_CHANNEL_IO_FROZEN, &bp->state); + if (netif_running(netdev)) bnxt_close(netdev); @@ -12878,7 +12881,7 @@ static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev) { struct net_device *netdev = pci_get_drvdata(pdev); struct bnxt *bp = netdev_priv(netdev); - int err = 0; + int err = 0, off; pci_ers_result_t result = PCI_ERS_RESULT_DISCONNECT; netdev_info(bp->dev, "PCI Slot Reset\n"); @@ -12890,6 +12893,20 @@ static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev) "Cannot re-enable PCI device after reset.\n"); } else { pci_set_master(pdev); + /* Upon fatal error, our device internal logic that latches to + * BAR value is getting reset and will restore only upon + * rewritting the BARs. + * + * As pci_restore_state() does not re-write the BARs if the + * value is same as saved value earlier, driver needs to + * write the BARs to 0 to force restore, in case of fatal error. + */ + if (test_and_clear_bit(BNXT_STATE_PCI_CHANNEL_IO_FROZEN, + &bp->state)) { + for (off = PCI_BASE_ADDRESS_0; + off <= PCI_BASE_ADDRESS_5; off += 4) + pci_write_config_dword(bp->pdev, off, 0); + } pci_restore_state(pdev); pci_save_state(pdev); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 21ef1c21f602..47b3c3127879 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -1781,6 +1781,7 @@ struct bnxt { #define BNXT_STATE_ABORT_ERR 5 #define BNXT_STATE_FW_FATAL_COND 6 #define BNXT_STATE_DRV_REGISTERED 7 +#define BNXT_STATE_PCI_CHANNEL_IO_FROZEN 8 #define BNXT_NO_FW_ACCESS(bp) \ (test_bit(BNXT_STATE_FW_FATAL_COND, &(bp)->state) || \ -- cgit v1.2.3 From a1301f08c5acf992d9c1fafddc84c3a822844b04 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 26 Oct 2020 00:18:20 -0400 Subject: bnxt_en: Check abort error state in bnxt_open_nic(). bnxt_open_nic() is called during configuration changes that require the NIC to be closed and then opened. This call is protected by rtnl_lock. Firmware reset can be happening at the same time. Only critical portions of the entire firmware reset sequence are protected by the rtnl_lock. It is possible that bnxt_open_nic() can be called when the firmware reset sequence is aborting. In that case, bnxt_open_nic() needs to check if the ABORT_ERR flag is set and abort if it is. The configuration change that resulted in the bnxt_open_nic() call will fail but the NIC will be brought to a consistent IF_DOWN state. Without this patch, if bnxt_open_nic() were to continue in this error state, it may crash like this: [ 1648.659736] BUG: unable to handle kernel NULL pointer dereference at (null) [ 1648.659768] IP: [] bnxt_alloc_mem+0x50a/0x1140 [bnxt_en] [ 1648.659796] PGD 101e1b3067 PUD 101e1b2067 PMD 0 [ 1648.659813] Oops: 0000 [#1] SMP [ 1648.659825] Modules linked in: xt_CHECKSUM iptable_mangle ipt_MASQUERADE nf_nat_masquerade_ipv4 iptable_nat nf_nat_ipv4 nf_nat nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack ipt_REJECT nf_reject_ipv4 tun bridge stp llc ebtable_filter ebtables ip6table_filter ip6_tables iptable_filter sunrpc dell_smbios dell_wmi_descriptor dcdbas amd64_edac_mod edac_mce_amd kvm_amd kvm irqbypass crc32_pclmul ghash_clmulni_intel aesni_intel lrw gf128mul glue_helper ablk_helper vfat cryptd fat pcspkr ipmi_ssif sg k10temp i2c_piix4 wmi ipmi_si ipmi_devintf ipmi_msghandler tpm_crb acpi_power_meter sch_fq_codel ip_tables xfs libcrc32c sd_mod crc_t10dif crct10dif_generic mgag200 i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ttm ahci drm libahci megaraid_sas crct10dif_pclmul crct10dif_common [ 1648.660063] tg3 libata crc32c_intel bnxt_en(OE) drm_panel_orientation_quirks devlink ptp pps_core dm_mirror dm_region_hash dm_log dm_mod fuse [ 1648.660105] CPU: 13 PID: 3867 Comm: ethtool Kdump: loaded Tainted: G OE ------------ 3.10.0-1152.el7.x86_64 #1 [ 1648.660911] Hardware name: Dell Inc. PowerEdge R7515/0R4CNN, BIOS 1.2.14 01/28/2020 [ 1648.661662] task: ffff94e64cbc9080 ti: ffff94f55df1c000 task.ti: ffff94f55df1c000 [ 1648.662409] RIP: 0010:[] [] bnxt_alloc_mem+0x50a/0x1140 [bnxt_en] [ 1648.663171] RSP: 0018:ffff94f55df1fba8 EFLAGS: 00010202 [ 1648.663927] RAX: 0000000000000000 RBX: ffff94e6827e0000 RCX: 0000000000000000 [ 1648.664684] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff94e6827e08c0 [ 1648.665433] RBP: ffff94f55df1fc20 R08: 00000000000001ff R09: 0000000000000008 [ 1648.666184] R10: 0000000000000d53 R11: ffff94f55df1f7ce R12: ffff94e6827e08c0 [ 1648.666940] R13: ffff94e6827e08c0 R14: ffff94e6827e08c0 R15: ffffffffb9115e40 [ 1648.667695] FS: 00007f8aadba5740(0000) GS:ffff94f57eb40000(0000) knlGS:0000000000000000 [ 1648.668447] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 1648.669202] CR2: 0000000000000000 CR3: 0000001022772000 CR4: 0000000000340fe0 [ 1648.669966] Call Trace: [ 1648.670730] [] ? bnxt_need_reserve_rings+0x9d/0x170 [bnxt_en] [ 1648.671496] [] __bnxt_open_nic+0x8a/0x9a0 [bnxt_en] [ 1648.672263] [] ? bnxt_close_nic+0x59/0x1b0 [bnxt_en] [ 1648.673031] [] bnxt_open_nic+0x1b/0x50 [bnxt_en] [ 1648.673793] [] bnxt_set_ringparam+0x6c/0xa0 [bnxt_en] [ 1648.674550] [] dev_ethtool+0x1334/0x21a0 [ 1648.675306] [] dev_ioctl+0x1ef/0x5f0 [ 1648.676061] [] sock_do_ioctl+0x4d/0x60 [ 1648.676810] [] sock_ioctl+0x1eb/0x2d0 [ 1648.677548] [] do_vfs_ioctl+0x3a0/0x5b0 [ 1648.678282] [] ? __do_page_fault+0x238/0x500 [ 1648.679016] [] SyS_ioctl+0xa1/0xc0 [ 1648.679745] [] system_call_fastpath+0x25/0x2a [ 1648.680461] Code: 9e 60 01 00 00 0f 1f 40 00 45 8b 8e 48 01 00 00 31 c9 45 85 c9 0f 8e 73 01 00 00 66 0f 1f 44 00 00 49 8b 86 a8 00 00 00 48 63 d1 <48> 8b 14 d0 48 85 d2 0f 84 46 01 00 00 41 8b 86 44 01 00 00 c7 [ 1648.681986] RIP [] bnxt_alloc_mem+0x50a/0x1140 [bnxt_en] [ 1648.682724] RSP [ 1648.683451] CR2: 0000000000000000 Fixes: ec5d31e3c15d ("bnxt_en: Handle firmware reset status during IF_UP.") Reviewed-by: Vasundhara Volam Reviewed-by: Pavan Chebbi Signed-off-by: Michael Chan Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 8012386b4a0f..0165f70dba74 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -9779,7 +9779,10 @@ int bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init) { int rc = 0; - rc = __bnxt_open_nic(bp, irq_re_init, link_re_init); + if (test_bit(BNXT_STATE_ABORT_ERR, &bp->state)) + rc = -EIO; + if (!rc) + rc = __bnxt_open_nic(bp, irq_re_init, link_re_init); if (rc) { netdev_err(bp->dev, "nic open fail (rc: %x)\n", rc); dev_close(bp->dev); -- cgit v1.2.3 From 825741b071722f1c8ad692cead562c4b5f5eaa93 Mon Sep 17 00:00:00 2001 From: Vasundhara Volam Date: Mon, 26 Oct 2020 00:18:21 -0400 Subject: bnxt_en: Send HWRM_FUNC_RESET fw command unconditionally. In the AER or firmware reset flow, if we are in fatal error state or if pci_channel_offline() is true, we don't send any commands to the firmware because the commands will likely not reach the firmware and most commands don't matter much because the firmware is likely to be reset imminently. However, the HWRM_FUNC_RESET command is different and we should always attempt to send it. In the AER flow for example, the .slot_reset() call will trigger this fw command and we need to try to send it to effect the proper reset. Fixes: b340dc680ed4 ("bnxt_en: Avoid sending firmware messages when AER error is detected.") Reviewed-by: Edwin Peer Signed-off-by: Vasundhara Volam Signed-off-by: Michael Chan Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 0165f70dba74..7975f59735d6 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -4352,7 +4352,8 @@ static int bnxt_hwrm_do_send_msg(struct bnxt *bp, void *msg, u32 msg_len, u32 bar_offset = BNXT_GRCPF_REG_CHIMP_COMM; u16 dst = BNXT_HWRM_CHNL_CHIMP; - if (BNXT_NO_FW_ACCESS(bp)) + if (BNXT_NO_FW_ACCESS(bp) && + le16_to_cpu(req->req_type) != HWRM_FUNC_RESET) return -EBUSY; if (msg_len > BNXT_HWRM_MAX_REQ_LEN) { -- cgit v1.2.3 From e3364c5ff3ff975b943a7bf47e21a2a4bf20f3fe Mon Sep 17 00:00:00 2001 From: Zenghui Yu Date: Fri, 23 Oct 2020 13:15:50 +0800 Subject: net: hns3: Clear the CMDQ registers before unmapping BAR region When unbinding the hns3 driver with the HNS3 VF, I got the following kernel panic: [ 265.709989] Unable to handle kernel paging request at virtual address ffff800054627000 [ 265.717928] Mem abort info: [ 265.720740] ESR = 0x96000047 [ 265.723810] EC = 0x25: DABT (current EL), IL = 32 bits [ 265.729126] SET = 0, FnV = 0 [ 265.732195] EA = 0, S1PTW = 0 [ 265.735351] Data abort info: [ 265.738227] ISV = 0, ISS = 0x00000047 [ 265.742071] CM = 0, WnR = 1 [ 265.745055] swapper pgtable: 4k pages, 48-bit VAs, pgdp=0000000009b54000 [ 265.751753] [ffff800054627000] pgd=0000202ffffff003, p4d=0000202ffffff003, pud=00002020020eb003, pmd=00000020a0dfc003, pte=0000000000000000 [ 265.764314] Internal error: Oops: 96000047 [#1] SMP [ 265.830357] CPU: 61 PID: 20319 Comm: bash Not tainted 5.9.0+ #206 [ 265.836423] Hardware name: Huawei TaiShan 2280 V2/BC82AMDDA, BIOS 1.05 09/18/2019 [ 265.843873] pstate: 80400009 (Nzcv daif +PAN -UAO -TCO BTYPE=--) [ 265.843890] pc : hclgevf_cmd_uninit+0xbc/0x300 [ 265.861988] lr : hclgevf_cmd_uninit+0xb0/0x300 [ 265.861992] sp : ffff80004c983b50 [ 265.881411] pmr_save: 000000e0 [ 265.884453] x29: ffff80004c983b50 x28: ffff20280bbce500 [ 265.889744] x27: 0000000000000000 x26: 0000000000000000 [ 265.895034] x25: ffff800011a1f000 x24: ffff800011a1fe90 [ 265.900325] x23: ffff0020ce9b00d8 x22: ffff0020ce9b0150 [ 265.905616] x21: ffff800010d70e90 x20: ffff800010d70e90 [ 265.910906] x19: ffff0020ce9b0080 x18: 0000000000000004 [ 265.916198] x17: 0000000000000000 x16: ffff800011ae32e8 [ 265.916201] x15: 0000000000000028 x14: 0000000000000002 [ 265.916204] x13: ffff800011ae32e8 x12: 0000000000012ad8 [ 265.946619] x11: ffff80004c983b50 x10: 0000000000000000 [ 265.951911] x9 : ffff8000115d0888 x8 : 0000000000000000 [ 265.951914] x7 : ffff800011890b20 x6 : c0000000ffff7fff [ 265.951917] x5 : ffff80004c983930 x4 : 0000000000000001 [ 265.951919] x3 : ffffa027eec1b000 x2 : 2b78ccbbff369100 [ 265.964487] x1 : 0000000000000000 x0 : ffff800054627000 [ 265.964491] Call trace: [ 265.964494] hclgevf_cmd_uninit+0xbc/0x300 [ 265.964496] hclgevf_uninit_ae_dev+0x9c/0xe8 [ 265.964501] hnae3_unregister_ae_dev+0xb0/0x130 [ 265.964516] hns3_remove+0x34/0x88 [hns3] [ 266.009683] pci_device_remove+0x48/0xf0 [ 266.009692] device_release_driver_internal+0x114/0x1e8 [ 266.030058] device_driver_detach+0x28/0x38 [ 266.034224] unbind_store+0xd4/0x108 [ 266.037784] drv_attr_store+0x40/0x58 [ 266.041435] sysfs_kf_write+0x54/0x80 [ 266.045081] kernfs_fop_write+0x12c/0x250 [ 266.049076] vfs_write+0xc4/0x248 [ 266.052378] ksys_write+0x74/0xf8 [ 266.055677] __arm64_sys_write+0x24/0x30 [ 266.059584] el0_svc_common.constprop.3+0x84/0x270 [ 266.064354] do_el0_svc+0x34/0xa0 [ 266.067658] el0_svc+0x38/0x40 [ 266.070700] el0_sync_handler+0x8c/0xb0 [ 266.074519] el0_sync+0x140/0x180 It looks like the BAR memory region had already been unmapped before we start clearing CMDQ registers in it, which is pretty bad and the kernel happily kills itself because of a Current EL Data Abort (on arm64). Moving the CMDQ uninitialization a bit early fixes the issue for me. Fixes: 862d969a3a4d ("net: hns3: do VF's pci re-initialization while PF doing FLR") Signed-off-by: Zenghui Yu Link: https://lore.kernel.org/r/20201023051550.793-1-yuzenghui@huawei.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c index 50c84c5e65d2..c8e3fdd5999c 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c @@ -3262,8 +3262,8 @@ static void hclgevf_uninit_hdev(struct hclgevf_dev *hdev) hclgevf_uninit_msi(hdev); } - hclgevf_pci_uninit(hdev); hclgevf_cmd_uninit(hdev); + hclgevf_pci_uninit(hdev); hclgevf_uninit_mac_list(hdev); } -- cgit v1.2.3 From 8685699c28d7452ff06d62b4692df985eb7301f0 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Fri, 23 Oct 2020 12:16:28 -0700 Subject: nvme: ignore zone validate errors on subsequent scans Revalidating nvme zoned namespaces requires IO commands, and there are controller states that prevent IO. For example, a sanitize in progress is required to fail all IO, but we don't want to remove a namespace we've previously added just because the controller is in such a state. Suppress the error in this case. Reported-by: Michael Nguyen Signed-off-by: Keith Busch Reviewed-by: Chaitanya Kulkarni Signed-off-by: Christoph Hellwig --- drivers/nvme/host/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index 95ef4943d8bd..376096bfc54a 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -2125,7 +2125,7 @@ static int nvme_update_ns_info(struct nvme_ns *ns, struct nvme_id_ns *id) if (blk_queue_is_zoned(ns->queue)) { ret = nvme_revalidate_zones(ns); - if (ret) + if (ret && !nvme_first_scan(ns->disk)) return ret; } -- cgit v1.2.3 From 25c1ca6ecaba3b751d3f7ff92d5cddff3b05f8d0 Mon Sep 17 00:00:00 2001 From: zhenwei pi Date: Sun, 25 Oct 2020 19:51:24 +0800 Subject: nvme-rdma: handle unexpected nvme completion data length Receiving a zero length message leads to the following warnings because the CQE is processed twice: refcount_t: underflow; use-after-free. WARNING: CPU: 0 PID: 0 at lib/refcount.c:28 RIP: 0010:refcount_warn_saturate+0xd9/0xe0 Call Trace: nvme_rdma_recv_done+0xf3/0x280 [nvme_rdma] __ib_process_cq+0x76/0x150 [ib_core] ... Sanity check the received data length, to avoids this. Thanks to Chao Leng & Sagi for suggestions. Signed-off-by: zhenwei pi Reviewed-by: Sagi Grimberg Signed-off-by: Christoph Hellwig --- drivers/nvme/host/rdma.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index aad829a2b50d..40a0a3b6476c 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -1768,6 +1768,14 @@ static void nvme_rdma_recv_done(struct ib_cq *cq, struct ib_wc *wc) return; } + /* sanity checking for received data length */ + if (unlikely(wc->byte_len < len)) { + dev_err(queue->ctrl->ctrl.device, + "Unexpected nvme completion length(%d)\n", wc->byte_len); + nvme_rdma_error_recovery(queue->ctrl); + return; + } + ib_dma_sync_single_for_cpu(ibdev, qe->dma, len, DMA_FROM_DEVICE); /* * AEN requests are special as they don't time out and can -- cgit v1.2.3 From caf1cbe3677576ef860f7bbba9074d0e6a0bcabb Mon Sep 17 00:00:00 2001 From: James Smart Date: Fri, 23 Oct 2020 15:26:04 -0700 Subject: nvme-fc: track error_recovery while connecting Whenever there are errors during CONNECTING, the driver recovers by aborting all outstanding ios and counts on the io completion to fail them and thus the connection/association they are on. However, the connection failure depends on a failure state from the core routines. Not all commands that are issued by the core routine are guaranteed to cause a failure of the core routine. They may be treated as a failure status and the status is then ignored. As such, whenever the transport enters error_recovery while CONNECTING, it will set a new flag indicating an association failed. The create_association routine which creates and initializes the controller, will monitor the state of the flag as well as the core routine error status and ensure the association fails if there was an error. Signed-off-by: James Smart Signed-off-by: Christoph Hellwig --- drivers/nvme/host/fc.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index 3c002bdcace3..cfb6ef74c742 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -146,7 +146,8 @@ struct nvme_fc_rport { /* fc_ctrl flags values - specified as bit positions */ #define ASSOC_ACTIVE 0 -#define FCCTRL_TERMIO 1 +#define ASSOC_FAILED 1 +#define FCCTRL_TERMIO 2 struct nvme_fc_ctrl { spinlock_t lock; @@ -2988,6 +2989,8 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl) ctrl->cnum, ctrl->lport->localport.port_name, ctrl->rport->remoteport.port_name, ctrl->ctrl.opts->subsysnqn); + clear_bit(ASSOC_FAILED, &ctrl->flags); + /* * Create the admin queue */ @@ -3016,7 +3019,7 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl) */ ret = nvme_enable_ctrl(&ctrl->ctrl); - if (ret) + if (ret || test_bit(ASSOC_FAILED, &ctrl->flags)) goto out_disconnect_admin_queue; ctrl->ctrl.max_segments = ctrl->lport->ops->max_sgl_segments; @@ -3026,7 +3029,7 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl) blk_mq_unquiesce_queue(ctrl->ctrl.admin_q); ret = nvme_init_identify(&ctrl->ctrl); - if (ret) + if (ret || test_bit(ASSOC_FAILED, &ctrl->flags)) goto out_disconnect_admin_queue; /* sanity checks */ @@ -3071,9 +3074,9 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl) ret = nvme_fc_create_io_queues(ctrl); else ret = nvme_fc_recreate_io_queues(ctrl); - if (ret) - goto out_term_aen_ops; } + if (ret || test_bit(ASSOC_FAILED, &ctrl->flags)) + goto out_term_aen_ops; changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE); @@ -3301,6 +3304,7 @@ __nvme_fc_terminate_io(struct nvme_fc_ctrl *ctrl) */ if (ctrl->ctrl.state == NVME_CTRL_CONNECTING) { __nvme_fc_abort_outstanding_ios(ctrl, true); + set_bit(ASSOC_FAILED, &ctrl->flags); return; } -- cgit v1.2.3 From 9c2bb2577d81b1a09f7e342e947986e55cad18e3 Mon Sep 17 00:00:00 2001 From: James Smart Date: Fri, 23 Oct 2020 15:27:50 -0700 Subject: nvme-fc: remove err_work work item err_work was created to handle errors (mainly I/O timeouts) while in CONNECTING state. The flag for err_work_active is also unneeded. Remove err_work_active and err_work. The actions to abort I/Os are moved inline to nvme_error_recovery(). Signed-off-by: James Smart Signed-off-by: Christoph Hellwig --- drivers/nvme/host/fc.c | 40 ++++++++++------------------------------ 1 file changed, 10 insertions(+), 30 deletions(-) diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index cfb6ef74c742..9cdb116b545b 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -158,7 +158,6 @@ struct nvme_fc_ctrl { u32 cnum; bool ioq_live; - atomic_t err_work_active; u64 association_id; struct nvmefc_ls_rcv_op *rcv_disconn; @@ -168,7 +167,6 @@ struct nvme_fc_ctrl { struct blk_mq_tag_set tag_set; struct delayed_work connect_work; - struct work_struct err_work; struct kref ref; unsigned long flags; @@ -2415,11 +2413,11 @@ nvme_fc_nvme_ctrl_freed(struct nvme_ctrl *nctrl) nvme_fc_ctrl_put(ctrl); } +static void __nvme_fc_terminate_io(struct nvme_fc_ctrl *ctrl); + static void nvme_fc_error_recovery(struct nvme_fc_ctrl *ctrl, char *errmsg) { - int active; - /* * if an error (io timeout, etc) while (re)connecting, * it's an error on creating the new association. @@ -2428,11 +2426,14 @@ nvme_fc_error_recovery(struct nvme_fc_ctrl *ctrl, char *errmsg) * ios hitting this path before things are cleaned up. */ if (ctrl->ctrl.state == NVME_CTRL_CONNECTING) { - active = atomic_xchg(&ctrl->err_work_active, 1); - if (!active && !queue_work(nvme_fc_wq, &ctrl->err_work)) { - atomic_set(&ctrl->err_work_active, 0); - WARN_ON(1); - } + __nvme_fc_terminate_io(ctrl); + + /* + * Rescheduling the connection after recovering + * from the io error is left to the reconnect work + * item, which is what should have stalled waiting on + * the io that had the error that scheduled this work. + */ return; } @@ -3240,7 +3241,6 @@ nvme_fc_delete_ctrl(struct nvme_ctrl *nctrl) { struct nvme_fc_ctrl *ctrl = to_fc_ctrl(nctrl); - cancel_work_sync(&ctrl->err_work); cancel_delayed_work_sync(&ctrl->connect_work); /* * kill the association on the link side. this will block @@ -3351,23 +3351,6 @@ nvme_fc_reset_ctrl_work(struct work_struct *work) ctrl->cnum); } -static void -nvme_fc_connect_err_work(struct work_struct *work) -{ - struct nvme_fc_ctrl *ctrl = - container_of(work, struct nvme_fc_ctrl, err_work); - - __nvme_fc_terminate_io(ctrl); - - atomic_set(&ctrl->err_work_active, 0); - - /* - * Rescheduling the connection after recovering - * from the io error is left to the reconnect work - * item, which is what should have stalled waiting on - * the io that had the error that scheduled this work. - */ -} static const struct nvme_ctrl_ops nvme_fc_ctrl_ops = { .name = "fc", @@ -3495,7 +3478,6 @@ nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts, ctrl->dev = lport->dev; ctrl->cnum = idx; ctrl->ioq_live = false; - atomic_set(&ctrl->err_work_active, 0); init_waitqueue_head(&ctrl->ioabort_wait); get_device(ctrl->dev); @@ -3503,7 +3485,6 @@ nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts, INIT_WORK(&ctrl->ctrl.reset_work, nvme_fc_reset_ctrl_work); INIT_DELAYED_WORK(&ctrl->connect_work, nvme_fc_connect_ctrl_work); - INIT_WORK(&ctrl->err_work, nvme_fc_connect_err_work); spin_lock_init(&ctrl->lock); /* io queue count */ @@ -3596,7 +3577,6 @@ nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts, fail_ctrl: nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_DELETING); cancel_work_sync(&ctrl->ctrl.reset_work); - cancel_work_sync(&ctrl->err_work); cancel_delayed_work_sync(&ctrl->connect_work); ctrl->ctrl.opts = NULL; -- cgit v1.2.3 From 95ced8a2c72dd48167adc4981c3c465c491f2f02 Mon Sep 17 00:00:00 2001 From: James Smart Date: Fri, 23 Oct 2020 15:27:51 -0700 Subject: nvme-fc: eliminate terminate_io use by nvme_fc_error_recovery nvme_fc_error_recovery() special cases handling when in CONNECTING state and calls __nvme_fc_terminate_io(). __nvme_fc_terminate_io() itself special cases CONNECTING state and calls the routine to abort outstanding ios. Simplify the sequence by putting the call to abort outstanding I/Os directly in nvme_fc_error_recovery. Move the location of __nvme_fc_abort_outstanding_ios(), and nvme_fc_terminate_exchange() which is called by it, to avoid adding function prototypes for nvme_fc_error_recovery(). Signed-off-by: James Smart Signed-off-by: Christoph Hellwig --- drivers/nvme/host/fc.c | 187 ++++++++++++++++++++++--------------------------- 1 file changed, 84 insertions(+), 103 deletions(-) diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index 9cdb116b545b..ffbfb0533cac 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -2413,27 +2413,97 @@ nvme_fc_nvme_ctrl_freed(struct nvme_ctrl *nctrl) nvme_fc_ctrl_put(ctrl); } -static void __nvme_fc_terminate_io(struct nvme_fc_ctrl *ctrl); +/* + * This routine is used by the transport when it needs to find active + * io on a queue that is to be terminated. The transport uses + * blk_mq_tagset_busy_itr() to find the busy requests, which then invoke + * this routine to kill them on a 1 by 1 basis. + * + * As FC allocates FC exchange for each io, the transport must contact + * the LLDD to terminate the exchange, thus releasing the FC exchange. + * After terminating the exchange the LLDD will call the transport's + * normal io done path for the request, but it will have an aborted + * status. The done path will return the io request back to the block + * layer with an error status. + */ +static bool +nvme_fc_terminate_exchange(struct request *req, void *data, bool reserved) +{ + struct nvme_ctrl *nctrl = data; + struct nvme_fc_ctrl *ctrl = to_fc_ctrl(nctrl); + struct nvme_fc_fcp_op *op = blk_mq_rq_to_pdu(req); + + __nvme_fc_abort_op(ctrl, op); + return true; +} + +/* + * This routine runs through all outstanding commands on the association + * and aborts them. This routine is typically be called by the + * delete_association routine. It is also called due to an error during + * reconnect. In that scenario, it is most likely a command that initializes + * the controller, including fabric Connect commands on io queues, that + * may have timed out or failed thus the io must be killed for the connect + * thread to see the error. + */ +static void +__nvme_fc_abort_outstanding_ios(struct nvme_fc_ctrl *ctrl, bool start_queues) +{ + /* + * If io queues are present, stop them and terminate all outstanding + * ios on them. As FC allocates FC exchange for each io, the + * transport must contact the LLDD to terminate the exchange, + * thus releasing the FC exchange. We use blk_mq_tagset_busy_itr() + * to tell us what io's are busy and invoke a transport routine + * to kill them with the LLDD. After terminating the exchange + * the LLDD will call the transport's normal io done path, but it + * will have an aborted status. The done path will return the + * io requests back to the block layer as part of normal completions + * (but with error status). + */ + if (ctrl->ctrl.queue_count > 1) { + nvme_stop_queues(&ctrl->ctrl); + blk_mq_tagset_busy_iter(&ctrl->tag_set, + nvme_fc_terminate_exchange, &ctrl->ctrl); + blk_mq_tagset_wait_completed_request(&ctrl->tag_set); + if (start_queues) + nvme_start_queues(&ctrl->ctrl); + } + + /* + * Other transports, which don't have link-level contexts bound + * to sqe's, would try to gracefully shutdown the controller by + * writing the registers for shutdown and polling (call + * nvme_shutdown_ctrl()). Given a bunch of i/o was potentially + * just aborted and we will wait on those contexts, and given + * there was no indication of how live the controlelr is on the + * link, don't send more io to create more contexts for the + * shutdown. Let the controller fail via keepalive failure if + * its still present. + */ + + /* + * clean up the admin queue. Same thing as above. + */ + blk_mq_quiesce_queue(ctrl->ctrl.admin_q); + blk_mq_tagset_busy_iter(&ctrl->admin_tag_set, + nvme_fc_terminate_exchange, &ctrl->ctrl); + blk_mq_tagset_wait_completed_request(&ctrl->admin_tag_set); +} static void nvme_fc_error_recovery(struct nvme_fc_ctrl *ctrl, char *errmsg) { /* - * if an error (io timeout, etc) while (re)connecting, - * it's an error on creating the new association. - * Start the error recovery thread if it hasn't already - * been started. It is expected there could be multiple - * ios hitting this path before things are cleaned up. + * if an error (io timeout, etc) while (re)connecting, the remote + * port requested terminating of the association (disconnect_ls) + * or an error (timeout or abort) occurred on an io while creating + * the controller. Abort any ios on the association and let the + * create_association error path resolve things. */ if (ctrl->ctrl.state == NVME_CTRL_CONNECTING) { - __nvme_fc_terminate_io(ctrl); - - /* - * Rescheduling the connection after recovering - * from the io error is left to the reconnect work - * item, which is what should have stalled waiting on - * the io that had the error that scheduled this work. - */ + __nvme_fc_abort_outstanding_ios(ctrl, true); + set_bit(ASSOC_FAILED, &ctrl->flags); return; } @@ -2747,30 +2817,6 @@ nvme_fc_complete_rq(struct request *rq) nvme_fc_ctrl_put(ctrl); } -/* - * This routine is used by the transport when it needs to find active - * io on a queue that is to be terminated. The transport uses - * blk_mq_tagset_busy_itr() to find the busy requests, which then invoke - * this routine to kill them on a 1 by 1 basis. - * - * As FC allocates FC exchange for each io, the transport must contact - * the LLDD to terminate the exchange, thus releasing the FC exchange. - * After terminating the exchange the LLDD will call the transport's - * normal io done path for the request, but it will have an aborted - * status. The done path will return the io request back to the block - * layer with an error status. - */ -static bool -nvme_fc_terminate_exchange(struct request *req, void *data, bool reserved) -{ - struct nvme_ctrl *nctrl = data; - struct nvme_fc_ctrl *ctrl = to_fc_ctrl(nctrl); - struct nvme_fc_fcp_op *op = blk_mq_rq_to_pdu(req); - - __nvme_fc_abort_op(ctrl, op); - return true; -} - static const struct blk_mq_ops nvme_fc_mq_ops = { .queue_rq = nvme_fc_queue_rq, @@ -3111,60 +3157,6 @@ out_free_queue: } -/* - * This routine runs through all outstanding commands on the association - * and aborts them. This routine is typically be called by the - * delete_association routine. It is also called due to an error during - * reconnect. In that scenario, it is most likely a command that initializes - * the controller, including fabric Connect commands on io queues, that - * may have timed out or failed thus the io must be killed for the connect - * thread to see the error. - */ -static void -__nvme_fc_abort_outstanding_ios(struct nvme_fc_ctrl *ctrl, bool start_queues) -{ - /* - * If io queues are present, stop them and terminate all outstanding - * ios on them. As FC allocates FC exchange for each io, the - * transport must contact the LLDD to terminate the exchange, - * thus releasing the FC exchange. We use blk_mq_tagset_busy_itr() - * to tell us what io's are busy and invoke a transport routine - * to kill them with the LLDD. After terminating the exchange - * the LLDD will call the transport's normal io done path, but it - * will have an aborted status. The done path will return the - * io requests back to the block layer as part of normal completions - * (but with error status). - */ - if (ctrl->ctrl.queue_count > 1) { - nvme_stop_queues(&ctrl->ctrl); - blk_mq_tagset_busy_iter(&ctrl->tag_set, - nvme_fc_terminate_exchange, &ctrl->ctrl); - blk_mq_tagset_wait_completed_request(&ctrl->tag_set); - if (start_queues) - nvme_start_queues(&ctrl->ctrl); - } - - /* - * Other transports, which don't have link-level contexts bound - * to sqe's, would try to gracefully shutdown the controller by - * writing the registers for shutdown and polling (call - * nvme_shutdown_ctrl()). Given a bunch of i/o was potentially - * just aborted and we will wait on those contexts, and given - * there was no indication of how live the controlelr is on the - * link, don't send more io to create more contexts for the - * shutdown. Let the controller fail via keepalive failure if - * its still present. - */ - - /* - * clean up the admin queue. Same thing as above. - */ - blk_mq_quiesce_queue(ctrl->ctrl.admin_q); - blk_mq_tagset_busy_iter(&ctrl->admin_tag_set, - nvme_fc_terminate_exchange, &ctrl->ctrl); - blk_mq_tagset_wait_completed_request(&ctrl->admin_tag_set); -} - /* * This routine stops operation of the controller on the host side. * On the host os stack side: Admin and IO queues are stopped, @@ -3297,17 +3289,6 @@ nvme_fc_reconnect_or_delete(struct nvme_fc_ctrl *ctrl, int status) static void __nvme_fc_terminate_io(struct nvme_fc_ctrl *ctrl) { - /* - * if state is CONNECTING - the error occurred as part of a - * reconnect attempt. Abort any ios on the association and - * let the create_association error paths resolve things. - */ - if (ctrl->ctrl.state == NVME_CTRL_CONNECTING) { - __nvme_fc_abort_outstanding_ios(ctrl, true); - set_bit(ASSOC_FAILED, &ctrl->flags); - return; - } - /* * For any other state, kill the association. As this routine * is a common io abort routine for resetting and such, after -- cgit v1.2.3 From ac9b820e713bdc8f340e7d65ca6a19b10b790f2a Mon Sep 17 00:00:00 2001 From: James Smart Date: Fri, 23 Oct 2020 15:27:52 -0700 Subject: nvme-fc: remove nvme_fc_terminate_io() __nvme_fc_terminate_io() is now called by only 1 place, in reset_work. Consoldate and move the functionality of terminate_io into reset_work. In reset_work, rather than calling the create_association directly, schedule the connect work element to do its thing. After scheduling, flush the connect work element to continue with semantic of not returning until connect has been attempted at least once. Signed-off-by: James Smart Signed-off-by: Christoph Hellwig --- drivers/nvme/host/fc.c | 49 ++++++++++++++++--------------------------------- 1 file changed, 16 insertions(+), 33 deletions(-) diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index ffbfb0533cac..f4c246462658 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -3287,49 +3287,32 @@ nvme_fc_reconnect_or_delete(struct nvme_fc_ctrl *ctrl, int status) } static void -__nvme_fc_terminate_io(struct nvme_fc_ctrl *ctrl) +nvme_fc_reset_ctrl_work(struct work_struct *work) { - /* - * For any other state, kill the association. As this routine - * is a common io abort routine for resetting and such, after - * the association is terminated, ensure that the state is set - * to CONNECTING. - */ + struct nvme_fc_ctrl *ctrl = + container_of(work, struct nvme_fc_ctrl, ctrl.reset_work); - nvme_stop_keep_alive(&ctrl->ctrl); + nvme_stop_ctrl(&ctrl->ctrl); /* will block will waiting for io to terminate */ nvme_fc_delete_association(ctrl); - if (ctrl->ctrl.state != NVME_CTRL_CONNECTING && - !nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) + if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) dev_err(ctrl->ctrl.device, "NVME-FC{%d}: error_recovery: Couldn't change state " "to CONNECTING\n", ctrl->cnum); -} - -static void -nvme_fc_reset_ctrl_work(struct work_struct *work) -{ - struct nvme_fc_ctrl *ctrl = - container_of(work, struct nvme_fc_ctrl, ctrl.reset_work); - int ret; - - __nvme_fc_terminate_io(ctrl); - nvme_stop_ctrl(&ctrl->ctrl); - - if (ctrl->rport->remoteport.port_state == FC_OBJSTATE_ONLINE) - ret = nvme_fc_create_association(ctrl); - else - ret = -ENOTCONN; - - if (ret) - nvme_fc_reconnect_or_delete(ctrl, ret); - else - dev_info(ctrl->ctrl.device, - "NVME-FC{%d}: controller reset complete\n", - ctrl->cnum); + if (ctrl->rport->remoteport.port_state == FC_OBJSTATE_ONLINE) { + if (!queue_delayed_work(nvme_wq, &ctrl->connect_work, 0)) { + dev_err(ctrl->ctrl.device, + "NVME-FC{%d}: failed to schedule connect " + "after reset\n", ctrl->cnum); + } else { + flush_delayed_work(&ctrl->connect_work); + } + } else { + nvme_fc_reconnect_or_delete(ctrl, -ENOTCONN); + } } -- cgit v1.2.3 From 3c3751f2daf6675f6b5bee83b792354c272f5bd2 Mon Sep 17 00:00:00 2001 From: Chaitanya Kulkarni Date: Thu, 22 Oct 2020 16:58:21 -0700 Subject: nvmet: fix a NULL pointer dereference when tracing the flush command When target side trace in turned on and flush command is issued from the host it results in the following Oops. [ 856.789724] BUG: kernel NULL pointer dereference, address: 0000000000000068 [ 856.790686] #PF: supervisor read access in kernel mode [ 856.791262] #PF: error_code(0x0000) - not-present page [ 856.791863] PGD 6d7110067 P4D 6d7110067 PUD 66f0ad067 PMD 0 [ 856.792527] Oops: 0000 [#1] SMP NOPTI [ 856.792950] CPU: 15 PID: 7034 Comm: nvme Tainted: G OE 5.9.0nvme-5.9+ #71 [ 856.793790] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.0-59-gc9ba5276e3214 [ 856.794956] RIP: 0010:trace_event_raw_event_nvmet_req_init+0x13e/0x170 [nvmet] [ 856.795734] Code: 41 5c 41 5d c3 31 d2 31 f6 e8 4e 9b b8 e0 e9 0e ff ff ff 49 8b 55 00 48 8b 38 8b 0 [ 856.797740] RSP: 0018:ffffc90001be3a60 EFLAGS: 00010246 [ 856.798375] RAX: 0000000000000000 RBX: ffff8887e7d2c01c RCX: 0000000000000000 [ 856.799234] RDX: 0000000000000020 RSI: 0000000057e70ea2 RDI: ffff8887e7d2c034 [ 856.800088] RBP: ffff88869f710578 R08: ffff888807500d40 R09: 00000000fffffffe [ 856.800951] R10: 0000000064c66670 R11: 00000000ef955201 R12: ffff8887e7d2c034 [ 856.801807] R13: ffff88869f7105c8 R14: 0000000000000040 R15: ffff88869f710440 [ 856.802667] FS: 00007f6a22bd8780(0000) GS:ffff888813a00000(0000) knlGS:0000000000000000 [ 856.803635] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 856.804367] CR2: 0000000000000068 CR3: 00000006d73e0000 CR4: 00000000003506e0 [ 856.805283] Call Trace: [ 856.805613] nvmet_req_init+0x27c/0x480 [nvmet] [ 856.806200] nvme_loop_queue_rq+0xcb/0x1d0 [nvme_loop] [ 856.806862] blk_mq_dispatch_rq_list+0x123/0x7b0 [ 856.807459] ? kvm_sched_clock_read+0x14/0x30 [ 856.808025] __blk_mq_sched_dispatch_requests+0xc7/0x170 [ 856.808708] blk_mq_sched_dispatch_requests+0x30/0x60 [ 856.809372] __blk_mq_run_hw_queue+0x70/0x100 [ 856.809935] __blk_mq_delay_run_hw_queue+0x156/0x170 [ 856.810574] blk_mq_run_hw_queue+0x86/0xe0 [ 856.811104] blk_mq_sched_insert_request+0xef/0x160 [ 856.811733] blk_execute_rq+0x69/0xc0 [ 856.812212] ? blk_mq_rq_ctx_init+0xd0/0x230 [ 856.812784] nvme_execute_passthru_rq+0x57/0x130 [nvme_core] [ 856.813461] nvme_submit_user_cmd+0xeb/0x300 [nvme_core] [ 856.814099] nvme_user_cmd.isra.82+0x11e/0x1a0 [nvme_core] [ 856.814752] blkdev_ioctl+0x1dc/0x2c0 [ 856.815197] block_ioctl+0x3f/0x50 [ 856.815606] __x64_sys_ioctl+0x84/0xc0 [ 856.816074] do_syscall_64+0x33/0x40 [ 856.816533] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 856.817168] RIP: 0033:0x7f6a222ed107 [ 856.817617] Code: 44 00 00 48 8b 05 81 cd 2c 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 8 [ 856.819901] RSP: 002b:00007ffca848f058 EFLAGS: 00000202 ORIG_RAX: 0000000000000010 [ 856.820846] RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007f6a222ed107 [ 856.821726] RDX: 00007ffca848f060 RSI: 00000000c0484e43 RDI: 0000000000000003 [ 856.822603] RBP: 0000000000000003 R08: 000000000000003f R09: 0000000000000005 [ 856.823478] R10: 00007ffca848ece0 R11: 0000000000000202 R12: 00007ffca84912d3 [ 856.824359] R13: 00007ffca848f4d0 R14: 0000000000000002 R15: 000000000067e900 [ 856.825236] Modules linked in: nvme_loop(OE) nvmet(OE) nvme_fabrics(OE) null_blk nvme(OE) nvme_corel Move the nvmet_req_init() tracepoint after we parse the command in nvmet_req_init() so that we can get rid of the duplicate nvmet_find_namespace() call. Rename __assign_disk_name() -> __assign_req_name(). Now that we call tracepoint after parsing the command simplify the newly added __assign_req_name() which fixes this bug. Signed-off-by: Chaitanya Kulkarni Signed-off-by: Christoph Hellwig --- drivers/nvme/target/core.c | 4 ++-- drivers/nvme/target/trace.h | 21 +++++++-------------- 2 files changed, 9 insertions(+), 16 deletions(-) diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c index aafcbc424b7a..957b39a82431 100644 --- a/drivers/nvme/target/core.c +++ b/drivers/nvme/target/core.c @@ -907,8 +907,6 @@ bool nvmet_req_init(struct nvmet_req *req, struct nvmet_cq *cq, req->error_loc = NVMET_NO_ERROR_LOC; req->error_slba = 0; - trace_nvmet_req_init(req, req->cmd); - /* no support for fused commands yet */ if (unlikely(flags & (NVME_CMD_FUSE_FIRST | NVME_CMD_FUSE_SECOND))) { req->error_loc = offsetof(struct nvme_common_command, flags); @@ -938,6 +936,8 @@ bool nvmet_req_init(struct nvmet_req *req, struct nvmet_cq *cq, if (status) goto fail; + trace_nvmet_req_init(req, req->cmd); + if (unlikely(!percpu_ref_tryget_live(&sq->ref))) { status = NVME_SC_INVALID_FIELD | NVME_SC_DNR; goto fail; diff --git a/drivers/nvme/target/trace.h b/drivers/nvme/target/trace.h index 0458046d6501..c14e3249a14d 100644 --- a/drivers/nvme/target/trace.h +++ b/drivers/nvme/target/trace.h @@ -46,19 +46,12 @@ static inline struct nvmet_ctrl *nvmet_req_to_ctrl(struct nvmet_req *req) return req->sq->ctrl; } -static inline void __assign_disk_name(char *name, struct nvmet_req *req, - bool init) +static inline void __assign_req_name(char *name, struct nvmet_req *req) { - struct nvmet_ctrl *ctrl = nvmet_req_to_ctrl(req); - struct nvmet_ns *ns; - - if ((init && req->sq->qid) || (!init && req->cq->qid)) { - ns = nvmet_find_namespace(ctrl, req->cmd->rw.nsid); - strncpy(name, ns->device_path, DISK_NAME_LEN); - return; - } - - memset(name, 0, DISK_NAME_LEN); + if (req->ns) + strncpy(name, req->ns->device_path, DISK_NAME_LEN); + else + memset(name, 0, DISK_NAME_LEN); } #endif @@ -81,7 +74,7 @@ TRACE_EVENT(nvmet_req_init, TP_fast_assign( __entry->cmd = cmd; __entry->ctrl = nvmet_req_to_ctrl(req); - __assign_disk_name(__entry->disk, req, true); + __assign_req_name(__entry->disk, req); __entry->qid = req->sq->qid; __entry->cid = cmd->common.command_id; __entry->opcode = cmd->common.opcode; @@ -121,7 +114,7 @@ TRACE_EVENT(nvmet_req_complete, __entry->cid = req->cqe->command_id; __entry->result = le64_to_cpu(req->cqe->result.u64); __entry->status = le16_to_cpu(req->cqe->status) >> 1; - __assign_disk_name(__entry->disk, req, false); + __assign_req_name(__entry->disk, req); ), TP_printk("nvmet%s: %sqid=%d, cmdid=%u, res=%#llx, status=%#x", __print_ctrl_name(__entry->ctrl), -- cgit v1.2.3 From 444d03badc72e246bbcbca4b9c922d70ef046934 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Tue, 27 Oct 2020 10:51:14 +0100 Subject: drm: kernel-doc: document drm_dp_set_subconnector_property() params Changeset e5b92773287c ("drm: report dp downstream port type as a subconnector property") added a new function to the kAPI, but didn't add any documentation for the parameters for drm_dp_set_subconnector_property(). Fixes: e5b92773287c ("drm: report dp downstream port type as a subconnector property") Signed-off-by: Mauro Carvalho Chehab Signed-off-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/0870be85a77bea4ba5cf1715010834289a4e10b1.1603791716.git.mchehab+huawei@kernel.org --- drivers/gpu/drm/drm_dp_helper.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c index 90807a6b415c..b1c71af88579 100644 --- a/drivers/gpu/drm/drm_dp_helper.c +++ b/drivers/gpu/drm/drm_dp_helper.c @@ -1028,7 +1028,8 @@ EXPORT_SYMBOL(drm_dp_downstream_debug); /** * drm_dp_subconnector_type() - get DP branch device type - * + * @dpcd: DisplayPort configuration data + * @port_cap: port capabilities */ enum drm_mode_subconnector drm_dp_subconnector_type(const u8 dpcd[DP_RECEIVER_CAP_SIZE], @@ -1079,6 +1080,10 @@ EXPORT_SYMBOL(drm_dp_subconnector_type); /** * drm_mode_set_dp_subconnector_property - set subconnector for DP connector + * @connector: connector to set property on + * @status: connector status + * @dpcd: DisplayPort configuration data + * @port_cap: port capabilities * * Called by a driver on every detect event. */ -- cgit v1.2.3 From 21a53bbd469b6df21c13f3a5aee1b935f032c03c Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Tue, 27 Oct 2020 10:51:15 +0100 Subject: drm/dp: fix kernel-doc warnings at drm_dp_helper.c As warned by kernel-doc: ./drivers/gpu/drm/drm_dp_helper.c:385: warning: Function parameter or member 'type' not described in 'drm_dp_downstream_is_type' ./drivers/gpu/drm/drm_dp_helper.c:886: warning: Function parameter or member 'dev' not described in 'drm_dp_downstream_mode' Some function parameters weren't documented. Fixes: 38784f6f8805 ("drm/dp: Add helpers to identify downstream facing port types") Reviewed-by: Lyude Paul Signed-off-by: Mauro Carvalho Chehab Signed-off-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/03c9c8ba3f492aca76e2b4836803219cd9c971cf.1603791716.git.mchehab+huawei@kernel.org --- drivers/gpu/drm/drm_dp_helper.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c index b1c71af88579..deeed73f4ed6 100644 --- a/drivers/gpu/drm/drm_dp_helper.c +++ b/drivers/gpu/drm/drm_dp_helper.c @@ -374,6 +374,10 @@ static bool is_edid_digital_input_dp(const struct edid *edid) * drm_dp_downstream_is_type() - is the downstream facing port of certain type? * @dpcd: DisplayPort configuration data * @port_cap: port capabilities + * @type: port type to be checked. Can be: + * %DP_DS_PORT_TYPE_DP, %DP_DS_PORT_TYPE_VGA, %DP_DS_PORT_TYPE_DVI, + * %DP_DS_PORT_TYPE_HDMI, %DP_DS_PORT_TYPE_NON_EDID, + * %DP_DS_PORT_TYPE_DP_DUALMODE or %DP_DS_PORT_TYPE_WIRELESS. * * Caveat: Only works with DPCD 1.1+ port caps. * @@ -870,6 +874,7 @@ EXPORT_SYMBOL(drm_dp_downstream_444_to_420_conversion); /** * drm_dp_downstream_mode() - return a mode for downstream facing port + * @dev: DRM device * @dpcd: DisplayPort configuration data * @port_cap: port capabilities * -- cgit v1.2.3 From 8d7d8c0afbeae253a2bc0a304ebbd21fa980467f Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Tue, 27 Oct 2020 10:51:16 +0100 Subject: drm/dp: fix a kernel-doc issue at drm_edid.c The name of the argument is different, causing those warnings: ./drivers/gpu/drm/drm_edid.c:3754: warning: Function parameter or member 'video_code' not described in 'drm_display_mode_from_cea_vic' ./drivers/gpu/drm/drm_edid.c:3754: warning: Excess function parameter 'vic' description in 'drm_display_mode_from_cea_vic' Fixes: 7af655bce275 ("drm/dp: Add drm_dp_downstream_mode()") Reviewed-by: Lyude Paul Signed-off-by: Mauro Carvalho Chehab Signed-off-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/7f4d6c3ff6df63ebd006eb90a5108006c23e2168.1603791716.git.mchehab+huawei@kernel.org --- drivers/gpu/drm/drm_edid.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index a82f37d44258..631125b46e04 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -3741,7 +3741,7 @@ drm_add_cmdb_modes(struct drm_connector *connector, u8 svd) /** * drm_display_mode_from_cea_vic() - return a mode for CEA VIC * @dev: DRM device - * @vic: CEA VIC of the mode + * @video_code: CEA VIC of the mode * * Creates a new mode matching the specified CEA VIC. * -- cgit v1.2.3 From 08989335e2b6b549ab20dd41ba2f9ca9782f3cd8 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Tue, 27 Oct 2020 10:51:28 +0100 Subject: drm: drm_edid: remove a duplicated kernel-doc declaration It is not possible to create cross-references for duplicated symbols. While Sphinx always detected it, on Sphinx 3 it generates warnings like this: .../Documentation/gpu/drm-kms-helpers:326: ../drivers/gpu/drm/drm_edid.c:1626: WARNING: Duplicate C declaration, also defined in 'gpu/drm-kms-helpers'. Declaration is 'bool drm_edid_are_equal (const struct edid *edid1, const struct edid *edid2)'. So, get rid of the duplicated kernel-doc markup. Signed-off-by: Mauro Carvalho Chehab Signed-off-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/9310f4074fa9d29cd3ad60684d86d0ace8dab7ae.1603791716.git.mchehab+huawei@kernel.org --- include/drm/drm_edid.h | 7 ------- 1 file changed, 7 deletions(-) diff --git a/include/drm/drm_edid.h b/include/drm/drm_edid.h index b27a0e2169c8..e97daf6ffbb1 100644 --- a/include/drm/drm_edid.h +++ b/include/drm/drm_edid.h @@ -359,13 +359,6 @@ drm_load_edid_firmware(struct drm_connector *connector) } #endif -/** - * drm_edid_are_equal - compare two edid blobs. - * @edid1: pointer to first blob - * @edid2: pointer to second blob - * This helper can be used during probing to determine if - * edid had changed. - */ bool drm_edid_are_equal(const struct edid *edid1, const struct edid *edid2); int -- cgit v1.2.3 From 7811a339da2e3b741dc19de50bad5a7728b6ef22 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Tue, 27 Oct 2020 10:51:29 +0100 Subject: drm: kernel-doc: add description for a new function parameter As reported by "make htmldocs": ./drivers/gpu/drm/drm_prime.c:808: warning: Function parameter or member 'dev' not described in 'drm_prime_pages_to_sg' Add a description for the new parameter. Fixes: 707d561f77b5 ("drm: allow limiting the scatter list size.") Signed-off-by: Mauro Carvalho Chehab Acked-by: Gerd Hoffmann Signed-off-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/9366f48e6e9c3ec2f31a3e68452a2b23a1089fce.1603791716.git.mchehab+huawei@kernel.org --- drivers/gpu/drm/drm_prime.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c index d6808f678db5..9f955f2010c2 100644 --- a/drivers/gpu/drm/drm_prime.c +++ b/drivers/gpu/drm/drm_prime.c @@ -794,6 +794,7 @@ static const struct dma_buf_ops drm_gem_prime_dmabuf_ops = { /** * drm_prime_pages_to_sg - converts a page array into an sg list + * @dev: DRM device * @pages: pointer to the array of page pointers to convert * @nr_pages: length of the page vector * -- cgit v1.2.3 From 38a8b32f467a9389ff413574968baa8777c77355 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Tue, 27 Oct 2020 10:51:31 +0100 Subject: drm: kernel-doc: drm_dp_helper.h: fix a typo Right now, kernel-doc generates a warning: ./include/drm/drm_dp_helper.h:1786: warning: Function parameter or member 'hbr2_reset' not described in 'drm_dp_phy_test_params' This is due to a typo: @hb2_reset -> @hbr2_reset Signed-off-by: Mauro Carvalho Chehab Signed-off-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/2a615cb38e951215bb1bddc2481ad323c9cf3fc9.1603791716.git.mchehab+huawei@kernel.org --- include/drm/drm_dp_helper.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/drm/drm_dp_helper.h b/include/drm/drm_dp_helper.h index da53aebb7230..a53243abd945 100644 --- a/include/drm/drm_dp_helper.h +++ b/include/drm/drm_dp_helper.h @@ -1836,7 +1836,7 @@ static inline void drm_dp_cec_unset_edid(struct drm_dp_aux *aux) * @link_rate: Requested Link rate from DPCD 0x219 * @num_lanes: Number of lanes requested by sing through DPCD 0x220 * @phy_pattern: DP Phy test pattern from DPCD 0x248 - * @hb2_reset: DP HBR2_COMPLIANCE_SCRAMBLER_RESET from DCPD 0x24A and 0x24B + * @hbr2_reset: DP HBR2_COMPLIANCE_SCRAMBLER_RESET from DCPD 0x24A and 0x24B * @custom80: DP Test_80BIT_CUSTOM_PATTERN from DPCDs 0x250 through 0x259 * @enhanced_frame_cap: flag for enhanced frame capability. */ -- cgit v1.2.3 From b52817e9de06a3af4ebefd6d244c9c750903d79c Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Tue, 27 Oct 2020 10:51:35 +0100 Subject: drm: drm_print.h: fix kernel-doc markups A kernel-doc markup should start with the identifier on its first line. Signed-off-by: Mauro Carvalho Chehab Signed-off-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/5b76c5625709aaaa3abee98faa620b9f3d27ff85.1603791716.git.mchehab+huawei@kernel.org --- include/drm/drm_print.h | 20 +++++++++++++++++--- 1 file changed, 17 insertions(+), 3 deletions(-) diff --git a/include/drm/drm_print.h b/include/drm/drm_print.h index 1c9417430d08..f32d179e139d 100644 --- a/include/drm/drm_print.h +++ b/include/drm/drm_print.h @@ -338,7 +338,7 @@ void drm_dev_dbg(const struct device *dev, enum drm_debug_category category, const char *format, ...); /** - * Error output. + * DRM_DEV_ERROR() - Error output. * * @dev: device pointer * @fmt: printf() like format string. @@ -347,10 +347,12 @@ void drm_dev_dbg(const struct device *dev, enum drm_debug_category category, drm_dev_printk(dev, KERN_ERR, "*ERROR* " fmt, ##__VA_ARGS__) /** - * Rate limited error output. Like DRM_ERROR() but won't flood the log. + * DRM_DEV_ERROR_RATELIMITED() - Rate limited error output. * * @dev: device pointer * @fmt: printf() like format string. + * + * Like DRM_ERROR() but won't flood the log. */ #define DRM_DEV_ERROR_RATELIMITED(dev, fmt, ...) \ ({ \ @@ -375,15 +377,27 @@ void drm_dev_dbg(const struct device *dev, enum drm_debug_category category, }) /** - * Debug output. + * DRM_DEV_DEBUG() - Debug output for generic drm code * * @dev: device pointer * @fmt: printf() like format string. */ #define DRM_DEV_DEBUG(dev, fmt, ...) \ drm_dev_dbg(dev, DRM_UT_CORE, fmt, ##__VA_ARGS__) +/** + * DRM_DEV_DEBUG_DRIVER() - Debug output for vendor specific part of the driver + * + * @dev: device pointer + * @fmt: printf() like format string. + */ #define DRM_DEV_DEBUG_DRIVER(dev, fmt, ...) \ drm_dev_dbg(dev, DRM_UT_DRIVER, fmt, ##__VA_ARGS__) +/** + * DRM_DEV_DEBUG_KMS() - Debug output for modesetting code + * + * @dev: device pointer + * @fmt: printf() like format string. + */ #define DRM_DEV_DEBUG_KMS(dev, fmt, ...) \ drm_dev_dbg(dev, DRM_UT_KMS, fmt, ##__VA_ARGS__) -- cgit v1.2.3 From 179f5dc36b0a1aa31538d7d8823deb65c39847b3 Mon Sep 17 00:00:00 2001 From: Alexander Sverdlin Date: Fri, 16 Oct 2020 12:18:57 +0200 Subject: staging: octeon: repair "fixed-link" support The PHYs must be registered once in device probe function, not in device open callback because it's only possible to register them once. Fixes: a25e278020bf ("staging: octeon: support fixed-link phys") Signed-off-by: Alexander Sverdlin Cc: stable Link: https://lore.kernel.org/r/20201016101858.11374-1-alexander.sverdlin@nokia.com Signed-off-by: Greg Kroah-Hartman --- drivers/staging/octeon/ethernet-mdio.c | 6 ------ drivers/staging/octeon/ethernet.c | 9 +++++++++ 2 files changed, 9 insertions(+), 6 deletions(-) diff --git a/drivers/staging/octeon/ethernet-mdio.c b/drivers/staging/octeon/ethernet-mdio.c index cfb673a52b25..0bf545849b11 100644 --- a/drivers/staging/octeon/ethernet-mdio.c +++ b/drivers/staging/octeon/ethernet-mdio.c @@ -147,12 +147,6 @@ int cvm_oct_phy_setup_device(struct net_device *dev) phy_node = of_parse_phandle(priv->of_node, "phy-handle", 0); if (!phy_node && of_phy_is_fixed_link(priv->of_node)) { - int rc; - - rc = of_phy_register_fixed_link(priv->of_node); - if (rc) - return rc; - phy_node = of_node_get(priv->of_node); } if (!phy_node) diff --git a/drivers/staging/octeon/ethernet.c b/drivers/staging/octeon/ethernet.c index 204f0b1e2739..5dea6e96ec90 100644 --- a/drivers/staging/octeon/ethernet.c +++ b/drivers/staging/octeon/ethernet.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include @@ -892,6 +893,14 @@ static int cvm_oct_probe(struct platform_device *pdev) break; } + if (priv->of_node && of_phy_is_fixed_link(priv->of_node)) { + if (of_phy_register_fixed_link(priv->of_node)) { + netdev_err(dev, "Failed to register fixed link for interface %d, port %d\n", + interface, priv->port); + dev->netdev_ops = NULL; + } + } + if (!dev->netdev_ops) { free_netdev(dev); } else if (register_netdev(dev) < 0) { -- cgit v1.2.3 From 49d28ebdf1e30d806410eefc7de0a7a1ca5d747c Mon Sep 17 00:00:00 2001 From: Alexander Sverdlin Date: Fri, 16 Oct 2020 16:56:30 +0200 Subject: staging: octeon: Drop on uncorrectable alignment or FCS error Currently in case of alignment or FCS error if the packet cannot be corrected it's still not dropped. Report the error properly and drop the packet while making the code around a little bit more readable. Fixes: 80ff0fd3ab64 ("Staging: Add octeon-ethernet driver files.") Signed-off-by: Alexander Sverdlin Cc: stable Link: https://lore.kernel.org/r/20201016145630.41852-1-alexander.sverdlin@nokia.com Signed-off-by: Greg Kroah-Hartman --- drivers/staging/octeon/ethernet-rx.c | 34 +++++++++++++++++++--------------- 1 file changed, 19 insertions(+), 15 deletions(-) diff --git a/drivers/staging/octeon/ethernet-rx.c b/drivers/staging/octeon/ethernet-rx.c index 2c16230f993c..9ebd665e5d42 100644 --- a/drivers/staging/octeon/ethernet-rx.c +++ b/drivers/staging/octeon/ethernet-rx.c @@ -69,15 +69,17 @@ static inline int cvm_oct_check_rcv_error(struct cvmx_wqe *work) else port = work->word1.cn38xx.ipprt; - if ((work->word2.snoip.err_code == 10) && (work->word1.len <= 64)) { + if ((work->word2.snoip.err_code == 10) && (work->word1.len <= 64)) /* * Ignore length errors on min size packets. Some * equipment incorrectly pads packets to 64+4FCS * instead of 60+4FCS. Note these packets still get * counted as frame errors. */ - } else if (work->word2.snoip.err_code == 5 || - work->word2.snoip.err_code == 7) { + return 0; + + if (work->word2.snoip.err_code == 5 || + work->word2.snoip.err_code == 7) { /* * We received a packet with either an alignment error * or a FCS error. This may be signalling that we are @@ -108,7 +110,10 @@ static inline int cvm_oct_check_rcv_error(struct cvmx_wqe *work) /* Port received 0xd5 preamble */ work->packet_ptr.s.addr += i + 1; work->word1.len -= i + 5; - } else if ((*ptr & 0xf) == 0xd) { + return 0; + } + + if ((*ptr & 0xf) == 0xd) { /* Port received 0xd preamble */ work->packet_ptr.s.addr += i; work->word1.len -= i + 4; @@ -118,21 +123,20 @@ static inline int cvm_oct_check_rcv_error(struct cvmx_wqe *work) ((*(ptr + 1) & 0xf) << 4); ptr++; } - } else { - printk_ratelimited("Port %d unknown preamble, packet dropped\n", - port); - cvm_oct_free_work(work); - return 1; + return 0; } + + printk_ratelimited("Port %d unknown preamble, packet dropped\n", + port); + cvm_oct_free_work(work); + return 1; } - } else { - printk_ratelimited("Port %d receive error code %d, packet dropped\n", - port, work->word2.snoip.err_code); - cvm_oct_free_work(work); - return 1; } - return 0; + printk_ratelimited("Port %d receive error code %d, packet dropped\n", + port, work->word2.snoip.err_code); + cvm_oct_free_work(work); + return 1; } static void copy_segments_to_skb(struct cvmx_wqe *work, struct sk_buff *skb) -- cgit v1.2.3 From 647a6002cb41d358d9ac5de101a8a6dc74748a59 Mon Sep 17 00:00:00 2001 From: Ian Abbott Date: Wed, 21 Oct 2020 13:21:42 +0100 Subject: staging: comedi: cb_pcidas: Allow 2-channel commands for AO subdevice The "cb_pcidas" driver supports asynchronous commands on the analog output (AO) subdevice for those boards that have an AO FIFO. The code (in `cb_pcidas_ao_check_chanlist()` and `cb_pcidas_ao_cmd()`) to validate and set up the command supports output to a single channel or to two channels simultaneously (the boards have two AO channels). However, the code in `cb_pcidas_auto_attach()` that initializes the subdevices neglects to initialize the AO subdevice's `len_chanlist` member, leaving it set to 0, but the Comedi core will "correct" it to 1 if the driver neglected to set it. This limits commands to use a single channel (either channel 0 or 1), but the limit should be two channels. Set the AO subdevice's `len_chanlist` member to be the same value as the `n_chan` member, which will be 2. Cc: Signed-off-by: Ian Abbott Link: https://lore.kernel.org/r/20201021122142.81628-1-abbotti@mev.co.uk Signed-off-by: Greg Kroah-Hartman --- drivers/staging/comedi/drivers/cb_pcidas.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/staging/comedi/drivers/cb_pcidas.c b/drivers/staging/comedi/drivers/cb_pcidas.c index 48ec2ee953dc..d740c4782775 100644 --- a/drivers/staging/comedi/drivers/cb_pcidas.c +++ b/drivers/staging/comedi/drivers/cb_pcidas.c @@ -1342,6 +1342,7 @@ static int cb_pcidas_auto_attach(struct comedi_device *dev, if (dev->irq && board->has_ao_fifo) { dev->write_subdev = s; s->subdev_flags |= SDF_CMD_WRITE; + s->len_chanlist = s->n_chan; s->do_cmdtest = cb_pcidas_ao_cmdtest; s->do_cmd = cb_pcidas_ao_cmd; s->cancel = cb_pcidas_ao_cancel; -- cgit v1.2.3 From b6ae84d648954fae096d94faea1ddb6518b27841 Mon Sep 17 00:00:00 2001 From: Seung-Woo Kim Date: Mon, 26 Oct 2020 18:55:50 +0900 Subject: staging: mmal-vchiq: Fix memory leak for vchiq_instance The vchiq_instance is allocated with vchiq_initialise() but never freed properly. Fix memory leak for the vchiq_instance. Reported-by: Jaehoon Chung Signed-off-by: Seung-Woo Kim Link: https://lore.kernel.org/r/1603706150-10806-1-git-send-email-sw0312.kim@samsung.com Signed-off-by: Greg Kroah-Hartman --- drivers/staging/vc04_services/vchiq-mmal/mmal-vchiq.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) diff --git a/drivers/staging/vc04_services/vchiq-mmal/mmal-vchiq.c b/drivers/staging/vc04_services/vchiq-mmal/mmal-vchiq.c index 3a4202551cfc..9097bcbd67d8 100644 --- a/drivers/staging/vc04_services/vchiq-mmal/mmal-vchiq.c +++ b/drivers/staging/vc04_services/vchiq-mmal/mmal-vchiq.c @@ -179,6 +179,9 @@ struct vchiq_mmal_instance { /* ordered workqueue to process all bulk operations */ struct workqueue_struct *bulk_wq; + + /* handle for a vchiq instance */ + struct vchiq_instance *vchiq_instance; }; static struct mmal_msg_context * @@ -1840,6 +1843,7 @@ int vchiq_mmal_finalise(struct vchiq_mmal_instance *instance) mutex_unlock(&instance->vchiq_mutex); + vchiq_shutdown(instance->vchiq_instance); flush_workqueue(instance->bulk_wq); destroy_workqueue(instance->bulk_wq); @@ -1856,6 +1860,7 @@ EXPORT_SYMBOL_GPL(vchiq_mmal_finalise); int vchiq_mmal_init(struct vchiq_mmal_instance **out_instance) { int status; + int err = -ENODEV; struct vchiq_mmal_instance *instance; static struct vchiq_instance *vchiq_instance; struct vchiq_service_params_kernel params = { @@ -1890,17 +1895,21 @@ int vchiq_mmal_init(struct vchiq_mmal_instance **out_instance) status = vchiq_connect(vchiq_instance); if (status) { pr_err("Failed to connect VCHI instance (status=%d)\n", status); - return -EIO; + err = -EIO; + goto err_shutdown_vchiq; } instance = kzalloc(sizeof(*instance), GFP_KERNEL); - if (!instance) - return -ENOMEM; + if (!instance) { + err = -ENOMEM; + goto err_shutdown_vchiq; + } mutex_init(&instance->vchiq_mutex); instance->bulk_scratch = vmalloc(PAGE_SIZE); + instance->vchiq_instance = vchiq_instance; mutex_init(&instance->context_map_lock); idr_init_base(&instance->context_map, 1); @@ -1932,7 +1941,9 @@ err_close_services: err_free: vfree(instance->bulk_scratch); kfree(instance); - return -ENODEV; +err_shutdown_vchiq: + vchiq_shutdown(vchiq_instance); + return err; } EXPORT_SYMBOL_GPL(vchiq_mmal_init); -- cgit v1.2.3 From d3a85737521a8570adf3a8171b4b3e42216686ec Mon Sep 17 00:00:00 2001 From: Jérôme Pouiller Date: Mon, 19 Oct 2020 18:06:03 +0200 Subject: staging: wfx: fix use of uninitialized pointer MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit With -Wuninitialized, the compiler complains: drivers/staging/wfx/data_tx.c:34:19: warning: variable 'band' is uninitialized when used here [-Wuninitialized] if (rate->idx >= band->n_bitrates) { ^~~~ Reported-by: kernel test robot Reported-by: Nathan Chancellor Fixes: 868fd970e187 ("staging: wfx: improve robustness of wfx_get_hw_rate()") Signed-off-by: Jérôme Pouiller Reviewed-by: Nathan Chancellor Link: https://lore.kernel.org/r/20201019160604.1609180-1-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman --- drivers/staging/wfx/data_tx.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/staging/wfx/data_tx.c b/drivers/staging/wfx/data_tx.c index 41f6a604a697..36b36ef39d05 100644 --- a/drivers/staging/wfx/data_tx.c +++ b/drivers/staging/wfx/data_tx.c @@ -31,13 +31,13 @@ static int wfx_get_hw_rate(struct wfx_dev *wdev, } return rate->idx + 14; } + // WFx only support 2GHz, else band information should be retrieved + // from ieee80211_tx_info + band = wdev->hw->wiphy->bands[NL80211_BAND_2GHZ]; if (rate->idx >= band->n_bitrates) { WARN(1, "wrong rate->idx value: %d", rate->idx); return -1; } - // WFx only support 2GHz, else band information should be retrieved - // from ieee80211_tx_info - band = wdev->hw->wiphy->bands[NL80211_BAND_2GHZ]; return band->bitrates[rate->idx].hw_value; } -- cgit v1.2.3 From df5fbca677c916d1959f814b41b841b65e450027 Mon Sep 17 00:00:00 2001 From: Jérôme Pouiller Date: Mon, 19 Oct 2020 18:06:04 +0200 Subject: staging: wfx: fix test on return value of gpiod_get_value() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The commit 8522d62e6bca ("staging: wfx: gpiod_get_value() can return an error") has changed the way the driver test the value returned by gpiod_get_value(). The new code was wrong. Fixes: 8522d62e6bca ("staging: wfx: gpiod_get_value() can return an error") Signed-off-by: Jérôme Pouiller Link: https://lore.kernel.org/r/20201019160604.1609180-2-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman --- drivers/staging/wfx/bh.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/staging/wfx/bh.c b/drivers/staging/wfx/bh.c index 2ffa587aefaa..ed53d0b45592 100644 --- a/drivers/staging/wfx/bh.c +++ b/drivers/staging/wfx/bh.c @@ -21,7 +21,7 @@ static void device_wakeup(struct wfx_dev *wdev) if (!wdev->pdata.gpio_wakeup) return; - if (gpiod_get_value_cansleep(wdev->pdata.gpio_wakeup) >= 0) + if (gpiod_get_value_cansleep(wdev->pdata.gpio_wakeup) > 0) return; if (wfx_api_older_than(wdev, 1, 4)) { -- cgit v1.2.3 From 7e97e4cbf30026b49b0145c3bfe06087958382c5 Mon Sep 17 00:00:00 2001 From: Jing Xiangfeng Date: Mon, 12 Oct 2020 21:24:04 +0800 Subject: staging: fieldbus: anybuss: jump to correct label in an error path In current code, controller_probe() misses to call ida_simple_remove() in an error path. Jump to correct label to fix it. Fixes: 17614978ed34 ("staging: fieldbus: anybus-s: support the Arcx anybus controller") Reviewed-by: Sven Van Asbroeck Signed-off-by: Jing Xiangfeng Cc: stable Link: https://lore.kernel.org/r/20201012132404.113031-1-jingxiangfeng@huawei.com Signed-off-by: Greg Kroah-Hartman --- drivers/staging/fieldbus/anybuss/arcx-anybus.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/staging/fieldbus/anybuss/arcx-anybus.c b/drivers/staging/fieldbus/anybuss/arcx-anybus.c index 5b8d0bae9ff3..b5fded15e8a6 100644 --- a/drivers/staging/fieldbus/anybuss/arcx-anybus.c +++ b/drivers/staging/fieldbus/anybuss/arcx-anybus.c @@ -293,7 +293,7 @@ static int controller_probe(struct platform_device *pdev) regulator = devm_regulator_register(dev, &can_power_desc, &config); if (IS_ERR(regulator)) { err = PTR_ERR(regulator); - goto out_reset; + goto out_ida; } /* make controller info visible to userspace */ cd->class_dev = kzalloc(sizeof(*cd->class_dev), GFP_KERNEL); -- cgit v1.2.3 From 761a8c58db6bc884994b28cd6d9707b467d680c1 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Fri, 23 Oct 2020 19:00:49 -0400 Subject: tracing, synthetic events: Replace buggy strcat() with seq_buf operations There was a memory corruption bug happening while running the synthetic event selftests: kmemleak: Cannot insert 0xffff8c196fa2afe5 into the object search tree (overlaps existing) CPU: 5 PID: 6866 Comm: ftracetest Tainted: G W 5.9.0-rc5-test+ #577 Hardware name: Hewlett-Packard HP Compaq Pro 6300 SFF/339A, BIOS K01 v03.03 07/14/2016 Call Trace: dump_stack+0x8d/0xc0 create_object.cold+0x3b/0x60 slab_post_alloc_hook+0x57/0x510 ? tracing_map_init+0x178/0x340 __kmalloc+0x1b1/0x390 tracing_map_init+0x178/0x340 event_hist_trigger_func+0x523/0xa40 trigger_process_regex+0xc5/0x110 event_trigger_write+0x71/0xd0 vfs_write+0xca/0x210 ksys_write+0x70/0xf0 do_syscall_64+0x33/0x40 entry_SYSCALL_64_after_hwframe+0x44/0xa9 RIP: 0033:0x7fef0a63a487 Code: 64 89 02 48 c7 c0 ff ff ff ff eb bb 0f 1f 80 00 00 00 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 48 89 54 24 18 48 89 74 24 RSP: 002b:00007fff76f18398 EFLAGS: 00000246 ORIG_RAX: 0000000000000001 RAX: ffffffffffffffda RBX: 0000000000000039 RCX: 00007fef0a63a487 RDX: 0000000000000039 RSI: 000055eb3b26d690 RDI: 0000000000000001 RBP: 000055eb3b26d690 R08: 000000000000000a R09: 0000000000000038 R10: 000055eb3b2cdb80 R11: 0000000000000246 R12: 0000000000000039 R13: 00007fef0a70b500 R14: 0000000000000039 R15: 00007fef0a70b700 kmemleak: Kernel memory leak detector disabled kmemleak: Object 0xffff8c196fa2afe0 (size 8): kmemleak: comm "ftracetest", pid 6866, jiffies 4295082531 kmemleak: min_count = 1 kmemleak: count = 0 kmemleak: flags = 0x1 kmemleak: checksum = 0 kmemleak: backtrace: __kmalloc+0x1b1/0x390 tracing_map_init+0x1be/0x340 event_hist_trigger_func+0x523/0xa40 trigger_process_regex+0xc5/0x110 event_trigger_write+0x71/0xd0 vfs_write+0xca/0x210 ksys_write+0x70/0xf0 do_syscall_64+0x33/0x40 entry_SYSCALL_64_after_hwframe+0x44/0xa9 The cause came down to a use of strcat() that was adding an string that was shorten, but the strcat() did not take that into account. strcat() is extremely dangerous as it does not care how big the buffer is. Replace it with seq_buf operations that prevent the buffer from being overwritten if what is being written is bigger than the buffer. Fixes: 10819e25799a ("tracing: Handle synthetic event array field type checking correctly") Reviewed-by: Tom Zanussi Tested-by: Tom Zanussi Signed-off-by: Steven Rostedt (VMware) --- kernel/trace/trace_events_synth.c | 36 ++++++++++++++++++++++-------------- 1 file changed, 22 insertions(+), 14 deletions(-) diff --git a/kernel/trace/trace_events_synth.c b/kernel/trace/trace_events_synth.c index 3212e2c653b3..84b7cab55291 100644 --- a/kernel/trace/trace_events_synth.c +++ b/kernel/trace/trace_events_synth.c @@ -585,6 +585,7 @@ static struct synth_field *parse_synth_field(int argc, const char **argv, struct synth_field *field; const char *prefix = NULL, *field_type = argv[0], *field_name, *array; int len, ret = 0; + struct seq_buf s; ssize_t size; if (field_type[0] == ';') @@ -630,13 +631,9 @@ static struct synth_field *parse_synth_field(int argc, const char **argv, field_type++; len = strlen(field_type) + 1; - if (array) { - int l = strlen(array); + if (array) + len += strlen(array); - if (l && array[l - 1] == ';') - l--; - len += l; - } if (prefix) len += strlen(prefix); @@ -645,14 +642,18 @@ static struct synth_field *parse_synth_field(int argc, const char **argv, ret = -ENOMEM; goto free; } + seq_buf_init(&s, field->type, len); if (prefix) - strcat(field->type, prefix); - strcat(field->type, field_type); + seq_buf_puts(&s, prefix); + seq_buf_puts(&s, field_type); if (array) { - strcat(field->type, array); - if (field->type[len - 1] == ';') - field->type[len - 1] = '\0'; + seq_buf_puts(&s, array); + if (s.buffer[s.len - 1] == ';') + s.len--; } + if (WARN_ON_ONCE(!seq_buf_buffer_left(&s))) + goto free; + s.buffer[s.len] = '\0'; size = synth_field_size(field->type); if (size < 0) { @@ -663,14 +664,21 @@ static struct synth_field *parse_synth_field(int argc, const char **argv, if (synth_field_is_string(field->type)) { char *type; - type = kzalloc(sizeof("__data_loc ") + strlen(field->type) + 1, GFP_KERNEL); + len = sizeof("__data_loc ") + strlen(field->type) + 1; + type = kzalloc(len, GFP_KERNEL); if (!type) { ret = -ENOMEM; goto free; } - strcat(type, "__data_loc "); - strcat(type, field->type); + seq_buf_init(&s, type, len); + seq_buf_puts(&s, "__data_loc "); + seq_buf_puts(&s, field->type); + + if (WARN_ON_ONCE(!seq_buf_buffer_left(&s))) + goto free; + s.buffer[s.len] = '\0'; + kfree(field->type); field->type = type; -- cgit v1.2.3 From fce52ad3483281930dc9215a68d69837f219f889 Mon Sep 17 00:00:00 2001 From: Dmitry Baryshkov Date: Tue, 27 Oct 2020 16:34:18 +0300 Subject: interconnect: qcom: use icc_sync state for sm8[12]50 In addition to the rest of Qcom interconnect drivers use icc_sync_state for SM8150/SM8250 interconnect drivers to notify the interconnect framework when all consumers are probed and there is no need to keep the bandwidth set to maximum anymore. Also move the BCM initialization before creating the nodes to set the max bandwidth in hardware for the initialization/probing stage. Signed-off-by: Dmitry Baryshkov Fixes: 7d3b0b0d8184 ("interconnect: qcom: Use icc_sync_state") Link: https://lore.kernel.org/r/20201027133418.976687-1-dmitry.baryshkov@linaro.org Signed-off-by: Georgi Djakov --- drivers/interconnect/qcom/sm8150.c | 7 ++++--- drivers/interconnect/qcom/sm8250.c | 7 ++++--- 2 files changed, 8 insertions(+), 6 deletions(-) diff --git a/drivers/interconnect/qcom/sm8150.c b/drivers/interconnect/qcom/sm8150.c index 9218efed04a0..c76b2c7f9b10 100644 --- a/drivers/interconnect/qcom/sm8150.c +++ b/drivers/interconnect/qcom/sm8150.c @@ -551,6 +551,9 @@ static int qnoc_probe(struct platform_device *pdev) return ret; } + for (i = 0; i < qp->num_bcms; i++) + qcom_icc_bcm_init(qp->bcms[i], &pdev->dev); + for (i = 0; i < num_nodes; i++) { size_t j; @@ -574,9 +577,6 @@ static int qnoc_probe(struct platform_device *pdev) } data->num_nodes = num_nodes; - for (i = 0; i < qp->num_bcms; i++) - qcom_icc_bcm_init(qp->bcms[i], &pdev->dev); - platform_set_drvdata(pdev, qp); return 0; @@ -627,6 +627,7 @@ static struct platform_driver qnoc_driver = { .driver = { .name = "qnoc-sm8150", .of_match_table = qnoc_of_match, + .sync_state = icc_sync_state, }, }; module_platform_driver(qnoc_driver); diff --git a/drivers/interconnect/qcom/sm8250.c b/drivers/interconnect/qcom/sm8250.c index 9b58946f7898..cc558fec74e3 100644 --- a/drivers/interconnect/qcom/sm8250.c +++ b/drivers/interconnect/qcom/sm8250.c @@ -567,6 +567,9 @@ static int qnoc_probe(struct platform_device *pdev) return ret; } + for (i = 0; i < qp->num_bcms; i++) + qcom_icc_bcm_init(qp->bcms[i], &pdev->dev); + for (i = 0; i < num_nodes; i++) { size_t j; @@ -590,9 +593,6 @@ static int qnoc_probe(struct platform_device *pdev) } data->num_nodes = num_nodes; - for (i = 0; i < qp->num_bcms; i++) - qcom_icc_bcm_init(qp->bcms[i], &pdev->dev); - platform_set_drvdata(pdev, qp); return 0; @@ -643,6 +643,7 @@ static struct platform_driver qnoc_driver = { .driver = { .name = "qnoc-sm8250", .of_match_table = qnoc_of_match, + .sync_state = icc_sync_state, }, }; module_platform_driver(qnoc_driver); -- cgit v1.2.3 From 0425e7badbdcb6e47109bfdfe494d86217e29009 Mon Sep 17 00:00:00 2001 From: Johannes Thumshirn Date: Thu, 22 Oct 2020 23:05:05 +0900 Subject: btrfs: don't fallback to buffered read if we don't need to Since we switched to the iomap infrastructure in b5ff9f1a96e8f ("btrfs: switch to iomap for direct IO") we're calling generic_file_buffered_read() directly and not via generic_file_read_iter() anymore. If the read could read everything there is no need to bother calling generic_file_buffered_read(), like it is handled in generic_file_read_iter(). If we call generic_file_buffered_read() in this case we can hit a situation where we do an invalid readahead and cause this UBSAN splat in fstest generic/091: run fstests generic/091 at 2020-10-21 10:52:32 ================================================================================ UBSAN: shift-out-of-bounds in ./include/linux/log2.h:57:13 shift exponent 64 is too large for 64-bit type 'long unsigned int' CPU: 0 PID: 656 Comm: fsx Not tainted 5.9.0-rc7+ #821 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.13.0-0-gf21b5a4-rebuilt.opensuse.org 04/01/2014 Call Trace: __dump_stack lib/dump_stack.c:77 dump_stack+0x57/0x70 lib/dump_stack.c:118 ubsan_epilogue+0x5/0x40 lib/ubsan.c:148 __ubsan_handle_shift_out_of_bounds.cold+0x61/0xe9 lib/ubsan.c:395 __roundup_pow_of_two ./include/linux/log2.h:57 get_init_ra_size mm/readahead.c:318 ondemand_readahead.cold+0x16/0x2c mm/readahead.c:530 generic_file_buffered_read+0x3ac/0x840 mm/filemap.c:2199 call_read_iter ./include/linux/fs.h:1876 new_sync_read+0x102/0x180 fs/read_write.c:415 vfs_read+0x11c/0x1a0 fs/read_write.c:481 ksys_read+0x4f/0xc0 fs/read_write.c:615 do_syscall_64+0x33/0x40 arch/x86/entry/common.c:46 entry_SYSCALL_64_after_hwframe+0x44/0xa9 arch/x86/entry/entry_64.S:118 RIP: 0033:0x7fe87fee992e RSP: 002b:00007ffe01605278 EFLAGS: 00000246 ORIG_RAX: 0000000000000000 RAX: ffffffffffffffda RBX: 000000000004f000 RCX: 00007fe87fee992e RDX: 0000000000004000 RSI: 0000000001677000 RDI: 0000000000000003 RBP: 000000000004f000 R08: 0000000000004000 R09: 000000000004f000 R10: 0000000000053000 R11: 0000000000000246 R12: 0000000000004000 R13: 0000000000000000 R14: 000000000007a120 R15: 0000000000000000 ================================================================================ BTRFS info (device nullb0): has skinny extents BTRFS info (device nullb0): ZONED mode enabled, zone size 268435456 B BTRFS info (device nullb0): enabling ssd optimizations Fixes: f85781fb505e ("btrfs: switch to iomap for direct IO") Reviewed-by: Goldwyn Rodrigues Signed-off-by: Johannes Thumshirn Signed-off-by: David Sterba --- fs/btrfs/file.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 0ff659455b1e..87355a38a654 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -3628,7 +3628,8 @@ static ssize_t btrfs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) inode_lock_shared(inode); ret = btrfs_direct_IO(iocb, to); inode_unlock_shared(inode); - if (ret < 0) + if (ret < 0 || !iov_iter_count(to) || + iocb->ki_pos >= i_size_read(file_inode(iocb->ki_filp))) return ret; } -- cgit v1.2.3 From d5c8238849e7bae6063dfc16c08ed62cee7ee688 Mon Sep 17 00:00:00 2001 From: Davidlohr Bueso Date: Wed, 21 Oct 2020 13:17:24 -0700 Subject: btrfs: convert data_seqcount to seqcount_mutex_t By doing so we can associate the sequence counter to the chunk_mutex for lockdep purposes (compiled-out otherwise), the mutex is otherwise used on the write side. Also avoid explicitly disabling preemption around the write region as it will now be done automatically by the seqcount machinery based on the lock type. Signed-off-by: Davidlohr Bueso Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/volumes.c | 2 +- fs/btrfs/volumes.h | 11 +++++------ 2 files changed, 6 insertions(+), 7 deletions(-) diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 1991bc5a6f59..b1e48078c318 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -431,7 +431,7 @@ static struct btrfs_device *__alloc_device(struct btrfs_fs_info *fs_info) atomic_set(&dev->reada_in_flight, 0); atomic_set(&dev->dev_stats_ccnt, 0); - btrfs_device_data_ordered_init(dev); + btrfs_device_data_ordered_init(dev, fs_info); INIT_RADIX_TREE(&dev->reada_zones, GFP_NOFS & ~__GFP_DIRECT_RECLAIM); INIT_RADIX_TREE(&dev->reada_extents, GFP_NOFS & ~__GFP_DIRECT_RECLAIM); extent_io_tree_init(fs_info, &dev->alloc_state, diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index f2177263748e..232f02bd214f 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -39,10 +39,10 @@ struct btrfs_io_geometry { #if BITS_PER_LONG==32 && defined(CONFIG_SMP) #include #define __BTRFS_NEED_DEVICE_DATA_ORDERED -#define btrfs_device_data_ordered_init(device) \ - seqcount_init(&device->data_seqcount) +#define btrfs_device_data_ordered_init(device, info) \ + seqcount_mutex_init(&device->data_seqcount, &info->chunk_mutex) #else -#define btrfs_device_data_ordered_init(device) do { } while (0) +#define btrfs_device_data_ordered_init(device, info) do { } while (0) #endif #define BTRFS_DEV_STATE_WRITEABLE (0) @@ -72,7 +72,8 @@ struct btrfs_device { blk_status_t last_flush_error; #ifdef __BTRFS_NEED_DEVICE_DATA_ORDERED - seqcount_t data_seqcount; + /* A seqcount_t with associated chunk_mutex (for lockdep) */ + seqcount_mutex_t data_seqcount; #endif /* the internal btrfs device id */ @@ -163,11 +164,9 @@ btrfs_device_get_##name(const struct btrfs_device *dev) \ static inline void \ btrfs_device_set_##name(struct btrfs_device *dev, u64 size) \ { \ - preempt_disable(); \ write_seqcount_begin(&dev->data_seqcount); \ dev->name = size; \ write_seqcount_end(&dev->data_seqcount); \ - preempt_enable(); \ } #elif BITS_PER_LONG==32 && defined(CONFIG_PREEMPTION) #define BTRFS_DEVICE_GETSET_FUNCS(name) \ -- cgit v1.2.3 From 9ce0af3e9573fb84c4c807183d13ea2a68271e4b Mon Sep 17 00:00:00 2001 From: Hoegeun Kwon Date: Tue, 27 Oct 2020 13:14:42 +0900 Subject: drm/vc4: drv: Add error handding for bind There is a problem that if vc4_drm bind fails, a memory leak occurs on the drm_property_create side. Add error handding for drm_mode_config. Signed-off-by: Hoegeun Kwon Signed-off-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20201027041442.30352-2-hoegeun.kwon@samsung.com --- drivers/gpu/drm/vc4/vc4_drv.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c index f1a5fd5dab6f..a17aa1db11b6 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.c +++ b/drivers/gpu/drm/vc4/vc4_drv.c @@ -314,6 +314,7 @@ unbind_all: component_unbind_all(dev, drm); gem_destroy: vc4_gem_destroy(drm); + drm_mode_config_cleanup(drm); vc4_bo_cache_destroy(drm); dev_put: drm_dev_put(drm); -- cgit v1.2.3 From 0bcd0a2be8c9ef39d84d167ff85359a49f7be175 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 27 Oct 2020 09:50:17 +0100 Subject: asm-generic: mark __{get,put}_user_fn as __always_inline Without the explicit __always_inline, some RISC-V configs place the functions out of line, triggering the BUILD_BUG_ON checks in the function. Fixes: 11129e8ed4d9 ("riscv: use memcpy based uaccess for nommu again") Reported-by: kernel test robot Signed-off-by: Christoph Hellwig Signed-off-by: Arnd Bergmann --- include/asm-generic/uaccess.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/include/asm-generic/uaccess.h b/include/asm-generic/uaccess.h index 45f9872fd747..4973328f3c6e 100644 --- a/include/asm-generic/uaccess.h +++ b/include/asm-generic/uaccess.h @@ -12,7 +12,8 @@ #ifdef CONFIG_UACCESS_MEMCPY #include -static inline int __get_user_fn(size_t size, const void __user *from, void *to) +static __always_inline int +__get_user_fn(size_t size, const void __user *from, void *to) { BUILD_BUG_ON(!__builtin_constant_p(size)); @@ -37,7 +38,8 @@ static inline int __get_user_fn(size_t size, const void __user *from, void *to) } #define __get_user_fn(sz, u, k) __get_user_fn(sz, u, k) -static inline int __put_user_fn(size_t size, void __user *to, void *from) +static __always_inline int +__put_user_fn(size_t size, void __user *to, void *from) { BUILD_BUG_ON(!__builtin_constant_p(size)); -- cgit v1.2.3 From 38dc5079da7081e8b09b14e255e9da82c451a531 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 26 Oct 2020 18:03:18 -0600 Subject: Fix compat regression in process_vm_rw() The removal of compat_process_vm_{readv,writev} didn't change process_vm_rw(), which always assumes it's not doing a compat syscall. Instead of passing in 'false' unconditionally for 'compat', make it conditional on in_compat_syscall(). [ Both Al and Christoph point out that trying to access a 64-bit process from a 32-bit one cannot work anyway, and is likely better prohibited, but that's a separate issue - Linus ] Fixes: c3973b401ef2 ("mm: remove compat_process_vm_{readv,writev}") Reported-and-tested-by: Kyle Huey Signed-off-by: Jens Axboe Acked-by: Al Viro Reviewed-by: Christoph Hellwig Signed-off-by: Linus Torvalds --- mm/process_vm_access.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/mm/process_vm_access.c b/mm/process_vm_access.c index fd12da80b6f2..05676722d9cd 100644 --- a/mm/process_vm_access.c +++ b/mm/process_vm_access.c @@ -273,7 +273,8 @@ static ssize_t process_vm_rw(pid_t pid, return rc; if (!iov_iter_count(&iter)) goto free_iov_l; - iov_r = iovec_from_user(rvec, riovcnt, UIO_FASTIOV, iovstack_r, false); + iov_r = iovec_from_user(rvec, riovcnt, UIO_FASTIOV, iovstack_r, + in_compat_syscall()); if (IS_ERR(iov_r)) { rc = PTR_ERR(iov_r); goto free_iov_l; -- cgit v1.2.3 From 6e9643a864aa4d532b0d467bacc18a15adf5ca82 Mon Sep 17 00:00:00 2001 From: Zhang Qilong Date: Fri, 23 Oct 2020 17:58:39 +0800 Subject: cpufreq: e_powersaver: remove unreachable break A 'break' following a 'return' statement is pointless, so remove it. Signed-off-by: Zhang Qilong Acked-by: Viresh Kumar [ rjw: Subject and changelog edits ] Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/e_powersaver.c | 1 - drivers/cpufreq/longhaul.c | 1 - 2 files changed, 2 deletions(-) diff --git a/drivers/cpufreq/e_powersaver.c b/drivers/cpufreq/e_powersaver.c index 776a58bab0ff..ab93bce8ae77 100644 --- a/drivers/cpufreq/e_powersaver.c +++ b/drivers/cpufreq/e_powersaver.c @@ -223,7 +223,6 @@ static int eps_cpu_init(struct cpufreq_policy *policy) case EPS_BRAND_C3: pr_cont("C3\n"); return -ENODEV; - break; } /* Enable Enhanced PowerSaver */ rdmsrl(MSR_IA32_MISC_ENABLE, val); diff --git a/drivers/cpufreq/longhaul.c b/drivers/cpufreq/longhaul.c index 123fb006810d..182a4dbca095 100644 --- a/drivers/cpufreq/longhaul.c +++ b/drivers/cpufreq/longhaul.c @@ -593,7 +593,6 @@ static void longhaul_setup_voltagescaling(void) break; default: return; - break; } if (min_vid_speed >= highest_speed) return; -- cgit v1.2.3 From db865272d9c4687520dc29f77e701a1b2669872f Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Fri, 23 Oct 2020 17:15:56 +0200 Subject: cpufreq: Avoid configuring old governors as default with intel_pstate Commit 33aa46f252c7 ("cpufreq: intel_pstate: Use passive mode by default without HWP") was meant to cause intel_pstate to be used in the passive mode with the schedutil governor on top of it, but it missed the case in which either "ondemand" or "conservative" was selected as the default governor in the existing kernel config, in which case the previous old governor configuration would be used, causing the default legacy governor to be used on top of intel_pstate instead of schedutil. Address this by preventing "ondemand" and "conservative" from being configured as the default cpufreq governor in the case when schedutil is the default choice for the default governor setting. [Note that the default cpufreq governor can still be set via the kernel command line if need be and that choice is not limited, so if anyone really wants to use one of the legacy governors by default, it can be achieved this way.] Fixes: 33aa46f252c7 ("cpufreq: intel_pstate: Use passive mode by default without HWP") Reported-by: Julia Lawall Cc: 5.8+ # 5.8+ Signed-off-by: Rafael J. Wysocki Acked-by: Viresh Kumar --- drivers/cpufreq/Kconfig | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig index 2c7171e0b001..85de313ddec2 100644 --- a/drivers/cpufreq/Kconfig +++ b/drivers/cpufreq/Kconfig @@ -71,6 +71,7 @@ config CPU_FREQ_DEFAULT_GOV_USERSPACE config CPU_FREQ_DEFAULT_GOV_ONDEMAND bool "ondemand" + depends on !(X86_INTEL_PSTATE && SMP) select CPU_FREQ_GOV_ONDEMAND select CPU_FREQ_GOV_PERFORMANCE help @@ -83,6 +84,7 @@ config CPU_FREQ_DEFAULT_GOV_ONDEMAND config CPU_FREQ_DEFAULT_GOV_CONSERVATIVE bool "conservative" + depends on !(X86_INTEL_PSTATE && SMP) select CPU_FREQ_GOV_CONSERVATIVE select CPU_FREQ_GOV_PERFORMANCE help -- cgit v1.2.3 From 1c534352f47fd83eb08075ac2474f707e74bf7f7 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Fri, 23 Oct 2020 17:35:19 +0200 Subject: cpufreq: Introduce CPUFREQ_NEED_UPDATE_LIMITS driver flag Generally, a cpufreq driver may need to update some internal upper and lower frequency boundaries on policy max and min changes, respectively, but currently this does not work if the target frequency does not change along with the policy limit. Namely, if the target frequency does not change along with the policy min or max, the "target_freq == policy->cur" check in __cpufreq_driver_target() prevents driver callbacks from being invoked and they do not even have a chance to update the corresponding internal boundary. This particularly affects the "powersave" and "performance" governors that always set the target frequency to one of the policy limits and it never changes when the other limit is updated. To allow cpufreq the drivers needing to update internal frequency boundaries on policy limits changes to avoid this issue, introduce a new driver flag, CPUFREQ_NEED_UPDATE_LIMITS, that (when set) will neutralize the check mentioned above. Signed-off-by: Rafael J. Wysocki Acked-by: Viresh Kumar --- drivers/cpufreq/cpufreq.c | 3 ++- include/linux/cpufreq.h | 10 +++++++++- 2 files changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index f4b60663efe6..ea58337fb65f 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -2187,7 +2187,8 @@ int __cpufreq_driver_target(struct cpufreq_policy *policy, * exactly same freq is called again and so we can save on few function * calls. */ - if (target_freq == policy->cur) + if (target_freq == policy->cur && + !(cpufreq_driver->flags & CPUFREQ_NEED_UPDATE_LIMITS)) return 0; /* Save last value to restore later on errors */ diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index fa37b1c66443..038ed83aab41 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -298,7 +298,7 @@ __ATTR(_name, 0644, show_##_name, store_##_name) struct cpufreq_driver { char name[CPUFREQ_NAME_LEN]; - u8 flags; + u16 flags; void *driver_data; /* needed by all drivers */ @@ -422,6 +422,14 @@ struct cpufreq_driver { */ #define CPUFREQ_IS_COOLING_DEV BIT(7) +/* + * Set by drivers that need to update internale upper and lower boundaries along + * with the target frequency and so the core and governors should also invoke + * the diver if the target frequency does not change, but the policy min or max + * may have changed. + */ +#define CPUFREQ_NEED_UPDATE_LIMITS BIT(8) + int cpufreq_register_driver(struct cpufreq_driver *driver_data); int cpufreq_unregister_driver(struct cpufreq_driver *driver_data); -- cgit v1.2.3 From e0be38ed4ab413ddd492118cf146369b86ee0ab5 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Fri, 23 Oct 2020 17:35:32 +0200 Subject: cpufreq: intel_pstate: Avoid missing HWP max updates in passive mode If the cpufreq policy max limit is changed when intel_pstate operates in the passive mode with HWP enabled and the "powersave" governor is used on top of it, the HWP max limit is not updated as appropriate. Namely, in the "powersave" governor case, the target P-state is always equal to the policy min limit, so if the latter does not change, intel_cpufreq_adjust_hwp() is not invoked to update the HWP Request MSR due to the "target_pstate != old_pstate" check in intel_cpufreq_update_pstate(), so the HWP max limit is not updated as a result. Also, if the CPUFREQ_NEED_UPDATE_LIMITS flag is not set for the driver and the target frequency does not change along with the policy max limit, the "target_freq == policy->cur" check in __cpufreq_driver_target() prevents the driver's ->target() callback from being invoked at all, so the HWP max limit is not updated. To prevent that occurring, set the CPUFREQ_NEED_UPDATE_LIMITS flag in the intel_cpufreq driver structure if HWP is enabled and modify intel_cpufreq_update_pstate() to do the "target_pstate != old_pstate" check only in the non-HWP case and let intel_cpufreq_adjust_hwp() always run in the HWP case (it will update HWP Request only if the cached value of the register is different from the new one including the limits, so if neither the target P-state value nor the max limit changes, the register write will still be avoided). Fixes: f6ebbcf08f37 ("cpufreq: intel_pstate: Implement passive mode with HWP enabled") Reported-by: Zhang Rui Cc: 5.9+ # 5.9+: 1c534352f47f cpufreq: Introduce CPUFREQ_NEED_UPDATE_LIMITS ... Signed-off-by: Rafael J. Wysocki Acked-by: Viresh Kumar Tested-by: Zhang Rui --- drivers/cpufreq/intel_pstate.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index 3c1455518738..b7a9779250aa 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -2568,14 +2568,12 @@ static int intel_cpufreq_update_pstate(struct cpudata *cpu, int target_pstate, int old_pstate = cpu->pstate.current_pstate; target_pstate = intel_pstate_prepare_request(cpu, target_pstate); - if (target_pstate != old_pstate) { + if (hwp_active) { + intel_cpufreq_adjust_hwp(cpu, target_pstate, fast_switch); + cpu->pstate.current_pstate = target_pstate; + } else if (target_pstate != old_pstate) { + intel_cpufreq_adjust_perf_ctl(cpu, target_pstate, fast_switch); cpu->pstate.current_pstate = target_pstate; - if (hwp_active) - intel_cpufreq_adjust_hwp(cpu, target_pstate, - fast_switch); - else - intel_cpufreq_adjust_perf_ctl(cpu, target_pstate, - fast_switch); } intel_cpufreq_trace(cpu, fast_switch ? INTEL_PSTATE_TRACE_FAST_SWITCH : @@ -3032,6 +3030,7 @@ static int __init intel_pstate_init(void) hwp_mode_bdw = id->driver_data; intel_pstate.attr = hwp_cpufreq_attrs; intel_cpufreq.attr = hwp_cpufreq_attrs; + intel_cpufreq.flags |= CPUFREQ_NEED_UPDATE_LIMITS; if (!default_driver) default_driver = &intel_pstate; -- cgit v1.2.3 From 4e0ba5577dba686f96c1c10ef4166380667fdec7 Mon Sep 17 00:00:00 2001 From: Chen Yu Date: Sun, 25 Oct 2020 00:29:53 +0800 Subject: intel_idle: Fix max_cstate for processor models without C-state tables Currently intel_idle driver gets the c-state information from ACPI _CST if the processor model is not recognized by it. However the c-state in _CST starts with index 1 which is different from the index in intel_idle driver's internal c-state table. While intel_idle_max_cstate_reached() was previously introduced to deal with intel_idle driver's internal c-state table, re-using this function directly on _CST is incorrect. Fix this by subtracting 1 from the index when checking max_cstate in the _CST case. For example, append intel_idle.max_cstate=1 in boot command line, Before the patch: grep . /sys/devices/system/cpu/cpu0/cpuidle/state*/name POLL After the patch: grep . /sys/devices/system/cpu/cpu0/cpuidle/state*/name /sys/devices/system/cpu/cpu0/cpuidle/state0/name:POLL /sys/devices/system/cpu/cpu0/cpuidle/state1/name:C1_ACPI Fixes: 18734958e9bf ("intel_idle: Use ACPI _CST for processor models without C-state tables") Reported-by: Pengfei Xu Cc: 5.6+ # 5.6+ Signed-off-by: Chen Yu [ rjw: Changelog edits ] Signed-off-by: Rafael J. Wysocki --- drivers/idle/intel_idle.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c index 56f5b8077cba..01bace49a962 100644 --- a/drivers/idle/intel_idle.c +++ b/drivers/idle/intel_idle.c @@ -1239,7 +1239,7 @@ static void __init intel_idle_init_cstates_acpi(struct cpuidle_driver *drv) struct acpi_processor_cx *cx; struct cpuidle_state *state; - if (intel_idle_max_cstate_reached(cstate)) + if (intel_idle_max_cstate_reached(cstate - 1)) break; cx = &acpi_state_table.states[cstate]; -- cgit v1.2.3 From 10105d0c9763f058f6a9a09f78397d5bf94dc94c Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 26 Oct 2020 17:30:28 -0400 Subject: drm/amdgpu/swsmu: drop smu i2c bus on navi1x Stop registering the SMU i2c bus on navi1x. This leads to instability issues when userspace processes mess with the bus and also seems to cause display stability issues in some cases. Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1314 Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1341 Reviewed-by: Evan Quan Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c | 25 ------------------------- 1 file changed, 25 deletions(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c index 8d8081c6bd38..9cf97744b67e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c @@ -2534,29 +2534,6 @@ static const struct i2c_algorithm navi10_i2c_algo = { .functionality = navi10_i2c_func, }; -static int navi10_i2c_control_init(struct smu_context *smu, struct i2c_adapter *control) -{ - struct amdgpu_device *adev = to_amdgpu_device(control); - int res; - - control->owner = THIS_MODULE; - control->class = I2C_CLASS_SPD; - control->dev.parent = &adev->pdev->dev; - control->algo = &navi10_i2c_algo; - snprintf(control->name, sizeof(control->name), "AMDGPU SMU"); - - res = i2c_add_adapter(control); - if (res) - DRM_ERROR("Failed to register hw i2c, err: %d\n", res); - - return res; -} - -static void navi10_i2c_control_fini(struct smu_context *smu, struct i2c_adapter *control) -{ - i2c_del_adapter(control); -} - static ssize_t navi10_get_gpu_metrics(struct smu_context *smu, void **table) { @@ -2687,8 +2664,6 @@ static const struct pptable_funcs navi10_ppt_funcs = { .set_default_dpm_table = navi10_set_default_dpm_table, .dpm_set_vcn_enable = navi10_dpm_set_vcn_enable, .dpm_set_jpeg_enable = navi10_dpm_set_jpeg_enable, - .i2c_init = navi10_i2c_control_init, - .i2c_fini = navi10_i2c_control_fini, .print_clk_levels = navi10_print_clk_levels, .force_clk_levels = navi10_force_clk_levels, .populate_umd_state_clk = navi10_populate_umd_state_clk, -- cgit v1.2.3 From e72d98445d089aa2d623a70b27ec9011be2c5166 Mon Sep 17 00:00:00 2001 From: Kenneth Feng Date: Wed, 21 Oct 2020 14:03:08 +0800 Subject: drm/amd/pm: fix the wrong fan speed in fan1_input fix the wrong fan speed in fan1_input when the fan control mode is manual. the fan speed value is not correct when we set manual mode to fan1_enalbe - 1. since the fan speed in the metrics table always reflects the real fan speed,we can fetch the fan speed for both auto and manual mode. Signed-off-by: Kenneth Feng Reviewed-by: Likun Gao Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c index 685a8a3b25d4..895d89bea7fa 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c @@ -1177,14 +1177,9 @@ static int sienna_cichlid_get_fan_speed_rpm(struct smu_context *smu, if (!speed) return -EINVAL; - switch (smu_v11_0_get_fan_control_mode(smu)) { - case AMD_FAN_CTRL_AUTO: - return sienna_cichlid_get_smu_metrics_data(smu, - METRICS_CURR_FANSPEED, - speed); - default: - return smu_v11_0_get_fan_speed_rpm(smu, speed); - } + return sienna_cichlid_get_smu_metrics_data(smu, + METRICS_CURR_FANSPEED, + speed); } static int sienna_cichlid_get_fan_parameters(struct smu_context *smu) -- cgit v1.2.3 From 65d437b83b2b8eb9ae47b6145399fc7683b028f7 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 26 Oct 2020 17:40:42 -0400 Subject: drm/amdgpu/pm: fix the fan speed in fan1_input in manual mode for navi1x It has been confirmed that the SMU metrics table should always reflect the current fan speed even in manual mode. Fixes: 3033e9f1c2de ("drm/amdgpu/swsmu: handle manual fan readback on SMU11") Reviewed-by: Evan Quan Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c index 9cf97744b67e..ef1a62e86a0e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c @@ -1361,14 +1361,9 @@ static int navi10_get_fan_speed_rpm(struct smu_context *smu, if (!speed) return -EINVAL; - switch (smu_v11_0_get_fan_control_mode(smu)) { - case AMD_FAN_CTRL_AUTO: - return navi10_get_smu_metrics_data(smu, - METRICS_CURR_FANSPEED, - speed); - default: - return smu_v11_0_get_fan_speed_rpm(smu, speed); - } + return navi10_get_smu_metrics_data(smu, + METRICS_CURR_FANSPEED, + speed); } static int navi10_get_fan_parameters(struct smu_context *smu) -- cgit v1.2.3 From 4d4ce8053bfac9a72b9094c6879119938efaa05d Mon Sep 17 00:00:00 2001 From: Jackie Zamow Date: Tue, 27 Oct 2020 07:43:19 -0500 Subject: PM: sleep: fix typo in kernel/power/process.c Fix a typo in a comment in freeze_processes(). Signed-off-by: Jackie Zamow [ rjw: Subject and changelog edits ] Signed-off-by: Rafael J. Wysocki --- kernel/power/process.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/power/process.c b/kernel/power/process.c index 4b6a54da7e65..45b054b7b5ec 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -146,7 +146,7 @@ int freeze_processes(void) BUG_ON(in_atomic()); /* - * Now that the whole userspace is frozen we need to disbale + * Now that the whole userspace is frozen we need to disable * the OOM killer to disallow any further interference with * killable tasks. There is no guarantee oom victims will * ever reach a point they go away we have to wait with a timeout. -- cgit v1.2.3 From 3e6631485fae70f474d5bd85cfaf0f113f61ccce Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Sun, 4 Oct 2020 19:57:20 -0700 Subject: vmlinux.lds.h: Keep .ctors.* with .ctors Under some circumstances, the compiler generates .ctors.* sections. This is seen doing a cross compile of x86_64 from a powerpc64el host: x86_64-linux-gnu-ld: warning: orphan section `.ctors.65435' from `kernel/trace/trace_clock.o' being placed in section `.ctors.65435' x86_64-linux-gnu-ld: warning: orphan section `.ctors.65435' from `kernel/trace/ftrace.o' being placed in section `.ctors.65435' x86_64-linux-gnu-ld: warning: orphan section `.ctors.65435' from `kernel/trace/ring_buffer.o' being placed in section `.ctors.65435' Include these orphans along with the regular .ctors section. Reported-by: Stephen Rothwell Tested-by: Stephen Rothwell Fixes: 83109d5d5fba ("x86/build: Warn on orphan section placement") Signed-off-by: Kees Cook Acked-by: Nick Desaulniers Link: https://lore.kernel.org/r/20201005025720.2599682-1-keescook@chromium.org --- include/asm-generic/vmlinux.lds.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index cd14444bf600..b2b3d81b1535 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -701,6 +701,7 @@ #ifdef CONFIG_CONSTRUCTORS #define KERNEL_CTORS() . = ALIGN(8); \ __ctors_start = .; \ + KEEP(*(SORT(.ctors.*))) \ KEEP(*(.ctors)) \ KEEP(*(SORT(.init_array.*))) \ KEEP(*(.init_array)) \ -- cgit v1.2.3 From d5dcce0c414fcbfe4c2037b66ac69ea5f9b3f75c Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Thu, 22 Oct 2020 21:40:59 +0300 Subject: device property: Keep secondary firmware node secondary by type Behind primary and secondary we understand the type of the nodes which might define their ordering. However, if primary node gone, we can't maintain the ordering by definition of the linked list. Thus, by ordering secondary node becomes first in the list. But in this case the meaning of it is still secondary (or auxiliary). The type of the node is maintained by the secondary pointer in it: secondary pointer Meaning NULL or valid primary node ERR_PTR(-ENODEV) secondary node So, if by some reason we do the following sequence of calls set_primary_fwnode(dev, NULL); set_primary_fwnode(dev, primary); we should preserve secondary node. This concept is supported by the description of set_primary_fwnode() along with implementation of set_secondary_fwnode(). Hence, fix the commit c15e1bdda436 to follow this as well. Fixes: c15e1bdda436 ("device property: Fix the secondary firmware node handling in set_primary_fwnode()") Cc: Ferry Toth Signed-off-by: Andy Shevchenko Reviewed-by: Heikki Krogerus Tested-by: Ferry Toth Cc: 5.9+ # 5.9+ Signed-off-by: Rafael J. Wysocki --- drivers/base/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/base/core.c b/drivers/base/core.c index c852f16c111b..41feab679fa1 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -4278,7 +4278,7 @@ void set_primary_fwnode(struct device *dev, struct fwnode_handle *fwnode) } else { if (fwnode_is_primary(fn)) { dev->fwnode = fn->secondary; - fn->secondary = NULL; + fn->secondary = ERR_PTR(-ENODEV); } else { dev->fwnode = NULL; } -- cgit v1.2.3 From 99aed9227073fb34ce2880cbc7063e04185a65e1 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Thu, 22 Oct 2020 21:41:00 +0300 Subject: device property: Don't clear secondary pointer for shared primary firmware node It appears that firmware nodes can be shared between devices. In such case when a (child) device is about to be deleted, its firmware node may be shared and ACPI_COMPANION_SET(..., NULL) call for it breaks the secondary link of the shared primary firmware node. In order to prevent that, check, if the device has a parent and parent's firmware node is shared with its child, and avoid crashing the link. Fixes: c15e1bdda436 ("device property: Fix the secondary firmware node handling in set_primary_fwnode()") Reported-by: Ferry Toth Signed-off-by: Andy Shevchenko Reviewed-by: Heikki Krogerus Tested-by: Ferry Toth Cc: 5.9+ # 5.9+ Signed-off-by: Rafael J. Wysocki --- drivers/base/core.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/base/core.c b/drivers/base/core.c index 41feab679fa1..78114ddac755 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -4264,6 +4264,7 @@ static inline bool fwnode_is_primary(struct fwnode_handle *fwnode) */ void set_primary_fwnode(struct device *dev, struct fwnode_handle *fwnode) { + struct device *parent = dev->parent; struct fwnode_handle *fn = dev->fwnode; if (fwnode) { @@ -4278,7 +4279,8 @@ void set_primary_fwnode(struct device *dev, struct fwnode_handle *fwnode) } else { if (fwnode_is_primary(fn)) { dev->fwnode = fn->secondary; - fn->secondary = ERR_PTR(-ENODEV); + if (!(parent && fn == parent->fwnode)) + fn->secondary = ERR_PTR(-ENODEV); } else { dev->fwnode = NULL; } -- cgit v1.2.3 From e510785f8aca4a7346497edd4d5aceefe5370960 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 23 Oct 2020 18:33:11 +0200 Subject: PNP: fix kernel-doc markups It sounds that there were function renames. Update the kernel-doc markups accordingly. Signed-off-by: Mauro Carvalho Chehab Signed-off-by: Rafael J. Wysocki --- drivers/pnp/core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/pnp/core.c b/drivers/pnp/core.c index 3bf18d718975..a50ab002e9e4 100644 --- a/drivers/pnp/core.c +++ b/drivers/pnp/core.c @@ -51,7 +51,7 @@ static void pnp_remove_protocol(struct pnp_protocol *protocol) } /** - * pnp_protocol_register - adds a pnp protocol to the pnp layer + * pnp_register_protocol - adds a pnp protocol to the pnp layer * @protocol: pointer to the corresponding pnp_protocol structure * * Ex protocols: ISAPNP, PNPBIOS, etc @@ -91,7 +91,7 @@ int pnp_register_protocol(struct pnp_protocol *protocol) } /** - * pnp_protocol_unregister - removes a pnp protocol from the pnp layer + * pnp_unregister_protocol - removes a pnp protocol from the pnp layer * @protocol: pointer to the corresponding pnp_protocol structure */ void pnp_unregister_protocol(struct pnp_protocol *protocol) -- cgit v1.2.3 From be0e97526455fefe3b7668b7fdf6cd247bf13318 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 26 Oct 2020 22:48:34 +0100 Subject: ACPI: dock: fix enum-conversion warning gcc points out a type mismatch: drivers/acpi/dock.c: In function 'hot_remove_dock_devices': drivers/acpi/dock.c:234:53: warning: implicit conversion from 'enum ' to 'enum dock_callback_type' [-Wenum-conversion] 234 | dock_hotplug_event(dd, ACPI_NOTIFY_EJECT_REQUEST, false); This is harmless because 'false' still has the correct numeric value, but passing DOCK_CALL_HANDLER documents better what is going on and avoids the warning. Fixes: 37f908778f20 ("ACPI / dock: Walk list in reverse order during removal of devices") Fixes: f09ce741a03a ("ACPI / dock / PCI: Drop ACPI dock notifier chain") Signed-off-by: Arnd Bergmann Reviewed-by: Hanjun Guo Signed-off-by: Rafael J. Wysocki --- drivers/acpi/dock.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/acpi/dock.c b/drivers/acpi/dock.c index 45d4b7b69de8..24e076f44d23 100644 --- a/drivers/acpi/dock.c +++ b/drivers/acpi/dock.c @@ -231,7 +231,8 @@ static void hot_remove_dock_devices(struct dock_station *ds) * between them). */ list_for_each_entry_reverse(dd, &ds->dependent_devices, list) - dock_hotplug_event(dd, ACPI_NOTIFY_EJECT_REQUEST, false); + dock_hotplug_event(dd, ACPI_NOTIFY_EJECT_REQUEST, + DOCK_CALL_HANDLER); list_for_each_entry_reverse(dd, &ds->dependent_devices, list) acpi_bus_trim(dd->adev); -- cgit v1.2.3 From c39866f268f89868df17724cd2262d121552d8c9 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Mon, 28 Sep 2020 15:48:54 -0700 Subject: arm/build: Always handle .ARM.exidx and .ARM.extab sections After turning on warnings for orphan section placement, enabling CONFIG_UNWINDER_FRAME_POINTER instead of CONFIG_UNWINDER_ARM causes thousands of warnings when clang + ld.lld are used: $ scripts/config --file arch/arm/configs/multi_v7_defconfig \ -d CONFIG_UNWINDER_ARM \ -e CONFIG_UNWINDER_FRAME_POINTER $ make -skj"$(nproc)" ARCH=arm CROSS_COMPILE=arm-linux-gnueabi- LLVM=1 defconfig zImage ld.lld: warning: init/built-in.a(main.o):(.ARM.extab) is being placed in '.ARM.extab' ld.lld: warning: init/built-in.a(main.o):(.ARM.extab.init.text) is being placed in '.ARM.extab.init.text' ld.lld: warning: init/built-in.a(main.o):(.ARM.extab.ref.text) is being placed in '.ARM.extab.ref.text' ld.lld: warning: init/built-in.a(do_mounts.o):(.ARM.extab.init.text) is being placed in '.ARM.extab.init.text' ld.lld: warning: init/built-in.a(do_mounts.o):(.ARM.extab) is being placed in '.ARM.extab' ld.lld: warning: init/built-in.a(do_mounts_rd.o):(.ARM.extab.init.text) is being placed in '.ARM.extab.init.text' ld.lld: warning: init/built-in.a(do_mounts_rd.o):(.ARM.extab) is being placed in '.ARM.extab' ld.lld: warning: init/built-in.a(do_mounts_initrd.o):(.ARM.extab.init.text) is being placed in '.ARM.extab.init.text' ld.lld: warning: init/built-in.a(initramfs.o):(.ARM.extab.init.text) is being placed in '.ARM.extab.init.text' ld.lld: warning: init/built-in.a(initramfs.o):(.ARM.extab) is being placed in '.ARM.extab' ld.lld: warning: init/built-in.a(calibrate.o):(.ARM.extab.init.text) is being placed in '.ARM.extab.init.text' ld.lld: warning: init/built-in.a(calibrate.o):(.ARM.extab) is being placed in '.ARM.extab' These sections are handled by the ARM_UNWIND_SECTIONS define, which is only added to the list of sections when CONFIG_ARM_UNWIND is set. CONFIG_ARM_UNWIND is a hidden symbol that is only selected when CONFIG_UNWINDER_ARM is set so CONFIG_UNWINDER_FRAME_POINTER never handles these sections. According to the help text of CONFIG_UNWINDER_ARM, these sections should be discarded so that the kernel image size is not affected. Fixes: 5a17850e251a ("arm/build: Warn on orphan section placement") Link: https://github.com/ClangBuiltLinux/linux/issues/1152 Reported-by: kernel test robot Signed-off-by: Nathan Chancellor Review-by: Nick Desaulniers Tested-by: Nick Desaulniers [kees: Made the discard slightly more specific] Signed-off-by: Kees Cook Link: https://lore.kernel.org/r/20200928224854.3224862-1-natechancellor@gmail.com --- arch/arm/kernel/vmlinux.lds.S | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S index 5f4922e858d0..f7f4620d59c3 100644 --- a/arch/arm/kernel/vmlinux.lds.S +++ b/arch/arm/kernel/vmlinux.lds.S @@ -40,6 +40,10 @@ SECTIONS ARM_DISCARD #ifndef CONFIG_SMP_ON_UP *(.alt.smp.init) +#endif +#ifndef CONFIG_ARM_UNWIND + *(.ARM.exidx) *(.ARM.exidx.*) + *(.ARM.extab) *(.ARM.extab.*) #endif } -- cgit v1.2.3 From 85f971b65a692b68181438e099b946cc06ed499b Mon Sep 17 00:00:00 2001 From: Zhang Qilong Date: Tue, 27 Oct 2020 21:49:01 +0800 Subject: ACPI: NFIT: Fix comparison to '-ENXIO' Initial value of rc is '-ENXIO', and we should use the initial value to check it. Signed-off-by: Zhang Qilong Reviewed-by: Pankaj Gupta Reviewed-by: Vishal Verma [ rjw: Subject edit ] Signed-off-by: Rafael J. Wysocki --- drivers/acpi/nfit/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c index 756227837b3b..3a3c209ed3d3 100644 --- a/drivers/acpi/nfit/core.c +++ b/drivers/acpi/nfit/core.c @@ -1564,7 +1564,7 @@ static ssize_t format1_show(struct device *dev, le16_to_cpu(nfit_dcr->dcr->code)); break; } - if (rc != ENXIO) + if (rc != -ENXIO) break; } mutex_unlock(&acpi_desc->init_mutex); -- cgit v1.2.3 From f78f63da916e04e0b12a96010e2dd551954b1db4 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Tue, 27 Oct 2020 19:22:46 +0100 Subject: mm/process_vm_access: Add missing #include MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit With e.g. m68k/defconfig: mm/process_vm_access.c: In function ‘process_vm_rw’: mm/process_vm_access.c:277:5: error: implicit declaration of function ‘in_compat_syscall’ [-Werror=implicit-function-declaration] 277 | in_compat_syscall()); | ^~~~~~~~~~~~~~~~~ Fix this by adding #include . Reported-by: noreply@ellerman.id.au Reported-by: damian Reported-by: Naresh Kamboju Fixes: 38dc5079da7081e8 ("Fix compat regression in process_vm_rw()") Signed-off-by: Geert Uytterhoeven Acked-by: Jens Axboe Signed-off-by: Linus Torvalds --- mm/process_vm_access.c | 1 + 1 file changed, 1 insertion(+) diff --git a/mm/process_vm_access.c b/mm/process_vm_access.c index 05676722d9cd..702250f148e7 100644 --- a/mm/process_vm_access.c +++ b/mm/process_vm_access.c @@ -5,6 +5,7 @@ * Copyright (C) 2010-2011 Christopher Yeoh , IBM Corp. */ +#include #include #include #include -- cgit v1.2.3 From 58d043690deb7c145f8a0bd07370dcaefa97a733 Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Tue, 27 Oct 2020 11:15:58 +0100 Subject: drm/vc4: hdmi: Avoid sleeping in atomic context When running the trigger hook, ALSA by default will take a spinlock, and thus will run the trigger hook in atomic context. However, our HDMI driver will send the infoframes as part of the trigger hook, and part of that process is to wait for a bit to be cleared for up to 100ms. To be nicer to the system, that wait has some usleep_range that interact poorly with the atomic context. There's several ways we can fix this, but the more obvious one is to make ALSA take a mutex instead by setting the nonatomic flag on the DAI link. That doesn't work though, since now the cyclic callback installed by the dmaengine helpers in ALSA will take a mutex, while that callback is run by dmaengine's virt-chan code in a tasklet where sleeping is not allowed either. Given the delay we need to poll the bit for, changing the usleep_range for a udelay and keep running it from a context where interrupts are disabled is not really a good option either. However, we can move the infoframe setup code in the hw_params hook, like is usually done in other HDMI controllers, that isn't protected by a spinlock and thus where we can sleep. Infoframes will be sent on a regular basis anyway, and since hw_params is where the audio parameters that end up in the infoframes are setup, this also makes a bit more sense. Fixes: bb7d78568814 ("drm/vc4: Add HDMI audio support") Suggested-by: Mark Brown Signed-off-by: Maxime Ripard Reviewed-by: Mark Brown Reviewed-by: Takashi Iwai Link: https://patchwork.freedesktop.org/patch/msgid/20201027101558.427256-1-maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_hdmi.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c index e8f99e290655..73ac52714e54 100644 --- a/drivers/gpu/drm/vc4/vc4_hdmi.c +++ b/drivers/gpu/drm/vc4/vc4_hdmi.c @@ -922,6 +922,7 @@ static int vc4_hdmi_audio_hw_params(struct snd_pcm_substream *substream, struct snd_soc_dai *dai) { struct vc4_hdmi *vc4_hdmi = dai_to_hdmi(dai); + struct drm_encoder *encoder = &vc4_hdmi->encoder.base.base; struct device *dev = &vc4_hdmi->pdev->dev; u32 audio_packet_config, channel_mask; u32 channel_map; @@ -981,6 +982,8 @@ static int vc4_hdmi_audio_hw_params(struct snd_pcm_substream *substream, HDMI_WRITE(HDMI_AUDIO_PACKET_CONFIG, audio_packet_config); vc4_hdmi_set_n_cts(vc4_hdmi); + vc4_hdmi_set_audio_infoframe(encoder); + return 0; } @@ -988,11 +991,9 @@ static int vc4_hdmi_audio_trigger(struct snd_pcm_substream *substream, int cmd, struct snd_soc_dai *dai) { struct vc4_hdmi *vc4_hdmi = dai_to_hdmi(dai); - struct drm_encoder *encoder = &vc4_hdmi->encoder.base.base; switch (cmd) { case SNDRV_PCM_TRIGGER_START: - vc4_hdmi_set_audio_infoframe(encoder); vc4_hdmi->audio.streaming = true; if (vc4_hdmi->variant->phy_rng_enable) -- cgit v1.2.3 From 06a17bbe1d47fec6232505c355b367797f6a635c Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 27 Oct 2020 09:39:04 +0000 Subject: afs: Fix copy_file_range() The prevention of splice-write without explicit ops made the copy_file_write() syscall to an afs file (as done by the generic/112 xfstest) fail with EINVAL. Fix by using iter_file_splice_write() for afs. Fixes: 36e2c7421f02 ("fs: don't allow splice read/write without explicit ops") Signed-off-by: David Howells Reviewed-by: Christoph Hellwig --- fs/afs/file.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/afs/file.c b/fs/afs/file.c index 371d1488cc54..91225421ad37 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -33,6 +33,7 @@ const struct file_operations afs_file_operations = { .write_iter = afs_file_write, .mmap = afs_file_mmap, .splice_read = generic_file_splice_read, + .splice_write = iter_file_splice_write, .fsync = afs_fsync, .lock = afs_lock, .flock = afs_flock, -- cgit v1.2.3 From acc080d15dde820bd39eb55a04f9a09c7ef52e67 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 27 Oct 2020 10:42:56 +0000 Subject: afs: Fix tracing deref-before-check The patch dca54a7bbb8c: "afs: Add tracing for cell refcount and active user count" from Oct 13, 2020, leads to the following Smatch complaint: fs/afs/cell.c:596 afs_unuse_cell() warn: variable dereferenced before check 'cell' (see line 592) Fix this by moving the retrieval of the cell debug ID to after the check of the validity of the cell pointer. Reported-by: Dan Carpenter Fixes: dca54a7bbb8c ("afs: Add tracing for cell refcount and active user count") Signed-off-by: David Howells cc: Dan Carpenter --- fs/afs/cell.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/afs/cell.c b/fs/afs/cell.c index 52233fa6195f..887b673f6223 100644 --- a/fs/afs/cell.c +++ b/fs/afs/cell.c @@ -589,7 +589,7 @@ struct afs_cell *afs_use_cell(struct afs_cell *cell, enum afs_cell_trace reason) */ void afs_unuse_cell(struct afs_net *net, struct afs_cell *cell, enum afs_cell_trace reason) { - unsigned int debug_id = cell->debug_id; + unsigned int debug_id; time64_t now, expire_delay; int u, a; @@ -604,6 +604,7 @@ void afs_unuse_cell(struct afs_net *net, struct afs_cell *cell, enum afs_cell_tr if (cell->vl_servers->nr_servers) expire_delay = afs_cell_gc_delay; + debug_id = cell->debug_id; u = atomic_read(&cell->ref); a = atomic_dec_return(&cell->active); trace_afs_cell(debug_id, u, a, reason); -- cgit v1.2.3 From 248c944e2159de4868bef558feea40214aaf8464 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 24 Aug 2020 11:58:12 +0300 Subject: afs: Fix a use after free in afs_xattr_get_acl() The "op" pointer is freed earlier when we call afs_put_operation(). Fixes: e49c7b2f6de7 ("afs: Build an abstraction around an "operation" concept") Signed-off-by: Dan Carpenter Signed-off-by: David Howells cc: Colin Ian King --- fs/afs/xattr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/afs/xattr.c b/fs/afs/xattr.c index 84f3c4f57531..38884d6c57cd 100644 --- a/fs/afs/xattr.c +++ b/fs/afs/xattr.c @@ -85,7 +85,7 @@ static int afs_xattr_get_acl(const struct xattr_handler *handler, if (acl->size <= size) memcpy(buffer, acl->data, acl->size); else - op->error = -ERANGE; + ret = -ERANGE; } } -- cgit v1.2.3 From d383e346f97d6bb0d654bb3d63c44ab106d92d29 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 22 Oct 2020 14:40:31 +0100 Subject: afs: Fix afs_launder_page to not clear PG_writeback Fix afs_launder_page() to not clear PG_writeback on the page it is laundering as the flag isn't set in this case. Fixes: 4343d00872e1 ("afs: Get rid of the afs_writeback record") Signed-off-by: David Howells --- fs/afs/internal.h | 1 + fs/afs/write.c | 10 ++++++---- 2 files changed, 7 insertions(+), 4 deletions(-) diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 81b0485fd22a..289f5dffa46f 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -812,6 +812,7 @@ struct afs_operation { pgoff_t last; /* last page in mapping to deal with */ unsigned first_offset; /* offset into mapping[first] */ unsigned last_to; /* amount of mapping[last] */ + bool laundering; /* Laundering page, PG_writeback not set */ } store; struct { struct iattr *attr; diff --git a/fs/afs/write.c b/fs/afs/write.c index da12abd6db21..b937ec047ec9 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -396,7 +396,8 @@ static void afs_store_data_success(struct afs_operation *op) op->ctime = op->file[0].scb.status.mtime_client; afs_vnode_commit_status(op, &op->file[0]); if (op->error == 0) { - afs_pages_written_back(vnode, op->store.first, op->store.last); + if (!op->store.laundering) + afs_pages_written_back(vnode, op->store.first, op->store.last); afs_stat_v(vnode, n_stores); atomic_long_add((op->store.last * PAGE_SIZE + op->store.last_to) - (op->store.first * PAGE_SIZE + op->store.first_offset), @@ -415,7 +416,7 @@ static const struct afs_operation_ops afs_store_data_operation = { */ static int afs_store_data(struct address_space *mapping, pgoff_t first, pgoff_t last, - unsigned offset, unsigned to) + unsigned offset, unsigned to, bool laundering) { struct afs_vnode *vnode = AFS_FS_I(mapping->host); struct afs_operation *op; @@ -448,6 +449,7 @@ static int afs_store_data(struct address_space *mapping, op->store.last = last; op->store.first_offset = offset; op->store.last_to = to; + op->store.laundering = laundering; op->mtime = vnode->vfs_inode.i_mtime; op->flags |= AFS_OPERATION_UNINTR; op->ops = &afs_store_data_operation; @@ -601,7 +603,7 @@ no_more: if (end > i_size) to = i_size & ~PAGE_MASK; - ret = afs_store_data(mapping, first, last, offset, to); + ret = afs_store_data(mapping, first, last, offset, to, false); switch (ret) { case 0: ret = count; @@ -921,7 +923,7 @@ int afs_launder_page(struct page *page) trace_afs_page_dirty(vnode, tracepoint_string("launder"), page->index, priv); - ret = afs_store_data(mapping, page->index, page->index, t, f); + ret = afs_store_data(mapping, page->index, page->index, t, f, true); } trace_afs_page_dirty(vnode, tracepoint_string("laundered"), -- cgit v1.2.3 From 2a9baf5ad4884108b3c6d56a50e8105ccf8a4ee7 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 27 Oct 2020 10:15:05 +0100 Subject: x86/debug: Fix BTF handling The SDM states that #DB clears DEBUGCTLMSR_BTF, this means that when the bit is set for userspace (TIF_BLOCKSTEP) and a kernel #DB happens first, the BTF bit meant for userspace execution is lost. Have the kernel #DB handler restore the BTF bit when it was requested for userspace. Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Thomas Gleixner Tested-by: Kyle Huey Link: https://lore.kernel.org/r/20201027093607.956147736@infradead.org --- arch/x86/kernel/traps.c | 28 +++++++++++++++++++++------- 1 file changed, 21 insertions(+), 7 deletions(-) diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 3c70fb34028b..b5208aa7351f 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -799,13 +799,6 @@ static __always_inline unsigned long debug_read_clear_dr6(void) */ current->thread.virtual_dr6 = 0; - /* - * The SDM says "The processor clears the BTF flag when it - * generates a debug exception." Clear TIF_BLOCKSTEP to keep - * TIF_BLOCKSTEP in sync with the hardware BTF flag. - */ - clear_thread_flag(TIF_BLOCKSTEP); - return dr6; } @@ -873,6 +866,20 @@ static __always_inline void exc_debug_kernel(struct pt_regs *regs, */ WARN_ON_ONCE(user_mode(regs)); + if (test_thread_flag(TIF_BLOCKSTEP)) { + /* + * The SDM says "The processor clears the BTF flag when it + * generates a debug exception." but PTRACE_BLOCKSTEP requested + * it for userspace, but we just took a kernel #DB, so re-set + * BTF. + */ + unsigned long debugctl; + + rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl); + debugctl |= DEBUGCTLMSR_BTF; + wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctl); + } + /* * Catch SYSENTER with TF set and clear DR_STEP. If this hit a * watchpoint at the same time then that will still be handled. @@ -935,6 +942,13 @@ static __always_inline void exc_debug_user(struct pt_regs *regs, irqentry_enter_from_user_mode(regs); instrumentation_begin(); + /* + * The SDM says "The processor clears the BTF flag when it + * generates a debug exception." Clear TIF_BLOCKSTEP to keep + * TIF_BLOCKSTEP in sync with the hardware BTF flag. + */ + clear_thread_flag(TIF_BLOCKSTEP); + /* * If dr6 has no reason to give us about the origin of this trap, * then it's very likely the result of an icebp/int01 trap. -- cgit v1.2.3 From a195f3d4528a2f88d6f986f6b1101775ad4891cf Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 27 Oct 2020 10:15:06 +0100 Subject: x86/debug: Only clear/set ->virtual_dr6 for userspace #DB The ->virtual_dr6 is the value used by ptrace_{get,set}_debugreg(6). A kernel #DB clearing it could mean spurious malfunction of ptrace() expectations. Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Thomas Gleixner Tested-by: Kyle Huey Link: https://lore.kernel.org/r/20201027093608.028952500@infradead.org --- arch/x86/kernel/traps.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index b5208aa7351f..32b2d39272e0 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -793,12 +793,6 @@ static __always_inline unsigned long debug_read_clear_dr6(void) set_debugreg(DR6_RESERVED, 6); dr6 ^= DR6_RESERVED; /* Flip to positive polarity */ - /* - * Clear the virtual DR6 value, ptrace routines will set bits here for - * things we want signals for. - */ - current->thread.virtual_dr6 = 0; - return dr6; } @@ -942,6 +936,12 @@ static __always_inline void exc_debug_user(struct pt_regs *regs, irqentry_enter_from_user_mode(regs); instrumentation_begin(); + /* + * Clear the virtual DR6 value, ptrace() routines will set bits here + * for things it wants signals for. + */ + current->thread.virtual_dr6 = 0; + /* * The SDM says "The processor clears the BTF flag when it * generates a debug exception." Clear TIF_BLOCKSTEP to keep -- cgit v1.2.3 From cb05143bdf428f280a5d519c82abf196d7871c11 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 27 Oct 2020 19:33:30 +0100 Subject: x86/debug: Fix DR_STEP vs ptrace_get_debugreg(6) Commit d53d9bc0cf78 ("x86/debug: Change thread.debugreg6 to thread.virtual_dr6") changed the semantics of the variable from random collection of bits, to exactly only those bits that ptrace() needs. Unfortunately this lost DR_STEP for PTRACE_{BLOCK,SINGLE}STEP. Furthermore, it turns out that userspace expects DR_STEP to be unconditionally available, even for manual TF usage outside of PTRACE_{BLOCK,SINGLE}_STEP. Fixes: d53d9bc0cf78 ("x86/debug: Change thread.debugreg6 to thread.virtual_dr6") Reported-by: Kyle Huey Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Thomas Gleixner Tested-by: Kyle Huey Link: https://lore.kernel.org/r/20201027183330.GM2628@hirez.programming.kicks-ass.net --- arch/x86/kernel/traps.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 32b2d39272e0..e19df6cde35d 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -937,10 +937,13 @@ static __always_inline void exc_debug_user(struct pt_regs *regs, instrumentation_begin(); /* - * Clear the virtual DR6 value, ptrace() routines will set bits here - * for things it wants signals for. + * Start the virtual/ptrace DR6 value with just the DR_STEP mask + * of the real DR6. ptrace_triggered() will set the DR_TRAPn bits. + * + * Userspace expects DR_STEP to be visible in ptrace_get_debugreg(6) + * even if it is not the result of PTRACE_SINGLESTEP. */ - current->thread.virtual_dr6 = 0; + current->thread.virtual_dr6 = (dr6 & DR_STEP); /* * The SDM says "The processor clears the BTF flag when it -- cgit v1.2.3 From 28e9dcd9172028263c8225c15c4e329e08475e89 Mon Sep 17 00:00:00 2001 From: Vinay Kumar Yadav Date: Mon, 26 Oct 2020 01:05:39 +0530 Subject: chelsio/chtls: fix deadlock issue In chtls_pass_establish() we hold child socket lock using bh_lock_sock and we are again trying bh_lock_sock in add_to_reap_list, causing deadlock. Remove bh_lock_sock in add_to_reap_list() as lock is already held. Fixes: cc35c88ae4db ("crypto : chtls - CPL handler definition") Signed-off-by: Vinay Kumar Yadav Link: https://lore.kernel.org/r/20201025193538.31112-1-vinay.yadav@chelsio.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c index ec4f79049a06..2fdcac0bcb14 100644 --- a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c +++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c @@ -1514,7 +1514,6 @@ static void add_to_reap_list(struct sock *sk) struct chtls_sock *csk = sk->sk_user_data; local_bh_disable(); - bh_lock_sock(sk); release_tcp_port(sk); /* release the port immediately */ spin_lock(&reap_list_lock); @@ -1523,7 +1522,6 @@ static void add_to_reap_list(struct sock *sk) if (!csk->passive_reap_next) schedule_work(&reap_task); spin_unlock(&reap_list_lock); - bh_unlock_sock(sk); local_bh_enable(); } -- cgit v1.2.3 From 6daa1da4e262b0cd52ef0acc1989ff22b5540264 Mon Sep 17 00:00:00 2001 From: Vinay Kumar Yadav Date: Mon, 26 Oct 2020 01:12:29 +0530 Subject: chelsio/chtls: fix memory leaks in CPL handlers CPL handler functions chtls_pass_open_rpl() and chtls_close_listsrv_rpl() should return CPL_RET_BUF_DONE so that caller function will do skb free to avoid leak. Fixes: cc35c88ae4db ("crypto : chtls - CPL handler definition") Signed-off-by: Vinay Kumar Yadav Link: https://lore.kernel.org/r/20201025194228.31271-1-vinay.yadav@chelsio.com Signed-off-by: Jakub Kicinski --- .../chelsio/inline_crypto/chtls/chtls_cm.c | 27 ++++++++++------------ 1 file changed, 12 insertions(+), 15 deletions(-) diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c index 2fdcac0bcb14..d581c4e623f8 100644 --- a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c +++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c @@ -772,14 +772,13 @@ static int chtls_pass_open_rpl(struct chtls_dev *cdev, struct sk_buff *skb) if (rpl->status != CPL_ERR_NONE) { pr_info("Unexpected PASS_OPEN_RPL status %u for STID %u\n", rpl->status, stid); - return CPL_RET_BUF_DONE; + } else { + cxgb4_free_stid(cdev->tids, stid, listen_ctx->lsk->sk_family); + sock_put(listen_ctx->lsk); + kfree(listen_ctx); + module_put(THIS_MODULE); } - cxgb4_free_stid(cdev->tids, stid, listen_ctx->lsk->sk_family); - sock_put(listen_ctx->lsk); - kfree(listen_ctx); - module_put(THIS_MODULE); - - return 0; + return CPL_RET_BUF_DONE; } static int chtls_close_listsrv_rpl(struct chtls_dev *cdev, struct sk_buff *skb) @@ -796,15 +795,13 @@ static int chtls_close_listsrv_rpl(struct chtls_dev *cdev, struct sk_buff *skb) if (rpl->status != CPL_ERR_NONE) { pr_info("Unexpected CLOSE_LISTSRV_RPL status %u for STID %u\n", rpl->status, stid); - return CPL_RET_BUF_DONE; + } else { + cxgb4_free_stid(cdev->tids, stid, listen_ctx->lsk->sk_family); + sock_put(listen_ctx->lsk); + kfree(listen_ctx); + module_put(THIS_MODULE); } - - cxgb4_free_stid(cdev->tids, stid, listen_ctx->lsk->sk_family); - sock_put(listen_ctx->lsk); - kfree(listen_ctx); - module_put(THIS_MODULE); - - return 0; + return CPL_RET_BUF_DONE; } static void chtls_purge_wr_queue(struct sock *sk) -- cgit v1.2.3 From 6c211809c87f0de939f3bd60ceec05338ae6eba1 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 26 Oct 2020 11:00:59 +0300 Subject: devlink: Fix some error codes These paths don't set the error codes. It's especially important in devlink_nl_region_notify_build() where it leads to a NULL dereference in the caller. Fixes: 544e7c33ec2f ("net: devlink: Add support for port regions") Signed-off-by: Dan Carpenter Link: https://lore.kernel.org/r/20201026080059.GA1628785@mwanda Signed-off-by: Jakub Kicinski --- net/core/devlink.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) diff --git a/net/core/devlink.c b/net/core/devlink.c index a578634052a3..925da7c0fcf3 100644 --- a/net/core/devlink.c +++ b/net/core/devlink.c @@ -4213,10 +4213,12 @@ static int devlink_nl_region_fill(struct sk_buff *msg, struct devlink *devlink, if (err) goto nla_put_failure; - if (region->port) - if (nla_put_u32(msg, DEVLINK_ATTR_PORT_INDEX, - region->port->index)) + if (region->port) { + err = nla_put_u32(msg, DEVLINK_ATTR_PORT_INDEX, + region->port->index); + if (err) goto nla_put_failure; + } err = nla_put_string(msg, DEVLINK_ATTR_REGION_NAME, region->ops->name); if (err) @@ -4265,10 +4267,12 @@ devlink_nl_region_notify_build(struct devlink_region *region, if (err) goto out_cancel_msg; - if (region->port) - if (nla_put_u32(msg, DEVLINK_ATTR_PORT_INDEX, - region->port->index)) + if (region->port) { + err = nla_put_u32(msg, DEVLINK_ATTR_PORT_INDEX, + region->port->index); + if (err) goto out_cancel_msg; + } err = nla_put_string(msg, DEVLINK_ATTR_REGION_NAME, region->ops->name); @@ -4962,10 +4966,12 @@ static int devlink_nl_cmd_region_read_dumpit(struct sk_buff *skb, if (err) goto nla_put_failure; - if (region->port) - if (nla_put_u32(skb, DEVLINK_ATTR_PORT_INDEX, - region->port->index)) + if (region->port) { + err = nla_put_u32(skb, DEVLINK_ATTR_PORT_INDEX, + region->port->index); + if (err) goto nla_put_failure; + } err = nla_put_string(skb, DEVLINK_ATTR_REGION_NAME, region_name); if (err) -- cgit v1.2.3 From 0d8cb9464a7d9c3e6349db3f7719a80f3793347e Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 26 Oct 2020 11:01:27 +0300 Subject: devlink: Unlock on error in dumpit() This needs to unlock before returning. Fixes: 544e7c33ec2f ("net: devlink: Add support for port regions") Signed-off-by: Dan Carpenter Link: https://lore.kernel.org/r/20201026080127.GB1628785@mwanda Signed-off-by: Jakub Kicinski --- net/core/devlink.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/core/devlink.c b/net/core/devlink.c index 925da7c0fcf3..a932d95be798 100644 --- a/net/core/devlink.c +++ b/net/core/devlink.c @@ -4919,8 +4919,10 @@ static int devlink_nl_cmd_region_read_dumpit(struct sk_buff *skb, index = nla_get_u32(info->attrs[DEVLINK_ATTR_PORT_INDEX]); port = devlink_port_get_by_index(devlink, index); - if (!port) - return -ENODEV; + if (!port) { + err = -ENODEV; + goto out_unlock; + } } region_name = nla_data(attrs[DEVLINK_ATTR_REGION_NAME]); -- cgit v1.2.3 From 68b9f0865b1ef545da180c57d54b82c94cb464a4 Mon Sep 17 00:00:00 2001 From: Andrew Gabbasov Date: Mon, 26 Oct 2020 05:21:30 -0500 Subject: ravb: Fix bit fields checking in ravb_hwtstamp_get() In the function ravb_hwtstamp_get() in ravb_main.c with the existing values for RAVB_RXTSTAMP_TYPE_V2_L2_EVENT (0x2) and RAVB_RXTSTAMP_TYPE_ALL (0x6) if (priv->tstamp_rx_ctrl & RAVB_RXTSTAMP_TYPE_V2_L2_EVENT) config.rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT; else if (priv->tstamp_rx_ctrl & RAVB_RXTSTAMP_TYPE_ALL) config.rx_filter = HWTSTAMP_FILTER_ALL; if the test on RAVB_RXTSTAMP_TYPE_ALL should be true, it will never be reached. This issue can be verified with 'hwtstamp_config' testing program (tools/testing/selftests/net/hwtstamp_config.c). Setting filter type to ALL and subsequent retrieving it gives incorrect value: $ hwtstamp_config eth0 OFF ALL flags = 0 tx_type = OFF rx_filter = ALL $ hwtstamp_config eth0 flags = 0 tx_type = OFF rx_filter = PTP_V2_L2_EVENT Correct this by converting if-else's to switch. Fixes: c156633f1353 ("Renesas Ethernet AVB driver proper") Reported-by: Julia Lawall Signed-off-by: Andrew Gabbasov Reviewed-by: Sergei Shtylyov Link: https://lore.kernel.org/r/20201026102130.29368-1-andrew_gabbasov@mentor.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/renesas/ravb_main.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 9c4df4ede011..bd30505fbc57 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -1744,12 +1744,16 @@ static int ravb_hwtstamp_get(struct net_device *ndev, struct ifreq *req) config.flags = 0; config.tx_type = priv->tstamp_tx_ctrl ? HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; - if (priv->tstamp_rx_ctrl & RAVB_RXTSTAMP_TYPE_V2_L2_EVENT) + switch (priv->tstamp_rx_ctrl & RAVB_RXTSTAMP_TYPE) { + case RAVB_RXTSTAMP_TYPE_V2_L2_EVENT: config.rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT; - else if (priv->tstamp_rx_ctrl & RAVB_RXTSTAMP_TYPE_ALL) + break; + case RAVB_RXTSTAMP_TYPE_ALL: config.rx_filter = HWTSTAMP_FILTER_ALL; - else + break; + default: config.rx_filter = HWTSTAMP_FILTER_NONE; + } return copy_to_user(req->ifr_data, &config, sizeof(config)) ? -EFAULT : 0; -- cgit v1.2.3 From 501b72ae2472a15a80c0f4063ee8341870e1ef55 Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Mon, 26 Oct 2020 11:29:45 +0100 Subject: net/sched: act_mpls: Add softdep on mpls_gso.ko TCA_MPLS_ACT_PUSH and TCA_MPLS_ACT_MAC_PUSH might be used on gso packets. Such packets will thus require mpls_gso.ko for segmentation. v2: Drop dependency on CONFIG_NET_MPLS_GSO in Kconfig (from Jakub and David). Fixes: 2a2ea50870ba ("net: sched: add mpls manipulation actions to TC") Signed-off-by: Guillaume Nault Link: https://lore.kernel.org/r/1f6cab15bbd15666795061c55563aaf6a386e90e.1603708007.git.gnault@redhat.com Signed-off-by: Jakub Kicinski --- net/sched/act_mpls.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/sched/act_mpls.c b/net/sched/act_mpls.c index f40bf9771cb9..5c7456e5b5cf 100644 --- a/net/sched/act_mpls.c +++ b/net/sched/act_mpls.c @@ -426,6 +426,7 @@ static void __exit mpls_cleanup_module(void) module_init(mpls_init_module); module_exit(mpls_cleanup_module); +MODULE_SOFTDEP("post: mpls_gso"); MODULE_AUTHOR("Netronome Systems "); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("MPLS manipulation actions"); -- cgit v1.2.3 From 2ac8af0967aaa2b67cb382727e784900d2f4d0da Mon Sep 17 00:00:00 2001 From: Thomas Bogendoerfer Date: Mon, 26 Oct 2020 11:42:21 +0100 Subject: ibmveth: Fix use of ibmveth in a bridge. The check for src mac address in ibmveth_is_packet_unsupported is wrong. Commit 6f2275433a2f wanted to shut down messages for loopback packets, but now suppresses bridged frames, which are accepted by the hypervisor otherwise bridging won't work at all. Fixes: 6f2275433a2f ("ibmveth: Detect unsupported packets before sending to the hypervisor") Signed-off-by: Michal Suchanek Signed-off-by: Thomas Bogendoerfer Link: https://lore.kernel.org/r/20201026104221.26570-1-msuchanek@suse.de Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ibm/ibmveth.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/drivers/net/ethernet/ibm/ibmveth.c b/drivers/net/ethernet/ibm/ibmveth.c index 7ef3369953b6..c3ec9ceed833 100644 --- a/drivers/net/ethernet/ibm/ibmveth.c +++ b/drivers/net/ethernet/ibm/ibmveth.c @@ -1031,12 +1031,6 @@ static int ibmveth_is_packet_unsupported(struct sk_buff *skb, ret = -EOPNOTSUPP; } - if (!ether_addr_equal(ether_header->h_source, netdev->dev_addr)) { - netdev_dbg(netdev, "source packet MAC address does not match veth device's, dropping packet.\n"); - netdev->stats.tx_dropped++; - ret = -EOPNOTSUPP; - } - return ret; } -- cgit v1.2.3 From d6535dca28859d8d9ef80894eb287b2ac35a32e8 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 26 Oct 2020 14:33:27 +0200 Subject: net: protect tcf_block_unbind with block lock The tcf_block_unbind() expects that the caller will take block->cb_lock before calling it, however the code took RTNL lock and dropped cb_lock instead. This causes to the following kernel panic. WARNING: CPU: 1 PID: 13524 at net/sched/cls_api.c:1488 tcf_block_unbind+0x2db/0x420 Modules linked in: mlx5_ib mlx5_core mlxfw ptp pps_core act_mirred act_tunnel_key cls_flower vxlan ip6_udp_tunnel udp_tunnel dummy sch_ingress openvswitch nsh xt_conntrack xt_MASQUERADE nf_conntrack_netlink nfnetlink xt_addrtype iptable_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 br_netfilter rpcrdma rdma_ucm ib_iser libiscsi scsi_transport_iscsi ib_umad ib_ipoib rdma_cm iw_cm ib_cm ib_uverbs ib_core overlay [last unloaded: mlxfw] CPU: 1 PID: 13524 Comm: test-ecmp-add-v Tainted: G W 5.9.0+ #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 RIP: 0010:tcf_block_unbind+0x2db/0x420 Code: ff 48 83 c4 40 5b 5d 41 5c 41 5d 41 5e 41 5f c3 49 8d bc 24 30 01 00 00 be ff ff ff ff e8 7d 7f 70 00 85 c0 0f 85 7b fd ff ff <0f> 0b e9 74 fd ff ff 48 c7 c7 dc 6a 24 84 e8 02 ec fe fe e9 55 fd RSP: 0018:ffff888117d17968 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff88812f713c00 RCX: 1ffffffff0848d5b RDX: 0000000000000001 RSI: ffff88814fbc8130 RDI: ffff888107f2b878 RBP: 1ffff11022fa2f3f R08: 0000000000000000 R09: ffffffff84115a87 R10: fffffbfff0822b50 R11: ffff888107f2b898 R12: ffff88814fbc8000 R13: ffff88812f713c10 R14: ffff888117d17a38 R15: ffff88814fbc80c0 FS: 00007f6593d36740(0000) GS:ffff8882a4f00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00005607a00758f8 CR3: 0000000131aea006 CR4: 0000000000170ea0 Call Trace: tc_block_indr_cleanup+0x3e0/0x5a0 ? tcf_block_unbind+0x420/0x420 ? __mutex_unlock_slowpath+0xe7/0x610 flow_indr_dev_unregister+0x5e2/0x930 ? mlx5e_restore_tunnel+0xdf0/0xdf0 [mlx5_core] ? mlx5e_restore_tunnel+0xdf0/0xdf0 [mlx5_core] ? flow_indr_block_cb_alloc+0x3c0/0x3c0 ? mlx5_db_free+0x37c/0x4b0 [mlx5_core] mlx5e_cleanup_rep_tx+0x8b/0xc0 [mlx5_core] mlx5e_detach_netdev+0xe5/0x120 [mlx5_core] mlx5e_vport_rep_unload+0x155/0x260 [mlx5_core] esw_offloads_disable+0x227/0x2b0 [mlx5_core] mlx5_eswitch_disable_locked.cold+0x38e/0x699 [mlx5_core] mlx5_eswitch_disable+0x94/0xf0 [mlx5_core] mlx5_device_disable_sriov+0x183/0x1f0 [mlx5_core] mlx5_core_sriov_configure+0xfd/0x230 [mlx5_core] sriov_numvfs_store+0x261/0x2f0 ? sriov_drivers_autoprobe_store+0x110/0x110 ? sysfs_file_ops+0x170/0x170 ? sysfs_file_ops+0x117/0x170 ? sysfs_file_ops+0x170/0x170 kernfs_fop_write+0x1ff/0x3f0 ? rcu_read_lock_any_held+0x6e/0x90 vfs_write+0x1f3/0x620 ksys_write+0xf9/0x1d0 ? __x64_sys_read+0xb0/0xb0 ? lockdep_hardirqs_on_prepare+0x273/0x3f0 ? syscall_enter_from_user_mode+0x1d/0x50 do_syscall_64+0x2d/0x40 entry_SYSCALL_64_after_hwframe+0x44/0xa9 <...> ---[ end trace bfdd028ada702879 ]--- Fixes: 0fdcf78d5973 ("net: use flow_indr_dev_setup_offload()") Signed-off-by: Leon Romanovsky Link: https://lore.kernel.org/r/20201026123327.1141066-1-leon@kernel.org Signed-off-by: Jakub Kicinski --- net/sched/cls_api.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index faeabff283a2..838b3fd94d77 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -652,12 +652,12 @@ static void tc_block_indr_cleanup(struct flow_block_cb *block_cb) block_cb->indr.binder_type, &block->flow_block, tcf_block_shared(block), &extack); + rtnl_lock(); down_write(&block->cb_lock); list_del(&block_cb->driver_list); list_move(&block_cb->list, &bo.cb_list); - up_write(&block->cb_lock); - rtnl_lock(); tcf_block_unbind(block, &bo); + up_write(&block->cb_lock); rtnl_unlock(); } -- cgit v1.2.3 From 35bc10b2eafbb701064b94f283b77c54d3304842 Mon Sep 17 00:00:00 2001 From: Kanchan Joshi Date: Mon, 28 Sep 2020 15:25:49 +0530 Subject: null_blk: synchronization fix for zoned device Parallel write,read,zone-mgmt operations accessing/altering zone state and write-pointer may get into race. Avoid the situation by using a new spinlock for zoned device. Concurrent zone-appends (on a zone) returning same write-pointer issue is also avoided using this lock. Cc: stable@vger.kernel.org Fixes: e0489ed5daeb ("null_blk: Support REQ_OP_ZONE_APPEND") Signed-off-by: Kanchan Joshi Reviewed-by: Damien Le Moal Signed-off-by: Jens Axboe --- drivers/block/null_blk.h | 1 + drivers/block/null_blk_zoned.c | 22 ++++++++++++++++++---- 2 files changed, 19 insertions(+), 4 deletions(-) diff --git a/drivers/block/null_blk.h b/drivers/block/null_blk.h index d2e7db43a52a..3176b269b822 100644 --- a/drivers/block/null_blk.h +++ b/drivers/block/null_blk.h @@ -47,6 +47,7 @@ struct nullb_device { unsigned int nr_zones_closed; struct blk_zone *zones; sector_t zone_size_sects; + spinlock_t zone_lock; unsigned long size; /* device size in MB */ unsigned long completion_nsec; /* time in ns to complete a request */ diff --git a/drivers/block/null_blk_zoned.c b/drivers/block/null_blk_zoned.c index 7d94f2d47a6a..98056c88926b 100644 --- a/drivers/block/null_blk_zoned.c +++ b/drivers/block/null_blk_zoned.c @@ -45,6 +45,7 @@ int null_init_zoned_dev(struct nullb_device *dev, struct request_queue *q) if (!dev->zones) return -ENOMEM; + spin_lock_init(&dev->zone_lock); if (dev->zone_nr_conv >= dev->nr_zones) { dev->zone_nr_conv = dev->nr_zones - 1; pr_info("changed the number of conventional zones to %u", @@ -149,8 +150,11 @@ int null_report_zones(struct gendisk *disk, sector_t sector, * So use a local copy to avoid corruption of the device zone * array. */ + spin_lock_irq(&dev->zone_lock); memcpy(&zone, &dev->zones[first_zone + i], sizeof(struct blk_zone)); + spin_unlock_irq(&dev->zone_lock); + error = cb(&zone, i, data); if (error) return error; @@ -499,18 +503,28 @@ static blk_status_t null_zone_mgmt(struct nullb_cmd *cmd, enum req_opf op, blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd, enum req_opf op, sector_t sector, sector_t nr_sectors) { + blk_status_t sts; + struct nullb_device *dev = cmd->nq->dev; + + spin_lock_irq(&dev->zone_lock); switch (op) { case REQ_OP_WRITE: - return null_zone_write(cmd, sector, nr_sectors, false); + sts = null_zone_write(cmd, sector, nr_sectors, false); + break; case REQ_OP_ZONE_APPEND: - return null_zone_write(cmd, sector, nr_sectors, true); + sts = null_zone_write(cmd, sector, nr_sectors, true); + break; case REQ_OP_ZONE_RESET: case REQ_OP_ZONE_RESET_ALL: case REQ_OP_ZONE_OPEN: case REQ_OP_ZONE_CLOSE: case REQ_OP_ZONE_FINISH: - return null_zone_mgmt(cmd, op, sector); + sts = null_zone_mgmt(cmd, op, sector); + break; default: - return null_process_cmd(cmd, op, sector, nr_sectors); + sts = null_process_cmd(cmd, op, sector, nr_sectors); } + spin_unlock_irq(&dev->zone_lock); + + return sts; } -- cgit v1.2.3 From defe40af1a7143a0538d7c3e87224459eea0a877 Mon Sep 17 00:00:00 2001 From: Pawel Laszczak Date: Mon, 12 Oct 2020 08:45:48 +0200 Subject: usb: cdns3: Variable 'length' set but not used Patch removes not used variable 'length' from cdns3_wa2_descmiss_copy_data function. Fixes: 141e70fef4ee ("usb: cdns3: gadget: need to handle sg case for workaround 2 case") Acked-by: Roger Quadros Signed-off-by: Pawel Laszczak Signed-off-by: Peter Chen --- drivers/usb/cdns3/gadget.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/usb/cdns3/gadget.c b/drivers/usb/cdns3/gadget.c index 6e7b70a2e352..692acf7b9b14 100644 --- a/drivers/usb/cdns3/gadget.c +++ b/drivers/usb/cdns3/gadget.c @@ -506,7 +506,6 @@ static void cdns3_wa2_descmiss_copy_data(struct cdns3_endpoint *priv_ep, while (!list_empty(&priv_ep->wa2_descmiss_req_list)) { int chunk_end; - int length; descmiss_priv_req = cdns3_next_priv_request(&priv_ep->wa2_descmiss_req_list); @@ -517,7 +516,6 @@ static void cdns3_wa2_descmiss_copy_data(struct cdns3_endpoint *priv_ep, break; chunk_end = descmiss_priv_req->flags & REQUEST_INTERNAL_CH; - length = request->actual + descmiss_req->actual; request->status = descmiss_req->status; __cdns3_descmiss_copy_data(request, descmiss_req); list_del_init(&descmiss_priv_req->list); -- cgit v1.2.3 From 011fde48394b7dc8dfd6660d1013b26a00157b80 Mon Sep 17 00:00:00 2001 From: Yangbo Lu Date: Tue, 20 Oct 2020 16:11:16 +0800 Subject: mmc: sdhci-of-esdhc: make sure delay chain locked for HS400 For eMMC HS400 mode initialization, the DLL reset is a required step if DLL is enabled to use previously, like in bootloader. This step has not been documented in reference manual, but the RM will be fixed sooner or later. This patch is to add the step of DLL reset, and make sure delay chain locked for HS400. Signed-off-by: Yangbo Lu Acked-by: Adrian Hunter Link: https://lore.kernel.org/r/20201020081116.20918-1-yangbo.lu@nxp.com Fixes: 54e08d9a95ca ("mmc: sdhci-of-esdhc: add hs400 mode support") Cc: stable@vger.kernel.org Signed-off-by: Ulf Hansson --- drivers/mmc/host/sdhci-esdhc.h | 2 ++ drivers/mmc/host/sdhci-of-esdhc.c | 17 +++++++++++++++++ 2 files changed, 19 insertions(+) diff --git a/drivers/mmc/host/sdhci-esdhc.h b/drivers/mmc/host/sdhci-esdhc.h index a30796e79b1c..6de02f09c322 100644 --- a/drivers/mmc/host/sdhci-esdhc.h +++ b/drivers/mmc/host/sdhci-esdhc.h @@ -5,6 +5,7 @@ * Copyright (c) 2007 Freescale Semiconductor, Inc. * Copyright (c) 2009 MontaVista Software, Inc. * Copyright (c) 2010 Pengutronix e.K. + * Copyright 2020 NXP * Author: Wolfram Sang */ @@ -88,6 +89,7 @@ /* DLL Config 0 Register */ #define ESDHC_DLLCFG0 0x160 #define ESDHC_DLL_ENABLE 0x80000000 +#define ESDHC_DLL_RESET 0x40000000 #define ESDHC_DLL_FREQ_SEL 0x08000000 /* DLL Config 1 Register */ diff --git a/drivers/mmc/host/sdhci-of-esdhc.c b/drivers/mmc/host/sdhci-of-esdhc.c index baf7801a1804..bb094459196a 100644 --- a/drivers/mmc/host/sdhci-of-esdhc.c +++ b/drivers/mmc/host/sdhci-of-esdhc.c @@ -4,6 +4,7 @@ * * Copyright (c) 2007, 2010, 2012 Freescale Semiconductor, Inc. * Copyright (c) 2009 MontaVista Software, Inc. + * Copyright 2020 NXP * * Authors: Xiaobo Xie * Anton Vorontsov @@ -19,6 +20,7 @@ #include #include #include +#include #include #include #include "sdhci-pltfm.h" @@ -743,6 +745,21 @@ static void esdhc_of_set_clock(struct sdhci_host *host, unsigned int clock) if (host->mmc->actual_clock == MMC_HS200_MAX_DTR) temp |= ESDHC_DLL_FREQ_SEL; sdhci_writel(host, temp, ESDHC_DLLCFG0); + + temp |= ESDHC_DLL_RESET; + sdhci_writel(host, temp, ESDHC_DLLCFG0); + udelay(1); + temp &= ~ESDHC_DLL_RESET; + sdhci_writel(host, temp, ESDHC_DLLCFG0); + + /* Wait max 20 ms */ + if (read_poll_timeout(sdhci_readl, temp, + temp & ESDHC_DLL_STS_SLV_LOCK, + 10, 20000, false, + host, ESDHC_DLLSTAT0)) + pr_err("%s: timeout for delay chain lock.\n", + mmc_hostname(host->mmc)); + temp = sdhci_readl(host, ESDHC_TBCTL); sdhci_writel(host, temp | ESDHC_HS400_WNDW_ADJUST, ESDHC_TBCTL); -- cgit v1.2.3 From 1de111b51b829bcf01d2e57971f8fd07a665fa3f Mon Sep 17 00:00:00 2001 From: Stephen Boyd Date: Fri, 23 Oct 2020 08:47:50 -0700 Subject: KVM: arm64: ARM_SMCCC_ARCH_WORKAROUND_1 doesn't return SMCCC_RET_NOT_REQUIRED According to the SMCCC spec[1](7.5.2 Discovery) the ARM_SMCCC_ARCH_WORKAROUND_1 function id only returns 0, 1, and SMCCC_RET_NOT_SUPPORTED. 0 is "workaround required and safe to call this function" 1 is "workaround not required but safe to call this function" SMCCC_RET_NOT_SUPPORTED is "might be vulnerable or might not be, who knows, I give up!" SMCCC_RET_NOT_SUPPORTED might as well mean "workaround required, except calling this function may not work because it isn't implemented in some cases". Wonderful. We map this SMC call to 0 is SPECTRE_MITIGATED 1 is SPECTRE_UNAFFECTED SMCCC_RET_NOT_SUPPORTED is SPECTRE_VULNERABLE For KVM hypercalls (hvc), we've implemented this function id to return SMCCC_RET_NOT_SUPPORTED, 0, and SMCCC_RET_NOT_REQUIRED. One of those isn't supposed to be there. Per the code we call arm64_get_spectre_v2_state() to figure out what to return for this feature discovery call. 0 is SPECTRE_MITIGATED SMCCC_RET_NOT_REQUIRED is SPECTRE_UNAFFECTED SMCCC_RET_NOT_SUPPORTED is SPECTRE_VULNERABLE Let's clean this up so that KVM tells the guest this mapping: 0 is SPECTRE_MITIGATED 1 is SPECTRE_UNAFFECTED SMCCC_RET_NOT_SUPPORTED is SPECTRE_VULNERABLE Note: SMCCC_RET_NOT_AFFECTED is 1 but isn't part of the SMCCC spec Fixes: c118bbb52743 ("arm64: KVM: Propagate full Spectre v2 workaround state to KVM guests") Signed-off-by: Stephen Boyd Acked-by: Marc Zyngier Acked-by: Will Deacon Cc: Andre Przywara Cc: Steven Price Cc: Marc Zyngier Cc: stable@vger.kernel.org Link: https://developer.arm.com/documentation/den0028/latest [1] Link: https://lore.kernel.org/r/20201023154751.1973872-1-swboyd@chromium.org Signed-off-by: Will Deacon --- arch/arm64/kernel/proton-pack.c | 2 -- arch/arm64/kvm/hypercalls.c | 2 +- include/linux/arm-smccc.h | 2 ++ 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm64/kernel/proton-pack.c b/arch/arm64/kernel/proton-pack.c index 25f3c80b5ffe..c18eb7d41274 100644 --- a/arch/arm64/kernel/proton-pack.c +++ b/arch/arm64/kernel/proton-pack.c @@ -135,8 +135,6 @@ static enum mitigation_state spectre_v2_get_cpu_hw_mitigation_state(void) return SPECTRE_VULNERABLE; } -#define SMCCC_ARCH_WORKAROUND_RET_UNAFFECTED (1) - static enum mitigation_state spectre_v2_get_cpu_fw_mitigation_state(void) { int ret; diff --git a/arch/arm64/kvm/hypercalls.c b/arch/arm64/kvm/hypercalls.c index 9824025ccc5c..25ea4ecb6449 100644 --- a/arch/arm64/kvm/hypercalls.c +++ b/arch/arm64/kvm/hypercalls.c @@ -31,7 +31,7 @@ int kvm_hvc_call_handler(struct kvm_vcpu *vcpu) val = SMCCC_RET_SUCCESS; break; case SPECTRE_UNAFFECTED: - val = SMCCC_RET_NOT_REQUIRED; + val = SMCCC_ARCH_WORKAROUND_RET_UNAFFECTED; break; } break; diff --git a/include/linux/arm-smccc.h b/include/linux/arm-smccc.h index 885c9ffc835c..f860645f6512 100644 --- a/include/linux/arm-smccc.h +++ b/include/linux/arm-smccc.h @@ -87,6 +87,8 @@ ARM_SMCCC_SMC_32, \ 0, 0x7fff) +#define SMCCC_ARCH_WORKAROUND_RET_UNAFFECTED 1 + /* Paravirtualised time calls (defined by ARM DEN0057A) */ #define ARM_SMCCC_HV_PV_TIME_FEATURES \ ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL, \ -- cgit v1.2.3 From f49a51bfdc8ea717c97ccd4cc98b7e6daaa5553a Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Tue, 27 Oct 2020 22:49:22 +0100 Subject: drm/shme-helpers: Fix dma_buf_mmap forwarding bug MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When we forward an mmap to the dma_buf exporter, they get to own everything. Unfortunately drm_gem_mmap_obj() overwrote vma->vm_private_data after the driver callback, wreaking the exporter complete. This was noticed because vb2_common_vm_close blew up on mali gpu with panfrost after commit 26d3ac3cb04d ("drm/shmem-helpers: Redirect mmap for imported dma-buf"). Unfortunately drm_gem_mmap_obj also acquires a surplus reference that we need to drop in shmem helpers, which is a bit of a mislayer situation. Maybe the entire dma_buf_mmap forwarding should be pulled into core gem code. Note that the only two other drivers which forward mmap in their own code (etnaviv and exynos) get this somewhat right by overwriting the gem mmap code. But they seem to still have the leak. This might be a good excuse to move these drivers over to shmem helpers completely. Reviewed-by: Boris Brezillon Acked-by: Christian König Cc: Christian König Cc: Sumit Semwal Cc: Lucas Stach Cc: Russell King Cc: Christian Gmeiner Cc: Inki Dae Cc: Joonyoung Shim Cc: Seung-Woo Kim Cc: Kyungmin Park Fixes: 26d3ac3cb04d ("drm/shmem-helpers: Redirect mmap for imported dma-buf") Cc: Boris Brezillon Cc: Thomas Zimmermann Cc: Gerd Hoffmann Cc: Rob Herring Cc: dri-devel@lists.freedesktop.org Cc: linux-media@vger.kernel.org Cc: linaro-mm-sig@lists.linaro.org Cc: # v5.9+ Reported-and-tested-by: piotr.oniszczuk@gmail.com Cc: piotr.oniszczuk@gmail.com Signed-off-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20201027214922.3566743-1-daniel.vetter@ffwll.ch --- drivers/gpu/drm/drm_gem.c | 4 ++-- drivers/gpu/drm/drm_gem_shmem_helper.c | 7 ++++++- 2 files changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index 19d73868490e..69c2c079d803 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -1085,6 +1085,8 @@ int drm_gem_mmap_obj(struct drm_gem_object *obj, unsigned long obj_size, */ drm_gem_object_get(obj); + vma->vm_private_data = obj; + if (obj->funcs && obj->funcs->mmap) { ret = obj->funcs->mmap(obj, vma); if (ret) { @@ -1107,8 +1109,6 @@ int drm_gem_mmap_obj(struct drm_gem_object *obj, unsigned long obj_size, vma->vm_page_prot = pgprot_decrypted(vma->vm_page_prot); } - vma->vm_private_data = obj; - return 0; } EXPORT_SYMBOL(drm_gem_mmap_obj); diff --git a/drivers/gpu/drm/drm_gem_shmem_helper.c b/drivers/gpu/drm/drm_gem_shmem_helper.c index d77c9f8ff26c..e00616d94f26 100644 --- a/drivers/gpu/drm/drm_gem_shmem_helper.c +++ b/drivers/gpu/drm/drm_gem_shmem_helper.c @@ -593,8 +593,13 @@ int drm_gem_shmem_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma) /* Remove the fake offset */ vma->vm_pgoff -= drm_vma_node_start(&obj->vma_node); - if (obj->import_attach) + if (obj->import_attach) { + /* Drop the reference drm_gem_mmap_obj() acquired.*/ + drm_gem_object_put(obj); + vma->vm_private_data = NULL; + return dma_buf_mmap(obj->dma_buf, vma, 0); + } shmem = to_drm_gem_shmem_obj(obj); -- cgit v1.2.3 From 063a3d67693ecde7f24ca00d4a131883edab1e73 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 23 Oct 2020 14:24:12 +0300 Subject: usb: typec: stusb160x: fix some signedness bugs These variables are enums but in this situation GCC will treat them as unsigned so the conditions are never true. Fixes: da0cb6310094 ("usb: typec: add support for STUSB160x Type-C controller family") Signed-off-by: Dan Carpenter Link: https://lore.kernel.org/r/20201023112412.GD282278@mwanda Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/stusb160x.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/usb/typec/stusb160x.c b/drivers/usb/typec/stusb160x.c index ce0bd7b3ad88..7143852ce94d 100644 --- a/drivers/usb/typec/stusb160x.c +++ b/drivers/usb/typec/stusb160x.c @@ -545,7 +545,7 @@ static int stusb160x_get_fw_caps(struct stusb160x *chip, ret = fwnode_property_read_string(fwnode, "power-role", &cap_str); if (!ret) { chip->port_type = typec_find_port_power_role(cap_str); - if (chip->port_type < 0) { + if ((int)chip->port_type < 0) { ret = chip->port_type; return ret; } @@ -567,9 +567,10 @@ static int stusb160x_get_fw_caps(struct stusb160x *chip, if (!ret) { chip->pwr_opmode = typec_find_pwr_opmode(cap_str); /* Power delivery not yet supported */ - if (chip->pwr_opmode < 0 || + if ((int)chip->pwr_opmode < 0 || chip->pwr_opmode == TYPEC_PWR_MODE_PD) { - ret = chip->pwr_opmode < 0 ? chip->pwr_opmode : -EINVAL; + ret = (int)chip->pwr_opmode < 0 ? chip->pwr_opmode : + -EINVAL; dev_err(chip->dev, "bad power operation mode: %d\n", chip->pwr_opmode); return ret; -- cgit v1.2.3 From 071ba4cc559de47160761b9500b72e8fa09d923d Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Mon, 26 Oct 2020 11:25:49 -0300 Subject: RDMA: Add rdma_connect_locked() There are two flows for handling RDMA_CM_EVENT_ROUTE_RESOLVED, either the handler triggers a completion and another thread does rdma_connect() or the handler directly calls rdma_connect(). In all cases rdma_connect() needs to hold the handler_mutex, but when handler's are invoked this is already held by the core code. This causes ULPs using the 2nd method to deadlock. Provide a rdma_connect_locked() and have all ULPs call it from their handlers. Link: https://lore.kernel.org/r/0-v2-53c22d5c1405+33-rdma_connect_locking_jgg@nvidia.com Reported-and-tested-by: Guoqing Jiang Fixes: 2a7cec538169 ("RDMA/cma: Fix locking for the RDMA_CM_CONNECT state") Acked-by: Santosh Shilimkar Acked-by: Jack Wang Reviewed-by: Christoph Hellwig Reviewed-by: Max Gurtovoy Reviewed-by: Sagi Grimberg Signed-off-by: Jason Gunthorpe --- drivers/infiniband/core/cma.c | 48 +++++++++++++++++++++++++------- drivers/infiniband/ulp/iser/iser_verbs.c | 2 +- drivers/infiniband/ulp/rtrs/rtrs-clt.c | 4 +-- drivers/nvme/host/rdma.c | 4 +-- include/rdma/rdma_cm.h | 14 ++-------- net/rds/ib_cm.c | 5 ++-- 6 files changed, 48 insertions(+), 29 deletions(-) diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 7c2ab1f2fbea..a77750b8954d 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -405,10 +405,10 @@ static int cma_comp_exch(struct rdma_id_private *id_priv, /* * The FSM uses a funny double locking where state is protected by both * the handler_mutex and the spinlock. State is not allowed to change - * away from a handler_mutex protected value without also holding + * to/from a handler_mutex protected value without also holding * handler_mutex. */ - if (comp == RDMA_CM_CONNECT) + if (comp == RDMA_CM_CONNECT || exch == RDMA_CM_CONNECT) lockdep_assert_held(&id_priv->handler_mutex); spin_lock_irqsave(&id_priv->lock, flags); @@ -4038,17 +4038,23 @@ out: return ret; } -int rdma_connect(struct rdma_cm_id *id, struct rdma_conn_param *conn_param) +/** + * rdma_connect_locked - Initiate an active connection request. + * @id: Connection identifier to connect. + * @conn_param: Connection information used for connected QPs. + * + * Same as rdma_connect() but can only be called from the + * RDMA_CM_EVENT_ROUTE_RESOLVED handler callback. + */ +int rdma_connect_locked(struct rdma_cm_id *id, + struct rdma_conn_param *conn_param) { struct rdma_id_private *id_priv = container_of(id, struct rdma_id_private, id); int ret; - mutex_lock(&id_priv->handler_mutex); - if (!cma_comp_exch(id_priv, RDMA_CM_ROUTE_RESOLVED, RDMA_CM_CONNECT)) { - ret = -EINVAL; - goto err_unlock; - } + if (!cma_comp_exch(id_priv, RDMA_CM_ROUTE_RESOLVED, RDMA_CM_CONNECT)) + return -EINVAL; if (!id->qp) { id_priv->qp_num = conn_param->qp_num; @@ -4066,11 +4072,33 @@ int rdma_connect(struct rdma_cm_id *id, struct rdma_conn_param *conn_param) ret = -ENOSYS; if (ret) goto err_state; - mutex_unlock(&id_priv->handler_mutex); return 0; err_state: cma_comp_exch(id_priv, RDMA_CM_CONNECT, RDMA_CM_ROUTE_RESOLVED); -err_unlock: + return ret; +} +EXPORT_SYMBOL(rdma_connect_locked); + +/** + * rdma_connect - Initiate an active connection request. + * @id: Connection identifier to connect. + * @conn_param: Connection information used for connected QPs. + * + * Users must have resolved a route for the rdma_cm_id to connect with by having + * called rdma_resolve_route before calling this routine. + * + * This call will either connect to a remote QP or obtain remote QP information + * for unconnected rdma_cm_id's. The actual operation is based on the + * rdma_cm_id's port space. + */ +int rdma_connect(struct rdma_cm_id *id, struct rdma_conn_param *conn_param) +{ + struct rdma_id_private *id_priv = + container_of(id, struct rdma_id_private, id); + int ret; + + mutex_lock(&id_priv->handler_mutex); + ret = rdma_connect_locked(id, conn_param); mutex_unlock(&id_priv->handler_mutex); return ret; } diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c index 2f3ebc0a75d9..2bd18b006893 100644 --- a/drivers/infiniband/ulp/iser/iser_verbs.c +++ b/drivers/infiniband/ulp/iser/iser_verbs.c @@ -620,7 +620,7 @@ static void iser_route_handler(struct rdma_cm_id *cma_id) conn_param.private_data = (void *)&req_hdr; conn_param.private_data_len = sizeof(struct iser_cm_hdr); - ret = rdma_connect(cma_id, &conn_param); + ret = rdma_connect_locked(cma_id, &conn_param); if (ret) { iser_err("failure connecting: %d\n", ret); goto failure; diff --git a/drivers/infiniband/ulp/rtrs/rtrs-clt.c b/drivers/infiniband/ulp/rtrs/rtrs-clt.c index 776e89231c52..f298adc02acb 100644 --- a/drivers/infiniband/ulp/rtrs/rtrs-clt.c +++ b/drivers/infiniband/ulp/rtrs/rtrs-clt.c @@ -1674,9 +1674,9 @@ static int rtrs_rdma_route_resolved(struct rtrs_clt_con *con) uuid_copy(&msg.sess_uuid, &sess->s.uuid); uuid_copy(&msg.paths_uuid, &clt->paths_uuid); - err = rdma_connect(con->c.cm_id, ¶m); + err = rdma_connect_locked(con->c.cm_id, ¶m); if (err) - rtrs_err(clt, "rdma_connect(): %d\n", err); + rtrs_err(clt, "rdma_connect_locked(): %d\n", err); return err; } diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index aad829a2b50d..8bbc48cc45dc 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -1890,10 +1890,10 @@ static int nvme_rdma_route_resolved(struct nvme_rdma_queue *queue) priv.hsqsize = cpu_to_le16(queue->ctrl->ctrl.sqsize); } - ret = rdma_connect(queue->cm_id, ¶m); + ret = rdma_connect_locked(queue->cm_id, ¶m); if (ret) { dev_err(ctrl->ctrl.device, - "rdma_connect failed (%d).\n", ret); + "rdma_connect_locked failed (%d).\n", ret); goto out_destroy_queue_ib; } diff --git a/include/rdma/rdma_cm.h b/include/rdma/rdma_cm.h index c672ae1da26b..32a67af18415 100644 --- a/include/rdma/rdma_cm.h +++ b/include/rdma/rdma_cm.h @@ -227,19 +227,9 @@ void rdma_destroy_qp(struct rdma_cm_id *id); int rdma_init_qp_attr(struct rdma_cm_id *id, struct ib_qp_attr *qp_attr, int *qp_attr_mask); -/** - * rdma_connect - Initiate an active connection request. - * @id: Connection identifier to connect. - * @conn_param: Connection information used for connected QPs. - * - * Users must have resolved a route for the rdma_cm_id to connect with - * by having called rdma_resolve_route before calling this routine. - * - * This call will either connect to a remote QP or obtain remote QP - * information for unconnected rdma_cm_id's. The actual operation is - * based on the rdma_cm_id's port space. - */ int rdma_connect(struct rdma_cm_id *id, struct rdma_conn_param *conn_param); +int rdma_connect_locked(struct rdma_cm_id *id, + struct rdma_conn_param *conn_param); int rdma_connect_ece(struct rdma_cm_id *id, struct rdma_conn_param *conn_param, struct rdma_ucm_ece *ece); diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c index 06603dd1c8aa..b36b60668b1d 100644 --- a/net/rds/ib_cm.c +++ b/net/rds/ib_cm.c @@ -956,9 +956,10 @@ int rds_ib_cm_initiate_connect(struct rdma_cm_id *cm_id, bool isv6) rds_ib_cm_fill_conn_param(conn, &conn_param, &dp, conn->c_proposed_version, UINT_MAX, UINT_MAX, isv6); - ret = rdma_connect(cm_id, &conn_param); + ret = rdma_connect_locked(cm_id, &conn_param); if (ret) - rds_ib_conn_error(conn, "rdma_connect failed (%d)\n", ret); + rds_ib_conn_error(conn, "rdma_connect_locked failed (%d)\n", + ret); out: /* Beware - returning non-zero tells the rdma_cm to destroy -- cgit v1.2.3 From cbdc0f54560f94c2205ddbebb5464d65868af0d8 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 23 Oct 2020 18:33:18 +0200 Subject: usb: fix kernel-doc markups There is a common comment marked, instead, with kernel-doc notation. Also, some identifiers have different names between their prototypes and the kernel-doc markup. Signed-off-by: Mauro Carvalho Chehab Acked-by: Felipe Balbi Link: https://lore.kernel.org/r/0b964be3884def04fcd20ea5c12cb90d0014871c.1603469755.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- drivers/usb/dwc3/core.c | 2 +- drivers/usb/dwc3/core.h | 2 +- drivers/usb/gadget/composite.c | 2 +- drivers/usb/typec/mux.c | 2 +- include/linux/usb/composite.h | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c index bdf0925da6b6..841daec70b6e 100644 --- a/drivers/usb/dwc3/core.c +++ b/drivers/usb/dwc3/core.c @@ -1,5 +1,5 @@ // SPDX-License-Identifier: GPL-2.0 -/** +/* * core.c - DesignWare USB3 DRD Controller Core file * * Copyright (C) 2010-2011 Texas Instruments Incorporated - https://www.ti.com diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h index 74323b10a64a..2f95f08ca511 100644 --- a/drivers/usb/dwc3/core.h +++ b/drivers/usb/dwc3/core.h @@ -1277,7 +1277,7 @@ struct dwc3_event_type { #define DWC3_DEPEVT_EPCMDCMPLT 0x07 /** - * struct dwc3_event_depvt - Device Endpoint Events + * struct dwc3_event_depevt - Device Endpoint Events * @one_bit: indicates this is an endpoint event (not used) * @endpoint_number: number of the endpoint * @endpoint_event: The event we have: diff --git a/drivers/usb/gadget/composite.c b/drivers/usb/gadget/composite.c index 05b176c82cc5..c6d455f2bb92 100644 --- a/drivers/usb/gadget/composite.c +++ b/drivers/usb/gadget/composite.c @@ -1245,7 +1245,7 @@ int usb_string_id(struct usb_composite_dev *cdev) EXPORT_SYMBOL_GPL(usb_string_id); /** - * usb_string_ids() - allocate unused string IDs in batch + * usb_string_ids_tab() - allocate unused string IDs in batch * @cdev: the device whose string descriptor IDs are being allocated * @str: an array of usb_string objects to assign numbers to * Context: single threaded during gadget setup diff --git a/drivers/usb/typec/mux.c b/drivers/usb/typec/mux.c index b069a5122aaa..cf720e944aaa 100644 --- a/drivers/usb/typec/mux.c +++ b/drivers/usb/typec/mux.c @@ -71,7 +71,7 @@ struct typec_switch *fwnode_typec_switch_get(struct fwnode_handle *fwnode) EXPORT_SYMBOL_GPL(fwnode_typec_switch_get); /** - * typec_put_switch - Release USB Type-C orientation switch + * typec_switch_put - Release USB Type-C orientation switch * @sw: USB Type-C orientation switch * * Decrement reference count for @sw. diff --git a/include/linux/usb/composite.h b/include/linux/usb/composite.h index 2040696d75b6..a2d229ab63ba 100644 --- a/include/linux/usb/composite.h +++ b/include/linux/usb/composite.h @@ -437,7 +437,7 @@ static inline struct usb_composite_driver *to_cdriver( #define OS_STRING_IDX 0xEE /** - * struct usb_composite_device - represents one composite usb gadget + * struct usb_composite_dev - represents one composite usb gadget * @gadget: read-only, abstracts the gadget's usb peripheral controller * @req: used for control responses; buffer is pre-allocated * @os_desc_req: used for OS descriptors responses; buffer is pre-allocated -- cgit v1.2.3 From 3cd54a618834430a26a648d880dd83d740f2ae30 Mon Sep 17 00:00:00 2001 From: Ran Wang Date: Sat, 10 Oct 2020 14:03:08 +0800 Subject: usb: host: fsl-mph-dr-of: check return of dma_set_mask() fsl_usb2_device_register() should stop init if dma_set_mask() return error. Fixes: cae058610465 ("drivers/usb/host: fsl: Set DMA_MASK of usb platform device") Reviewed-by: Peter Chen Signed-off-by: Ran Wang Link: https://lore.kernel.org/r/20201010060308.33693-1-ran.wang_1@nxp.com Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/fsl-mph-dr-of.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/usb/host/fsl-mph-dr-of.c b/drivers/usb/host/fsl-mph-dr-of.c index ae8f60f6e6a5..44a7e58a26e3 100644 --- a/drivers/usb/host/fsl-mph-dr-of.c +++ b/drivers/usb/host/fsl-mph-dr-of.c @@ -94,10 +94,13 @@ static struct platform_device *fsl_usb2_device_register( pdev->dev.coherent_dma_mask = ofdev->dev.coherent_dma_mask; - if (!pdev->dev.dma_mask) + if (!pdev->dev.dma_mask) { pdev->dev.dma_mask = &ofdev->dev.coherent_dma_mask; - else - dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); + } else { + retval = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); + if (retval) + goto error; + } retval = platform_device_add_data(pdev, pdata, sizeof(*pdata)); if (retval) -- cgit v1.2.3 From 38203b8385bf6283537162bde7d499f830964711 Mon Sep 17 00:00:00 2001 From: Jerome Brunet Date: Mon, 19 Oct 2020 19:07:02 +0200 Subject: usb: cdc-acm: fix cooldown mechanism Commit a4e7279cd1d1 ("cdc-acm: introduce a cool down") is causing regression if there is some USB error, such as -EPROTO. This has been reported on some samples of the Odroid-N2 using the Combee II Zibgee USB dongle. > struct acm *acm = container_of(work, struct acm, work) is incorrect in case of a delayed work and causes warnings, usually from the workqueue: > WARNING: CPU: 0 PID: 0 at kernel/workqueue.c:1474 __queue_work+0x480/0x528. When this happens, USB eventually stops working completely after a while. Also the ACM_ERROR_DELAY bit is never set, so the cooldown mechanism previously introduced cannot be triggered and acm_submit_read_urb() is never called. This changes makes the cdc-acm driver use a single delayed work, fixing the pointer arithmetic in acm_softint() and set the ACM_ERROR_DELAY when the cooldown mechanism appear to be needed. Fixes: a4e7279cd1d1 ("cdc-acm: introduce a cool down") Cc: Oliver Neukum Reported-by: Pascal Vizeli Acked-by: Oliver Neukum Signed-off-by: Jerome Brunet Link: https://lore.kernel.org/r/20201019170702.150534-1-jbrunet@baylibre.com Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/class/cdc-acm.c | 12 +++++------- drivers/usb/class/cdc-acm.h | 3 +-- 2 files changed, 6 insertions(+), 9 deletions(-) diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index 30ef946a8e1a..1e7568867910 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -508,6 +508,7 @@ static void acm_read_bulk_callback(struct urb *urb) "%s - cooling babbling device\n", __func__); usb_mark_last_busy(acm->dev); set_bit(rb->index, &acm->urbs_in_error_delay); + set_bit(ACM_ERROR_DELAY, &acm->flags); cooldown = true; break; default: @@ -533,7 +534,7 @@ static void acm_read_bulk_callback(struct urb *urb) if (stopped || stalled || cooldown) { if (stalled) - schedule_work(&acm->work); + schedule_delayed_work(&acm->dwork, 0); else if (cooldown) schedule_delayed_work(&acm->dwork, HZ / 2); return; @@ -563,13 +564,13 @@ static void acm_write_bulk(struct urb *urb) acm_write_done(acm, wb); spin_unlock_irqrestore(&acm->write_lock, flags); set_bit(EVENT_TTY_WAKEUP, &acm->flags); - schedule_work(&acm->work); + schedule_delayed_work(&acm->dwork, 0); } static void acm_softint(struct work_struct *work) { int i; - struct acm *acm = container_of(work, struct acm, work); + struct acm *acm = container_of(work, struct acm, dwork.work); if (test_bit(EVENT_RX_STALL, &acm->flags)) { smp_mb(); /* against acm_suspend() */ @@ -585,7 +586,7 @@ static void acm_softint(struct work_struct *work) if (test_and_clear_bit(ACM_ERROR_DELAY, &acm->flags)) { for (i = 0; i < acm->rx_buflimit; i++) if (test_and_clear_bit(i, &acm->urbs_in_error_delay)) - acm_submit_read_urb(acm, i, GFP_NOIO); + acm_submit_read_urb(acm, i, GFP_KERNEL); } if (test_and_clear_bit(EVENT_TTY_WAKEUP, &acm->flags)) @@ -1351,7 +1352,6 @@ made_compressed_probe: acm->ctrlsize = ctrlsize; acm->readsize = readsize; acm->rx_buflimit = num_rx_buf; - INIT_WORK(&acm->work, acm_softint); INIT_DELAYED_WORK(&acm->dwork, acm_softint); init_waitqueue_head(&acm->wioctl); spin_lock_init(&acm->write_lock); @@ -1561,7 +1561,6 @@ static void acm_disconnect(struct usb_interface *intf) } acm_kill_urbs(acm); - cancel_work_sync(&acm->work); cancel_delayed_work_sync(&acm->dwork); tty_unregister_device(acm_tty_driver, acm->minor); @@ -1604,7 +1603,6 @@ static int acm_suspend(struct usb_interface *intf, pm_message_t message) return 0; acm_kill_urbs(acm); - cancel_work_sync(&acm->work); cancel_delayed_work_sync(&acm->dwork); acm->urbs_in_error_delay = 0; diff --git a/drivers/usb/class/cdc-acm.h b/drivers/usb/class/cdc-acm.h index 9dce179d031b..8aef5eb769a0 100644 --- a/drivers/usb/class/cdc-acm.h +++ b/drivers/usb/class/cdc-acm.h @@ -112,8 +112,7 @@ struct acm { # define ACM_ERROR_DELAY 3 unsigned long urbs_in_error_delay; /* these need to be restarted after a delay */ struct usb_cdc_line_coding line; /* bits, stop, parity */ - struct work_struct work; /* work queue entry for various purposes*/ - struct delayed_work dwork; /* for cool downs needed in error recovery */ + struct delayed_work dwork; /* work queue entry for various purposes */ unsigned int ctrlin; /* input control lines (DCD, DSR, RI, break, overruns) */ unsigned int ctrlout; /* output control lines (DTR, RTS) */ struct async_icount iocount; /* counters for control line changes */ -- cgit v1.2.3 From 2d9c6442a9c81f4f8dee678d0b3c183173ab1e2d Mon Sep 17 00:00:00 2001 From: Li Jun Date: Mon, 12 Oct 2020 19:03:12 +0800 Subject: usb: typec: tcpm: reset hard_reset_count for any disconnect MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Current tcpm_detach() only reset hard_reset_count if port->attached is true, this may cause this counter clear is missed if the CC disconnect event is generated after tcpm_port_reset() is done by other events, e.g. VBUS off comes first before CC disconect for a power sink, in that case the first tcpm_detach() will only clear port->attached flag but leave hard_reset_count there because tcpm_port_is_disconnected() is still false, then later tcpm_detach() by CC disconnect will directly return due to port->attached is cleared, finally this will result tcpm will not try hard reset or error recovery for later attach. ChiYuan reported this issue on his platform with below tcpm trace: After power sink session setup after hard reset 2 times, detach from the power source and then attach: [ 4848.046358] VBUS off [ 4848.046384] state change SNK_READY -> SNK_UNATTACHED [ 4848.050908] Setting voltage/current limit 0 mV 0 mA [ 4848.050936] polarity 0 [ 4848.052593] Requesting mux state 0, usb-role 0, orientation 0 [ 4848.053222] Start toggling [ 4848.086500] state change SNK_UNATTACHED -> TOGGLING [ 4848.089983] CC1: 0 -> 0, CC2: 3 -> 3 [state TOGGLING, polarity 0, connected] [ 4848.089993] state change TOGGLING -> SNK_ATTACH_WAIT [ 4848.090031] pending state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED @200 ms [ 4848.141162] CC1: 0 -> 0, CC2: 3 -> 0 [state SNK_ATTACH_WAIT, polarity 0, disconnected] [ 4848.141170] state change SNK_ATTACH_WAIT -> SNK_ATTACH_WAIT [ 4848.141184] pending state change SNK_ATTACH_WAIT -> SNK_UNATTACHED @20 ms [ 4848.163156] state change SNK_ATTACH_WAIT -> SNK_UNATTACHED [delayed 20 ms] [ 4848.163162] Start toggling [ 4848.216918] CC1: 0 -> 0, CC2: 0 -> 3 [state TOGGLING, polarity 0, connected] [ 4848.216954] state change TOGGLING -> SNK_ATTACH_WAIT [ 4848.217080] pending state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED @200 ms [ 4848.231771] CC1: 0 -> 0, CC2: 3 -> 0 [state SNK_ATTACH_WAIT, polarity 0, disconnected] [ 4848.231800] state change SNK_ATTACH_WAIT -> SNK_ATTACH_WAIT [ 4848.231857] pending state change SNK_ATTACH_WAIT -> SNK_UNATTACHED @20 ms [ 4848.256022] state change SNK_ATTACH_WAIT -> SNK_UNATTACHED [delayed20 ms] [ 4848.256049] Start toggling [ 4848.871148] VBUS on [ 4848.885324] CC1: 0 -> 0, CC2: 0 -> 3 [state TOGGLING, polarity 0, connected] [ 4848.885372] state change TOGGLING -> SNK_ATTACH_WAIT [ 4848.885548] pending state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED @200 ms [ 4849.088240] state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED [delayed200 ms] [ 4849.088284] state change SNK_DEBOUNCED -> SNK_ATTACHED [ 4849.088291] polarity 1 [ 4849.088769] Requesting mux state 1, usb-role 2, orientation 2 [ 4849.088895] state change SNK_ATTACHED -> SNK_STARTUP [ 4849.088907] state change SNK_STARTUP -> SNK_DISCOVERY [ 4849.088915] Setting voltage/current limit 5000 mV 0 mA [ 4849.088927] vbus=0 charge:=1 [ 4849.090505] state change SNK_DISCOVERY -> SNK_WAIT_CAPABILITIES [ 4849.090828] pending state change SNK_WAIT_CAPABILITIES -> SNK_READY @240 ms [ 4849.335878] state change SNK_WAIT_CAPABILITIES -> SNK_READY [delayed240 ms] this patch fix this issue by clear hard_reset_count at any cases of cc disconnect, í.e. don't check port->attached flag. Fixes: 4b4e02c83167 ("typec: tcpm: Move out of staging") Cc: stable@vger.kernel.org Reported-and-tested-by: ChiYuan Huang Reviewed-by: Guenter Roeck Reviewed-by: Heikki Krogerus Signed-off-by: Li Jun Link: https://lore.kernel.org/r/1602500592-3817-1-git-send-email-jun.li@nxp.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/tcpm/tcpm.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/usb/typec/tcpm/tcpm.c b/drivers/usb/typec/tcpm/tcpm.c index 55535c4f66bf..a6fae1f86505 100644 --- a/drivers/usb/typec/tcpm/tcpm.c +++ b/drivers/usb/typec/tcpm/tcpm.c @@ -2890,6 +2890,9 @@ static void tcpm_reset_port(struct tcpm_port *port) static void tcpm_detach(struct tcpm_port *port) { + if (tcpm_port_is_disconnected(port)) + port->hard_reset_count = 0; + if (!port->attached) return; @@ -2898,9 +2901,6 @@ static void tcpm_detach(struct tcpm_port *port) port->tcpc->set_bist_data(port->tcpc, false); } - if (tcpm_port_is_disconnected(port)) - port->hard_reset_count = 0; - tcpm_reset_port(port); } -- cgit v1.2.3 From b9c6f782150c9992308f4474f920586eaaca4cf1 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 23 Oct 2020 14:40:17 +0300 Subject: usb: typec: stusb160x: fix an IS_ERR() vs NULL check in probe The typec_register_port() function doesn't return NULL, it returns error pointers. Fixes: da0cb6310094 ("usb: typec: add support for STUSB160x Type-C controller family") Signed-off-by: Dan Carpenter Reviewed-by: Heikki Krogerus Link: https://lore.kernel.org/r/20201023114017.GE18329@kadam Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/stusb160x.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/usb/typec/stusb160x.c b/drivers/usb/typec/stusb160x.c index 7143852ce94d..da7f1957bcb3 100644 --- a/drivers/usb/typec/stusb160x.c +++ b/drivers/usb/typec/stusb160x.c @@ -730,8 +730,8 @@ static int stusb160x_probe(struct i2c_client *client) } chip->port = typec_register_port(chip->dev, &chip->capability); - if (!chip->port) { - ret = -ENODEV; + if (IS_ERR(chip->port)) { + ret = PTR_ERR(chip->port); goto all_reg_disable; } -- cgit v1.2.3 From 32d174d2d5eb318c34ff36771adefabdf227c186 Mon Sep 17 00:00:00 2001 From: Tang Bin Date: Mon, 26 Oct 2020 17:06:57 +0800 Subject: usb: host: ehci-tegra: Fix error handling in tegra_ehci_probe() If the function platform_get_irq() failed, the negative value returned will not be detected here. So fix error handling in tegra_ehci_probe(). Fixes: 79ad3b5add4a ("usb: host: Add EHCI driver for NVIDIA Tegra SoCs") Acked-by: Alan Stern Acked-by: Thierry Reding Signed-off-by: Tang Bin Link: https://lore.kernel.org/r/20201026090657.49988-1-tangbin@cmss.chinamobile.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/ehci-tegra.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/usb/host/ehci-tegra.c b/drivers/usb/host/ehci-tegra.c index e077b2ca53c5..869d9c4de5fc 100644 --- a/drivers/usb/host/ehci-tegra.c +++ b/drivers/usb/host/ehci-tegra.c @@ -479,8 +479,8 @@ static int tegra_ehci_probe(struct platform_device *pdev) u_phy->otg->host = hcd_to_bus(hcd); irq = platform_get_irq(pdev, 0); - if (!irq) { - err = -ENODEV; + if (irq < 0) { + err = irq; goto cleanup_phy; } -- cgit v1.2.3 From 0942d59b0af46511d59dbf5bd69ec4a64d1a854c Mon Sep 17 00:00:00 2001 From: Bastien Nocera Date: Thu, 22 Oct 2020 09:55:20 -0400 Subject: usbcore: Check both id_table and match() when both available From: Bastien Nocera When a USB device driver has both an id_table and a match() function, make sure to check both to find a match, first matching the id_table, then checking the match() function. This makes it possible to have module autoloading done through the id_table when devices are plugged in, before checking for further device eligibility in the match() function. Cc: # 5.8 Cc: Greg Kroah-Hartman Cc: Alan Stern Co-developed-by: M. Vefa Bicakci Tested-by: Bastien Nocera Signed-off-by: Bastien Nocera Signed-off-by: M. Vefa Bicakci Tested-by: Pan (Pany) YUAN Link: https://lore.kernel.org/r/20201022135521.375211-2-m.v.b@runbox.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/driver.c | 30 +++++++++++++++++++++--------- drivers/usb/core/generic.c | 4 +--- drivers/usb/core/usb.h | 2 ++ 3 files changed, 24 insertions(+), 12 deletions(-) diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c index 98b7449c11f3..4dfa44d6cc3c 100644 --- a/drivers/usb/core/driver.c +++ b/drivers/usb/core/driver.c @@ -839,6 +839,22 @@ const struct usb_device_id *usb_device_match_id(struct usb_device *udev, return NULL; } +bool usb_driver_applicable(struct usb_device *udev, + struct usb_device_driver *udrv) +{ + if (udrv->id_table && udrv->match) + return usb_device_match_id(udev, udrv->id_table) != NULL && + udrv->match(udev); + + if (udrv->id_table) + return usb_device_match_id(udev, udrv->id_table) != NULL; + + if (udrv->match) + return udrv->match(udev); + + return false; +} + static int usb_device_match(struct device *dev, struct device_driver *drv) { /* devices and interfaces are handled separately */ @@ -853,17 +869,14 @@ static int usb_device_match(struct device *dev, struct device_driver *drv) udev = to_usb_device(dev); udrv = to_usb_device_driver(drv); - if (udrv->id_table) - return usb_device_match_id(udev, udrv->id_table) != NULL; - - if (udrv->match) - return udrv->match(udev); - /* If the device driver under consideration does not have a * id_table or a match function, then let the driver's probe * function decide. */ - return 1; + if (!udrv->id_table && !udrv->match) + return 1; + + return usb_driver_applicable(udev, udrv); } else if (is_usb_interface(dev)) { struct usb_interface *intf; @@ -941,8 +954,7 @@ static int __usb_bus_reprobe_drivers(struct device *dev, void *data) return 0; udev = to_usb_device(dev); - if (usb_device_match_id(udev, new_udriver->id_table) == NULL && - (!new_udriver->match || new_udriver->match(udev) == 0)) + if (!usb_driver_applicable(udev, new_udriver)) return 0; ret = device_reprobe(dev); diff --git a/drivers/usb/core/generic.c b/drivers/usb/core/generic.c index 22c887f5c497..26f9fb9f67ca 100644 --- a/drivers/usb/core/generic.c +++ b/drivers/usb/core/generic.c @@ -205,9 +205,7 @@ static int __check_for_non_generic_match(struct device_driver *drv, void *data) udrv = to_usb_device_driver(drv); if (udrv == &usb_generic_driver) return 0; - if (usb_device_match_id(udev, udrv->id_table) != NULL) - return 1; - return (udrv->match && udrv->match(udev)); + return usb_driver_applicable(udev, udrv); } static bool usb_generic_driver_match(struct usb_device *udev) diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h index c893f54a3420..82538daac8b8 100644 --- a/drivers/usb/core/usb.h +++ b/drivers/usb/core/usb.h @@ -74,6 +74,8 @@ extern int usb_match_device(struct usb_device *dev, const struct usb_device_id *id); extern const struct usb_device_id *usb_device_match_id(struct usb_device *udev, const struct usb_device_id *id); +extern bool usb_driver_applicable(struct usb_device *udev, + struct usb_device_driver *udrv); extern void usb_forced_unbind_intf(struct usb_interface *intf); extern void usb_unbind_and_rebind_marked_interfaces(struct usb_device *udev); -- cgit v1.2.3 From 0cb686692fd200db12dcfb8231e793c1c98aec41 Mon Sep 17 00:00:00 2001 From: Bastien Nocera Date: Thu, 22 Oct 2020 09:55:21 -0400 Subject: USB: apple-mfi-fastcharge: don't probe unhandled devices From: Bastien Nocera Contrary to the comment above the id table, we didn't implement a match function. This meant that every single Apple device that was already plugged in to the computer would have its device driver reprobed when the apple-mfi-fastcharge driver was loaded, eg. the SD card reader could be reprobed when the apple-mfi-fastcharge after pivoting root during boot up and the module became available. Make sure that the driver probe isn't being run for unsupported devices by adding a match function that checks the product ID, in addition to the id_table checking the vendor ID. Link: https://bugzilla.redhat.com/show_bug.cgi?id=1878347 Link: https://lore.kernel.org/linux-usb/CAE3RAxt0WhBEz8zkHrVO5RiyEOasayy1QUAjsv-pB0fAbY1GSw@mail.gmail.com/ Fixes: 249fa8217b84 ("USB: Add driver to control USB fast charge for iOS devices") Cc: # 5.8 Cc: Greg Kroah-Hartman Cc: Alan Stern [m.v.b: Add Link and Reported-by tags to the commit message] Reported-by: Pany Tested-by: Pan (Pany) YUAN Tested-by: Bastien Nocera Signed-off-by: Bastien Nocera Signed-off-by: M. Vefa Bicakci Link: https://lore.kernel.org/r/20201022135521.375211-3-m.v.b@runbox.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/misc/apple-mfi-fastcharge.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) diff --git a/drivers/usb/misc/apple-mfi-fastcharge.c b/drivers/usb/misc/apple-mfi-fastcharge.c index b403094a6b3a..579d8c84de42 100644 --- a/drivers/usb/misc/apple-mfi-fastcharge.c +++ b/drivers/usb/misc/apple-mfi-fastcharge.c @@ -163,17 +163,23 @@ static const struct power_supply_desc apple_mfi_fc_desc = { .property_is_writeable = apple_mfi_fc_property_is_writeable }; +static bool mfi_fc_match(struct usb_device *udev) +{ + int idProduct; + + idProduct = le16_to_cpu(udev->descriptor.idProduct); + /* See comment above mfi_fc_id_table[] */ + return (idProduct >= 0x1200 && idProduct <= 0x12ff); +} + static int mfi_fc_probe(struct usb_device *udev) { struct power_supply_config battery_cfg = {}; struct mfi_device *mfi = NULL; - int err, idProduct; + int err; - idProduct = le16_to_cpu(udev->descriptor.idProduct); - /* See comment above mfi_fc_id_table[] */ - if (idProduct < 0x1200 || idProduct > 0x12ff) { + if (!mfi_fc_match(udev)) return -ENODEV; - } mfi = kzalloc(sizeof(struct mfi_device), GFP_KERNEL); if (!mfi) { @@ -220,6 +226,7 @@ static struct usb_device_driver mfi_fc_driver = { .probe = mfi_fc_probe, .disconnect = mfi_fc_disconnect, .id_table = mfi_fc_id_table, + .match = mfi_fc_match, .generic_subclass = 1, }; -- cgit v1.2.3 From 82776f6c75a90e1d2103e689b84a689de8f1aa02 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 18 Oct 2020 09:42:04 +0100 Subject: tty: serial: 21285: fix lockup on open Commit 293f89959483 ("tty: serial: 21285: stop using the unused[] variable from struct uart_port") introduced a bug which stops the transmit interrupt being disabled when there are no characters to transmit - disabling the transmit interrupt at the interrupt controller is the only way to stop an interrupt storm. If this interrupt is not disabled when there are no transmit characters, we end up with an interrupt storm which prevents the machine making forward progress. Fixes: 293f89959483 ("tty: serial: 21285: stop using the unused[] variable from struct uart_port") Signed-off-by: Russell King Cc: stable Link: https://lore.kernel.org/r/E1kU4GS-0006lE-OO@rmk-PC.armlinux.org.uk Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/21285.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/tty/serial/21285.c b/drivers/tty/serial/21285.c index 718e010fcb04..09baef4ccc39 100644 --- a/drivers/tty/serial/21285.c +++ b/drivers/tty/serial/21285.c @@ -50,25 +50,25 @@ static const char serial21285_name[] = "Footbridge UART"; static bool is_enabled(struct uart_port *port, int bit) { - unsigned long private_data = (unsigned long)port->private_data; + unsigned long *private_data = (unsigned long *)&port->private_data; - if (test_bit(bit, &private_data)) + if (test_bit(bit, private_data)) return true; return false; } static void enable(struct uart_port *port, int bit) { - unsigned long private_data = (unsigned long)port->private_data; + unsigned long *private_data = (unsigned long *)&port->private_data; - set_bit(bit, &private_data); + set_bit(bit, private_data); } static void disable(struct uart_port *port, int bit) { - unsigned long private_data = (unsigned long)port->private_data; + unsigned long *private_data = (unsigned long *)&port->private_data; - clear_bit(bit, &private_data); + clear_bit(bit, private_data); } #define is_tx_enabled(port) is_enabled(port, tx_enabled_bit) -- cgit v1.2.3 From c97f2a6fb3dfbfbbc88edc8ea62ef2b944e18849 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 23 Oct 2020 04:34:29 +0300 Subject: tty: serial: fsl_lpuart: LS1021A has a FIFO size of 16 words, like LS1028A MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Prior to the commit that this one fixes, the FIFO size was derived from the read-only register LPUARTx_FIFO[TXFIFOSIZE] using the following formula: TX FIFO size = 2 ^ (LPUARTx_FIFO[TXFIFOSIZE] - 1) The documentation for LS1021A is a mess. Under chapter 26.1.3 LS1021A LPUART module special consideration, it mentions TXFIFO_SZ and RXFIFO_SZ being equal to 4, and in the register description for LPUARTx_FIFO, it shows the out-of-reset value of TXFIFOSIZE and RXFIFOSIZE fields as "011", even though these registers read as "101" in reality. And when LPUART on LS1021A was working, the "101" value did correspond to "16 datawords", by applying the formula above, even though the documentation is wrong again (!!!!) and says that "101" means 64 datawords (hint: it doesn't). So the "new" formula created by commit f77ebb241ce0 has all the premises of being wrong for LS1021A, because it relied only on false data and no actual experimentation. Interestingly, in commit c2f448cff22a ("tty: serial: fsl_lpuart: add LS1028A support"), Michael Walle applied a workaround to this by manually setting the FIFO widths for LS1028A. It looks like the same values are used by LS1021A as well, in fact. When the driver thinks that it has a deeper FIFO than it really has, getty (user space) output gets truncated. Many thanks to Michael for pointing out where to look. Fixes: f77ebb241ce0 ("tty: serial: fsl_lpuart: correct the FIFO depth size") Suggested-by: Michael Walle Signed-off-by: Vladimir Oltean Link: https://lore.kernel.org/r/20201023013429.3551026-1-vladimir.oltean@nxp.com Reviewed-by：Fugang Duan Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/fsl_lpuart.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/tty/serial/fsl_lpuart.c b/drivers/tty/serial/fsl_lpuart.c index ff4b88c637d0..bd047e1f9bea 100644 --- a/drivers/tty/serial/fsl_lpuart.c +++ b/drivers/tty/serial/fsl_lpuart.c @@ -314,9 +314,10 @@ MODULE_DEVICE_TABLE(of, lpuart_dt_ids); /* Forward declare this for the dma callbacks*/ static void lpuart_dma_tx_complete(void *arg); -static inline bool is_ls1028a_lpuart(struct lpuart_port *sport) +static inline bool is_layerscape_lpuart(struct lpuart_port *sport) { - return sport->devtype == LS1028A_LPUART; + return (sport->devtype == LS1021A_LPUART || + sport->devtype == LS1028A_LPUART); } static inline bool is_imx8qxp_lpuart(struct lpuart_port *sport) @@ -1701,11 +1702,11 @@ static int lpuart32_startup(struct uart_port *port) UARTFIFO_FIFOSIZE_MASK); /* - * The LS1028A has a fixed length of 16 words. Although it supports the - * RX/TXSIZE fields their encoding is different. Eg the reference manual - * states 0b101 is 16 words. + * The LS1021A and LS1028A have a fixed FIFO depth of 16 words. + * Although they support the RX/TXSIZE fields, their encoding is + * different. Eg the reference manual states 0b101 is 16 words. */ - if (is_ls1028a_lpuart(sport)) { + if (is_layerscape_lpuart(sport)) { sport->rxfifo_size = 16; sport->txfifo_size = 16; sport->port.fifosize = sport->txfifo_size; -- cgit v1.2.3 From 6ca03f90527e499dd5e32d6522909e2ad390896b Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Mon, 19 Oct 2020 10:55:16 +0200 Subject: vt: keyboard, simplify vt_kdgkbsent Use 'strlen' of the string, add one for NUL terminator and simply do 'copy_to_user' instead of the explicit 'for' loop. This makes the KDGKBSENT case more compact. The only thing we need to take care about is NULL 'func_table[i]'. Use an empty string in that case. The original check for overflow could never trigger as the func_buf strings are always shorter or equal to 'struct kbsentry's. Cc: Signed-off-by: Jiri Slaby Link: https://lore.kernel.org/r/20201019085517.10176-1-jslaby@suse.cz Signed-off-by: Greg Kroah-Hartman --- drivers/tty/vt/keyboard.c | 28 +++++++++------------------- 1 file changed, 9 insertions(+), 19 deletions(-) diff --git a/drivers/tty/vt/keyboard.c b/drivers/tty/vt/keyboard.c index 0db53b5b3acf..7bfa95c3252c 100644 --- a/drivers/tty/vt/keyboard.c +++ b/drivers/tty/vt/keyboard.c @@ -1995,9 +1995,7 @@ out: int vt_do_kdgkb_ioctl(int cmd, struct kbsentry __user *user_kdgkb, int perm) { struct kbsentry *kbs; - char *p; u_char *q; - u_char __user *up; int sz, fnw_sz; int delta; char *first_free, *fj, *fnw; @@ -2023,23 +2021,15 @@ int vt_do_kdgkb_ioctl(int cmd, struct kbsentry __user *user_kdgkb, int perm) i = array_index_nospec(kbs->kb_func, MAX_NR_FUNC); switch (cmd) { - case KDGKBSENT: - sz = sizeof(kbs->kb_string) - 1; /* sz should have been - a struct member */ - up = user_kdgkb->kb_string; - p = func_table[i]; - if(p) - for ( ; *p && sz; p++, sz--) - if (put_user(*p, up++)) { - ret = -EFAULT; - goto reterr; - } - if (put_user('\0', up)) { - ret = -EFAULT; - goto reterr; - } - kfree(kbs); - return ((p && *p) ? -EOVERFLOW : 0); + case KDGKBSENT: { + /* size should have been a struct member */ + unsigned char *from = func_table[i] ? : ""; + + ret = copy_to_user(user_kdgkb->kb_string, from, + strlen(from) + 1) ? -EFAULT : 0; + + goto reterr; + } case KDSKBSENT: if (!perm) { ret = -EPERM; -- cgit v1.2.3 From 82e61c3909db51d91b9d3e2071557b6435018b80 Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Mon, 19 Oct 2020 10:55:17 +0200 Subject: vt: keyboard, extend func_buf_lock to readers Both read-side users of func_table/func_buf need locking. Without that, one can easily confuse the code by repeatedly setting altering strings like: while (1) for (a = 0; a < 2; a++) { struct kbsentry kbs = {}; strcpy((char *)kbs.kb_string, a ? ".\n" : "88888\n"); ioctl(fd, KDSKBSENT, &kbs); } When that program runs, one can get unexpected output by holding F1 (note the unxpected period on the last line): . 88888 .8888 So protect all accesses to 'func_table' (and func_buf) by preexisting 'func_buf_lock'. It is easy in 'k_fn' handler as 'puts_queue' is expected not to sleep. On the other hand, KDGKBSENT needs a local (atomic) copy of the string because copy_to_user can sleep. Use already allocated, but unused 'kbs->kb_string' for that purpose. Note that the program above needs at least CAP_SYS_TTY_CONFIG. This depends on the previous patch and on the func_buf_lock lock added in commit 46ca3f735f34 (tty/vt: fix write/write race in ioctl(KDSKBSENT) handler) in 5.2. Likely fixes CVE-2020-25656. Cc: Reported-by: Minh Yuan Signed-off-by: Jiri Slaby Link: https://lore.kernel.org/r/20201019085517.10176-2-jslaby@suse.cz Signed-off-by: Greg Kroah-Hartman --- drivers/tty/vt/keyboard.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/drivers/tty/vt/keyboard.c b/drivers/tty/vt/keyboard.c index 7bfa95c3252c..78acc270e39a 100644 --- a/drivers/tty/vt/keyboard.c +++ b/drivers/tty/vt/keyboard.c @@ -743,8 +743,13 @@ static void k_fn(struct vc_data *vc, unsigned char value, char up_flag) return; if ((unsigned)value < ARRAY_SIZE(func_table)) { + unsigned long flags; + + spin_lock_irqsave(&func_buf_lock, flags); if (func_table[value]) puts_queue(vc, func_table[value]); + spin_unlock_irqrestore(&func_buf_lock, flags); + } else pr_err("k_fn called with value=%d\n", value); } @@ -1991,7 +1996,7 @@ out: #undef s #undef v -/* FIXME: This one needs untangling and locking */ +/* FIXME: This one needs untangling */ int vt_do_kdgkb_ioctl(int cmd, struct kbsentry __user *user_kdgkb, int perm) { struct kbsentry *kbs; @@ -2023,10 +2028,14 @@ int vt_do_kdgkb_ioctl(int cmd, struct kbsentry __user *user_kdgkb, int perm) switch (cmd) { case KDGKBSENT: { /* size should have been a struct member */ - unsigned char *from = func_table[i] ? : ""; + ssize_t len = sizeof(user_kdgkb->kb_string); + + spin_lock_irqsave(&func_buf_lock, flags); + len = strlcpy(kbs->kb_string, func_table[i] ? : "", len); + spin_unlock_irqrestore(&func_buf_lock, flags); - ret = copy_to_user(user_kdgkb->kb_string, from, - strlen(from) + 1) ? -EFAULT : 0; + ret = copy_to_user(user_kdgkb->kb_string, kbs->kb_string, + len + 1) ? -EFAULT : 0; goto reterr; } -- cgit v1.2.3 From d54654790302ccaa72589380dce060d376ef8716 Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Mon, 26 Oct 2020 06:54:19 +0100 Subject: vt_ioctl: fix GIO_UNIMAP regression In commit 5ba127878722, we shuffled with the check of 'perm'. But my brain somehow inverted the condition in 'do_unimap_ioctl' (I thought it is ||, not &&), so GIO_UNIMAP stopped working completely. Move the 'perm' checks back to do_unimap_ioctl and do them right again. In fact, this reverts this part of code to the pre-5ba127878722 state. Except 'perm' is now a bool. Fixes: 5ba127878722 ("vt_ioctl: move perm checks level up") Cc: stable@vger.kernel.org Reported-by: Fabian Vogt Signed-off-by: Jiri Slaby Link: https://lore.kernel.org/r/20201026055419.30518-1-jslaby@suse.cz Signed-off-by: Greg Kroah-Hartman --- drivers/tty/vt/vt_ioctl.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/drivers/tty/vt/vt_ioctl.c b/drivers/tty/vt/vt_ioctl.c index 0a33b8ababe3..2321775ef098 100644 --- a/drivers/tty/vt/vt_ioctl.c +++ b/drivers/tty/vt/vt_ioctl.c @@ -549,7 +549,7 @@ static int vt_io_fontreset(struct console_font_op *op) } static inline int do_unimap_ioctl(int cmd, struct unimapdesc __user *user_ud, - struct vc_data *vc) + bool perm, struct vc_data *vc) { struct unimapdesc tmp; @@ -557,9 +557,11 @@ static inline int do_unimap_ioctl(int cmd, struct unimapdesc __user *user_ud, return -EFAULT; switch (cmd) { case PIO_UNIMAP: + if (!perm) + return -EPERM; return con_set_unimap(vc, tmp.entry_ct, tmp.entries); case GIO_UNIMAP: - if (fg_console != vc->vc_num) + if (!perm && fg_console != vc->vc_num) return -EPERM; return con_get_unimap(vc, tmp.entry_ct, &(user_ud->entry_ct), tmp.entries); @@ -639,10 +641,7 @@ static int vt_io_ioctl(struct vc_data *vc, unsigned int cmd, void __user *up, case PIO_UNIMAP: case GIO_UNIMAP: - if (!perm) - return -EPERM; - - return do_unimap_ioctl(cmd, up, vc); + return do_unimap_ioctl(cmd, up, perm, vc); default: return -ENOIOCTLCMD; -- cgit v1.2.3 From a2267f8a52eea9096861affd463f691be0f0e8c9 Mon Sep 17 00:00:00 2001 From: Alok Prasad Date: Wed, 21 Oct 2020 11:50:08 +0000 Subject: RDMA/qedr: Fix memory leak in iWARP CM Fixes memory leak in iWARP CM Fixes: e411e0587e0d ("RDMA/qedr: Add iWARP connection management functions") Link: https://lore.kernel.org/r/20201021115008.28138-1-palok@marvell.com Signed-off-by: Michal Kalderon Signed-off-by: Igor Russkikh Signed-off-by: Alok Prasad Signed-off-by: Jason Gunthorpe --- drivers/infiniband/hw/qedr/qedr_iw_cm.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/hw/qedr/qedr_iw_cm.c b/drivers/infiniband/hw/qedr/qedr_iw_cm.c index c7169d2c69e5..c4bc58736e48 100644 --- a/drivers/infiniband/hw/qedr/qedr_iw_cm.c +++ b/drivers/infiniband/hw/qedr/qedr_iw_cm.c @@ -727,6 +727,7 @@ int qedr_iw_destroy_listen(struct iw_cm_id *cm_id) listener->qed_handle); cm_id->rem_ref(cm_id); + kfree(listener); return rc; } -- cgit v1.2.3 From 136a4dfe87cd34db2153330aa84dbbc210eb36b6 Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Mon, 26 Oct 2020 22:28:40 +0100 Subject: ACPI: button: Drop no longer necessary Acer SW5-012 lid_init_state quirk Commit 78a5b53e9fb4 ("Input: soc_button_array - work around DSDTs which modify the irqflags") adds a workaround for DSDTs with a _LID method which play tricks with the irqflags, assuming that the OS is using an irq-type of IRQ_TYPE_LEVEL_LOW. Now that this workaround is in place, we no longer need to disable the lid functionality on the Acer SW5-012. Fixes: 78a5b53e9fb4 ("Input: soc_button_array - work around DSDTs which modify the irqflags") Signed-off-by: Hans de Goede Signed-off-by: Rafael J. Wysocki --- drivers/acpi/button.c | 13 ------------- 1 file changed, 13 deletions(-) diff --git a/drivers/acpi/button.c b/drivers/acpi/button.c index da4b125ab4c3..0761529cac05 100644 --- a/drivers/acpi/button.c +++ b/drivers/acpi/button.c @@ -73,19 +73,6 @@ MODULE_DEVICE_TABLE(acpi, button_device_ids); /* Please keep this list sorted alphabetically by vendor and model */ static const struct dmi_system_id dmi_lid_quirks[] = { - { - /* - * Acer Switch 10 SW5-012. _LID method messes with home and - * power button GPIO IRQ settings causing an interrupt storm on - * both GPIOs. This is unfixable without a DSDT override, so we - * have to disable the lid-switch functionality altogether :| - */ - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "Acer"), - DMI_MATCH(DMI_PRODUCT_NAME, "Aspire SW5-012"), - }, - .driver_data = (void *)(long)ACPI_BUTTON_LID_INIT_DISABLED, - }, { /* GP-electronic T701, _LID method points to a floating GPIO */ .matches = { -- cgit v1.2.3 From 332576e69a046b9b33f49bd0c3c071cc98215b60 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 26 Oct 2020 20:37:46 +0100 Subject: arm64: avoid -Woverride-init warning The icache_policy_str[] definition causes a warning when extra warning flags are enabled: arch/arm64/kernel/cpuinfo.c:38:26: warning: initialized field overwritten [-Woverride-init] 38 | [ICACHE_POLICY_VIPT] = "VIPT", | ^~~~~~ arch/arm64/kernel/cpuinfo.c:38:26: note: (near initialization for 'icache_policy_str[2]') arch/arm64/kernel/cpuinfo.c:39:26: warning: initialized field overwritten [-Woverride-init] 39 | [ICACHE_POLICY_PIPT] = "PIPT", | ^~~~~~ arch/arm64/kernel/cpuinfo.c:39:26: note: (near initialization for 'icache_policy_str[3]') arch/arm64/kernel/cpuinfo.c:40:27: warning: initialized field overwritten [-Woverride-init] 40 | [ICACHE_POLICY_VPIPT] = "VPIPT", | ^~~~~~~ arch/arm64/kernel/cpuinfo.c:40:27: note: (near initialization for 'icache_policy_str[0]') There is no real need for the default initializer here, as printing a NULL string is harmless. Rewrite the logic to have an explicit reserved value for the only one that uses the default value. This partially reverts the commit that removed ICACHE_POLICY_AIVIVT. Fixes: 155433cb365e ("arm64: cache: Remove support for ASID-tagged VIVT I-caches") Signed-off-by: Arnd Bergmann Link: https://lore.kernel.org/r/20201026193807.3816388-1-arnd@kernel.org Signed-off-by: Will Deacon --- arch/arm64/include/asm/cache.h | 1 + arch/arm64/kernel/cpuinfo.c | 7 ++++--- 2 files changed, 5 insertions(+), 3 deletions(-) diff --git a/arch/arm64/include/asm/cache.h b/arch/arm64/include/asm/cache.h index 0ac3e06a2118..63d43b5f82f6 100644 --- a/arch/arm64/include/asm/cache.h +++ b/arch/arm64/include/asm/cache.h @@ -24,6 +24,7 @@ #define CTR_L1IP(ctr) (((ctr) >> CTR_L1IP_SHIFT) & CTR_L1IP_MASK) #define ICACHE_POLICY_VPIPT 0 +#define ICACHE_POLICY_RESERVED 1 #define ICACHE_POLICY_VIPT 2 #define ICACHE_POLICY_PIPT 3 diff --git a/arch/arm64/kernel/cpuinfo.c b/arch/arm64/kernel/cpuinfo.c index 6a7bb3729d60..77605aec25fe 100644 --- a/arch/arm64/kernel/cpuinfo.c +++ b/arch/arm64/kernel/cpuinfo.c @@ -34,10 +34,10 @@ DEFINE_PER_CPU(struct cpuinfo_arm64, cpu_data); static struct cpuinfo_arm64 boot_cpu_data; static const char *icache_policy_str[] = { - [0 ... ICACHE_POLICY_PIPT] = "RESERVED/UNKNOWN", + [ICACHE_POLICY_VPIPT] = "VPIPT", + [ICACHE_POLICY_RESERVED] = "RESERVED/UNKNOWN", [ICACHE_POLICY_VIPT] = "VIPT", [ICACHE_POLICY_PIPT] = "PIPT", - [ICACHE_POLICY_VPIPT] = "VPIPT", }; unsigned long __icache_flags; @@ -334,10 +334,11 @@ static void cpuinfo_detect_icache_policy(struct cpuinfo_arm64 *info) case ICACHE_POLICY_VPIPT: set_bit(ICACHEF_VPIPT, &__icache_flags); break; - default: + case ICACHE_POLICY_RESERVED: case ICACHE_POLICY_VIPT: /* Assume aliasing */ set_bit(ICACHEF_ALIASING, &__icache_flags); + break; } pr_info("Detected %s I-cache on CPU%d\n", icache_policy_str[l1ip], cpu); -- cgit v1.2.3 From 5bc7c1156f3f16d7e244abaf8c80566af32cb092 Mon Sep 17 00:00:00 2001 From: Vincenzo Frascino Date: Mon, 26 Oct 2020 12:12:43 +0000 Subject: kselftest/arm64: Fix check_buffer_fill test The check_buffer_fill test reports the error below because the test plan is not declared correctly: # Planned tests != run tests (0 != 20) Fix the test adding the correct test plan declaration. Fixes: e9b60476bea0 ("kselftest/arm64: Add utilities and a test to validate mte memory") Signed-off-by: Vincenzo Frascino Cc: Shuah Khan Cc: Catalin Marinas Cc: Will Deacon Cc: Gabor Kertesz Cc: Amit Daniel Kachhap Link: https://lore.kernel.org/r/20201026121248.2340-2-vincenzo.frascino@arm.com Signed-off-by: Will Deacon --- tools/testing/selftests/arm64/mte/check_buffer_fill.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tools/testing/selftests/arm64/mte/check_buffer_fill.c b/tools/testing/selftests/arm64/mte/check_buffer_fill.c index 242635d79035..c9fa141ebdcc 100644 --- a/tools/testing/selftests/arm64/mte/check_buffer_fill.c +++ b/tools/testing/selftests/arm64/mte/check_buffer_fill.c @@ -417,6 +417,9 @@ int main(int argc, char *argv[]) /* Register SIGSEGV handler */ mte_register_signal(SIGSEGV, mte_default_handler); + /* Set test plan */ + ksft_set_plan(20); + /* Buffer by byte tests */ evaluate_test(check_buffer_by_byte(USE_MMAP, MTE_SYNC_ERR), "Check buffer correctness by byte with sync err mode and mmap memory\n"); -- cgit v1.2.3 From 041fa41f5422ee275a8f60a182e1b32d196a1a2a Mon Sep 17 00:00:00 2001 From: Vincenzo Frascino Date: Mon, 26 Oct 2020 12:12:44 +0000 Subject: kselftest/arm64: Fix check_tags_inclusion test The check_tags_inclusion test reports the error below because the test plan is not declared correctly: # Planned tests != run tests (0 != 4) Fix the test adding the correct test plan declaration. Fixes: f3b2a26ca78d ("kselftest/arm64: Verify mte tag inclusion via prctl") Signed-off-by: Vincenzo Frascino Cc: Shuah Khan Cc: Catalin Marinas Cc: Will Deacon Cc: Gabor Kertesz Cc: Amit Daniel Kachhap Link: https://lore.kernel.org/r/20201026121248.2340-3-vincenzo.frascino@arm.com Signed-off-by: Will Deacon --- tools/testing/selftests/arm64/mte/check_tags_inclusion.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tools/testing/selftests/arm64/mte/check_tags_inclusion.c b/tools/testing/selftests/arm64/mte/check_tags_inclusion.c index 94d245a0ed56..deaef1f61076 100644 --- a/tools/testing/selftests/arm64/mte/check_tags_inclusion.c +++ b/tools/testing/selftests/arm64/mte/check_tags_inclusion.c @@ -170,6 +170,9 @@ int main(int argc, char *argv[]) /* Register SIGSEGV handler */ mte_register_signal(SIGSEGV, mte_default_handler); + /* Set test plan */ + ksft_set_plan(4); + evaluate_test(check_single_included_tags(USE_MMAP, MTE_SYNC_ERR), "Check an included tag value with sync mode\n"); evaluate_test(check_multiple_included_tags(USE_MMAP, MTE_SYNC_ERR), -- cgit v1.2.3 From 386cf789fa6dc4e0f35d25cba9469d62321e7180 Mon Sep 17 00:00:00 2001 From: Vincenzo Frascino Date: Mon, 26 Oct 2020 12:12:45 +0000 Subject: kselftest/arm64: Fix check_child_memory test The check_child_memory test reports the error below because the test plan is not declared correctly: # Planned tests != run tests (0 != 12) Fix the test adding the correct test plan declaration. Fixes: dfe537cf4718 ("kselftest/arm64: Check forked child mte memory accessibility") Signed-off-by: Vincenzo Frascino Cc: Shuah Khan Cc: Catalin Marinas Cc: Will Deacon Cc: Gabor Kertesz Cc: Amit Daniel Kachhap Link: https://lore.kernel.org/r/20201026121248.2340-4-vincenzo.frascino@arm.com Signed-off-by: Will Deacon --- tools/testing/selftests/arm64/mte/check_child_memory.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tools/testing/selftests/arm64/mte/check_child_memory.c b/tools/testing/selftests/arm64/mte/check_child_memory.c index 97bebdecd29e..43bd94f853ba 100644 --- a/tools/testing/selftests/arm64/mte/check_child_memory.c +++ b/tools/testing/selftests/arm64/mte/check_child_memory.c @@ -163,6 +163,9 @@ int main(int argc, char *argv[]) mte_register_signal(SIGSEGV, mte_default_handler); mte_register_signal(SIGBUS, mte_default_handler); + /* Set test plan */ + ksft_set_plan(12); + evaluate_test(check_child_memory_mapping(USE_MMAP, MTE_SYNC_ERR, MAP_PRIVATE), "Check child anonymous memory with private mapping, precise mode and mmap memory\n"); evaluate_test(check_child_memory_mapping(USE_MMAP, MTE_SYNC_ERR, MAP_SHARED), -- cgit v1.2.3 From 7419390a466e20d4a00cecdb4a68af47d0b21ca5 Mon Sep 17 00:00:00 2001 From: Vincenzo Frascino Date: Mon, 26 Oct 2020 12:12:46 +0000 Subject: kselftest/arm64: Fix check_mmap_options test The check_mmap_options test reports the error below because the test plan is not declared correctly: # Planned tests != run tests (0 != 22) Fix the test adding the correct test plan declaration. Fixes: 53ec81d23213 ("kselftest/arm64: Verify all different mmap MTE options") Signed-off-by: Vincenzo Frascino Cc: Shuah Khan Cc: Catalin Marinas Cc: Will Deacon Cc: Gabor Kertesz Cc: Amit Daniel Kachhap Link: https://lore.kernel.org/r/20201026121248.2340-5-vincenzo.frascino@arm.com Signed-off-by: Will Deacon --- tools/testing/selftests/arm64/mte/check_mmap_options.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/tools/testing/selftests/arm64/mte/check_mmap_options.c b/tools/testing/selftests/arm64/mte/check_mmap_options.c index 33b13b86199b..a04b12c21ac9 100644 --- a/tools/testing/selftests/arm64/mte/check_mmap_options.c +++ b/tools/testing/selftests/arm64/mte/check_mmap_options.c @@ -205,7 +205,11 @@ int main(int argc, char *argv[]) mte_register_signal(SIGBUS, mte_default_handler); mte_register_signal(SIGSEGV, mte_default_handler); + /* Set test plan */ + ksft_set_plan(22); + mte_enable_pstate_tco(); + evaluate_test(check_anonymous_memory_mapping(USE_MMAP, MTE_SYNC_ERR, MAP_PRIVATE, TAG_CHECK_OFF), "Check anonymous memory with private mapping, sync error mode, mmap memory and tag check off\n"); evaluate_test(check_file_memory_mapping(USE_MPROTECT, MTE_SYNC_ERR, MAP_PRIVATE, TAG_CHECK_OFF), -- cgit v1.2.3 From cbb268af05de9beedc6994f4d90628f451d996d7 Mon Sep 17 00:00:00 2001 From: Vincenzo Frascino Date: Mon, 26 Oct 2020 12:12:47 +0000 Subject: kselftest/arm64: Fix check_ksm_options test The check_ksm_options test reports the error below because the test plan is not declared correctly: # Planned tests != run tests (0 != 4) Fix the test adding the correct test plan declaration. Fixes: f981d8fa2646 ("kselftest/arm64: Verify KSM page merge for MTE pages") Signed-off-by: Vincenzo Frascino Cc: Shuah Khan Cc: Catalin Marinas Cc: Will Deacon Cc: Gabor Kertesz Cc: Amit Daniel Kachhap Link: https://lore.kernel.org/r/20201026121248.2340-6-vincenzo.frascino@arm.com Signed-off-by: Will Deacon --- tools/testing/selftests/arm64/mte/check_ksm_options.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/tools/testing/selftests/arm64/mte/check_ksm_options.c b/tools/testing/selftests/arm64/mte/check_ksm_options.c index bc41ae630c86..3b23c4d61d38 100644 --- a/tools/testing/selftests/arm64/mte/check_ksm_options.c +++ b/tools/testing/selftests/arm64/mte/check_ksm_options.c @@ -140,6 +140,10 @@ int main(int argc, char *argv[]) /* Register signal handlers */ mte_register_signal(SIGBUS, mte_default_handler); mte_register_signal(SIGSEGV, mte_default_handler); + + /* Set test plan */ + ksft_set_plan(4); + /* Enable KSM */ mte_ksm_setup(); -- cgit v1.2.3 From 493b35db0548956bd598a54aea0b2bb1c6227e2f Mon Sep 17 00:00:00 2001 From: Vincenzo Frascino Date: Mon, 26 Oct 2020 12:12:48 +0000 Subject: kselftest/arm64: Fix check_user_mem test The check_user_mem test reports the error below because the test plan is not declared correctly: # Planned tests != run tests (0 != 4) Fix the test adding the correct test plan declaration. Fixes: 4dafc08d0ba4 ("kselftest/arm64: Check mte tagged user address in kernel") Signed-off-by: Vincenzo Frascino Cc: Shuah Khan Cc: Catalin Marinas Cc: Will Deacon Cc: Gabor Kertesz Cc: Amit Daniel Kachhap Link: https://lore.kernel.org/r/20201026121248.2340-7-vincenzo.frascino@arm.com Signed-off-by: Will Deacon --- tools/testing/selftests/arm64/mte/check_user_mem.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/tools/testing/selftests/arm64/mte/check_user_mem.c b/tools/testing/selftests/arm64/mte/check_user_mem.c index 594e98e76880..4bfa80f2a8c3 100644 --- a/tools/testing/selftests/arm64/mte/check_user_mem.c +++ b/tools/testing/selftests/arm64/mte/check_user_mem.c @@ -92,9 +92,13 @@ int main(int argc, char *argv[]) err = mte_default_setup(); if (err) return err; + /* Register signal handlers */ mte_register_signal(SIGSEGV, mte_default_handler); + /* Set test plan */ + ksft_set_plan(4); + evaluate_test(check_usermem_access_fault(USE_MMAP, MTE_SYNC_ERR, MAP_PRIVATE), "Check memory access from kernel in sync mode, private mapping and mmap memory\n"); evaluate_test(check_usermem_access_fault(USE_MMAP, MTE_SYNC_ERR, MAP_SHARED), -- cgit v1.2.3 From 5f692a81b4e923d0e856d35a708e53b97e8f8755 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Wed, 28 Oct 2020 14:33:32 +0100 Subject: arm64: vmlinux.lds: account for spurious empty .igot.plt sections Now that we started making the linker warn about orphan sections (input sections that are not explicitly consumed by an output section), some configurations produce the following warning: aarch64-linux-gnu-ld: warning: orphan section `.igot.plt' from `arch/arm64/kernel/head.o' being placed in section `.igot.plt' It could be any file that triggers this - head.o is simply the first input file in the link - and the resulting .igot.plt section never actually appears in vmlinux as it turns out to be empty. So let's add .igot.plt to our collection of input sections to disregard unless they are empty. Signed-off-by: Ard Biesheuvel Cc: Jessica Yu Cc: Kees Cook Cc: Geert Uytterhoeven Cc: Nick Desaulniers Link: https://lore.kernel.org/r/20201028133332.5571-1-ardb@kernel.org Signed-off-by: Will Deacon --- arch/arm64/kernel/vmlinux.lds.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/kernel/vmlinux.lds.S b/arch/arm64/kernel/vmlinux.lds.S index 6d78c041fdf6..1bda604f4c70 100644 --- a/arch/arm64/kernel/vmlinux.lds.S +++ b/arch/arm64/kernel/vmlinux.lds.S @@ -278,7 +278,7 @@ SECTIONS * explicitly check instead of blindly discarding. */ .plt : { - *(.plt) *(.plt.*) *(.iplt) *(.igot) + *(.plt) *(.plt.*) *(.iplt) *(.igot .igot.plt) } ASSERT(SIZEOF(.plt) == 0, "Unexpected run-time procedure linkages detected!") -- cgit v1.2.3 From 4977d121bc9bc5138d4d48b85469123001859573 Mon Sep 17 00:00:00 2001 From: Naohiro Aota Date: Wed, 28 Oct 2020 16:25:36 +0900 Subject: block: advance iov_iter on bio_add_hw_page failure When the bio's size reaches max_append_sectors, bio_add_hw_page returns 0 then __bio_iov_append_get_pages returns -EINVAL. This is an expected result of building a small enough bio not to be split in the IO path. However, iov_iter is not advanced in this case, causing the same pages are filled for the bio again and again. Fix the case by properly advancing the iov_iter for already processed pages. Fixes: 0512a75b98f8 ("block: Introduce REQ_OP_ZONE_APPEND") Cc: stable@vger.kernel.org # 5.8+ Reviewed-by: Johannes Thumshirn Signed-off-by: Naohiro Aota Signed-off-by: Jens Axboe --- block/bio.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/block/bio.c b/block/bio.c index e6e26d7a1ffb..fa01bef35bb1 100644 --- a/block/bio.c +++ b/block/bio.c @@ -1044,6 +1044,7 @@ static int __bio_iov_append_get_pages(struct bio *bio, struct iov_iter *iter) ssize_t size, left; unsigned len, i; size_t offset; + int ret = 0; if (WARN_ON_ONCE(!max_append_sectors)) return 0; @@ -1066,15 +1067,17 @@ static int __bio_iov_append_get_pages(struct bio *bio, struct iov_iter *iter) len = min_t(size_t, PAGE_SIZE - offset, left); if (bio_add_hw_page(q, bio, page, len, offset, - max_append_sectors, &same_page) != len) - return -EINVAL; + max_append_sectors, &same_page) != len) { + ret = -EINVAL; + break; + } if (same_page) put_page(page); offset = 0; } - iov_iter_advance(iter, size); - return 0; + iov_iter_advance(iter, size - left); + return ret; } /** -- cgit v1.2.3 From 8e4c309f9f33b76c09daa02b796ef87918eee494 Mon Sep 17 00:00:00 2001 From: Sascha Hauer Date: Fri, 8 May 2020 07:28:19 +0200 Subject: ata: sata_nv: Fix retrieving of active qcs MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit ata_qc_complete_multiple() has to be called with the tags physically active, that is the hw tag is at bit 0. ap->qc_active has the same tag at bit ATA_TAG_INTERNAL instead, so call ata_qc_get_active() to fix that up. This is done in the vein of 8385d756e114 ("libata: Fix retrieving of active qcs"). Fixes: 28361c403683 ("libata: add extra internal command") Tested-by: Pali Rohár Signed-off-by: Sascha Hauer Signed-off-by: Jens Axboe --- drivers/ata/sata_nv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c index eb9dc14e5147..20190f66ced9 100644 --- a/drivers/ata/sata_nv.c +++ b/drivers/ata/sata_nv.c @@ -2100,7 +2100,7 @@ static int nv_swncq_sdbfis(struct ata_port *ap) pp->dhfis_bits &= ~done_mask; pp->dmafis_bits &= ~done_mask; pp->sdbfis_bits |= done_mask; - ata_qc_complete_multiple(ap, ap->qc_active ^ done_mask); + ata_qc_complete_multiple(ap, ata_qc_get_active(ap) ^ done_mask); if (!ap->qc_active) { DPRINTK("over\n"); -- cgit v1.2.3 From a2d50c1c77aa879af24f9f67b33186737b3d4885 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Tue, 27 Oct 2020 08:32:06 +0100 Subject: arm64: efi: increase EFI PE/COFF header padding to 64 KB Commit 76085aff29f5 ("efi/libstub/arm64: align PE/COFF sections to segment alignment") increased the PE/COFF section alignment to match the minimum segment alignment of the kernel image, which ensures that the kernel does not need to be moved around in memory by the EFI stub if it was built as relocatable. However, the first PE/COFF section starts at _stext, which is only 4 KB aligned, and so the section layout is inconsistent. Existing EFI loaders seem to care little about this, but it is better to clean this up. So let's pad the header to 64 KB to match the PE/COFF section alignment. Fixes: 76085aff29f5 ("efi/libstub/arm64: align PE/COFF sections to segment alignment") Signed-off-by: Ard Biesheuvel Link: https://lore.kernel.org/r/20201027073209.2897-2-ardb@kernel.org Signed-off-by: Will Deacon --- arch/arm64/kernel/efi-header.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/kernel/efi-header.S b/arch/arm64/kernel/efi-header.S index df67c0f2a077..a71844fb923e 100644 --- a/arch/arm64/kernel/efi-header.S +++ b/arch/arm64/kernel/efi-header.S @@ -147,6 +147,6 @@ efi_debug_entry: * correctly at this alignment, we must ensure that .text is * placed at a 4k boundary in the Image to begin with. */ - .align 12 + .balign SEGMENT_ALIGN efi_header_end: .endm -- cgit v1.2.3 From 18e09253ac68b333ec44c502b5c434a697e2d4af Mon Sep 17 00:00:00 2001 From: Pascal Paillet Date: Wed, 28 Oct 2020 12:47:07 +0100 Subject: ARM: dts: stm32: Describe Vin power supply on stm32mp15xx-dkx board Add description for Vin power supply and for peripherals that are supplied by Vin. Signed-off-by: Pascal Paillet Signed-off-by: Alexandre Torgue --- arch/arm/boot/dts/stm32mp15xx-dkx.dtsi | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/arch/arm/boot/dts/stm32mp15xx-dkx.dtsi b/arch/arm/boot/dts/stm32mp15xx-dkx.dtsi index a5307745719a..93398cfae97e 100644 --- a/arch/arm/boot/dts/stm32mp15xx-dkx.dtsi +++ b/arch/arm/boot/dts/stm32mp15xx-dkx.dtsi @@ -80,6 +80,14 @@ dais = <&sai2a_port &sai2b_port &i2s2_port>; status = "okay"; }; + + vin: vin { + compatible = "regulator-fixed"; + regulator-name = "vin"; + regulator-min-microvolt = <5000000>; + regulator-max-microvolt = <5000000>; + regulator-always-on; + }; }; &adc { @@ -240,9 +248,18 @@ regulators { compatible = "st,stpmic1-regulators"; + buck1-supply = <&vin>; + buck2-supply = <&vin>; + buck3-supply = <&vin>; + buck4-supply = <&vin>; ldo1-supply = <&v3v3>; + ldo2-supply = <&vin>; ldo3-supply = <&vdd_ddr>; + ldo4-supply = <&vin>; + ldo5-supply = <&vin>; ldo6-supply = <&v3v3>; + vref_ddr-supply = <&vin>; + boost-supply = <&vin>; pwr_sw1-supply = <&bst_out>; pwr_sw2-supply = <&bst_out>; -- cgit v1.2.3 From 728a7e225ad807b4b4de3be3fb653424199f8a76 Mon Sep 17 00:00:00 2001 From: Pascal Paillet Date: Wed, 28 Oct 2020 12:49:16 +0100 Subject: ARM: dts: stm32: Describe Vin power supply on stm32mp157c-edx board Add description for Vin power supply and for peripherals that are supplied by Vin. Signed-off-by: Pascal Paillet Signed-off-by: Alexandre Torgue --- arch/arm/boot/dts/stm32mp157c-ed1.dts | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/arch/arm/boot/dts/stm32mp157c-ed1.dts b/arch/arm/boot/dts/stm32mp157c-ed1.dts index ca109dc18238..2e77ccec3fc1 100644 --- a/arch/arm/boot/dts/stm32mp157c-ed1.dts +++ b/arch/arm/boot/dts/stm32mp157c-ed1.dts @@ -89,6 +89,14 @@ states = <1800000 0x1>, <2900000 0x0>; }; + + vin: vin { + compatible = "regulator-fixed"; + regulator-name = "vin"; + regulator-min-microvolt = <5000000>; + regulator-max-microvolt = <5000000>; + regulator-always-on; + }; }; &adc { @@ -150,11 +158,18 @@ regulators { compatible = "st,stpmic1-regulators"; + buck1-supply = <&vin>; + buck2-supply = <&vin>; + buck3-supply = <&vin>; + buck4-supply = <&vin>; ldo1-supply = <&v3v3>; ldo2-supply = <&v3v3>; ldo3-supply = <&vdd_ddr>; + ldo4-supply = <&vin>; ldo5-supply = <&v3v3>; ldo6-supply = <&v3v3>; + vref_ddr-supply = <&vin>; + boost-supply = <&vin>; pwr_sw1-supply = <&bst_out>; pwr_sw2-supply = <&bst_out>; -- cgit v1.2.3 From 13150bc5416f45234c955e5bed91623d178c6117 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Tue, 27 Oct 2020 16:11:32 +0100 Subject: module: use hidden visibility for weak symbol references Geert reports that commit be2881824ae9eb92 ("arm64/build: Assert for unwanted sections") results in build errors on arm64 for configurations that have CONFIG_MODULES disabled. The commit in question added ASSERT()s to the arm64 linker script to ensure that linker generated sections such as .got.plt etc are empty, but as it turns out, there are corner cases where the linker does emit content into those sections. More specifically, weak references to function symbols (which can remain unsatisfied, and can therefore not be emitted as relative references) will be emitted as GOT and PLT entries when linking the kernel in PIE mode (which is the case when CONFIG_RELOCATABLE is enabled, which is on by default). What happens is that code such as struct device *(*fn)(struct device *dev); struct device *iommu_device; fn = symbol_get(mdev_get_iommu_device); if (fn) { iommu_device = fn(dev); essentially gets converted into the following when CONFIG_MODULES is off: struct device *iommu_device; if (&mdev_get_iommu_device) { iommu_device = mdev_get_iommu_device(dev); where mdev_get_iommu_device is emitted as a weak symbol reference into the object file. The first reference is decorated with an ordinary ABS64 data relocation (which yields 0x0 if the reference remains unsatisfied). However, the indirect call is turned into a direct call covered by a R_AARCH64_CALL26 relocation, which is converted into a call via a PLT entry taking the target address from the associated GOT entry. Given that such GOT and PLT entries are unnecessary for fully linked binaries such as the kernel, let's give these weak symbol references hidden visibility, so that the linker knows that the weak reference via R_AARCH64_CALL26 can simply remain unsatisfied. Signed-off-by: Ard Biesheuvel Tested-by: Geert Uytterhoeven Reviewed-by: Fangrui Song Acked-by: Jessica Yu Cc: Jessica Yu Cc: Kees Cook Cc: Geert Uytterhoeven Cc: Nick Desaulniers Link: https://lore.kernel.org/r/20201027151132.14066-1-ardb@kernel.org Signed-off-by: Will Deacon --- include/linux/module.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/module.h b/include/linux/module.h index 7ccdf87f376f..6264617bab4d 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -740,7 +740,7 @@ static inline bool within_module(unsigned long addr, const struct module *mod) } /* Get/put a kernel symbol (calls should be symmetric) */ -#define symbol_get(x) ({ extern typeof(x) x __attribute__((weak)); &(x); }) +#define symbol_get(x) ({ extern typeof(x) x __attribute__((weak,visibility("hidden"))); &(x); }) #define symbol_put(x) do { } while (0) #define symbol_put_addr(x) do { } while (0) -- cgit v1.2.3 From 921c7ebd1337d1a46783d7e15a850e12aed2eaa0 Mon Sep 17 00:00:00 2001 From: Mateusz Nosek Date: Sun, 27 Sep 2020 02:08:58 +0200 Subject: futex: Fix incorrect should_fail_futex() handling If should_futex_fail() returns true in futex_wake_pi(), then the 'ret' variable is set to -EFAULT and then immediately overwritten. So the failure injection is non-functional. Fix it by actually leaving the function and returning -EFAULT. The Fixes tag is kinda blury because the initial commit which introduced failure injection was already sloppy, but the below mentioned commit broke it completely. [ tglx: Massaged changelog ] Fixes: 6b4f4bc9cb22 ("locking/futex: Allow low-level atomic operations to return -EAGAIN") Signed-off-by: Mateusz Nosek Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/r/20200927000858.24219-1-mateusznosek0@gmail.com --- kernel/futex.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/kernel/futex.c b/kernel/futex.c index a5876694a60e..39681bf8b06c 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -1502,8 +1502,10 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_pi_state *pi_ */ newval = FUTEX_WAITERS | task_pid_vnr(new_owner); - if (unlikely(should_fail_futex(true))) + if (unlikely(should_fail_futex(true))) { ret = -EFAULT; + goto out_unlock; + } ret = cmpxchg_futex_value_locked(&curval, uaddr, uval, newval); if (!ret && (curval != uval)) { -- cgit v1.2.3 From 00d4394792418f8fe968f0cb22557053c6310010 Mon Sep 17 00:00:00 2001 From: Tom Rix Date: Tue, 27 Oct 2020 11:59:34 -0700 Subject: cpufreq: speedstep: remove unneeded semicolon A semicolon is not needed after a switch statement. Signed-off-by: Tom Rix Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/speedstep-lib.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/cpufreq/speedstep-lib.c b/drivers/cpufreq/speedstep-lib.c index a13a2d1e444e..0b66df4ed513 100644 --- a/drivers/cpufreq/speedstep-lib.c +++ b/drivers/cpufreq/speedstep-lib.c @@ -240,7 +240,7 @@ unsigned int speedstep_get_frequency(enum speedstep_processor processor) return pentium3_get_frequency(processor); default: return 0; - }; + } return 0; } EXPORT_SYMBOL_GPL(speedstep_get_frequency); -- cgit v1.2.3 From 5112e9a540fb4a63184850c6570f30ae13b5f5e3 Mon Sep 17 00:00:00 2001 From: Harshad Shirwadkar Date: Fri, 23 Oct 2020 09:13:39 -0700 Subject: ext4: fix double locking in ext4_fc_commit_dentry_updates() Fixed double locking of sbi->s_fc_lock in the above function as reported by kernel-test-robot. Signed-off-by: Harshad Shirwadkar Link: https://lore.kernel.org/r/20201023161339.1449437-1-harshadshirwadkar@gmail.com Signed-off-by: Theodore Ts'o --- fs/ext4/fast_commit.c | 1 - 1 file changed, 1 deletion(-) diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c index 447c8d93f480..1d72f8f13a05 100644 --- a/fs/ext4/fast_commit.c +++ b/fs/ext4/fast_commit.c @@ -964,7 +964,6 @@ static int ext4_fc_commit_dentry_updates(journal_t *journal, u32 *crc) fc_dentry->fcd_parent, fc_dentry->fcd_ino, fc_dentry->fcd_name.len, fc_dentry->fcd_name.name, crc)) { - spin_lock(&sbi->s_fc_lock); ret = -ENOSPC; goto lock_and_exit; } -- cgit v1.2.3 From d0520df7240f7117226e871453e1b20d12a9614a Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Mon, 26 Oct 2020 21:49:13 -0700 Subject: ext4: properly check for dirty state in ext4_inode_datasync_dirty() ext4_inode_datasync_dirty() needs to return 'true' if the inode is dirty, 'false' otherwise, but the logic seems to be incorrectly changed by commit aa75f4d3daae ("ext4: main fast-commit commit path"). This introduces a problem with swap files that are always failing to be activated, showing this error in dmesg: [ 34.406479] swapon: file is not committed Simple test case to reproduce the problem: # fallocate -l 8G swapfile # chmod 0600 swapfile # mkswap swapfile # swapon swapfile Fix the logic to return the proper state of the inode. Link: https://lore.kernel.org/lkml/20201024131333.GA32124@xps-13-7390 Fixes: 8016e29f4362 ("ext4: fast commit recovery path") Signed-off-by: Andrea Righi Signed-off-by: Harshad Shirwadkar Link: https://lore.kernel.org/r/20201027044915.2553163-1-harshadshirwadkar@gmail.com Signed-off-by: Theodore Ts'o --- fs/ext4/inode.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 03c2253005f0..520a0209451e 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -3307,10 +3307,12 @@ static bool ext4_inode_datasync_dirty(struct inode *inode) if (journal) { if (jbd2_transaction_committed(journal, - EXT4_I(inode)->i_datasync_tid)) - return true; - return atomic_read(&EXT4_SB(inode->i_sb)->s_fc_subtid) >= - EXT4_I(inode)->i_fc_committed_subtid; + EXT4_I(inode)->i_datasync_tid)) + return false; + if (test_opt2(inode->i_sb, JOURNAL_FAST_COMMIT)) + return atomic_read(&EXT4_SB(inode->i_sb)->s_fc_subtid) < + EXT4_I(inode)->i_fc_committed_subtid; + return true; } /* Any metadata buffers to write? */ -- cgit v1.2.3 From e029c5f2798720b463e8df0e184a4d1036311b43 Mon Sep 17 00:00:00 2001 From: Harshad Shirwadkar Date: Mon, 26 Oct 2020 21:49:14 -0700 Subject: ext4: make num of fast commit blocks configurable This patch reserves a field in the jbd2 superblock for number of fast commit blocks. When this value is non-zero, Ext4 uses this field to set the number of fast commit blocks. Fixes: 6866d7b3f2bb ("ext4/jbd2: add fast commit initialization") Signed-off-by: Harshad Shirwadkar Link: https://lore.kernel.org/r/20201027044915.2553163-2-harshadshirwadkar@gmail.com Signed-off-by: Theodore Ts'o --- fs/ext4/fast_commit.c | 12 +++++++++++- include/linux/jbd2.h | 5 ++++- 2 files changed, 15 insertions(+), 2 deletions(-) diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c index 1d72f8f13a05..dd7931400474 100644 --- a/fs/ext4/fast_commit.c +++ b/fs/ext4/fast_commit.c @@ -2077,6 +2077,8 @@ static int ext4_fc_replay(journal_t *journal, struct buffer_head *bh, void ext4_fc_init(struct super_block *sb, journal_t *journal) { + int num_fc_blocks; + /* * We set replay callback even if fast commit disabled because we may * could still have fast commit blocks that need to be replayed even if @@ -2086,7 +2088,15 @@ void ext4_fc_init(struct super_block *sb, journal_t *journal) if (!test_opt2(sb, JOURNAL_FAST_COMMIT)) return; journal->j_fc_cleanup_callback = ext4_fc_cleanup; - if (jbd2_fc_init(journal, EXT4_NUM_FC_BLKS)) { + if (!buffer_uptodate(journal->j_sb_buffer) + && ext4_read_bh_lock(journal->j_sb_buffer, REQ_META | REQ_PRIO, + true)) { + ext4_msg(sb, KERN_ERR, "I/O error on journal"); + return; + } + num_fc_blocks = be32_to_cpu(journal->j_superblock->s_num_fc_blks); + if (jbd2_fc_init(journal, num_fc_blocks ? num_fc_blocks : + EXT4_NUM_FC_BLKS)) { pr_warn("Error while enabling fast commits, turning off."); ext4_clear_feature_fast_commit(sb); } diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index fb3d71ad6eea..7e88bbc16ffb 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -263,7 +263,10 @@ typedef struct journal_superblock_s /* 0x0050 */ __u8 s_checksum_type; /* checksum type */ __u8 s_padding2[3]; - __u32 s_padding[42]; +/* 0x0054 */ + __be32 s_num_fc_blks; /* Number of fast commit blocks */ +/* 0x0058 */ + __u32 s_padding[41]; __be32 s_checksum; /* crc32c(superblock) */ /* 0x0100 */ -- cgit v1.2.3 From ababea77bc50e004f7cc2792f0ec8d26b7574c3e Mon Sep 17 00:00:00 2001 From: Harshad Shirwadkar Date: Mon, 26 Oct 2020 21:49:15 -0700 Subject: ext4: use s_mount_flags instead of s_mount_state for fast commit state Ext4's fast commit related transient states should use sb->s_mount_flags instead of persistent sb->s_mount_state. Fixes: 8016e29f4362 ("ext4: fast commit recovery path") Signed-off-by: Harshad Shirwadkar Link: https://lore.kernel.org/r/20201027044915.2553163-3-harshadshirwadkar@gmail.com Signed-off-by: Theodore Ts'o --- fs/ext4/ext4.h | 8 ++++---- fs/ext4/fast_commit.c | 18 +++++++++--------- fs/ext4/super.c | 4 ++-- 3 files changed, 15 insertions(+), 15 deletions(-) diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 254d1c26bea8..2337e443fa30 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -1166,10 +1166,6 @@ struct ext4_inode_info { #define EXT4_VALID_FS 0x0001 /* Unmounted cleanly */ #define EXT4_ERROR_FS 0x0002 /* Errors detected */ #define EXT4_ORPHAN_FS 0x0004 /* Orphans being recovered */ -#define EXT4_FC_INELIGIBLE 0x0008 /* Fast commit ineligible */ -#define EXT4_FC_COMMITTING 0x0010 /* File system underoing a fast - * commit. - */ #define EXT4_FC_REPLAY 0x0020 /* Fast commit replay ongoing */ /* @@ -1431,6 +1427,10 @@ struct ext4_super_block { */ #define EXT4_MF_MNTDIR_SAMPLED 0x0001 #define EXT4_MF_FS_ABORTED 0x0002 /* Fatal error detected */ +#define EXT4_MF_FC_INELIGIBLE 0x0004 /* Fast commit ineligible */ +#define EXT4_MF_FC_COMMITTING 0x0008 /* File system underoing a fast + * commit. + */ #ifdef CONFIG_FS_ENCRYPTION #define DUMMY_ENCRYPTION_ENABLED(sbi) ((sbi)->s_dummy_enc_policy.policy != NULL) diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c index dd7931400474..3ee43fd6d5aa 100644 --- a/fs/ext4/fast_commit.c +++ b/fs/ext4/fast_commit.c @@ -269,7 +269,7 @@ void ext4_fc_mark_ineligible(struct super_block *sb, int reason) (EXT4_SB(sb)->s_mount_state & EXT4_FC_REPLAY)) return; - sbi->s_mount_state |= EXT4_FC_INELIGIBLE; + sbi->s_mount_flags |= EXT4_MF_FC_INELIGIBLE; WARN_ON(reason >= EXT4_FC_REASON_MAX); sbi->s_fc_stats.fc_ineligible_reason_count[reason]++; } @@ -292,7 +292,7 @@ void ext4_fc_start_ineligible(struct super_block *sb, int reason) } /* - * Stop a fast commit ineligible update. We set EXT4_FC_INELIGIBLE flag here + * Stop a fast commit ineligible update. We set EXT4_MF_FC_INELIGIBLE flag here * to ensure that after stopping the ineligible update, at least one full * commit takes place. */ @@ -302,13 +302,13 @@ void ext4_fc_stop_ineligible(struct super_block *sb) (EXT4_SB(sb)->s_mount_state & EXT4_FC_REPLAY)) return; - EXT4_SB(sb)->s_mount_state |= EXT4_FC_INELIGIBLE; + EXT4_SB(sb)->s_mount_flags |= EXT4_MF_FC_INELIGIBLE; atomic_dec(&EXT4_SB(sb)->s_fc_ineligible_updates); } static inline int ext4_fc_is_ineligible(struct super_block *sb) { - return (EXT4_SB(sb)->s_mount_state & EXT4_FC_INELIGIBLE) || + return (EXT4_SB(sb)->s_mount_flags & EXT4_MF_FC_INELIGIBLE) || atomic_read(&EXT4_SB(sb)->s_fc_ineligible_updates); } @@ -358,7 +358,7 @@ static int ext4_fc_track_template( spin_lock(&sbi->s_fc_lock); if (list_empty(&EXT4_I(inode)->i_fc_list)) list_add_tail(&EXT4_I(inode)->i_fc_list, - (sbi->s_mount_state & EXT4_FC_COMMITTING) ? + (sbi->s_mount_flags & EXT4_MF_FC_COMMITTING) ? &sbi->s_fc_q[FC_Q_STAGING] : &sbi->s_fc_q[FC_Q_MAIN]); spin_unlock(&sbi->s_fc_lock); @@ -411,7 +411,7 @@ static int __track_dentry_update(struct inode *inode, void *arg, bool update) node->fcd_name.len = dentry->d_name.len; spin_lock(&sbi->s_fc_lock); - if (sbi->s_mount_state & EXT4_FC_COMMITTING) + if (sbi->s_mount_flags & EXT4_MF_FC_COMMITTING) list_add_tail(&node->fcd_list, &sbi->s_fc_dentry_q[FC_Q_STAGING]); else @@ -846,7 +846,7 @@ static int ext4_fc_submit_inode_data_all(journal_t *journal) int ret = 0; spin_lock(&sbi->s_fc_lock); - sbi->s_mount_state |= EXT4_FC_COMMITTING; + sbi->s_mount_flags |= EXT4_MF_FC_COMMITTING; list_for_each(pos, &sbi->s_fc_q[FC_Q_MAIN]) { ei = list_entry(pos, struct ext4_inode_info, i_fc_list); ext4_set_inode_state(&ei->vfs_inode, EXT4_STATE_FC_COMMITTING); @@ -1190,8 +1190,8 @@ static void ext4_fc_cleanup(journal_t *journal, int full) list_splice_init(&sbi->s_fc_q[FC_Q_STAGING], &sbi->s_fc_q[FC_Q_STAGING]); - sbi->s_mount_state &= ~EXT4_FC_COMMITTING; - sbi->s_mount_state &= ~EXT4_FC_INELIGIBLE; + sbi->s_mount_flags &= ~EXT4_MF_FC_COMMITTING; + sbi->s_mount_flags &= ~EXT4_MF_FC_INELIGIBLE; if (full) sbi->s_fc_bytes = 0; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 03373471131c..d92de21212e9 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -4777,8 +4777,8 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) INIT_LIST_HEAD(&sbi->s_fc_dentry_q[FC_Q_MAIN]); INIT_LIST_HEAD(&sbi->s_fc_dentry_q[FC_Q_STAGING]); sbi->s_fc_bytes = 0; - sbi->s_mount_state &= ~EXT4_FC_INELIGIBLE; - sbi->s_mount_state &= ~EXT4_FC_COMMITTING; + sbi->s_mount_flags &= ~EXT4_MF_FC_INELIGIBLE; + sbi->s_mount_flags &= ~EXT4_MF_FC_COMMITTING; spin_lock_init(&sbi->s_fc_lock); memset(&sbi->s_fc_stats, 0, sizeof(sbi->s_fc_stats)); sbi->s_fc_replay_state.fc_regions = NULL; -- cgit v1.2.3 From ea4b01d9b81f5f381fc6832bc31046878a2d1a5d Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Tue, 27 Oct 2020 10:51:27 +0100 Subject: jbd2: fix a kernel-doc markup The kernel-doc markup that documents _fc_replay_callback is missing an asterisk, causing this warning: ../include/linux/jbd2.h:1271: warning: Function parameter or member 'j_fc_replay_callback' not described in 'journal_s' When building the docs. Fixes: 609f928af48f ("jbd2: fast commit recovery path") Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/6055927ada2015b55b413cdd2670533bdc9a8da2.1603791716.git.mchehab+huawei@kernel.org Signed-off-by: Theodore Ts'o --- include/linux/jbd2.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 7e88bbc16ffb..1d5566af48ac 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1256,7 +1256,7 @@ struct journal_s */ void (*j_fc_cleanup_callback)(struct journal_s *journal, int); - /* + /** * @j_fc_replay_callback: * * File-system specific function that performs replay of a fast -- cgit v1.2.3 From b5b18160a3e7a9f55e3528d77051670cca6d9314 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 27 Oct 2020 14:27:51 +0100 Subject: ext4: fix mmap write protection for data=journal mode Commit afb585a97f81 "ext4: data=journal: write-protect pages on j_submit_inode_data_buffers()") added calls ext4_jbd2_inode_add_write() to track inode ranges whose mappings need to get write-protected during transaction commits. However the added calls use wrong start of a range (0 instead of page offset) and so write protection is not necessarily effective. Use correct range start to fix the problem. Fixes: afb585a97f81 ("ext4: data=journal: write-protect pages on j_submit_inode_data_buffers()") Signed-off-by: Jan Kara Link: https://lore.kernel.org/r/20201027132751.29858-1-jack@suse.cz Signed-off-by: Theodore Ts'o --- fs/ext4/inode.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 520a0209451e..b96a18679a27 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1918,7 +1918,7 @@ static int __ext4_journalled_writepage(struct page *page, } if (ret == 0) ret = err; - err = ext4_jbd2_inode_add_write(handle, inode, 0, len); + err = ext4_jbd2_inode_add_write(handle, inode, page_offset(page), len); if (ret == 0) ret = err; EXT4_I(inode)->i_datasync_tid = handle->h_transaction->t_tid; @@ -6159,7 +6159,8 @@ retry_alloc: if (ext4_walk_page_buffers(handle, page_buffers(page), 0, len, NULL, write_end_fn)) goto out_error; - if (ext4_jbd2_inode_add_write(handle, inode, 0, len)) + if (ext4_jbd2_inode_add_write(handle, inode, + page_offset(page), len)) goto out_error; ext4_set_inode_state(inode, EXT4_STATE_JDATA); } else { -- cgit v1.2.3 From 8c9be1e58a8dc03ef815ca51ece95c5e6862b25e Mon Sep 17 00:00:00 2001 From: Harshad Shirwadkar Date: Tue, 27 Oct 2020 13:43:42 -0700 Subject: ext4: use IS_ERR() for error checking of path With this fix, fast commit recovery code uses IS_ERR() for path returned by ext4_find_extent. Fixes: 8016e29f4362 ("ext4: fast commit recovery path") Reported-by: Dan Carpenter Signed-off-by: Harshad Shirwadkar Link: https://lore.kernel.org/r/20201027204342.2794949-1-harshadshirwadkar@gmail.com Signed-off-by: Theodore Ts'o --- fs/ext4/fast_commit.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c index 3ee43fd6d5aa..8d43058386c3 100644 --- a/fs/ext4/fast_commit.c +++ b/fs/ext4/fast_commit.c @@ -1616,8 +1616,10 @@ static int ext4_fc_replay_add_range(struct super_block *sb, if (ret == 0) { /* Range is not mapped */ path = ext4_find_extent(inode, cur, NULL, 0); - if (!path) - continue; + if (IS_ERR(path)) { + iput(inode); + return 0; + } memset(&newex, 0, sizeof(newex)); newex.ee_block = cpu_to_le32(cur); ext4_ext_store_pblock( -- cgit v1.2.3 From d7dce9e08595e80bf8039a81794809c66fe26431 Mon Sep 17 00:00:00 2001 From: yangerkun Date: Wed, 28 Oct 2020 13:56:17 +0800 Subject: ext4: do not use extent after put_bh ext4_ext_search_right() will read more extent blocks and call put_bh after we get the information we need. However, ret_ex will break this and may cause use-after-free once pagecache has been freed. Fix it by copying the extent structure if needed. Signed-off-by: yangerkun Link: https://lore.kernel.org/r/20201028055617.2569255-1-yangerkun@huawei.com Signed-off-by: Theodore Ts'o Cc: stable@kernel.org --- fs/ext4/extents.c | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 559100f3e23c..57cfa28919a9 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -1471,16 +1471,16 @@ static int ext4_ext_search_left(struct inode *inode, } /* - * search the closest allocated block to the right for *logical - * and returns it at @logical + it's physical address at @phys - * if *logical is the largest allocated block, the function - * returns 0 at @phys - * return value contains 0 (success) or error code + * Search the closest allocated block to the right for *logical + * and returns it at @logical + it's physical address at @phys. + * If not exists, return 0 and @phys is set to 0. We will return + * 1 which means we found an allocated block and ret_ex is valid. + * Or return a (< 0) error code. */ static int ext4_ext_search_right(struct inode *inode, struct ext4_ext_path *path, ext4_lblk_t *logical, ext4_fsblk_t *phys, - struct ext4_extent **ret_ex) + struct ext4_extent *ret_ex) { struct buffer_head *bh = NULL; struct ext4_extent_header *eh; @@ -1574,10 +1574,11 @@ got_index: found_extent: *logical = le32_to_cpu(ex->ee_block); *phys = ext4_ext_pblock(ex); - *ret_ex = ex; + if (ret_ex) + *ret_ex = *ex; if (bh) put_bh(bh); - return 0; + return 1; } /* @@ -2868,8 +2869,8 @@ again: */ lblk = ex_end + 1; err = ext4_ext_search_right(inode, path, &lblk, &pblk, - &ex); - if (err) + NULL); + if (err < 0) goto out; if (pblk) { partial.pclu = EXT4_B2C(sbi, pblk); @@ -4039,7 +4040,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, struct ext4_map_blocks *map, int flags) { struct ext4_ext_path *path = NULL; - struct ext4_extent newex, *ex, *ex2; + struct ext4_extent newex, *ex, ex2; struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); ext4_fsblk_t newblock = 0, pblk; int err = 0, depth, ret; @@ -4175,15 +4176,14 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, if (err) goto out; ar.lright = map->m_lblk; - ex2 = NULL; err = ext4_ext_search_right(inode, path, &ar.lright, &ar.pright, &ex2); - if (err) + if (err < 0) goto out; /* Check if the extent after searching to the right implies a * cluster we can use. */ - if ((sbi->s_cluster_ratio > 1) && ex2 && - get_implied_cluster_alloc(inode->i_sb, map, ex2, path)) { + if ((sbi->s_cluster_ratio > 1) && err && + get_implied_cluster_alloc(inode->i_sb, map, &ex2, path)) { ar.len = allocated = map->m_len; newblock = map->m_pblk; goto got_allocated_blocks; -- cgit v1.2.3 From f8f4acb6cded4e455b2d390ce2221391fc3f09ee Mon Sep 17 00:00:00 2001 From: Daniel Rosenberg Date: Wed, 28 Oct 2020 05:08:20 +0000 Subject: ext4: use generic casefolding support This switches ext4 over to the generic support provided in libfs. Since casefolded dentries behave the same in ext4 and f2fs, we decrease the maintenance burden by unifying them, and any optimizations will immediately apply to both. Signed-off-by: Daniel Rosenberg Reviewed-by: Eric Biggers Link: https://lore.kernel.org/r/20201028050820.1636571-1-drosen@google.com Signed-off-by: Theodore Ts'o --- fs/ext4/dir.c | 64 ++------------------------------------------------------- fs/ext4/ext4.h | 12 ----------- fs/ext4/hash.c | 2 +- fs/ext4/namei.c | 20 ++++++++---------- fs/ext4/super.c | 12 +++++------ 5 files changed, 17 insertions(+), 93 deletions(-) diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index 5b81f3b080ee..ca50c90adc4c 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -669,68 +669,8 @@ const struct file_operations ext4_dir_operations = { }; #ifdef CONFIG_UNICODE -static int ext4_d_compare(const struct dentry *dentry, unsigned int len, - const char *str, const struct qstr *name) -{ - struct qstr qstr = {.name = str, .len = len }; - const struct dentry *parent = READ_ONCE(dentry->d_parent); - const struct inode *inode = d_inode_rcu(parent); - char strbuf[DNAME_INLINE_LEN]; - - if (!inode || !IS_CASEFOLDED(inode) || - !EXT4_SB(inode->i_sb)->s_encoding) { - if (len != name->len) - return -1; - return memcmp(str, name->name, len); - } - - /* - * If the dentry name is stored in-line, then it may be concurrently - * modified by a rename. If this happens, the VFS will eventually retry - * the lookup, so it doesn't matter what ->d_compare() returns. - * However, it's unsafe to call utf8_strncasecmp() with an unstable - * string. Therefore, we have to copy the name into a temporary buffer. - */ - if (len <= DNAME_INLINE_LEN - 1) { - memcpy(strbuf, str, len); - strbuf[len] = 0; - qstr.name = strbuf; - /* prevent compiler from optimizing out the temporary buffer */ - barrier(); - } - - return ext4_ci_compare(inode, name, &qstr, false); -} - -static int ext4_d_hash(const struct dentry *dentry, struct qstr *str) -{ - const struct ext4_sb_info *sbi = EXT4_SB(dentry->d_sb); - const struct unicode_map *um = sbi->s_encoding; - const struct inode *inode = d_inode_rcu(dentry); - unsigned char *norm; - int len, ret = 0; - - if (!inode || !IS_CASEFOLDED(inode) || !um) - return 0; - - norm = kmalloc(PATH_MAX, GFP_ATOMIC); - if (!norm) - return -ENOMEM; - - len = utf8_casefold(um, str, norm, PATH_MAX); - if (len < 0) { - if (ext4_has_strict_mode(sbi)) - ret = -EINVAL; - goto out; - } - str->hash = full_name_hash(dentry, norm, len); -out: - kfree(norm); - return ret; -} - const struct dentry_operations ext4_dentry_ops = { - .d_hash = ext4_d_hash, - .d_compare = ext4_d_compare, + .d_hash = generic_ci_d_hash, + .d_compare = generic_ci_d_compare, }; #endif diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 2337e443fa30..45fcdbf538d1 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -1443,14 +1443,6 @@ struct ext4_super_block { #define EXT4_ENC_UTF8_12_1 1 -/* - * Flags for ext4_sb_info.s_encoding_flags. - */ -#define EXT4_ENC_STRICT_MODE_FL (1 << 0) - -#define ext4_has_strict_mode(sbi) \ - (sbi->s_encoding_flags & EXT4_ENC_STRICT_MODE_FL) - /* * fourth extended-fs super-block data in memory */ @@ -1500,10 +1492,6 @@ struct ext4_sb_info { struct kobject s_kobj; struct completion s_kobj_unregister; struct super_block *s_sb; -#ifdef CONFIG_UNICODE - struct unicode_map *s_encoding; - __u16 s_encoding_flags; -#endif /* Journaling */ struct journal_s *s_journal; diff --git a/fs/ext4/hash.c b/fs/ext4/hash.c index 2924261226e0..a92eb79de0cc 100644 --- a/fs/ext4/hash.c +++ b/fs/ext4/hash.c @@ -275,7 +275,7 @@ int ext4fs_dirhash(const struct inode *dir, const char *name, int len, struct dx_hash_info *hinfo) { #ifdef CONFIG_UNICODE - const struct unicode_map *um = EXT4_SB(dir->i_sb)->s_encoding; + const struct unicode_map *um = dir->i_sb->s_encoding; int r, dlen; unsigned char *buff; struct qstr qstr = {.name = name, .len = len }; diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 5159830dacb8..f458d1d81d96 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -1285,8 +1285,8 @@ static void dx_insert_block(struct dx_frame *frame, u32 hash, ext4_lblk_t block) int ext4_ci_compare(const struct inode *parent, const struct qstr *name, const struct qstr *entry, bool quick) { - const struct ext4_sb_info *sbi = EXT4_SB(parent->i_sb); - const struct unicode_map *um = sbi->s_encoding; + const struct super_block *sb = parent->i_sb; + const struct unicode_map *um = sb->s_encoding; int ret; if (quick) @@ -1298,7 +1298,7 @@ int ext4_ci_compare(const struct inode *parent, const struct qstr *name, /* Handle invalid character sequence as either an error * or as an opaque byte sequence. */ - if (ext4_has_strict_mode(sbi)) + if (sb_has_strict_encoding(sb)) return -EINVAL; if (name->len != entry->len) @@ -1315,7 +1315,7 @@ void ext4_fname_setup_ci_filename(struct inode *dir, const struct qstr *iname, { int len; - if (!IS_CASEFOLDED(dir) || !EXT4_SB(dir->i_sb)->s_encoding) { + if (!IS_CASEFOLDED(dir) || !dir->i_sb->s_encoding) { cf_name->name = NULL; return; } @@ -1324,7 +1324,7 @@ void ext4_fname_setup_ci_filename(struct inode *dir, const struct qstr *iname, if (!cf_name->name) return; - len = utf8_casefold(EXT4_SB(dir->i_sb)->s_encoding, + len = utf8_casefold(dir->i_sb->s_encoding, iname, cf_name->name, EXT4_NAME_LEN); if (len <= 0) { @@ -1361,7 +1361,7 @@ static inline bool ext4_match(const struct inode *parent, #endif #ifdef CONFIG_UNICODE - if (EXT4_SB(parent->i_sb)->s_encoding && IS_CASEFOLDED(parent)) { + if (parent->i_sb->s_encoding && IS_CASEFOLDED(parent)) { if (fname->cf_name.name) { struct qstr cf = {.name = fname->cf_name.name, .len = fname->cf_name.len}; @@ -2180,9 +2180,6 @@ static int ext4_add_entry(handle_t *handle, struct dentry *dentry, struct buffer_head *bh = NULL; struct ext4_dir_entry_2 *de; struct super_block *sb; -#ifdef CONFIG_UNICODE - struct ext4_sb_info *sbi; -#endif struct ext4_filename fname; int retval; int dx_fallback=0; @@ -2199,9 +2196,8 @@ static int ext4_add_entry(handle_t *handle, struct dentry *dentry, return -EINVAL; #ifdef CONFIG_UNICODE - sbi = EXT4_SB(sb); - if (ext4_has_strict_mode(sbi) && IS_CASEFOLDED(dir) && - sbi->s_encoding && utf8_validate(sbi->s_encoding, &dentry->d_name)) + if (sb_has_strict_encoding(sb) && IS_CASEFOLDED(dir) && + sb->s_encoding && utf8_validate(sb->s_encoding, &dentry->d_name)) return -EINVAL; #endif diff --git a/fs/ext4/super.c b/fs/ext4/super.c index d92de21212e9..0f5a31498393 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1288,7 +1288,7 @@ static void ext4_put_super(struct super_block *sb) fs_put_dax(sbi->s_daxdev); fscrypt_free_dummy_policy(&sbi->s_dummy_enc_policy); #ifdef CONFIG_UNICODE - utf8_unload(sbi->s_encoding); + utf8_unload(sb->s_encoding); #endif kfree(sbi); } @@ -4303,7 +4303,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) goto failed_mount; #ifdef CONFIG_UNICODE - if (ext4_has_feature_casefold(sb) && !sbi->s_encoding) { + if (ext4_has_feature_casefold(sb) && !sb->s_encoding) { const struct ext4_sb_encodings *encoding_info; struct unicode_map *encoding; __u16 encoding_flags; @@ -4334,8 +4334,8 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) "%s-%s with flags 0x%hx", encoding_info->name, encoding_info->version?:"\b", encoding_flags); - sbi->s_encoding = encoding; - sbi->s_encoding_flags = encoding_flags; + sb->s_encoding = encoding; + sb->s_encoding_flags = encoding_flags; } #endif @@ -4975,7 +4975,7 @@ no_journal: } #ifdef CONFIG_UNICODE - if (sbi->s_encoding) + if (sb->s_encoding) sb->s_d_op = &ext4_dentry_ops; #endif @@ -5184,7 +5184,7 @@ failed_mount: crypto_free_shash(sbi->s_chksum_driver); #ifdef CONFIG_UNICODE - utf8_unload(sbi->s_encoding); + utf8_unload(sb->s_encoding); #endif #ifdef CONFIG_QUOTA -- cgit v1.2.3 From 6694875ef8045cdb1e6712ee9b68fe08763507d8 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Wed, 28 Oct 2020 13:39:13 -0400 Subject: ext4: indicate that fast_commit is available via /sys/fs/ext4/feature/... Signed-off-by: Theodore Ts'o --- fs/ext4/sysfs.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c index 5ff33d18996a..4e27fe6ed3ae 100644 --- a/fs/ext4/sysfs.c +++ b/fs/ext4/sysfs.c @@ -315,6 +315,7 @@ EXT4_ATTR_FEATURE(casefold); EXT4_ATTR_FEATURE(verity); #endif EXT4_ATTR_FEATURE(metadata_csum_seed); +EXT4_ATTR_FEATURE(fast_commit); static struct attribute *ext4_feat_attrs[] = { ATTR_LIST(lazy_itable_init), @@ -331,6 +332,7 @@ static struct attribute *ext4_feat_attrs[] = { ATTR_LIST(verity), #endif ATTR_LIST(metadata_csum_seed), + ATTR_LIST(fast_commit), NULL, }; ATTRIBUTE_GROUPS(ext4_feat); -- cgit v1.2.3 From ef5dd6a0c828b6fbd9d595e5772fcb51ff86697e Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Wed, 28 Oct 2020 14:55:24 +0000 Subject: arm64: mte: Document that user PSTATE.TCO is ignored by kernel uaccess On exception entry, the kernel explicitly resets the PSTATE.TCO (tag check override) so that any kernel memory accesses will be checked (the bit is restored on exception return). This has the side-effect that the uaccess routines will not honour the PSTATE.TCO that may have been set by the user prior to a syscall. There is no issue in practice since PSTATE.TCO is expected to be used only for brief periods in specific routines (e.g. garbage collection). To control the tag checking mode of the uaccess routines, the user will have to invoke a corresponding prctl() call. Document the kernel behaviour w.r.t. PSTATE.TCO accordingly. Signed-off-by: Catalin Marinas Fixes: df9d7a22dd21 ("arm64: mte: Add Memory Tagging Extension documentation") Reviewed-by: Vincenzo Frascino Cc: Will Deacon Cc: Szabolcs Nagy Signed-off-by: Will Deacon --- Documentation/arm64/memory-tagging-extension.rst | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/Documentation/arm64/memory-tagging-extension.rst b/Documentation/arm64/memory-tagging-extension.rst index 034d37c605e8..b540178a93f8 100644 --- a/Documentation/arm64/memory-tagging-extension.rst +++ b/Documentation/arm64/memory-tagging-extension.rst @@ -102,7 +102,9 @@ applications. system call) are not checked if the user thread tag checking mode is ``PR_MTE_TCF_NONE`` or ``PR_MTE_TCF_ASYNC``. If the tag checking mode is ``PR_MTE_TCF_SYNC``, the kernel makes a best effort to check its user -address accesses, however it cannot always guarantee it. +address accesses, however it cannot always guarantee it. Kernel accesses +to user addresses are always performed with an effective ``PSTATE.TCO`` +value of zero, regardless of the user configuration. Excluding Tags in the ``IRG``, ``ADDG`` and ``SUBG`` instructions ----------------------------------------------------------------- -- cgit v1.2.3 From 5a169bf04cd2bfdbac967d12eb5b70915b29d7ee Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Mon, 19 Oct 2020 15:55:56 +0100 Subject: x86/kvm: Reserve KVM_FEATURE_MSI_EXT_DEST_ID No functional change; just reserve the feature bit for now so that VMMs can start to implement it. This will allow the host to indicate that MSI emulation supports 15-bit destination IDs, allowing up to 32768 CPUs without interrupt remapping. cf. https://patchwork.kernel.org/patch/11816693/ for qemu Signed-off-by: David Woodhouse Acked-by: Paolo Bonzini Message-Id: <4cd59bed05f4b7410d3d1ffd1e997ab53683874d.camel@infradead.org> Signed-off-by: Paolo Bonzini --- Documentation/virt/kvm/cpuid.rst | 4 ++++ arch/x86/include/uapi/asm/kvm_para.h | 1 + 2 files changed, 5 insertions(+) diff --git a/Documentation/virt/kvm/cpuid.rst b/Documentation/virt/kvm/cpuid.rst index 7d81c0aa4a59..cf62162d4be2 100644 --- a/Documentation/virt/kvm/cpuid.rst +++ b/Documentation/virt/kvm/cpuid.rst @@ -92,6 +92,10 @@ KVM_FEATURE_ASYNC_PF_INT 14 guest checks this feature bit async pf acknowledgment msr 0x4b564d07. +KVM_FEATURE_MSI_EXT_DEST_ID 15 guest checks this feature bit + before using extended destination + ID bits in MSI address bits 11-5. + KVM_FEATURE_CLOCKSOURCE_STABLE_BIT 24 host will warn if no guest-side per-cpu warps are expected in kvmclock diff --git a/arch/x86/include/uapi/asm/kvm_para.h b/arch/x86/include/uapi/asm/kvm_para.h index 812e9b4c1114..950afebfba88 100644 --- a/arch/x86/include/uapi/asm/kvm_para.h +++ b/arch/x86/include/uapi/asm/kvm_para.h @@ -32,6 +32,7 @@ #define KVM_FEATURE_POLL_CONTROL 12 #define KVM_FEATURE_PV_SCHED_YIELD 13 #define KVM_FEATURE_ASYNC_PF_INT 14 +#define KVM_FEATURE_MSI_EXT_DEST_ID 15 #define KVM_HINTS_REALTIME 0 -- cgit v1.2.3 From 80ade22c06ca115b81dd168e99479c8e09843513 Mon Sep 17 00:00:00 2001 From: Sudeep Dutt Date: Tue, 27 Oct 2020 20:14:15 -0700 Subject: misc: mic: remove the MIC drivers This patch removes the MIC drivers from the kernel tree since the corresponding devices have been discontinued. Removing the dma and char-misc changes in one patch and merging via the char-misc tree is best to avoid any potential build breakage. Cc: Nikhil Rao Reviewed-by: Ashutosh Dixit Signed-off-by: Sudeep Dutt Acked-By: Vinod Koul Reviewed-by: Sherry Sun Link: https://lore.kernel.org/r/8c1443136563de34699d2c084df478181c205db4.1603854416.git.sudeep.dutt@intel.com Signed-off-by: Greg Kroah-Hartman --- Documentation/misc-devices/mic/index.rst | 16 - Documentation/misc-devices/mic/mic_overview.rst | 85 - Documentation/misc-devices/mic/scif_overview.rst | 108 -- MAINTAINERS | 16 - drivers/dma/Kconfig | 18 - drivers/dma/Makefile | 1 - drivers/dma/mic_x100_dma.c | 770 --------- drivers/dma/mic_x100_dma.h | 275 --- drivers/misc/Kconfig | 1 - drivers/misc/Makefile | 1 - drivers/misc/mic/Kconfig | 141 -- drivers/misc/mic/Makefile | 12 - drivers/misc/mic/bus/Makefile | 9 - drivers/misc/mic/bus/cosm_bus.c | 130 -- drivers/misc/mic/bus/cosm_bus.h | 125 -- drivers/misc/mic/bus/mic_bus.c | 194 --- drivers/misc/mic/bus/scif_bus.c | 201 --- drivers/misc/mic/bus/scif_bus.h | 125 -- drivers/misc/mic/bus/vop_bus.c | 194 --- drivers/misc/mic/bus/vop_bus.h | 129 -- drivers/misc/mic/card/Makefile | 11 - drivers/misc/mic/card/mic_debugfs.c | 85 - drivers/misc/mic/card/mic_device.c | 417 ----- drivers/misc/mic/card/mic_device.h | 137 -- drivers/misc/mic/card/mic_x100.c | 347 ---- drivers/misc/mic/card/mic_x100.h | 37 - drivers/misc/mic/common/mic_dev.h | 55 - drivers/misc/mic/cosm/Makefile | 11 - drivers/misc/mic/cosm/cosm_debugfs.c | 116 -- drivers/misc/mic/cosm/cosm_main.c | 382 ----- drivers/misc/mic/cosm/cosm_main.h | 61 - drivers/misc/mic/cosm/cosm_scif_server.c | 399 ----- drivers/misc/mic/cosm/cosm_sysfs.c | 449 ----- drivers/misc/mic/cosm_client/Makefile | 8 - drivers/misc/mic/cosm_client/cosm_scif_client.c | 269 --- drivers/misc/mic/host/Makefile | 12 - drivers/misc/mic/host/mic_boot.c | 588 ------- drivers/misc/mic/host/mic_debugfs.c | 149 -- drivers/misc/mic/host/mic_device.h | 157 -- drivers/misc/mic/host/mic_intr.c | 635 ------- drivers/misc/mic/host/mic_intr.h | 137 -- drivers/misc/mic/host/mic_main.c | 335 ---- drivers/misc/mic/host/mic_smpt.c | 427 ----- drivers/misc/mic/host/mic_smpt.h | 87 - drivers/misc/mic/host/mic_x100.c | 585 ------- drivers/misc/mic/host/mic_x100.h | 77 - drivers/misc/mic/scif/Makefile | 21 - drivers/misc/mic/scif/scif_api.c | 1485 ----------------- drivers/misc/mic/scif/scif_debugfs.c | 116 -- drivers/misc/mic/scif/scif_dma.c | 1940 ---------------------- drivers/misc/mic/scif/scif_epd.c | 357 ---- drivers/misc/mic/scif/scif_epd.h | 200 --- drivers/misc/mic/scif/scif_fd.c | 462 ------ drivers/misc/mic/scif/scif_fence.c | 783 --------- drivers/misc/mic/scif/scif_main.c | 351 ---- drivers/misc/mic/scif/scif_main.h | 274 --- drivers/misc/mic/scif/scif_map.h | 127 -- drivers/misc/mic/scif/scif_mmap.c | 690 -------- drivers/misc/mic/scif/scif_nm.c | 229 --- drivers/misc/mic/scif/scif_nodeqp.c | 1349 --------------- drivers/misc/mic/scif/scif_nodeqp.h | 221 --- drivers/misc/mic/scif/scif_peer_bus.c | 175 -- drivers/misc/mic/scif/scif_peer_bus.h | 23 - drivers/misc/mic/scif/scif_ports.c | 116 -- drivers/misc/mic/scif/scif_rb.c | 240 --- drivers/misc/mic/scif/scif_rb.h | 100 -- drivers/misc/mic/scif/scif_rma.c | 1760 -------------------- drivers/misc/mic/scif/scif_rma.h | 477 ------ drivers/misc/mic/scif/scif_rma_list.c | 282 ---- drivers/misc/mic/scif/scif_rma_list.h | 48 - drivers/misc/mic/vop/Makefile | 10 - drivers/misc/mic/vop/vop_debugfs.c | 184 -- drivers/misc/mic/vop/vop_main.c | 784 --------- drivers/misc/mic/vop/vop_main.h | 158 -- drivers/misc/mic/vop/vop_vringh.c | 1166 ------------- include/linux/mic_bus.h | 100 -- include/linux/scif.h | 1339 --------------- include/uapi/linux/mic_common.h | 235 --- include/uapi/linux/mic_ioctl.h | 77 - samples/mic/mpssd/.gitignore | 2 - samples/mic/mpssd/Makefile | 28 - samples/mic/mpssd/micctrl | 162 -- samples/mic/mpssd/mpss | 189 --- samples/mic/mpssd/mpssd.c | 1815 -------------------- samples/mic/mpssd/mpssd.h | 89 - samples/mic/mpssd/sysfs.c | 91 - 86 files changed, 26779 deletions(-) delete mode 100644 Documentation/misc-devices/mic/index.rst delete mode 100644 Documentation/misc-devices/mic/mic_overview.rst delete mode 100644 Documentation/misc-devices/mic/scif_overview.rst delete mode 100644 drivers/dma/mic_x100_dma.c delete mode 100644 drivers/dma/mic_x100_dma.h delete mode 100644 drivers/misc/mic/Kconfig delete mode 100644 drivers/misc/mic/Makefile delete mode 100644 drivers/misc/mic/bus/Makefile delete mode 100644 drivers/misc/mic/bus/cosm_bus.c delete mode 100644 drivers/misc/mic/bus/cosm_bus.h delete mode 100644 drivers/misc/mic/bus/mic_bus.c delete mode 100644 drivers/misc/mic/bus/scif_bus.c delete mode 100644 drivers/misc/mic/bus/scif_bus.h delete mode 100644 drivers/misc/mic/bus/vop_bus.c delete mode 100644 drivers/misc/mic/bus/vop_bus.h delete mode 100644 drivers/misc/mic/card/Makefile delete mode 100644 drivers/misc/mic/card/mic_debugfs.c delete mode 100644 drivers/misc/mic/card/mic_device.c delete mode 100644 drivers/misc/mic/card/mic_device.h delete mode 100644 drivers/misc/mic/card/mic_x100.c delete mode 100644 drivers/misc/mic/card/mic_x100.h delete mode 100644 drivers/misc/mic/common/mic_dev.h delete mode 100644 drivers/misc/mic/cosm/Makefile delete mode 100644 drivers/misc/mic/cosm/cosm_debugfs.c delete mode 100644 drivers/misc/mic/cosm/cosm_main.c delete mode 100644 drivers/misc/mic/cosm/cosm_main.h delete mode 100644 drivers/misc/mic/cosm/cosm_scif_server.c delete mode 100644 drivers/misc/mic/cosm/cosm_sysfs.c delete mode 100644 drivers/misc/mic/cosm_client/Makefile delete mode 100644 drivers/misc/mic/cosm_client/cosm_scif_client.c delete mode 100644 drivers/misc/mic/host/Makefile delete mode 100644 drivers/misc/mic/host/mic_boot.c delete mode 100644 drivers/misc/mic/host/mic_debugfs.c delete mode 100644 drivers/misc/mic/host/mic_device.h delete mode 100644 drivers/misc/mic/host/mic_intr.c delete mode 100644 drivers/misc/mic/host/mic_intr.h delete mode 100644 drivers/misc/mic/host/mic_main.c delete mode 100644 drivers/misc/mic/host/mic_smpt.c delete mode 100644 drivers/misc/mic/host/mic_smpt.h delete mode 100644 drivers/misc/mic/host/mic_x100.c delete mode 100644 drivers/misc/mic/host/mic_x100.h delete mode 100644 drivers/misc/mic/scif/Makefile delete mode 100644 drivers/misc/mic/scif/scif_api.c delete mode 100644 drivers/misc/mic/scif/scif_debugfs.c delete mode 100644 drivers/misc/mic/scif/scif_dma.c delete mode 100644 drivers/misc/mic/scif/scif_epd.c delete mode 100644 drivers/misc/mic/scif/scif_epd.h delete mode 100644 drivers/misc/mic/scif/scif_fd.c delete mode 100644 drivers/misc/mic/scif/scif_fence.c delete mode 100644 drivers/misc/mic/scif/scif_main.c delete mode 100644 drivers/misc/mic/scif/scif_main.h delete mode 100644 drivers/misc/mic/scif/scif_map.h delete mode 100644 drivers/misc/mic/scif/scif_mmap.c delete mode 100644 drivers/misc/mic/scif/scif_nm.c delete mode 100644 drivers/misc/mic/scif/scif_nodeqp.c delete mode 100644 drivers/misc/mic/scif/scif_nodeqp.h delete mode 100644 drivers/misc/mic/scif/scif_peer_bus.c delete mode 100644 drivers/misc/mic/scif/scif_peer_bus.h delete mode 100644 drivers/misc/mic/scif/scif_ports.c delete mode 100644 drivers/misc/mic/scif/scif_rb.c delete mode 100644 drivers/misc/mic/scif/scif_rb.h delete mode 100644 drivers/misc/mic/scif/scif_rma.c delete mode 100644 drivers/misc/mic/scif/scif_rma.h delete mode 100644 drivers/misc/mic/scif/scif_rma_list.c delete mode 100644 drivers/misc/mic/scif/scif_rma_list.h delete mode 100644 drivers/misc/mic/vop/Makefile delete mode 100644 drivers/misc/mic/vop/vop_debugfs.c delete mode 100644 drivers/misc/mic/vop/vop_main.c delete mode 100644 drivers/misc/mic/vop/vop_main.h delete mode 100644 drivers/misc/mic/vop/vop_vringh.c delete mode 100644 include/linux/mic_bus.h delete mode 100644 include/linux/scif.h delete mode 100644 include/uapi/linux/mic_common.h delete mode 100644 include/uapi/linux/mic_ioctl.h delete mode 100644 samples/mic/mpssd/.gitignore delete mode 100644 samples/mic/mpssd/Makefile delete mode 100755 samples/mic/mpssd/micctrl delete mode 100755 samples/mic/mpssd/mpss delete mode 100644 samples/mic/mpssd/mpssd.c delete mode 100644 samples/mic/mpssd/mpssd.h delete mode 100644 samples/mic/mpssd/sysfs.c diff --git a/Documentation/misc-devices/mic/index.rst b/Documentation/misc-devices/mic/index.rst deleted file mode 100644 index 3a8d06367ef1..000000000000 --- a/Documentation/misc-devices/mic/index.rst +++ /dev/null @@ -1,16 +0,0 @@ -============================================= -Intel Many Integrated Core (MIC) architecture -============================================= - -.. toctree:: - :maxdepth: 1 - - mic_overview - scif_overview - -.. only:: subproject and html - - Indices - ======= - - * :ref:`genindex` diff --git a/Documentation/misc-devices/mic/mic_overview.rst b/Documentation/misc-devices/mic/mic_overview.rst deleted file mode 100644 index 17d956bdaf7c..000000000000 --- a/Documentation/misc-devices/mic/mic_overview.rst +++ /dev/null @@ -1,85 +0,0 @@ -====================================================== -Intel Many Integrated Core (MIC) architecture overview -====================================================== - -An Intel MIC X100 device is a PCIe form factor add-in coprocessor -card based on the Intel Many Integrated Core (MIC) architecture -that runs a Linux OS. It is a PCIe endpoint in a platform and therefore -implements the three required standard address spaces i.e. configuration, -memory and I/O. The host OS loads a device driver as is typical for -PCIe devices. The card itself runs a bootstrap after reset that -transfers control to the card OS downloaded from the host driver. The -host driver supports OSPM suspend and resume operations. It shuts down -the card during suspend and reboots the card OS during resume. -The card OS as shipped by Intel is a Linux kernel with modifications -for the X100 devices. - -Since it is a PCIe card, it does not have the ability to host hardware -devices for networking, storage and console. We provide these devices -on X100 coprocessors thus enabling a self-bootable equivalent -environment for applications. A key benefit of our solution is that it -leverages the standard virtio framework for network, disk and console -devices, though in our case the virtio framework is used across a PCIe -bus. A Virtio Over PCIe (VOP) driver allows creating user space -backends or devices on the host which are used to probe virtio drivers -for these devices on the MIC card. The existing VRINGH infrastructure -in the kernel is used to access virtio rings from the host. The card -VOP driver allows card virtio drivers to communicate with their user -space backends on the host via a device page. Ring 3 apps on the host -can add, remove and configure virtio devices. A thin MIC specific -virtio_config_ops is implemented which is borrowed heavily from -previous similar implementations in lguest and s390. - -MIC PCIe card has a dma controller with 8 channels. These channels are -shared between the host s/w and the card s/w. 0 to 3 are used by host -and 4 to 7 by card. As the dma device doesn't show up as PCIe device, -a virtual bus called mic bus is created and virtual dma devices are -created on it by the host/card drivers. On host the channels are private -and used only by the host driver to transfer data for the virtio devices. - -The Symmetric Communication Interface (SCIF (pronounced as skiff)) is a -low level communications API across PCIe currently implemented for MIC. -More details are available at scif_overview.txt. - -The Coprocessor State Management (COSM) driver on the host allows for -boot, shutdown and reset of Intel MIC devices. It communicates with a COSM -"client" driver on the MIC cards over SCIF to perform these functions. - -Here is a block diagram of the various components described above. The -virtio backends are situated on the host rather than the card given better -single threaded performance for the host compared to MIC, the ability of -the host to initiate DMA's to/from the card using the MIC DMA engine and -the fact that the virtio block storage backend can only be on the host:: - - +----------+ | +----------+ - | Card OS | | | Host OS | - +----------+ | +----------+ - | - +-------+ +--------+ +------+ | +---------+ +--------+ +--------+ - | Virtio| |Virtio | |Virtio| | |Virtio | |Virtio | |Virtio | - | Net | |Console | |Block | | |Net | |Console | |Block | - | Driver| |Driver | |Driver| | |backend | |backend | |backend | - +---+---+ +---+----+ +--+---+ | +---------+ +----+---+ +--------+ - | | | | | | | - | | | |User | | | - | | | |------|------------|--+------|------- - +---------+---------+ |Kernel | - | | | - +---------+ +---+----+ +------+ | +------+ +------+ +--+---+ +-------+ - |MIC DMA | | VOP | | SCIF | | | SCIF | | COSM | | VOP | |MIC DMA| - +---+-----+ +---+----+ +--+---+ | +--+---+ +--+---+ +------+ +----+--+ - | | | | | | | - +---+-----+ +---+----+ +--+---+ | +--+---+ +--+---+ +------+ +----+--+ - |MIC | | VOP | |SCIF | | |SCIF | | COSM | | VOP | | MIC | - |HW Bus | | HW Bus| |HW Bus| | |HW Bus| | Bus | |HW Bus| |HW Bus | - +---------+ +--------+ +--+---+ | +--+---+ +------+ +------+ +-------+ - | | | | | | | - | +-----------+--+ | | | +---------------+ | - | |Intel MIC | | | | |Intel MIC | | - | |Card Driver | | | | |Host Driver | | - +---+--------------+------+ | +----+---------------+-----+ - | | | - +-------------------------------------------------------------+ - | | - | PCIe Bus | - +-------------------------------------------------------------+ diff --git a/Documentation/misc-devices/mic/scif_overview.rst b/Documentation/misc-devices/mic/scif_overview.rst deleted file mode 100644 index 4c8ad9e43706..000000000000 --- a/Documentation/misc-devices/mic/scif_overview.rst +++ /dev/null @@ -1,108 +0,0 @@ -======================================== -Symmetric Communication Interface (SCIF) -======================================== - -The Symmetric Communication Interface (SCIF (pronounced as skiff)) is a low -level communications API across PCIe currently implemented for MIC. Currently -SCIF provides inter-node communication within a single host platform, where a -node is a MIC Coprocessor or Xeon based host. SCIF abstracts the details of -communicating over the PCIe bus while providing an API that is symmetric -across all the nodes in the PCIe network. An important design objective for SCIF -is to deliver the maximum possible performance given the communication -abilities of the hardware. SCIF has been used to implement an offload compiler -runtime and OFED support for MPI implementations for MIC coprocessors. - -SCIF API Components -=================== - -The SCIF API has the following parts: - -1. Connection establishment using a client server model -2. Byte stream messaging intended for short messages -3. Node enumeration to determine online nodes -4. Poll semantics for detection of incoming connections and messages -5. Memory registration to pin down pages -6. Remote memory mapping for low latency CPU accesses via mmap -7. Remote DMA (RDMA) for high bandwidth DMA transfers -8. Fence APIs for RDMA synchronization - -SCIF exposes the notion of a connection which can be used by peer processes on -nodes in a SCIF PCIe "network" to share memory "windows" and to communicate. A -process in a SCIF node initiates a SCIF connection to a peer process on a -different node via a SCIF "endpoint". SCIF endpoints support messaging APIs -which are similar to connection oriented socket APIs. Connected SCIF endpoints -can also register local memory which is followed by data transfer using either -DMA, CPU copies or remote memory mapping via mmap. SCIF supports both user and -kernel mode clients which are functionally equivalent. - -SCIF Performance for MIC -======================== - -DMA bandwidth comparison between the TCP (over ethernet over PCIe) stack versus -SCIF shows the performance advantages of SCIF for HPC applications and -runtimes:: - - Comparison of TCP and SCIF based BW - - Throughput (GB/sec) - 8 + PCIe Bandwidth ****** - + TCP ###### - 7 + ************************************** SCIF %%%%%% - | %%%%%%%%%%%%%%%%%%% - 6 + %%%% - | %% - | %%% - 5 + %% - | %% - 4 + %% - | %% - 3 + %% - | % - 2 + %% - | %% - | % - 1 + - + ###################################### - 0 +++---+++--+--+-+--+--+-++-+--+-++-+--+-++-+- - 1 10 100 1000 10000 100000 - Transfer Size (KBytes) - -SCIF allows memory sharing via mmap(..) between processes on different PCIe -nodes and thus provides bare-metal PCIe latency. The round trip SCIF mmap -latency from the host to an x100 MIC for an 8 byte message is 0.44 usecs. - -SCIF has a user space library which is a thin IOCTL wrapper providing a user -space API similar to the kernel API in scif.h. The SCIF user space library -is distributed @ https://software.intel.com/en-us/mic-developer - -Here is some pseudo code for an example of how two applications on two PCIe -nodes would typically use the SCIF API:: - - Process A (on node A) Process B (on node B) - - /* get online node information */ - scif_get_node_ids(..) scif_get_node_ids(..) - scif_open(..) scif_open(..) - scif_bind(..) scif_bind(..) - scif_listen(..) - scif_accept(..) scif_connect(..) - /* SCIF connection established */ - - /* Send and receive short messages */ - scif_send(..)/scif_recv(..) scif_send(..)/scif_recv(..) - - /* Register memory */ - scif_register(..) scif_register(..) - - /* RDMA */ - scif_readfrom(..)/scif_writeto(..) scif_readfrom(..)/scif_writeto(..) - - /* Fence DMAs */ - scif_fence_signal(..) scif_fence_signal(..) - - mmap(..) mmap(..) - - /* Access remote registered memory */ - - /* Close the endpoints */ - scif_close(..) scif_close(..) diff --git a/MAINTAINERS b/MAINTAINERS index e73636b75f29..9289a9b43a51 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -8976,22 +8976,6 @@ S: Supported W: https://01.org/linux-acpi F: drivers/platform/x86/intel_menlow.c -INTEL MIC DRIVERS (mic) -M: Sudeep Dutt -M: Ashutosh Dixit -S: Supported -W: https://github.com/sudeepdutt/mic -W: http://software.intel.com/en-us/mic-developer -F: Documentation/misc-devices/mic/ -F: drivers/dma/mic_x100_dma.c -F: drivers/dma/mic_x100_dma.h -F: drivers/misc/mic/ -F: include/linux/mic_bus.h -F: include/linux/scif.h -F: include/uapi/linux/mic_common.h -F: include/uapi/linux/mic_ioctl.h -F: include/uapi/linux/scif_ioctl.h - INTEL P-Unit IPC DRIVER M: Zha Qipeng L: platform-driver-x86@vger.kernel.org diff --git a/drivers/dma/Kconfig b/drivers/dma/Kconfig index 518a1437862a..90284ffda58a 100644 --- a/drivers/dma/Kconfig +++ b/drivers/dma/Kconfig @@ -318,24 +318,6 @@ config INTEL_IOP_ADMA help Enable support for the Intel(R) IOP Series RAID engines. -config INTEL_MIC_X100_DMA - tristate "Intel MIC X100 DMA Driver" - depends on 64BIT && X86 && INTEL_MIC_BUS - select DMA_ENGINE - help - This enables DMA support for the Intel Many Integrated Core - (MIC) family of PCIe form factor coprocessor X100 devices that - run a 64 bit Linux OS. This driver will be used by both MIC - host and card drivers. - - If you are building host kernel with a MIC device or a card - kernel for a MIC device, then say M (recommended) or Y, else - say N. If unsure say N. - - More information about the Intel MIC family as well as the Linux - OS and tools for MIC to use with this driver are available from - . - config K3_DMA tristate "Hisilicon K3 DMA support" depends on ARCH_HI3xxx || ARCH_HISI || COMPILE_TEST diff --git a/drivers/dma/Makefile b/drivers/dma/Makefile index e60f81331d4c..948a8da05f8b 100644 --- a/drivers/dma/Makefile +++ b/drivers/dma/Makefile @@ -44,7 +44,6 @@ obj-$(CONFIG_INTEL_IDMA64) += idma64.o obj-$(CONFIG_INTEL_IOATDMA) += ioat/ obj-$(CONFIG_INTEL_IDXD) += idxd/ obj-$(CONFIG_INTEL_IOP_ADMA) += iop-adma.o -obj-$(CONFIG_INTEL_MIC_X100_DMA) += mic_x100_dma.o obj-$(CONFIG_K3_DMA) += k3dma.o obj-$(CONFIG_LPC18XX_DMAMUX) += lpc18xx-dmamux.o obj-$(CONFIG_MILBEAUT_HDMAC) += milbeaut-hdmac.o diff --git a/drivers/dma/mic_x100_dma.c b/drivers/dma/mic_x100_dma.c deleted file mode 100644 index fea8608a7810..000000000000 --- a/drivers/dma/mic_x100_dma.c +++ /dev/null @@ -1,770 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel MIC X100 DMA Driver. - * - * Adapted from IOAT dma driver. - */ -#include -#include -#include -#include - -#include "mic_x100_dma.h" - -#define MIC_DMA_MAX_XFER_SIZE_CARD (1 * 1024 * 1024 -\ - MIC_DMA_ALIGN_BYTES) -#define MIC_DMA_MAX_XFER_SIZE_HOST (1 * 1024 * 1024 >> 1) -#define MIC_DMA_DESC_TYPE_SHIFT 60 -#define MIC_DMA_MEMCPY_LEN_SHIFT 46 -#define MIC_DMA_STAT_INTR_SHIFT 59 - -/* high-water mark for pushing dma descriptors */ -static int mic_dma_pending_level = 4; - -/* Status descriptor is used to write a 64 bit value to a memory location */ -enum mic_dma_desc_format_type { - MIC_DMA_MEMCPY = 1, - MIC_DMA_STATUS, -}; - -static inline u32 mic_dma_hw_ring_inc(u32 val) -{ - return (val + 1) % MIC_DMA_DESC_RX_SIZE; -} - -static inline u32 mic_dma_hw_ring_dec(u32 val) -{ - return val ? val - 1 : MIC_DMA_DESC_RX_SIZE - 1; -} - -static inline void mic_dma_hw_ring_inc_head(struct mic_dma_chan *ch) -{ - ch->head = mic_dma_hw_ring_inc(ch->head); -} - -/* Prepare a memcpy desc */ -static inline void mic_dma_memcpy_desc(struct mic_dma_desc *desc, - dma_addr_t src_phys, dma_addr_t dst_phys, u64 size) -{ - u64 qw0, qw1; - - qw0 = src_phys; - qw0 |= (size >> MIC_DMA_ALIGN_SHIFT) << MIC_DMA_MEMCPY_LEN_SHIFT; - qw1 = MIC_DMA_MEMCPY; - qw1 <<= MIC_DMA_DESC_TYPE_SHIFT; - qw1 |= dst_phys; - desc->qw0 = qw0; - desc->qw1 = qw1; -} - -/* Prepare a status desc. with @data to be written at @dst_phys */ -static inline void mic_dma_prep_status_desc(struct mic_dma_desc *desc, u64 data, - dma_addr_t dst_phys, bool generate_intr) -{ - u64 qw0, qw1; - - qw0 = data; - qw1 = (u64) MIC_DMA_STATUS << MIC_DMA_DESC_TYPE_SHIFT | dst_phys; - if (generate_intr) - qw1 |= (1ULL << MIC_DMA_STAT_INTR_SHIFT); - desc->qw0 = qw0; - desc->qw1 = qw1; -} - -static void mic_dma_cleanup(struct mic_dma_chan *ch) -{ - struct dma_async_tx_descriptor *tx; - u32 tail; - u32 last_tail; - - spin_lock(&ch->cleanup_lock); - tail = mic_dma_read_cmp_cnt(ch); - /* - * This is the barrier pair for smp_wmb() in fn. - * mic_dma_tx_submit_unlock. It's required so that we read the - * updated cookie value from tx->cookie. - */ - smp_rmb(); - for (last_tail = ch->last_tail; tail != last_tail;) { - tx = &ch->tx_array[last_tail]; - if (tx->cookie) { - dma_cookie_complete(tx); - dmaengine_desc_get_callback_invoke(tx, NULL); - tx->callback = NULL; - } - last_tail = mic_dma_hw_ring_inc(last_tail); - } - /* finish all completion callbacks before incrementing tail */ - smp_mb(); - ch->last_tail = last_tail; - spin_unlock(&ch->cleanup_lock); -} - -static u32 mic_dma_ring_count(u32 head, u32 tail) -{ - u32 count; - - if (head >= tail) - count = (tail - 0) + (MIC_DMA_DESC_RX_SIZE - head); - else - count = tail - head; - return count - 1; -} - -/* Returns the num. of free descriptors on success, -ENOMEM on failure */ -static int mic_dma_avail_desc_ring_space(struct mic_dma_chan *ch, int required) -{ - struct device *dev = mic_dma_ch_to_device(ch); - u32 count; - - count = mic_dma_ring_count(ch->head, ch->last_tail); - if (count < required) { - mic_dma_cleanup(ch); - count = mic_dma_ring_count(ch->head, ch->last_tail); - } - - if (count < required) { - dev_dbg(dev, "Not enough desc space"); - dev_dbg(dev, "%s %d required=%u, avail=%u\n", - __func__, __LINE__, required, count); - return -ENOMEM; - } else { - return count; - } -} - -/* Program memcpy descriptors into the descriptor ring and update s/w head ptr*/ -static int mic_dma_prog_memcpy_desc(struct mic_dma_chan *ch, dma_addr_t src, - dma_addr_t dst, size_t len) -{ - size_t current_transfer_len; - size_t max_xfer_size = to_mic_dma_dev(ch)->max_xfer_size; - /* 3 is added to make sure we have enough space for status desc */ - int num_desc = len / max_xfer_size + 3; - int ret; - - if (len % max_xfer_size) - num_desc++; - - ret = mic_dma_avail_desc_ring_space(ch, num_desc); - if (ret < 0) - return ret; - do { - current_transfer_len = min(len, max_xfer_size); - mic_dma_memcpy_desc(&ch->desc_ring[ch->head], - src, dst, current_transfer_len); - mic_dma_hw_ring_inc_head(ch); - len -= current_transfer_len; - dst = dst + current_transfer_len; - src = src + current_transfer_len; - } while (len > 0); - return 0; -} - -/* It's a h/w quirk and h/w needs 2 status descriptors for every status desc */ -static void mic_dma_prog_intr(struct mic_dma_chan *ch) -{ - mic_dma_prep_status_desc(&ch->desc_ring[ch->head], 0, - ch->status_dest_micpa, false); - mic_dma_hw_ring_inc_head(ch); - mic_dma_prep_status_desc(&ch->desc_ring[ch->head], 0, - ch->status_dest_micpa, true); - mic_dma_hw_ring_inc_head(ch); -} - -/* Wrapper function to program memcpy descriptors/status descriptors */ -static int mic_dma_do_dma(struct mic_dma_chan *ch, int flags, dma_addr_t src, - dma_addr_t dst, size_t len) -{ - if (len && -ENOMEM == mic_dma_prog_memcpy_desc(ch, src, dst, len)) { - return -ENOMEM; - } else { - /* 3 is the maximum number of status descriptors */ - int ret = mic_dma_avail_desc_ring_space(ch, 3); - - if (ret < 0) - return ret; - } - - /* Above mic_dma_prog_memcpy_desc() makes sure we have enough space */ - if (flags & DMA_PREP_FENCE) { - mic_dma_prep_status_desc(&ch->desc_ring[ch->head], 0, - ch->status_dest_micpa, false); - mic_dma_hw_ring_inc_head(ch); - } - - if (flags & DMA_PREP_INTERRUPT) - mic_dma_prog_intr(ch); - - return 0; -} - -static inline void mic_dma_issue_pending(struct dma_chan *ch) -{ - struct mic_dma_chan *mic_ch = to_mic_dma_chan(ch); - - spin_lock(&mic_ch->issue_lock); - /* - * Write to head triggers h/w to act on the descriptors. - * On MIC, writing the same head value twice causes - * a h/w error. On second write, h/w assumes we filled - * the entire ring & overwrote some of the descriptors. - */ - if (mic_ch->issued == mic_ch->submitted) - goto out; - mic_ch->issued = mic_ch->submitted; - /* - * make descriptor updates visible before advancing head, - * this is purposefully not smp_wmb() since we are also - * publishing the descriptor updates to a dma device - */ - wmb(); - mic_dma_write_reg(mic_ch, MIC_DMA_REG_DHPR, mic_ch->issued); -out: - spin_unlock(&mic_ch->issue_lock); -} - -static inline void mic_dma_update_pending(struct mic_dma_chan *ch) -{ - if (mic_dma_ring_count(ch->issued, ch->submitted) - > mic_dma_pending_level) - mic_dma_issue_pending(&ch->api_ch); -} - -static dma_cookie_t mic_dma_tx_submit_unlock(struct dma_async_tx_descriptor *tx) -{ - struct mic_dma_chan *mic_ch = to_mic_dma_chan(tx->chan); - dma_cookie_t cookie; - - dma_cookie_assign(tx); - cookie = tx->cookie; - /* - * We need an smp write barrier here because another CPU might see - * an update to submitted and update h/w head even before we - * assigned a cookie to this tx. - */ - smp_wmb(); - mic_ch->submitted = mic_ch->head; - spin_unlock(&mic_ch->prep_lock); - mic_dma_update_pending(mic_ch); - return cookie; -} - -static inline struct dma_async_tx_descriptor * -allocate_tx(struct mic_dma_chan *ch) -{ - u32 idx = mic_dma_hw_ring_dec(ch->head); - struct dma_async_tx_descriptor *tx = &ch->tx_array[idx]; - - dma_async_tx_descriptor_init(tx, &ch->api_ch); - tx->tx_submit = mic_dma_tx_submit_unlock; - return tx; -} - -/* Program a status descriptor with dst as address and value to be written */ -static struct dma_async_tx_descriptor * -mic_dma_prep_status_lock(struct dma_chan *ch, dma_addr_t dst, u64 src_val, - unsigned long flags) -{ - struct mic_dma_chan *mic_ch = to_mic_dma_chan(ch); - int result; - - spin_lock(&mic_ch->prep_lock); - result = mic_dma_avail_desc_ring_space(mic_ch, 4); - if (result < 0) - goto error; - mic_dma_prep_status_desc(&mic_ch->desc_ring[mic_ch->head], src_val, dst, - false); - mic_dma_hw_ring_inc_head(mic_ch); - result = mic_dma_do_dma(mic_ch, flags, 0, 0, 0); - if (result < 0) - goto error; - - return allocate_tx(mic_ch); -error: - dev_err(mic_dma_ch_to_device(mic_ch), - "Error enqueueing dma status descriptor, error=%d\n", result); - spin_unlock(&mic_ch->prep_lock); - return NULL; -} - -/* - * Prepare a memcpy descriptor to be added to the ring. - * Note that the temporary descriptor adds an extra overhead of copying the - * descriptor to ring. So, we copy directly to the descriptor ring - */ -static struct dma_async_tx_descriptor * -mic_dma_prep_memcpy_lock(struct dma_chan *ch, dma_addr_t dma_dest, - dma_addr_t dma_src, size_t len, unsigned long flags) -{ - struct mic_dma_chan *mic_ch = to_mic_dma_chan(ch); - struct device *dev = mic_dma_ch_to_device(mic_ch); - int result; - - if (!len && !flags) - return NULL; - - spin_lock(&mic_ch->prep_lock); - result = mic_dma_do_dma(mic_ch, flags, dma_src, dma_dest, len); - if (result >= 0) - return allocate_tx(mic_ch); - dev_err(dev, "Error enqueueing dma, error=%d\n", result); - spin_unlock(&mic_ch->prep_lock); - return NULL; -} - -static struct dma_async_tx_descriptor * -mic_dma_prep_interrupt_lock(struct dma_chan *ch, unsigned long flags) -{ - struct mic_dma_chan *mic_ch = to_mic_dma_chan(ch); - int ret; - - spin_lock(&mic_ch->prep_lock); - ret = mic_dma_do_dma(mic_ch, flags, 0, 0, 0); - if (!ret) - return allocate_tx(mic_ch); - spin_unlock(&mic_ch->prep_lock); - return NULL; -} - -/* Return the status of the transaction */ -static enum dma_status -mic_dma_tx_status(struct dma_chan *ch, dma_cookie_t cookie, - struct dma_tx_state *txstate) -{ - struct mic_dma_chan *mic_ch = to_mic_dma_chan(ch); - - if (DMA_COMPLETE != dma_cookie_status(ch, cookie, txstate)) - mic_dma_cleanup(mic_ch); - - return dma_cookie_status(ch, cookie, txstate); -} - -static irqreturn_t mic_dma_thread_fn(int irq, void *data) -{ - mic_dma_cleanup((struct mic_dma_chan *)data); - return IRQ_HANDLED; -} - -static irqreturn_t mic_dma_intr_handler(int irq, void *data) -{ - struct mic_dma_chan *ch = ((struct mic_dma_chan *)data); - - mic_dma_ack_interrupt(ch); - return IRQ_WAKE_THREAD; -} - -static int mic_dma_alloc_desc_ring(struct mic_dma_chan *ch) -{ - u64 desc_ring_size = MIC_DMA_DESC_RX_SIZE * sizeof(*ch->desc_ring); - struct device *dev = &to_mbus_device(ch)->dev; - - desc_ring_size = ALIGN(desc_ring_size, MIC_DMA_ALIGN_BYTES); - ch->desc_ring = kzalloc(desc_ring_size, GFP_KERNEL); - - if (!ch->desc_ring) - return -ENOMEM; - - ch->desc_ring_micpa = dma_map_single(dev, ch->desc_ring, - desc_ring_size, DMA_BIDIRECTIONAL); - if (dma_mapping_error(dev, ch->desc_ring_micpa)) - goto map_error; - - ch->tx_array = vzalloc(array_size(MIC_DMA_DESC_RX_SIZE, - sizeof(*ch->tx_array))); - if (!ch->tx_array) - goto tx_error; - return 0; -tx_error: - dma_unmap_single(dev, ch->desc_ring_micpa, desc_ring_size, - DMA_BIDIRECTIONAL); -map_error: - kfree(ch->desc_ring); - return -ENOMEM; -} - -static void mic_dma_free_desc_ring(struct mic_dma_chan *ch) -{ - u64 desc_ring_size = MIC_DMA_DESC_RX_SIZE * sizeof(*ch->desc_ring); - - vfree(ch->tx_array); - desc_ring_size = ALIGN(desc_ring_size, MIC_DMA_ALIGN_BYTES); - dma_unmap_single(&to_mbus_device(ch)->dev, ch->desc_ring_micpa, - desc_ring_size, DMA_BIDIRECTIONAL); - kfree(ch->desc_ring); - ch->desc_ring = NULL; -} - -static void mic_dma_free_status_dest(struct mic_dma_chan *ch) -{ - dma_unmap_single(&to_mbus_device(ch)->dev, ch->status_dest_micpa, - L1_CACHE_BYTES, DMA_BIDIRECTIONAL); - kfree(ch->status_dest); -} - -static int mic_dma_alloc_status_dest(struct mic_dma_chan *ch) -{ - struct device *dev = &to_mbus_device(ch)->dev; - - ch->status_dest = kzalloc(L1_CACHE_BYTES, GFP_KERNEL); - if (!ch->status_dest) - return -ENOMEM; - ch->status_dest_micpa = dma_map_single(dev, ch->status_dest, - L1_CACHE_BYTES, DMA_BIDIRECTIONAL); - if (dma_mapping_error(dev, ch->status_dest_micpa)) { - kfree(ch->status_dest); - ch->status_dest = NULL; - return -ENOMEM; - } - return 0; -} - -static int mic_dma_check_chan(struct mic_dma_chan *ch) -{ - if (mic_dma_read_reg(ch, MIC_DMA_REG_DCHERR) || - mic_dma_read_reg(ch, MIC_DMA_REG_DSTAT) & MIC_DMA_CHAN_QUIESCE) { - mic_dma_disable_chan(ch); - mic_dma_chan_mask_intr(ch); - dev_err(mic_dma_ch_to_device(ch), - "%s %d error setting up mic dma chan %d\n", - __func__, __LINE__, ch->ch_num); - return -EBUSY; - } - return 0; -} - -static int mic_dma_chan_setup(struct mic_dma_chan *ch) -{ - if (MIC_DMA_CHAN_MIC == ch->owner) - mic_dma_chan_set_owner(ch); - mic_dma_disable_chan(ch); - mic_dma_chan_mask_intr(ch); - mic_dma_write_reg(ch, MIC_DMA_REG_DCHERRMSK, 0); - mic_dma_chan_set_desc_ring(ch); - ch->last_tail = mic_dma_read_reg(ch, MIC_DMA_REG_DTPR); - ch->head = ch->last_tail; - ch->issued = 0; - mic_dma_chan_unmask_intr(ch); - mic_dma_enable_chan(ch); - return mic_dma_check_chan(ch); -} - -static void mic_dma_chan_destroy(struct mic_dma_chan *ch) -{ - mic_dma_disable_chan(ch); - mic_dma_chan_mask_intr(ch); -} - -static int mic_dma_setup_irq(struct mic_dma_chan *ch) -{ - ch->cookie = - to_mbus_hw_ops(ch)->request_threaded_irq(to_mbus_device(ch), - mic_dma_intr_handler, mic_dma_thread_fn, - "mic dma_channel", ch, ch->ch_num); - return PTR_ERR_OR_ZERO(ch->cookie); -} - -static inline void mic_dma_free_irq(struct mic_dma_chan *ch) -{ - to_mbus_hw_ops(ch)->free_irq(to_mbus_device(ch), ch->cookie, ch); -} - -static int mic_dma_chan_init(struct mic_dma_chan *ch) -{ - int ret = mic_dma_alloc_desc_ring(ch); - - if (ret) - goto ring_error; - ret = mic_dma_alloc_status_dest(ch); - if (ret) - goto status_error; - ret = mic_dma_chan_setup(ch); - if (ret) - goto chan_error; - return ret; -chan_error: - mic_dma_free_status_dest(ch); -status_error: - mic_dma_free_desc_ring(ch); -ring_error: - return ret; -} - -static int mic_dma_drain_chan(struct mic_dma_chan *ch) -{ - struct dma_async_tx_descriptor *tx; - int err = 0; - dma_cookie_t cookie; - - tx = mic_dma_prep_memcpy_lock(&ch->api_ch, 0, 0, 0, DMA_PREP_FENCE); - if (!tx) { - err = -ENOMEM; - goto error; - } - - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) - err = -ENOMEM; - else - err = dma_sync_wait(&ch->api_ch, cookie); - if (err) { - dev_err(mic_dma_ch_to_device(ch), "%s %d TO chan 0x%x\n", - __func__, __LINE__, ch->ch_num); - err = -EIO; - } -error: - mic_dma_cleanup(ch); - return err; -} - -static inline void mic_dma_chan_uninit(struct mic_dma_chan *ch) -{ - mic_dma_chan_destroy(ch); - mic_dma_cleanup(ch); - mic_dma_free_status_dest(ch); - mic_dma_free_desc_ring(ch); -} - -static int mic_dma_init(struct mic_dma_device *mic_dma_dev, - enum mic_dma_chan_owner owner) -{ - int i, first_chan = mic_dma_dev->start_ch; - struct mic_dma_chan *ch; - int ret; - - for (i = first_chan; i < first_chan + MIC_DMA_NUM_CHAN; i++) { - ch = &mic_dma_dev->mic_ch[i]; - ch->ch_num = i; - ch->owner = owner; - spin_lock_init(&ch->cleanup_lock); - spin_lock_init(&ch->prep_lock); - spin_lock_init(&ch->issue_lock); - ret = mic_dma_setup_irq(ch); - if (ret) - goto error; - } - return 0; -error: - for (i = i - 1; i >= first_chan; i--) - mic_dma_free_irq(ch); - return ret; -} - -static void mic_dma_uninit(struct mic_dma_device *mic_dma_dev) -{ - int i, first_chan = mic_dma_dev->start_ch; - struct mic_dma_chan *ch; - - for (i = first_chan; i < first_chan + MIC_DMA_NUM_CHAN; i++) { - ch = &mic_dma_dev->mic_ch[i]; - mic_dma_free_irq(ch); - } -} - -static int mic_dma_alloc_chan_resources(struct dma_chan *ch) -{ - int ret = mic_dma_chan_init(to_mic_dma_chan(ch)); - if (ret) - return ret; - return MIC_DMA_DESC_RX_SIZE; -} - -static void mic_dma_free_chan_resources(struct dma_chan *ch) -{ - struct mic_dma_chan *mic_ch = to_mic_dma_chan(ch); - mic_dma_drain_chan(mic_ch); - mic_dma_chan_uninit(mic_ch); -} - -/* Set the fn. handlers and register the dma device with dma api */ -static int mic_dma_register_dma_device(struct mic_dma_device *mic_dma_dev, - enum mic_dma_chan_owner owner) -{ - int i, first_chan = mic_dma_dev->start_ch; - - dma_cap_zero(mic_dma_dev->dma_dev.cap_mask); - /* - * This dma engine is not capable of host memory to host memory - * transfers - */ - dma_cap_set(DMA_MEMCPY, mic_dma_dev->dma_dev.cap_mask); - - if (MIC_DMA_CHAN_HOST == owner) - dma_cap_set(DMA_PRIVATE, mic_dma_dev->dma_dev.cap_mask); - mic_dma_dev->dma_dev.device_alloc_chan_resources = - mic_dma_alloc_chan_resources; - mic_dma_dev->dma_dev.device_free_chan_resources = - mic_dma_free_chan_resources; - mic_dma_dev->dma_dev.device_tx_status = mic_dma_tx_status; - mic_dma_dev->dma_dev.device_prep_dma_memcpy = mic_dma_prep_memcpy_lock; - mic_dma_dev->dma_dev.device_prep_dma_imm_data = - mic_dma_prep_status_lock; - mic_dma_dev->dma_dev.device_prep_dma_interrupt = - mic_dma_prep_interrupt_lock; - mic_dma_dev->dma_dev.device_issue_pending = mic_dma_issue_pending; - mic_dma_dev->dma_dev.copy_align = MIC_DMA_ALIGN_SHIFT; - INIT_LIST_HEAD(&mic_dma_dev->dma_dev.channels); - for (i = first_chan; i < first_chan + MIC_DMA_NUM_CHAN; i++) { - mic_dma_dev->mic_ch[i].api_ch.device = &mic_dma_dev->dma_dev; - dma_cookie_init(&mic_dma_dev->mic_ch[i].api_ch); - list_add_tail(&mic_dma_dev->mic_ch[i].api_ch.device_node, - &mic_dma_dev->dma_dev.channels); - } - return dmaenginem_async_device_register(&mic_dma_dev->dma_dev); -} - -/* - * Initializes dma channels and registers the dma device with the - * dma engine api. - */ -static struct mic_dma_device *mic_dma_dev_reg(struct mbus_device *mbdev, - enum mic_dma_chan_owner owner) -{ - struct mic_dma_device *mic_dma_dev; - int ret; - struct device *dev = &mbdev->dev; - - mic_dma_dev = devm_kzalloc(dev, sizeof(*mic_dma_dev), GFP_KERNEL); - if (!mic_dma_dev) { - ret = -ENOMEM; - goto alloc_error; - } - mic_dma_dev->mbdev = mbdev; - mic_dma_dev->dma_dev.dev = dev; - mic_dma_dev->mmio = mbdev->mmio_va; - if (MIC_DMA_CHAN_HOST == owner) { - mic_dma_dev->start_ch = 0; - mic_dma_dev->max_xfer_size = MIC_DMA_MAX_XFER_SIZE_HOST; - } else { - mic_dma_dev->start_ch = 4; - mic_dma_dev->max_xfer_size = MIC_DMA_MAX_XFER_SIZE_CARD; - } - ret = mic_dma_init(mic_dma_dev, owner); - if (ret) - goto init_error; - ret = mic_dma_register_dma_device(mic_dma_dev, owner); - if (ret) - goto reg_error; - return mic_dma_dev; -reg_error: - mic_dma_uninit(mic_dma_dev); -init_error: - mic_dma_dev = NULL; -alloc_error: - dev_err(dev, "Error at %s %d ret=%d\n", __func__, __LINE__, ret); - return mic_dma_dev; -} - -static void mic_dma_dev_unreg(struct mic_dma_device *mic_dma_dev) -{ - mic_dma_uninit(mic_dma_dev); -} - -/* DEBUGFS CODE */ -static int mic_dma_reg_show(struct seq_file *s, void *pos) -{ - struct mic_dma_device *mic_dma_dev = s->private; - int i, chan_num, first_chan = mic_dma_dev->start_ch; - struct mic_dma_chan *ch; - - seq_printf(s, "SBOX_DCR: %#x\n", - mic_dma_mmio_read(&mic_dma_dev->mic_ch[first_chan], - MIC_DMA_SBOX_BASE + MIC_DMA_SBOX_DCR)); - seq_puts(s, "DMA Channel Registers\n"); - seq_printf(s, "%-10s| %-10s %-10s %-10s %-10s %-10s", - "Channel", "DCAR", "DTPR", "DHPR", "DRAR_HI", "DRAR_LO"); - seq_printf(s, " %-11s %-14s %-10s\n", "DCHERR", "DCHERRMSK", "DSTAT"); - for (i = first_chan; i < first_chan + MIC_DMA_NUM_CHAN; i++) { - ch = &mic_dma_dev->mic_ch[i]; - chan_num = ch->ch_num; - seq_printf(s, "%-10i| %-#10x %-#10x %-#10x %-#10x", - chan_num, - mic_dma_read_reg(ch, MIC_DMA_REG_DCAR), - mic_dma_read_reg(ch, MIC_DMA_REG_DTPR), - mic_dma_read_reg(ch, MIC_DMA_REG_DHPR), - mic_dma_read_reg(ch, MIC_DMA_REG_DRAR_HI)); - seq_printf(s, " %-#10x %-#10x %-#14x %-#10x\n", - mic_dma_read_reg(ch, MIC_DMA_REG_DRAR_LO), - mic_dma_read_reg(ch, MIC_DMA_REG_DCHERR), - mic_dma_read_reg(ch, MIC_DMA_REG_DCHERRMSK), - mic_dma_read_reg(ch, MIC_DMA_REG_DSTAT)); - } - return 0; -} - -DEFINE_SHOW_ATTRIBUTE(mic_dma_reg); - -/* Debugfs parent dir */ -static struct dentry *mic_dma_dbg; - -static int mic_dma_driver_probe(struct mbus_device *mbdev) -{ - struct mic_dma_device *mic_dma_dev; - enum mic_dma_chan_owner owner; - - if (MBUS_DEV_DMA_MIC == mbdev->id.device) - owner = MIC_DMA_CHAN_MIC; - else - owner = MIC_DMA_CHAN_HOST; - - mic_dma_dev = mic_dma_dev_reg(mbdev, owner); - dev_set_drvdata(&mbdev->dev, mic_dma_dev); - - if (mic_dma_dbg) { - mic_dma_dev->dbg_dir = debugfs_create_dir(dev_name(&mbdev->dev), - mic_dma_dbg); - debugfs_create_file("mic_dma_reg", 0444, mic_dma_dev->dbg_dir, - mic_dma_dev, &mic_dma_reg_fops); - } - return 0; -} - -static void mic_dma_driver_remove(struct mbus_device *mbdev) -{ - struct mic_dma_device *mic_dma_dev; - - mic_dma_dev = dev_get_drvdata(&mbdev->dev); - debugfs_remove_recursive(mic_dma_dev->dbg_dir); - mic_dma_dev_unreg(mic_dma_dev); -} - -static struct mbus_device_id id_table[] = { - {MBUS_DEV_DMA_MIC, MBUS_DEV_ANY_ID}, - {MBUS_DEV_DMA_HOST, MBUS_DEV_ANY_ID}, - {0}, -}; - -static struct mbus_driver mic_dma_driver = { - .driver.name = KBUILD_MODNAME, - .driver.owner = THIS_MODULE, - .id_table = id_table, - .probe = mic_dma_driver_probe, - .remove = mic_dma_driver_remove, -}; - -static int __init mic_x100_dma_init(void) -{ - int rc = mbus_register_driver(&mic_dma_driver); - if (rc) - return rc; - mic_dma_dbg = debugfs_create_dir(KBUILD_MODNAME, NULL); - return 0; -} - -static void __exit mic_x100_dma_exit(void) -{ - debugfs_remove_recursive(mic_dma_dbg); - mbus_unregister_driver(&mic_dma_driver); -} - -module_init(mic_x100_dma_init); -module_exit(mic_x100_dma_exit); - -MODULE_DEVICE_TABLE(mbus, id_table); -MODULE_AUTHOR("Intel Corporation"); -MODULE_DESCRIPTION("Intel(R) MIC X100 DMA Driver"); -MODULE_LICENSE("GPL v2"); diff --git a/drivers/dma/mic_x100_dma.h b/drivers/dma/mic_x100_dma.h deleted file mode 100644 index 68ef43a91714..000000000000 --- a/drivers/dma/mic_x100_dma.h +++ /dev/null @@ -1,275 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel MIC X100 DMA Driver. - * - * Adapted from IOAT dma driver. - */ -#ifndef _MIC_X100_DMA_H_ -#define _MIC_X100_DMA_H_ - -#include -#include -#include -#include -#include -#include -#include - -#include "dmaengine.h" - -/* - * MIC has a total of 8 dma channels. - * Four channels are assigned for host SW use & the remaining for MIC SW. - * MIC DMA transfer size & addresses need to be 64 byte aligned. - */ -#define MIC_DMA_MAX_NUM_CHAN 8 -#define MIC_DMA_NUM_CHAN 4 -#define MIC_DMA_ALIGN_SHIFT DMAENGINE_ALIGN_64_BYTES -#define MIC_DMA_ALIGN_BYTES (1 << MIC_DMA_ALIGN_SHIFT) -#define MIC_DMA_DESC_RX_SIZE (128 * 1024 - 4) - -/* - * Register descriptions - * All the registers are 32 bit registers. - * DCR is a global register and all others are per-channel. - * DCR - bits 0, 2, 4, 6, 8, 10, 12, 14 - enable bits for channels 0 to 7 - * bits 1, 3, 5, 7, 9, 11, 13, 15 - owner bits for channels 0 to 7 - * DCAR - bit 24 & 25 interrupt masks for mic owned & host owned channels - * DHPR - head of the descriptor ring updated by s/w - * DTPR - tail of the descriptor ring updated by h/w - * DRAR_LO - lower 32 bits of descriptor ring's mic address - * DRAR_HI - 3:0 - remaining 4 bits of descriptor ring's mic address - * 20:4 descriptor ring size - * 25:21 mic smpt entry number - * DSTAT - 16:0 h/w completion count; 31:28 dma engine status - * DCHERR - this register is non-zero on error - * DCHERRMSK - interrupt mask register - */ -#define MIC_DMA_HW_CMP_CNT_MASK 0x1ffff -#define MIC_DMA_CHAN_QUIESCE 0x20000000 -#define MIC_DMA_SBOX_BASE 0x00010000 -#define MIC_DMA_SBOX_DCR 0x0000A280 -#define MIC_DMA_SBOX_CH_BASE 0x0001A000 -#define MIC_DMA_SBOX_CHAN_OFF 0x40 -#define MIC_DMA_SBOX_DCAR_IM0 (0x1 << 24) -#define MIC_DMA_SBOX_DCAR_IM1 (0x1 << 25) -#define MIC_DMA_SBOX_DRARHI_SYS_MASK (0x1 << 26) -#define MIC_DMA_REG_DCAR 0 -#define MIC_DMA_REG_DHPR 4 -#define MIC_DMA_REG_DTPR 8 -#define MIC_DMA_REG_DRAR_LO 20 -#define MIC_DMA_REG_DRAR_HI 24 -#define MIC_DMA_REG_DSTAT 32 -#define MIC_DMA_REG_DCHERR 44 -#define MIC_DMA_REG_DCHERRMSK 48 - -/* HW dma desc */ -struct mic_dma_desc { - u64 qw0; - u64 qw1; -}; - -enum mic_dma_chan_owner { - MIC_DMA_CHAN_MIC = 0, - MIC_DMA_CHAN_HOST -}; - -/* - * mic_dma_chan - channel specific information - * @ch_num: channel number - * @owner: owner of this channel - * @last_tail: cached value of descriptor ring tail - * @head: index of next descriptor in desc_ring - * @issued: hardware notification point - * @submitted: index that will be used to submit descriptors to h/w - * @api_ch: dma engine api channel - * @desc_ring: dma descriptor ring - * @desc_ring_micpa: mic physical address of desc_ring - * @status_dest: destination for status (fence) descriptor - * @status_dest_micpa: mic address for status_dest, - * DMA controller uses this address - * @tx_array: array of async_tx - * @cleanup_lock: lock held when processing completed tx - * @prep_lock: lock held in prep_memcpy & released in tx_submit - * @issue_lock: lock used to synchronize writes to head - * @cookie: mic_irq cookie used with mic irq request - */ -struct mic_dma_chan { - int ch_num; - enum mic_dma_chan_owner owner; - u32 last_tail; - u32 head; - u32 issued; - u32 submitted; - struct dma_chan api_ch; - struct mic_dma_desc *desc_ring; - dma_addr_t desc_ring_micpa; - u64 *status_dest; - dma_addr_t status_dest_micpa; - struct dma_async_tx_descriptor *tx_array; - spinlock_t cleanup_lock; - spinlock_t prep_lock; - spinlock_t issue_lock; - struct mic_irq *cookie; -}; - -/* - * struct mic_dma_device - per mic device - * @mic_ch: dma channels - * @dma_dev: underlying dma device - * @mbdev: mic bus dma device - * @mmio: virtual address of the mmio space - * @dbg_dir: debugfs directory - * @start_ch: first channel number that can be used - * @max_xfer_size: maximum transfer size per dma descriptor - */ -struct mic_dma_device { - struct mic_dma_chan mic_ch[MIC_DMA_MAX_NUM_CHAN]; - struct dma_device dma_dev; - struct mbus_device *mbdev; - void __iomem *mmio; - struct dentry *dbg_dir; - int start_ch; - size_t max_xfer_size; -}; - -static inline struct mic_dma_chan *to_mic_dma_chan(struct dma_chan *ch) -{ - return container_of(ch, struct mic_dma_chan, api_ch); -} - -static inline struct mic_dma_device *to_mic_dma_dev(struct mic_dma_chan *ch) -{ - return - container_of((const typeof(((struct mic_dma_device *)0)->mic_ch)*) - (ch - ch->ch_num), struct mic_dma_device, mic_ch); -} - -static inline struct mbus_device *to_mbus_device(struct mic_dma_chan *ch) -{ - return to_mic_dma_dev(ch)->mbdev; -} - -static inline struct mbus_hw_ops *to_mbus_hw_ops(struct mic_dma_chan *ch) -{ - return to_mbus_device(ch)->hw_ops; -} - -static inline struct device *mic_dma_ch_to_device(struct mic_dma_chan *ch) -{ - return to_mic_dma_dev(ch)->dma_dev.dev; -} - -static inline void __iomem *mic_dma_chan_to_mmio(struct mic_dma_chan *ch) -{ - return to_mic_dma_dev(ch)->mmio; -} - -static inline u32 mic_dma_read_reg(struct mic_dma_chan *ch, u32 reg) -{ - return ioread32(mic_dma_chan_to_mmio(ch) + MIC_DMA_SBOX_CH_BASE + - ch->ch_num * MIC_DMA_SBOX_CHAN_OFF + reg); -} - -static inline void mic_dma_write_reg(struct mic_dma_chan *ch, u32 reg, u32 val) -{ - iowrite32(val, mic_dma_chan_to_mmio(ch) + MIC_DMA_SBOX_CH_BASE + - ch->ch_num * MIC_DMA_SBOX_CHAN_OFF + reg); -} - -static inline u32 mic_dma_mmio_read(struct mic_dma_chan *ch, u32 offset) -{ - return ioread32(mic_dma_chan_to_mmio(ch) + offset); -} - -static inline void mic_dma_mmio_write(struct mic_dma_chan *ch, u32 val, - u32 offset) -{ - iowrite32(val, mic_dma_chan_to_mmio(ch) + offset); -} - -static inline u32 mic_dma_read_cmp_cnt(struct mic_dma_chan *ch) -{ - return mic_dma_read_reg(ch, MIC_DMA_REG_DSTAT) & - MIC_DMA_HW_CMP_CNT_MASK; -} - -static inline void mic_dma_chan_set_owner(struct mic_dma_chan *ch) -{ - u32 dcr = mic_dma_mmio_read(ch, MIC_DMA_SBOX_BASE + MIC_DMA_SBOX_DCR); - u32 chan_num = ch->ch_num; - - dcr = (dcr & ~(0x1 << (chan_num * 2))) | (ch->owner << (chan_num * 2)); - mic_dma_mmio_write(ch, dcr, MIC_DMA_SBOX_BASE + MIC_DMA_SBOX_DCR); -} - -static inline void mic_dma_enable_chan(struct mic_dma_chan *ch) -{ - u32 dcr = mic_dma_mmio_read(ch, MIC_DMA_SBOX_BASE + MIC_DMA_SBOX_DCR); - - dcr |= 2 << (ch->ch_num << 1); - mic_dma_mmio_write(ch, dcr, MIC_DMA_SBOX_BASE + MIC_DMA_SBOX_DCR); -} - -static inline void mic_dma_disable_chan(struct mic_dma_chan *ch) -{ - u32 dcr = mic_dma_mmio_read(ch, MIC_DMA_SBOX_BASE + MIC_DMA_SBOX_DCR); - - dcr &= ~(2 << (ch->ch_num << 1)); - mic_dma_mmio_write(ch, dcr, MIC_DMA_SBOX_BASE + MIC_DMA_SBOX_DCR); -} - -static void mic_dma_chan_set_desc_ring(struct mic_dma_chan *ch) -{ - u32 drar_hi; - dma_addr_t desc_ring_micpa = ch->desc_ring_micpa; - - drar_hi = (MIC_DMA_DESC_RX_SIZE & 0x1ffff) << 4; - if (MIC_DMA_CHAN_MIC == ch->owner) { - drar_hi |= (desc_ring_micpa >> 32) & 0xf; - } else { - drar_hi |= MIC_DMA_SBOX_DRARHI_SYS_MASK; - drar_hi |= ((desc_ring_micpa >> 34) - & 0x1f) << 21; - drar_hi |= (desc_ring_micpa >> 32) & 0x3; - } - mic_dma_write_reg(ch, MIC_DMA_REG_DRAR_LO, (u32) desc_ring_micpa); - mic_dma_write_reg(ch, MIC_DMA_REG_DRAR_HI, drar_hi); -} - -static inline void mic_dma_chan_mask_intr(struct mic_dma_chan *ch) -{ - u32 dcar = mic_dma_read_reg(ch, MIC_DMA_REG_DCAR); - - if (MIC_DMA_CHAN_MIC == ch->owner) - dcar |= MIC_DMA_SBOX_DCAR_IM0; - else - dcar |= MIC_DMA_SBOX_DCAR_IM1; - mic_dma_write_reg(ch, MIC_DMA_REG_DCAR, dcar); -} - -static inline void mic_dma_chan_unmask_intr(struct mic_dma_chan *ch) -{ - u32 dcar = mic_dma_read_reg(ch, MIC_DMA_REG_DCAR); - - if (MIC_DMA_CHAN_MIC == ch->owner) - dcar &= ~MIC_DMA_SBOX_DCAR_IM0; - else - dcar &= ~MIC_DMA_SBOX_DCAR_IM1; - mic_dma_write_reg(ch, MIC_DMA_REG_DCAR, dcar); -} - -static void mic_dma_ack_interrupt(struct mic_dma_chan *ch) -{ - if (MIC_DMA_CHAN_MIC == ch->owner) { - /* HW errata */ - mic_dma_chan_mask_intr(ch); - mic_dma_chan_unmask_intr(ch); - } - to_mbus_hw_ops(ch)->ack_interrupt(to_mbus_device(ch), ch->ch_num); -} -#endif diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig index d5ce8082b0a0..fafa8b0d8099 100644 --- a/drivers/misc/Kconfig +++ b/drivers/misc/Kconfig @@ -474,7 +474,6 @@ source "drivers/misc/lis3lv02d/Kconfig" source "drivers/misc/altera-stapl/Kconfig" source "drivers/misc/mei/Kconfig" source "drivers/misc/vmw_vmci/Kconfig" -source "drivers/misc/mic/Kconfig" source "drivers/misc/genwqe/Kconfig" source "drivers/misc/echo/Kconfig" source "drivers/misc/cxl/Kconfig" diff --git a/drivers/misc/Makefile b/drivers/misc/Makefile index 2521359e8ef7..d23231e73330 100644 --- a/drivers/misc/Makefile +++ b/drivers/misc/Makefile @@ -46,7 +46,6 @@ obj-$(CONFIG_VMWARE_VMCI) += vmw_vmci/ obj-$(CONFIG_LATTICE_ECP3_CONFIG) += lattice-ecp3-config.o obj-$(CONFIG_SRAM) += sram.o obj-$(CONFIG_SRAM_EXEC) += sram-exec.o -obj-y += mic/ obj-$(CONFIG_GENWQE) += genwqe/ obj-$(CONFIG_ECHO) += echo/ obj-$(CONFIG_CXL_BASE) += cxl/ diff --git a/drivers/misc/mic/Kconfig b/drivers/misc/mic/Kconfig deleted file mode 100644 index 8a7c2c5711d5..000000000000 --- a/drivers/misc/mic/Kconfig +++ /dev/null @@ -1,141 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0-only -menu "Intel MIC & related support" - -config INTEL_MIC_BUS - tristate "Intel MIC Bus Driver" - depends on 64BIT && PCI && X86 - select DMA_OPS - help - This option is selected by any driver which registers a - device or driver on the MIC Bus, such as CONFIG_INTEL_MIC_HOST, - CONFIG_INTEL_MIC_CARD, CONFIG_INTEL_MIC_X100_DMA etc. - - If you are building a host/card kernel with an Intel MIC device - then say M (recommended) or Y, else say N. If unsure say N. - - More information about the Intel MIC family as well as the Linux - OS and tools for MIC to use with this driver are available from - . - -config SCIF_BUS - tristate "SCIF Bus Driver" - depends on 64BIT && PCI && X86 - select DMA_OPS - help - This option is selected by any driver which registers a - device or driver on the SCIF Bus, such as CONFIG_INTEL_MIC_HOST - and CONFIG_INTEL_MIC_CARD. - - If you are building a host/card kernel with an Intel MIC device - then say M (recommended) or Y, else say N. If unsure say N. - - More information about the Intel MIC family as well as the Linux - OS and tools for MIC to use with this driver are available from - . - -config VOP_BUS - tristate "VOP Bus Driver" - depends on HAS_DMA - select DMA_OPS - help - This option is selected by any driver which registers a - device or driver on the VOP Bus, such as CONFIG_INTEL_MIC_HOST - and CONFIG_INTEL_MIC_CARD. - - If you are building a host/card kernel with an Intel MIC device - then say M (recommended) or Y, else say N. If unsure say N. - - More information about the Intel MIC family as well as the Linux - OS and tools for MIC to use with this driver are available from - . - -config INTEL_MIC_HOST - tristate "Intel MIC Host Driver" - depends on 64BIT && PCI && X86 - depends on INTEL_MIC_BUS && SCIF_BUS && MIC_COSM && VOP_BUS - select DMA_OPS - help - This enables Host Driver support for the Intel Many Integrated - Core (MIC) family of PCIe form factor coprocessor devices that - run a 64 bit Linux OS. The driver manages card OS state and - enables communication between host and card. Intel MIC X100 - devices are currently supported. - - If you are building a host kernel with an Intel MIC device then - say M (recommended) or Y, else say N. If unsure say N. - - More information about the Intel MIC family as well as the Linux - OS and tools for MIC to use with this driver are available from - . - -config INTEL_MIC_CARD - tristate "Intel MIC Card Driver" - depends on 64BIT && X86 - depends on INTEL_MIC_BUS && SCIF_BUS && MIC_COSM && VOP_BUS - select VIRTIO - help - This enables card driver support for the Intel Many Integrated - Core (MIC) device family. The card driver communicates shutdown/ - crash events to the host and allows registration/configuration of - virtio devices. Intel MIC X100 devices are currently supported. - - If you are building a card kernel for an Intel MIC device then - say M (recommended) or Y, else say N. If unsure say N. - - For more information see - . - -config SCIF - tristate "SCIF Driver" - depends on 64BIT && PCI && X86 && SCIF_BUS && IOMMU_SUPPORT - select IOMMU_IOVA - help - This enables SCIF Driver support for the Intel Many Integrated - Core (MIC) family of PCIe form factor coprocessor devices that - run a 64 bit Linux OS. The Symmetric Communication Interface - (SCIF (pronounced as skiff)) is a low level communications API - across PCIe currently implemented for MIC. - - If you are building a host kernel with an Intel MIC device then - say M (recommended) or Y, else say N. If unsure say N. - - More information about the Intel MIC family as well as the Linux - OS and tools for MIC to use with this driver are available from - . - -config MIC_COSM - tristate "Intel MIC Coprocessor State Management (COSM) Drivers" - depends on 64BIT && PCI && X86 && SCIF - help - This enables COSM driver support for the Intel Many - Integrated Core (MIC) family of PCIe form factor coprocessor - devices. COSM drivers implement functions such as boot, - shutdown, reset and reboot of MIC devices. - - If you are building a host kernel with an Intel MIC device then - say M (recommended) or Y, else say N. If unsure say N. - - More information about the Intel MIC family as well as the Linux - OS and tools for MIC to use with this driver are available from - . - -config VOP - tristate "VOP Driver" - depends on VOP_BUS - select VHOST_RING - select VIRTIO - help - This enables VOP (Virtio over PCIe) Driver support for the Intel - Many Integrated Core (MIC) family of PCIe form factor coprocessor - devices. The VOP driver allows virtio drivers, e.g. net, console - and block drivers, on the card connect to user space virtio - devices on the host. - - If you are building a host kernel with an Intel MIC device then - say M (recommended) or Y, else say N. If unsure say N. - - More information about the Intel MIC family as well as the Linux - OS and tools for MIC to use with this driver are available from - . - -endmenu diff --git a/drivers/misc/mic/Makefile b/drivers/misc/mic/Makefile deleted file mode 100644 index 1a43622b183f..000000000000 --- a/drivers/misc/mic/Makefile +++ /dev/null @@ -1,12 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0 -# -# Makefile - Intel MIC Linux driver. -# Copyright(c) 2013, Intel Corporation. -# -obj-$(CONFIG_INTEL_MIC_HOST) += host/ -obj-$(CONFIG_INTEL_MIC_CARD) += card/ -obj-y += bus/ -obj-$(CONFIG_SCIF) += scif/ -obj-$(CONFIG_MIC_COSM) += cosm/ -obj-$(CONFIG_MIC_COSM) += cosm_client/ -obj-$(CONFIG_VOP) += vop/ diff --git a/drivers/misc/mic/bus/Makefile b/drivers/misc/mic/bus/Makefile deleted file mode 100644 index 0a6aa21b2f67..000000000000 --- a/drivers/misc/mic/bus/Makefile +++ /dev/null @@ -1,9 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0-only -# -# Makefile - Intel MIC Linux driver. -# Copyright(c) 2014, Intel Corporation. -# -obj-$(CONFIG_INTEL_MIC_BUS) += mic_bus.o -obj-$(CONFIG_SCIF_BUS) += scif_bus.o -obj-$(CONFIG_MIC_COSM) += cosm_bus.o -obj-$(CONFIG_VOP_BUS) += vop_bus.o diff --git a/drivers/misc/mic/bus/cosm_bus.c b/drivers/misc/mic/bus/cosm_bus.c deleted file mode 100644 index 5f2141c71738..000000000000 --- a/drivers/misc/mic/bus/cosm_bus.c +++ /dev/null @@ -1,130 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2015 Intel Corporation. - * - * Intel MIC COSM Bus Driver - */ -#include -#include -#include -#include "cosm_bus.h" - -/* Unique numbering for cosm devices. */ -static DEFINE_IDA(cosm_index_ida); - -static int cosm_dev_probe(struct device *d) -{ - struct cosm_device *dev = dev_to_cosm(d); - struct cosm_driver *drv = drv_to_cosm(dev->dev.driver); - - return drv->probe(dev); -} - -static int cosm_dev_remove(struct device *d) -{ - struct cosm_device *dev = dev_to_cosm(d); - struct cosm_driver *drv = drv_to_cosm(dev->dev.driver); - - drv->remove(dev); - return 0; -} - -static struct bus_type cosm_bus = { - .name = "cosm_bus", - .probe = cosm_dev_probe, - .remove = cosm_dev_remove, -}; - -int cosm_register_driver(struct cosm_driver *driver) -{ - driver->driver.bus = &cosm_bus; - return driver_register(&driver->driver); -} -EXPORT_SYMBOL_GPL(cosm_register_driver); - -void cosm_unregister_driver(struct cosm_driver *driver) -{ - driver_unregister(&driver->driver); -} -EXPORT_SYMBOL_GPL(cosm_unregister_driver); - -static inline void cosm_release_dev(struct device *d) -{ - struct cosm_device *cdev = dev_to_cosm(d); - - kfree(cdev); -} - -struct cosm_device * -cosm_register_device(struct device *pdev, struct cosm_hw_ops *hw_ops) -{ - struct cosm_device *cdev; - int ret; - - cdev = kzalloc(sizeof(*cdev), GFP_KERNEL); - if (!cdev) - return ERR_PTR(-ENOMEM); - - cdev->dev.parent = pdev; - cdev->dev.release = cosm_release_dev; - cdev->hw_ops = hw_ops; - dev_set_drvdata(&cdev->dev, cdev); - cdev->dev.bus = &cosm_bus; - - /* Assign a unique device index and hence name */ - ret = ida_simple_get(&cosm_index_ida, 0, 0, GFP_KERNEL); - if (ret < 0) - goto free_cdev; - - cdev->index = ret; - cdev->dev.id = ret; - dev_set_name(&cdev->dev, "cosm-dev%u", cdev->index); - - ret = device_register(&cdev->dev); - if (ret) - goto ida_remove; - return cdev; -ida_remove: - ida_simple_remove(&cosm_index_ida, cdev->index); -free_cdev: - put_device(&cdev->dev); - return ERR_PTR(ret); -} -EXPORT_SYMBOL_GPL(cosm_register_device); - -void cosm_unregister_device(struct cosm_device *dev) -{ - int index = dev->index; /* save for after device release */ - - device_unregister(&dev->dev); - ida_simple_remove(&cosm_index_ida, index); -} -EXPORT_SYMBOL_GPL(cosm_unregister_device); - -struct cosm_device *cosm_find_cdev_by_id(int id) -{ - struct device *dev = subsys_find_device_by_id(&cosm_bus, id, NULL); - - return dev ? container_of(dev, struct cosm_device, dev) : NULL; -} -EXPORT_SYMBOL_GPL(cosm_find_cdev_by_id); - -static int __init cosm_init(void) -{ - return bus_register(&cosm_bus); -} - -static void __exit cosm_exit(void) -{ - bus_unregister(&cosm_bus); - ida_destroy(&cosm_index_ida); -} - -core_initcall(cosm_init); -module_exit(cosm_exit); - -MODULE_AUTHOR("Intel Corporation"); -MODULE_DESCRIPTION("Intel(R) MIC card OS state management bus driver"); -MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/bus/cosm_bus.h b/drivers/misc/mic/bus/cosm_bus.h deleted file mode 100644 index d50d7aea168d..000000000000 --- a/drivers/misc/mic/bus/cosm_bus.h +++ /dev/null @@ -1,125 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2015 Intel Corporation. - * - * Intel MIC COSM Bus Driver - */ -#ifndef _COSM_BUS_H_ -#define _COSM_BUS_H_ - -#include -#include -#include "../common/mic_dev.h" - -/** - * cosm_device - representation of a cosm device - * - * @attr_group: Pointer to list of sysfs attribute groups. - * @sdev: Device for sysfs entries. - * @state: MIC state. - * @prev_state: MIC state previous to MIC_RESETTING - * @shutdown_status: MIC status reported by card for shutdown/crashes. - * @shutdown_status_int: Internal shutdown status maintained by the driver - * @cosm_mutex: Mutex for synchronizing access to data structures. - * @reset_trigger_work: Work for triggering reset requests. - * @scif_work: Work for handling per device SCIF connections - * @cmdline: Kernel command line. - * @firmware: Firmware file name. - * @ramdisk: Ramdisk file name. - * @bootmode: Boot mode i.e. "linux" or "elf" for flash updates. - * @log_buf_addr: Log buffer address for MIC. - * @log_buf_len: Log buffer length address for MIC. - * @state_sysfs: Sysfs dirent for notifying ring 3 about MIC state changes. - * @hw_ops: the hardware bus ops for this device. - * @dev: underlying device. - * @index: unique position on the cosm bus - * @dbg_dir: debug fs directory - * @newepd: new endpoint from scif accept to be assigned to this cdev - * @epd: SCIF endpoint for this cdev - * @heartbeat_watchdog_enable: if heartbeat watchdog is enabled for this cdev - * @sysfs_heartbeat_enable: sysfs setting for disabling heartbeat notification - */ -struct cosm_device { - const struct attribute_group **attr_group; - struct device *sdev; - u8 state; - u8 prev_state; - u8 shutdown_status; - u8 shutdown_status_int; - struct mutex cosm_mutex; - struct work_struct reset_trigger_work; - struct work_struct scif_work; - char *cmdline; - char *firmware; - char *ramdisk; - char *bootmode; - void *log_buf_addr; - int *log_buf_len; - struct kernfs_node *state_sysfs; - struct cosm_hw_ops *hw_ops; - struct device dev; - int index; - struct dentry *dbg_dir; - scif_epd_t newepd; - scif_epd_t epd; - bool heartbeat_watchdog_enable; - bool sysfs_heartbeat_enable; -}; - -/** - * cosm_driver - operations for a cosm driver - * - * @driver: underlying device driver (populate name and owner). - * @probe: the function to call when a device is found. Returns 0 or -errno. - * @remove: the function to call when a device is removed. - */ -struct cosm_driver { - struct device_driver driver; - int (*probe)(struct cosm_device *dev); - void (*remove)(struct cosm_device *dev); -}; - -/** - * cosm_hw_ops - cosm bus ops - * - * @reset: trigger MIC reset - * @force_reset: force MIC reset - * @post_reset: inform MIC reset is complete - * @ready: is MIC ready for OS download - * @start: boot MIC - * @stop: prepare MIC for reset - * @family: return MIC HW family string - * @stepping: return MIC HW stepping string - * @aper: return MIC PCIe aperture - */ -struct cosm_hw_ops { - void (*reset)(struct cosm_device *cdev); - void (*force_reset)(struct cosm_device *cdev); - void (*post_reset)(struct cosm_device *cdev, enum mic_states state); - bool (*ready)(struct cosm_device *cdev); - int (*start)(struct cosm_device *cdev, int id); - void (*stop)(struct cosm_device *cdev, bool force); - ssize_t (*family)(struct cosm_device *cdev, char *buf); - ssize_t (*stepping)(struct cosm_device *cdev, char *buf); - struct mic_mw *(*aper)(struct cosm_device *cdev); -}; - -struct cosm_device * -cosm_register_device(struct device *pdev, struct cosm_hw_ops *hw_ops); -void cosm_unregister_device(struct cosm_device *dev); -int cosm_register_driver(struct cosm_driver *drv); -void cosm_unregister_driver(struct cosm_driver *drv); -struct cosm_device *cosm_find_cdev_by_id(int id); - -static inline struct cosm_device *dev_to_cosm(struct device *dev) -{ - return container_of(dev, struct cosm_device, dev); -} - -static inline struct cosm_driver *drv_to_cosm(struct device_driver *drv) -{ - return container_of(drv, struct cosm_driver, driver); -} -#endif /* _COSM_BUS_H */ diff --git a/drivers/misc/mic/bus/mic_bus.c b/drivers/misc/mic/bus/mic_bus.c deleted file mode 100644 index a08cb29692a8..000000000000 --- a/drivers/misc/mic/bus/mic_bus.c +++ /dev/null @@ -1,194 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel MIC Bus driver. - * - * This implementation is very similar to the the virtio bus driver - * implementation @ drivers/virtio/virtio.c - */ -#include -#include -#include -#include -#include - -static ssize_t device_show(struct device *d, - struct device_attribute *attr, char *buf) -{ - struct mbus_device *dev = dev_to_mbus(d); - return sprintf(buf, "0x%04x\n", dev->id.device); -} -static DEVICE_ATTR_RO(device); - -static ssize_t vendor_show(struct device *d, - struct device_attribute *attr, char *buf) -{ - struct mbus_device *dev = dev_to_mbus(d); - return sprintf(buf, "0x%04x\n", dev->id.vendor); -} -static DEVICE_ATTR_RO(vendor); - -static ssize_t modalias_show(struct device *d, - struct device_attribute *attr, char *buf) -{ - struct mbus_device *dev = dev_to_mbus(d); - return sprintf(buf, "mbus:d%08Xv%08X\n", - dev->id.device, dev->id.vendor); -} -static DEVICE_ATTR_RO(modalias); - -static struct attribute *mbus_dev_attrs[] = { - &dev_attr_device.attr, - &dev_attr_vendor.attr, - &dev_attr_modalias.attr, - NULL, -}; -ATTRIBUTE_GROUPS(mbus_dev); - -static inline int mbus_id_match(const struct mbus_device *dev, - const struct mbus_device_id *id) -{ - if (id->device != dev->id.device && id->device != MBUS_DEV_ANY_ID) - return 0; - - return id->vendor == MBUS_DEV_ANY_ID || id->vendor == dev->id.vendor; -} - -/* - * This looks through all the IDs a driver claims to support. If any of them - * match, we return 1 and the kernel will call mbus_dev_probe(). - */ -static int mbus_dev_match(struct device *dv, struct device_driver *dr) -{ - unsigned int i; - struct mbus_device *dev = dev_to_mbus(dv); - const struct mbus_device_id *ids; - - ids = drv_to_mbus(dr)->id_table; - for (i = 0; ids[i].device; i++) - if (mbus_id_match(dev, &ids[i])) - return 1; - return 0; -} - -static int mbus_uevent(struct device *dv, struct kobj_uevent_env *env) -{ - struct mbus_device *dev = dev_to_mbus(dv); - - return add_uevent_var(env, "MODALIAS=mbus:d%08Xv%08X", - dev->id.device, dev->id.vendor); -} - -static int mbus_dev_probe(struct device *d) -{ - int err; - struct mbus_device *dev = dev_to_mbus(d); - struct mbus_driver *drv = drv_to_mbus(dev->dev.driver); - - err = drv->probe(dev); - if (!err) - if (drv->scan) - drv->scan(dev); - return err; -} - -static int mbus_dev_remove(struct device *d) -{ - struct mbus_device *dev = dev_to_mbus(d); - struct mbus_driver *drv = drv_to_mbus(dev->dev.driver); - - drv->remove(dev); - return 0; -} - -static struct bus_type mic_bus = { - .name = "mic_bus", - .match = mbus_dev_match, - .dev_groups = mbus_dev_groups, - .uevent = mbus_uevent, - .probe = mbus_dev_probe, - .remove = mbus_dev_remove, -}; - -int mbus_register_driver(struct mbus_driver *driver) -{ - driver->driver.bus = &mic_bus; - return driver_register(&driver->driver); -} -EXPORT_SYMBOL_GPL(mbus_register_driver); - -void mbus_unregister_driver(struct mbus_driver *driver) -{ - driver_unregister(&driver->driver); -} -EXPORT_SYMBOL_GPL(mbus_unregister_driver); - -static void mbus_release_dev(struct device *d) -{ - struct mbus_device *mbdev = dev_to_mbus(d); - kfree(mbdev); -} - -struct mbus_device * -mbus_register_device(struct device *pdev, int id, const struct dma_map_ops *dma_ops, - struct mbus_hw_ops *hw_ops, int index, - void __iomem *mmio_va) -{ - int ret; - struct mbus_device *mbdev; - - mbdev = kzalloc(sizeof(*mbdev), GFP_KERNEL); - if (!mbdev) - return ERR_PTR(-ENOMEM); - - mbdev->mmio_va = mmio_va; - mbdev->dev.parent = pdev; - mbdev->id.device = id; - mbdev->id.vendor = MBUS_DEV_ANY_ID; - mbdev->dev.dma_ops = dma_ops; - mbdev->dev.dma_mask = &mbdev->dev.coherent_dma_mask; - dma_set_mask(&mbdev->dev, DMA_BIT_MASK(64)); - mbdev->dev.release = mbus_release_dev; - mbdev->hw_ops = hw_ops; - mbdev->dev.bus = &mic_bus; - mbdev->index = index; - dev_set_name(&mbdev->dev, "mbus-dev%u", mbdev->index); - /* - * device_register() causes the bus infrastructure to look for a - * matching driver. - */ - ret = device_register(&mbdev->dev); - if (ret) - goto free_mbdev; - return mbdev; -free_mbdev: - put_device(&mbdev->dev); - return ERR_PTR(ret); -} -EXPORT_SYMBOL_GPL(mbus_register_device); - -void mbus_unregister_device(struct mbus_device *mbdev) -{ - device_unregister(&mbdev->dev); -} -EXPORT_SYMBOL_GPL(mbus_unregister_device); - -static int __init mbus_init(void) -{ - return bus_register(&mic_bus); -} - -static void __exit mbus_exit(void) -{ - bus_unregister(&mic_bus); -} - -core_initcall(mbus_init); -module_exit(mbus_exit); - -MODULE_AUTHOR("Intel Corporation"); -MODULE_DESCRIPTION("Intel(R) MIC Bus driver"); -MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/bus/scif_bus.c b/drivers/misc/mic/bus/scif_bus.c deleted file mode 100644 index ad7c3604f151..000000000000 --- a/drivers/misc/mic/bus/scif_bus.c +++ /dev/null @@ -1,201 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel Symmetric Communications Interface Bus driver. - */ -#include -#include -#include -#include - -#include "scif_bus.h" - -static ssize_t device_show(struct device *d, - struct device_attribute *attr, char *buf) -{ - struct scif_hw_dev *dev = dev_to_scif(d); - - return sprintf(buf, "0x%04x\n", dev->id.device); -} -static DEVICE_ATTR_RO(device); - -static ssize_t vendor_show(struct device *d, - struct device_attribute *attr, char *buf) -{ - struct scif_hw_dev *dev = dev_to_scif(d); - - return sprintf(buf, "0x%04x\n", dev->id.vendor); -} -static DEVICE_ATTR_RO(vendor); - -static ssize_t modalias_show(struct device *d, - struct device_attribute *attr, char *buf) -{ - struct scif_hw_dev *dev = dev_to_scif(d); - - return sprintf(buf, "scif:d%08Xv%08X\n", - dev->id.device, dev->id.vendor); -} -static DEVICE_ATTR_RO(modalias); - -static struct attribute *scif_dev_attrs[] = { - &dev_attr_device.attr, - &dev_attr_vendor.attr, - &dev_attr_modalias.attr, - NULL, -}; -ATTRIBUTE_GROUPS(scif_dev); - -static inline int scif_id_match(const struct scif_hw_dev *dev, - const struct scif_hw_dev_id *id) -{ - if (id->device != dev->id.device && id->device != SCIF_DEV_ANY_ID) - return 0; - - return id->vendor == SCIF_DEV_ANY_ID || id->vendor == dev->id.vendor; -} - -/* - * This looks through all the IDs a driver claims to support. If any of them - * match, we return 1 and the kernel will call scif_dev_probe(). - */ -static int scif_dev_match(struct device *dv, struct device_driver *dr) -{ - unsigned int i; - struct scif_hw_dev *dev = dev_to_scif(dv); - const struct scif_hw_dev_id *ids; - - ids = drv_to_scif(dr)->id_table; - for (i = 0; ids[i].device; i++) - if (scif_id_match(dev, &ids[i])) - return 1; - return 0; -} - -static int scif_uevent(struct device *dv, struct kobj_uevent_env *env) -{ - struct scif_hw_dev *dev = dev_to_scif(dv); - - return add_uevent_var(env, "MODALIAS=scif:d%08Xv%08X", - dev->id.device, dev->id.vendor); -} - -static int scif_dev_probe(struct device *d) -{ - struct scif_hw_dev *dev = dev_to_scif(d); - struct scif_driver *drv = drv_to_scif(dev->dev.driver); - - return drv->probe(dev); -} - -static int scif_dev_remove(struct device *d) -{ - struct scif_hw_dev *dev = dev_to_scif(d); - struct scif_driver *drv = drv_to_scif(dev->dev.driver); - - drv->remove(dev); - return 0; -} - -static struct bus_type scif_bus = { - .name = "scif_bus", - .match = scif_dev_match, - .dev_groups = scif_dev_groups, - .uevent = scif_uevent, - .probe = scif_dev_probe, - .remove = scif_dev_remove, -}; - -int scif_register_driver(struct scif_driver *driver) -{ - driver->driver.bus = &scif_bus; - return driver_register(&driver->driver); -} -EXPORT_SYMBOL_GPL(scif_register_driver); - -void scif_unregister_driver(struct scif_driver *driver) -{ - driver_unregister(&driver->driver); -} -EXPORT_SYMBOL_GPL(scif_unregister_driver); - -static void scif_release_dev(struct device *d) -{ - struct scif_hw_dev *sdev = dev_to_scif(d); - - kfree(sdev); -} - -struct scif_hw_dev * -scif_register_device(struct device *pdev, int id, const struct dma_map_ops *dma_ops, - struct scif_hw_ops *hw_ops, u8 dnode, u8 snode, - struct mic_mw *mmio, struct mic_mw *aper, void *dp, - void __iomem *rdp, struct dma_chan **chan, int num_chan, - bool card_rel_da) -{ - int ret; - struct scif_hw_dev *sdev; - - sdev = kzalloc(sizeof(*sdev), GFP_KERNEL); - if (!sdev) - return ERR_PTR(-ENOMEM); - - sdev->dev.parent = pdev; - sdev->id.device = id; - sdev->id.vendor = SCIF_DEV_ANY_ID; - sdev->dev.dma_ops = dma_ops; - sdev->dev.release = scif_release_dev; - sdev->hw_ops = hw_ops; - sdev->dnode = dnode; - sdev->snode = snode; - dev_set_drvdata(&sdev->dev, sdev); - sdev->dev.bus = &scif_bus; - sdev->mmio = mmio; - sdev->aper = aper; - sdev->dp = dp; - sdev->rdp = rdp; - sdev->dev.dma_mask = &sdev->dev.coherent_dma_mask; - dma_set_mask(&sdev->dev, DMA_BIT_MASK(64)); - sdev->dma_ch = chan; - sdev->num_dma_ch = num_chan; - sdev->card_rel_da = card_rel_da; - dev_set_name(&sdev->dev, "scif-dev%u", sdev->dnode); - /* - * device_register() causes the bus infrastructure to look for a - * matching driver. - */ - ret = device_register(&sdev->dev); - if (ret) - goto free_sdev; - return sdev; -free_sdev: - put_device(&sdev->dev); - return ERR_PTR(ret); -} -EXPORT_SYMBOL_GPL(scif_register_device); - -void scif_unregister_device(struct scif_hw_dev *sdev) -{ - device_unregister(&sdev->dev); -} -EXPORT_SYMBOL_GPL(scif_unregister_device); - -static int __init scif_init(void) -{ - return bus_register(&scif_bus); -} - -static void __exit scif_exit(void) -{ - bus_unregister(&scif_bus); -} - -core_initcall(scif_init); -module_exit(scif_exit); - -MODULE_AUTHOR("Intel Corporation"); -MODULE_DESCRIPTION("Intel(R) SCIF Bus driver"); -MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/bus/scif_bus.h b/drivers/misc/mic/bus/scif_bus.h deleted file mode 100644 index 4981eb56f879..000000000000 --- a/drivers/misc/mic/bus/scif_bus.h +++ /dev/null @@ -1,125 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel Symmetric Communications Interface Bus driver. - */ -#ifndef _SCIF_BUS_H_ -#define _SCIF_BUS_H_ -/* - * Everything a scif driver needs to work with any particular scif - * hardware abstraction layer. - */ -#include - -#include -#include "../common/mic_dev.h" - -struct scif_hw_dev_id { - u32 device; - u32 vendor; -}; - -#define MIC_SCIF_DEV 1 -#define SCIF_DEV_ANY_ID 0xffffffff - -/** - * scif_hw_dev - representation of a hardware device abstracted for scif - * @hw_ops: the hardware ops supported by this device - * @id: the device type identification (used to match it with a driver) - * @mmio: MMIO memory window - * @aper: Aperture memory window - * @dev: underlying device - * @dnode - The destination node which this device will communicate with. - * @snode - The source node for this device. - * @dp - Self device page - * @rdp - Remote device page - * @dma_ch - Array of DMA channels - * @num_dma_ch - Number of DMA channels available - * @card_rel_da - Set to true if DMA addresses programmed in the DMA engine - * are relative to the card point of view - */ -struct scif_hw_dev { - struct scif_hw_ops *hw_ops; - struct scif_hw_dev_id id; - struct mic_mw *mmio; - struct mic_mw *aper; - struct device dev; - u8 dnode; - u8 snode; - void *dp; - void __iomem *rdp; - struct dma_chan **dma_ch; - int num_dma_ch; - bool card_rel_da; -}; - -/** - * scif_driver - operations for a scif I/O driver - * @driver: underlying device driver (populate name and owner). - * @id_table: the ids serviced by this driver. - * @probe: the function to call when a device is found. Returns 0 or -errno. - * @remove: the function to call when a device is removed. - */ -struct scif_driver { - struct device_driver driver; - const struct scif_hw_dev_id *id_table; - int (*probe)(struct scif_hw_dev *dev); - void (*remove)(struct scif_hw_dev *dev); -}; - -/** - * scif_hw_ops - Hardware operations for accessing a SCIF device on the SCIF bus. - * - * @next_db: Obtain the next available doorbell. - * @request_irq: Request an interrupt on a particular doorbell. - * @free_irq: Free an interrupt requested previously. - * @ack_interrupt: acknowledge an interrupt in the ISR. - * @send_intr: Send an interrupt to the remote node on a specified doorbell. - * @send_p2p_intr: Send an interrupt to the peer node on a specified doorbell - * which is specifically targeted for a peer to peer node. - * @remap: Map a buffer with the specified physical address and length. - * @unmap: Unmap a buffer previously mapped. - */ -struct scif_hw_ops { - int (*next_db)(struct scif_hw_dev *sdev); - struct mic_irq * (*request_irq)(struct scif_hw_dev *sdev, - irqreturn_t (*func)(int irq, - void *data), - const char *name, void *data, - int db); - void (*free_irq)(struct scif_hw_dev *sdev, - struct mic_irq *cookie, void *data); - void (*ack_interrupt)(struct scif_hw_dev *sdev, int num); - void (*send_intr)(struct scif_hw_dev *sdev, int db); - void (*send_p2p_intr)(struct scif_hw_dev *sdev, int db, - struct mic_mw *mw); - void __iomem * (*remap)(struct scif_hw_dev *sdev, - phys_addr_t pa, size_t len); - void (*unmap)(struct scif_hw_dev *sdev, void __iomem *va); -}; - -int scif_register_driver(struct scif_driver *driver); -void scif_unregister_driver(struct scif_driver *driver); -struct scif_hw_dev * -scif_register_device(struct device *pdev, int id, - const struct dma_map_ops *dma_ops, - struct scif_hw_ops *hw_ops, u8 dnode, u8 snode, - struct mic_mw *mmio, struct mic_mw *aper, - void *dp, void __iomem *rdp, - struct dma_chan **chan, int num_chan, - bool card_rel_da); -void scif_unregister_device(struct scif_hw_dev *sdev); - -static inline struct scif_hw_dev *dev_to_scif(struct device *dev) -{ - return container_of(dev, struct scif_hw_dev, dev); -} - -static inline struct scif_driver *drv_to_scif(struct device_driver *drv) -{ - return container_of(drv, struct scif_driver, driver); -} -#endif /* _SCIF_BUS_H */ diff --git a/drivers/misc/mic/bus/vop_bus.c b/drivers/misc/mic/bus/vop_bus.c deleted file mode 100644 index 6935ddca1bd5..000000000000 --- a/drivers/misc/mic/bus/vop_bus.c +++ /dev/null @@ -1,194 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2016 Intel Corporation. - * - * Intel Virtio Over PCIe (VOP) Bus driver. - */ -#include -#include -#include -#include - -#include "vop_bus.h" - -static ssize_t device_show(struct device *d, - struct device_attribute *attr, char *buf) -{ - struct vop_device *dev = dev_to_vop(d); - - return sprintf(buf, "0x%04x\n", dev->id.device); -} -static DEVICE_ATTR_RO(device); - -static ssize_t vendor_show(struct device *d, - struct device_attribute *attr, char *buf) -{ - struct vop_device *dev = dev_to_vop(d); - - return sprintf(buf, "0x%04x\n", dev->id.vendor); -} -static DEVICE_ATTR_RO(vendor); - -static ssize_t modalias_show(struct device *d, - struct device_attribute *attr, char *buf) -{ - struct vop_device *dev = dev_to_vop(d); - - return sprintf(buf, "vop:d%08Xv%08X\n", - dev->id.device, dev->id.vendor); -} -static DEVICE_ATTR_RO(modalias); - -static struct attribute *vop_dev_attrs[] = { - &dev_attr_device.attr, - &dev_attr_vendor.attr, - &dev_attr_modalias.attr, - NULL, -}; -ATTRIBUTE_GROUPS(vop_dev); - -static inline int vop_id_match(const struct vop_device *dev, - const struct vop_device_id *id) -{ - if (id->device != dev->id.device && id->device != VOP_DEV_ANY_ID) - return 0; - - return id->vendor == VOP_DEV_ANY_ID || id->vendor == dev->id.vendor; -} - -/* - * This looks through all the IDs a driver claims to support. If any of them - * match, we return 1 and the kernel will call vop_dev_probe(). - */ -static int vop_dev_match(struct device *dv, struct device_driver *dr) -{ - unsigned int i; - struct vop_device *dev = dev_to_vop(dv); - const struct vop_device_id *ids; - - ids = drv_to_vop(dr)->id_table; - for (i = 0; ids[i].device; i++) - if (vop_id_match(dev, &ids[i])) - return 1; - return 0; -} - -static int vop_uevent(struct device *dv, struct kobj_uevent_env *env) -{ - struct vop_device *dev = dev_to_vop(dv); - - return add_uevent_var(env, "MODALIAS=vop:d%08Xv%08X", - dev->id.device, dev->id.vendor); -} - -static int vop_dev_probe(struct device *d) -{ - struct vop_device *dev = dev_to_vop(d); - struct vop_driver *drv = drv_to_vop(dev->dev.driver); - - return drv->probe(dev); -} - -static int vop_dev_remove(struct device *d) -{ - struct vop_device *dev = dev_to_vop(d); - struct vop_driver *drv = drv_to_vop(dev->dev.driver); - - drv->remove(dev); - return 0; -} - -static struct bus_type vop_bus = { - .name = "vop_bus", - .match = vop_dev_match, - .dev_groups = vop_dev_groups, - .uevent = vop_uevent, - .probe = vop_dev_probe, - .remove = vop_dev_remove, -}; - -int vop_register_driver(struct vop_driver *driver) -{ - driver->driver.bus = &vop_bus; - return driver_register(&driver->driver); -} -EXPORT_SYMBOL_GPL(vop_register_driver); - -void vop_unregister_driver(struct vop_driver *driver) -{ - driver_unregister(&driver->driver); -} -EXPORT_SYMBOL_GPL(vop_unregister_driver); - -static void vop_release_dev(struct device *d) -{ - struct vop_device *dev = dev_to_vop(d); - - kfree(dev); -} - -struct vop_device * -vop_register_device(struct device *pdev, int id, - const struct dma_map_ops *dma_ops, - struct vop_hw_ops *hw_ops, u8 dnode, struct mic_mw *aper, - struct dma_chan *chan) -{ - int ret; - struct vop_device *vdev; - - vdev = kzalloc(sizeof(*vdev), GFP_KERNEL); - if (!vdev) - return ERR_PTR(-ENOMEM); - - vdev->dev.parent = pdev; - vdev->id.device = id; - vdev->id.vendor = VOP_DEV_ANY_ID; - vdev->dev.dma_ops = dma_ops; - vdev->dev.dma_mask = &vdev->dev.coherent_dma_mask; - dma_set_mask(&vdev->dev, DMA_BIT_MASK(64)); - vdev->dev.release = vop_release_dev; - vdev->hw_ops = hw_ops; - vdev->dev.bus = &vop_bus; - vdev->dnode = dnode; - vdev->aper = aper; - vdev->dma_ch = chan; - vdev->index = dnode - 1; - dev_set_name(&vdev->dev, "vop-dev%u", vdev->index); - /* - * device_register() causes the bus infrastructure to look for a - * matching driver. - */ - ret = device_register(&vdev->dev); - if (ret) - goto free_vdev; - return vdev; -free_vdev: - put_device(&vdev->dev); - return ERR_PTR(ret); -} -EXPORT_SYMBOL_GPL(vop_register_device); - -void vop_unregister_device(struct vop_device *dev) -{ - device_unregister(&dev->dev); -} -EXPORT_SYMBOL_GPL(vop_unregister_device); - -static int __init vop_init(void) -{ - return bus_register(&vop_bus); -} - -static void __exit vop_exit(void) -{ - bus_unregister(&vop_bus); -} - -core_initcall(vop_init); -module_exit(vop_exit); - -MODULE_AUTHOR("Intel Corporation"); -MODULE_DESCRIPTION("Intel(R) VOP Bus driver"); -MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/bus/vop_bus.h b/drivers/misc/mic/bus/vop_bus.h deleted file mode 100644 index 4fa02808c1e2..000000000000 --- a/drivers/misc/mic/bus/vop_bus.h +++ /dev/null @@ -1,129 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2016 Intel Corporation. - * - * Intel Virtio over PCIe Bus driver. - */ -#ifndef _VOP_BUS_H_ -#define _VOP_BUS_H_ -/* - * Everything a vop driver needs to work with any particular vop - * implementation. - */ -#include -#include - -#include "../common/mic_dev.h" - -struct vop_device_id { - u32 device; - u32 vendor; -}; - -#define VOP_DEV_TRNSP 1 -#define VOP_DEV_ANY_ID 0xffffffff -/* - * Size of the internal buffer used during DMA's as an intermediate buffer - * for copy to/from user. Must be an integral number of pages. - */ -#define VOP_INT_DMA_BUF_SIZE PAGE_ALIGN(64 * 1024ULL) - -/** - * vop_device - representation of a device using vop - * @hw_ops: the hardware ops supported by this device. - * @id: the device type identification (used to match it with a driver). - * @dev: underlying device. - * @dnode - The destination node which this device will communicate with. - * @aper: Aperture memory window - * @dma_ch - DMA channel - * @index: unique position on the vop bus - */ -struct vop_device { - struct vop_hw_ops *hw_ops; - struct vop_device_id id; - struct device dev; - u8 dnode; - struct mic_mw *aper; - struct dma_chan *dma_ch; - int index; -}; - -/** - * vop_driver - operations for a vop I/O driver - * @driver: underlying device driver (populate name and owner). - * @id_table: the ids serviced by this driver. - * @probe: the function to call when a device is found. Returns 0 or -errno. - * @remove: the function to call when a device is removed. - */ -struct vop_driver { - struct device_driver driver; - const struct vop_device_id *id_table; - int (*probe)(struct vop_device *dev); - void (*remove)(struct vop_device *dev); -}; - -/** - * vop_hw_ops - Hardware operations for accessing a VOP device on the VOP bus. - * - * @next_db: Obtain the next available doorbell. - * @request_irq: Request an interrupt on a particular doorbell. - * @free_irq: Free an interrupt requested previously. - * @ack_interrupt: acknowledge an interrupt in the ISR. - * @get_remote_dp: Get access to the virtio device page used by the remote - * node to add/remove/configure virtio devices. - * @get_dp: Get access to the virtio device page used by the self - * node to add/remove/configure virtio devices. - * @send_intr: Send an interrupt to the peer node on a specified doorbell. - * @remap: Map a buffer with the specified DMA address and length. - * @unmap: Unmap a buffer previously mapped. - * @dma_filter: The DMA filter function to use for obtaining access to - * a DMA channel on the peer node. - */ -struct vop_hw_ops { - int (*next_db)(struct vop_device *vpdev); - struct mic_irq *(*request_irq)(struct vop_device *vpdev, - irqreturn_t (*func)(int irq, void *data), - const char *name, void *data, - int intr_src); - void (*free_irq)(struct vop_device *vpdev, - struct mic_irq *cookie, void *data); - void (*ack_interrupt)(struct vop_device *vpdev, int num); - void __iomem * (*get_remote_dp)(struct vop_device *vpdev); - void * (*get_dp)(struct vop_device *vpdev); - void (*send_intr)(struct vop_device *vpdev, int db); - void __iomem * (*remap)(struct vop_device *vpdev, - dma_addr_t pa, size_t len); - void (*unmap)(struct vop_device *vpdev, void __iomem *va); -}; - -struct vop_device * -vop_register_device(struct device *pdev, int id, - const struct dma_map_ops *dma_ops, - struct vop_hw_ops *hw_ops, u8 dnode, struct mic_mw *aper, - struct dma_chan *chan); -void vop_unregister_device(struct vop_device *dev); -int vop_register_driver(struct vop_driver *drv); -void vop_unregister_driver(struct vop_driver *drv); - -/* - * module_vop_driver() - Helper macro for drivers that don't do - * anything special in module init/exit. This eliminates a lot of - * boilerplate. Each module may only use this macro once, and - * calling it replaces module_init() and module_exit() - */ -#define module_vop_driver(__vop_driver) \ - module_driver(__vop_driver, vop_register_driver, \ - vop_unregister_driver) - -static inline struct vop_device *dev_to_vop(struct device *dev) -{ - return container_of(dev, struct vop_device, dev); -} - -static inline struct vop_driver *drv_to_vop(struct device_driver *drv) -{ - return container_of(drv, struct vop_driver, driver); -} -#endif /* _VOP_BUS_H */ diff --git a/drivers/misc/mic/card/Makefile b/drivers/misc/mic/card/Makefile deleted file mode 100644 index 921a7e7e0fbd..000000000000 --- a/drivers/misc/mic/card/Makefile +++ /dev/null @@ -1,11 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0 -# -# Makefile - Intel MIC Linux driver. -# Copyright(c) 2013, Intel Corporation. -# -ccflags-y += -DINTEL_MIC_CARD - -obj-$(CONFIG_INTEL_MIC_CARD) += mic_card.o -mic_card-y += mic_x100.o -mic_card-y += mic_device.o -mic_card-y += mic_debugfs.o diff --git a/drivers/misc/mic/card/mic_debugfs.c b/drivers/misc/mic/card/mic_debugfs.c deleted file mode 100644 index 4c326e8f4d99..000000000000 --- a/drivers/misc/mic/card/mic_debugfs.c +++ /dev/null @@ -1,85 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Disclaimer: The codes contained in these modules may be specific to - * the Intel Software Development Platform codenamed: Knights Ferry, and - * the Intel product codenamed: Knights Corner, and are not backward - * compatible with other Intel products. Additionally, Intel will NOT - * support the codes or instruction set in future products. - * - * Intel MIC Card driver. - */ -#include -#include -#include -#include -#include - -#include "../common/mic_dev.h" -#include "mic_device.h" - -/* Debugfs parent dir */ -static struct dentry *mic_dbg; - -/* - * mic_intr_show - Send interrupts to host. - */ -static int mic_intr_show(struct seq_file *s, void *unused) -{ - struct mic_driver *mdrv = s->private; - struct mic_device *mdev = &mdrv->mdev; - - mic_send_intr(mdev, 0); - msleep(1000); - mic_send_intr(mdev, 1); - msleep(1000); - mic_send_intr(mdev, 2); - msleep(1000); - mic_send_intr(mdev, 3); - msleep(1000); - - return 0; -} - -DEFINE_SHOW_ATTRIBUTE(mic_intr); - -/* - * mic_create_card_debug_dir - Initialize MIC debugfs entries. - */ -void __init mic_create_card_debug_dir(struct mic_driver *mdrv) -{ - if (!mic_dbg) - return; - - mdrv->dbg_dir = debugfs_create_dir(mdrv->name, mic_dbg); - - debugfs_create_file("intr_test", 0444, mdrv->dbg_dir, mdrv, - &mic_intr_fops); -} - -/* - * mic_delete_card_debug_dir - Uninitialize MIC debugfs entries. - */ -void mic_delete_card_debug_dir(struct mic_driver *mdrv) -{ - debugfs_remove_recursive(mdrv->dbg_dir); -} - -/* - * mic_init_card_debugfs - Initialize global debugfs entry. - */ -void __init mic_init_card_debugfs(void) -{ - mic_dbg = debugfs_create_dir(KBUILD_MODNAME, NULL); -} - -/* - * mic_exit_card_debugfs - Uninitialize global debugfs entry - */ -void mic_exit_card_debugfs(void) -{ - debugfs_remove(mic_dbg); -} diff --git a/drivers/misc/mic/card/mic_device.c b/drivers/misc/mic/card/mic_device.c deleted file mode 100644 index a15606259bdc..000000000000 --- a/drivers/misc/mic/card/mic_device.c +++ /dev/null @@ -1,417 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Disclaimer: The codes contained in these modules may be specific to - * the Intel Software Development Platform codenamed: Knights Ferry, and - * the Intel product codenamed: Knights Corner, and are not backward - * compatible with other Intel products. Additionally, Intel will NOT - * support the codes or instruction set in future products. - * - * Intel MIC Card driver. - */ -#include -#include -#include -#include -#include -#include - -#include -#include "../common/mic_dev.h" -#include "mic_device.h" - -static struct mic_driver *g_drv; - -static int __init mic_dp_init(void) -{ - struct mic_driver *mdrv = g_drv; - struct mic_device *mdev = &mdrv->mdev; - struct mic_bootparam __iomem *bootparam; - u64 lo, hi, dp_dma_addr; - u32 magic; - - lo = mic_read_spad(&mdrv->mdev, MIC_DPLO_SPAD); - hi = mic_read_spad(&mdrv->mdev, MIC_DPHI_SPAD); - - dp_dma_addr = lo | (hi << 32); - mdrv->dp = mic_card_map(mdev, dp_dma_addr, MIC_DP_SIZE); - if (!mdrv->dp) { - dev_err(mdrv->dev, "Cannot remap Aperture BAR\n"); - return -ENOMEM; - } - bootparam = mdrv->dp; - magic = ioread32(&bootparam->magic); - if (MIC_MAGIC != magic) { - dev_err(mdrv->dev, "bootparam magic mismatch 0x%x\n", magic); - return -EIO; - } - return 0; -} - -/* Uninitialize the device page */ -static void mic_dp_uninit(void) -{ - mic_card_unmap(&g_drv->mdev, g_drv->dp); -} - -/** - * mic_request_card_irq - request an irq. - * - * @handler: interrupt handler passed to request_threaded_irq. - * @thread_fn: thread fn. passed to request_threaded_irq. - * @name: The ASCII name of the callee requesting the irq. - * @data: private data that is returned back when calling the - * function handler. - * @index: The doorbell index of the requester. - * - * returns: The cookie that is transparent to the caller. Passed - * back when calling mic_free_irq. An appropriate error code - * is returned on failure. Caller needs to use IS_ERR(return_val) - * to check for failure and PTR_ERR(return_val) to obtained the - * error code. - * - */ -struct mic_irq * -mic_request_card_irq(irq_handler_t handler, - irq_handler_t thread_fn, const char *name, - void *data, int index) -{ - int rc = 0; - unsigned long cookie; - struct mic_driver *mdrv = g_drv; - - rc = request_threaded_irq(mic_db_to_irq(mdrv, index), handler, - thread_fn, 0, name, data); - if (rc) { - dev_err(mdrv->dev, "request_threaded_irq failed rc = %d\n", rc); - goto err; - } - mdrv->irq_info.irq_usage_count[index]++; - cookie = index; - return (struct mic_irq *)cookie; -err: - return ERR_PTR(rc); -} - -/** - * mic_free_card_irq - free irq. - * - * @cookie: cookie obtained during a successful call to mic_request_threaded_irq - * @data: private data specified by the calling function during the - * mic_request_threaded_irq - * - * returns: none. - */ -void mic_free_card_irq(struct mic_irq *cookie, void *data) -{ - int index; - struct mic_driver *mdrv = g_drv; - - index = (unsigned long)cookie & 0xFFFFU; - free_irq(mic_db_to_irq(mdrv, index), data); - mdrv->irq_info.irq_usage_count[index]--; -} - -/** - * mic_next_card_db - Get the doorbell with minimum usage count. - * - * Returns the irq index. - */ -int mic_next_card_db(void) -{ - int i; - int index = 0; - struct mic_driver *mdrv = g_drv; - - for (i = 0; i < mdrv->intr_info.num_intr; i++) { - if (mdrv->irq_info.irq_usage_count[i] < - mdrv->irq_info.irq_usage_count[index]) - index = i; - } - - return index; -} - -/** - * mic_init_irq - Initialize irq information. - * - * Returns 0 in success. Appropriate error code on failure. - */ -static int mic_init_irq(void) -{ - struct mic_driver *mdrv = g_drv; - - mdrv->irq_info.irq_usage_count = kzalloc((sizeof(u32) * - mdrv->intr_info.num_intr), - GFP_KERNEL); - if (!mdrv->irq_info.irq_usage_count) - return -ENOMEM; - return 0; -} - -/** - * mic_uninit_irq - Uninitialize irq information. - * - * None. - */ -static void mic_uninit_irq(void) -{ - struct mic_driver *mdrv = g_drv; - - kfree(mdrv->irq_info.irq_usage_count); -} - -static inline struct mic_driver *scdev_to_mdrv(struct scif_hw_dev *scdev) -{ - return dev_get_drvdata(scdev->dev.parent); -} - -static struct mic_irq * -___mic_request_irq(struct scif_hw_dev *scdev, - irqreturn_t (*func)(int irq, void *data), - const char *name, void *data, - int db) -{ - return mic_request_card_irq(func, NULL, name, data, db); -} - -static void -___mic_free_irq(struct scif_hw_dev *scdev, - struct mic_irq *cookie, void *data) -{ - return mic_free_card_irq(cookie, data); -} - -static void ___mic_ack_interrupt(struct scif_hw_dev *scdev, int num) -{ - struct mic_driver *mdrv = scdev_to_mdrv(scdev); - - mic_ack_interrupt(&mdrv->mdev); -} - -static int ___mic_next_db(struct scif_hw_dev *scdev) -{ - return mic_next_card_db(); -} - -static void ___mic_send_intr(struct scif_hw_dev *scdev, int db) -{ - struct mic_driver *mdrv = scdev_to_mdrv(scdev); - - mic_send_intr(&mdrv->mdev, db); -} - -static void ___mic_send_p2p_intr(struct scif_hw_dev *scdev, int db, - struct mic_mw *mw) -{ - mic_send_p2p_intr(db, mw); -} - -static void __iomem * -___mic_ioremap(struct scif_hw_dev *scdev, - phys_addr_t pa, size_t len) -{ - struct mic_driver *mdrv = scdev_to_mdrv(scdev); - - return mic_card_map(&mdrv->mdev, pa, len); -} - -static void ___mic_iounmap(struct scif_hw_dev *scdev, void __iomem *va) -{ - struct mic_driver *mdrv = scdev_to_mdrv(scdev); - - mic_card_unmap(&mdrv->mdev, va); -} - -static struct scif_hw_ops scif_hw_ops = { - .request_irq = ___mic_request_irq, - .free_irq = ___mic_free_irq, - .ack_interrupt = ___mic_ack_interrupt, - .next_db = ___mic_next_db, - .send_intr = ___mic_send_intr, - .send_p2p_intr = ___mic_send_p2p_intr, - .remap = ___mic_ioremap, - .unmap = ___mic_iounmap, -}; - -static inline struct mic_driver *vpdev_to_mdrv(struct vop_device *vpdev) -{ - return dev_get_drvdata(vpdev->dev.parent); -} - -static struct mic_irq * -__mic_request_irq(struct vop_device *vpdev, - irqreturn_t (*func)(int irq, void *data), - const char *name, void *data, int intr_src) -{ - return mic_request_card_irq(func, NULL, name, data, intr_src); -} - -static void __mic_free_irq(struct vop_device *vpdev, - struct mic_irq *cookie, void *data) -{ - return mic_free_card_irq(cookie, data); -} - -static void __mic_ack_interrupt(struct vop_device *vpdev, int num) -{ - struct mic_driver *mdrv = vpdev_to_mdrv(vpdev); - - mic_ack_interrupt(&mdrv->mdev); -} - -static int __mic_next_db(struct vop_device *vpdev) -{ - return mic_next_card_db(); -} - -static void __iomem *__mic_get_remote_dp(struct vop_device *vpdev) -{ - struct mic_driver *mdrv = vpdev_to_mdrv(vpdev); - - return mdrv->dp; -} - -static void __mic_send_intr(struct vop_device *vpdev, int db) -{ - struct mic_driver *mdrv = vpdev_to_mdrv(vpdev); - - mic_send_intr(&mdrv->mdev, db); -} - -static void __iomem *__mic_ioremap(struct vop_device *vpdev, - dma_addr_t pa, size_t len) -{ - struct mic_driver *mdrv = vpdev_to_mdrv(vpdev); - - return mic_card_map(&mdrv->mdev, pa, len); -} - -static void __mic_iounmap(struct vop_device *vpdev, void __iomem *va) -{ - struct mic_driver *mdrv = vpdev_to_mdrv(vpdev); - - mic_card_unmap(&mdrv->mdev, va); -} - -static struct vop_hw_ops vop_hw_ops = { - .request_irq = __mic_request_irq, - .free_irq = __mic_free_irq, - .ack_interrupt = __mic_ack_interrupt, - .next_db = __mic_next_db, - .get_remote_dp = __mic_get_remote_dp, - .send_intr = __mic_send_intr, - .remap = __mic_ioremap, - .unmap = __mic_iounmap, -}; - -static int mic_request_dma_chans(struct mic_driver *mdrv) -{ - dma_cap_mask_t mask; - struct dma_chan *chan; - - dma_cap_zero(mask); - dma_cap_set(DMA_MEMCPY, mask); - - do { - chan = dma_request_channel(mask, NULL, NULL); - if (chan) { - mdrv->dma_ch[mdrv->num_dma_ch++] = chan; - if (mdrv->num_dma_ch >= MIC_MAX_DMA_CHAN) - break; - } - } while (chan); - dev_info(mdrv->dev, "DMA channels # %d\n", mdrv->num_dma_ch); - return mdrv->num_dma_ch; -} - -static void mic_free_dma_chans(struct mic_driver *mdrv) -{ - int i = 0; - - for (i = 0; i < mdrv->num_dma_ch; i++) { - dma_release_channel(mdrv->dma_ch[i]); - mdrv->dma_ch[i] = NULL; - } - mdrv->num_dma_ch = 0; -} - -/* - * mic_driver_init - MIC driver initialization tasks. - * - * Returns 0 in success. Appropriate error code on failure. - */ -int __init mic_driver_init(struct mic_driver *mdrv) -{ - int rc; - struct mic_bootparam __iomem *bootparam; - u8 node_id; - - g_drv = mdrv; - /* Unloading the card module is not supported. */ - if (!try_module_get(mdrv->dev->driver->owner)) { - rc = -ENODEV; - goto done; - } - rc = mic_dp_init(); - if (rc) - goto put; - rc = mic_init_irq(); - if (rc) - goto dp_uninit; - if (!mic_request_dma_chans(mdrv)) { - rc = -ENODEV; - goto irq_uninit; - } - mdrv->vpdev = vop_register_device(mdrv->dev, VOP_DEV_TRNSP, - NULL, &vop_hw_ops, 0, - NULL, mdrv->dma_ch[0]); - if (IS_ERR(mdrv->vpdev)) { - rc = PTR_ERR(mdrv->vpdev); - goto dma_free; - } - bootparam = mdrv->dp; - node_id = ioread8(&bootparam->node_id); - mdrv->scdev = scif_register_device(mdrv->dev, MIC_SCIF_DEV, - NULL, &scif_hw_ops, - 0, node_id, &mdrv->mdev.mmio, NULL, - NULL, mdrv->dp, mdrv->dma_ch, - mdrv->num_dma_ch, true); - if (IS_ERR(mdrv->scdev)) { - rc = PTR_ERR(mdrv->scdev); - goto vop_remove; - } - mic_create_card_debug_dir(mdrv); -done: - return rc; -vop_remove: - vop_unregister_device(mdrv->vpdev); -dma_free: - mic_free_dma_chans(mdrv); -irq_uninit: - mic_uninit_irq(); -dp_uninit: - mic_dp_uninit(); -put: - module_put(mdrv->dev->driver->owner); - return rc; -} - -/* - * mic_driver_uninit - MIC driver uninitialization tasks. - * - * Returns None - */ -void mic_driver_uninit(struct mic_driver *mdrv) -{ - mic_delete_card_debug_dir(mdrv); - scif_unregister_device(mdrv->scdev); - vop_unregister_device(mdrv->vpdev); - mic_free_dma_chans(mdrv); - mic_uninit_irq(); - mic_dp_uninit(); - module_put(mdrv->dev->driver->owner); -} diff --git a/drivers/misc/mic/card/mic_device.h b/drivers/misc/mic/card/mic_device.h deleted file mode 100644 index d6cc69a235a3..000000000000 --- a/drivers/misc/mic/card/mic_device.h +++ /dev/null @@ -1,137 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Disclaimer: The codes contained in these modules may be specific to - * the Intel Software Development Platform codenamed: Knights Ferry, and - * the Intel product codenamed: Knights Corner, and are not backward - * compatible with other Intel products. Additionally, Intel will NOT - * support the codes or instruction set in future products. - * - * Intel MIC Card driver. - */ -#ifndef _MIC_CARD_DEVICE_H_ -#define _MIC_CARD_DEVICE_H_ - -#include -#include -#include -#include -#include "../bus/scif_bus.h" -#include "../bus/vop_bus.h" - -/** - * struct mic_intr_info - Contains h/w specific interrupt sources info - * - * @num_intr: The number of irqs available - */ -struct mic_intr_info { - u32 num_intr; -}; - -/** - * struct mic_irq_info - OS specific irq information - * - * @irq_usage_count: usage count array tracking the number of sources - * assigned for each irq. - */ -struct mic_irq_info { - int *irq_usage_count; -}; - -/** - * struct mic_device - MIC device information. - * - * @mmio: MMIO bar information. - */ -struct mic_device { - struct mic_mw mmio; -}; - -/** - * struct mic_driver - MIC card driver information. - * - * @name: Name for MIC driver. - * @dbg_dir: debugfs directory of this MIC device. - * @dev: The device backing this MIC. - * @dp: The pointer to the virtio device page. - * @mdev: MIC device information for the host. - * @hotplug_work: Hot plug work for adding/removing virtio devices. - * @irq_info: The OS specific irq information - * @intr_info: H/W specific interrupt information. - * @dma_mbdev: dma device on the MIC virtual bus. - * @dma_ch - Array of DMA channels - * @num_dma_ch - Number of DMA channels available - * @scdev: SCIF device on the SCIF virtual bus. - * @vpdev: Virtio over PCIe device on the VOP virtual bus. - */ -struct mic_driver { - char name[20]; - struct dentry *dbg_dir; - struct device *dev; - void __iomem *dp; - struct mic_device mdev; - struct work_struct hotplug_work; - struct mic_irq_info irq_info; - struct mic_intr_info intr_info; - struct mbus_device *dma_mbdev; - struct dma_chan *dma_ch[MIC_MAX_DMA_CHAN]; - int num_dma_ch; - struct scif_hw_dev *scdev; - struct vop_device *vpdev; -}; - -/** - * struct mic_irq - opaque pointer used as cookie - */ -struct mic_irq; - -/** - * mic_mmio_read - read from an MMIO register. - * @mw: MMIO register base virtual address. - * @offset: register offset. - * - * RETURNS: register value. - */ -static inline u32 mic_mmio_read(struct mic_mw *mw, u32 offset) -{ - return ioread32(mw->va + offset); -} - -/** - * mic_mmio_write - write to an MMIO register. - * @mw: MMIO register base virtual address. - * @val: the data value to put into the register - * @offset: register offset. - * - * RETURNS: none. - */ -static inline void -mic_mmio_write(struct mic_mw *mw, u32 val, u32 offset) -{ - iowrite32(val, mw->va + offset); -} - -int mic_driver_init(struct mic_driver *mdrv); -void mic_driver_uninit(struct mic_driver *mdrv); -int mic_next_card_db(void); -struct mic_irq * -mic_request_card_irq(irq_handler_t handler, irq_handler_t thread_fn, - const char *name, void *data, int db); -void mic_free_card_irq(struct mic_irq *cookie, void *data); -u32 mic_read_spad(struct mic_device *mdev, unsigned int idx); -void mic_send_intr(struct mic_device *mdev, int doorbell); -void mic_send_p2p_intr(int doorbell, struct mic_mw *mw); -int mic_db_to_irq(struct mic_driver *mdrv, int db); -u32 mic_ack_interrupt(struct mic_device *mdev); -void mic_hw_intr_init(struct mic_driver *mdrv); -void __iomem * -mic_card_map(struct mic_device *mdev, dma_addr_t addr, size_t size); -void mic_card_unmap(struct mic_device *mdev, void __iomem *addr); -void __init mic_create_card_debug_dir(struct mic_driver *mdrv); -void mic_delete_card_debug_dir(struct mic_driver *mdrv); -void __init mic_init_card_debugfs(void); -void mic_exit_card_debugfs(void); -#endif diff --git a/drivers/misc/mic/card/mic_x100.c b/drivers/misc/mic/card/mic_x100.c deleted file mode 100644 index c8bff2916d3d..000000000000 --- a/drivers/misc/mic/card/mic_x100.c +++ /dev/null @@ -1,347 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Disclaimer: The codes contained in these modules may be specific to - * the Intel Software Development Platform codenamed: Knights Ferry, and - * the Intel product codenamed: Knights Corner, and are not backward - * compatible with other Intel products. Additionally, Intel will NOT - * support the codes or instruction set in future products. - * - * Intel MIC Card driver. - */ -#include -#include -#include - -#include "../common/mic_dev.h" -#include "mic_device.h" -#include "mic_x100.h" - -static const char mic_driver_name[] = "mic"; - -static struct mic_driver g_drv; - -/** - * mic_read_spad - read from the scratchpad register - * @mdev: pointer to mic_device instance - * @idx: index to scratchpad register, 0 based - * - * This function allows reading of the 32bit scratchpad register. - * - * RETURNS: An appropriate -ERRNO error value on error, or zero for success. - */ -u32 mic_read_spad(struct mic_device *mdev, unsigned int idx) -{ - return mic_mmio_read(&mdev->mmio, - MIC_X100_SBOX_BASE_ADDRESS + - MIC_X100_SBOX_SPAD0 + idx * 4); -} - -/** - * __mic_send_intr - Send interrupt to Host. - * @mdev: pointer to mic_device instance - * @doorbell: Doorbell number. - */ -void mic_send_intr(struct mic_device *mdev, int doorbell) -{ - struct mic_mw *mw = &mdev->mmio; - - if (doorbell > MIC_X100_MAX_DOORBELL_IDX) - return; - /* Ensure that the interrupt is ordered w.r.t previous stores. */ - wmb(); - mic_mmio_write(mw, MIC_X100_SBOX_SDBIC0_DBREQ_BIT, - MIC_X100_SBOX_BASE_ADDRESS + - (MIC_X100_SBOX_SDBIC0 + (4 * doorbell))); -} - -/* - * mic_x100_send_sbox_intr - Send an MIC_X100_SBOX interrupt to MIC. - */ -static void mic_x100_send_sbox_intr(struct mic_mw *mw, int doorbell) -{ - u64 apic_icr_offset = MIC_X100_SBOX_APICICR0 + doorbell * 8; - u32 apicicr_low = mic_mmio_read(mw, MIC_X100_SBOX_BASE_ADDRESS + - apic_icr_offset); - - /* for MIC we need to make sure we "hit" the send_icr bit (13) */ - apicicr_low = (apicicr_low | (1 << 13)); - /* - * Ensure that the interrupt is ordered w.r.t. previous stores - * to main memory. Fence instructions are not implemented in X100 - * since execution is in order but a compiler barrier is still - * required. - */ - wmb(); - mic_mmio_write(mw, apicicr_low, - MIC_X100_SBOX_BASE_ADDRESS + apic_icr_offset); -} - -static void mic_x100_send_rdmasr_intr(struct mic_mw *mw, int doorbell) -{ - int rdmasr_offset = MIC_X100_SBOX_RDMASR0 + (doorbell << 2); - /* - * Ensure that the interrupt is ordered w.r.t. previous stores - * to main memory. Fence instructions are not implemented in X100 - * since execution is in order but a compiler barrier is still - * required. - */ - wmb(); - mic_mmio_write(mw, 0, MIC_X100_SBOX_BASE_ADDRESS + rdmasr_offset); -} - -/** - * mic_ack_interrupt - Device specific interrupt handling. - * @mdev: pointer to mic_device instance - * - * Returns: bitmask of doorbell events triggered. - */ -u32 mic_ack_interrupt(struct mic_device *mdev) -{ - return 0; -} - -static inline int mic_get_sbox_irq(int db) -{ - return MIC_X100_IRQ_BASE + db; -} - -static inline int mic_get_rdmasr_irq(int index) -{ - return MIC_X100_RDMASR_IRQ_BASE + index; -} - -void mic_send_p2p_intr(int db, struct mic_mw *mw) -{ - int rdmasr_index; - - if (db < MIC_X100_NUM_SBOX_IRQ) { - mic_x100_send_sbox_intr(mw, db); - } else { - rdmasr_index = db - MIC_X100_NUM_SBOX_IRQ; - mic_x100_send_rdmasr_intr(mw, rdmasr_index); - } -} - -/** - * mic_hw_intr_init - Initialize h/w specific interrupt - * information. - * @mdrv: pointer to mic_driver - */ -void mic_hw_intr_init(struct mic_driver *mdrv) -{ - mdrv->intr_info.num_intr = MIC_X100_NUM_SBOX_IRQ + - MIC_X100_NUM_RDMASR_IRQ; -} - -/** - * mic_db_to_irq - Retrieve irq number corresponding to a doorbell. - * @mdrv: pointer to mic_driver - * @db: The doorbell obtained for which the irq is needed. Doorbell - * may correspond to an sbox doorbell or an rdmasr index. - * - * Returns the irq corresponding to the doorbell. - */ -int mic_db_to_irq(struct mic_driver *mdrv, int db) -{ - int rdmasr_index; - - /* - * The total number of doorbell interrupts on the card are 16. Indices - * 0-8 falls in the SBOX category and 8-15 fall in the RDMASR category. - */ - if (db < MIC_X100_NUM_SBOX_IRQ) { - return mic_get_sbox_irq(db); - } else { - rdmasr_index = db - MIC_X100_NUM_SBOX_IRQ; - return mic_get_rdmasr_irq(rdmasr_index); - } -} - -/* - * mic_card_map - Allocate virtual address for a remote memory region. - * @mdev: pointer to mic_device instance. - * @addr: Remote DMA address. - * @size: Size of the region. - * - * Returns: Virtual address backing the remote memory region. - */ -void __iomem * -mic_card_map(struct mic_device *mdev, dma_addr_t addr, size_t size) -{ - return ioremap(addr, size); -} - -/* - * mic_card_unmap - Unmap the virtual address for a remote memory region. - * @mdev: pointer to mic_device instance. - * @addr: Virtual address for remote memory region. - * - * Returns: None. - */ -void mic_card_unmap(struct mic_device *mdev, void __iomem *addr) -{ - iounmap(addr); -} - -static inline struct mic_driver *mbdev_to_mdrv(struct mbus_device *mbdev) -{ - return dev_get_drvdata(mbdev->dev.parent); -} - -static struct mic_irq * -_mic_request_threaded_irq(struct mbus_device *mbdev, - irq_handler_t handler, irq_handler_t thread_fn, - const char *name, void *data, int intr_src) -{ - int rc = 0; - unsigned int irq = intr_src; - unsigned long cookie = irq; - - rc = request_threaded_irq(irq, handler, thread_fn, 0, name, data); - if (rc) { - dev_err(mbdev_to_mdrv(mbdev)->dev, - "request_threaded_irq failed rc = %d\n", rc); - return ERR_PTR(rc); - } - return (struct mic_irq *)cookie; -} - -static void _mic_free_irq(struct mbus_device *mbdev, - struct mic_irq *cookie, void *data) -{ - unsigned long irq = (unsigned long)cookie; - free_irq(irq, data); -} - -static void _mic_ack_interrupt(struct mbus_device *mbdev, int num) -{ - mic_ack_interrupt(&mbdev_to_mdrv(mbdev)->mdev); -} - -static struct mbus_hw_ops mbus_hw_ops = { - .request_threaded_irq = _mic_request_threaded_irq, - .free_irq = _mic_free_irq, - .ack_interrupt = _mic_ack_interrupt, -}; - -static int __init mic_probe(struct platform_device *pdev) -{ - struct mic_driver *mdrv = &g_drv; - struct mic_device *mdev = &mdrv->mdev; - int rc = 0; - - mdrv->dev = &pdev->dev; - snprintf(mdrv->name, sizeof(mic_driver_name), mic_driver_name); - - /* FIXME: use dma_set_mask_and_coherent() and check result */ - dma_coerce_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64)); - - mdev->mmio.pa = MIC_X100_MMIO_BASE; - mdev->mmio.len = MIC_X100_MMIO_LEN; - mdev->mmio.va = devm_ioremap(&pdev->dev, MIC_X100_MMIO_BASE, - MIC_X100_MMIO_LEN); - if (!mdev->mmio.va) { - dev_err(&pdev->dev, "Cannot remap MMIO BAR\n"); - rc = -EIO; - goto done; - } - mic_hw_intr_init(mdrv); - platform_set_drvdata(pdev, mdrv); - mdrv->dma_mbdev = mbus_register_device(mdrv->dev, MBUS_DEV_DMA_MIC, - NULL, &mbus_hw_ops, 0, - mdrv->mdev.mmio.va); - if (IS_ERR(mdrv->dma_mbdev)) { - rc = PTR_ERR(mdrv->dma_mbdev); - dev_err(&pdev->dev, "mbus_add_device failed rc %d\n", rc); - goto done; - } - rc = mic_driver_init(mdrv); - if (rc) { - dev_err(&pdev->dev, "mic_driver_init failed rc %d\n", rc); - goto remove_dma; - } -done: - return rc; -remove_dma: - mbus_unregister_device(mdrv->dma_mbdev); - return rc; -} - -static int mic_remove(struct platform_device *pdev) -{ - struct mic_driver *mdrv = &g_drv; - - mic_driver_uninit(mdrv); - mbus_unregister_device(mdrv->dma_mbdev); - return 0; -} - -static void mic_platform_shutdown(struct platform_device *pdev) -{ - mic_remove(pdev); -} - -static struct platform_driver __refdata mic_platform_driver = { - .probe = mic_probe, - .remove = mic_remove, - .shutdown = mic_platform_shutdown, - .driver = { - .name = mic_driver_name, - }, -}; - -static struct platform_device *mic_platform_dev; - -static int __init mic_init(void) -{ - int ret; - struct cpuinfo_x86 *c = &cpu_data(0); - - if (!(c->x86 == 11 && c->x86_model == 1)) { - ret = -ENODEV; - pr_err("%s not running on X100 ret %d\n", __func__, ret); - goto done; - } - - request_module("mic_x100_dma"); - mic_init_card_debugfs(); - - mic_platform_dev = platform_device_register_simple(mic_driver_name, - 0, NULL, 0); - ret = PTR_ERR_OR_ZERO(mic_platform_dev); - if (ret) { - pr_err("platform_device_register_full ret %d\n", ret); - goto cleanup_debugfs; - } - ret = platform_driver_register(&mic_platform_driver); - if (ret) { - pr_err("platform_driver_register ret %d\n", ret); - goto device_unregister; - } - return ret; - -device_unregister: - platform_device_unregister(mic_platform_dev); -cleanup_debugfs: - mic_exit_card_debugfs(); -done: - return ret; -} - -static void __exit mic_exit(void) -{ - platform_driver_unregister(&mic_platform_driver); - platform_device_unregister(mic_platform_dev); - mic_exit_card_debugfs(); -} - -module_init(mic_init); -module_exit(mic_exit); - -MODULE_AUTHOR("Intel Corporation"); -MODULE_DESCRIPTION("Intel(R) MIC X100 Card driver"); -MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/card/mic_x100.h b/drivers/misc/mic/card/mic_x100.h deleted file mode 100644 index 46644dde0c07..000000000000 --- a/drivers/misc/mic/card/mic_x100.h +++ /dev/null @@ -1,37 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Disclaimer: The codes contained in these modules may be specific to - * the Intel Software Development Platform codenamed: Knights Ferry, and - * the Intel product codenamed: Knights Corner, and are not backward - * compatible with other Intel products. Additionally, Intel will NOT - * support the codes or instruction set in future products. - * - * Intel MIC Card driver. - */ -#ifndef _MIC_X100_CARD_H_ -#define _MIC_X100_CARD_H_ - -#define MIC_X100_MMIO_BASE 0x08007C0000ULL -#define MIC_X100_MMIO_LEN 0x00020000ULL -#define MIC_X100_SBOX_BASE_ADDRESS 0x00010000ULL - -#define MIC_X100_SBOX_SPAD0 0x0000AB20 -#define MIC_X100_SBOX_SDBIC0 0x0000CC90 -#define MIC_X100_SBOX_SDBIC0_DBREQ_BIT 0x80000000 -#define MIC_X100_SBOX_RDMASR0 0x0000B180 -#define MIC_X100_SBOX_APICICR0 0x0000A9D0 - -#define MIC_X100_MAX_DOORBELL_IDX 8 - -#define MIC_X100_NUM_SBOX_IRQ 8 -#define MIC_X100_NUM_RDMASR_IRQ 8 -#define MIC_X100_SBOX_IRQ_BASE 0 -#define MIC_X100_RDMASR_IRQ_BASE 17 - -#define MIC_X100_IRQ_BASE 26 - -#endif diff --git a/drivers/misc/mic/common/mic_dev.h b/drivers/misc/mic/common/mic_dev.h deleted file mode 100644 index f94f08df0260..000000000000 --- a/drivers/misc/mic/common/mic_dev.h +++ /dev/null @@ -1,55 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Intel MIC driver. - */ -#ifndef __MIC_DEV_H__ -#define __MIC_DEV_H__ - -/* The maximum number of MIC devices supported in a single host system. */ -#define MIC_MAX_NUM_DEVS 128 - -/** - * enum mic_hw_family - The hardware family to which a device belongs. - */ -enum mic_hw_family { - MIC_FAMILY_X100 = 0, - MIC_FAMILY_X200, - MIC_FAMILY_UNKNOWN, - MIC_FAMILY_LAST -}; - -/** - * struct mic_mw - MIC memory window - * - * @pa: Base physical address. - * @va: Base ioremap'd virtual address. - * @len: Size of the memory window. - */ -struct mic_mw { - phys_addr_t pa; - void __iomem *va; - resource_size_t len; -}; - -/* - * Scratch pad register offsets used by the host to communicate - * device page DMA address to the card. - */ -#define MIC_DPLO_SPAD 14 -#define MIC_DPHI_SPAD 15 - -/* - * These values are supposed to be in the config_change field of the - * device page when the host sends a config change interrupt to the card. - */ -#define MIC_VIRTIO_PARAM_DEV_REMOVE 0x1 -#define MIC_VIRTIO_PARAM_CONFIG_CHANGED 0x2 - -/* Maximum number of DMA channels */ -#define MIC_MAX_DMA_CHAN 4 - -#endif diff --git a/drivers/misc/mic/cosm/Makefile b/drivers/misc/mic/cosm/Makefile deleted file mode 100644 index 97d74cb12030..000000000000 --- a/drivers/misc/mic/cosm/Makefile +++ /dev/null @@ -1,11 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0 -# -# Makefile - Intel MIC Coprocessor State Management (COSM) Driver -# Copyright(c) 2015, Intel Corporation. -# -obj-$(CONFIG_MIC_COSM) += mic_cosm.o - -mic_cosm-objs := cosm_main.o -mic_cosm-objs += cosm_debugfs.o -mic_cosm-objs += cosm_sysfs.o -mic_cosm-objs += cosm_scif_server.o diff --git a/drivers/misc/mic/cosm/cosm_debugfs.c b/drivers/misc/mic/cosm/cosm_debugfs.c deleted file mode 100644 index cb55653cf1f9..000000000000 --- a/drivers/misc/mic/cosm/cosm_debugfs.c +++ /dev/null @@ -1,116 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2015 Intel Corporation. - * - * Intel MIC Coprocessor State Management (COSM) Driver - */ - -#include -#include -#include -#include "cosm_main.h" - -/* Debugfs parent dir */ -static struct dentry *cosm_dbg; - -/* - * log_buf_show - Display MIC kernel log buffer - * - * log_buf addr/len is read from System.map by user space - * and populated in sysfs entries. - */ -static int log_buf_show(struct seq_file *s, void *unused) -{ - void __iomem *log_buf_va; - int __iomem *log_buf_len_va; - struct cosm_device *cdev = s->private; - void *kva; - int size; - u64 aper_offset; - - if (!cdev || !cdev->log_buf_addr || !cdev->log_buf_len) - goto done; - - mutex_lock(&cdev->cosm_mutex); - switch (cdev->state) { - case MIC_BOOTING: - case MIC_ONLINE: - case MIC_SHUTTING_DOWN: - break; - default: - goto unlock; - } - - /* - * Card kernel will never be relocated and any kernel text/data mapping - * can be translated to phys address by subtracting __START_KERNEL_map. - */ - aper_offset = (u64)cdev->log_buf_len - __START_KERNEL_map; - log_buf_len_va = cdev->hw_ops->aper(cdev)->va + aper_offset; - aper_offset = (u64)cdev->log_buf_addr - __START_KERNEL_map; - log_buf_va = cdev->hw_ops->aper(cdev)->va + aper_offset; - - size = ioread32(log_buf_len_va); - kva = kmalloc(size, GFP_KERNEL); - if (!kva) - goto unlock; - - memcpy_fromio(kva, log_buf_va, size); - seq_write(s, kva, size); - kfree(kva); -unlock: - mutex_unlock(&cdev->cosm_mutex); -done: - return 0; -} - -DEFINE_SHOW_ATTRIBUTE(log_buf); - -/* - * force_reset_show - Force MIC reset - * - * Invokes the force_reset COSM bus op instead of the standard reset - * op in case a force reset of the MIC device is required - */ -static int force_reset_show(struct seq_file *s, void *pos) -{ - struct cosm_device *cdev = s->private; - - cosm_stop(cdev, true); - return 0; -} - -DEFINE_SHOW_ATTRIBUTE(force_reset); - -void cosm_create_debug_dir(struct cosm_device *cdev) -{ - char name[16]; - - if (!cosm_dbg) - return; - - scnprintf(name, sizeof(name), "mic%d", cdev->index); - cdev->dbg_dir = debugfs_create_dir(name, cosm_dbg); - - debugfs_create_file("log_buf", 0444, cdev->dbg_dir, cdev, - &log_buf_fops); - debugfs_create_file("force_reset", 0444, cdev->dbg_dir, cdev, - &force_reset_fops); -} - -void cosm_delete_debug_dir(struct cosm_device *cdev) -{ - debugfs_remove_recursive(cdev->dbg_dir); -} - -void cosm_init_debugfs(void) -{ - cosm_dbg = debugfs_create_dir(KBUILD_MODNAME, NULL); -} - -void cosm_exit_debugfs(void) -{ - debugfs_remove(cosm_dbg); -} diff --git a/drivers/misc/mic/cosm/cosm_main.c b/drivers/misc/mic/cosm/cosm_main.c deleted file mode 100644 index ebb0eac43754..000000000000 --- a/drivers/misc/mic/cosm/cosm_main.c +++ /dev/null @@ -1,382 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2015 Intel Corporation. - * - * Intel MIC Coprocessor State Management (COSM) Driver - */ - -#include -#include -#include -#include -#include -#include "cosm_main.h" - -static const char cosm_driver_name[] = "mic"; - -/* COSM ID allocator */ -static struct ida g_cosm_ida; -/* Class of MIC devices for sysfs accessibility. */ -static struct class *g_cosm_class; -/* Number of MIC devices */ -static atomic_t g_num_dev; - -/** - * cosm_hw_reset - Issue a HW reset for the MIC device - * @cdev: pointer to cosm_device instance - * @force: force a MIC to reset even if it is already reset and ready - */ -static void cosm_hw_reset(struct cosm_device *cdev, bool force) -{ - int i; - -#define MIC_RESET_TO (45) - if (force && cdev->hw_ops->force_reset) - cdev->hw_ops->force_reset(cdev); - else - cdev->hw_ops->reset(cdev); - - for (i = 0; i < MIC_RESET_TO; i++) { - if (cdev->hw_ops->ready(cdev)) { - cosm_set_state(cdev, MIC_READY); - return; - } - /* - * Resets typically take 10s of seconds to complete. - * Since an MMIO read is required to check if the - * firmware is ready or not, a 1 second delay works nicely. - */ - msleep(1000); - } - cosm_set_state(cdev, MIC_RESET_FAILED); -} - -/** - * cosm_start - Start the MIC - * @cdev: pointer to cosm_device instance - * - * This function prepares an MIC for boot and initiates boot. - * RETURNS: An appropriate -ERRNO error value on error, or 0 for success. - */ -int cosm_start(struct cosm_device *cdev) -{ - const struct cred *orig_cred; - struct cred *override_cred; - int rc; - - mutex_lock(&cdev->cosm_mutex); - if (!cdev->bootmode) { - dev_err(&cdev->dev, "%s %d bootmode not set\n", - __func__, __LINE__); - rc = -EINVAL; - goto unlock_ret; - } -retry: - if (cdev->state != MIC_READY) { - dev_err(&cdev->dev, "%s %d MIC state not READY\n", - __func__, __LINE__); - rc = -EINVAL; - goto unlock_ret; - } - if (!cdev->hw_ops->ready(cdev)) { - cosm_hw_reset(cdev, false); - /* - * The state will either be MIC_READY if the reset succeeded - * or MIC_RESET_FAILED if the firmware reset failed. - */ - goto retry; - } - - /* - * Set credentials to root to allow non-root user to download initramsfs - * with 600 permissions - */ - override_cred = prepare_creds(); - if (!override_cred) { - dev_err(&cdev->dev, "%s %d prepare_creds failed\n", - __func__, __LINE__); - rc = -ENOMEM; - goto unlock_ret; - } - override_cred->fsuid = GLOBAL_ROOT_UID; - orig_cred = override_creds(override_cred); - - rc = cdev->hw_ops->start(cdev, cdev->index); - - revert_creds(orig_cred); - put_cred(override_cred); - if (rc) - goto unlock_ret; - - /* - * If linux is being booted, card is treated 'online' only - * when the scif interface in the card is up. If anything else - * is booted, we set card to 'online' immediately. - */ - if (!strcmp(cdev->bootmode, "linux")) - cosm_set_state(cdev, MIC_BOOTING); - else - cosm_set_state(cdev, MIC_ONLINE); -unlock_ret: - mutex_unlock(&cdev->cosm_mutex); - if (rc) - dev_err(&cdev->dev, "cosm_start failed rc %d\n", rc); - return rc; -} - -/** - * cosm_stop - Prepare the MIC for reset and trigger reset - * @cdev: pointer to cosm_device instance - * @force: force a MIC to reset even if it is already reset and ready. - * - * RETURNS: None - */ -void cosm_stop(struct cosm_device *cdev, bool force) -{ - mutex_lock(&cdev->cosm_mutex); - if (cdev->state != MIC_READY || force) { - /* - * Don't call hw_ops if they have been called previously. - * stop(..) calls device_unregister and will crash the system if - * called multiple times. - */ - u8 state = cdev->state == MIC_RESETTING ? - cdev->prev_state : cdev->state; - bool call_hw_ops = state != MIC_RESET_FAILED && - state != MIC_READY; - - if (cdev->state != MIC_RESETTING) - cosm_set_state(cdev, MIC_RESETTING); - cdev->heartbeat_watchdog_enable = false; - if (call_hw_ops) - cdev->hw_ops->stop(cdev, force); - cosm_hw_reset(cdev, force); - cosm_set_shutdown_status(cdev, MIC_NOP); - if (call_hw_ops && cdev->hw_ops->post_reset) - cdev->hw_ops->post_reset(cdev, cdev->state); - } - mutex_unlock(&cdev->cosm_mutex); - flush_work(&cdev->scif_work); -} - -/** - * cosm_reset_trigger_work - Trigger MIC reset - * @work: The work structure - * - * This work is scheduled whenever the host wants to reset the MIC. - */ -static void cosm_reset_trigger_work(struct work_struct *work) -{ - struct cosm_device *cdev = container_of(work, struct cosm_device, - reset_trigger_work); - cosm_stop(cdev, false); -} - -/** - * cosm_reset - Schedule MIC reset - * @cdev: pointer to cosm_device instance - * - * RETURNS: An -EINVAL if the card is already READY or 0 for success. - */ -int cosm_reset(struct cosm_device *cdev) -{ - int rc = 0; - - mutex_lock(&cdev->cosm_mutex); - if (cdev->state != MIC_READY) { - if (cdev->state != MIC_RESETTING) { - cdev->prev_state = cdev->state; - cosm_set_state(cdev, MIC_RESETTING); - schedule_work(&cdev->reset_trigger_work); - } - } else { - dev_err(&cdev->dev, "%s %d MIC is READY\n", __func__, __LINE__); - rc = -EINVAL; - } - mutex_unlock(&cdev->cosm_mutex); - return rc; -} - -/** - * cosm_shutdown - Initiate MIC shutdown. - * @cdev: pointer to cosm_device instance - * - * RETURNS: None - */ -int cosm_shutdown(struct cosm_device *cdev) -{ - struct cosm_msg msg = { .id = COSM_MSG_SHUTDOWN }; - int rc = 0; - - mutex_lock(&cdev->cosm_mutex); - if (cdev->state != MIC_ONLINE) { - rc = -EINVAL; - dev_err(&cdev->dev, "%s %d skipping shutdown in state: %s\n", - __func__, __LINE__, cosm_state_string[cdev->state]); - goto err; - } - - if (!cdev->epd) { - rc = -ENOTCONN; - dev_err(&cdev->dev, "%s %d scif endpoint not connected rc %d\n", - __func__, __LINE__, rc); - goto err; - } - - rc = scif_send(cdev->epd, &msg, sizeof(msg), SCIF_SEND_BLOCK); - if (rc < 0) { - dev_err(&cdev->dev, "%s %d scif_send failed rc %d\n", - __func__, __LINE__, rc); - goto err; - } - cdev->heartbeat_watchdog_enable = false; - cosm_set_state(cdev, MIC_SHUTTING_DOWN); - rc = 0; -err: - mutex_unlock(&cdev->cosm_mutex); - return rc; -} - -static int cosm_driver_probe(struct cosm_device *cdev) -{ - int rc; - - /* Initialize SCIF server at first probe */ - if (atomic_add_return(1, &g_num_dev) == 1) { - rc = cosm_scif_init(); - if (rc) - goto scif_exit; - } - mutex_init(&cdev->cosm_mutex); - INIT_WORK(&cdev->reset_trigger_work, cosm_reset_trigger_work); - INIT_WORK(&cdev->scif_work, cosm_scif_work); - cdev->sysfs_heartbeat_enable = true; - cosm_sysfs_init(cdev); - cdev->sdev = device_create_with_groups(g_cosm_class, cdev->dev.parent, - MKDEV(0, cdev->index), cdev, cdev->attr_group, - "mic%d", cdev->index); - if (IS_ERR(cdev->sdev)) { - rc = PTR_ERR(cdev->sdev); - dev_err(&cdev->dev, "device_create_with_groups failed rc %d\n", - rc); - goto scif_exit; - } - - cdev->state_sysfs = sysfs_get_dirent(cdev->sdev->kobj.sd, - "state"); - if (!cdev->state_sysfs) { - rc = -ENODEV; - dev_err(&cdev->dev, "sysfs_get_dirent failed rc %d\n", rc); - goto destroy_device; - } - cosm_create_debug_dir(cdev); - return 0; -destroy_device: - device_destroy(g_cosm_class, MKDEV(0, cdev->index)); -scif_exit: - if (atomic_dec_and_test(&g_num_dev)) - cosm_scif_exit(); - return rc; -} - -static void cosm_driver_remove(struct cosm_device *cdev) -{ - cosm_delete_debug_dir(cdev); - sysfs_put(cdev->state_sysfs); - device_destroy(g_cosm_class, MKDEV(0, cdev->index)); - flush_work(&cdev->reset_trigger_work); - cosm_stop(cdev, false); - if (atomic_dec_and_test(&g_num_dev)) - cosm_scif_exit(); - - /* These sysfs entries might have allocated */ - kfree(cdev->cmdline); - kfree(cdev->firmware); - kfree(cdev->ramdisk); - kfree(cdev->bootmode); -} - -static int cosm_suspend(struct device *dev) -{ - struct cosm_device *cdev = dev_to_cosm(dev); - - mutex_lock(&cdev->cosm_mutex); - switch (cdev->state) { - /** - * Suspend/freeze hooks in userspace have already shutdown the card. - * Card should be 'ready' in most cases. It is however possible that - * some userspace application initiated a boot. In those cases, we - * simply reset the card. - */ - case MIC_ONLINE: - case MIC_BOOTING: - case MIC_SHUTTING_DOWN: - mutex_unlock(&cdev->cosm_mutex); - cosm_stop(cdev, false); - break; - default: - mutex_unlock(&cdev->cosm_mutex); - break; - } - return 0; -} - -static const struct dev_pm_ops cosm_pm_ops = { - .suspend = cosm_suspend, - .freeze = cosm_suspend -}; - -static struct cosm_driver cosm_driver = { - .driver = { - .name = KBUILD_MODNAME, - .owner = THIS_MODULE, - .pm = &cosm_pm_ops, - }, - .probe = cosm_driver_probe, - .remove = cosm_driver_remove -}; - -static int __init cosm_init(void) -{ - int ret; - - cosm_init_debugfs(); - - g_cosm_class = class_create(THIS_MODULE, cosm_driver_name); - if (IS_ERR(g_cosm_class)) { - ret = PTR_ERR(g_cosm_class); - pr_err("class_create failed ret %d\n", ret); - goto cleanup_debugfs; - } - - ida_init(&g_cosm_ida); - ret = cosm_register_driver(&cosm_driver); - if (ret) { - pr_err("cosm_register_driver failed ret %d\n", ret); - goto ida_destroy; - } - return 0; -ida_destroy: - ida_destroy(&g_cosm_ida); - class_destroy(g_cosm_class); -cleanup_debugfs: - cosm_exit_debugfs(); - return ret; -} - -static void __exit cosm_exit(void) -{ - cosm_unregister_driver(&cosm_driver); - ida_destroy(&g_cosm_ida); - class_destroy(g_cosm_class); - cosm_exit_debugfs(); -} - -module_init(cosm_init); -module_exit(cosm_exit); - -MODULE_AUTHOR("Intel Corporation"); -MODULE_DESCRIPTION("Intel(R) MIC Coprocessor State Management (COSM) Driver"); -MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/cosm/cosm_main.h b/drivers/misc/mic/cosm/cosm_main.h deleted file mode 100644 index 5188ad245814..000000000000 --- a/drivers/misc/mic/cosm/cosm_main.h +++ /dev/null @@ -1,61 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2015 Intel Corporation. - * - * Intel MIC Coprocessor State Management (COSM) Driver - */ -#ifndef _COSM_COSM_H_ -#define _COSM_COSM_H_ - -#include -#include "../bus/cosm_bus.h" - -#define COSM_HEARTBEAT_SEND_SEC 30 -#define SCIF_COSM_LISTEN_PORT 201 - -/** - * enum COSM msg id's - * @COSM_MSG_SHUTDOWN: host->card trigger shutdown - * @COSM_MSG_SYNC_TIME: host->card send host time to card to sync time - * @COSM_MSG_HEARTBEAT: card->host heartbeat - * @COSM_MSG_SHUTDOWN_STATUS: card->host with shutdown status as payload - */ -enum cosm_msg_id { - COSM_MSG_SHUTDOWN, - COSM_MSG_SYNC_TIME, - COSM_MSG_HEARTBEAT, - COSM_MSG_SHUTDOWN_STATUS, -}; - -struct cosm_msg { - u64 id; - union { - u64 shutdown_status; - struct { - u64 tv_sec; - u64 tv_nsec; - } timespec; - }; -}; - -extern const char * const cosm_state_string[]; -extern const char * const cosm_shutdown_status_string[]; - -void cosm_sysfs_init(struct cosm_device *cdev); -int cosm_start(struct cosm_device *cdev); -void cosm_stop(struct cosm_device *cdev, bool force); -int cosm_reset(struct cosm_device *cdev); -int cosm_shutdown(struct cosm_device *cdev); -void cosm_set_state(struct cosm_device *cdev, u8 state); -void cosm_set_shutdown_status(struct cosm_device *cdev, u8 status); -void cosm_init_debugfs(void); -void cosm_exit_debugfs(void); -void cosm_create_debug_dir(struct cosm_device *cdev); -void cosm_delete_debug_dir(struct cosm_device *cdev); -int cosm_scif_init(void); -void cosm_scif_exit(void); -void cosm_scif_work(struct work_struct *work); - -#endif diff --git a/drivers/misc/mic/cosm/cosm_scif_server.c b/drivers/misc/mic/cosm/cosm_scif_server.c deleted file mode 100644 index 7baec9fd8756..000000000000 --- a/drivers/misc/mic/cosm/cosm_scif_server.c +++ /dev/null @@ -1,399 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2015 Intel Corporation. - * - * Intel MIC Coprocessor State Management (COSM) Driver - */ -#include -#include - -#include "cosm_main.h" - -/* - * The COSM driver uses SCIF to communicate between the management node and the - * MIC cards. SCIF is used to (a) Send a shutdown command to the card (b) - * receive a shutdown status back from the card upon completion of shutdown and - * (c) receive periodic heartbeat messages from the card used to deduce if the - * card has crashed. - * - * A COSM server consisting of a SCIF listening endpoint waits for incoming - * connections from the card. Upon acceptance of the connection, a separate - * work-item is scheduled to handle SCIF message processing for that card. The - * life-time of this work-item is therefore the time from which the connection - * from a card is accepted to the time at which the connection is closed. A new - * work-item starts each time the card boots and is alive till the card (a) - * shuts down (b) is reset (c) crashes (d) cosm_client driver on the card is - * unloaded. - * - * From the point of view of COSM interactions with SCIF during card - * shutdown, reset and crash are as follows: - * - * Card shutdown - * ------------- - * 1. COSM client on the card invokes orderly_poweroff() in response to SHUTDOWN - * message from the host. - * 2. Card driver shutdown callback invokes scif_unregister_device(..) resulting - * in scif_remove(..) getting called on the card - * 3. scif_remove -> scif_stop -> scif_handle_remove_node -> - * scif_peer_unregister_device -> device_unregister for the host peer device - * 4. During device_unregister remove(..) method of cosm_client is invoked which - * closes the COSM SCIF endpoint on the card. This results in a SCIF_DISCNCT - * message being sent to host SCIF. SCIF_DISCNCT message processing on the - * host SCIF sets the host COSM SCIF endpoint state to DISCONNECTED and wakes - * up the host COSM thread blocked in scif_poll(..) resulting in - * scif_poll(..) returning EPOLLHUP. - * 5. On the card, scif_peer_release_dev is next called which results in an - * SCIF_EXIT message being sent to the host and after receiving the - * SCIF_EXIT_ACK from the host the peer device teardown on the card is - * complete. - * 6. As part of the SCIF_EXIT message processing on the host, host sends a - * SCIF_REMOVE_NODE to itself corresponding to the card being removed. This - * starts a similar SCIF peer device teardown sequence on the host - * corresponding to the card being shut down. - * - * Card reset - * ---------- - * The case of interest here is when the card has not been previously shut down - * since most of the steps below are skipped in that case: - - * 1. cosm_stop(..) invokes hw_ops->stop(..) method of the base PCIe driver - * which unregisters the SCIF HW device resulting in scif_remove(..) being - * called on the host. - * 2. scif_remove(..) calls scif_disconnect_node(..) which results in a - * SCIF_EXIT message being sent to the card. - * 3. The card executes scif_stop() as part of SCIF_EXIT message - * processing. This results in the COSM endpoint on the card being closed and - * the SCIF host peer device on the card getting unregistered similar to - * steps 3, 4 and 5 for the card shutdown case above. scif_poll(..) on the - * host returns EPOLLHUP as a result. - * 4. On the host, card peer device unregister and SCIF HW remove(..) also - * subsequently complete. - * - * Card crash - * ---------- - * If a reset is issued after the card has crashed, there is no SCIF_DISCNT - * message from the card which would result in scif_poll(..) returning - * EPOLLHUP. In this case when the host SCIF driver sends a SCIF_REMOVE_NODE - * message to itself resulting in the card SCIF peer device being unregistered, - * this results in a scif_peer_release_dev -> scif_cleanup_scifdev-> - * scif_invalidate_ep call sequence which sets the endpoint state to - * DISCONNECTED and results in scif_poll(..) returning EPOLLHUP. - */ - -#define COSM_SCIF_BACKLOG 16 -#define COSM_HEARTBEAT_CHECK_DELTA_SEC 10 -#define COSM_HEARTBEAT_TIMEOUT_SEC \ - (COSM_HEARTBEAT_SEND_SEC + COSM_HEARTBEAT_CHECK_DELTA_SEC) -#define COSM_HEARTBEAT_TIMEOUT_MSEC (COSM_HEARTBEAT_TIMEOUT_SEC * MSEC_PER_SEC) - -static struct task_struct *server_thread; -static scif_epd_t listen_epd; - -/* Publish MIC card's shutdown status to user space MIC daemon */ -static void cosm_update_mic_status(struct cosm_device *cdev) -{ - if (cdev->shutdown_status_int != MIC_NOP) { - cosm_set_shutdown_status(cdev, cdev->shutdown_status_int); - cdev->shutdown_status_int = MIC_NOP; - } -} - -/* Store MIC card's shutdown status internally when it is received */ -static void cosm_shutdown_status_int(struct cosm_device *cdev, - enum mic_status shutdown_status) -{ - switch (shutdown_status) { - case MIC_HALTED: - case MIC_POWER_OFF: - case MIC_RESTART: - case MIC_CRASHED: - break; - default: - dev_err(&cdev->dev, "%s %d Unexpected shutdown_status %d\n", - __func__, __LINE__, shutdown_status); - return; - }; - cdev->shutdown_status_int = shutdown_status; - cdev->heartbeat_watchdog_enable = false; - - if (cdev->state != MIC_SHUTTING_DOWN) - cosm_set_state(cdev, MIC_SHUTTING_DOWN); -} - -/* Non-blocking recv. Read and process all available messages */ -static void cosm_scif_recv(struct cosm_device *cdev) -{ - struct cosm_msg msg; - int rc; - - while (1) { - rc = scif_recv(cdev->epd, &msg, sizeof(msg), 0); - if (!rc) { - break; - } else if (rc < 0) { - dev_dbg(&cdev->dev, "%s: %d rc %d\n", - __func__, __LINE__, rc); - break; - } - dev_dbg(&cdev->dev, "%s: %d rc %d id 0x%llx\n", - __func__, __LINE__, rc, msg.id); - - switch (msg.id) { - case COSM_MSG_SHUTDOWN_STATUS: - cosm_shutdown_status_int(cdev, msg.shutdown_status); - break; - case COSM_MSG_HEARTBEAT: - /* Nothing to do, heartbeat only unblocks scif_poll */ - break; - default: - dev_err(&cdev->dev, "%s: %d unknown msg.id %lld\n", - __func__, __LINE__, msg.id); - break; - } - } -} - -/* Publish crashed status for this MIC card */ -static void cosm_set_crashed(struct cosm_device *cdev) -{ - dev_err(&cdev->dev, "node alive timeout\n"); - cosm_shutdown_status_int(cdev, MIC_CRASHED); - cosm_update_mic_status(cdev); -} - -/* Send host time to the MIC card to sync system time between host and MIC */ -static void cosm_send_time(struct cosm_device *cdev) -{ - struct cosm_msg msg = { .id = COSM_MSG_SYNC_TIME }; - struct timespec64 ts; - int rc; - - ktime_get_real_ts64(&ts); - msg.timespec.tv_sec = ts.tv_sec; - msg.timespec.tv_nsec = ts.tv_nsec; - - rc = scif_send(cdev->epd, &msg, sizeof(msg), SCIF_SEND_BLOCK); - if (rc < 0) - dev_err(&cdev->dev, "%s %d scif_send failed rc %d\n", - __func__, __LINE__, rc); -} - -/* - * Close this cosm_device's endpoint after its peer endpoint on the card has - * been closed. In all cases except MIC card crash EPOLLHUP on the host is - * triggered by the client's endpoint being closed. - */ -static void cosm_scif_close(struct cosm_device *cdev) -{ - /* - * Because SHUTDOWN_STATUS message is sent by the MIC cards in the - * reboot notifier when shutdown is still not complete, we notify mpssd - * to reset the card when SCIF endpoint is closed. - */ - cosm_update_mic_status(cdev); - scif_close(cdev->epd); - cdev->epd = NULL; - dev_dbg(&cdev->dev, "%s %d\n", __func__, __LINE__); -} - -/* - * Set card state to ONLINE when a new SCIF connection from a MIC card is - * received. Normally the state is BOOTING when the connection comes in, but can - * be ONLINE if cosm_client driver on the card was unloaded and then reloaded. - */ -static int cosm_set_online(struct cosm_device *cdev) -{ - int rc = 0; - - if (MIC_BOOTING == cdev->state || MIC_ONLINE == cdev->state) { - cdev->heartbeat_watchdog_enable = cdev->sysfs_heartbeat_enable; - cdev->epd = cdev->newepd; - if (cdev->state == MIC_BOOTING) - cosm_set_state(cdev, MIC_ONLINE); - cosm_send_time(cdev); - dev_dbg(&cdev->dev, "%s %d\n", __func__, __LINE__); - } else { - dev_warn(&cdev->dev, "%s %d not going online in state: %s\n", - __func__, __LINE__, cosm_state_string[cdev->state]); - rc = -EINVAL; - } - /* Drop reference acquired by bus_find_device in the server thread */ - put_device(&cdev->dev); - return rc; -} - -/* - * Work function for handling work for a SCIF connection from a particular MIC - * card. It first sets the card state to ONLINE and then calls scif_poll to - * block on activity such as incoming messages on the SCIF endpoint. When the - * endpoint is closed, the work function exits, completing its life cycle, from - * MIC card boot to card shutdown/reset/crash. - */ -void cosm_scif_work(struct work_struct *work) -{ - struct cosm_device *cdev = container_of(work, struct cosm_device, - scif_work); - struct scif_pollepd pollepd; - int rc; - - mutex_lock(&cdev->cosm_mutex); - if (cosm_set_online(cdev)) - goto exit; - - while (1) { - pollepd.epd = cdev->epd; - pollepd.events = EPOLLIN; - - /* Drop the mutex before blocking in scif_poll(..) */ - mutex_unlock(&cdev->cosm_mutex); - /* poll(..) with timeout on our endpoint */ - rc = scif_poll(&pollepd, 1, COSM_HEARTBEAT_TIMEOUT_MSEC); - mutex_lock(&cdev->cosm_mutex); - if (rc < 0) { - dev_err(&cdev->dev, "%s %d scif_poll rc %d\n", - __func__, __LINE__, rc); - continue; - } - - /* There is a message from the card */ - if (pollepd.revents & EPOLLIN) - cosm_scif_recv(cdev); - - /* The peer endpoint is closed or this endpoint disconnected */ - if (pollepd.revents & EPOLLHUP) { - cosm_scif_close(cdev); - break; - } - - /* Did we timeout from poll? */ - if (!rc && cdev->heartbeat_watchdog_enable) - cosm_set_crashed(cdev); - } -exit: - dev_dbg(&cdev->dev, "%s %d exiting\n", __func__, __LINE__); - mutex_unlock(&cdev->cosm_mutex); -} - -/* - * COSM SCIF server thread function. Accepts incoming SCIF connections from MIC - * cards, finds the correct cosm_device to associate that connection with and - * schedules individual work items for each MIC card. - */ -static int cosm_scif_server(void *unused) -{ - struct cosm_device *cdev; - scif_epd_t newepd; - struct scif_port_id port_id; - int rc; - - allow_signal(SIGKILL); - - while (!kthread_should_stop()) { - rc = scif_accept(listen_epd, &port_id, &newepd, - SCIF_ACCEPT_SYNC); - if (rc < 0) { - if (-ERESTARTSYS != rc) - pr_err("%s %d rc %d\n", __func__, __LINE__, rc); - continue; - } - - /* - * Associate the incoming connection with a particular - * cosm_device, COSM device ID == SCIF node ID - 1 - */ - cdev = cosm_find_cdev_by_id(port_id.node - 1); - if (!cdev) - continue; - cdev->newepd = newepd; - schedule_work(&cdev->scif_work); - } - - pr_debug("%s %d Server thread stopped\n", __func__, __LINE__); - return 0; -} - -static int cosm_scif_listen(void) -{ - int rc; - - listen_epd = scif_open(); - if (!listen_epd) { - pr_err("%s %d scif_open failed\n", __func__, __LINE__); - return -ENOMEM; - } - - rc = scif_bind(listen_epd, SCIF_COSM_LISTEN_PORT); - if (rc < 0) { - pr_err("%s %d scif_bind failed rc %d\n", - __func__, __LINE__, rc); - goto err; - } - - rc = scif_listen(listen_epd, COSM_SCIF_BACKLOG); - if (rc < 0) { - pr_err("%s %d scif_listen rc %d\n", __func__, __LINE__, rc); - goto err; - } - pr_debug("%s %d listen_epd set up\n", __func__, __LINE__); - return 0; -err: - scif_close(listen_epd); - listen_epd = NULL; - return rc; -} - -static void cosm_scif_listen_exit(void) -{ - pr_debug("%s %d closing listen_epd\n", __func__, __LINE__); - if (listen_epd) { - scif_close(listen_epd); - listen_epd = NULL; - } -} - -/* - * Create a listening SCIF endpoint and a server kthread which accepts incoming - * SCIF connections from MIC cards - */ -int cosm_scif_init(void) -{ - int rc = cosm_scif_listen(); - - if (rc) { - pr_err("%s %d cosm_scif_listen rc %d\n", - __func__, __LINE__, rc); - goto err; - } - - server_thread = kthread_run(cosm_scif_server, NULL, "cosm_server"); - if (IS_ERR(server_thread)) { - rc = PTR_ERR(server_thread); - pr_err("%s %d kthread_run rc %d\n", __func__, __LINE__, rc); - goto listen_exit; - } - return 0; -listen_exit: - cosm_scif_listen_exit(); -err: - return rc; -} - -/* Stop the running server thread and close the listening SCIF endpoint */ -void cosm_scif_exit(void) -{ - int rc; - - if (!IS_ERR_OR_NULL(server_thread)) { - rc = send_sig(SIGKILL, server_thread, 0); - if (rc) { - pr_err("%s %d send_sig rc %d\n", - __func__, __LINE__, rc); - return; - } - kthread_stop(server_thread); - } - - cosm_scif_listen_exit(); -} diff --git a/drivers/misc/mic/cosm/cosm_sysfs.c b/drivers/misc/mic/cosm/cosm_sysfs.c deleted file mode 100644 index e6dac967c1af..000000000000 --- a/drivers/misc/mic/cosm/cosm_sysfs.c +++ /dev/null @@ -1,449 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2015 Intel Corporation. - * - * Intel MIC Coprocessor State Management (COSM) Driver - */ -#include -#include "cosm_main.h" - -/* - * A state-to-string lookup table, for exposing a human readable state - * via sysfs. Always keep in sync with enum cosm_states - */ -const char * const cosm_state_string[] = { - [MIC_READY] = "ready", - [MIC_BOOTING] = "booting", - [MIC_ONLINE] = "online", - [MIC_SHUTTING_DOWN] = "shutting_down", - [MIC_RESETTING] = "resetting", - [MIC_RESET_FAILED] = "reset_failed", -}; - -/* - * A shutdown-status-to-string lookup table, for exposing a human - * readable state via sysfs. Always keep in sync with enum cosm_shutdown_status - */ -const char * const cosm_shutdown_status_string[] = { - [MIC_NOP] = "nop", - [MIC_CRASHED] = "crashed", - [MIC_HALTED] = "halted", - [MIC_POWER_OFF] = "poweroff", - [MIC_RESTART] = "restart", -}; - -void cosm_set_shutdown_status(struct cosm_device *cdev, u8 shutdown_status) -{ - dev_dbg(&cdev->dev, "Shutdown Status %s -> %s\n", - cosm_shutdown_status_string[cdev->shutdown_status], - cosm_shutdown_status_string[shutdown_status]); - cdev->shutdown_status = shutdown_status; -} - -void cosm_set_state(struct cosm_device *cdev, u8 state) -{ - dev_dbg(&cdev->dev, "State %s -> %s\n", - cosm_state_string[cdev->state], - cosm_state_string[state]); - cdev->state = state; - sysfs_notify_dirent(cdev->state_sysfs); -} - -static ssize_t -family_show(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - - if (!cdev) - return -EINVAL; - - return cdev->hw_ops->family(cdev, buf); -} -static DEVICE_ATTR_RO(family); - -static ssize_t -stepping_show(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - - if (!cdev) - return -EINVAL; - - return cdev->hw_ops->stepping(cdev, buf); -} -static DEVICE_ATTR_RO(stepping); - -static ssize_t -state_show(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - - if (!cdev || cdev->state >= MIC_LAST) - return -EINVAL; - - return scnprintf(buf, PAGE_SIZE, "%s\n", - cosm_state_string[cdev->state]); -} - -static ssize_t -state_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - int rc; - - if (!cdev) - return -EINVAL; - - if (sysfs_streq(buf, "boot")) { - rc = cosm_start(cdev); - goto done; - } - if (sysfs_streq(buf, "reset")) { - rc = cosm_reset(cdev); - goto done; - } - - if (sysfs_streq(buf, "shutdown")) { - rc = cosm_shutdown(cdev); - goto done; - } - rc = -EINVAL; -done: - if (rc) - count = rc; - return count; -} -static DEVICE_ATTR_RW(state); - -static ssize_t shutdown_status_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - - if (!cdev || cdev->shutdown_status >= MIC_STATUS_LAST) - return -EINVAL; - - return scnprintf(buf, PAGE_SIZE, "%s\n", - cosm_shutdown_status_string[cdev->shutdown_status]); -} -static DEVICE_ATTR_RO(shutdown_status); - -static ssize_t -heartbeat_enable_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - - if (!cdev) - return -EINVAL; - - return scnprintf(buf, PAGE_SIZE, "%d\n", cdev->sysfs_heartbeat_enable); -} - -static ssize_t -heartbeat_enable_store(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - int enable; - int ret; - - if (!cdev) - return -EINVAL; - - mutex_lock(&cdev->cosm_mutex); - ret = kstrtoint(buf, 10, &enable); - if (ret) - goto unlock; - - cdev->sysfs_heartbeat_enable = enable; - /* if state is not online, cdev->heartbeat_watchdog_enable is 0 */ - if (cdev->state == MIC_ONLINE) - cdev->heartbeat_watchdog_enable = enable; - ret = count; -unlock: - mutex_unlock(&cdev->cosm_mutex); - return ret; -} -static DEVICE_ATTR_RW(heartbeat_enable); - -static ssize_t -cmdline_show(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - char *cmdline; - - if (!cdev) - return -EINVAL; - - cmdline = cdev->cmdline; - - if (cmdline) - return scnprintf(buf, PAGE_SIZE, "%s\n", cmdline); - return 0; -} - -static ssize_t -cmdline_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - - if (!cdev) - return -EINVAL; - - mutex_lock(&cdev->cosm_mutex); - kfree(cdev->cmdline); - - cdev->cmdline = kmalloc(count + 1, GFP_KERNEL); - if (!cdev->cmdline) { - count = -ENOMEM; - goto unlock; - } - - strncpy(cdev->cmdline, buf, count); - - if (cdev->cmdline[count - 1] == '\n') - cdev->cmdline[count - 1] = '\0'; - else - cdev->cmdline[count] = '\0'; -unlock: - mutex_unlock(&cdev->cosm_mutex); - return count; -} -static DEVICE_ATTR_RW(cmdline); - -static ssize_t -firmware_show(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - char *firmware; - - if (!cdev) - return -EINVAL; - - firmware = cdev->firmware; - - if (firmware) - return scnprintf(buf, PAGE_SIZE, "%s\n", firmware); - return 0; -} - -static ssize_t -firmware_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - - if (!cdev) - return -EINVAL; - - mutex_lock(&cdev->cosm_mutex); - kfree(cdev->firmware); - - cdev->firmware = kmalloc(count + 1, GFP_KERNEL); - if (!cdev->firmware) { - count = -ENOMEM; - goto unlock; - } - strncpy(cdev->firmware, buf, count); - - if (cdev->firmware[count - 1] == '\n') - cdev->firmware[count - 1] = '\0'; - else - cdev->firmware[count] = '\0'; -unlock: - mutex_unlock(&cdev->cosm_mutex); - return count; -} -static DEVICE_ATTR_RW(firmware); - -static ssize_t -ramdisk_show(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - char *ramdisk; - - if (!cdev) - return -EINVAL; - - ramdisk = cdev->ramdisk; - - if (ramdisk) - return scnprintf(buf, PAGE_SIZE, "%s\n", ramdisk); - return 0; -} - -static ssize_t -ramdisk_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - - if (!cdev) - return -EINVAL; - - mutex_lock(&cdev->cosm_mutex); - kfree(cdev->ramdisk); - - cdev->ramdisk = kmalloc(count + 1, GFP_KERNEL); - if (!cdev->ramdisk) { - count = -ENOMEM; - goto unlock; - } - - strncpy(cdev->ramdisk, buf, count); - - if (cdev->ramdisk[count - 1] == '\n') - cdev->ramdisk[count - 1] = '\0'; - else - cdev->ramdisk[count] = '\0'; -unlock: - mutex_unlock(&cdev->cosm_mutex); - return count; -} -static DEVICE_ATTR_RW(ramdisk); - -static ssize_t -bootmode_show(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - char *bootmode; - - if (!cdev) - return -EINVAL; - - bootmode = cdev->bootmode; - - if (bootmode) - return scnprintf(buf, PAGE_SIZE, "%s\n", bootmode); - return 0; -} - -static ssize_t -bootmode_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - - if (!cdev) - return -EINVAL; - - if (!sysfs_streq(buf, "linux") && !sysfs_streq(buf, "flash")) - return -EINVAL; - - mutex_lock(&cdev->cosm_mutex); - kfree(cdev->bootmode); - - cdev->bootmode = kmalloc(count + 1, GFP_KERNEL); - if (!cdev->bootmode) { - count = -ENOMEM; - goto unlock; - } - - strncpy(cdev->bootmode, buf, count); - - if (cdev->bootmode[count - 1] == '\n') - cdev->bootmode[count - 1] = '\0'; - else - cdev->bootmode[count] = '\0'; -unlock: - mutex_unlock(&cdev->cosm_mutex); - return count; -} -static DEVICE_ATTR_RW(bootmode); - -static ssize_t -log_buf_addr_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - - if (!cdev) - return -EINVAL; - - return scnprintf(buf, PAGE_SIZE, "%p\n", cdev->log_buf_addr); -} - -static ssize_t -log_buf_addr_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - int ret; - unsigned long addr; - - if (!cdev) - return -EINVAL; - - ret = kstrtoul(buf, 16, &addr); - if (ret) - goto exit; - - cdev->log_buf_addr = (void *)addr; - ret = count; -exit: - return ret; -} -static DEVICE_ATTR_RW(log_buf_addr); - -static ssize_t -log_buf_len_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - - if (!cdev) - return -EINVAL; - - return scnprintf(buf, PAGE_SIZE, "%p\n", cdev->log_buf_len); -} - -static ssize_t -log_buf_len_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct cosm_device *cdev = dev_get_drvdata(dev); - int ret; - unsigned long addr; - - if (!cdev) - return -EINVAL; - - ret = kstrtoul(buf, 16, &addr); - if (ret) - goto exit; - - cdev->log_buf_len = (int *)addr; - ret = count; -exit: - return ret; -} -static DEVICE_ATTR_RW(log_buf_len); - -static struct attribute *cosm_default_attrs[] = { - &dev_attr_family.attr, - &dev_attr_stepping.attr, - &dev_attr_state.attr, - &dev_attr_shutdown_status.attr, - &dev_attr_heartbeat_enable.attr, - &dev_attr_cmdline.attr, - &dev_attr_firmware.attr, - &dev_attr_ramdisk.attr, - &dev_attr_bootmode.attr, - &dev_attr_log_buf_addr.attr, - &dev_attr_log_buf_len.attr, - - NULL -}; - -ATTRIBUTE_GROUPS(cosm_default); - -void cosm_sysfs_init(struct cosm_device *cdev) -{ - cdev->attr_group = cosm_default_groups; -} diff --git a/drivers/misc/mic/cosm_client/Makefile b/drivers/misc/mic/cosm_client/Makefile deleted file mode 100644 index 5b62270bc2ab..000000000000 --- a/drivers/misc/mic/cosm_client/Makefile +++ /dev/null @@ -1,8 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0-only -# -# Makefile - Intel MIC COSM Client Driver -# Copyright(c) 2015, Intel Corporation. -# -obj-$(CONFIG_MIC_COSM) += cosm_client.o - -cosm_client-objs += cosm_scif_client.o diff --git a/drivers/misc/mic/cosm_client/cosm_scif_client.c b/drivers/misc/mic/cosm_client/cosm_scif_client.c deleted file mode 100644 index a03213dd9319..000000000000 --- a/drivers/misc/mic/cosm_client/cosm_scif_client.c +++ /dev/null @@ -1,269 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2015 Intel Corporation. - * - * Intel MIC COSM Client Driver - */ -#include -#include -#include -#include -#include - -#include "../cosm/cosm_main.h" - -#define COSM_SCIF_MAX_RETRIES 10 -#define COSM_HEARTBEAT_SEND_MSEC (COSM_HEARTBEAT_SEND_SEC * MSEC_PER_SEC) - -static struct task_struct *client_thread; -static scif_epd_t client_epd; -static struct scif_peer_dev *client_spdev; - -/* - * Reboot notifier: receives shutdown status from the OS and communicates it - * back to the COSM process on the host - */ -static int cosm_reboot_event(struct notifier_block *this, unsigned long event, - void *ptr) -{ - struct cosm_msg msg = { .id = COSM_MSG_SHUTDOWN_STATUS }; - int rc; - - event = (event == SYS_RESTART) ? SYSTEM_RESTART : event; - dev_info(&client_spdev->dev, "%s %d received event %ld\n", - __func__, __LINE__, event); - - msg.shutdown_status = event; - rc = scif_send(client_epd, &msg, sizeof(msg), SCIF_SEND_BLOCK); - if (rc < 0) - dev_err(&client_spdev->dev, "%s %d scif_send rc %d\n", - __func__, __LINE__, rc); - - return NOTIFY_DONE; -} - -static struct notifier_block cosm_reboot = { - .notifier_call = cosm_reboot_event, -}; - -/* Set system time from timespec value received from the host */ -static void cosm_set_time(struct cosm_msg *msg) -{ - struct timespec64 ts = { - .tv_sec = msg->timespec.tv_sec, - .tv_nsec = msg->timespec.tv_nsec, - }; - int rc = do_settimeofday64(&ts); - - if (rc) - dev_err(&client_spdev->dev, "%s: %d settimeofday rc %d\n", - __func__, __LINE__, rc); -} - -/* COSM client receive message processing */ -static void cosm_client_recv(void) -{ - struct cosm_msg msg; - int rc; - - while (1) { - rc = scif_recv(client_epd, &msg, sizeof(msg), 0); - if (!rc) { - return; - } else if (rc < 0) { - dev_err(&client_spdev->dev, "%s: %d rc %d\n", - __func__, __LINE__, rc); - return; - } - - dev_dbg(&client_spdev->dev, "%s: %d rc %d id 0x%llx\n", - __func__, __LINE__, rc, msg.id); - - switch (msg.id) { - case COSM_MSG_SYNC_TIME: - cosm_set_time(&msg); - break; - case COSM_MSG_SHUTDOWN: - orderly_poweroff(true); - break; - default: - dev_err(&client_spdev->dev, "%s: %d unknown id %lld\n", - __func__, __LINE__, msg.id); - break; - } - } -} - -/* Initiate connection to the COSM server on the host */ -static int cosm_scif_connect(void) -{ - struct scif_port_id port_id; - int i, rc; - - client_epd = scif_open(); - if (!client_epd) { - dev_err(&client_spdev->dev, "%s %d scif_open failed\n", - __func__, __LINE__); - return -ENOMEM; - } - - port_id.node = 0; - port_id.port = SCIF_COSM_LISTEN_PORT; - - for (i = 0; i < COSM_SCIF_MAX_RETRIES; i++) { - rc = scif_connect(client_epd, &port_id); - if (rc < 0) - msleep(1000); - else - break; - } - - if (rc < 0) { - dev_err(&client_spdev->dev, "%s %d scif_connect rc %d\n", - __func__, __LINE__, rc); - scif_close(client_epd); - client_epd = NULL; - } - return rc < 0 ? rc : 0; -} - -/* Close host SCIF connection */ -static void cosm_scif_connect_exit(void) -{ - if (client_epd) { - scif_close(client_epd); - client_epd = NULL; - } -} - -/* - * COSM SCIF client thread function: waits for messages from the host and sends - * a heartbeat to the host - */ -static int cosm_scif_client(void *unused) -{ - struct cosm_msg msg = { .id = COSM_MSG_HEARTBEAT }; - struct scif_pollepd pollepd; - int rc; - - allow_signal(SIGKILL); - - while (!kthread_should_stop()) { - pollepd.epd = client_epd; - pollepd.events = EPOLLIN; - - rc = scif_poll(&pollepd, 1, COSM_HEARTBEAT_SEND_MSEC); - if (rc < 0) { - if (-EINTR != rc) - dev_err(&client_spdev->dev, - "%s %d scif_poll rc %d\n", - __func__, __LINE__, rc); - continue; - } - - if (pollepd.revents & EPOLLIN) - cosm_client_recv(); - - msg.id = COSM_MSG_HEARTBEAT; - rc = scif_send(client_epd, &msg, sizeof(msg), SCIF_SEND_BLOCK); - if (rc < 0) - dev_err(&client_spdev->dev, "%s %d scif_send rc %d\n", - __func__, __LINE__, rc); - } - - dev_dbg(&client_spdev->dev, "%s %d Client thread stopped\n", - __func__, __LINE__); - return 0; -} - -static void cosm_scif_probe(struct scif_peer_dev *spdev) -{ - int rc; - - dev_dbg(&spdev->dev, "%s %d: dnode %d\n", - __func__, __LINE__, spdev->dnode); - - /* We are only interested in the host with spdev->dnode == 0 */ - if (spdev->dnode) - return; - - client_spdev = spdev; - rc = cosm_scif_connect(); - if (rc) - goto exit; - - rc = register_reboot_notifier(&cosm_reboot); - if (rc) { - dev_err(&spdev->dev, - "reboot notifier registration failed rc %d\n", rc); - goto connect_exit; - } - - client_thread = kthread_run(cosm_scif_client, NULL, "cosm_client"); - if (IS_ERR(client_thread)) { - rc = PTR_ERR(client_thread); - dev_err(&spdev->dev, "%s %d kthread_run rc %d\n", - __func__, __LINE__, rc); - goto unreg_reboot; - } - return; -unreg_reboot: - unregister_reboot_notifier(&cosm_reboot); -connect_exit: - cosm_scif_connect_exit(); -exit: - client_spdev = NULL; -} - -static void cosm_scif_remove(struct scif_peer_dev *spdev) -{ - int rc; - - dev_dbg(&spdev->dev, "%s %d: dnode %d\n", - __func__, __LINE__, spdev->dnode); - - if (spdev->dnode) - return; - - if (!IS_ERR_OR_NULL(client_thread)) { - rc = send_sig(SIGKILL, client_thread, 0); - if (rc) { - pr_err("%s %d send_sig rc %d\n", - __func__, __LINE__, rc); - return; - } - kthread_stop(client_thread); - } - unregister_reboot_notifier(&cosm_reboot); - cosm_scif_connect_exit(); - client_spdev = NULL; -} - -static struct scif_client scif_client_cosm = { - .name = KBUILD_MODNAME, - .probe = cosm_scif_probe, - .remove = cosm_scif_remove, -}; - -static int __init cosm_client_init(void) -{ - int rc = scif_client_register(&scif_client_cosm); - - if (rc) - pr_err("scif_client_register failed rc %d\n", rc); - return rc; -} - -static void __exit cosm_client_exit(void) -{ - scif_client_unregister(&scif_client_cosm); -} - -module_init(cosm_client_init); -module_exit(cosm_client_exit); - -MODULE_AUTHOR("Intel Corporation"); -MODULE_DESCRIPTION("Intel(R) MIC card OS state management client driver"); -MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/host/Makefile b/drivers/misc/mic/host/Makefile deleted file mode 100644 index 25f153367980..000000000000 --- a/drivers/misc/mic/host/Makefile +++ /dev/null @@ -1,12 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0 -# -# Makefile - Intel MIC Linux driver. -# Copyright(c) 2013, Intel Corporation. -# -obj-$(CONFIG_INTEL_MIC_HOST) += mic_host.o -mic_host-objs := mic_main.o -mic_host-objs += mic_x100.o -mic_host-objs += mic_smpt.o -mic_host-objs += mic_intr.o -mic_host-objs += mic_boot.o -mic_host-objs += mic_debugfs.o diff --git a/drivers/misc/mic/host/mic_boot.c b/drivers/misc/mic/host/mic_boot.c deleted file mode 100644 index 8cb85b8b3e19..000000000000 --- a/drivers/misc/mic/host/mic_boot.c +++ /dev/null @@ -1,588 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Intel MIC Host driver. - */ -#include -#include -#include -#include -#include -#include -#include -#include "../bus/scif_bus.h" -#include "../bus/vop_bus.h" -#include "../common/mic_dev.h" -#include "mic_device.h" -#include "mic_smpt.h" - -static inline struct mic_device *vpdev_to_mdev(struct device *dev) -{ - return dev_get_drvdata(dev->parent); -} - -static dma_addr_t -_mic_dma_map_page(struct device *dev, struct page *page, - unsigned long offset, size_t size, - enum dma_data_direction dir, unsigned long attrs) -{ - void *va = phys_to_virt(page_to_phys(page)) + offset; - struct mic_device *mdev = vpdev_to_mdev(dev); - - return mic_map_single(mdev, va, size); -} - -static void _mic_dma_unmap_page(struct device *dev, dma_addr_t dma_addr, - size_t size, enum dma_data_direction dir, - unsigned long attrs) -{ - struct mic_device *mdev = vpdev_to_mdev(dev); - - mic_unmap_single(mdev, dma_addr, size); -} - -static const struct dma_map_ops _mic_dma_ops = { - .map_page = _mic_dma_map_page, - .unmap_page = _mic_dma_unmap_page, -}; - -static struct mic_irq * -__mic_request_irq(struct vop_device *vpdev, - irqreturn_t (*func)(int irq, void *data), - const char *name, void *data, int intr_src) -{ - struct mic_device *mdev = vpdev_to_mdev(&vpdev->dev); - - return mic_request_threaded_irq(mdev, func, NULL, name, data, - intr_src, MIC_INTR_DB); -} - -static void __mic_free_irq(struct vop_device *vpdev, - struct mic_irq *cookie, void *data) -{ - struct mic_device *mdev = vpdev_to_mdev(&vpdev->dev); - - mic_free_irq(mdev, cookie, data); -} - -static void __mic_ack_interrupt(struct vop_device *vpdev, int num) -{ - struct mic_device *mdev = vpdev_to_mdev(&vpdev->dev); - - mdev->ops->intr_workarounds(mdev); -} - -static int __mic_next_db(struct vop_device *vpdev) -{ - struct mic_device *mdev = vpdev_to_mdev(&vpdev->dev); - - return mic_next_db(mdev); -} - -static void *__mic_get_dp(struct vop_device *vpdev) -{ - struct mic_device *mdev = vpdev_to_mdev(&vpdev->dev); - - return mdev->dp; -} - -static void __iomem *__mic_get_remote_dp(struct vop_device *vpdev) -{ - return NULL; -} - -static void __mic_send_intr(struct vop_device *vpdev, int db) -{ - struct mic_device *mdev = vpdev_to_mdev(&vpdev->dev); - - mdev->ops->send_intr(mdev, db); -} - -static void __iomem *__mic_ioremap(struct vop_device *vpdev, - dma_addr_t pa, size_t len) -{ - struct mic_device *mdev = vpdev_to_mdev(&vpdev->dev); - - return mdev->aper.va + pa; -} - -static void __mic_iounmap(struct vop_device *vpdev, void __iomem *va) -{ - /* nothing to do */ -} - -static struct vop_hw_ops vop_hw_ops = { - .request_irq = __mic_request_irq, - .free_irq = __mic_free_irq, - .ack_interrupt = __mic_ack_interrupt, - .next_db = __mic_next_db, - .get_dp = __mic_get_dp, - .get_remote_dp = __mic_get_remote_dp, - .send_intr = __mic_send_intr, - .remap = __mic_ioremap, - .unmap = __mic_iounmap, -}; - -static inline struct mic_device *scdev_to_mdev(struct scif_hw_dev *scdev) -{ - return dev_get_drvdata(scdev->dev.parent); -} - -static void *__mic_dma_alloc(struct device *dev, size_t size, - dma_addr_t *dma_handle, gfp_t gfp, - unsigned long attrs) -{ - struct scif_hw_dev *scdev = dev_get_drvdata(dev); - struct mic_device *mdev = scdev_to_mdev(scdev); - dma_addr_t tmp; - void *va = kzalloc(size, gfp); - - if (va) { - tmp = mic_map_single(mdev, va, size); - if (dma_mapping_error(dev, tmp)) { - kfree(va); - va = NULL; - } else { - *dma_handle = tmp; - } - } - return va; -} - -static void __mic_dma_free(struct device *dev, size_t size, void *vaddr, - dma_addr_t dma_handle, unsigned long attrs) -{ - struct scif_hw_dev *scdev = dev_get_drvdata(dev); - struct mic_device *mdev = scdev_to_mdev(scdev); - - mic_unmap_single(mdev, dma_handle, size); - kfree(vaddr); -} - -static dma_addr_t -__mic_dma_map_page(struct device *dev, struct page *page, unsigned long offset, - size_t size, enum dma_data_direction dir, - unsigned long attrs) -{ - void *va = phys_to_virt(page_to_phys(page)) + offset; - struct scif_hw_dev *scdev = dev_get_drvdata(dev); - struct mic_device *mdev = scdev_to_mdev(scdev); - - return mic_map_single(mdev, va, size); -} - -static void -__mic_dma_unmap_page(struct device *dev, dma_addr_t dma_addr, - size_t size, enum dma_data_direction dir, - unsigned long attrs) -{ - struct scif_hw_dev *scdev = dev_get_drvdata(dev); - struct mic_device *mdev = scdev_to_mdev(scdev); - - mic_unmap_single(mdev, dma_addr, size); -} - -static int __mic_dma_map_sg(struct device *dev, struct scatterlist *sg, - int nents, enum dma_data_direction dir, - unsigned long attrs) -{ - struct scif_hw_dev *scdev = dev_get_drvdata(dev); - struct mic_device *mdev = scdev_to_mdev(scdev); - struct scatterlist *s; - int i, j, ret; - dma_addr_t da; - - ret = dma_map_sg(&mdev->pdev->dev, sg, nents, dir); - if (ret <= 0) - return 0; - - for_each_sg(sg, s, nents, i) { - da = mic_map(mdev, sg_dma_address(s) + s->offset, s->length); - if (!da) - goto err; - sg_dma_address(s) = da; - } - return nents; -err: - for_each_sg(sg, s, i, j) { - mic_unmap(mdev, sg_dma_address(s), s->length); - sg_dma_address(s) = mic_to_dma_addr(mdev, sg_dma_address(s)); - } - dma_unmap_sg(&mdev->pdev->dev, sg, nents, dir); - return 0; -} - -static void __mic_dma_unmap_sg(struct device *dev, - struct scatterlist *sg, int nents, - enum dma_data_direction dir, - unsigned long attrs) -{ - struct scif_hw_dev *scdev = dev_get_drvdata(dev); - struct mic_device *mdev = scdev_to_mdev(scdev); - struct scatterlist *s; - dma_addr_t da; - int i; - - for_each_sg(sg, s, nents, i) { - da = mic_to_dma_addr(mdev, sg_dma_address(s)); - mic_unmap(mdev, sg_dma_address(s), s->length); - sg_dma_address(s) = da; - } - dma_unmap_sg(&mdev->pdev->dev, sg, nents, dir); -} - -static const struct dma_map_ops __mic_dma_ops = { - .alloc = __mic_dma_alloc, - .free = __mic_dma_free, - .map_page = __mic_dma_map_page, - .unmap_page = __mic_dma_unmap_page, - .map_sg = __mic_dma_map_sg, - .unmap_sg = __mic_dma_unmap_sg, -}; - -static struct mic_irq * -___mic_request_irq(struct scif_hw_dev *scdev, - irqreturn_t (*func)(int irq, void *data), - const char *name, - void *data, int db) -{ - struct mic_device *mdev = scdev_to_mdev(scdev); - - return mic_request_threaded_irq(mdev, func, NULL, name, data, - db, MIC_INTR_DB); -} - -static void -___mic_free_irq(struct scif_hw_dev *scdev, - struct mic_irq *cookie, void *data) -{ - struct mic_device *mdev = scdev_to_mdev(scdev); - - mic_free_irq(mdev, cookie, data); -} - -static void ___mic_ack_interrupt(struct scif_hw_dev *scdev, int num) -{ - struct mic_device *mdev = scdev_to_mdev(scdev); - - mdev->ops->intr_workarounds(mdev); -} - -static int ___mic_next_db(struct scif_hw_dev *scdev) -{ - struct mic_device *mdev = scdev_to_mdev(scdev); - - return mic_next_db(mdev); -} - -static void ___mic_send_intr(struct scif_hw_dev *scdev, int db) -{ - struct mic_device *mdev = scdev_to_mdev(scdev); - - mdev->ops->send_intr(mdev, db); -} - -static void __iomem *___mic_ioremap(struct scif_hw_dev *scdev, - phys_addr_t pa, size_t len) -{ - struct mic_device *mdev = scdev_to_mdev(scdev); - - return mdev->aper.va + pa; -} - -static void ___mic_iounmap(struct scif_hw_dev *scdev, void __iomem *va) -{ - /* nothing to do */ -} - -static struct scif_hw_ops scif_hw_ops = { - .request_irq = ___mic_request_irq, - .free_irq = ___mic_free_irq, - .ack_interrupt = ___mic_ack_interrupt, - .next_db = ___mic_next_db, - .send_intr = ___mic_send_intr, - .remap = ___mic_ioremap, - .unmap = ___mic_iounmap, -}; - -static inline struct mic_device *mbdev_to_mdev(struct mbus_device *mbdev) -{ - return dev_get_drvdata(mbdev->dev.parent); -} - -static dma_addr_t -mic_dma_map_page(struct device *dev, struct page *page, - unsigned long offset, size_t size, enum dma_data_direction dir, - unsigned long attrs) -{ - void *va = phys_to_virt(page_to_phys(page)) + offset; - struct mic_device *mdev = dev_get_drvdata(dev->parent); - - return mic_map_single(mdev, va, size); -} - -static void -mic_dma_unmap_page(struct device *dev, dma_addr_t dma_addr, - size_t size, enum dma_data_direction dir, - unsigned long attrs) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - mic_unmap_single(mdev, dma_addr, size); -} - -static const struct dma_map_ops mic_dma_ops = { - .map_page = mic_dma_map_page, - .unmap_page = mic_dma_unmap_page, -}; - -static struct mic_irq * -_mic_request_threaded_irq(struct mbus_device *mbdev, - irq_handler_t handler, irq_handler_t thread_fn, - const char *name, void *data, int intr_src) -{ - return mic_request_threaded_irq(mbdev_to_mdev(mbdev), handler, - thread_fn, name, data, - intr_src, MIC_INTR_DMA); -} - -static void _mic_free_irq(struct mbus_device *mbdev, - struct mic_irq *cookie, void *data) -{ - mic_free_irq(mbdev_to_mdev(mbdev), cookie, data); -} - -static void _mic_ack_interrupt(struct mbus_device *mbdev, int num) -{ - struct mic_device *mdev = mbdev_to_mdev(mbdev); - mdev->ops->intr_workarounds(mdev); -} - -static struct mbus_hw_ops mbus_hw_ops = { - .request_threaded_irq = _mic_request_threaded_irq, - .free_irq = _mic_free_irq, - .ack_interrupt = _mic_ack_interrupt, -}; - -/* Initialize the MIC bootparams */ -void mic_bootparam_init(struct mic_device *mdev) -{ - struct mic_bootparam *bootparam = mdev->dp; - - bootparam->magic = cpu_to_le32(MIC_MAGIC); - bootparam->h2c_config_db = -1; - bootparam->node_id = mdev->id + 1; - bootparam->scif_host_dma_addr = 0x0; - bootparam->scif_card_dma_addr = 0x0; - bootparam->c2h_scif_db = -1; - bootparam->h2c_scif_db = -1; -} - -static inline struct mic_device *cosmdev_to_mdev(struct cosm_device *cdev) -{ - return dev_get_drvdata(cdev->dev.parent); -} - -static void _mic_reset(struct cosm_device *cdev) -{ - struct mic_device *mdev = cosmdev_to_mdev(cdev); - - mdev->ops->reset_fw_ready(mdev); - mdev->ops->reset(mdev); -} - -static bool _mic_ready(struct cosm_device *cdev) -{ - struct mic_device *mdev = cosmdev_to_mdev(cdev); - - return mdev->ops->is_fw_ready(mdev); -} - -/** - * mic_request_dma_chans - Request DMA channels - * @mdev: pointer to mic_device instance - * - * returns number of DMA channels acquired - */ -static int mic_request_dma_chans(struct mic_device *mdev) -{ - dma_cap_mask_t mask; - struct dma_chan *chan; - - dma_cap_zero(mask); - dma_cap_set(DMA_MEMCPY, mask); - - do { - chan = dma_request_channel(mask, mdev->ops->dma_filter, - &mdev->pdev->dev); - if (chan) { - mdev->dma_ch[mdev->num_dma_ch++] = chan; - if (mdev->num_dma_ch >= MIC_MAX_DMA_CHAN) - break; - } - } while (chan); - dev_info(&mdev->pdev->dev, "DMA channels # %d\n", mdev->num_dma_ch); - return mdev->num_dma_ch; -} - -/** - * mic_free_dma_chans - release DMA channels - * @mdev: pointer to mic_device instance - * - * returns none - */ -static void mic_free_dma_chans(struct mic_device *mdev) -{ - int i = 0; - - for (i = 0; i < mdev->num_dma_ch; i++) { - dma_release_channel(mdev->dma_ch[i]); - mdev->dma_ch[i] = NULL; - } - mdev->num_dma_ch = 0; -} - -/** - * _mic_start - Start the MIC. - * @cdev: pointer to cosm_device instance - * @id: MIC device id/index provided by COSM used in other drivers like SCIF - * - * This function prepares an MIC for boot and initiates boot. - * RETURNS: An appropriate -ERRNO error value on error, or zero for success. - * - * For all cosm_hw_ops the caller holds a mutex to ensure serialization. - */ -static int _mic_start(struct cosm_device *cdev, int id) -{ - struct mic_device *mdev = cosmdev_to_mdev(cdev); - int rc; - - mic_bootparam_init(mdev); - mdev->dma_mbdev = mbus_register_device(&mdev->pdev->dev, - MBUS_DEV_DMA_HOST, &mic_dma_ops, - &mbus_hw_ops, id, mdev->mmio.va); - if (IS_ERR(mdev->dma_mbdev)) { - rc = PTR_ERR(mdev->dma_mbdev); - goto unlock_ret; - } - if (!mic_request_dma_chans(mdev)) { - rc = -ENODEV; - goto dma_remove; - } - mdev->scdev = scif_register_device(&mdev->pdev->dev, MIC_SCIF_DEV, - &__mic_dma_ops, &scif_hw_ops, - id + 1, 0, &mdev->mmio, - &mdev->aper, mdev->dp, NULL, - mdev->dma_ch, mdev->num_dma_ch, - true); - if (IS_ERR(mdev->scdev)) { - rc = PTR_ERR(mdev->scdev); - goto dma_free; - } - - mdev->vpdev = vop_register_device(&mdev->pdev->dev, - VOP_DEV_TRNSP, &_mic_dma_ops, - &vop_hw_ops, id + 1, &mdev->aper, - mdev->dma_ch[0]); - if (IS_ERR(mdev->vpdev)) { - rc = PTR_ERR(mdev->vpdev); - goto scif_remove; - } - - rc = mdev->ops->load_mic_fw(mdev, NULL); - if (rc) - goto vop_remove; - mic_smpt_restore(mdev); - mic_intr_restore(mdev); - mdev->intr_ops->enable_interrupts(mdev); - mdev->ops->write_spad(mdev, MIC_DPLO_SPAD, mdev->dp_dma_addr); - mdev->ops->write_spad(mdev, MIC_DPHI_SPAD, mdev->dp_dma_addr >> 32); - mdev->ops->send_firmware_intr(mdev); - goto unlock_ret; -vop_remove: - vop_unregister_device(mdev->vpdev); -scif_remove: - scif_unregister_device(mdev->scdev); -dma_free: - mic_free_dma_chans(mdev); -dma_remove: - mbus_unregister_device(mdev->dma_mbdev); -unlock_ret: - return rc; -} - -/** - * _mic_stop - Prepare the MIC for reset and trigger reset. - * @cdev: pointer to cosm_device instance - * @force: force a MIC to reset even if it is already offline. - * - * RETURNS: None. - */ -static void _mic_stop(struct cosm_device *cdev, bool force) -{ - struct mic_device *mdev = cosmdev_to_mdev(cdev); - - /* - * Since SCIF handles card shutdown and reset (using COSM), it will - * will be the first to be registered and the last to be - * unregistered. - */ - vop_unregister_device(mdev->vpdev); - scif_unregister_device(mdev->scdev); - mic_free_dma_chans(mdev); - mbus_unregister_device(mdev->dma_mbdev); - mic_bootparam_init(mdev); -} - -static ssize_t _mic_family(struct cosm_device *cdev, char *buf) -{ - struct mic_device *mdev = cosmdev_to_mdev(cdev); - static const char *family[MIC_FAMILY_LAST] = { "x100", "Unknown" }; - - return scnprintf(buf, PAGE_SIZE, "%s\n", family[mdev->family]); -} - -static ssize_t _mic_stepping(struct cosm_device *cdev, char *buf) -{ - struct mic_device *mdev = cosmdev_to_mdev(cdev); - const char *string = "??"; - - switch (mdev->stepping) { - case MIC_A0_STEP: - string = "A0"; - break; - case MIC_B0_STEP: - string = "B0"; - break; - case MIC_B1_STEP: - string = "B1"; - break; - case MIC_C0_STEP: - string = "C0"; - break; - default: - break; - } - return scnprintf(buf, PAGE_SIZE, "%s\n", string); -} - -static struct mic_mw *_mic_aper(struct cosm_device *cdev) -{ - struct mic_device *mdev = cosmdev_to_mdev(cdev); - - return &mdev->aper; -} - -struct cosm_hw_ops cosm_hw_ops = { - .reset = _mic_reset, - .force_reset = _mic_reset, - .post_reset = NULL, - .ready = _mic_ready, - .start = _mic_start, - .stop = _mic_stop, - .family = _mic_family, - .stepping = _mic_stepping, - .aper = _mic_aper, -}; diff --git a/drivers/misc/mic/host/mic_debugfs.c b/drivers/misc/mic/host/mic_debugfs.c deleted file mode 100644 index ffda740e20d5..000000000000 --- a/drivers/misc/mic/host/mic_debugfs.c +++ /dev/null @@ -1,149 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Intel MIC Host driver. - */ -#include -#include -#include - -#include -#include "../common/mic_dev.h" -#include "mic_device.h" -#include "mic_smpt.h" - -/* Debugfs parent dir */ -static struct dentry *mic_dbg; - -static int mic_smpt_show(struct seq_file *s, void *pos) -{ - int i; - struct mic_device *mdev = s->private; - unsigned long flags; - - seq_printf(s, "MIC %-2d |%-10s| %-14s %-10s\n", - mdev->id, "SMPT entry", "SW DMA addr", "RefCount"); - seq_puts(s, "====================================================\n"); - - if (mdev->smpt) { - struct mic_smpt_info *smpt_info = mdev->smpt; - spin_lock_irqsave(&smpt_info->smpt_lock, flags); - for (i = 0; i < smpt_info->info.num_reg; i++) { - seq_printf(s, "%9s|%-10d| %-#14llx %-10lld\n", - " ", i, smpt_info->entry[i].dma_addr, - smpt_info->entry[i].ref_count); - } - spin_unlock_irqrestore(&smpt_info->smpt_lock, flags); - } - seq_puts(s, "====================================================\n"); - return 0; -} - -DEFINE_SHOW_ATTRIBUTE(mic_smpt); - -static int mic_post_code_show(struct seq_file *s, void *pos) -{ - struct mic_device *mdev = s->private; - u32 reg = mdev->ops->get_postcode(mdev); - - seq_printf(s, "%c%c", reg & 0xff, (reg >> 8) & 0xff); - return 0; -} - -DEFINE_SHOW_ATTRIBUTE(mic_post_code); - -static int mic_msi_irq_info_show(struct seq_file *s, void *pos) -{ - struct mic_device *mdev = s->private; - int reg; - int i, j; - u16 entry; - u16 vector; - struct pci_dev *pdev = mdev->pdev; - - if (pci_dev_msi_enabled(pdev)) { - for (i = 0; i < mdev->irq_info.num_vectors; i++) { - if (pdev->msix_enabled) { - entry = mdev->irq_info.msix_entries[i].entry; - vector = mdev->irq_info.msix_entries[i].vector; - } else { - entry = 0; - vector = pdev->irq; - } - - reg = mdev->intr_ops->read_msi_to_src_map(mdev, entry); - - seq_printf(s, "%s %-10d %s %-10d MXAR[%d]: %08X\n", - "IRQ:", vector, "Entry:", entry, i, reg); - - seq_printf(s, "%-10s", "offset:"); - for (j = (MIC_NUM_OFFSETS - 1); j >= 0; j--) - seq_printf(s, "%4d ", j); - seq_puts(s, "\n"); - - - seq_printf(s, "%-10s", "count:"); - for (j = (MIC_NUM_OFFSETS - 1); j >= 0; j--) - seq_printf(s, "%4d ", - (mdev->irq_info.mic_msi_map[i] & - BIT(j)) ? 1 : 0); - seq_puts(s, "\n\n"); - } - } else { - seq_puts(s, "MSI/MSIx interrupts not enabled\n"); - } - - return 0; -} - -DEFINE_SHOW_ATTRIBUTE(mic_msi_irq_info); - -/* - * mic_create_debug_dir - Initialize MIC debugfs entries. - */ -void mic_create_debug_dir(struct mic_device *mdev) -{ - char name[16]; - - if (!mic_dbg) - return; - - scnprintf(name, sizeof(name), "mic%d", mdev->id); - mdev->dbg_dir = debugfs_create_dir(name, mic_dbg); - - debugfs_create_file("smpt", 0444, mdev->dbg_dir, mdev, - &mic_smpt_fops); - - debugfs_create_file("post_code", 0444, mdev->dbg_dir, mdev, - &mic_post_code_fops); - - debugfs_create_file("msi_irq_info", 0444, mdev->dbg_dir, mdev, - &mic_msi_irq_info_fops); -} - -/* - * mic_delete_debug_dir - Uninitialize MIC debugfs entries. - */ -void mic_delete_debug_dir(struct mic_device *mdev) -{ - debugfs_remove_recursive(mdev->dbg_dir); -} - -/* - * mic_init_debugfs - Initialize global debugfs entry. - */ -void __init mic_init_debugfs(void) -{ - mic_dbg = debugfs_create_dir(KBUILD_MODNAME, NULL); -} - -/* - * mic_exit_debugfs - Uninitialize global debugfs entry - */ -void mic_exit_debugfs(void) -{ - debugfs_remove(mic_dbg); -} diff --git a/drivers/misc/mic/host/mic_device.h b/drivers/misc/mic/host/mic_device.h deleted file mode 100644 index 41bcd308ae59..000000000000 --- a/drivers/misc/mic/host/mic_device.h +++ /dev/null @@ -1,157 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Intel MIC Host driver. - */ -#ifndef _MIC_DEVICE_H_ -#define _MIC_DEVICE_H_ - -#include -#include -#include -#include -#include -#include -#include -#include "../bus/scif_bus.h" -#include "../bus/vop_bus.h" -#include "../bus/cosm_bus.h" -#include "mic_intr.h" - -/** - * enum mic_stepping - MIC stepping ids. - */ -enum mic_stepping { - MIC_A0_STEP = 0x0, - MIC_B0_STEP = 0x10, - MIC_B1_STEP = 0x11, - MIC_C0_STEP = 0x20, -}; - -extern struct cosm_hw_ops cosm_hw_ops; - -/** - * struct mic_device - MIC device information for each card. - * - * @mmio: MMIO bar information. - * @aper: Aperture bar information. - * @family: The MIC family to which this device belongs. - * @ops: MIC HW specific operations. - * @id: The unique device id for this MIC device. - * @stepping: Stepping ID. - * @pdev: Underlying PCI device. - * @mic_mutex: Mutex for synchronizing access to mic_device. - * @intr_ops: HW specific interrupt operations. - * @smpt_ops: Hardware specific SMPT operations. - * @smpt: MIC SMPT information. - * @intr_info: H/W specific interrupt information. - * @irq_info: The OS specific irq information - * @dbg_dir: debugfs directory of this MIC device. - * @bootaddr: MIC boot address. - * @dp: virtio device page - * @dp_dma_addr: virtio device page DMA address. - * @dma_mbdev: MIC BUS DMA device. - * @dma_ch - Array of DMA channels - * @num_dma_ch - Number of DMA channels available - * @scdev: SCIF device on the SCIF virtual bus. - * @vpdev: Virtio over PCIe device on the VOP virtual bus. - * @cosm_dev: COSM device - */ -struct mic_device { - struct mic_mw mmio; - struct mic_mw aper; - enum mic_hw_family family; - struct mic_hw_ops *ops; - int id; - enum mic_stepping stepping; - struct pci_dev *pdev; - struct mutex mic_mutex; - struct mic_hw_intr_ops *intr_ops; - struct mic_smpt_ops *smpt_ops; - struct mic_smpt_info *smpt; - struct mic_intr_info *intr_info; - struct mic_irq_info irq_info; - struct dentry *dbg_dir; - u32 bootaddr; - void *dp; - dma_addr_t dp_dma_addr; - struct mbus_device *dma_mbdev; - struct dma_chan *dma_ch[MIC_MAX_DMA_CHAN]; - int num_dma_ch; - struct scif_hw_dev *scdev; - struct vop_device *vpdev; - struct cosm_device *cosm_dev; -}; - -/** - * struct mic_hw_ops - MIC HW specific operations. - * @aper_bar: Aperture bar resource number. - * @mmio_bar: MMIO bar resource number. - * @read_spad: Read from scratch pad register. - * @write_spad: Write to scratch pad register. - * @send_intr: Send an interrupt for a particular doorbell on the card. - * @ack_interrupt: Hardware specific operations to ack the h/w on - * receipt of an interrupt. - * @intr_workarounds: Hardware specific workarounds needed after - * handling an interrupt. - * @reset: Reset the remote processor. - * @reset_fw_ready: Reset firmware ready field. - * @is_fw_ready: Check if firmware is ready for OS download. - * @send_firmware_intr: Send an interrupt to the card firmware. - * @load_mic_fw: Load firmware segments required to boot the card - * into card memory. This includes the kernel, command line, ramdisk etc. - * @get_postcode: Get post code status from firmware. - * @dma_filter: DMA filter function to be used. - */ -struct mic_hw_ops { - u8 aper_bar; - u8 mmio_bar; - u32 (*read_spad)(struct mic_device *mdev, unsigned int idx); - void (*write_spad)(struct mic_device *mdev, unsigned int idx, u32 val); - void (*send_intr)(struct mic_device *mdev, int doorbell); - u32 (*ack_interrupt)(struct mic_device *mdev); - void (*intr_workarounds)(struct mic_device *mdev); - void (*reset)(struct mic_device *mdev); - void (*reset_fw_ready)(struct mic_device *mdev); - bool (*is_fw_ready)(struct mic_device *mdev); - void (*send_firmware_intr)(struct mic_device *mdev); - int (*load_mic_fw)(struct mic_device *mdev, const char *buf); - u32 (*get_postcode)(struct mic_device *mdev); - bool (*dma_filter)(struct dma_chan *chan, void *param); -}; - -/** - * mic_mmio_read - read from an MMIO register. - * @mw: MMIO register base virtual address. - * @offset: register offset. - * - * RETURNS: register value. - */ -static inline u32 mic_mmio_read(struct mic_mw *mw, u32 offset) -{ - return ioread32(mw->va + offset); -} - -/** - * mic_mmio_write - write to an MMIO register. - * @mw: MMIO register base virtual address. - * @val: the data value to put into the register - * @offset: register offset. - * - * RETURNS: none. - */ -static inline void -mic_mmio_write(struct mic_mw *mw, u32 val, u32 offset) -{ - iowrite32(val, mw->va + offset); -} - -void mic_bootparam_init(struct mic_device *mdev); -void mic_create_debug_dir(struct mic_device *dev); -void mic_delete_debug_dir(struct mic_device *dev); -void __init mic_init_debugfs(void); -void mic_exit_debugfs(void); -#endif diff --git a/drivers/misc/mic/host/mic_intr.c b/drivers/misc/mic/host/mic_intr.c deleted file mode 100644 index 85b3221b5d40..000000000000 --- a/drivers/misc/mic/host/mic_intr.c +++ /dev/null @@ -1,635 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Intel MIC Host driver. - */ -#include -#include - -#include "../common/mic_dev.h" -#include "mic_device.h" - -static irqreturn_t mic_thread_fn(int irq, void *dev) -{ - struct mic_device *mdev = dev; - struct mic_intr_info *intr_info = mdev->intr_info; - struct mic_irq_info *irq_info = &mdev->irq_info; - struct mic_intr_cb *intr_cb; - struct pci_dev *pdev = mdev->pdev; - int i; - - spin_lock(&irq_info->mic_thread_lock); - for (i = intr_info->intr_start_idx[MIC_INTR_DB]; - i < intr_info->intr_len[MIC_INTR_DB]; i++) - if (test_and_clear_bit(i, &irq_info->mask)) { - list_for_each_entry(intr_cb, &irq_info->cb_list[i], - list) - if (intr_cb->thread_fn) - intr_cb->thread_fn(pdev->irq, - intr_cb->data); - } - spin_unlock(&irq_info->mic_thread_lock); - return IRQ_HANDLED; -} -/** - * mic_interrupt - Generic interrupt handler for - * MSI and INTx based interrupts. - * @irq: interrupt to handle (unused) - * @dev: pointer to the mic_device instance - */ -static irqreturn_t mic_interrupt(int irq, void *dev) -{ - struct mic_device *mdev = dev; - struct mic_intr_info *intr_info = mdev->intr_info; - struct mic_irq_info *irq_info = &mdev->irq_info; - struct mic_intr_cb *intr_cb; - struct pci_dev *pdev = mdev->pdev; - u32 mask; - int i; - - mask = mdev->ops->ack_interrupt(mdev); - if (!mask) - return IRQ_NONE; - - spin_lock(&irq_info->mic_intr_lock); - for (i = intr_info->intr_start_idx[MIC_INTR_DB]; - i < intr_info->intr_len[MIC_INTR_DB]; i++) - if (mask & BIT(i)) { - list_for_each_entry(intr_cb, &irq_info->cb_list[i], - list) - if (intr_cb->handler) - intr_cb->handler(pdev->irq, - intr_cb->data); - set_bit(i, &irq_info->mask); - } - spin_unlock(&irq_info->mic_intr_lock); - return IRQ_WAKE_THREAD; -} - -/* Return the interrupt offset from the index. Index is 0 based. */ -static u16 mic_map_src_to_offset(struct mic_device *mdev, - int intr_src, enum mic_intr_type type) -{ - if (type >= MIC_NUM_INTR_TYPES) - return MIC_NUM_OFFSETS; - if (intr_src >= mdev->intr_info->intr_len[type]) - return MIC_NUM_OFFSETS; - - return mdev->intr_info->intr_start_idx[type] + intr_src; -} - -/* Return next available msix_entry. */ -static struct msix_entry *mic_get_available_vector(struct mic_device *mdev) -{ - int i; - struct mic_irq_info *info = &mdev->irq_info; - - for (i = 0; i < info->num_vectors; i++) - if (!info->mic_msi_map[i]) - return &info->msix_entries[i]; - return NULL; -} - -/** - * mic_register_intr_callback - Register a callback handler for the - * given source id. - * - * @mdev: pointer to the mic_device instance - * @idx: The source id to be registered. - * @handler: The function to be called when the source id receives - * the interrupt. - * @thread_fn: thread fn. corresponding to the handler - * @data: Private data of the requester. - * Return the callback structure that was registered or an - * appropriate error on failure. - */ -static struct mic_intr_cb *mic_register_intr_callback(struct mic_device *mdev, - u8 idx, irq_handler_t handler, irq_handler_t thread_fn, - void *data) -{ - struct mic_intr_cb *intr_cb; - unsigned long flags; - int rc; - intr_cb = kmalloc(sizeof(*intr_cb), GFP_KERNEL); - - if (!intr_cb) - return ERR_PTR(-ENOMEM); - - intr_cb->handler = handler; - intr_cb->thread_fn = thread_fn; - intr_cb->data = data; - intr_cb->cb_id = ida_simple_get(&mdev->irq_info.cb_ida, - 0, 0, GFP_KERNEL); - if (intr_cb->cb_id < 0) { - rc = intr_cb->cb_id; - goto ida_fail; - } - - spin_lock(&mdev->irq_info.mic_thread_lock); - spin_lock_irqsave(&mdev->irq_info.mic_intr_lock, flags); - list_add_tail(&intr_cb->list, &mdev->irq_info.cb_list[idx]); - spin_unlock_irqrestore(&mdev->irq_info.mic_intr_lock, flags); - spin_unlock(&mdev->irq_info.mic_thread_lock); - - return intr_cb; -ida_fail: - kfree(intr_cb); - return ERR_PTR(rc); -} - -/** - * mic_unregister_intr_callback - Unregister the callback handler - * identified by its callback id. - * - * @mdev: pointer to the mic_device instance - * @idx: The callback structure id to be unregistered. - * Return the source id that was unregistered or MIC_NUM_OFFSETS if no - * such callback handler was found. - */ -static u8 mic_unregister_intr_callback(struct mic_device *mdev, u32 idx) -{ - struct list_head *pos, *tmp; - struct mic_intr_cb *intr_cb; - unsigned long flags; - int i; - - spin_lock(&mdev->irq_info.mic_thread_lock); - spin_lock_irqsave(&mdev->irq_info.mic_intr_lock, flags); - for (i = 0; i < MIC_NUM_OFFSETS; i++) { - list_for_each_safe(pos, tmp, &mdev->irq_info.cb_list[i]) { - intr_cb = list_entry(pos, struct mic_intr_cb, list); - if (intr_cb->cb_id == idx) { - list_del(pos); - ida_simple_remove(&mdev->irq_info.cb_ida, - intr_cb->cb_id); - kfree(intr_cb); - spin_unlock_irqrestore( - &mdev->irq_info.mic_intr_lock, flags); - spin_unlock(&mdev->irq_info.mic_thread_lock); - return i; - } - } - } - spin_unlock_irqrestore(&mdev->irq_info.mic_intr_lock, flags); - spin_unlock(&mdev->irq_info.mic_thread_lock); - return MIC_NUM_OFFSETS; -} - -/** - * mic_setup_msix - Initializes MSIx interrupts. - * - * @mdev: pointer to mic_device instance - * @pdev: PCI device structure - * - * RETURNS: An appropriate -ERRNO error value on error, or zero for success. - */ -static int mic_setup_msix(struct mic_device *mdev, struct pci_dev *pdev) -{ - int rc, i; - int entry_size = sizeof(*mdev->irq_info.msix_entries); - - mdev->irq_info.msix_entries = kmalloc_array(MIC_MIN_MSIX, - entry_size, GFP_KERNEL); - if (!mdev->irq_info.msix_entries) { - rc = -ENOMEM; - goto err_nomem1; - } - - for (i = 0; i < MIC_MIN_MSIX; i++) - mdev->irq_info.msix_entries[i].entry = i; - - rc = pci_enable_msix_exact(pdev, mdev->irq_info.msix_entries, - MIC_MIN_MSIX); - if (rc) { - dev_dbg(&pdev->dev, "Error enabling MSIx. rc = %d\n", rc); - goto err_enable_msix; - } - - mdev->irq_info.num_vectors = MIC_MIN_MSIX; - mdev->irq_info.mic_msi_map = kzalloc((sizeof(u32) * - mdev->irq_info.num_vectors), GFP_KERNEL); - - if (!mdev->irq_info.mic_msi_map) { - rc = -ENOMEM; - goto err_nomem2; - } - - dev_dbg(&mdev->pdev->dev, - "%d MSIx irqs setup\n", mdev->irq_info.num_vectors); - return 0; -err_nomem2: - pci_disable_msix(pdev); -err_enable_msix: - kfree(mdev->irq_info.msix_entries); -err_nomem1: - mdev->irq_info.num_vectors = 0; - return rc; -} - -/** - * mic_setup_callbacks - Initialize data structures needed - * to handle callbacks. - * - * @mdev: pointer to mic_device instance - */ -static int mic_setup_callbacks(struct mic_device *mdev) -{ - int i; - - mdev->irq_info.cb_list = kmalloc_array(MIC_NUM_OFFSETS, - sizeof(*mdev->irq_info.cb_list), - GFP_KERNEL); - if (!mdev->irq_info.cb_list) - return -ENOMEM; - - for (i = 0; i < MIC_NUM_OFFSETS; i++) - INIT_LIST_HEAD(&mdev->irq_info.cb_list[i]); - ida_init(&mdev->irq_info.cb_ida); - spin_lock_init(&mdev->irq_info.mic_intr_lock); - spin_lock_init(&mdev->irq_info.mic_thread_lock); - return 0; -} - -/** - * mic_release_callbacks - Uninitialize data structures needed - * to handle callbacks. - * - * @mdev: pointer to mic_device instance - */ -static void mic_release_callbacks(struct mic_device *mdev) -{ - unsigned long flags; - struct list_head *pos, *tmp; - struct mic_intr_cb *intr_cb; - int i; - - spin_lock(&mdev->irq_info.mic_thread_lock); - spin_lock_irqsave(&mdev->irq_info.mic_intr_lock, flags); - for (i = 0; i < MIC_NUM_OFFSETS; i++) { - if (list_empty(&mdev->irq_info.cb_list[i])) - break; - - list_for_each_safe(pos, tmp, &mdev->irq_info.cb_list[i]) { - intr_cb = list_entry(pos, struct mic_intr_cb, list); - list_del(pos); - ida_simple_remove(&mdev->irq_info.cb_ida, - intr_cb->cb_id); - kfree(intr_cb); - } - } - spin_unlock_irqrestore(&mdev->irq_info.mic_intr_lock, flags); - spin_unlock(&mdev->irq_info.mic_thread_lock); - ida_destroy(&mdev->irq_info.cb_ida); - kfree(mdev->irq_info.cb_list); -} - -/** - * mic_setup_msi - Initializes MSI interrupts. - * - * @mdev: pointer to mic_device instance - * @pdev: PCI device structure - * - * RETURNS: An appropriate -ERRNO error value on error, or zero for success. - */ -static int mic_setup_msi(struct mic_device *mdev, struct pci_dev *pdev) -{ - int rc; - - rc = pci_enable_msi(pdev); - if (rc) { - dev_dbg(&pdev->dev, "Error enabling MSI. rc = %d\n", rc); - return rc; - } - - mdev->irq_info.num_vectors = 1; - mdev->irq_info.mic_msi_map = kzalloc((sizeof(u32) * - mdev->irq_info.num_vectors), GFP_KERNEL); - - if (!mdev->irq_info.mic_msi_map) { - rc = -ENOMEM; - goto err_nomem1; - } - - rc = mic_setup_callbacks(mdev); - if (rc) { - dev_err(&pdev->dev, "Error setting up callbacks\n"); - goto err_nomem2; - } - - rc = request_threaded_irq(pdev->irq, mic_interrupt, mic_thread_fn, - 0, "mic-msi", mdev); - if (rc) { - dev_err(&pdev->dev, "Error allocating MSI interrupt\n"); - goto err_irq_req_fail; - } - - dev_dbg(&pdev->dev, "%d MSI irqs setup\n", mdev->irq_info.num_vectors); - return 0; -err_irq_req_fail: - mic_release_callbacks(mdev); -err_nomem2: - kfree(mdev->irq_info.mic_msi_map); -err_nomem1: - pci_disable_msi(pdev); - mdev->irq_info.num_vectors = 0; - return rc; -} - -/** - * mic_setup_intx - Initializes legacy interrupts. - * - * @mdev: pointer to mic_device instance - * @pdev: PCI device structure - * - * RETURNS: An appropriate -ERRNO error value on error, or zero for success. - */ -static int mic_setup_intx(struct mic_device *mdev, struct pci_dev *pdev) -{ - int rc; - - /* Enable intx */ - pci_intx(pdev, 1); - rc = mic_setup_callbacks(mdev); - if (rc) { - dev_err(&pdev->dev, "Error setting up callbacks\n"); - goto err_nomem; - } - - rc = request_threaded_irq(pdev->irq, mic_interrupt, mic_thread_fn, - IRQF_SHARED, "mic-intx", mdev); - if (rc) - goto err; - - dev_dbg(&pdev->dev, "intx irq setup\n"); - return 0; -err: - mic_release_callbacks(mdev); -err_nomem: - return rc; -} - -/** - * mic_next_db - Retrieve the next doorbell interrupt source id. - * The id is picked sequentially from the available pool of - * doorlbell ids. - * - * @mdev: pointer to the mic_device instance. - * - * Returns the next doorbell interrupt source. - */ -int mic_next_db(struct mic_device *mdev) -{ - int next_db; - - next_db = mdev->irq_info.next_avail_src % - mdev->intr_info->intr_len[MIC_INTR_DB]; - mdev->irq_info.next_avail_src++; - return next_db; -} - -#define COOKIE_ID_SHIFT 16 -#define GET_ENTRY(cookie) ((cookie) & 0xFFFF) -#define GET_OFFSET(cookie) ((cookie) >> COOKIE_ID_SHIFT) -#define MK_COOKIE(x, y) ((x) | (y) << COOKIE_ID_SHIFT) - -/** - * mic_request_threaded_irq - request an irq. mic_mutex needs - * to be held before calling this function. - * - * @mdev: pointer to mic_device instance - * @handler: The callback function that handles the interrupt. - * The function needs to call ack_interrupts - * (mdev->ops->ack_interrupt(mdev)) when handling the interrupts. - * @thread_fn: thread fn required by request_threaded_irq. - * @name: The ASCII name of the callee requesting the irq. - * @data: private data that is returned back when calling the - * function handler. - * @intr_src: The source id of the requester. Its the doorbell id - * for Doorbell interrupts and DMA channel id for DMA interrupts. - * @type: The type of interrupt. Values defined in mic_intr_type - * - * returns: The cookie that is transparent to the caller. Passed - * back when calling mic_free_irq. An appropriate error code - * is returned on failure. Caller needs to use IS_ERR(return_val) - * to check for failure and PTR_ERR(return_val) to obtained the - * error code. - * - */ -struct mic_irq * -mic_request_threaded_irq(struct mic_device *mdev, - irq_handler_t handler, irq_handler_t thread_fn, - const char *name, void *data, int intr_src, - enum mic_intr_type type) -{ - u16 offset; - int rc = 0; - struct msix_entry *msix = NULL; - unsigned long cookie = 0; - u16 entry; - struct mic_intr_cb *intr_cb; - struct pci_dev *pdev = mdev->pdev; - - offset = mic_map_src_to_offset(mdev, intr_src, type); - if (offset >= MIC_NUM_OFFSETS) { - dev_err(&mdev->pdev->dev, - "Error mapping index %d to a valid source id.\n", - intr_src); - rc = -EINVAL; - goto err; - } - - if (mdev->irq_info.num_vectors > 1) { - msix = mic_get_available_vector(mdev); - if (!msix) { - dev_err(&mdev->pdev->dev, - "No MSIx vectors available for use.\n"); - rc = -ENOSPC; - goto err; - } - - rc = request_threaded_irq(msix->vector, handler, thread_fn, - 0, name, data); - if (rc) { - dev_dbg(&mdev->pdev->dev, - "request irq failed rc = %d\n", rc); - goto err; - } - entry = msix->entry; - mdev->irq_info.mic_msi_map[entry] |= BIT(offset); - mdev->intr_ops->program_msi_to_src_map(mdev, - entry, offset, true); - cookie = MK_COOKIE(entry, offset); - dev_dbg(&mdev->pdev->dev, "irq: %d assigned for src: %d\n", - msix->vector, intr_src); - } else { - intr_cb = mic_register_intr_callback(mdev, offset, handler, - thread_fn, data); - if (IS_ERR(intr_cb)) { - dev_err(&mdev->pdev->dev, - "No available callback entries for use\n"); - rc = PTR_ERR(intr_cb); - goto err; - } - - entry = 0; - if (pci_dev_msi_enabled(pdev)) { - mdev->irq_info.mic_msi_map[entry] |= (1 << offset); - mdev->intr_ops->program_msi_to_src_map(mdev, - entry, offset, true); - } - cookie = MK_COOKIE(entry, intr_cb->cb_id); - dev_dbg(&mdev->pdev->dev, "callback %d registered for src: %d\n", - intr_cb->cb_id, intr_src); - } - return (struct mic_irq *)cookie; -err: - return ERR_PTR(rc); -} - -/** - * mic_free_irq - free irq. mic_mutex - * needs to be held before calling this function. - * - * @mdev: pointer to mic_device instance - * @cookie: cookie obtained during a successful call to mic_request_threaded_irq - * @data: private data specified by the calling function during the - * mic_request_threaded_irq - * - * returns: none. - */ -void mic_free_irq(struct mic_device *mdev, - struct mic_irq *cookie, void *data) -{ - u32 offset; - u32 entry; - u8 src_id; - unsigned int irq; - struct pci_dev *pdev = mdev->pdev; - - entry = GET_ENTRY((unsigned long)cookie); - offset = GET_OFFSET((unsigned long)cookie); - if (mdev->irq_info.num_vectors > 1) { - if (entry >= mdev->irq_info.num_vectors) { - dev_warn(&mdev->pdev->dev, - "entry %d should be < num_irq %d\n", - entry, mdev->irq_info.num_vectors); - return; - } - irq = mdev->irq_info.msix_entries[entry].vector; - free_irq(irq, data); - mdev->irq_info.mic_msi_map[entry] &= ~(BIT(offset)); - mdev->intr_ops->program_msi_to_src_map(mdev, - entry, offset, false); - - dev_dbg(&mdev->pdev->dev, "irq: %d freed\n", irq); - } else { - irq = pdev->irq; - src_id = mic_unregister_intr_callback(mdev, offset); - if (src_id >= MIC_NUM_OFFSETS) { - dev_warn(&mdev->pdev->dev, "Error unregistering callback\n"); - return; - } - if (pci_dev_msi_enabled(pdev)) { - mdev->irq_info.mic_msi_map[entry] &= ~(BIT(src_id)); - mdev->intr_ops->program_msi_to_src_map(mdev, - entry, src_id, false); - } - dev_dbg(&mdev->pdev->dev, "callback %d unregistered for src: %d\n", - offset, src_id); - } -} - -/** - * mic_setup_interrupts - Initializes interrupts. - * - * @mdev: pointer to mic_device instance - * @pdev: PCI device structure - * - * RETURNS: An appropriate -ERRNO error value on error, or zero for success. - */ -int mic_setup_interrupts(struct mic_device *mdev, struct pci_dev *pdev) -{ - int rc; - - rc = mic_setup_msix(mdev, pdev); - if (!rc) - goto done; - - rc = mic_setup_msi(mdev, pdev); - if (!rc) - goto done; - - rc = mic_setup_intx(mdev, pdev); - if (rc) { - dev_err(&mdev->pdev->dev, "no usable interrupts\n"); - return rc; - } -done: - mdev->intr_ops->enable_interrupts(mdev); - return 0; -} - -/** - * mic_free_interrupts - Frees interrupts setup by mic_setup_interrupts - * - * @mdev: pointer to mic_device instance - * @pdev: PCI device structure - * - * returns none. - */ -void mic_free_interrupts(struct mic_device *mdev, struct pci_dev *pdev) -{ - int i; - - mdev->intr_ops->disable_interrupts(mdev); - if (mdev->irq_info.num_vectors > 1) { - for (i = 0; i < mdev->irq_info.num_vectors; i++) { - if (mdev->irq_info.mic_msi_map[i]) - dev_warn(&pdev->dev, "irq %d may still be in use.\n", - mdev->irq_info.msix_entries[i].vector); - } - kfree(mdev->irq_info.mic_msi_map); - kfree(mdev->irq_info.msix_entries); - pci_disable_msix(pdev); - } else { - if (pci_dev_msi_enabled(pdev)) { - free_irq(pdev->irq, mdev); - kfree(mdev->irq_info.mic_msi_map); - pci_disable_msi(pdev); - } else { - free_irq(pdev->irq, mdev); - } - mic_release_callbacks(mdev); - } -} - -/** - * mic_intr_restore - Restore MIC interrupt registers. - * - * @mdev: pointer to mic_device instance. - * - * Restore the interrupt registers to values previously - * stored in the SW data structures. mic_mutex needs to - * be held before calling this function. - * - * returns None. - */ -void mic_intr_restore(struct mic_device *mdev) -{ - int entry, offset; - struct pci_dev *pdev = mdev->pdev; - - if (!pci_dev_msi_enabled(pdev)) - return; - - for (entry = 0; entry < mdev->irq_info.num_vectors; entry++) { - for (offset = 0; offset < MIC_NUM_OFFSETS; offset++) { - if (mdev->irq_info.mic_msi_map[entry] & BIT(offset)) - mdev->intr_ops->program_msi_to_src_map(mdev, - entry, offset, true); - } - } -} diff --git a/drivers/misc/mic/host/mic_intr.h b/drivers/misc/mic/host/mic_intr.h deleted file mode 100644 index b14ba818006f..000000000000 --- a/drivers/misc/mic/host/mic_intr.h +++ /dev/null @@ -1,137 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Intel MIC Host driver. - */ -#ifndef _MIC_INTR_H_ -#define _MIC_INTR_H_ - -#include -#include -/* - * The minimum number of msix vectors required for normal operation. - * 3 for virtio network, console and block devices. - * 1 for card shutdown notifications. - * 4 for host owned DMA channels. - * 1 for SCIF - */ -#define MIC_MIN_MSIX 9 -#define MIC_NUM_OFFSETS 32 - -/** - * mic_intr_source - The type of source that will generate - * the interrupt.The number of types needs to be in sync with - * MIC_NUM_INTR_TYPES - * - * MIC_INTR_DB: The source is a doorbell - * MIC_INTR_DMA: The source is a DMA channel - * MIC_INTR_ERR: The source is an error interrupt e.g. SBOX ERR - * MIC_NUM_INTR_TYPES: Total number of interrupt sources. - */ -enum mic_intr_type { - MIC_INTR_DB = 0, - MIC_INTR_DMA, - MIC_INTR_ERR, - MIC_NUM_INTR_TYPES -}; - -/** - * struct mic_intr_info - Contains h/w specific interrupt sources - * information. - * - * @intr_start_idx: Contains the starting indexes of the - * interrupt types. - * @intr_len: Contains the length of the interrupt types. - */ -struct mic_intr_info { - u16 intr_start_idx[MIC_NUM_INTR_TYPES]; - u16 intr_len[MIC_NUM_INTR_TYPES]; -}; - -/** - * struct mic_irq_info - OS specific irq information - * - * @next_avail_src: next available doorbell that can be assigned. - * @msix_entries: msix entries allocated while setting up MSI-x - * @mic_msi_map: The MSI/MSI-x mapping information. - * @num_vectors: The number of MSI/MSI-x vectors that have been allocated. - * @cb_ida: callback ID allocator to track the callbacks registered. - * @mic_intr_lock: spinlock to protect the interrupt callback list. - * @mic_thread_lock: spinlock to protect the thread callback list. - * This lock is used to protect against thread_fn while - * mic_intr_lock is used to protect against interrupt handler. - * @cb_list: Array of callback lists one for each source. - * @mask: Mask used by the main thread fn to call the underlying thread fns. - */ -struct mic_irq_info { - int next_avail_src; - struct msix_entry *msix_entries; - u32 *mic_msi_map; - u16 num_vectors; - struct ida cb_ida; - spinlock_t mic_intr_lock; - spinlock_t mic_thread_lock; - struct list_head *cb_list; - unsigned long mask; -}; - -/** - * struct mic_intr_cb - Interrupt callback structure. - * - * @handler: The callback function - * @thread_fn: The thread_fn. - * @data: Private data of the requester. - * @cb_id: The callback id. Identifies this callback. - * @list: list head pointing to the next callback structure. - */ -struct mic_intr_cb { - irq_handler_t handler; - irq_handler_t thread_fn; - void *data; - int cb_id; - struct list_head list; -}; - -/** - * struct mic_irq - opaque pointer used as cookie - */ -struct mic_irq; - -/* Forward declaration */ -struct mic_device; - -/** - * struct mic_hw_intr_ops: MIC HW specific interrupt operations - * @intr_init: Initialize H/W specific interrupt information. - * @enable_interrupts: Enable interrupts from the hardware. - * @disable_interrupts: Disable interrupts from the hardware. - * @program_msi_to_src_map: Update MSI mapping registers with - * irq information. - * @read_msi_to_src_map: Read MSI mapping registers containing - * irq information. - */ -struct mic_hw_intr_ops { - void (*intr_init)(struct mic_device *mdev); - void (*enable_interrupts)(struct mic_device *mdev); - void (*disable_interrupts)(struct mic_device *mdev); - void (*program_msi_to_src_map) (struct mic_device *mdev, - int idx, int intr_src, bool set); - u32 (*read_msi_to_src_map) (struct mic_device *mdev, - int idx); -}; - -int mic_next_db(struct mic_device *mdev); -struct mic_irq * -mic_request_threaded_irq(struct mic_device *mdev, - irq_handler_t handler, irq_handler_t thread_fn, - const char *name, void *data, int intr_src, - enum mic_intr_type type); -void mic_free_irq(struct mic_device *mdev, - struct mic_irq *cookie, void *data); -int mic_setup_interrupts(struct mic_device *mdev, struct pci_dev *pdev); -void mic_free_interrupts(struct mic_device *mdev, struct pci_dev *pdev); -void mic_intr_restore(struct mic_device *mdev); -#endif diff --git a/drivers/misc/mic/host/mic_main.c b/drivers/misc/mic/host/mic_main.c deleted file mode 100644 index ea4608527ea0..000000000000 --- a/drivers/misc/mic/host/mic_main.c +++ /dev/null @@ -1,335 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Intel MIC Host driver. - */ -#include -#include -#include -#include - -#include -#include "../common/mic_dev.h" -#include "mic_device.h" -#include "mic_x100.h" -#include "mic_smpt.h" - -static const char mic_driver_name[] = "mic"; - -static const struct pci_device_id mic_pci_tbl[] = { - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2250)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2251)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2252)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2253)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2254)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2255)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2256)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2257)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2258)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2259)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_225a)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_225b)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_225c)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_225d)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_225e)}, - - /* required last entry */ - { 0, } -}; - -MODULE_DEVICE_TABLE(pci, mic_pci_tbl); - -/* ID allocator for MIC devices */ -static struct ida g_mic_ida; - -/* Initialize the device page */ -static int mic_dp_init(struct mic_device *mdev) -{ - mdev->dp = kzalloc(MIC_DP_SIZE, GFP_KERNEL); - if (!mdev->dp) - return -ENOMEM; - - mdev->dp_dma_addr = mic_map_single(mdev, - mdev->dp, MIC_DP_SIZE); - if (mic_map_error(mdev->dp_dma_addr)) { - kfree(mdev->dp); - dev_err(&mdev->pdev->dev, "%s %d err %d\n", - __func__, __LINE__, -ENOMEM); - return -ENOMEM; - } - mdev->ops->write_spad(mdev, MIC_DPLO_SPAD, mdev->dp_dma_addr); - mdev->ops->write_spad(mdev, MIC_DPHI_SPAD, mdev->dp_dma_addr >> 32); - return 0; -} - -/* Uninitialize the device page */ -static void mic_dp_uninit(struct mic_device *mdev) -{ - mic_unmap_single(mdev, mdev->dp_dma_addr, MIC_DP_SIZE); - kfree(mdev->dp); -} - -/** - * mic_ops_init: Initialize HW specific operation tables. - * - * @mdev: pointer to mic_device instance - * - * returns none. - */ -static void mic_ops_init(struct mic_device *mdev) -{ - switch (mdev->family) { - case MIC_FAMILY_X100: - mdev->ops = &mic_x100_ops; - mdev->intr_ops = &mic_x100_intr_ops; - mdev->smpt_ops = &mic_x100_smpt_ops; - break; - default: - break; - } -} - -/** - * mic_get_family - Determine hardware family to which this MIC belongs. - * - * @pdev: The pci device structure - * - * returns family. - */ -static enum mic_hw_family mic_get_family(struct pci_dev *pdev) -{ - enum mic_hw_family family; - - switch (pdev->device) { - case MIC_X100_PCI_DEVICE_2250: - case MIC_X100_PCI_DEVICE_2251: - case MIC_X100_PCI_DEVICE_2252: - case MIC_X100_PCI_DEVICE_2253: - case MIC_X100_PCI_DEVICE_2254: - case MIC_X100_PCI_DEVICE_2255: - case MIC_X100_PCI_DEVICE_2256: - case MIC_X100_PCI_DEVICE_2257: - case MIC_X100_PCI_DEVICE_2258: - case MIC_X100_PCI_DEVICE_2259: - case MIC_X100_PCI_DEVICE_225a: - case MIC_X100_PCI_DEVICE_225b: - case MIC_X100_PCI_DEVICE_225c: - case MIC_X100_PCI_DEVICE_225d: - case MIC_X100_PCI_DEVICE_225e: - family = MIC_FAMILY_X100; - break; - default: - family = MIC_FAMILY_UNKNOWN; - break; - } - return family; -} - -/** - * mic_device_init - Allocates and initializes the MIC device structure - * - * @mdev: pointer to mic_device instance - * @pdev: The pci device structure - * - * returns none. - */ -static void -mic_device_init(struct mic_device *mdev, struct pci_dev *pdev) -{ - mdev->pdev = pdev; - mdev->family = mic_get_family(pdev); - mdev->stepping = pdev->revision; - mic_ops_init(mdev); - mutex_init(&mdev->mic_mutex); - mdev->irq_info.next_avail_src = 0; -} - -/** - * mic_probe - Device Initialization Routine - * - * @pdev: PCI device structure - * @ent: entry in mic_pci_tbl - * - * returns 0 on success, < 0 on failure. - */ -static int mic_probe(struct pci_dev *pdev, - const struct pci_device_id *ent) -{ - int rc; - struct mic_device *mdev; - - mdev = kzalloc(sizeof(*mdev), GFP_KERNEL); - if (!mdev) { - rc = -ENOMEM; - goto mdev_alloc_fail; - } - mdev->id = ida_simple_get(&g_mic_ida, 0, MIC_MAX_NUM_DEVS, GFP_KERNEL); - if (mdev->id < 0) { - rc = mdev->id; - dev_err(&pdev->dev, "ida_simple_get failed rc %d\n", rc); - goto ida_fail; - } - - mic_device_init(mdev, pdev); - - rc = pci_enable_device(pdev); - if (rc) { - dev_err(&pdev->dev, "failed to enable pci device.\n"); - goto ida_remove; - } - - pci_set_master(pdev); - - rc = pci_request_regions(pdev, mic_driver_name); - if (rc) { - dev_err(&pdev->dev, "failed to get pci regions.\n"); - goto disable_device; - } - - rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(64)); - if (rc) { - dev_err(&pdev->dev, "Cannot set DMA mask\n"); - goto release_regions; - } - - mdev->mmio.pa = pci_resource_start(pdev, mdev->ops->mmio_bar); - mdev->mmio.len = pci_resource_len(pdev, mdev->ops->mmio_bar); - mdev->mmio.va = pci_ioremap_bar(pdev, mdev->ops->mmio_bar); - if (!mdev->mmio.va) { - dev_err(&pdev->dev, "Cannot remap MMIO BAR\n"); - rc = -EIO; - goto release_regions; - } - - mdev->aper.pa = pci_resource_start(pdev, mdev->ops->aper_bar); - mdev->aper.len = pci_resource_len(pdev, mdev->ops->aper_bar); - mdev->aper.va = ioremap_wc(mdev->aper.pa, mdev->aper.len); - if (!mdev->aper.va) { - dev_err(&pdev->dev, "Cannot remap Aperture BAR\n"); - rc = -EIO; - goto unmap_mmio; - } - - mdev->intr_ops->intr_init(mdev); - rc = mic_setup_interrupts(mdev, pdev); - if (rc) { - dev_err(&pdev->dev, "mic_setup_interrupts failed %d\n", rc); - goto unmap_aper; - } - rc = mic_smpt_init(mdev); - if (rc) { - dev_err(&pdev->dev, "smpt_init failed %d\n", rc); - goto free_interrupts; - } - - pci_set_drvdata(pdev, mdev); - - rc = mic_dp_init(mdev); - if (rc) { - dev_err(&pdev->dev, "mic_dp_init failed rc %d\n", rc); - goto smpt_uninit; - } - mic_bootparam_init(mdev); - mic_create_debug_dir(mdev); - - mdev->cosm_dev = cosm_register_device(&mdev->pdev->dev, &cosm_hw_ops); - if (IS_ERR(mdev->cosm_dev)) { - rc = PTR_ERR(mdev->cosm_dev); - dev_err(&pdev->dev, "cosm_add_device failed rc %d\n", rc); - goto cleanup_debug_dir; - } - return 0; -cleanup_debug_dir: - mic_delete_debug_dir(mdev); - mic_dp_uninit(mdev); -smpt_uninit: - mic_smpt_uninit(mdev); -free_interrupts: - mic_free_interrupts(mdev, pdev); -unmap_aper: - iounmap(mdev->aper.va); -unmap_mmio: - iounmap(mdev->mmio.va); -release_regions: - pci_release_regions(pdev); -disable_device: - pci_disable_device(pdev); -ida_remove: - ida_simple_remove(&g_mic_ida, mdev->id); -ida_fail: - kfree(mdev); -mdev_alloc_fail: - dev_err(&pdev->dev, "Probe failed rc %d\n", rc); - return rc; -} - -/** - * mic_remove - Device Removal Routine - * mic_remove is called by the PCI subsystem to alert the driver - * that it should release a PCI device. - * - * @pdev: PCI device structure - */ -static void mic_remove(struct pci_dev *pdev) -{ - struct mic_device *mdev; - - mdev = pci_get_drvdata(pdev); - if (!mdev) - return; - - cosm_unregister_device(mdev->cosm_dev); - mic_delete_debug_dir(mdev); - mic_dp_uninit(mdev); - mic_smpt_uninit(mdev); - mic_free_interrupts(mdev, pdev); - iounmap(mdev->aper.va); - iounmap(mdev->mmio.va); - pci_release_regions(pdev); - pci_disable_device(pdev); - ida_simple_remove(&g_mic_ida, mdev->id); - kfree(mdev); -} - -static struct pci_driver mic_driver = { - .name = mic_driver_name, - .id_table = mic_pci_tbl, - .probe = mic_probe, - .remove = mic_remove -}; - -static int __init mic_init(void) -{ - int ret; - - request_module("mic_x100_dma"); - mic_init_debugfs(); - ida_init(&g_mic_ida); - ret = pci_register_driver(&mic_driver); - if (ret) { - pr_err("pci_register_driver failed ret %d\n", ret); - goto cleanup_debugfs; - } - return 0; -cleanup_debugfs: - ida_destroy(&g_mic_ida); - mic_exit_debugfs(); - return ret; -} - -static void __exit mic_exit(void) -{ - pci_unregister_driver(&mic_driver); - ida_destroy(&g_mic_ida); - mic_exit_debugfs(); -} - -module_init(mic_init); -module_exit(mic_exit); - -MODULE_AUTHOR("Intel Corporation"); -MODULE_DESCRIPTION("Intel(R) MIC X100 Host driver"); -MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/host/mic_smpt.c b/drivers/misc/mic/host/mic_smpt.c deleted file mode 100644 index 50d1bebecd54..000000000000 --- a/drivers/misc/mic/host/mic_smpt.c +++ /dev/null @@ -1,427 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Intel MIC Host driver. - */ -#include - -#include "../common/mic_dev.h" -#include "mic_device.h" -#include "mic_smpt.h" - -static inline u64 mic_system_page_mask(struct mic_device *mdev) -{ - return (1ULL << mdev->smpt->info.page_shift) - 1ULL; -} - -static inline u8 mic_sys_addr_to_smpt(struct mic_device *mdev, dma_addr_t pa) -{ - return (pa - mdev->smpt->info.base) >> mdev->smpt->info.page_shift; -} - -static inline u64 mic_smpt_to_pa(struct mic_device *mdev, u8 index) -{ - return mdev->smpt->info.base + (index * mdev->smpt->info.page_size); -} - -static inline u64 mic_smpt_offset(struct mic_device *mdev, dma_addr_t pa) -{ - return pa & mic_system_page_mask(mdev); -} - -static inline u64 mic_smpt_align_low(struct mic_device *mdev, dma_addr_t pa) -{ - return ALIGN(pa - mic_system_page_mask(mdev), - mdev->smpt->info.page_size); -} - -static inline u64 mic_smpt_align_high(struct mic_device *mdev, dma_addr_t pa) -{ - return ALIGN(pa, mdev->smpt->info.page_size); -} - -/* Total Cumulative system memory accessible by MIC across all SMPT entries */ -static inline u64 mic_max_system_memory(struct mic_device *mdev) -{ - return mdev->smpt->info.num_reg * mdev->smpt->info.page_size; -} - -/* Maximum system memory address accessible by MIC */ -static inline u64 mic_max_system_addr(struct mic_device *mdev) -{ - return mdev->smpt->info.base + mic_max_system_memory(mdev) - 1ULL; -} - -/* Check if the DMA address is a MIC system memory address */ -static inline bool -mic_is_system_addr(struct mic_device *mdev, dma_addr_t pa) -{ - return pa >= mdev->smpt->info.base && pa <= mic_max_system_addr(mdev); -} - -/* Populate an SMPT entry and update the reference counts. */ -static void mic_add_smpt_entry(int spt, s64 *ref, u64 addr, - int entries, struct mic_device *mdev) -{ - struct mic_smpt_info *smpt_info = mdev->smpt; - int i; - - for (i = spt; i < spt + entries; i++, - addr += smpt_info->info.page_size) { - if (!smpt_info->entry[i].ref_count && - (smpt_info->entry[i].dma_addr != addr)) { - mdev->smpt_ops->set(mdev, addr, i); - smpt_info->entry[i].dma_addr = addr; - } - smpt_info->entry[i].ref_count += ref[i - spt]; - } -} - -/* - * Find an available MIC address in MIC SMPT address space - * for a given DMA address and size. - */ -static dma_addr_t mic_smpt_op(struct mic_device *mdev, u64 dma_addr, - int entries, s64 *ref, size_t size) -{ - int spt; - int ae = 0; - int i; - unsigned long flags; - dma_addr_t mic_addr = 0; - dma_addr_t addr = dma_addr; - struct mic_smpt_info *smpt_info = mdev->smpt; - - spin_lock_irqsave(&smpt_info->smpt_lock, flags); - - /* find existing entries */ - for (i = 0; i < smpt_info->info.num_reg; i++) { - if (smpt_info->entry[i].dma_addr == addr) { - ae++; - addr += smpt_info->info.page_size; - } else if (ae) /* cannot find contiguous entries */ - goto not_found; - - if (ae == entries) - goto found; - } - - /* find free entry */ - for (ae = 0, i = 0; i < smpt_info->info.num_reg; i++) { - ae = (smpt_info->entry[i].ref_count == 0) ? ae + 1 : 0; - if (ae == entries) - goto found; - } - -not_found: - spin_unlock_irqrestore(&smpt_info->smpt_lock, flags); - return mic_addr; - -found: - spt = i - entries + 1; - mic_addr = mic_smpt_to_pa(mdev, spt); - mic_add_smpt_entry(spt, ref, dma_addr, entries, mdev); - smpt_info->map_count++; - smpt_info->ref_count += (s64)size; - spin_unlock_irqrestore(&smpt_info->smpt_lock, flags); - return mic_addr; -} - -/* - * Returns number of smpt entries needed for dma_addr to dma_addr + size - * also returns the reference count array for each of those entries - * and the starting smpt address - */ -static int mic_get_smpt_ref_count(struct mic_device *mdev, dma_addr_t dma_addr, - size_t size, s64 *ref, u64 *smpt_start) -{ - u64 start = dma_addr; - u64 end = dma_addr + size; - int i = 0; - - while (start < end) { - ref[i++] = min(mic_smpt_align_high(mdev, start + 1), - end) - start; - start = mic_smpt_align_high(mdev, start + 1); - } - - if (smpt_start) - *smpt_start = mic_smpt_align_low(mdev, dma_addr); - - return i; -} - -/* - * mic_to_dma_addr - Converts a MIC address to a DMA address. - * - * @mdev: pointer to mic_device instance. - * @mic_addr: MIC address. - * - * returns a DMA address. - */ -dma_addr_t mic_to_dma_addr(struct mic_device *mdev, dma_addr_t mic_addr) -{ - struct mic_smpt_info *smpt_info = mdev->smpt; - int spt; - dma_addr_t dma_addr; - - if (!mic_is_system_addr(mdev, mic_addr)) { - dev_err(&mdev->pdev->dev, - "mic_addr is invalid. mic_addr = 0x%llx\n", mic_addr); - return -EINVAL; - } - spt = mic_sys_addr_to_smpt(mdev, mic_addr); - dma_addr = smpt_info->entry[spt].dma_addr + - mic_smpt_offset(mdev, mic_addr); - return dma_addr; -} - -/** - * mic_map - Maps a DMA address to a MIC physical address. - * - * @mdev: pointer to mic_device instance. - * @dma_addr: DMA address. - * @size: Size of the region to be mapped. - * - * This API converts the DMA address provided to a DMA address understood - * by MIC. Caller should check for errors by calling mic_map_error(..). - * - * returns DMA address as required by MIC. - */ -dma_addr_t mic_map(struct mic_device *mdev, dma_addr_t dma_addr, size_t size) -{ - dma_addr_t mic_addr = 0; - int num_entries; - s64 *ref; - u64 smpt_start; - - if (!size || size > mic_max_system_memory(mdev)) - return mic_addr; - - ref = kmalloc_array(mdev->smpt->info.num_reg, sizeof(s64), GFP_ATOMIC); - if (!ref) - return mic_addr; - - num_entries = mic_get_smpt_ref_count(mdev, dma_addr, size, - ref, &smpt_start); - - /* Set the smpt table appropriately and get 16G aligned mic address */ - mic_addr = mic_smpt_op(mdev, smpt_start, num_entries, ref, size); - - kfree(ref); - - /* - * If mic_addr is zero then its an error case - * since mic_addr can never be zero. - * else generate mic_addr by adding the 16G offset in dma_addr - */ - if (!mic_addr && MIC_FAMILY_X100 == mdev->family) { - dev_err(&mdev->pdev->dev, - "mic_map failed dma_addr 0x%llx size 0x%lx\n", - dma_addr, size); - return mic_addr; - } else { - return mic_addr + mic_smpt_offset(mdev, dma_addr); - } -} - -/** - * mic_unmap - Unmaps a MIC physical address. - * - * @mdev: pointer to mic_device instance. - * @mic_addr: MIC physical address. - * @size: Size of the region to be unmapped. - * - * This API unmaps the mappings created by mic_map(..). - * - * returns None. - */ -void mic_unmap(struct mic_device *mdev, dma_addr_t mic_addr, size_t size) -{ - struct mic_smpt_info *smpt_info = mdev->smpt; - s64 *ref; - int num_smpt; - int spt; - int i; - unsigned long flags; - - if (!size) - return; - - if (!mic_is_system_addr(mdev, mic_addr)) { - dev_err(&mdev->pdev->dev, - "invalid address: 0x%llx\n", mic_addr); - return; - } - - spt = mic_sys_addr_to_smpt(mdev, mic_addr); - ref = kmalloc_array(mdev->smpt->info.num_reg, sizeof(s64), GFP_ATOMIC); - if (!ref) - return; - - /* Get number of smpt entries to be mapped, ref count array */ - num_smpt = mic_get_smpt_ref_count(mdev, mic_addr, size, ref, NULL); - - spin_lock_irqsave(&smpt_info->smpt_lock, flags); - smpt_info->unmap_count++; - smpt_info->ref_count -= (s64)size; - - for (i = spt; i < spt + num_smpt; i++) { - smpt_info->entry[i].ref_count -= ref[i - spt]; - if (smpt_info->entry[i].ref_count < 0) - dev_warn(&mdev->pdev->dev, - "ref count for entry %d is negative\n", i); - } - spin_unlock_irqrestore(&smpt_info->smpt_lock, flags); - kfree(ref); -} - -/** - * mic_map_single - Maps a virtual address to a MIC physical address. - * - * @mdev: pointer to mic_device instance. - * @va: Kernel direct mapped virtual address. - * @size: Size of the region to be mapped. - * - * This API calls pci_map_single(..) for the direct mapped virtual address - * and then converts the DMA address provided to a DMA address understood - * by MIC. Caller should check for errors by calling mic_map_error(..). - * - * returns DMA address as required by MIC. - */ -dma_addr_t mic_map_single(struct mic_device *mdev, void *va, size_t size) -{ - dma_addr_t mic_addr = 0; - struct pci_dev *pdev = mdev->pdev; - dma_addr_t dma_addr = - pci_map_single(pdev, va, size, PCI_DMA_BIDIRECTIONAL); - - if (!pci_dma_mapping_error(pdev, dma_addr)) { - mic_addr = mic_map(mdev, dma_addr, size); - if (!mic_addr) { - dev_err(&mdev->pdev->dev, - "mic_map failed dma_addr 0x%llx size 0x%lx\n", - dma_addr, size); - pci_unmap_single(pdev, dma_addr, - size, PCI_DMA_BIDIRECTIONAL); - } - } - return mic_addr; -} - -/** - * mic_unmap_single - Unmaps a MIC physical address. - * - * @mdev: pointer to mic_device instance. - * @mic_addr: MIC physical address. - * @size: Size of the region to be unmapped. - * - * This API unmaps the mappings created by mic_map_single(..). - * - * returns None. - */ -void -mic_unmap_single(struct mic_device *mdev, dma_addr_t mic_addr, size_t size) -{ - struct pci_dev *pdev = mdev->pdev; - dma_addr_t dma_addr = mic_to_dma_addr(mdev, mic_addr); - mic_unmap(mdev, mic_addr, size); - pci_unmap_single(pdev, dma_addr, size, PCI_DMA_BIDIRECTIONAL); -} - -/** - * mic_smpt_init - Initialize MIC System Memory Page Tables. - * - * @mdev: pointer to mic_device instance. - * - * returns 0 for success and -errno for error. - */ -int mic_smpt_init(struct mic_device *mdev) -{ - int i, err = 0; - dma_addr_t dma_addr; - struct mic_smpt_info *smpt_info; - - mdev->smpt = kmalloc(sizeof(*mdev->smpt), GFP_KERNEL); - if (!mdev->smpt) - return -ENOMEM; - - smpt_info = mdev->smpt; - mdev->smpt_ops->init(mdev); - smpt_info->entry = kmalloc_array(smpt_info->info.num_reg, - sizeof(*smpt_info->entry), GFP_KERNEL); - if (!smpt_info->entry) { - err = -ENOMEM; - goto free_smpt; - } - spin_lock_init(&smpt_info->smpt_lock); - for (i = 0; i < smpt_info->info.num_reg; i++) { - dma_addr = i * smpt_info->info.page_size; - smpt_info->entry[i].dma_addr = dma_addr; - smpt_info->entry[i].ref_count = 0; - mdev->smpt_ops->set(mdev, dma_addr, i); - } - smpt_info->ref_count = 0; - smpt_info->map_count = 0; - smpt_info->unmap_count = 0; - return 0; -free_smpt: - kfree(smpt_info); - return err; -} - -/** - * mic_smpt_uninit - UnInitialize MIC System Memory Page Tables. - * - * @mdev: pointer to mic_device instance. - * - * returns None. - */ -void mic_smpt_uninit(struct mic_device *mdev) -{ - struct mic_smpt_info *smpt_info = mdev->smpt; - int i; - - dev_dbg(&mdev->pdev->dev, - "nodeid %d SMPT ref count %lld map %lld unmap %lld\n", - mdev->id, smpt_info->ref_count, - smpt_info->map_count, smpt_info->unmap_count); - - for (i = 0; i < smpt_info->info.num_reg; i++) { - dev_dbg(&mdev->pdev->dev, - "SMPT entry[%d] dma_addr = 0x%llx ref_count = %lld\n", - i, smpt_info->entry[i].dma_addr, - smpt_info->entry[i].ref_count); - if (smpt_info->entry[i].ref_count) - dev_warn(&mdev->pdev->dev, - "ref count for entry %d is not zero\n", i); - } - kfree(smpt_info->entry); - kfree(smpt_info); -} - -/** - * mic_smpt_restore - Restore MIC System Memory Page Tables. - * - * @mdev: pointer to mic_device instance. - * - * Restore the SMPT registers to values previously stored in the - * SW data structures. Some MIC steppings lose register state - * across resets and this API should be called for performing - * a restore operation if required. - * - * returns None. - */ -void mic_smpt_restore(struct mic_device *mdev) -{ - int i; - dma_addr_t dma_addr; - - for (i = 0; i < mdev->smpt->info.num_reg; i++) { - dma_addr = mdev->smpt->entry[i].dma_addr; - mdev->smpt_ops->set(mdev, dma_addr, i); - } -} diff --git a/drivers/misc/mic/host/mic_smpt.h b/drivers/misc/mic/host/mic_smpt.h deleted file mode 100644 index 3b1ec14a9d81..000000000000 --- a/drivers/misc/mic/host/mic_smpt.h +++ /dev/null @@ -1,87 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Intel MIC Host driver. - */ -#ifndef MIC_SMPT_H -#define MIC_SMPT_H -/** - * struct mic_smpt_ops - MIC HW specific SMPT operations. - * @init: Initialize hardware specific SMPT information in mic_smpt_hw_info. - * @set: Set the value for a particular SMPT entry. - */ -struct mic_smpt_ops { - void (*init)(struct mic_device *mdev); - void (*set)(struct mic_device *mdev, dma_addr_t dma_addr, u8 index); -}; - -/** - * struct mic_smpt - MIC SMPT entry information. - * @dma_addr: Base DMA address for this SMPT entry. - * @ref_count: Number of active mappings for this SMPT entry in bytes. - */ -struct mic_smpt { - dma_addr_t dma_addr; - s64 ref_count; -}; - -/** - * struct mic_smpt_hw_info - MIC SMPT hardware specific information. - * @num_reg: Number of SMPT registers. - * @page_shift: System memory page shift. - * @page_size: System memory page size. - * @base: System address base. - */ -struct mic_smpt_hw_info { - u8 num_reg; - u8 page_shift; - u64 page_size; - u64 base; -}; - -/** - * struct mic_smpt_info - MIC SMPT information. - * @entry: Array of SMPT entries. - * @smpt_lock: Spin lock protecting access to SMPT data structures. - * @info: Hardware specific SMPT information. - * @ref_count: Number of active SMPT mappings (for debug). - * @map_count: Number of SMPT mappings created (for debug). - * @unmap_count: Number of SMPT mappings destroyed (for debug). - */ -struct mic_smpt_info { - struct mic_smpt *entry; - spinlock_t smpt_lock; - struct mic_smpt_hw_info info; - s64 ref_count; - s64 map_count; - s64 unmap_count; -}; - -dma_addr_t mic_map_single(struct mic_device *mdev, void *va, size_t size); -void mic_unmap_single(struct mic_device *mdev, - dma_addr_t mic_addr, size_t size); -dma_addr_t mic_map(struct mic_device *mdev, - dma_addr_t dma_addr, size_t size); -void mic_unmap(struct mic_device *mdev, dma_addr_t mic_addr, size_t size); -dma_addr_t mic_to_dma_addr(struct mic_device *mdev, dma_addr_t mic_addr); - -/** - * mic_map_error - Check a MIC address for errors. - * - * @mdev: pointer to mic_device instance. - * - * returns Whether there was an error during mic_map..(..) APIs. - */ -static inline bool mic_map_error(dma_addr_t mic_addr) -{ - return !mic_addr; -} - -int mic_smpt_init(struct mic_device *mdev); -void mic_smpt_uninit(struct mic_device *mdev); -void mic_smpt_restore(struct mic_device *mdev); - -#endif diff --git a/drivers/misc/mic/host/mic_x100.c b/drivers/misc/mic/host/mic_x100.c deleted file mode 100644 index f5536c1ad607..000000000000 --- a/drivers/misc/mic/host/mic_x100.c +++ /dev/null @@ -1,585 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Intel MIC Host driver. - */ -#include -#include -#include -#include -#include - -#include "../common/mic_dev.h" -#include "mic_device.h" -#include "mic_x100.h" -#include "mic_smpt.h" - -static const u16 mic_x100_intr_init[] = { - MIC_X100_DOORBELL_IDX_START, - MIC_X100_DMA_IDX_START, - MIC_X100_ERR_IDX_START, - MIC_X100_NUM_DOORBELL, - MIC_X100_NUM_DMA, - MIC_X100_NUM_ERR, -}; - -/** - * mic_x100_write_spad - write to the scratchpad register - * @mdev: pointer to mic_device instance - * @idx: index to the scratchpad register, 0 based - * @val: the data value to put into the register - * - * This function allows writing of a 32bit value to the indexed scratchpad - * register. - * - * RETURNS: none. - */ -static void -mic_x100_write_spad(struct mic_device *mdev, unsigned int idx, u32 val) -{ - dev_dbg(&mdev->pdev->dev, "Writing 0x%x to scratch pad index %d\n", - val, idx); - mic_mmio_write(&mdev->mmio, val, - MIC_X100_SBOX_BASE_ADDRESS + - MIC_X100_SBOX_SPAD0 + idx * 4); -} - -/** - * mic_x100_read_spad - read from the scratchpad register - * @mdev: pointer to mic_device instance - * @idx: index to scratchpad register, 0 based - * - * This function allows reading of the 32bit scratchpad register. - * - * RETURNS: An appropriate -ERRNO error value on error, or zero for success. - */ -static u32 -mic_x100_read_spad(struct mic_device *mdev, unsigned int idx) -{ - u32 val = mic_mmio_read(&mdev->mmio, - MIC_X100_SBOX_BASE_ADDRESS + - MIC_X100_SBOX_SPAD0 + idx * 4); - - dev_dbg(&mdev->pdev->dev, - "Reading 0x%x from scratch pad index %d\n", val, idx); - return val; -} - -/** - * mic_x100_enable_interrupts - Enable interrupts. - * @mdev: pointer to mic_device instance - */ -static void mic_x100_enable_interrupts(struct mic_device *mdev) -{ - u32 reg; - struct mic_mw *mw = &mdev->mmio; - u32 sice0 = MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_SICE0; - u32 siac0 = MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_SIAC0; - - reg = mic_mmio_read(mw, sice0); - reg |= MIC_X100_SBOX_DBR_BITS(0xf) | MIC_X100_SBOX_DMA_BITS(0xff); - mic_mmio_write(mw, reg, sice0); - - /* - * Enable auto-clear when enabling interrupts. Applicable only for - * MSI-x. Legacy and MSI mode cannot have auto-clear enabled. - */ - if (mdev->irq_info.num_vectors > 1) { - reg = mic_mmio_read(mw, siac0); - reg |= MIC_X100_SBOX_DBR_BITS(0xf) | - MIC_X100_SBOX_DMA_BITS(0xff); - mic_mmio_write(mw, reg, siac0); - } -} - -/** - * mic_x100_disable_interrupts - Disable interrupts. - * @mdev: pointer to mic_device instance - */ -static void mic_x100_disable_interrupts(struct mic_device *mdev) -{ - u32 reg; - struct mic_mw *mw = &mdev->mmio; - u32 sice0 = MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_SICE0; - u32 siac0 = MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_SIAC0; - u32 sicc0 = MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_SICC0; - - reg = mic_mmio_read(mw, sice0); - mic_mmio_write(mw, reg, sicc0); - - if (mdev->irq_info.num_vectors > 1) { - reg = mic_mmio_read(mw, siac0); - reg &= ~(MIC_X100_SBOX_DBR_BITS(0xf) | - MIC_X100_SBOX_DMA_BITS(0xff)); - mic_mmio_write(mw, reg, siac0); - } -} - -/** - * mic_x100_send_sbox_intr - Send an MIC_X100_SBOX interrupt to MIC. - * @mdev: pointer to mic_device instance - * @doorbell: doorbell number - */ -static void mic_x100_send_sbox_intr(struct mic_device *mdev, - int doorbell) -{ - struct mic_mw *mw = &mdev->mmio; - u64 apic_icr_offset = MIC_X100_SBOX_APICICR0 + doorbell * 8; - u32 apicicr_low = mic_mmio_read(mw, MIC_X100_SBOX_BASE_ADDRESS + - apic_icr_offset); - - /* for MIC we need to make sure we "hit" the send_icr bit (13) */ - apicicr_low = (apicicr_low | (1 << 13)); - - /* Ensure that the interrupt is ordered w.r.t. previous stores. */ - wmb(); - mic_mmio_write(mw, apicicr_low, - MIC_X100_SBOX_BASE_ADDRESS + apic_icr_offset); -} - -/** - * mic_x100_send_rdmasr_intr - Send an RDMASR interrupt to MIC. - * @mdev: pointer to mic_device instance - * @doorbell: doorbell number - */ -static void mic_x100_send_rdmasr_intr(struct mic_device *mdev, - int doorbell) -{ - int rdmasr_offset = MIC_X100_SBOX_RDMASR0 + (doorbell << 2); - /* Ensure that the interrupt is ordered w.r.t. previous stores. */ - wmb(); - mic_mmio_write(&mdev->mmio, 0, - MIC_X100_SBOX_BASE_ADDRESS + rdmasr_offset); -} - -/** - * __mic_x100_send_intr - Send interrupt to MIC. - * @mdev: pointer to mic_device instance - * @doorbell: doorbell number. - */ -static void mic_x100_send_intr(struct mic_device *mdev, int doorbell) -{ - int rdmasr_db; - if (doorbell < MIC_X100_NUM_SBOX_IRQ) { - mic_x100_send_sbox_intr(mdev, doorbell); - } else { - rdmasr_db = doorbell - MIC_X100_NUM_SBOX_IRQ; - mic_x100_send_rdmasr_intr(mdev, rdmasr_db); - } -} - -/** - * mic_x100_ack_interrupt - Read the interrupt sources register and - * clear it. This function will be called in the MSI/INTx case. - * @mdev: Pointer to mic_device instance. - * - * Returns: bitmask of interrupt sources triggered. - */ -static u32 mic_x100_ack_interrupt(struct mic_device *mdev) -{ - u32 sicr0 = MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_SICR0; - u32 reg = mic_mmio_read(&mdev->mmio, sicr0); - mic_mmio_write(&mdev->mmio, reg, sicr0); - return reg; -} - -/** - * mic_x100_intr_workarounds - These hardware specific workarounds are - * to be invoked everytime an interrupt is handled. - * @mdev: Pointer to mic_device instance. - * - * Returns: none - */ -static void mic_x100_intr_workarounds(struct mic_device *mdev) -{ - struct mic_mw *mw = &mdev->mmio; - - /* Clear pending bit array. */ - if (MIC_A0_STEP == mdev->stepping) - mic_mmio_write(mw, 1, MIC_X100_SBOX_BASE_ADDRESS + - MIC_X100_SBOX_MSIXPBACR); - - if (mdev->stepping >= MIC_B0_STEP) - mdev->intr_ops->enable_interrupts(mdev); -} - -/** - * mic_x100_hw_intr_init - Initialize h/w specific interrupt - * information. - * @mdev: pointer to mic_device instance - */ -static void mic_x100_hw_intr_init(struct mic_device *mdev) -{ - mdev->intr_info = (struct mic_intr_info *)mic_x100_intr_init; -} - -/** - * mic_x100_read_msi_to_src_map - read from the MSI mapping registers - * @mdev: pointer to mic_device instance - * @idx: index to the mapping register, 0 based - * - * This function allows reading of the 32bit MSI mapping register. - * - * RETURNS: The value in the register. - */ -static u32 -mic_x100_read_msi_to_src_map(struct mic_device *mdev, int idx) -{ - return mic_mmio_read(&mdev->mmio, - MIC_X100_SBOX_BASE_ADDRESS + - MIC_X100_SBOX_MXAR0 + idx * 4); -} - -/** - * mic_x100_program_msi_to_src_map - program the MSI mapping registers - * @mdev: pointer to mic_device instance - * @idx: index to the mapping register, 0 based - * @offset: The bit offset in the register that needs to be updated. - * @set: boolean specifying if the bit in the specified offset needs - * to be set or cleared. - * - * RETURNS: None. - */ -static void -mic_x100_program_msi_to_src_map(struct mic_device *mdev, - int idx, int offset, bool set) -{ - unsigned long reg; - struct mic_mw *mw = &mdev->mmio; - u32 mxar = MIC_X100_SBOX_BASE_ADDRESS + - MIC_X100_SBOX_MXAR0 + idx * 4; - - reg = mic_mmio_read(mw, mxar); - if (set) - __set_bit(offset, ®); - else - __clear_bit(offset, ®); - mic_mmio_write(mw, reg, mxar); -} - -/* - * mic_x100_reset_fw_ready - Reset Firmware ready status field. - * @mdev: pointer to mic_device instance - */ -static void mic_x100_reset_fw_ready(struct mic_device *mdev) -{ - mdev->ops->write_spad(mdev, MIC_X100_DOWNLOAD_INFO, 0); -} - -/* - * mic_x100_is_fw_ready - Check if firmware is ready. - * @mdev: pointer to mic_device instance - */ -static bool mic_x100_is_fw_ready(struct mic_device *mdev) -{ - u32 scratch2 = mdev->ops->read_spad(mdev, MIC_X100_DOWNLOAD_INFO); - return MIC_X100_SPAD2_DOWNLOAD_STATUS(scratch2) ? true : false; -} - -/** - * mic_x100_get_apic_id - Get bootstrap APIC ID. - * @mdev: pointer to mic_device instance - */ -static u32 mic_x100_get_apic_id(struct mic_device *mdev) -{ - u32 scratch2 = 0; - - scratch2 = mdev->ops->read_spad(mdev, MIC_X100_DOWNLOAD_INFO); - return MIC_X100_SPAD2_APIC_ID(scratch2); -} - -/** - * mic_x100_send_firmware_intr - Send an interrupt to the firmware on MIC. - * @mdev: pointer to mic_device instance - */ -static void mic_x100_send_firmware_intr(struct mic_device *mdev) -{ - u32 apicicr_low; - u64 apic_icr_offset = MIC_X100_SBOX_APICICR7; - int vector = MIC_X100_BSP_INTERRUPT_VECTOR; - struct mic_mw *mw = &mdev->mmio; - - /* - * For MIC we need to make sure we "hit" - * the send_icr bit (13). - */ - apicicr_low = (vector | (1 << 13)); - - mic_mmio_write(mw, mic_x100_get_apic_id(mdev), - MIC_X100_SBOX_BASE_ADDRESS + apic_icr_offset + 4); - - /* Ensure that the interrupt is ordered w.r.t. previous stores. */ - wmb(); - mic_mmio_write(mw, apicicr_low, - MIC_X100_SBOX_BASE_ADDRESS + apic_icr_offset); -} - -/** - * mic_x100_hw_reset - Reset the MIC device. - * @mdev: pointer to mic_device instance - */ -static void mic_x100_hw_reset(struct mic_device *mdev) -{ - u32 reset_reg; - u32 rgcr = MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_RGCR; - struct mic_mw *mw = &mdev->mmio; - - /* Ensure that the reset is ordered w.r.t. previous loads and stores */ - mb(); - /* Trigger reset */ - reset_reg = mic_mmio_read(mw, rgcr); - reset_reg |= 0x1; - mic_mmio_write(mw, reset_reg, rgcr); - /* - * It seems we really want to delay at least 1 second - * after touching reset to prevent a lot of problems. - */ - msleep(1000); -} - -/** - * mic_x100_load_command_line - Load command line to MIC. - * @mdev: pointer to mic_device instance - * @fw: the firmware image - * - * RETURNS: An appropriate -ERRNO error value on error, or zero for success. - */ -static int -mic_x100_load_command_line(struct mic_device *mdev, const struct firmware *fw) -{ - u32 len = 0; - u32 boot_mem; - char *buf; - void __iomem *cmd_line_va = mdev->aper.va + mdev->bootaddr + fw->size; -#define CMDLINE_SIZE 2048 - - boot_mem = mdev->aper.len >> 20; - buf = kzalloc(CMDLINE_SIZE, GFP_KERNEL); - if (!buf) - return -ENOMEM; - - len += scnprintf(buf, CMDLINE_SIZE - len, - " mem=%dM", boot_mem); - if (mdev->cosm_dev->cmdline) - scnprintf(buf + len, CMDLINE_SIZE - len, " %s", - mdev->cosm_dev->cmdline); - memcpy_toio(cmd_line_va, buf, strlen(buf) + 1); - kfree(buf); - return 0; -} - -/** - * mic_x100_load_ramdisk - Load ramdisk to MIC. - * @mdev: pointer to mic_device instance - * - * RETURNS: An appropriate -ERRNO error value on error, or zero for success. - */ -static int -mic_x100_load_ramdisk(struct mic_device *mdev) -{ - const struct firmware *fw; - int rc; - struct boot_params __iomem *bp = mdev->aper.va + mdev->bootaddr; - - rc = request_firmware(&fw, mdev->cosm_dev->ramdisk, &mdev->pdev->dev); - if (rc < 0) { - dev_err(&mdev->pdev->dev, - "ramdisk request_firmware failed: %d %s\n", - rc, mdev->cosm_dev->ramdisk); - goto error; - } - /* - * Typically the bootaddr for card OS is 64M - * so copy over the ramdisk @ 128M. - */ - memcpy_toio(mdev->aper.va + (mdev->bootaddr << 1), fw->data, fw->size); - iowrite32(mdev->bootaddr << 1, &bp->hdr.ramdisk_image); - iowrite32(fw->size, &bp->hdr.ramdisk_size); - release_firmware(fw); -error: - return rc; -} - -/** - * mic_x100_get_boot_addr - Get MIC boot address. - * @mdev: pointer to mic_device instance - * - * This function is called during firmware load to determine - * the address at which the OS should be downloaded in card - * memory i.e. GDDR. - * RETURNS: An appropriate -ERRNO error value on error, or zero for success. - */ -static int -mic_x100_get_boot_addr(struct mic_device *mdev) -{ - u32 scratch2, boot_addr; - int rc = 0; - - scratch2 = mdev->ops->read_spad(mdev, MIC_X100_DOWNLOAD_INFO); - boot_addr = MIC_X100_SPAD2_DOWNLOAD_ADDR(scratch2); - dev_dbg(&mdev->pdev->dev, "%s %d boot_addr 0x%x\n", - __func__, __LINE__, boot_addr); - if (boot_addr > (1 << 31)) { - dev_err(&mdev->pdev->dev, - "incorrect bootaddr 0x%x\n", - boot_addr); - rc = -EINVAL; - goto error; - } - mdev->bootaddr = boot_addr; -error: - return rc; -} - -/** - * mic_x100_load_firmware - Load firmware to MIC. - * @mdev: pointer to mic_device instance - * @buf: buffer containing boot string including firmware/ramdisk path. - * - * RETURNS: An appropriate -ERRNO error value on error, or zero for success. - */ -static int -mic_x100_load_firmware(struct mic_device *mdev, const char *buf) -{ - int rc; - const struct firmware *fw; - - rc = mic_x100_get_boot_addr(mdev); - if (rc) - return rc; - /* load OS */ - rc = request_firmware(&fw, mdev->cosm_dev->firmware, &mdev->pdev->dev); - if (rc < 0) { - dev_err(&mdev->pdev->dev, - "ramdisk request_firmware failed: %d %s\n", - rc, mdev->cosm_dev->firmware); - return rc; - } - if (mdev->bootaddr > mdev->aper.len - fw->size) { - rc = -EINVAL; - dev_err(&mdev->pdev->dev, "%s %d rc %d bootaddr 0x%x\n", - __func__, __LINE__, rc, mdev->bootaddr); - goto error; - } - memcpy_toio(mdev->aper.va + mdev->bootaddr, fw->data, fw->size); - mdev->ops->write_spad(mdev, MIC_X100_FW_SIZE, fw->size); - if (!strcmp(mdev->cosm_dev->bootmode, "flash")) { - rc = -EINVAL; - dev_err(&mdev->pdev->dev, "%s %d rc %d\n", - __func__, __LINE__, rc); - goto error; - } - /* load command line */ - rc = mic_x100_load_command_line(mdev, fw); - if (rc) { - dev_err(&mdev->pdev->dev, "%s %d rc %d\n", - __func__, __LINE__, rc); - goto error; - } - release_firmware(fw); - /* load ramdisk */ - if (mdev->cosm_dev->ramdisk) - rc = mic_x100_load_ramdisk(mdev); - - return rc; - -error: - release_firmware(fw); - return rc; -} - -/** - * mic_x100_get_postcode - Get postcode status from firmware. - * @mdev: pointer to mic_device instance - * - * RETURNS: postcode. - */ -static u32 mic_x100_get_postcode(struct mic_device *mdev) -{ - return mic_mmio_read(&mdev->mmio, MIC_X100_POSTCODE); -} - -/** - * mic_x100_smpt_set - Update an SMPT entry with a DMA address. - * @mdev: pointer to mic_device instance - * @dma_addr: DMA address to use - * @index: entry to write to - * - * RETURNS: none. - */ -static void -mic_x100_smpt_set(struct mic_device *mdev, dma_addr_t dma_addr, u8 index) -{ -#define SNOOP_ON (0 << 0) -#define SNOOP_OFF (1 << 0) -/* - * Sbox Smpt Reg Bits: - * Bits 31:2 Host address - * Bits 1 RSVD - * Bits 0 No snoop - */ -#define BUILD_SMPT(NO_SNOOP, HOST_ADDR) \ - (u32)(((HOST_ADDR) << 2) | ((NO_SNOOP) & 0x01)) - - uint32_t smpt_reg_val = BUILD_SMPT(SNOOP_ON, - dma_addr >> mdev->smpt->info.page_shift); - mic_mmio_write(&mdev->mmio, smpt_reg_val, - MIC_X100_SBOX_BASE_ADDRESS + - MIC_X100_SBOX_SMPT00 + (4 * index)); -} - -/** - * mic_x100_smpt_hw_init - Initialize SMPT X100 specific fields. - * @mdev: pointer to mic_device instance - * - * RETURNS: none. - */ -static void mic_x100_smpt_hw_init(struct mic_device *mdev) -{ - struct mic_smpt_hw_info *info = &mdev->smpt->info; - - info->num_reg = 32; - info->page_shift = 34; - info->page_size = (1ULL << info->page_shift); - info->base = 0x8000000000ULL; -} - -struct mic_smpt_ops mic_x100_smpt_ops = { - .init = mic_x100_smpt_hw_init, - .set = mic_x100_smpt_set, -}; - -static bool mic_x100_dma_filter(struct dma_chan *chan, void *param) -{ - if (chan->device->dev->parent == (struct device *)param) - return true; - return false; -} - -struct mic_hw_ops mic_x100_ops = { - .aper_bar = MIC_X100_APER_BAR, - .mmio_bar = MIC_X100_MMIO_BAR, - .read_spad = mic_x100_read_spad, - .write_spad = mic_x100_write_spad, - .send_intr = mic_x100_send_intr, - .ack_interrupt = mic_x100_ack_interrupt, - .intr_workarounds = mic_x100_intr_workarounds, - .reset = mic_x100_hw_reset, - .reset_fw_ready = mic_x100_reset_fw_ready, - .is_fw_ready = mic_x100_is_fw_ready, - .send_firmware_intr = mic_x100_send_firmware_intr, - .load_mic_fw = mic_x100_load_firmware, - .get_postcode = mic_x100_get_postcode, - .dma_filter = mic_x100_dma_filter, -}; - -struct mic_hw_intr_ops mic_x100_intr_ops = { - .intr_init = mic_x100_hw_intr_init, - .enable_interrupts = mic_x100_enable_interrupts, - .disable_interrupts = mic_x100_disable_interrupts, - .program_msi_to_src_map = mic_x100_program_msi_to_src_map, - .read_msi_to_src_map = mic_x100_read_msi_to_src_map, -}; diff --git a/drivers/misc/mic/host/mic_x100.h b/drivers/misc/mic/host/mic_x100.h deleted file mode 100644 index aebcaed6fa72..000000000000 --- a/drivers/misc/mic/host/mic_x100.h +++ /dev/null @@ -1,77 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Intel MIC Host driver. - */ -#ifndef _MIC_X100_HW_H_ -#define _MIC_X100_HW_H_ - -#define MIC_X100_PCI_DEVICE_2250 0x2250 -#define MIC_X100_PCI_DEVICE_2251 0x2251 -#define MIC_X100_PCI_DEVICE_2252 0x2252 -#define MIC_X100_PCI_DEVICE_2253 0x2253 -#define MIC_X100_PCI_DEVICE_2254 0x2254 -#define MIC_X100_PCI_DEVICE_2255 0x2255 -#define MIC_X100_PCI_DEVICE_2256 0x2256 -#define MIC_X100_PCI_DEVICE_2257 0x2257 -#define MIC_X100_PCI_DEVICE_2258 0x2258 -#define MIC_X100_PCI_DEVICE_2259 0x2259 -#define MIC_X100_PCI_DEVICE_225a 0x225a -#define MIC_X100_PCI_DEVICE_225b 0x225b -#define MIC_X100_PCI_DEVICE_225c 0x225c -#define MIC_X100_PCI_DEVICE_225d 0x225d -#define MIC_X100_PCI_DEVICE_225e 0x225e - -#define MIC_X100_APER_BAR 0 -#define MIC_X100_MMIO_BAR 4 - -#define MIC_X100_SBOX_BASE_ADDRESS 0x00010000 -#define MIC_X100_SBOX_SPAD0 0x0000AB20 -#define MIC_X100_SBOX_SICR0_DBR(x) ((x) & 0xf) -#define MIC_X100_SBOX_SICR0_DMA(x) (((x) >> 8) & 0xff) -#define MIC_X100_SBOX_SICE0_DBR(x) ((x) & 0xf) -#define MIC_X100_SBOX_DBR_BITS(x) ((x) & 0xf) -#define MIC_X100_SBOX_SICE0_DMA(x) (((x) >> 8) & 0xff) -#define MIC_X100_SBOX_DMA_BITS(x) (((x) & 0xff) << 8) - -#define MIC_X100_SBOX_APICICR0 0x0000A9D0 -#define MIC_X100_SBOX_SICR0 0x00009004 -#define MIC_X100_SBOX_SICE0 0x0000900C -#define MIC_X100_SBOX_SICC0 0x00009010 -#define MIC_X100_SBOX_SIAC0 0x00009014 -#define MIC_X100_SBOX_MSIXPBACR 0x00009084 -#define MIC_X100_SBOX_MXAR0 0x00009044 -#define MIC_X100_SBOX_SMPT00 0x00003100 -#define MIC_X100_SBOX_RDMASR0 0x0000B180 - -#define MIC_X100_DOORBELL_IDX_START 0 -#define MIC_X100_NUM_DOORBELL 4 -#define MIC_X100_DMA_IDX_START 8 -#define MIC_X100_NUM_DMA 8 -#define MIC_X100_ERR_IDX_START 30 -#define MIC_X100_NUM_ERR 1 - -#define MIC_X100_NUM_SBOX_IRQ 8 -#define MIC_X100_NUM_RDMASR_IRQ 8 -#define MIC_X100_RDMASR_IRQ_BASE 17 -#define MIC_X100_SPAD2_DOWNLOAD_STATUS(x) ((x) & 0x1) -#define MIC_X100_SPAD2_APIC_ID(x) (((x) >> 1) & 0x1ff) -#define MIC_X100_SPAD2_DOWNLOAD_ADDR(x) ((x) & 0xfffff000) -#define MIC_X100_SBOX_APICICR7 0x0000AA08 -#define MIC_X100_SBOX_RGCR 0x00004010 -#define MIC_X100_SBOX_SDBIC0 0x0000CC90 -#define MIC_X100_DOWNLOAD_INFO 2 -#define MIC_X100_FW_SIZE 5 -#define MIC_X100_POSTCODE 0x242c - -/* Host->Card(bootstrap) Interrupt Vector */ -#define MIC_X100_BSP_INTERRUPT_VECTOR 229 - -extern struct mic_hw_ops mic_x100_ops; -extern struct mic_smpt_ops mic_x100_smpt_ops; -extern struct mic_hw_intr_ops mic_x100_intr_ops; - -#endif diff --git a/drivers/misc/mic/scif/Makefile b/drivers/misc/mic/scif/Makefile deleted file mode 100644 index ff372555d118..000000000000 --- a/drivers/misc/mic/scif/Makefile +++ /dev/null @@ -1,21 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0 -# -# Makefile - SCIF driver. -# Copyright(c) 2014, Intel Corporation. -# -obj-$(CONFIG_SCIF) += scif.o -scif-objs := scif_main.o -scif-objs += scif_peer_bus.o -scif-objs += scif_ports.o -scif-objs += scif_debugfs.o -scif-objs += scif_fd.o -scif-objs += scif_api.o -scif-objs += scif_epd.o -scif-objs += scif_rb.o -scif-objs += scif_nodeqp.o -scif-objs += scif_nm.o -scif-objs += scif_dma.o -scif-objs += scif_fence.o -scif-objs += scif_mmap.o -scif-objs += scif_rma.o -scif-objs += scif_rma_list.o diff --git a/drivers/misc/mic/scif/scif_api.c b/drivers/misc/mic/scif/scif_api.c deleted file mode 100644 index 304d6c833712..000000000000 --- a/drivers/misc/mic/scif/scif_api.c +++ /dev/null @@ -1,1485 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel SCIF driver. - */ -#include -#include "scif_main.h" -#include "scif_map.h" - -static const char * const scif_ep_states[] = { - "Unbound", - "Bound", - "Listening", - "Connected", - "Connecting", - "Mapping", - "Closing", - "Close Listening", - "Disconnected", - "Zombie"}; - -enum conn_async_state { - ASYNC_CONN_IDLE = 1, /* ep setup for async connect */ - ASYNC_CONN_INPROGRESS, /* async connect in progress */ - ASYNC_CONN_FLUSH_WORK /* async work flush in progress */ -}; - -/* - * File operations for anonymous inode file associated with a SCIF endpoint, - * used in kernel mode SCIF poll. Kernel mode SCIF poll calls portions of the - * poll API in the kernel and these take in a struct file *. Since a struct - * file is not available to kernel mode SCIF, it uses an anonymous file for - * this purpose. - */ -const struct file_operations scif_anon_fops = { - .owner = THIS_MODULE, -}; - -scif_epd_t scif_open(void) -{ - struct scif_endpt *ep; - int err; - - might_sleep(); - ep = kzalloc(sizeof(*ep), GFP_KERNEL); - if (!ep) - goto err_ep_alloc; - - ep->qp_info.qp = kzalloc(sizeof(*ep->qp_info.qp), GFP_KERNEL); - if (!ep->qp_info.qp) - goto err_qp_alloc; - - err = scif_anon_inode_getfile(ep); - if (err) - goto err_anon_inode; - - spin_lock_init(&ep->lock); - mutex_init(&ep->sendlock); - mutex_init(&ep->recvlock); - - scif_rma_ep_init(ep); - ep->state = SCIFEP_UNBOUND; - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI open: ep %p success\n", ep); - return ep; - -err_anon_inode: - kfree(ep->qp_info.qp); -err_qp_alloc: - kfree(ep); -err_ep_alloc: - return NULL; -} -EXPORT_SYMBOL_GPL(scif_open); - -/* - * scif_disconnect_ep - Disconnects the endpoint if found - * @epd: The end point returned from scif_open() - */ -static struct scif_endpt *scif_disconnect_ep(struct scif_endpt *ep) -{ - struct scifmsg msg; - struct scif_endpt *fep = NULL; - struct scif_endpt *tmpep; - struct list_head *pos, *tmpq; - int err; - - /* - * Wake up any threads blocked in send()/recv() before closing - * out the connection. Grabbing and releasing the send/recv lock - * will ensure that any blocked senders/receivers have exited for - * Ring 0 endpoints. It is a Ring 0 bug to call send/recv after - * close. Ring 3 endpoints are not affected since close will not - * be called while there are IOCTLs executing. - */ - wake_up_interruptible(&ep->sendwq); - wake_up_interruptible(&ep->recvwq); - mutex_lock(&ep->sendlock); - mutex_unlock(&ep->sendlock); - mutex_lock(&ep->recvlock); - mutex_unlock(&ep->recvlock); - - /* Remove from the connected list */ - mutex_lock(&scif_info.connlock); - list_for_each_safe(pos, tmpq, &scif_info.connected) { - tmpep = list_entry(pos, struct scif_endpt, list); - if (tmpep == ep) { - list_del(pos); - fep = tmpep; - spin_lock(&ep->lock); - break; - } - } - - if (!fep) { - /* - * The other side has completed the disconnect before - * the end point can be removed from the list. Therefore - * the ep lock is not locked, traverse the disconnected - * list to find the endpoint and release the conn lock. - */ - list_for_each_safe(pos, tmpq, &scif_info.disconnected) { - tmpep = list_entry(pos, struct scif_endpt, list); - if (tmpep == ep) { - list_del(pos); - break; - } - } - mutex_unlock(&scif_info.connlock); - return NULL; - } - - init_completion(&ep->discon); - msg.uop = SCIF_DISCNCT; - msg.src = ep->port; - msg.dst = ep->peer; - msg.payload[0] = (u64)ep; - msg.payload[1] = ep->remote_ep; - - err = scif_nodeqp_send(ep->remote_dev, &msg); - spin_unlock(&ep->lock); - mutex_unlock(&scif_info.connlock); - - if (!err) - /* Wait for the remote node to respond with SCIF_DISCNT_ACK */ - wait_for_completion_timeout(&ep->discon, - SCIF_NODE_ALIVE_TIMEOUT); - return ep; -} - -int scif_close(scif_epd_t epd) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - struct scif_endpt *tmpep; - struct list_head *pos, *tmpq; - enum scif_epd_state oldstate; - bool flush_conn; - - dev_dbg(scif_info.mdev.this_device, "SCIFAPI close: ep %p %s\n", - ep, scif_ep_states[ep->state]); - might_sleep(); - spin_lock(&ep->lock); - flush_conn = (ep->conn_async_state == ASYNC_CONN_INPROGRESS); - spin_unlock(&ep->lock); - - if (flush_conn) - flush_work(&scif_info.conn_work); - - spin_lock(&ep->lock); - oldstate = ep->state; - - ep->state = SCIFEP_CLOSING; - - switch (oldstate) { - case SCIFEP_ZOMBIE: - dev_err(scif_info.mdev.this_device, - "SCIFAPI close: zombie state unexpected\n"); - fallthrough; - case SCIFEP_DISCONNECTED: - spin_unlock(&ep->lock); - scif_unregister_all_windows(epd); - /* Remove from the disconnected list */ - mutex_lock(&scif_info.connlock); - list_for_each_safe(pos, tmpq, &scif_info.disconnected) { - tmpep = list_entry(pos, struct scif_endpt, list); - if (tmpep == ep) { - list_del(pos); - break; - } - } - mutex_unlock(&scif_info.connlock); - break; - case SCIFEP_UNBOUND: - case SCIFEP_BOUND: - case SCIFEP_CONNECTING: - spin_unlock(&ep->lock); - break; - case SCIFEP_MAPPING: - case SCIFEP_CONNECTED: - case SCIFEP_CLOSING: - { - spin_unlock(&ep->lock); - scif_unregister_all_windows(epd); - scif_disconnect_ep(ep); - break; - } - case SCIFEP_LISTENING: - case SCIFEP_CLLISTEN: - { - struct scif_conreq *conreq; - struct scifmsg msg; - struct scif_endpt *aep; - - spin_unlock(&ep->lock); - mutex_lock(&scif_info.eplock); - - /* remove from listen list */ - list_for_each_safe(pos, tmpq, &scif_info.listen) { - tmpep = list_entry(pos, struct scif_endpt, list); - if (tmpep == ep) - list_del(pos); - } - /* Remove any dangling accepts */ - while (ep->acceptcnt) { - aep = list_first_entry(&ep->li_accept, - struct scif_endpt, liacceptlist); - list_del(&aep->liacceptlist); - scif_put_port(aep->port.port); - list_for_each_safe(pos, tmpq, &scif_info.uaccept) { - tmpep = list_entry(pos, struct scif_endpt, - miacceptlist); - if (tmpep == aep) { - list_del(pos); - break; - } - } - mutex_unlock(&scif_info.eplock); - mutex_lock(&scif_info.connlock); - list_for_each_safe(pos, tmpq, &scif_info.connected) { - tmpep = list_entry(pos, - struct scif_endpt, list); - if (tmpep == aep) { - list_del(pos); - break; - } - } - list_for_each_safe(pos, tmpq, &scif_info.disconnected) { - tmpep = list_entry(pos, - struct scif_endpt, list); - if (tmpep == aep) { - list_del(pos); - break; - } - } - mutex_unlock(&scif_info.connlock); - scif_teardown_ep(aep); - mutex_lock(&scif_info.eplock); - scif_add_epd_to_zombie_list(aep, SCIF_EPLOCK_HELD); - ep->acceptcnt--; - } - - spin_lock(&ep->lock); - mutex_unlock(&scif_info.eplock); - - /* Remove and reject any pending connection requests. */ - while (ep->conreqcnt) { - conreq = list_first_entry(&ep->conlist, - struct scif_conreq, list); - list_del(&conreq->list); - - msg.uop = SCIF_CNCT_REJ; - msg.dst.node = conreq->msg.src.node; - msg.dst.port = conreq->msg.src.port; - msg.payload[0] = conreq->msg.payload[0]; - msg.payload[1] = conreq->msg.payload[1]; - /* - * No Error Handling on purpose for scif_nodeqp_send(). - * If the remote node is lost we still want free the - * connection requests on the self node. - */ - scif_nodeqp_send(&scif_dev[conreq->msg.src.node], - &msg); - ep->conreqcnt--; - kfree(conreq); - } - - spin_unlock(&ep->lock); - /* If a kSCIF accept is waiting wake it up */ - wake_up_interruptible(&ep->conwq); - break; - } - } - scif_put_port(ep->port.port); - scif_anon_inode_fput(ep); - scif_teardown_ep(ep); - scif_add_epd_to_zombie_list(ep, !SCIF_EPLOCK_HELD); - return 0; -} -EXPORT_SYMBOL_GPL(scif_close); - -/** - * scif_flush() - Wakes up any blocking accepts. The endpoint will no longer - * accept new connections. - * @epd: The end point returned from scif_open() - */ -int __scif_flush(scif_epd_t epd) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - - switch (ep->state) { - case SCIFEP_LISTENING: - { - ep->state = SCIFEP_CLLISTEN; - - /* If an accept is waiting wake it up */ - wake_up_interruptible(&ep->conwq); - break; - } - default: - break; - } - return 0; -} - -int scif_bind(scif_epd_t epd, u16 pn) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - int ret = 0; - int tmp; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI bind: ep %p %s requested port number %d\n", - ep, scif_ep_states[ep->state], pn); - if (pn) { - /* - * Similar to IETF RFC 1700, SCIF ports below - * SCIF_ADMIN_PORT_END can only be bound by system (or root) - * processes or by processes executed by privileged users. - */ - if (pn < SCIF_ADMIN_PORT_END && !capable(CAP_SYS_ADMIN)) { - ret = -EACCES; - goto scif_bind_admin_exit; - } - } - - spin_lock(&ep->lock); - if (ep->state == SCIFEP_BOUND) { - ret = -EINVAL; - goto scif_bind_exit; - } else if (ep->state != SCIFEP_UNBOUND) { - ret = -EISCONN; - goto scif_bind_exit; - } - - if (pn) { - tmp = scif_rsrv_port(pn); - if (tmp != pn) { - ret = -EINVAL; - goto scif_bind_exit; - } - } else { - ret = scif_get_new_port(); - if (ret < 0) - goto scif_bind_exit; - pn = ret; - } - - ep->state = SCIFEP_BOUND; - ep->port.node = scif_info.nodeid; - ep->port.port = pn; - ep->conn_async_state = ASYNC_CONN_IDLE; - ret = pn; - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI bind: bound to port number %d\n", pn); -scif_bind_exit: - spin_unlock(&ep->lock); -scif_bind_admin_exit: - return ret; -} -EXPORT_SYMBOL_GPL(scif_bind); - -int scif_listen(scif_epd_t epd, int backlog) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI listen: ep %p %s\n", ep, scif_ep_states[ep->state]); - spin_lock(&ep->lock); - switch (ep->state) { - case SCIFEP_ZOMBIE: - case SCIFEP_CLOSING: - case SCIFEP_CLLISTEN: - case SCIFEP_UNBOUND: - case SCIFEP_DISCONNECTED: - spin_unlock(&ep->lock); - return -EINVAL; - case SCIFEP_LISTENING: - case SCIFEP_CONNECTED: - case SCIFEP_CONNECTING: - case SCIFEP_MAPPING: - spin_unlock(&ep->lock); - return -EISCONN; - case SCIFEP_BOUND: - break; - } - - ep->state = SCIFEP_LISTENING; - ep->backlog = backlog; - - ep->conreqcnt = 0; - ep->acceptcnt = 0; - INIT_LIST_HEAD(&ep->conlist); - init_waitqueue_head(&ep->conwq); - INIT_LIST_HEAD(&ep->li_accept); - spin_unlock(&ep->lock); - - /* - * Listen status is complete so delete the qp information not needed - * on a listen before placing on the list of listening ep's - */ - scif_teardown_ep(ep); - ep->qp_info.qp = NULL; - - mutex_lock(&scif_info.eplock); - list_add_tail(&ep->list, &scif_info.listen); - mutex_unlock(&scif_info.eplock); - return 0; -} -EXPORT_SYMBOL_GPL(scif_listen); - -/* - ************************************************************************ - * SCIF connection flow: - * - * 1) A SCIF listening endpoint can call scif_accept(..) to wait for SCIF - * connections via a SCIF_CNCT_REQ message - * 2) A SCIF endpoint can initiate a SCIF connection by calling - * scif_connect(..) which calls scif_setup_qp_connect(..) which - * allocates the local qp for the endpoint ring buffer and then sends - * a SCIF_CNCT_REQ to the remote node and waits for a SCIF_CNCT_GNT or - * a SCIF_CNCT_REJ message - * 3) The peer node handles a SCIF_CNCT_REQ via scif_cnctreq_resp(..) which - * wakes up any threads blocked in step 1 or sends a SCIF_CNCT_REJ - * message otherwise - * 4) A thread blocked waiting for incoming connections allocates its local - * endpoint QP and ring buffer following which it sends a SCIF_CNCT_GNT - * and waits for a SCIF_CNCT_GNT(N)ACK. If the allocation fails then - * the node sends a SCIF_CNCT_REJ message - * 5) Upon receipt of a SCIF_CNCT_GNT or a SCIF_CNCT_REJ message the - * connecting endpoint is woken up as part of handling - * scif_cnctgnt_resp(..) following which it maps the remote endpoints' - * QP, updates its outbound QP and sends a SCIF_CNCT_GNTACK message on - * success or a SCIF_CNCT_GNTNACK message on failure and completes - * the scif_connect(..) API - * 6) Upon receipt of a SCIF_CNCT_GNT(N)ACK the accepting endpoint blocked - * in step 4 is woken up and completes the scif_accept(..) API - * 7) The SCIF connection is now established between the two SCIF endpoints. - */ -static int scif_conn_func(struct scif_endpt *ep) -{ - int err = 0; - struct scifmsg msg; - struct device *spdev; - - err = scif_reserve_dma_chan(ep); - if (err) { - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, err); - ep->state = SCIFEP_BOUND; - goto connect_error_simple; - } - /* Initiate the first part of the endpoint QP setup */ - err = scif_setup_qp_connect(ep->qp_info.qp, &ep->qp_info.qp_offset, - SCIF_ENDPT_QP_SIZE, ep->remote_dev); - if (err) { - dev_err(&ep->remote_dev->sdev->dev, - "%s err %d qp_offset 0x%llx\n", - __func__, err, ep->qp_info.qp_offset); - ep->state = SCIFEP_BOUND; - goto connect_error_simple; - } - - spdev = scif_get_peer_dev(ep->remote_dev); - if (IS_ERR(spdev)) { - err = PTR_ERR(spdev); - goto cleanup_qp; - } - /* Format connect message and send it */ - msg.src = ep->port; - msg.dst = ep->conn_port; - msg.uop = SCIF_CNCT_REQ; - msg.payload[0] = (u64)ep; - msg.payload[1] = ep->qp_info.qp_offset; - err = _scif_nodeqp_send(ep->remote_dev, &msg); - if (err) - goto connect_error_dec; - scif_put_peer_dev(spdev); - /* - * Wait for the remote node to respond with SCIF_CNCT_GNT or - * SCIF_CNCT_REJ message. - */ - err = wait_event_timeout(ep->conwq, ep->state != SCIFEP_CONNECTING, - SCIF_NODE_ALIVE_TIMEOUT); - if (!err) { - dev_err(&ep->remote_dev->sdev->dev, - "%s %d timeout\n", __func__, __LINE__); - ep->state = SCIFEP_BOUND; - } - spdev = scif_get_peer_dev(ep->remote_dev); - if (IS_ERR(spdev)) { - err = PTR_ERR(spdev); - goto cleanup_qp; - } - if (ep->state == SCIFEP_MAPPING) { - err = scif_setup_qp_connect_response(ep->remote_dev, - ep->qp_info.qp, - ep->qp_info.gnt_pld); - /* - * If the resource to map the queue are not available then - * we need to tell the other side to terminate the accept - */ - if (err) { - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, err); - msg.uop = SCIF_CNCT_GNTNACK; - msg.payload[0] = ep->remote_ep; - _scif_nodeqp_send(ep->remote_dev, &msg); - ep->state = SCIFEP_BOUND; - goto connect_error_dec; - } - - msg.uop = SCIF_CNCT_GNTACK; - msg.payload[0] = ep->remote_ep; - err = _scif_nodeqp_send(ep->remote_dev, &msg); - if (err) { - ep->state = SCIFEP_BOUND; - goto connect_error_dec; - } - ep->state = SCIFEP_CONNECTED; - mutex_lock(&scif_info.connlock); - list_add_tail(&ep->list, &scif_info.connected); - mutex_unlock(&scif_info.connlock); - dev_dbg(&ep->remote_dev->sdev->dev, - "SCIFAPI connect: ep %p connected\n", ep); - } else if (ep->state == SCIFEP_BOUND) { - dev_dbg(&ep->remote_dev->sdev->dev, - "SCIFAPI connect: ep %p connection refused\n", ep); - err = -ECONNREFUSED; - goto connect_error_dec; - } - scif_put_peer_dev(spdev); - return err; -connect_error_dec: - scif_put_peer_dev(spdev); -cleanup_qp: - scif_cleanup_ep_qp(ep); -connect_error_simple: - return err; -} - -/* - * scif_conn_handler: - * - * Workqueue handler for servicing non-blocking SCIF connect - * - */ -void scif_conn_handler(struct work_struct *work) -{ - struct scif_endpt *ep; - - do { - ep = NULL; - spin_lock(&scif_info.nb_connect_lock); - if (!list_empty(&scif_info.nb_connect_list)) { - ep = list_first_entry(&scif_info.nb_connect_list, - struct scif_endpt, conn_list); - list_del(&ep->conn_list); - } - spin_unlock(&scif_info.nb_connect_lock); - if (ep) { - ep->conn_err = scif_conn_func(ep); - wake_up_interruptible(&ep->conn_pend_wq); - } - } while (ep); -} - -int __scif_connect(scif_epd_t epd, struct scif_port_id *dst, bool non_block) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - int err = 0; - struct scif_dev *remote_dev; - struct device *spdev; - - dev_dbg(scif_info.mdev.this_device, "SCIFAPI connect: ep %p %s\n", ep, - scif_ep_states[ep->state]); - - if (!scif_dev || dst->node > scif_info.maxid) - return -ENODEV; - - might_sleep(); - - remote_dev = &scif_dev[dst->node]; - spdev = scif_get_peer_dev(remote_dev); - if (IS_ERR(spdev)) { - err = PTR_ERR(spdev); - return err; - } - - spin_lock(&ep->lock); - switch (ep->state) { - case SCIFEP_ZOMBIE: - case SCIFEP_CLOSING: - err = -EINVAL; - break; - case SCIFEP_DISCONNECTED: - if (ep->conn_async_state == ASYNC_CONN_INPROGRESS) - ep->conn_async_state = ASYNC_CONN_FLUSH_WORK; - else - err = -EINVAL; - break; - case SCIFEP_LISTENING: - case SCIFEP_CLLISTEN: - err = -EOPNOTSUPP; - break; - case SCIFEP_CONNECTING: - case SCIFEP_MAPPING: - if (ep->conn_async_state == ASYNC_CONN_INPROGRESS) - err = -EINPROGRESS; - else - err = -EISCONN; - break; - case SCIFEP_CONNECTED: - if (ep->conn_async_state == ASYNC_CONN_INPROGRESS) - ep->conn_async_state = ASYNC_CONN_FLUSH_WORK; - else - err = -EISCONN; - break; - case SCIFEP_UNBOUND: - err = scif_get_new_port(); - if (err < 0) - break; - ep->port.port = err; - ep->port.node = scif_info.nodeid; - ep->conn_async_state = ASYNC_CONN_IDLE; - fallthrough; - case SCIFEP_BOUND: - /* - * If a non-blocking connect has been already initiated - * (conn_async_state is either ASYNC_CONN_INPROGRESS or - * ASYNC_CONN_FLUSH_WORK), the end point could end up in - * SCIF_BOUND due an error in the connection process - * (e.g., connection refused) If conn_async_state is - * ASYNC_CONN_INPROGRESS - transition to ASYNC_CONN_FLUSH_WORK - * so that the error status can be collected. If the state is - * already ASYNC_CONN_FLUSH_WORK - then set the error to - * EINPROGRESS since some other thread is waiting to collect - * error status. - */ - if (ep->conn_async_state == ASYNC_CONN_INPROGRESS) { - ep->conn_async_state = ASYNC_CONN_FLUSH_WORK; - } else if (ep->conn_async_state == ASYNC_CONN_FLUSH_WORK) { - err = -EINPROGRESS; - } else { - ep->conn_port = *dst; - init_waitqueue_head(&ep->sendwq); - init_waitqueue_head(&ep->recvwq); - init_waitqueue_head(&ep->conwq); - ep->conn_async_state = 0; - - if (unlikely(non_block)) - ep->conn_async_state = ASYNC_CONN_INPROGRESS; - } - break; - } - - if (err || ep->conn_async_state == ASYNC_CONN_FLUSH_WORK) - goto connect_simple_unlock1; - - ep->state = SCIFEP_CONNECTING; - ep->remote_dev = &scif_dev[dst->node]; - ep->qp_info.qp->magic = SCIFEP_MAGIC; - if (ep->conn_async_state == ASYNC_CONN_INPROGRESS) { - init_waitqueue_head(&ep->conn_pend_wq); - spin_lock(&scif_info.nb_connect_lock); - list_add_tail(&ep->conn_list, &scif_info.nb_connect_list); - spin_unlock(&scif_info.nb_connect_lock); - err = -EINPROGRESS; - schedule_work(&scif_info.conn_work); - } -connect_simple_unlock1: - spin_unlock(&ep->lock); - scif_put_peer_dev(spdev); - if (err) { - return err; - } else if (ep->conn_async_state == ASYNC_CONN_FLUSH_WORK) { - flush_work(&scif_info.conn_work); - err = ep->conn_err; - spin_lock(&ep->lock); - ep->conn_async_state = ASYNC_CONN_IDLE; - spin_unlock(&ep->lock); - } else { - err = scif_conn_func(ep); - } - return err; -} - -int scif_connect(scif_epd_t epd, struct scif_port_id *dst) -{ - return __scif_connect(epd, dst, false); -} -EXPORT_SYMBOL_GPL(scif_connect); - -/* - * scif_accept() - Accept a connection request from the remote node - * - * The function accepts a connection request from the remote node. Successful - * complete is indicate by a new end point being created and passed back - * to the caller for future reference. - * - * Upon successful complete a zero will be returned and the peer information - * will be filled in. - * - * If the end point is not in the listening state -EINVAL will be returned. - * - * If during the connection sequence resource allocation fails the -ENOMEM - * will be returned. - * - * If the function is called with the ASYNC flag set and no connection requests - * are pending it will return -EAGAIN. - * - * If the remote side is not sending any connection requests the caller may - * terminate this function with a signal. If so a -EINTR will be returned. - */ -int scif_accept(scif_epd_t epd, struct scif_port_id *peer, - scif_epd_t *newepd, int flags) -{ - struct scif_endpt *lep = (struct scif_endpt *)epd; - struct scif_endpt *cep; - struct scif_conreq *conreq; - struct scifmsg msg; - int err; - struct device *spdev; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI accept: ep %p %s\n", lep, scif_ep_states[lep->state]); - - if (flags & ~SCIF_ACCEPT_SYNC) - return -EINVAL; - - if (!peer || !newepd) - return -EINVAL; - - might_sleep(); - spin_lock(&lep->lock); - if (lep->state != SCIFEP_LISTENING) { - spin_unlock(&lep->lock); - return -EINVAL; - } - - if (!lep->conreqcnt && !(flags & SCIF_ACCEPT_SYNC)) { - /* No connection request present and we do not want to wait */ - spin_unlock(&lep->lock); - return -EAGAIN; - } - - lep->files = current->files; -retry_connection: - spin_unlock(&lep->lock); - /* Wait for the remote node to send us a SCIF_CNCT_REQ */ - err = wait_event_interruptible(lep->conwq, - (lep->conreqcnt || - (lep->state != SCIFEP_LISTENING))); - if (err) - return err; - - if (lep->state != SCIFEP_LISTENING) - return -EINTR; - - spin_lock(&lep->lock); - - if (!lep->conreqcnt) - goto retry_connection; - - /* Get the first connect request off the list */ - conreq = list_first_entry(&lep->conlist, struct scif_conreq, list); - list_del(&conreq->list); - lep->conreqcnt--; - spin_unlock(&lep->lock); - - /* Fill in the peer information */ - peer->node = conreq->msg.src.node; - peer->port = conreq->msg.src.port; - - cep = kzalloc(sizeof(*cep), GFP_KERNEL); - if (!cep) { - err = -ENOMEM; - goto scif_accept_error_epalloc; - } - spin_lock_init(&cep->lock); - mutex_init(&cep->sendlock); - mutex_init(&cep->recvlock); - cep->state = SCIFEP_CONNECTING; - cep->remote_dev = &scif_dev[peer->node]; - cep->remote_ep = conreq->msg.payload[0]; - - scif_rma_ep_init(cep); - - err = scif_reserve_dma_chan(cep); - if (err) { - dev_err(scif_info.mdev.this_device, - "%s %d err %d\n", __func__, __LINE__, err); - goto scif_accept_error_qpalloc; - } - - cep->qp_info.qp = kzalloc(sizeof(*cep->qp_info.qp), GFP_KERNEL); - if (!cep->qp_info.qp) { - err = -ENOMEM; - goto scif_accept_error_qpalloc; - } - - err = scif_anon_inode_getfile(cep); - if (err) - goto scif_accept_error_anon_inode; - - cep->qp_info.qp->magic = SCIFEP_MAGIC; - spdev = scif_get_peer_dev(cep->remote_dev); - if (IS_ERR(spdev)) { - err = PTR_ERR(spdev); - goto scif_accept_error_map; - } - err = scif_setup_qp_accept(cep->qp_info.qp, &cep->qp_info.qp_offset, - conreq->msg.payload[1], SCIF_ENDPT_QP_SIZE, - cep->remote_dev); - if (err) { - dev_dbg(&cep->remote_dev->sdev->dev, - "SCIFAPI accept: ep %p new %p scif_setup_qp_accept %d qp_offset 0x%llx\n", - lep, cep, err, cep->qp_info.qp_offset); - scif_put_peer_dev(spdev); - goto scif_accept_error_map; - } - - cep->port.node = lep->port.node; - cep->port.port = lep->port.port; - cep->peer.node = peer->node; - cep->peer.port = peer->port; - init_waitqueue_head(&cep->sendwq); - init_waitqueue_head(&cep->recvwq); - init_waitqueue_head(&cep->conwq); - - msg.uop = SCIF_CNCT_GNT; - msg.src = cep->port; - msg.payload[0] = cep->remote_ep; - msg.payload[1] = cep->qp_info.qp_offset; - msg.payload[2] = (u64)cep; - - err = _scif_nodeqp_send(cep->remote_dev, &msg); - scif_put_peer_dev(spdev); - if (err) - goto scif_accept_error_map; -retry: - /* Wait for the remote node to respond with SCIF_CNCT_GNT(N)ACK */ - err = wait_event_timeout(cep->conwq, cep->state != SCIFEP_CONNECTING, - SCIF_NODE_ACCEPT_TIMEOUT); - if (!err && scifdev_alive(cep)) - goto retry; - err = !err ? -ENODEV : 0; - if (err) - goto scif_accept_error_map; - kfree(conreq); - - spin_lock(&cep->lock); - - if (cep->state == SCIFEP_CLOSING) { - /* - * Remote failed to allocate resources and NAKed the grant. - * There is at this point nothing referencing the new end point. - */ - spin_unlock(&cep->lock); - scif_teardown_ep(cep); - kfree(cep); - - /* If call with sync flag then go back and wait. */ - if (flags & SCIF_ACCEPT_SYNC) { - spin_lock(&lep->lock); - goto retry_connection; - } - return -EAGAIN; - } - - scif_get_port(cep->port.port); - *newepd = (scif_epd_t)cep; - spin_unlock(&cep->lock); - return 0; -scif_accept_error_map: - scif_anon_inode_fput(cep); -scif_accept_error_anon_inode: - scif_teardown_ep(cep); -scif_accept_error_qpalloc: - kfree(cep); -scif_accept_error_epalloc: - msg.uop = SCIF_CNCT_REJ; - msg.dst.node = conreq->msg.src.node; - msg.dst.port = conreq->msg.src.port; - msg.payload[0] = conreq->msg.payload[0]; - msg.payload[1] = conreq->msg.payload[1]; - scif_nodeqp_send(&scif_dev[conreq->msg.src.node], &msg); - kfree(conreq); - return err; -} -EXPORT_SYMBOL_GPL(scif_accept); - -/* - * scif_msg_param_check: - * @epd: The end point returned from scif_open() - * @len: Length to receive - * @flags: blocking or non blocking - * - * Validate parameters for messaging APIs scif_send(..)/scif_recv(..). - */ -static inline int scif_msg_param_check(scif_epd_t epd, int len, int flags) -{ - int ret = -EINVAL; - - if (len < 0) - goto err_ret; - if (flags && (!(flags & SCIF_RECV_BLOCK))) - goto err_ret; - ret = 0; -err_ret: - return ret; -} - -static int _scif_send(scif_epd_t epd, void *msg, int len, int flags) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - struct scifmsg notif_msg; - int curr_xfer_len = 0, sent_len = 0, write_count; - int ret = 0; - struct scif_qp *qp = ep->qp_info.qp; - - if (flags & SCIF_SEND_BLOCK) - might_sleep(); - - spin_lock(&ep->lock); - while (sent_len != len && SCIFEP_CONNECTED == ep->state) { - write_count = scif_rb_space(&qp->outbound_q); - if (write_count) { - /* Best effort to send as much data as possible */ - curr_xfer_len = min(len - sent_len, write_count); - ret = scif_rb_write(&qp->outbound_q, msg, - curr_xfer_len); - if (ret < 0) - break; - /* Success. Update write pointer */ - scif_rb_commit(&qp->outbound_q); - /* - * Send a notification to the peer about the - * produced data message. - */ - notif_msg.src = ep->port; - notif_msg.uop = SCIF_CLIENT_SENT; - notif_msg.payload[0] = ep->remote_ep; - ret = _scif_nodeqp_send(ep->remote_dev, ¬if_msg); - if (ret) - break; - sent_len += curr_xfer_len; - msg = msg + curr_xfer_len; - continue; - } - curr_xfer_len = min(len - sent_len, SCIF_ENDPT_QP_SIZE - 1); - /* Not enough RB space. return for the Non Blocking case */ - if (!(flags & SCIF_SEND_BLOCK)) - break; - - spin_unlock(&ep->lock); - /* Wait for a SCIF_CLIENT_RCVD message in the Blocking case */ - ret = - wait_event_interruptible(ep->sendwq, - (SCIFEP_CONNECTED != ep->state) || - (scif_rb_space(&qp->outbound_q) >= - curr_xfer_len)); - spin_lock(&ep->lock); - if (ret) - break; - } - if (sent_len) - ret = sent_len; - else if (!ret && SCIFEP_CONNECTED != ep->state) - ret = SCIFEP_DISCONNECTED == ep->state ? - -ECONNRESET : -ENOTCONN; - spin_unlock(&ep->lock); - return ret; -} - -static int _scif_recv(scif_epd_t epd, void *msg, int len, int flags) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - struct scifmsg notif_msg; - int curr_recv_len = 0, remaining_len = len, read_count; - int ret = 0; - struct scif_qp *qp = ep->qp_info.qp; - - if (flags & SCIF_RECV_BLOCK) - might_sleep(); - spin_lock(&ep->lock); - while (remaining_len && (SCIFEP_CONNECTED == ep->state || - SCIFEP_DISCONNECTED == ep->state)) { - read_count = scif_rb_count(&qp->inbound_q, remaining_len); - if (read_count) { - /* - * Best effort to recv as much data as there - * are bytes to read in the RB particularly - * important for the Non Blocking case. - */ - curr_recv_len = min(remaining_len, read_count); - scif_rb_get_next(&qp->inbound_q, msg, curr_recv_len); - if (ep->state == SCIFEP_CONNECTED) { - /* - * Update the read pointer only if the endpoint - * is still connected else the read pointer - * might no longer exist since the peer has - * freed resources! - */ - scif_rb_update_read_ptr(&qp->inbound_q); - /* - * Send a notification to the peer about the - * consumed data message only if the EP is in - * SCIFEP_CONNECTED state. - */ - notif_msg.src = ep->port; - notif_msg.uop = SCIF_CLIENT_RCVD; - notif_msg.payload[0] = ep->remote_ep; - ret = _scif_nodeqp_send(ep->remote_dev, - ¬if_msg); - if (ret) - break; - } - remaining_len -= curr_recv_len; - msg = msg + curr_recv_len; - continue; - } - /* - * Bail out now if the EP is in SCIFEP_DISCONNECTED state else - * we will keep looping forever. - */ - if (ep->state == SCIFEP_DISCONNECTED) - break; - /* - * Return in the Non Blocking case if there is no data - * to read in this iteration. - */ - if (!(flags & SCIF_RECV_BLOCK)) - break; - curr_recv_len = min(remaining_len, SCIF_ENDPT_QP_SIZE - 1); - spin_unlock(&ep->lock); - /* - * Wait for a SCIF_CLIENT_SEND message in the blocking case - * or until other side disconnects. - */ - ret = - wait_event_interruptible(ep->recvwq, - SCIFEP_CONNECTED != ep->state || - scif_rb_count(&qp->inbound_q, - curr_recv_len) - >= curr_recv_len); - spin_lock(&ep->lock); - if (ret) - break; - } - if (len - remaining_len) - ret = len - remaining_len; - else if (!ret && ep->state != SCIFEP_CONNECTED) - ret = ep->state == SCIFEP_DISCONNECTED ? - -ECONNRESET : -ENOTCONN; - spin_unlock(&ep->lock); - return ret; -} - -/** - * scif_user_send() - Send data to connection queue - * @epd: The end point returned from scif_open() - * @msg: Address to place data - * @len: Length to receive - * @flags: blocking or non blocking - * - * This function is called from the driver IOCTL entry point - * only and is a wrapper for _scif_send(). - */ -int scif_user_send(scif_epd_t epd, void __user *msg, int len, int flags) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - int err = 0; - int sent_len = 0; - char *tmp; - int loop_len; - int chunk_len = min(len, (1 << (MAX_ORDER + PAGE_SHIFT - 1))); - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI send (U): ep %p %s\n", ep, scif_ep_states[ep->state]); - if (!len) - return 0; - - err = scif_msg_param_check(epd, len, flags); - if (err) - goto send_err; - - tmp = kmalloc(chunk_len, GFP_KERNEL); - if (!tmp) { - err = -ENOMEM; - goto send_err; - } - /* - * Grabbing the lock before breaking up the transfer in - * multiple chunks is required to ensure that messages do - * not get fragmented and reordered. - */ - mutex_lock(&ep->sendlock); - while (sent_len != len) { - loop_len = len - sent_len; - loop_len = min(chunk_len, loop_len); - if (copy_from_user(tmp, msg, loop_len)) { - err = -EFAULT; - goto send_free_err; - } - err = _scif_send(epd, tmp, loop_len, flags); - if (err < 0) - goto send_free_err; - sent_len += err; - msg += err; - if (err != loop_len) - goto send_free_err; - } -send_free_err: - mutex_unlock(&ep->sendlock); - kfree(tmp); -send_err: - return err < 0 ? err : sent_len; -} - -/** - * scif_user_recv() - Receive data from connection queue - * @epd: The end point returned from scif_open() - * @msg: Address to place data - * @len: Length to receive - * @flags: blocking or non blocking - * - * This function is called from the driver IOCTL entry point - * only and is a wrapper for _scif_recv(). - */ -int scif_user_recv(scif_epd_t epd, void __user *msg, int len, int flags) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - int err = 0; - int recv_len = 0; - char *tmp; - int loop_len; - int chunk_len = min(len, (1 << (MAX_ORDER + PAGE_SHIFT - 1))); - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI recv (U): ep %p %s\n", ep, scif_ep_states[ep->state]); - if (!len) - return 0; - - err = scif_msg_param_check(epd, len, flags); - if (err) - goto recv_err; - - tmp = kmalloc(chunk_len, GFP_KERNEL); - if (!tmp) { - err = -ENOMEM; - goto recv_err; - } - /* - * Grabbing the lock before breaking up the transfer in - * multiple chunks is required to ensure that messages do - * not get fragmented and reordered. - */ - mutex_lock(&ep->recvlock); - while (recv_len != len) { - loop_len = len - recv_len; - loop_len = min(chunk_len, loop_len); - err = _scif_recv(epd, tmp, loop_len, flags); - if (err < 0) - goto recv_free_err; - if (copy_to_user(msg, tmp, err)) { - err = -EFAULT; - goto recv_free_err; - } - recv_len += err; - msg += err; - if (err != loop_len) - goto recv_free_err; - } -recv_free_err: - mutex_unlock(&ep->recvlock); - kfree(tmp); -recv_err: - return err < 0 ? err : recv_len; -} - -/** - * scif_send() - Send data to connection queue - * @epd: The end point returned from scif_open() - * @msg: Address to place data - * @len: Length to receive - * @flags: blocking or non blocking - * - * This function is called from the kernel mode only and is - * a wrapper for _scif_send(). - */ -int scif_send(scif_epd_t epd, void *msg, int len, int flags) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - int ret; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI send (K): ep %p %s\n", ep, scif_ep_states[ep->state]); - if (!len) - return 0; - - ret = scif_msg_param_check(epd, len, flags); - if (ret) - return ret; - if (!ep->remote_dev) - return -ENOTCONN; - /* - * Grab the mutex lock in the blocking case only - * to ensure messages do not get fragmented/reordered. - * The non blocking mode is protected using spin locks - * in _scif_send(). - */ - if (flags & SCIF_SEND_BLOCK) - mutex_lock(&ep->sendlock); - - ret = _scif_send(epd, msg, len, flags); - - if (flags & SCIF_SEND_BLOCK) - mutex_unlock(&ep->sendlock); - return ret; -} -EXPORT_SYMBOL_GPL(scif_send); - -/** - * scif_recv() - Receive data from connection queue - * @epd: The end point returned from scif_open() - * @msg: Address to place data - * @len: Length to receive - * @flags: blocking or non blocking - * - * This function is called from the kernel mode only and is - * a wrapper for _scif_recv(). - */ -int scif_recv(scif_epd_t epd, void *msg, int len, int flags) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - int ret; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI recv (K): ep %p %s\n", ep, scif_ep_states[ep->state]); - if (!len) - return 0; - - ret = scif_msg_param_check(epd, len, flags); - if (ret) - return ret; - /* - * Grab the mutex lock in the blocking case only - * to ensure messages do not get fragmented/reordered. - * The non blocking mode is protected using spin locks - * in _scif_send(). - */ - if (flags & SCIF_RECV_BLOCK) - mutex_lock(&ep->recvlock); - - ret = _scif_recv(epd, msg, len, flags); - - if (flags & SCIF_RECV_BLOCK) - mutex_unlock(&ep->recvlock); - - return ret; -} -EXPORT_SYMBOL_GPL(scif_recv); - -static inline void _scif_poll_wait(struct file *f, wait_queue_head_t *wq, - poll_table *p, struct scif_endpt *ep) -{ - /* - * Because poll_wait makes a GFP_KERNEL allocation, give up the lock - * and regrab it afterwards. Because the endpoint state might have - * changed while the lock was given up, the state must be checked - * again after re-acquiring the lock. The code in __scif_pollfd(..) - * does this. - */ - spin_unlock(&ep->lock); - poll_wait(f, wq, p); - spin_lock(&ep->lock); -} - -__poll_t -__scif_pollfd(struct file *f, poll_table *wait, struct scif_endpt *ep) -{ - __poll_t mask = 0; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI pollfd: ep %p %s\n", ep, scif_ep_states[ep->state]); - - spin_lock(&ep->lock); - - /* Endpoint is waiting for a non-blocking connect to complete */ - if (ep->conn_async_state == ASYNC_CONN_INPROGRESS) { - _scif_poll_wait(f, &ep->conn_pend_wq, wait, ep); - if (ep->conn_async_state == ASYNC_CONN_INPROGRESS) { - if (ep->state == SCIFEP_CONNECTED || - ep->state == SCIFEP_DISCONNECTED || - ep->conn_err) - mask |= EPOLLOUT; - goto exit; - } - } - - /* Endpoint is listening for incoming connection requests */ - if (ep->state == SCIFEP_LISTENING) { - _scif_poll_wait(f, &ep->conwq, wait, ep); - if (ep->state == SCIFEP_LISTENING) { - if (ep->conreqcnt) - mask |= EPOLLIN; - goto exit; - } - } - - /* Endpoint is connected or disconnected */ - if (ep->state == SCIFEP_CONNECTED || ep->state == SCIFEP_DISCONNECTED) { - if (poll_requested_events(wait) & EPOLLIN) - _scif_poll_wait(f, &ep->recvwq, wait, ep); - if (poll_requested_events(wait) & EPOLLOUT) - _scif_poll_wait(f, &ep->sendwq, wait, ep); - if (ep->state == SCIFEP_CONNECTED || - ep->state == SCIFEP_DISCONNECTED) { - /* Data can be read without blocking */ - if (scif_rb_count(&ep->qp_info.qp->inbound_q, 1)) - mask |= EPOLLIN; - /* Data can be written without blocking */ - if (scif_rb_space(&ep->qp_info.qp->outbound_q)) - mask |= EPOLLOUT; - /* Return EPOLLHUP if endpoint is disconnected */ - if (ep->state == SCIFEP_DISCONNECTED) - mask |= EPOLLHUP; - goto exit; - } - } - - /* Return EPOLLERR if the endpoint is in none of the above states */ - mask |= EPOLLERR; -exit: - spin_unlock(&ep->lock); - return mask; -} - -/** - * scif_poll() - Kernel mode SCIF poll - * @ufds: Array of scif_pollepd structures containing the end points - * and events to poll on - * @nfds: Size of the ufds array - * @timeout_msecs: Timeout in msecs, -ve implies infinite timeout - * - * The code flow in this function is based on do_poll(..) in select.c - * - * Returns the number of endpoints which have pending events or 0 in - * the event of a timeout. If a signal is used for wake up, -EINTR is - * returned. - */ -int -scif_poll(struct scif_pollepd *ufds, unsigned int nfds, long timeout_msecs) -{ - struct poll_wqueues table; - poll_table *pt; - int i, count = 0, timed_out = timeout_msecs == 0; - __poll_t mask; - u64 timeout = timeout_msecs < 0 ? MAX_SCHEDULE_TIMEOUT - : msecs_to_jiffies(timeout_msecs); - - poll_initwait(&table); - pt = &table.pt; - while (1) { - for (i = 0; i < nfds; i++) { - pt->_key = ufds[i].events | EPOLLERR | EPOLLHUP; - mask = __scif_pollfd(ufds[i].epd->anon, - pt, ufds[i].epd); - mask &= ufds[i].events | EPOLLERR | EPOLLHUP; - if (mask) { - count++; - pt->_qproc = NULL; - } - ufds[i].revents = mask; - } - pt->_qproc = NULL; - if (!count) { - count = table.error; - if (signal_pending(current)) - count = -EINTR; - } - if (count || timed_out) - break; - - if (!schedule_timeout_interruptible(timeout)) - timed_out = 1; - } - poll_freewait(&table); - return count; -} -EXPORT_SYMBOL_GPL(scif_poll); - -int scif_get_node_ids(u16 *nodes, int len, u16 *self) -{ - int online = 0; - int offset = 0; - int node; - - if (!scif_is_mgmt_node()) - scif_get_node_info(); - - *self = scif_info.nodeid; - mutex_lock(&scif_info.conflock); - len = min_t(int, len, scif_info.total); - for (node = 0; node <= scif_info.maxid; node++) { - if (_scifdev_alive(&scif_dev[node])) { - online++; - if (offset < len) - nodes[offset++] = node; - } - } - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI get_node_ids total %d online %d filled in %d nodes\n", - scif_info.total, online, offset); - mutex_unlock(&scif_info.conflock); - - return online; -} -EXPORT_SYMBOL_GPL(scif_get_node_ids); - -static int scif_add_client_dev(struct device *dev, struct subsys_interface *si) -{ - struct scif_client *client = - container_of(si, struct scif_client, si); - struct scif_peer_dev *spdev = - container_of(dev, struct scif_peer_dev, dev); - - if (client->probe) - client->probe(spdev); - return 0; -} - -static void scif_remove_client_dev(struct device *dev, - struct subsys_interface *si) -{ - struct scif_client *client = - container_of(si, struct scif_client, si); - struct scif_peer_dev *spdev = - container_of(dev, struct scif_peer_dev, dev); - - if (client->remove) - client->remove(spdev); -} - -void scif_client_unregister(struct scif_client *client) -{ - subsys_interface_unregister(&client->si); -} -EXPORT_SYMBOL_GPL(scif_client_unregister); - -int scif_client_register(struct scif_client *client) -{ - struct subsys_interface *si = &client->si; - - si->name = client->name; - si->subsys = &scif_peer_bus; - si->add_dev = scif_add_client_dev; - si->remove_dev = scif_remove_client_dev; - - return subsys_interface_register(&client->si); -} -EXPORT_SYMBOL_GPL(scif_client_register); diff --git a/drivers/misc/mic/scif/scif_debugfs.c b/drivers/misc/mic/scif/scif_debugfs.c deleted file mode 100644 index 8fe38e7ca6e6..000000000000 --- a/drivers/misc/mic/scif/scif_debugfs.c +++ /dev/null @@ -1,116 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel SCIF driver. - */ -#include -#include - -#include "../common/mic_dev.h" -#include "scif_main.h" - -/* Debugfs parent dir */ -static struct dentry *scif_dbg; - -static int scif_dev_show(struct seq_file *s, void *unused) -{ - int node; - - seq_printf(s, "Total Nodes %d Self Node Id %d Maxid %d\n", - scif_info.total, scif_info.nodeid, - scif_info.maxid); - - if (!scif_dev) - return 0; - - seq_printf(s, "%-16s\t%-16s\n", "node_id", "state"); - - for (node = 0; node <= scif_info.maxid; node++) - seq_printf(s, "%-16d\t%-16s\n", scif_dev[node].node, - _scifdev_alive(&scif_dev[node]) ? - "Running" : "Offline"); - return 0; -} - -DEFINE_SHOW_ATTRIBUTE(scif_dev); - -static void scif_display_window(struct scif_window *window, struct seq_file *s) -{ - int j; - struct scatterlist *sg; - scif_pinned_pages_t pin = window->pinned_pages; - - seq_printf(s, "window %p type %d temp %d offset 0x%llx ", - window, window->type, window->temp, window->offset); - seq_printf(s, "nr_pages 0x%llx nr_contig_chunks 0x%x prot %d ", - window->nr_pages, window->nr_contig_chunks, window->prot); - seq_printf(s, "ref_count %d magic 0x%llx peer_window 0x%llx ", - window->ref_count, window->magic, window->peer_window); - seq_printf(s, "unreg_state 0x%x va_for_temp 0x%lx\n", - window->unreg_state, window->va_for_temp); - - for (j = 0; j < window->nr_contig_chunks; j++) - seq_printf(s, "page[%d] dma_addr 0x%llx num_pages 0x%llx\n", j, - window->dma_addr[j], window->num_pages[j]); - - if (window->type == SCIF_WINDOW_SELF && pin) - for (j = 0; j < window->nr_pages; j++) - seq_printf(s, "page[%d] = pinned_pages %p address %p\n", - j, pin->pages[j], - page_address(pin->pages[j])); - - if (window->st) - for_each_sg(window->st->sgl, sg, window->st->nents, j) - seq_printf(s, "sg[%d] dma addr 0x%llx length 0x%x\n", - j, sg_dma_address(sg), sg_dma_len(sg)); -} - -static void scif_display_all_windows(struct list_head *head, struct seq_file *s) -{ - struct list_head *item; - struct scif_window *window; - - list_for_each(item, head) { - window = list_entry(item, struct scif_window, list); - scif_display_window(window, s); - } -} - -static int scif_rma_show(struct seq_file *s, void *unused) -{ - struct scif_endpt *ep; - struct list_head *pos; - - mutex_lock(&scif_info.connlock); - list_for_each(pos, &scif_info.connected) { - ep = list_entry(pos, struct scif_endpt, list); - seq_printf(s, "ep %p self windows\n", ep); - mutex_lock(&ep->rma_info.rma_lock); - scif_display_all_windows(&ep->rma_info.reg_list, s); - seq_printf(s, "ep %p remote windows\n", ep); - scif_display_all_windows(&ep->rma_info.remote_reg_list, s); - mutex_unlock(&ep->rma_info.rma_lock); - } - mutex_unlock(&scif_info.connlock); - return 0; -} - -DEFINE_SHOW_ATTRIBUTE(scif_rma); - -void __init scif_init_debugfs(void) -{ - scif_dbg = debugfs_create_dir(KBUILD_MODNAME, NULL); - - debugfs_create_file("scif_dev", 0444, scif_dbg, NULL, &scif_dev_fops); - debugfs_create_file("scif_rma", 0444, scif_dbg, NULL, &scif_rma_fops); - debugfs_create_u8("en_msg_log", 0666, scif_dbg, &scif_info.en_msg_log); - debugfs_create_u8("p2p_enable", 0666, scif_dbg, &scif_info.p2p_enable); -} - -void scif_exit_debugfs(void) -{ - debugfs_remove_recursive(scif_dbg); -} diff --git a/drivers/misc/mic/scif/scif_dma.c b/drivers/misc/mic/scif/scif_dma.c deleted file mode 100644 index 401b98e5ad79..000000000000 --- a/drivers/misc/mic/scif/scif_dma.c +++ /dev/null @@ -1,1940 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2015 Intel Corporation. - * - * Intel SCIF driver. - */ -#include "scif_main.h" -#include "scif_map.h" - -/* - * struct scif_dma_comp_cb - SCIF DMA completion callback - * - * @dma_completion_func: DMA completion callback - * @cb_cookie: DMA completion callback cookie - * @temp_buf: Temporary buffer - * @temp_buf_to_free: Temporary buffer to be freed - * @is_cache: Is a kmem_cache allocated buffer - * @dst_offset: Destination registration offset - * @dst_window: Destination registration window - * @len: Length of the temp buffer - * @temp_phys: DMA address of the temp buffer - * @sdev: The SCIF device - * @header_padding: padding for cache line alignment - */ -struct scif_dma_comp_cb { - void (*dma_completion_func)(void *cookie); - void *cb_cookie; - u8 *temp_buf; - u8 *temp_buf_to_free; - bool is_cache; - s64 dst_offset; - struct scif_window *dst_window; - size_t len; - dma_addr_t temp_phys; - struct scif_dev *sdev; - int header_padding; -}; - -/** - * struct scif_copy_work - Work for DMA copy - * - * @src_offset: Starting source offset - * @dst_offset: Starting destination offset - * @src_window: Starting src registered window - * @dst_window: Starting dst registered window - * @loopback: true if this is a loopback DMA transfer - * @len: Length of the transfer - * @comp_cb: DMA copy completion callback - * @remote_dev: The remote SCIF peer device - * @fence_type: polling or interrupt based - * @ordered: is this a tail byte ordered DMA transfer - */ -struct scif_copy_work { - s64 src_offset; - s64 dst_offset; - struct scif_window *src_window; - struct scif_window *dst_window; - int loopback; - size_t len; - struct scif_dma_comp_cb *comp_cb; - struct scif_dev *remote_dev; - int fence_type; - bool ordered; -}; - -/** - * scif_reserve_dma_chan: - * @ep: Endpoint Descriptor. - * - * This routine reserves a DMA channel for a particular - * endpoint. All DMA transfers for an endpoint are always - * programmed on the same DMA channel. - */ -int scif_reserve_dma_chan(struct scif_endpt *ep) -{ - int err = 0; - struct scif_dev *scifdev; - struct scif_hw_dev *sdev; - struct dma_chan *chan; - - /* Loopback DMAs are not supported on the management node */ - if (!scif_info.nodeid && scifdev_self(ep->remote_dev)) - return 0; - if (scif_info.nodeid) - scifdev = &scif_dev[0]; - else - scifdev = ep->remote_dev; - sdev = scifdev->sdev; - if (!sdev->num_dma_ch) - return -ENODEV; - chan = sdev->dma_ch[scifdev->dma_ch_idx]; - scifdev->dma_ch_idx = (scifdev->dma_ch_idx + 1) % sdev->num_dma_ch; - mutex_lock(&ep->rma_info.rma_lock); - ep->rma_info.dma_chan = chan; - mutex_unlock(&ep->rma_info.rma_lock); - return err; -} - -#ifdef CONFIG_MMU_NOTIFIER -/* - * scif_rma_destroy_tcw: - * - * This routine destroys temporary cached windows - */ -static -void __scif_rma_destroy_tcw(struct scif_mmu_notif *mmn, - u64 start, u64 len) -{ - struct list_head *item, *tmp; - struct scif_window *window; - u64 start_va, end_va; - u64 end = start + len; - - if (end <= start) - return; - - list_for_each_safe(item, tmp, &mmn->tc_reg_list) { - window = list_entry(item, struct scif_window, list); - if (!len) - break; - start_va = window->va_for_temp; - end_va = start_va + (window->nr_pages << PAGE_SHIFT); - if (start < start_va && end <= start_va) - break; - if (start >= end_va) - continue; - __scif_rma_destroy_tcw_helper(window); - } -} - -static void scif_rma_destroy_tcw(struct scif_mmu_notif *mmn, u64 start, u64 len) -{ - struct scif_endpt *ep = mmn->ep; - - spin_lock(&ep->rma_info.tc_lock); - __scif_rma_destroy_tcw(mmn, start, len); - spin_unlock(&ep->rma_info.tc_lock); -} - -static void scif_rma_destroy_tcw_ep(struct scif_endpt *ep) -{ - struct list_head *item, *tmp; - struct scif_mmu_notif *mmn; - - list_for_each_safe(item, tmp, &ep->rma_info.mmn_list) { - mmn = list_entry(item, struct scif_mmu_notif, list); - scif_rma_destroy_tcw(mmn, 0, ULONG_MAX); - } -} - -static void __scif_rma_destroy_tcw_ep(struct scif_endpt *ep) -{ - struct list_head *item, *tmp; - struct scif_mmu_notif *mmn; - - spin_lock(&ep->rma_info.tc_lock); - list_for_each_safe(item, tmp, &ep->rma_info.mmn_list) { - mmn = list_entry(item, struct scif_mmu_notif, list); - __scif_rma_destroy_tcw(mmn, 0, ULONG_MAX); - } - spin_unlock(&ep->rma_info.tc_lock); -} - -static bool scif_rma_tc_can_cache(struct scif_endpt *ep, size_t cur_bytes) -{ - if ((cur_bytes >> PAGE_SHIFT) > scif_info.rma_tc_limit) - return false; - if ((atomic_read(&ep->rma_info.tcw_total_pages) - + (cur_bytes >> PAGE_SHIFT)) > - scif_info.rma_tc_limit) { - dev_info(scif_info.mdev.this_device, - "%s %d total=%d, current=%zu reached max\n", - __func__, __LINE__, - atomic_read(&ep->rma_info.tcw_total_pages), - (1 + (cur_bytes >> PAGE_SHIFT))); - scif_rma_destroy_tcw_invalid(); - __scif_rma_destroy_tcw_ep(ep); - } - return true; -} - -static void scif_mmu_notifier_release(struct mmu_notifier *mn, - struct mm_struct *mm) -{ - struct scif_mmu_notif *mmn; - - mmn = container_of(mn, struct scif_mmu_notif, ep_mmu_notifier); - scif_rma_destroy_tcw(mmn, 0, ULONG_MAX); - schedule_work(&scif_info.misc_work); -} - -static int scif_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn, - const struct mmu_notifier_range *range) -{ - struct scif_mmu_notif *mmn; - - mmn = container_of(mn, struct scif_mmu_notif, ep_mmu_notifier); - scif_rma_destroy_tcw(mmn, range->start, range->end - range->start); - - return 0; -} - -static void scif_mmu_notifier_invalidate_range_end(struct mmu_notifier *mn, - const struct mmu_notifier_range *range) -{ - /* - * Nothing to do here, everything needed was done in - * invalidate_range_start. - */ -} - -static const struct mmu_notifier_ops scif_mmu_notifier_ops = { - .release = scif_mmu_notifier_release, - .clear_flush_young = NULL, - .invalidate_range_start = scif_mmu_notifier_invalidate_range_start, - .invalidate_range_end = scif_mmu_notifier_invalidate_range_end}; - -static void scif_ep_unregister_mmu_notifier(struct scif_endpt *ep) -{ - struct scif_endpt_rma_info *rma = &ep->rma_info; - struct scif_mmu_notif *mmn = NULL; - struct list_head *item, *tmp; - - mutex_lock(&ep->rma_info.mmn_lock); - list_for_each_safe(item, tmp, &rma->mmn_list) { - mmn = list_entry(item, struct scif_mmu_notif, list); - mmu_notifier_unregister(&mmn->ep_mmu_notifier, mmn->mm); - list_del(item); - kfree(mmn); - } - mutex_unlock(&ep->rma_info.mmn_lock); -} - -static void scif_init_mmu_notifier(struct scif_mmu_notif *mmn, - struct mm_struct *mm, struct scif_endpt *ep) -{ - mmn->ep = ep; - mmn->mm = mm; - mmn->ep_mmu_notifier.ops = &scif_mmu_notifier_ops; - INIT_LIST_HEAD(&mmn->list); - INIT_LIST_HEAD(&mmn->tc_reg_list); -} - -static struct scif_mmu_notif * -scif_find_mmu_notifier(struct mm_struct *mm, struct scif_endpt_rma_info *rma) -{ - struct scif_mmu_notif *mmn; - - list_for_each_entry(mmn, &rma->mmn_list, list) - if (mmn->mm == mm) - return mmn; - return NULL; -} - -static struct scif_mmu_notif * -scif_add_mmu_notifier(struct mm_struct *mm, struct scif_endpt *ep) -{ - struct scif_mmu_notif *mmn - = kzalloc(sizeof(*mmn), GFP_KERNEL); - - if (!mmn) - return ERR_PTR(-ENOMEM); - - scif_init_mmu_notifier(mmn, current->mm, ep); - if (mmu_notifier_register(&mmn->ep_mmu_notifier, current->mm)) { - kfree(mmn); - return ERR_PTR(-EBUSY); - } - list_add(&mmn->list, &ep->rma_info.mmn_list); - return mmn; -} - -/* - * Called from the misc thread to destroy temporary cached windows and - * unregister the MMU notifier for the SCIF endpoint. - */ -void scif_mmu_notif_handler(struct work_struct *work) -{ - struct list_head *pos, *tmpq; - struct scif_endpt *ep; -restart: - scif_rma_destroy_tcw_invalid(); - spin_lock(&scif_info.rmalock); - list_for_each_safe(pos, tmpq, &scif_info.mmu_notif_cleanup) { - ep = list_entry(pos, struct scif_endpt, mmu_list); - list_del(&ep->mmu_list); - spin_unlock(&scif_info.rmalock); - scif_rma_destroy_tcw_ep(ep); - scif_ep_unregister_mmu_notifier(ep); - goto restart; - } - spin_unlock(&scif_info.rmalock); -} - -static bool scif_is_set_reg_cache(int flags) -{ - return !!(flags & SCIF_RMA_USECACHE); -} -#else -static struct scif_mmu_notif * -scif_find_mmu_notifier(struct mm_struct *mm, - struct scif_endpt_rma_info *rma) -{ - return NULL; -} - -static struct scif_mmu_notif * -scif_add_mmu_notifier(struct mm_struct *mm, struct scif_endpt *ep) -{ - return NULL; -} - -void scif_mmu_notif_handler(struct work_struct *work) -{ -} - -static bool scif_is_set_reg_cache(int flags) -{ - return false; -} - -static bool scif_rma_tc_can_cache(struct scif_endpt *ep, size_t cur_bytes) -{ - return false; -} -#endif - -/** - * scif_register_temp: - * @epd: End Point Descriptor. - * @addr: virtual address to/from which to copy - * @len: length of range to copy - * @prot: read/write protection - * @out_offset: computed offset returned by reference. - * @out_window: allocated registered window returned by reference. - * - * Create a temporary registered window. The peer will not know about this - * window. This API is used for scif_vreadfrom()/scif_vwriteto() API's. - */ -static int -scif_register_temp(scif_epd_t epd, unsigned long addr, size_t len, int prot, - off_t *out_offset, struct scif_window **out_window) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - int err; - scif_pinned_pages_t pinned_pages; - size_t aligned_len; - - aligned_len = ALIGN(len, PAGE_SIZE); - - err = __scif_pin_pages((void *)(addr & PAGE_MASK), - aligned_len, &prot, 0, &pinned_pages); - if (err) - return err; - - pinned_pages->prot = prot; - - /* Compute the offset for this registration */ - err = scif_get_window_offset(ep, 0, 0, - aligned_len >> PAGE_SHIFT, - (s64 *)out_offset); - if (err) - goto error_unpin; - - /* Allocate and prepare self registration window */ - *out_window = scif_create_window(ep, aligned_len >> PAGE_SHIFT, - *out_offset, true); - if (!*out_window) { - scif_free_window_offset(ep, NULL, *out_offset); - err = -ENOMEM; - goto error_unpin; - } - - (*out_window)->pinned_pages = pinned_pages; - (*out_window)->nr_pages = pinned_pages->nr_pages; - (*out_window)->prot = pinned_pages->prot; - - (*out_window)->va_for_temp = addr & PAGE_MASK; - err = scif_map_window(ep->remote_dev, *out_window); - if (err) { - /* Something went wrong! Rollback */ - scif_destroy_window(ep, *out_window); - *out_window = NULL; - } else { - *out_offset |= (addr - (*out_window)->va_for_temp); - } - return err; -error_unpin: - if (err) - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, err); - scif_unpin_pages(pinned_pages); - return err; -} - -#define SCIF_DMA_TO (3 * HZ) - -/* - * scif_sync_dma - Program a DMA without an interrupt descriptor - * - * @dev - The address of the pointer to the device instance used - * for DMA registration. - * @chan - DMA channel to be used. - * @sync_wait: Wait for DMA to complete? - * - * Return 0 on success and -errno on error. - */ -static int scif_sync_dma(struct scif_hw_dev *sdev, struct dma_chan *chan, - bool sync_wait) -{ - int err = 0; - struct dma_async_tx_descriptor *tx = NULL; - enum dma_ctrl_flags flags = DMA_PREP_FENCE; - dma_cookie_t cookie; - struct dma_device *ddev; - - if (!chan) { - err = -EIO; - dev_err(&sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - return err; - } - ddev = chan->device; - - tx = ddev->device_prep_dma_memcpy(chan, 0, 0, 0, flags); - if (!tx) { - err = -ENOMEM; - dev_err(&sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - goto release; - } - cookie = tx->tx_submit(tx); - - if (dma_submit_error(cookie)) { - err = -ENOMEM; - dev_err(&sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - goto release; - } - if (!sync_wait) { - dma_async_issue_pending(chan); - } else { - if (dma_sync_wait(chan, cookie) == DMA_COMPLETE) { - err = 0; - } else { - err = -EIO; - dev_err(&sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - } - } -release: - return err; -} - -static void scif_dma_callback(void *arg) -{ - struct completion *done = (struct completion *)arg; - - complete(done); -} - -#define SCIF_DMA_SYNC_WAIT true -#define SCIF_DMA_POLL BIT(0) -#define SCIF_DMA_INTR BIT(1) - -/* - * scif_async_dma - Program a DMA with an interrupt descriptor - * - * @dev - The address of the pointer to the device instance used - * for DMA registration. - * @chan - DMA channel to be used. - * Return 0 on success and -errno on error. - */ -static int scif_async_dma(struct scif_hw_dev *sdev, struct dma_chan *chan) -{ - int err = 0; - struct dma_device *ddev; - struct dma_async_tx_descriptor *tx = NULL; - enum dma_ctrl_flags flags = DMA_PREP_INTERRUPT | DMA_PREP_FENCE; - DECLARE_COMPLETION_ONSTACK(done_wait); - dma_cookie_t cookie; - enum dma_status status; - - if (!chan) { - err = -EIO; - dev_err(&sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - return err; - } - ddev = chan->device; - - tx = ddev->device_prep_dma_memcpy(chan, 0, 0, 0, flags); - if (!tx) { - err = -ENOMEM; - dev_err(&sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - goto release; - } - reinit_completion(&done_wait); - tx->callback = scif_dma_callback; - tx->callback_param = &done_wait; - cookie = tx->tx_submit(tx); - - if (dma_submit_error(cookie)) { - err = -ENOMEM; - dev_err(&sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - goto release; - } - dma_async_issue_pending(chan); - - err = wait_for_completion_timeout(&done_wait, SCIF_DMA_TO); - if (!err) { - err = -EIO; - dev_err(&sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - goto release; - } - err = 0; - status = dma_async_is_tx_complete(chan, cookie, NULL, NULL); - if (status != DMA_COMPLETE) { - err = -EIO; - dev_err(&sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - goto release; - } -release: - return err; -} - -/* - * scif_drain_dma_poll - Drain all outstanding DMA operations for a particular - * DMA channel via polling. - * - * @sdev - The SCIF device - * @chan - DMA channel - * Return 0 on success and -errno on error. - */ -static int scif_drain_dma_poll(struct scif_hw_dev *sdev, struct dma_chan *chan) -{ - if (!chan) - return -EINVAL; - return scif_sync_dma(sdev, chan, SCIF_DMA_SYNC_WAIT); -} - -/* - * scif_drain_dma_intr - Drain all outstanding DMA operations for a particular - * DMA channel via interrupt based blocking wait. - * - * @sdev - The SCIF device - * @chan - DMA channel - * Return 0 on success and -errno on error. - */ -int scif_drain_dma_intr(struct scif_hw_dev *sdev, struct dma_chan *chan) -{ - if (!chan) - return -EINVAL; - return scif_async_dma(sdev, chan); -} - -/** - * scif_rma_destroy_windows: - * - * This routine destroys all windows queued for cleanup - */ -void scif_rma_destroy_windows(void) -{ - struct list_head *item, *tmp; - struct scif_window *window; - struct scif_endpt *ep; - struct dma_chan *chan; - - might_sleep(); -restart: - spin_lock(&scif_info.rmalock); - list_for_each_safe(item, tmp, &scif_info.rma) { - window = list_entry(item, struct scif_window, - list); - ep = (struct scif_endpt *)window->ep; - chan = ep->rma_info.dma_chan; - - list_del_init(&window->list); - spin_unlock(&scif_info.rmalock); - if (!chan || !scifdev_alive(ep) || - !scif_drain_dma_intr(ep->remote_dev->sdev, - ep->rma_info.dma_chan)) - /* Remove window from global list */ - window->unreg_state = OP_COMPLETED; - else - dev_warn(&ep->remote_dev->sdev->dev, - "DMA engine hung?\n"); - if (window->unreg_state == OP_COMPLETED) { - if (window->type == SCIF_WINDOW_SELF) - scif_destroy_window(ep, window); - else - scif_destroy_remote_window(window); - atomic_dec(&ep->rma_info.tw_refcount); - } - goto restart; - } - spin_unlock(&scif_info.rmalock); -} - -/** - * scif_rma_destroy_tcw: - * - * This routine destroys temporary cached registered windows - * which have been queued for cleanup. - */ -void scif_rma_destroy_tcw_invalid(void) -{ - struct list_head *item, *tmp; - struct scif_window *window; - struct scif_endpt *ep; - struct dma_chan *chan; - - might_sleep(); -restart: - spin_lock(&scif_info.rmalock); - list_for_each_safe(item, tmp, &scif_info.rma_tc) { - window = list_entry(item, struct scif_window, list); - ep = (struct scif_endpt *)window->ep; - chan = ep->rma_info.dma_chan; - list_del_init(&window->list); - spin_unlock(&scif_info.rmalock); - mutex_lock(&ep->rma_info.rma_lock); - if (!chan || !scifdev_alive(ep) || - !scif_drain_dma_intr(ep->remote_dev->sdev, - ep->rma_info.dma_chan)) { - atomic_sub(window->nr_pages, - &ep->rma_info.tcw_total_pages); - scif_destroy_window(ep, window); - atomic_dec(&ep->rma_info.tcw_refcount); - } else { - dev_warn(&ep->remote_dev->sdev->dev, - "DMA engine hung?\n"); - } - mutex_unlock(&ep->rma_info.rma_lock); - goto restart; - } - spin_unlock(&scif_info.rmalock); -} - -static inline -void *_get_local_va(off_t off, struct scif_window *window, size_t len) -{ - int page_nr = (off - window->offset) >> PAGE_SHIFT; - off_t page_off = off & ~PAGE_MASK; - void *va = NULL; - - if (window->type == SCIF_WINDOW_SELF) { - struct page **pages = window->pinned_pages->pages; - - va = page_address(pages[page_nr]) + page_off; - } - return va; -} - -static inline -void *ioremap_remote(off_t off, struct scif_window *window, - size_t len, struct scif_dev *dev, - struct scif_window_iter *iter) -{ - dma_addr_t phys = scif_off_to_dma_addr(window, off, NULL, iter); - - /* - * If the DMA address is not card relative then we need the DMA - * addresses to be an offset into the bar. The aperture base was already - * added so subtract it here since scif_ioremap is going to add it again - */ - if (!scifdev_self(dev) && window->type == SCIF_WINDOW_PEER && - dev->sdev->aper && !dev->sdev->card_rel_da) - phys = phys - dev->sdev->aper->pa; - return scif_ioremap(phys, len, dev); -} - -static inline void -iounmap_remote(void *virt, size_t size, struct scif_copy_work *work) -{ - scif_iounmap(virt, size, work->remote_dev); -} - -/* - * Takes care of ordering issue caused by - * 1. Hardware: Only in the case of cpu copy from mgmt node to card - * because of WC memory. - * 2. Software: If memcpy reorders copy instructions for optimization. - * This could happen at both mgmt node and card. - */ -static inline void -scif_ordered_memcpy_toio(char *dst, const char *src, size_t count) -{ - if (!count) - return; - - memcpy_toio((void __iomem __force *)dst, src, --count); - /* Order the last byte with the previous stores */ - wmb(); - *(dst + count) = *(src + count); -} - -static inline void scif_unaligned_cpy_toio(char *dst, const char *src, - size_t count, bool ordered) -{ - if (ordered) - scif_ordered_memcpy_toio(dst, src, count); - else - memcpy_toio((void __iomem __force *)dst, src, count); -} - -static inline -void scif_ordered_memcpy_fromio(char *dst, const char *src, size_t count) -{ - if (!count) - return; - - memcpy_fromio(dst, (void __iomem __force *)src, --count); - /* Order the last byte with the previous loads */ - rmb(); - *(dst + count) = *(src + count); -} - -static inline void scif_unaligned_cpy_fromio(char *dst, const char *src, - size_t count, bool ordered) -{ - if (ordered) - scif_ordered_memcpy_fromio(dst, src, count); - else - memcpy_fromio(dst, (void __iomem __force *)src, count); -} - -#define SCIF_RMA_ERROR_CODE (~(dma_addr_t)0x0) - -/* - * scif_off_to_dma_addr: - * Obtain the dma_addr given the window and the offset. - * @window: Registered window. - * @off: Window offset. - * @nr_bytes: Return the number of contiguous bytes till next DMA addr index. - * @index: Return the index of the dma_addr array found. - * @start_off: start offset of index of the dma addr array found. - * The nr_bytes provides the callee an estimate of the maximum possible - * DMA xfer possible while the index/start_off provide faster lookups - * for the next iteration. - */ -dma_addr_t scif_off_to_dma_addr(struct scif_window *window, s64 off, - size_t *nr_bytes, struct scif_window_iter *iter) -{ - int i, page_nr; - s64 start, end; - off_t page_off; - - if (window->nr_pages == window->nr_contig_chunks) { - page_nr = (off - window->offset) >> PAGE_SHIFT; - page_off = off & ~PAGE_MASK; - - if (nr_bytes) - *nr_bytes = PAGE_SIZE - page_off; - return window->dma_addr[page_nr] | page_off; - } - if (iter) { - i = iter->index; - start = iter->offset; - } else { - i = 0; - start = window->offset; - } - for (; i < window->nr_contig_chunks; i++) { - end = start + (window->num_pages[i] << PAGE_SHIFT); - if (off >= start && off < end) { - if (iter) { - iter->index = i; - iter->offset = start; - } - if (nr_bytes) - *nr_bytes = end - off; - return (window->dma_addr[i] + (off - start)); - } - start += (window->num_pages[i] << PAGE_SHIFT); - } - dev_err(scif_info.mdev.this_device, - "%s %d BUG. Addr not found? window %p off 0x%llx\n", - __func__, __LINE__, window, off); - return SCIF_RMA_ERROR_CODE; -} - -/* - * Copy between rma window and temporary buffer - */ -static void scif_rma_local_cpu_copy(s64 offset, struct scif_window *window, - u8 *temp, size_t rem_len, bool to_temp) -{ - void *window_virt; - size_t loop_len; - int offset_in_page; - s64 end_offset; - - offset_in_page = offset & ~PAGE_MASK; - loop_len = PAGE_SIZE - offset_in_page; - - if (rem_len < loop_len) - loop_len = rem_len; - - window_virt = _get_local_va(offset, window, loop_len); - if (!window_virt) - return; - if (to_temp) - memcpy(temp, window_virt, loop_len); - else - memcpy(window_virt, temp, loop_len); - - offset += loop_len; - temp += loop_len; - rem_len -= loop_len; - - end_offset = window->offset + - (window->nr_pages << PAGE_SHIFT); - while (rem_len) { - if (offset == end_offset) { - window = list_next_entry(window, list); - end_offset = window->offset + - (window->nr_pages << PAGE_SHIFT); - } - loop_len = min(PAGE_SIZE, rem_len); - window_virt = _get_local_va(offset, window, loop_len); - if (!window_virt) - return; - if (to_temp) - memcpy(temp, window_virt, loop_len); - else - memcpy(window_virt, temp, loop_len); - offset += loop_len; - temp += loop_len; - rem_len -= loop_len; - } -} - -/** - * scif_rma_completion_cb: - * @data: RMA cookie - * - * RMA interrupt completion callback. - */ -static void scif_rma_completion_cb(void *data) -{ - struct scif_dma_comp_cb *comp_cb = data; - - /* Free DMA Completion CB. */ - if (comp_cb->dst_window) - scif_rma_local_cpu_copy(comp_cb->dst_offset, - comp_cb->dst_window, - comp_cb->temp_buf + - comp_cb->header_padding, - comp_cb->len, false); - scif_unmap_single(comp_cb->temp_phys, comp_cb->sdev, - SCIF_KMEM_UNALIGNED_BUF_SIZE); - if (comp_cb->is_cache) - kmem_cache_free(unaligned_cache, - comp_cb->temp_buf_to_free); - else - kfree(comp_cb->temp_buf_to_free); -} - -/* Copies between temporary buffer and offsets provided in work */ -static int -scif_rma_list_dma_copy_unaligned(struct scif_copy_work *work, - u8 *temp, struct dma_chan *chan, - bool src_local) -{ - struct scif_dma_comp_cb *comp_cb = work->comp_cb; - dma_addr_t window_dma_addr, temp_dma_addr; - dma_addr_t temp_phys = comp_cb->temp_phys; - size_t loop_len, nr_contig_bytes = 0, remaining_len = work->len; - int offset_in_ca, ret = 0; - s64 end_offset, offset; - struct scif_window *window; - void *window_virt_addr; - size_t tail_len; - struct dma_async_tx_descriptor *tx; - struct dma_device *dev = chan->device; - dma_cookie_t cookie; - - if (src_local) { - offset = work->dst_offset; - window = work->dst_window; - } else { - offset = work->src_offset; - window = work->src_window; - } - - offset_in_ca = offset & (L1_CACHE_BYTES - 1); - if (offset_in_ca) { - loop_len = L1_CACHE_BYTES - offset_in_ca; - loop_len = min(loop_len, remaining_len); - window_virt_addr = ioremap_remote(offset, window, - loop_len, - work->remote_dev, - NULL); - if (!window_virt_addr) - return -ENOMEM; - if (src_local) - scif_unaligned_cpy_toio(window_virt_addr, temp, - loop_len, - work->ordered && - !(remaining_len - loop_len)); - else - scif_unaligned_cpy_fromio(temp, window_virt_addr, - loop_len, work->ordered && - !(remaining_len - loop_len)); - iounmap_remote(window_virt_addr, loop_len, work); - - offset += loop_len; - temp += loop_len; - temp_phys += loop_len; - remaining_len -= loop_len; - } - - offset_in_ca = offset & ~PAGE_MASK; - end_offset = window->offset + - (window->nr_pages << PAGE_SHIFT); - - tail_len = remaining_len & (L1_CACHE_BYTES - 1); - remaining_len -= tail_len; - while (remaining_len) { - if (offset == end_offset) { - window = list_next_entry(window, list); - end_offset = window->offset + - (window->nr_pages << PAGE_SHIFT); - } - if (scif_is_mgmt_node()) - temp_dma_addr = temp_phys; - else - /* Fix if we ever enable IOMMU on the card */ - temp_dma_addr = (dma_addr_t)virt_to_phys(temp); - window_dma_addr = scif_off_to_dma_addr(window, offset, - &nr_contig_bytes, - NULL); - loop_len = min(nr_contig_bytes, remaining_len); - if (src_local) { - if (work->ordered && !tail_len && - !(remaining_len - loop_len) && - loop_len != L1_CACHE_BYTES) { - /* - * Break up the last chunk of the transfer into - * two steps. if there is no tail to guarantee - * DMA ordering. SCIF_DMA_POLLING inserts - * a status update descriptor in step 1 which - * acts as a double sided synchronization fence - * for the DMA engine to ensure that the last - * cache line in step 2 is updated last. - */ - /* Step 1) DMA: Body Length - L1_CACHE_BYTES. */ - tx = - dev->device_prep_dma_memcpy(chan, - window_dma_addr, - temp_dma_addr, - loop_len - - L1_CACHE_BYTES, - DMA_PREP_FENCE); - if (!tx) { - ret = -ENOMEM; - goto err; - } - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - ret = -ENOMEM; - goto err; - } - dma_async_issue_pending(chan); - offset += (loop_len - L1_CACHE_BYTES); - temp_dma_addr += (loop_len - L1_CACHE_BYTES); - window_dma_addr += (loop_len - L1_CACHE_BYTES); - remaining_len -= (loop_len - L1_CACHE_BYTES); - loop_len = remaining_len; - - /* Step 2) DMA: L1_CACHE_BYTES */ - tx = - dev->device_prep_dma_memcpy(chan, - window_dma_addr, - temp_dma_addr, - loop_len, 0); - if (!tx) { - ret = -ENOMEM; - goto err; - } - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - ret = -ENOMEM; - goto err; - } - dma_async_issue_pending(chan); - } else { - tx = - dev->device_prep_dma_memcpy(chan, - window_dma_addr, - temp_dma_addr, - loop_len, 0); - if (!tx) { - ret = -ENOMEM; - goto err; - } - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - ret = -ENOMEM; - goto err; - } - dma_async_issue_pending(chan); - } - } else { - tx = dev->device_prep_dma_memcpy(chan, temp_dma_addr, - window_dma_addr, loop_len, 0); - if (!tx) { - ret = -ENOMEM; - goto err; - } - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - ret = -ENOMEM; - goto err; - } - dma_async_issue_pending(chan); - } - offset += loop_len; - temp += loop_len; - temp_phys += loop_len; - remaining_len -= loop_len; - offset_in_ca = 0; - } - if (tail_len) { - if (offset == end_offset) { - window = list_next_entry(window, list); - end_offset = window->offset + - (window->nr_pages << PAGE_SHIFT); - } - window_virt_addr = ioremap_remote(offset, window, tail_len, - work->remote_dev, - NULL); - if (!window_virt_addr) - return -ENOMEM; - /* - * The CPU copy for the tail bytes must be initiated only once - * previous DMA transfers for this endpoint have completed - * to guarantee ordering. - */ - if (work->ordered) { - struct scif_dev *rdev = work->remote_dev; - - ret = scif_drain_dma_intr(rdev->sdev, chan); - if (ret) - return ret; - } - if (src_local) - scif_unaligned_cpy_toio(window_virt_addr, temp, - tail_len, work->ordered); - else - scif_unaligned_cpy_fromio(temp, window_virt_addr, - tail_len, work->ordered); - iounmap_remote(window_virt_addr, tail_len, work); - } - tx = dev->device_prep_dma_memcpy(chan, 0, 0, 0, DMA_PREP_INTERRUPT); - if (!tx) { - ret = -ENOMEM; - return ret; - } - tx->callback = &scif_rma_completion_cb; - tx->callback_param = comp_cb; - cookie = tx->tx_submit(tx); - - if (dma_submit_error(cookie)) { - ret = -ENOMEM; - return ret; - } - dma_async_issue_pending(chan); - return 0; -err: - dev_err(scif_info.mdev.this_device, - "%s %d Desc Prog Failed ret %d\n", - __func__, __LINE__, ret); - return ret; -} - -/* - * _scif_rma_list_dma_copy_aligned: - * - * Traverse all the windows and perform DMA copy. - */ -static int _scif_rma_list_dma_copy_aligned(struct scif_copy_work *work, - struct dma_chan *chan) -{ - dma_addr_t src_dma_addr, dst_dma_addr; - size_t loop_len, remaining_len, src_contig_bytes = 0; - size_t dst_contig_bytes = 0; - struct scif_window_iter src_win_iter; - struct scif_window_iter dst_win_iter; - s64 end_src_offset, end_dst_offset; - struct scif_window *src_window = work->src_window; - struct scif_window *dst_window = work->dst_window; - s64 src_offset = work->src_offset, dst_offset = work->dst_offset; - int ret = 0; - struct dma_async_tx_descriptor *tx; - struct dma_device *dev = chan->device; - dma_cookie_t cookie; - - remaining_len = work->len; - - scif_init_window_iter(src_window, &src_win_iter); - scif_init_window_iter(dst_window, &dst_win_iter); - end_src_offset = src_window->offset + - (src_window->nr_pages << PAGE_SHIFT); - end_dst_offset = dst_window->offset + - (dst_window->nr_pages << PAGE_SHIFT); - while (remaining_len) { - if (src_offset == end_src_offset) { - src_window = list_next_entry(src_window, list); - end_src_offset = src_window->offset + - (src_window->nr_pages << PAGE_SHIFT); - scif_init_window_iter(src_window, &src_win_iter); - } - if (dst_offset == end_dst_offset) { - dst_window = list_next_entry(dst_window, list); - end_dst_offset = dst_window->offset + - (dst_window->nr_pages << PAGE_SHIFT); - scif_init_window_iter(dst_window, &dst_win_iter); - } - - /* compute dma addresses for transfer */ - src_dma_addr = scif_off_to_dma_addr(src_window, src_offset, - &src_contig_bytes, - &src_win_iter); - dst_dma_addr = scif_off_to_dma_addr(dst_window, dst_offset, - &dst_contig_bytes, - &dst_win_iter); - loop_len = min(src_contig_bytes, dst_contig_bytes); - loop_len = min(loop_len, remaining_len); - if (work->ordered && !(remaining_len - loop_len)) { - /* - * Break up the last chunk of the transfer into two - * steps to ensure that the last byte in step 2 is - * updated last. - */ - /* Step 1) DMA: Body Length - 1 */ - tx = dev->device_prep_dma_memcpy(chan, dst_dma_addr, - src_dma_addr, - loop_len - 1, - DMA_PREP_FENCE); - if (!tx) { - ret = -ENOMEM; - goto err; - } - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - ret = -ENOMEM; - goto err; - } - src_offset += (loop_len - 1); - dst_offset += (loop_len - 1); - src_dma_addr += (loop_len - 1); - dst_dma_addr += (loop_len - 1); - remaining_len -= (loop_len - 1); - loop_len = remaining_len; - - /* Step 2) DMA: 1 BYTES */ - tx = dev->device_prep_dma_memcpy(chan, dst_dma_addr, - src_dma_addr, loop_len, 0); - if (!tx) { - ret = -ENOMEM; - goto err; - } - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - ret = -ENOMEM; - goto err; - } - dma_async_issue_pending(chan); - } else { - tx = dev->device_prep_dma_memcpy(chan, dst_dma_addr, - src_dma_addr, loop_len, 0); - if (!tx) { - ret = -ENOMEM; - goto err; - } - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - ret = -ENOMEM; - goto err; - } - } - src_offset += loop_len; - dst_offset += loop_len; - remaining_len -= loop_len; - } - return ret; -err: - dev_err(scif_info.mdev.this_device, - "%s %d Desc Prog Failed ret %d\n", - __func__, __LINE__, ret); - return ret; -} - -/* - * scif_rma_list_dma_copy_aligned: - * - * Traverse all the windows and perform DMA copy. - */ -static int scif_rma_list_dma_copy_aligned(struct scif_copy_work *work, - struct dma_chan *chan) -{ - dma_addr_t src_dma_addr, dst_dma_addr; - size_t loop_len, remaining_len, tail_len, src_contig_bytes = 0; - size_t dst_contig_bytes = 0; - int src_cache_off; - s64 end_src_offset, end_dst_offset; - struct scif_window_iter src_win_iter; - struct scif_window_iter dst_win_iter; - void *src_virt, *dst_virt; - struct scif_window *src_window = work->src_window; - struct scif_window *dst_window = work->dst_window; - s64 src_offset = work->src_offset, dst_offset = work->dst_offset; - int ret = 0; - struct dma_async_tx_descriptor *tx; - struct dma_device *dev = chan->device; - dma_cookie_t cookie; - - remaining_len = work->len; - scif_init_window_iter(src_window, &src_win_iter); - scif_init_window_iter(dst_window, &dst_win_iter); - - src_cache_off = src_offset & (L1_CACHE_BYTES - 1); - if (src_cache_off != 0) { - /* Head */ - loop_len = L1_CACHE_BYTES - src_cache_off; - loop_len = min(loop_len, remaining_len); - src_dma_addr = __scif_off_to_dma_addr(src_window, src_offset); - dst_dma_addr = __scif_off_to_dma_addr(dst_window, dst_offset); - if (src_window->type == SCIF_WINDOW_SELF) - src_virt = _get_local_va(src_offset, src_window, - loop_len); - else - src_virt = ioremap_remote(src_offset, src_window, - loop_len, - work->remote_dev, NULL); - if (!src_virt) - return -ENOMEM; - if (dst_window->type == SCIF_WINDOW_SELF) - dst_virt = _get_local_va(dst_offset, dst_window, - loop_len); - else - dst_virt = ioremap_remote(dst_offset, dst_window, - loop_len, - work->remote_dev, NULL); - if (!dst_virt) { - if (src_window->type != SCIF_WINDOW_SELF) - iounmap_remote(src_virt, loop_len, work); - return -ENOMEM; - } - if (src_window->type == SCIF_WINDOW_SELF) - scif_unaligned_cpy_toio(dst_virt, src_virt, loop_len, - remaining_len == loop_len ? - work->ordered : false); - else - scif_unaligned_cpy_fromio(dst_virt, src_virt, loop_len, - remaining_len == loop_len ? - work->ordered : false); - if (src_window->type != SCIF_WINDOW_SELF) - iounmap_remote(src_virt, loop_len, work); - if (dst_window->type != SCIF_WINDOW_SELF) - iounmap_remote(dst_virt, loop_len, work); - src_offset += loop_len; - dst_offset += loop_len; - remaining_len -= loop_len; - } - - end_src_offset = src_window->offset + - (src_window->nr_pages << PAGE_SHIFT); - end_dst_offset = dst_window->offset + - (dst_window->nr_pages << PAGE_SHIFT); - tail_len = remaining_len & (L1_CACHE_BYTES - 1); - remaining_len -= tail_len; - while (remaining_len) { - if (src_offset == end_src_offset) { - src_window = list_next_entry(src_window, list); - end_src_offset = src_window->offset + - (src_window->nr_pages << PAGE_SHIFT); - scif_init_window_iter(src_window, &src_win_iter); - } - if (dst_offset == end_dst_offset) { - dst_window = list_next_entry(dst_window, list); - end_dst_offset = dst_window->offset + - (dst_window->nr_pages << PAGE_SHIFT); - scif_init_window_iter(dst_window, &dst_win_iter); - } - - /* compute dma addresses for transfer */ - src_dma_addr = scif_off_to_dma_addr(src_window, src_offset, - &src_contig_bytes, - &src_win_iter); - dst_dma_addr = scif_off_to_dma_addr(dst_window, dst_offset, - &dst_contig_bytes, - &dst_win_iter); - loop_len = min(src_contig_bytes, dst_contig_bytes); - loop_len = min(loop_len, remaining_len); - if (work->ordered && !tail_len && - !(remaining_len - loop_len)) { - /* - * Break up the last chunk of the transfer into two - * steps. if there is no tail to gurantee DMA ordering. - * Passing SCIF_DMA_POLLING inserts a status update - * descriptor in step 1 which acts as a double sided - * synchronization fence for the DMA engine to ensure - * that the last cache line in step 2 is updated last. - */ - /* Step 1) DMA: Body Length - L1_CACHE_BYTES. */ - tx = dev->device_prep_dma_memcpy(chan, dst_dma_addr, - src_dma_addr, - loop_len - - L1_CACHE_BYTES, - DMA_PREP_FENCE); - if (!tx) { - ret = -ENOMEM; - goto err; - } - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - ret = -ENOMEM; - goto err; - } - dma_async_issue_pending(chan); - src_offset += (loop_len - L1_CACHE_BYTES); - dst_offset += (loop_len - L1_CACHE_BYTES); - src_dma_addr += (loop_len - L1_CACHE_BYTES); - dst_dma_addr += (loop_len - L1_CACHE_BYTES); - remaining_len -= (loop_len - L1_CACHE_BYTES); - loop_len = remaining_len; - - /* Step 2) DMA: L1_CACHE_BYTES */ - tx = dev->device_prep_dma_memcpy(chan, dst_dma_addr, - src_dma_addr, - loop_len, 0); - if (!tx) { - ret = -ENOMEM; - goto err; - } - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - ret = -ENOMEM; - goto err; - } - dma_async_issue_pending(chan); - } else { - tx = dev->device_prep_dma_memcpy(chan, dst_dma_addr, - src_dma_addr, - loop_len, 0); - if (!tx) { - ret = -ENOMEM; - goto err; - } - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - ret = -ENOMEM; - goto err; - } - dma_async_issue_pending(chan); - } - src_offset += loop_len; - dst_offset += loop_len; - remaining_len -= loop_len; - } - remaining_len = tail_len; - if (remaining_len) { - loop_len = remaining_len; - if (src_offset == end_src_offset) - src_window = list_next_entry(src_window, list); - if (dst_offset == end_dst_offset) - dst_window = list_next_entry(dst_window, list); - - src_dma_addr = __scif_off_to_dma_addr(src_window, src_offset); - dst_dma_addr = __scif_off_to_dma_addr(dst_window, dst_offset); - /* - * The CPU copy for the tail bytes must be initiated only once - * previous DMA transfers for this endpoint have completed to - * guarantee ordering. - */ - if (work->ordered) { - struct scif_dev *rdev = work->remote_dev; - - ret = scif_drain_dma_poll(rdev->sdev, chan); - if (ret) - return ret; - } - if (src_window->type == SCIF_WINDOW_SELF) - src_virt = _get_local_va(src_offset, src_window, - loop_len); - else - src_virt = ioremap_remote(src_offset, src_window, - loop_len, - work->remote_dev, NULL); - if (!src_virt) - return -ENOMEM; - - if (dst_window->type == SCIF_WINDOW_SELF) - dst_virt = _get_local_va(dst_offset, dst_window, - loop_len); - else - dst_virt = ioremap_remote(dst_offset, dst_window, - loop_len, - work->remote_dev, NULL); - if (!dst_virt) { - if (src_window->type != SCIF_WINDOW_SELF) - iounmap_remote(src_virt, loop_len, work); - return -ENOMEM; - } - - if (src_window->type == SCIF_WINDOW_SELF) - scif_unaligned_cpy_toio(dst_virt, src_virt, loop_len, - work->ordered); - else - scif_unaligned_cpy_fromio(dst_virt, src_virt, - loop_len, work->ordered); - if (src_window->type != SCIF_WINDOW_SELF) - iounmap_remote(src_virt, loop_len, work); - - if (dst_window->type != SCIF_WINDOW_SELF) - iounmap_remote(dst_virt, loop_len, work); - remaining_len -= loop_len; - } - return ret; -err: - dev_err(scif_info.mdev.this_device, - "%s %d Desc Prog Failed ret %d\n", - __func__, __LINE__, ret); - return ret; -} - -/* - * scif_rma_list_cpu_copy: - * - * Traverse all the windows and perform CPU copy. - */ -static int scif_rma_list_cpu_copy(struct scif_copy_work *work) -{ - void *src_virt, *dst_virt; - size_t loop_len, remaining_len; - int src_page_off, dst_page_off; - s64 src_offset = work->src_offset, dst_offset = work->dst_offset; - struct scif_window *src_window = work->src_window; - struct scif_window *dst_window = work->dst_window; - s64 end_src_offset, end_dst_offset; - int ret = 0; - struct scif_window_iter src_win_iter; - struct scif_window_iter dst_win_iter; - - remaining_len = work->len; - - scif_init_window_iter(src_window, &src_win_iter); - scif_init_window_iter(dst_window, &dst_win_iter); - while (remaining_len) { - src_page_off = src_offset & ~PAGE_MASK; - dst_page_off = dst_offset & ~PAGE_MASK; - loop_len = min(PAGE_SIZE - - max(src_page_off, dst_page_off), - remaining_len); - - if (src_window->type == SCIF_WINDOW_SELF) - src_virt = _get_local_va(src_offset, src_window, - loop_len); - else - src_virt = ioremap_remote(src_offset, src_window, - loop_len, - work->remote_dev, - &src_win_iter); - if (!src_virt) { - ret = -ENOMEM; - goto error; - } - - if (dst_window->type == SCIF_WINDOW_SELF) - dst_virt = _get_local_va(dst_offset, dst_window, - loop_len); - else - dst_virt = ioremap_remote(dst_offset, dst_window, - loop_len, - work->remote_dev, - &dst_win_iter); - if (!dst_virt) { - if (src_window->type == SCIF_WINDOW_PEER) - iounmap_remote(src_virt, loop_len, work); - ret = -ENOMEM; - goto error; - } - - if (work->loopback) { - memcpy(dst_virt, src_virt, loop_len); - } else { - if (src_window->type == SCIF_WINDOW_SELF) - memcpy_toio((void __iomem __force *)dst_virt, - src_virt, loop_len); - else - memcpy_fromio(dst_virt, - (void __iomem __force *)src_virt, - loop_len); - } - if (src_window->type == SCIF_WINDOW_PEER) - iounmap_remote(src_virt, loop_len, work); - - if (dst_window->type == SCIF_WINDOW_PEER) - iounmap_remote(dst_virt, loop_len, work); - - src_offset += loop_len; - dst_offset += loop_len; - remaining_len -= loop_len; - if (remaining_len) { - end_src_offset = src_window->offset + - (src_window->nr_pages << PAGE_SHIFT); - end_dst_offset = dst_window->offset + - (dst_window->nr_pages << PAGE_SHIFT); - if (src_offset == end_src_offset) { - src_window = list_next_entry(src_window, list); - scif_init_window_iter(src_window, - &src_win_iter); - } - if (dst_offset == end_dst_offset) { - dst_window = list_next_entry(dst_window, list); - scif_init_window_iter(dst_window, - &dst_win_iter); - } - } - } -error: - return ret; -} - -static int scif_rma_list_dma_copy_wrapper(struct scif_endpt *epd, - struct scif_copy_work *work, - struct dma_chan *chan, off_t loffset) -{ - int src_cache_off, dst_cache_off; - s64 src_offset = work->src_offset, dst_offset = work->dst_offset; - u8 *temp = NULL; - bool src_local = true; - struct scif_dma_comp_cb *comp_cb; - int err; - - if (is_dma_copy_aligned(chan->device, 1, 1, 1)) - return _scif_rma_list_dma_copy_aligned(work, chan); - - src_cache_off = src_offset & (L1_CACHE_BYTES - 1); - dst_cache_off = dst_offset & (L1_CACHE_BYTES - 1); - - if (dst_cache_off == src_cache_off) - return scif_rma_list_dma_copy_aligned(work, chan); - - if (work->loopback) - return scif_rma_list_cpu_copy(work); - src_local = work->src_window->type == SCIF_WINDOW_SELF; - - /* Allocate dma_completion cb */ - comp_cb = kzalloc(sizeof(*comp_cb), GFP_KERNEL); - if (!comp_cb) - goto error; - - work->comp_cb = comp_cb; - comp_cb->cb_cookie = comp_cb; - comp_cb->dma_completion_func = &scif_rma_completion_cb; - - if (work->len + (L1_CACHE_BYTES << 1) < SCIF_KMEM_UNALIGNED_BUF_SIZE) { - comp_cb->is_cache = false; - /* Allocate padding bytes to align to a cache line */ - temp = kmalloc(work->len + (L1_CACHE_BYTES << 1), - GFP_KERNEL); - if (!temp) - goto free_comp_cb; - comp_cb->temp_buf_to_free = temp; - /* kmalloc(..) does not guarantee cache line alignment */ - if (!IS_ALIGNED((u64)temp, L1_CACHE_BYTES)) - temp = PTR_ALIGN(temp, L1_CACHE_BYTES); - } else { - comp_cb->is_cache = true; - temp = kmem_cache_alloc(unaligned_cache, GFP_KERNEL); - if (!temp) - goto free_comp_cb; - comp_cb->temp_buf_to_free = temp; - } - - if (src_local) { - temp += dst_cache_off; - scif_rma_local_cpu_copy(work->src_offset, work->src_window, - temp, work->len, true); - } else { - comp_cb->dst_window = work->dst_window; - comp_cb->dst_offset = work->dst_offset; - work->src_offset = work->src_offset - src_cache_off; - comp_cb->len = work->len; - work->len = ALIGN(work->len + src_cache_off, L1_CACHE_BYTES); - comp_cb->header_padding = src_cache_off; - } - comp_cb->temp_buf = temp; - - err = scif_map_single(&comp_cb->temp_phys, temp, - work->remote_dev, SCIF_KMEM_UNALIGNED_BUF_SIZE); - if (err) - goto free_temp_buf; - comp_cb->sdev = work->remote_dev; - if (scif_rma_list_dma_copy_unaligned(work, temp, chan, src_local) < 0) - goto free_temp_buf; - if (!src_local) - work->fence_type = SCIF_DMA_INTR; - return 0; -free_temp_buf: - if (comp_cb->is_cache) - kmem_cache_free(unaligned_cache, comp_cb->temp_buf_to_free); - else - kfree(comp_cb->temp_buf_to_free); -free_comp_cb: - kfree(comp_cb); -error: - return -ENOMEM; -} - -/** - * scif_rma_copy: - * @epd: end point descriptor. - * @loffset: offset in local registered address space to/from which to copy - * @addr: user virtual address to/from which to copy - * @len: length of range to copy - * @roffset: offset in remote registered address space to/from which to copy - * @flags: flags - * @dir: LOCAL->REMOTE or vice versa. - * @last_chunk: true if this is the last chunk of a larger transfer - * - * Validate parameters, check if src/dst registered ranges requested for copy - * are valid and initiate either CPU or DMA copy. - */ -static int scif_rma_copy(scif_epd_t epd, off_t loffset, unsigned long addr, - size_t len, off_t roffset, int flags, - enum scif_rma_dir dir, bool last_chunk) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - struct scif_rma_req remote_req; - struct scif_rma_req req; - struct scif_window *local_window = NULL; - struct scif_window *remote_window = NULL; - struct scif_copy_work copy_work; - bool loopback; - int err = 0; - struct dma_chan *chan; - struct scif_mmu_notif *mmn = NULL; - bool cache = false; - struct device *spdev; - - err = scif_verify_epd(ep); - if (err) - return err; - - if (flags && !(flags & (SCIF_RMA_USECPU | SCIF_RMA_USECACHE | - SCIF_RMA_SYNC | SCIF_RMA_ORDERED))) - return -EINVAL; - - loopback = scifdev_self(ep->remote_dev) ? true : false; - copy_work.fence_type = ((flags & SCIF_RMA_SYNC) && last_chunk) ? - SCIF_DMA_POLL : 0; - copy_work.ordered = !!((flags & SCIF_RMA_ORDERED) && last_chunk); - - /* Use CPU for Mgmt node <-> Mgmt node copies */ - if (loopback && scif_is_mgmt_node()) { - flags |= SCIF_RMA_USECPU; - copy_work.fence_type = 0x0; - } - - cache = scif_is_set_reg_cache(flags); - - remote_req.out_window = &remote_window; - remote_req.offset = roffset; - remote_req.nr_bytes = len; - /* - * If transfer is from local to remote then the remote window - * must be writeable and vice versa. - */ - remote_req.prot = dir == SCIF_LOCAL_TO_REMOTE ? VM_WRITE : VM_READ; - remote_req.type = SCIF_WINDOW_PARTIAL; - remote_req.head = &ep->rma_info.remote_reg_list; - - spdev = scif_get_peer_dev(ep->remote_dev); - if (IS_ERR(spdev)) { - err = PTR_ERR(spdev); - return err; - } - - if (addr && cache) { - mutex_lock(&ep->rma_info.mmn_lock); - mmn = scif_find_mmu_notifier(current->mm, &ep->rma_info); - if (!mmn) - mmn = scif_add_mmu_notifier(current->mm, ep); - mutex_unlock(&ep->rma_info.mmn_lock); - if (IS_ERR(mmn)) { - scif_put_peer_dev(spdev); - return PTR_ERR(mmn); - } - cache = cache && !scif_rma_tc_can_cache(ep, len); - } - mutex_lock(&ep->rma_info.rma_lock); - if (addr) { - req.out_window = &local_window; - req.nr_bytes = ALIGN(len + (addr & ~PAGE_MASK), - PAGE_SIZE); - req.va_for_temp = addr & PAGE_MASK; - req.prot = (dir == SCIF_LOCAL_TO_REMOTE ? - VM_READ : VM_WRITE | VM_READ); - /* Does a valid local window exist? */ - if (mmn) { - spin_lock(&ep->rma_info.tc_lock); - req.head = &mmn->tc_reg_list; - err = scif_query_tcw(ep, &req); - spin_unlock(&ep->rma_info.tc_lock); - } - if (!mmn || err) { - err = scif_register_temp(epd, req.va_for_temp, - req.nr_bytes, req.prot, - &loffset, &local_window); - if (err) { - mutex_unlock(&ep->rma_info.rma_lock); - goto error; - } - if (!cache) - goto skip_cache; - atomic_inc(&ep->rma_info.tcw_refcount); - atomic_add_return(local_window->nr_pages, - &ep->rma_info.tcw_total_pages); - if (mmn) { - spin_lock(&ep->rma_info.tc_lock); - scif_insert_tcw(local_window, - &mmn->tc_reg_list); - spin_unlock(&ep->rma_info.tc_lock); - } - } -skip_cache: - loffset = local_window->offset + - (addr - local_window->va_for_temp); - } else { - req.out_window = &local_window; - req.offset = loffset; - /* - * If transfer is from local to remote then the self window - * must be readable and vice versa. - */ - req.prot = dir == SCIF_LOCAL_TO_REMOTE ? VM_READ : VM_WRITE; - req.nr_bytes = len; - req.type = SCIF_WINDOW_PARTIAL; - req.head = &ep->rma_info.reg_list; - /* Does a valid local window exist? */ - err = scif_query_window(&req); - if (err) { - mutex_unlock(&ep->rma_info.rma_lock); - goto error; - } - } - - /* Does a valid remote window exist? */ - err = scif_query_window(&remote_req); - if (err) { - mutex_unlock(&ep->rma_info.rma_lock); - goto error; - } - - /* - * Prepare copy_work for submitting work to the DMA kernel thread - * or CPU copy routine. - */ - copy_work.len = len; - copy_work.loopback = loopback; - copy_work.remote_dev = ep->remote_dev; - if (dir == SCIF_LOCAL_TO_REMOTE) { - copy_work.src_offset = loffset; - copy_work.src_window = local_window; - copy_work.dst_offset = roffset; - copy_work.dst_window = remote_window; - } else { - copy_work.src_offset = roffset; - copy_work.src_window = remote_window; - copy_work.dst_offset = loffset; - copy_work.dst_window = local_window; - } - - if (flags & SCIF_RMA_USECPU) { - scif_rma_list_cpu_copy(©_work); - } else { - chan = ep->rma_info.dma_chan; - err = scif_rma_list_dma_copy_wrapper(epd, ©_work, - chan, loffset); - } - if (addr && !cache) - atomic_inc(&ep->rma_info.tw_refcount); - - mutex_unlock(&ep->rma_info.rma_lock); - - if (last_chunk) { - struct scif_dev *rdev = ep->remote_dev; - - if (copy_work.fence_type == SCIF_DMA_POLL) - err = scif_drain_dma_poll(rdev->sdev, - ep->rma_info.dma_chan); - else if (copy_work.fence_type == SCIF_DMA_INTR) - err = scif_drain_dma_intr(rdev->sdev, - ep->rma_info.dma_chan); - } - - if (addr && !cache) - scif_queue_for_cleanup(local_window, &scif_info.rma); - scif_put_peer_dev(spdev); - return err; -error: - if (err) { - if (addr && local_window && !cache) - scif_destroy_window(ep, local_window); - dev_err(scif_info.mdev.this_device, - "%s %d err %d len 0x%lx\n", - __func__, __LINE__, err, len); - } - scif_put_peer_dev(spdev); - return err; -} - -int scif_readfrom(scif_epd_t epd, off_t loffset, size_t len, - off_t roffset, int flags) -{ - int err; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI readfrom: ep %p loffset 0x%lx len 0x%lx offset 0x%lx flags 0x%x\n", - epd, loffset, len, roffset, flags); - if (scif_unaligned(loffset, roffset)) { - while (len > SCIF_MAX_UNALIGNED_BUF_SIZE) { - err = scif_rma_copy(epd, loffset, 0x0, - SCIF_MAX_UNALIGNED_BUF_SIZE, - roffset, flags, - SCIF_REMOTE_TO_LOCAL, false); - if (err) - goto readfrom_err; - loffset += SCIF_MAX_UNALIGNED_BUF_SIZE; - roffset += SCIF_MAX_UNALIGNED_BUF_SIZE; - len -= SCIF_MAX_UNALIGNED_BUF_SIZE; - } - } - err = scif_rma_copy(epd, loffset, 0x0, len, - roffset, flags, SCIF_REMOTE_TO_LOCAL, true); -readfrom_err: - return err; -} -EXPORT_SYMBOL_GPL(scif_readfrom); - -int scif_writeto(scif_epd_t epd, off_t loffset, size_t len, - off_t roffset, int flags) -{ - int err; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI writeto: ep %p loffset 0x%lx len 0x%lx roffset 0x%lx flags 0x%x\n", - epd, loffset, len, roffset, flags); - if (scif_unaligned(loffset, roffset)) { - while (len > SCIF_MAX_UNALIGNED_BUF_SIZE) { - err = scif_rma_copy(epd, loffset, 0x0, - SCIF_MAX_UNALIGNED_BUF_SIZE, - roffset, flags, - SCIF_LOCAL_TO_REMOTE, false); - if (err) - goto writeto_err; - loffset += SCIF_MAX_UNALIGNED_BUF_SIZE; - roffset += SCIF_MAX_UNALIGNED_BUF_SIZE; - len -= SCIF_MAX_UNALIGNED_BUF_SIZE; - } - } - err = scif_rma_copy(epd, loffset, 0x0, len, - roffset, flags, SCIF_LOCAL_TO_REMOTE, true); -writeto_err: - return err; -} -EXPORT_SYMBOL_GPL(scif_writeto); - -int scif_vreadfrom(scif_epd_t epd, void *addr, size_t len, - off_t roffset, int flags) -{ - int err; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI vreadfrom: ep %p addr %p len 0x%lx roffset 0x%lx flags 0x%x\n", - epd, addr, len, roffset, flags); - if (scif_unaligned((off_t __force)addr, roffset)) { - if (len > SCIF_MAX_UNALIGNED_BUF_SIZE) - flags &= ~SCIF_RMA_USECACHE; - - while (len > SCIF_MAX_UNALIGNED_BUF_SIZE) { - err = scif_rma_copy(epd, 0, (u64)addr, - SCIF_MAX_UNALIGNED_BUF_SIZE, - roffset, flags, - SCIF_REMOTE_TO_LOCAL, false); - if (err) - goto vreadfrom_err; - addr += SCIF_MAX_UNALIGNED_BUF_SIZE; - roffset += SCIF_MAX_UNALIGNED_BUF_SIZE; - len -= SCIF_MAX_UNALIGNED_BUF_SIZE; - } - } - err = scif_rma_copy(epd, 0, (u64)addr, len, - roffset, flags, SCIF_REMOTE_TO_LOCAL, true); -vreadfrom_err: - return err; -} -EXPORT_SYMBOL_GPL(scif_vreadfrom); - -int scif_vwriteto(scif_epd_t epd, void *addr, size_t len, - off_t roffset, int flags) -{ - int err; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI vwriteto: ep %p addr %p len 0x%lx roffset 0x%lx flags 0x%x\n", - epd, addr, len, roffset, flags); - if (scif_unaligned((off_t __force)addr, roffset)) { - if (len > SCIF_MAX_UNALIGNED_BUF_SIZE) - flags &= ~SCIF_RMA_USECACHE; - - while (len > SCIF_MAX_UNALIGNED_BUF_SIZE) { - err = scif_rma_copy(epd, 0, (u64)addr, - SCIF_MAX_UNALIGNED_BUF_SIZE, - roffset, flags, - SCIF_LOCAL_TO_REMOTE, false); - if (err) - goto vwriteto_err; - addr += SCIF_MAX_UNALIGNED_BUF_SIZE; - roffset += SCIF_MAX_UNALIGNED_BUF_SIZE; - len -= SCIF_MAX_UNALIGNED_BUF_SIZE; - } - } - err = scif_rma_copy(epd, 0, (u64)addr, len, - roffset, flags, SCIF_LOCAL_TO_REMOTE, true); -vwriteto_err: - return err; -} -EXPORT_SYMBOL_GPL(scif_vwriteto); diff --git a/drivers/misc/mic/scif/scif_epd.c b/drivers/misc/mic/scif/scif_epd.c deleted file mode 100644 index 426687f6696b..000000000000 --- a/drivers/misc/mic/scif/scif_epd.c +++ /dev/null @@ -1,357 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel SCIF driver. - */ -#include "scif_main.h" -#include "scif_map.h" - -void scif_cleanup_ep_qp(struct scif_endpt *ep) -{ - struct scif_qp *qp = ep->qp_info.qp; - - if (qp->outbound_q.rb_base) { - scif_iounmap((void *)qp->outbound_q.rb_base, - qp->outbound_q.size, ep->remote_dev); - qp->outbound_q.rb_base = NULL; - } - if (qp->remote_qp) { - scif_iounmap((void *)qp->remote_qp, - sizeof(struct scif_qp), ep->remote_dev); - qp->remote_qp = NULL; - } - if (qp->local_qp) { - scif_unmap_single(qp->local_qp, ep->remote_dev, - sizeof(struct scif_qp)); - qp->local_qp = 0x0; - } - if (qp->local_buf) { - scif_unmap_single(qp->local_buf, ep->remote_dev, - SCIF_ENDPT_QP_SIZE); - qp->local_buf = 0; - } -} - -void scif_teardown_ep(void *endpt) -{ - struct scif_endpt *ep = endpt; - struct scif_qp *qp = ep->qp_info.qp; - - if (qp) { - spin_lock(&ep->lock); - scif_cleanup_ep_qp(ep); - spin_unlock(&ep->lock); - kfree(qp->inbound_q.rb_base); - kfree(qp); - } -} - -/* - * Enqueue the endpoint to the zombie list for cleanup. - * The endpoint should not be accessed once this API returns. - */ -void scif_add_epd_to_zombie_list(struct scif_endpt *ep, bool eplock_held) -{ - if (!eplock_held) - mutex_lock(&scif_info.eplock); - spin_lock(&ep->lock); - ep->state = SCIFEP_ZOMBIE; - spin_unlock(&ep->lock); - list_add_tail(&ep->list, &scif_info.zombie); - scif_info.nr_zombies++; - if (!eplock_held) - mutex_unlock(&scif_info.eplock); - schedule_work(&scif_info.misc_work); -} - -static struct scif_endpt *scif_find_listen_ep(u16 port) -{ - struct scif_endpt *ep = NULL; - struct list_head *pos, *tmpq; - - mutex_lock(&scif_info.eplock); - list_for_each_safe(pos, tmpq, &scif_info.listen) { - ep = list_entry(pos, struct scif_endpt, list); - if (ep->port.port == port) { - mutex_unlock(&scif_info.eplock); - return ep; - } - } - mutex_unlock(&scif_info.eplock); - return NULL; -} - -void scif_cleanup_zombie_epd(void) -{ - struct list_head *pos, *tmpq; - struct scif_endpt *ep; - - mutex_lock(&scif_info.eplock); - list_for_each_safe(pos, tmpq, &scif_info.zombie) { - ep = list_entry(pos, struct scif_endpt, list); - if (scif_rma_ep_can_uninit(ep)) { - list_del(pos); - scif_info.nr_zombies--; - put_iova_domain(&ep->rma_info.iovad); - kfree(ep); - } - } - mutex_unlock(&scif_info.eplock); -} - -/** - * scif_cnctreq() - Respond to SCIF_CNCT_REQ interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * This message is initiated by the remote node to request a connection - * to the local node. This function looks for an end point in the - * listen state on the requested port id. - * - * If it finds a listening port it places the connect request on the - * listening end points queue and wakes up any pending accept calls. - * - * If it does not find a listening end point it sends a connection - * reject message to the remote node. - */ -void scif_cnctreq(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = NULL; - struct scif_conreq *conreq; - - conreq = kmalloc(sizeof(*conreq), GFP_KERNEL); - if (!conreq) - /* Lack of resources so reject the request. */ - goto conreq_sendrej; - - ep = scif_find_listen_ep(msg->dst.port); - if (!ep) - /* Send reject due to no listening ports */ - goto conreq_sendrej_free; - else - spin_lock(&ep->lock); - - if (ep->backlog <= ep->conreqcnt) { - /* Send reject due to too many pending requests */ - spin_unlock(&ep->lock); - goto conreq_sendrej_free; - } - - conreq->msg = *msg; - list_add_tail(&conreq->list, &ep->conlist); - ep->conreqcnt++; - wake_up_interruptible(&ep->conwq); - spin_unlock(&ep->lock); - return; - -conreq_sendrej_free: - kfree(conreq); -conreq_sendrej: - msg->uop = SCIF_CNCT_REJ; - scif_nodeqp_send(&scif_dev[msg->src.node], msg); -} - -/** - * scif_cnctgnt() - Respond to SCIF_CNCT_GNT interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * An accept() on the remote node has occurred and sent this message - * to indicate success. Place the end point in the MAPPING state and - * save the remote nodes memory information. Then wake up the connect - * request so it can finish. - */ -void scif_cnctgnt(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - - spin_lock(&ep->lock); - if (SCIFEP_CONNECTING == ep->state) { - ep->peer.node = msg->src.node; - ep->peer.port = msg->src.port; - ep->qp_info.gnt_pld = msg->payload[1]; - ep->remote_ep = msg->payload[2]; - ep->state = SCIFEP_MAPPING; - - wake_up(&ep->conwq); - } - spin_unlock(&ep->lock); -} - -/** - * scif_cnctgnt_ack() - Respond to SCIF_CNCT_GNTACK interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * The remote connection request has finished mapping the local memory. - * Place the connection in the connected state and wake up the pending - * accept() call. - */ -void scif_cnctgnt_ack(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - - mutex_lock(&scif_info.connlock); - spin_lock(&ep->lock); - /* New ep is now connected with all resources set. */ - ep->state = SCIFEP_CONNECTED; - list_add_tail(&ep->list, &scif_info.connected); - wake_up(&ep->conwq); - spin_unlock(&ep->lock); - mutex_unlock(&scif_info.connlock); -} - -/** - * scif_cnctgnt_nack() - Respond to SCIF_CNCT_GNTNACK interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * The remote connection request failed to map the local memory it was sent. - * Place the end point in the CLOSING state to indicate it and wake up - * the pending accept(); - */ -void scif_cnctgnt_nack(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - - spin_lock(&ep->lock); - ep->state = SCIFEP_CLOSING; - wake_up(&ep->conwq); - spin_unlock(&ep->lock); -} - -/** - * scif_cnctrej() - Respond to SCIF_CNCT_REJ interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * The remote end has rejected the connection request. Set the end - * point back to the bound state and wake up the pending connect(). - */ -void scif_cnctrej(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - - spin_lock(&ep->lock); - if (SCIFEP_CONNECTING == ep->state) { - ep->state = SCIFEP_BOUND; - wake_up(&ep->conwq); - } - spin_unlock(&ep->lock); -} - -/** - * scif_discnct() - Respond to SCIF_DISCNCT interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * The remote node has indicated close() has been called on its end - * point. Remove the local end point from the connected list, set its - * state to disconnected and ensure accesses to the remote node are - * shutdown. - * - * When all accesses to the remote end have completed then send a - * DISCNT_ACK to indicate it can remove its resources and complete - * the close routine. - */ -void scif_discnct(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = NULL; - struct scif_endpt *tmpep; - struct list_head *pos, *tmpq; - - mutex_lock(&scif_info.connlock); - list_for_each_safe(pos, tmpq, &scif_info.connected) { - tmpep = list_entry(pos, struct scif_endpt, list); - /* - * The local ep may have sent a disconnect and and been closed - * due to a message response time out. It may have been - * allocated again and formed a new connection so we want to - * check if the remote ep matches - */ - if (((u64)tmpep == msg->payload[1]) && - ((u64)tmpep->remote_ep == msg->payload[0])) { - list_del(pos); - ep = tmpep; - spin_lock(&ep->lock); - break; - } - } - - /* - * If the terminated end is not found then this side started closing - * before the other side sent the disconnect. If so the ep will no - * longer be on the connected list. Regardless the other side - * needs to be acked to let it know close is complete. - */ - if (!ep) { - mutex_unlock(&scif_info.connlock); - goto discnct_ack; - } - - ep->state = SCIFEP_DISCONNECTED; - list_add_tail(&ep->list, &scif_info.disconnected); - - wake_up_interruptible(&ep->sendwq); - wake_up_interruptible(&ep->recvwq); - spin_unlock(&ep->lock); - mutex_unlock(&scif_info.connlock); - -discnct_ack: - msg->uop = SCIF_DISCNT_ACK; - scif_nodeqp_send(&scif_dev[msg->src.node], msg); -} - -/** - * scif_discnct_ack() - Respond to SCIF_DISCNT_ACK interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * Remote side has indicated it has not more references to local resources - */ -void scif_discnt_ack(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - - spin_lock(&ep->lock); - ep->state = SCIFEP_DISCONNECTED; - spin_unlock(&ep->lock); - complete(&ep->discon); -} - -/** - * scif_clientsend() - Respond to SCIF_CLIENT_SEND interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * Remote side is confirming send or receive interrupt handling is complete. - */ -void scif_clientsend(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - - spin_lock(&ep->lock); - if (SCIFEP_CONNECTED == ep->state) - wake_up_interruptible(&ep->recvwq); - spin_unlock(&ep->lock); -} - -/** - * scif_clientrcvd() - Respond to SCIF_CLIENT_RCVD interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * Remote side is confirming send or receive interrupt handling is complete. - */ -void scif_clientrcvd(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - - spin_lock(&ep->lock); - if (SCIFEP_CONNECTED == ep->state) - wake_up_interruptible(&ep->sendwq); - spin_unlock(&ep->lock); -} diff --git a/drivers/misc/mic/scif/scif_epd.h b/drivers/misc/mic/scif/scif_epd.h deleted file mode 100644 index 0b9dfe1cc06c..000000000000 --- a/drivers/misc/mic/scif/scif_epd.h +++ /dev/null @@ -1,200 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel SCIF driver. - */ -#ifndef SCIF_EPD_H -#define SCIF_EPD_H - -#include -#include -#include - -#define SCIF_EPLOCK_HELD true - -enum scif_epd_state { - SCIFEP_UNBOUND, - SCIFEP_BOUND, - SCIFEP_LISTENING, - SCIFEP_CONNECTED, - SCIFEP_CONNECTING, - SCIFEP_MAPPING, - SCIFEP_CLOSING, - SCIFEP_CLLISTEN, - SCIFEP_DISCONNECTED, - SCIFEP_ZOMBIE -}; - -/* - * struct scif_conreq - Data structure added to the connection list. - * - * @msg: connection request message received - * @list: link to list of connection requests - */ -struct scif_conreq { - struct scifmsg msg; - struct list_head list; -}; - -/* Size of the RB for the Endpoint QP */ -#define SCIF_ENDPT_QP_SIZE 0x1000 - -/* - * scif_endpt_qp_info - SCIF endpoint queue pair - * - * @qp - Qpair for this endpoint - * @qp_offset - DMA address of the QP - * @gnt_pld - Payload in a SCIF_CNCT_GNT message containing the - * physical address of the remote_qp. - */ -struct scif_endpt_qp_info { - struct scif_qp *qp; - dma_addr_t qp_offset; - dma_addr_t gnt_pld; -}; - -/* - * struct scif_endpt - The SCIF endpoint data structure - * - * @state: end point state - * @lock: lock synchronizing access to endpoint fields like state etc - * @port: self port information - * @peer: peer port information - * @backlog: maximum pending connection requests - * @qp_info: Endpoint QP information for SCIF messaging - * @remote_dev: scifdev used by this endpt to communicate with remote node. - * @remote_ep: remote endpoint - * @conreqcnt: Keep track of number of connection requests. - * @files: Open file information used to match the id passed in with - * the flush routine. - * @conlist: list of connection requests - * @conwq: waitqueue for connection processing - * @discon: completion used during disconnection - * @sendwq: waitqueue used during sending messages - * @recvwq: waitqueue used during message receipt - * @sendlock: Synchronize ordering of messages sent - * @recvlock: Synchronize ordering of messages received - * @list: link to list of various endpoints like connected, listening etc - * @li_accept: pending ACCEPTREG - * @acceptcnt: pending ACCEPTREG cnt - * @liacceptlist: link to listen accept - * @miacceptlist: link to uaccept - * @listenep: associated listen ep - * @conn_work: Non blocking connect work - * @conn_port: Connection port - * @conn_err: Errors during connection - * @conn_async_state: Async connection - * @conn_pend_wq: Used by poll while waiting for incoming connections - * @conn_list: List of async connection requests - * @rma_info: Information for triggering SCIF RMA and DMA operations - * @mmu_list: link to list of MMU notifier cleanup work - * @anon: anonymous file for use in kernel mode scif poll - */ -struct scif_endpt { - enum scif_epd_state state; - spinlock_t lock; - struct scif_port_id port; - struct scif_port_id peer; - int backlog; - struct scif_endpt_qp_info qp_info; - struct scif_dev *remote_dev; - u64 remote_ep; - int conreqcnt; - struct files_struct *files; - struct list_head conlist; - wait_queue_head_t conwq; - struct completion discon; - wait_queue_head_t sendwq; - wait_queue_head_t recvwq; - struct mutex sendlock; - struct mutex recvlock; - struct list_head list; - struct list_head li_accept; - int acceptcnt; - struct list_head liacceptlist; - struct list_head miacceptlist; - struct scif_endpt *listenep; - struct scif_port_id conn_port; - int conn_err; - int conn_async_state; - wait_queue_head_t conn_pend_wq; - struct list_head conn_list; - struct scif_endpt_rma_info rma_info; - struct list_head mmu_list; - struct file *anon; -}; - -static inline int scifdev_alive(struct scif_endpt *ep) -{ - return _scifdev_alive(ep->remote_dev); -} - -/* - * scif_verify_epd: - * ep: SCIF endpoint - * - * Checks several generic error conditions and returns the - * appropriate error. - */ -static inline int scif_verify_epd(struct scif_endpt *ep) -{ - if (ep->state == SCIFEP_DISCONNECTED) - return -ECONNRESET; - - if (ep->state != SCIFEP_CONNECTED) - return -ENOTCONN; - - if (!scifdev_alive(ep)) - return -ENODEV; - - return 0; -} - -static inline int scif_anon_inode_getfile(scif_epd_t epd) -{ - epd->anon = anon_inode_getfile("scif", &scif_anon_fops, NULL, 0); - - return PTR_ERR_OR_ZERO(epd->anon); -} - -static inline void scif_anon_inode_fput(scif_epd_t epd) -{ - if (epd->anon) { - fput(epd->anon); - epd->anon = NULL; - } -} - -void scif_cleanup_zombie_epd(void); -void scif_teardown_ep(void *endpt); -void scif_cleanup_ep_qp(struct scif_endpt *ep); -void scif_add_epd_to_zombie_list(struct scif_endpt *ep, bool eplock_held); -void scif_get_node_info(void); -void scif_send_acks(struct scif_dev *dev); -void scif_conn_handler(struct work_struct *work); -int scif_rsrv_port(u16 port); -void scif_get_port(u16 port); -int scif_get_new_port(void); -void scif_put_port(u16 port); -int scif_user_send(scif_epd_t epd, void __user *msg, int len, int flags); -int scif_user_recv(scif_epd_t epd, void __user *msg, int len, int flags); -void scif_cnctreq(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_cnctgnt(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_cnctgnt_ack(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_cnctgnt_nack(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_cnctrej(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_discnct(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_discnt_ack(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_clientsend(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_clientrcvd(struct scif_dev *scifdev, struct scifmsg *msg); -int __scif_connect(scif_epd_t epd, struct scif_port_id *dst, bool non_block); -int __scif_flush(scif_epd_t epd); -int scif_mmap(struct vm_area_struct *vma, scif_epd_t epd); -__poll_t __scif_pollfd(struct file *f, poll_table *wait, - struct scif_endpt *ep); -int __scif_pin_pages(void *addr, size_t len, int *out_prot, - int map_flags, scif_pinned_pages_t *pages); -#endif /* SCIF_EPD_H */ diff --git a/drivers/misc/mic/scif/scif_fd.c b/drivers/misc/mic/scif/scif_fd.c deleted file mode 100644 index 3f08646cd78a..000000000000 --- a/drivers/misc/mic/scif/scif_fd.c +++ /dev/null @@ -1,462 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel SCIF driver. - */ -#include "scif_main.h" - -static int scif_fdopen(struct inode *inode, struct file *f) -{ - struct scif_endpt *priv = scif_open(); - - if (!priv) - return -ENOMEM; - f->private_data = priv; - return 0; -} - -static int scif_fdclose(struct inode *inode, struct file *f) -{ - struct scif_endpt *priv = f->private_data; - - return scif_close(priv); -} - -static int scif_fdmmap(struct file *f, struct vm_area_struct *vma) -{ - struct scif_endpt *priv = f->private_data; - - return scif_mmap(vma, priv); -} - -static __poll_t scif_fdpoll(struct file *f, poll_table *wait) -{ - struct scif_endpt *priv = f->private_data; - - return __scif_pollfd(f, wait, priv); -} - -static int scif_fdflush(struct file *f, fl_owner_t id) -{ - struct scif_endpt *ep = f->private_data; - - spin_lock(&ep->lock); - /* - * The listening endpoint stashes the open file information before - * waiting for incoming connections. The release callback would never be - * called if the application closed the endpoint, while waiting for - * incoming connections from a separate thread since the file descriptor - * reference count is bumped up in the accept IOCTL. Call the flush - * routine if the id matches the endpoint open file information so that - * the listening endpoint can be woken up and the fd released. - */ - if (ep->files == id) - __scif_flush(ep); - spin_unlock(&ep->lock); - return 0; -} - -static __always_inline void scif_err_debug(int err, const char *str) -{ - /* - * ENOTCONN is a common uninteresting error which is - * flooding debug messages to the console unnecessarily. - */ - if (err < 0 && err != -ENOTCONN) - dev_dbg(scif_info.mdev.this_device, "%s err %d\n", str, err); -} - -static long scif_fdioctl(struct file *f, unsigned int cmd, unsigned long arg) -{ - struct scif_endpt *priv = f->private_data; - void __user *argp = (void __user *)arg; - int err = 0; - struct scifioctl_msg request; - bool non_block = false; - - non_block = !!(f->f_flags & O_NONBLOCK); - - switch (cmd) { - case SCIF_BIND: - { - int pn; - - if (copy_from_user(&pn, argp, sizeof(pn))) - return -EFAULT; - - pn = scif_bind(priv, pn); - if (pn < 0) - return pn; - - if (copy_to_user(argp, &pn, sizeof(pn))) - return -EFAULT; - - return 0; - } - case SCIF_LISTEN: - return scif_listen(priv, arg); - case SCIF_CONNECT: - { - struct scifioctl_connect req; - struct scif_endpt *ep = (struct scif_endpt *)priv; - - if (copy_from_user(&req, argp, sizeof(req))) - return -EFAULT; - - err = __scif_connect(priv, &req.peer, non_block); - if (err < 0) - return err; - - req.self.node = ep->port.node; - req.self.port = ep->port.port; - - if (copy_to_user(argp, &req, sizeof(req))) - return -EFAULT; - - return 0; - } - /* - * Accept is done in two halves. The request ioctl does the basic - * functionality of accepting the request and returning the information - * about it including the internal ID of the end point. The register - * is done with the internal ID on a new file descriptor opened by the - * requesting process. - */ - case SCIF_ACCEPTREQ: - { - struct scifioctl_accept request; - scif_epd_t *ep = (scif_epd_t *)&request.endpt; - - if (copy_from_user(&request, argp, sizeof(request))) - return -EFAULT; - - err = scif_accept(priv, &request.peer, ep, request.flags); - if (err < 0) - return err; - - if (copy_to_user(argp, &request, sizeof(request))) { - scif_close(*ep); - return -EFAULT; - } - /* - * Add to the list of user mode eps where the second half - * of the accept is not yet completed. - */ - mutex_lock(&scif_info.eplock); - list_add_tail(&((*ep)->miacceptlist), &scif_info.uaccept); - list_add_tail(&((*ep)->liacceptlist), &priv->li_accept); - (*ep)->listenep = priv; - priv->acceptcnt++; - mutex_unlock(&scif_info.eplock); - - return 0; - } - case SCIF_ACCEPTREG: - { - struct scif_endpt *priv = f->private_data; - struct scif_endpt *newep; - struct scif_endpt *lisep; - struct scif_endpt *fep = NULL; - struct scif_endpt *tmpep; - struct list_head *pos, *tmpq; - - /* Finally replace the pointer to the accepted endpoint */ - if (copy_from_user(&newep, argp, sizeof(void *))) - return -EFAULT; - - /* Remove form the user accept queue */ - mutex_lock(&scif_info.eplock); - list_for_each_safe(pos, tmpq, &scif_info.uaccept) { - tmpep = list_entry(pos, - struct scif_endpt, miacceptlist); - if (tmpep == newep) { - list_del(pos); - fep = tmpep; - break; - } - } - - if (!fep) { - mutex_unlock(&scif_info.eplock); - return -ENOENT; - } - - lisep = newep->listenep; - list_for_each_safe(pos, tmpq, &lisep->li_accept) { - tmpep = list_entry(pos, - struct scif_endpt, liacceptlist); - if (tmpep == newep) { - list_del(pos); - lisep->acceptcnt--; - break; - } - } - - mutex_unlock(&scif_info.eplock); - - /* Free the resources automatically created from the open. */ - scif_anon_inode_fput(priv); - scif_teardown_ep(priv); - scif_add_epd_to_zombie_list(priv, !SCIF_EPLOCK_HELD); - f->private_data = newep; - return 0; - } - case SCIF_SEND: - { - struct scif_endpt *priv = f->private_data; - - if (copy_from_user(&request, argp, - sizeof(struct scifioctl_msg))) { - err = -EFAULT; - goto send_err; - } - err = scif_user_send(priv, (void __user *)request.msg, - request.len, request.flags); - if (err < 0) - goto send_err; - if (copy_to_user(& - ((struct scifioctl_msg __user *)argp)->out_len, - &err, sizeof(err))) { - err = -EFAULT; - goto send_err; - } - err = 0; -send_err: - scif_err_debug(err, "scif_send"); - return err; - } - case SCIF_RECV: - { - struct scif_endpt *priv = f->private_data; - - if (copy_from_user(&request, argp, - sizeof(struct scifioctl_msg))) { - err = -EFAULT; - goto recv_err; - } - - err = scif_user_recv(priv, (void __user *)request.msg, - request.len, request.flags); - if (err < 0) - goto recv_err; - - if (copy_to_user(& - ((struct scifioctl_msg __user *)argp)->out_len, - &err, sizeof(err))) { - err = -EFAULT; - goto recv_err; - } - err = 0; -recv_err: - scif_err_debug(err, "scif_recv"); - return err; - } - case SCIF_GET_NODEIDS: - { - struct scifioctl_node_ids node_ids; - int entries; - u16 *nodes; - void __user *unodes, *uself; - u16 self; - - if (copy_from_user(&node_ids, argp, sizeof(node_ids))) { - err = -EFAULT; - goto getnodes_err2; - } - - entries = min_t(int, scif_info.maxid, node_ids.len); - nodes = kmalloc_array(entries, sizeof(u16), GFP_KERNEL); - if (entries && !nodes) { - err = -ENOMEM; - goto getnodes_err2; - } - node_ids.len = scif_get_node_ids(nodes, entries, &self); - - unodes = (void __user *)node_ids.nodes; - if (copy_to_user(unodes, nodes, sizeof(u16) * entries)) { - err = -EFAULT; - goto getnodes_err1; - } - - uself = (void __user *)node_ids.self; - if (copy_to_user(uself, &self, sizeof(u16))) { - err = -EFAULT; - goto getnodes_err1; - } - - if (copy_to_user(argp, &node_ids, sizeof(node_ids))) { - err = -EFAULT; - goto getnodes_err1; - } -getnodes_err1: - kfree(nodes); -getnodes_err2: - return err; - } - case SCIF_REG: - { - struct scif_endpt *priv = f->private_data; - struct scifioctl_reg reg; - off_t ret; - - if (copy_from_user(®, argp, sizeof(reg))) { - err = -EFAULT; - goto reg_err; - } - if (reg.flags & SCIF_MAP_KERNEL) { - err = -EINVAL; - goto reg_err; - } - ret = scif_register(priv, (void *)reg.addr, reg.len, - reg.offset, reg.prot, reg.flags); - if (ret < 0) { - err = (int)ret; - goto reg_err; - } - - if (copy_to_user(&((struct scifioctl_reg __user *)argp) - ->out_offset, &ret, sizeof(reg.out_offset))) { - err = -EFAULT; - goto reg_err; - } - err = 0; -reg_err: - scif_err_debug(err, "scif_register"); - return err; - } - case SCIF_UNREG: - { - struct scif_endpt *priv = f->private_data; - struct scifioctl_unreg unreg; - - if (copy_from_user(&unreg, argp, sizeof(unreg))) { - err = -EFAULT; - goto unreg_err; - } - err = scif_unregister(priv, unreg.offset, unreg.len); -unreg_err: - scif_err_debug(err, "scif_unregister"); - return err; - } - case SCIF_READFROM: - { - struct scif_endpt *priv = f->private_data; - struct scifioctl_copy copy; - - if (copy_from_user(©, argp, sizeof(copy))) { - err = -EFAULT; - goto readfrom_err; - } - err = scif_readfrom(priv, copy.loffset, copy.len, copy.roffset, - copy.flags); -readfrom_err: - scif_err_debug(err, "scif_readfrom"); - return err; - } - case SCIF_WRITETO: - { - struct scif_endpt *priv = f->private_data; - struct scifioctl_copy copy; - - if (copy_from_user(©, argp, sizeof(copy))) { - err = -EFAULT; - goto writeto_err; - } - err = scif_writeto(priv, copy.loffset, copy.len, copy.roffset, - copy.flags); -writeto_err: - scif_err_debug(err, "scif_writeto"); - return err; - } - case SCIF_VREADFROM: - { - struct scif_endpt *priv = f->private_data; - struct scifioctl_copy copy; - - if (copy_from_user(©, argp, sizeof(copy))) { - err = -EFAULT; - goto vreadfrom_err; - } - err = scif_vreadfrom(priv, (void __force *)copy.addr, copy.len, - copy.roffset, copy.flags); -vreadfrom_err: - scif_err_debug(err, "scif_vreadfrom"); - return err; - } - case SCIF_VWRITETO: - { - struct scif_endpt *priv = f->private_data; - struct scifioctl_copy copy; - - if (copy_from_user(©, argp, sizeof(copy))) { - err = -EFAULT; - goto vwriteto_err; - } - err = scif_vwriteto(priv, (void __force *)copy.addr, copy.len, - copy.roffset, copy.flags); -vwriteto_err: - scif_err_debug(err, "scif_vwriteto"); - return err; - } - case SCIF_FENCE_MARK: - { - struct scif_endpt *priv = f->private_data; - struct scifioctl_fence_mark mark; - int tmp_mark = 0; - - if (copy_from_user(&mark, argp, sizeof(mark))) { - err = -EFAULT; - goto fence_mark_err; - } - err = scif_fence_mark(priv, mark.flags, &tmp_mark); - if (err) - goto fence_mark_err; - if (copy_to_user((void __user *)mark.mark, &tmp_mark, - sizeof(tmp_mark))) { - err = -EFAULT; - goto fence_mark_err; - } -fence_mark_err: - scif_err_debug(err, "scif_fence_mark"); - return err; - } - case SCIF_FENCE_WAIT: - { - struct scif_endpt *priv = f->private_data; - - err = scif_fence_wait(priv, arg); - scif_err_debug(err, "scif_fence_wait"); - return err; - } - case SCIF_FENCE_SIGNAL: - { - struct scif_endpt *priv = f->private_data; - struct scifioctl_fence_signal signal; - - if (copy_from_user(&signal, argp, sizeof(signal))) { - err = -EFAULT; - goto fence_signal_err; - } - - err = scif_fence_signal(priv, signal.loff, signal.lval, - signal.roff, signal.rval, signal.flags); -fence_signal_err: - scif_err_debug(err, "scif_fence_signal"); - return err; - } - } - return -EINVAL; -} - -const struct file_operations scif_fops = { - .open = scif_fdopen, - .release = scif_fdclose, - .unlocked_ioctl = scif_fdioctl, - .mmap = scif_fdmmap, - .poll = scif_fdpoll, - .flush = scif_fdflush, - .owner = THIS_MODULE, -}; diff --git a/drivers/misc/mic/scif/scif_fence.c b/drivers/misc/mic/scif/scif_fence.c deleted file mode 100644 index 4fedf6183951..000000000000 --- a/drivers/misc/mic/scif/scif_fence.c +++ /dev/null @@ -1,783 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2015 Intel Corporation. - * - * Intel SCIF driver. - */ - -#include "scif_main.h" - -/** - * scif_recv_mark: Handle SCIF_MARK request - * @scifdev: SCIF device - * @msg: Interrupt message - * - * The peer has requested a mark. - */ -void scif_recv_mark(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - int mark = 0; - int err; - - err = _scif_fence_mark(ep, &mark); - if (err) - msg->uop = SCIF_MARK_NACK; - else - msg->uop = SCIF_MARK_ACK; - msg->payload[0] = ep->remote_ep; - msg->payload[2] = mark; - scif_nodeqp_send(ep->remote_dev, msg); -} - -/** - * scif_recv_mark_resp: Handle SCIF_MARK_(N)ACK messages. - * @scifdev: SCIF device - * @msg: Interrupt message - * - * The peer has responded to a SCIF_MARK message. - */ -void scif_recv_mark_resp(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - struct scif_fence_info *fence_req = - (struct scif_fence_info *)msg->payload[1]; - - mutex_lock(&ep->rma_info.rma_lock); - if (msg->uop == SCIF_MARK_ACK) { - fence_req->state = OP_COMPLETED; - fence_req->dma_mark = (int)msg->payload[2]; - } else { - fence_req->state = OP_FAILED; - } - mutex_unlock(&ep->rma_info.rma_lock); - complete(&fence_req->comp); -} - -/** - * scif_recv_wait: Handle SCIF_WAIT request - * @scifdev: SCIF device - * @msg: Interrupt message - * - * The peer has requested waiting on a fence. - */ -void scif_recv_wait(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - struct scif_remote_fence_info *fence; - - /* - * Allocate structure for remote fence information and - * send a NACK if the allocation failed. The peer will - * return ENOMEM upon receiving a NACK. - */ - fence = kmalloc(sizeof(*fence), GFP_KERNEL); - if (!fence) { - msg->payload[0] = ep->remote_ep; - msg->uop = SCIF_WAIT_NACK; - scif_nodeqp_send(ep->remote_dev, msg); - return; - } - - /* Prepare the fence request */ - memcpy(&fence->msg, msg, sizeof(struct scifmsg)); - INIT_LIST_HEAD(&fence->list); - - /* Insert to the global remote fence request list */ - mutex_lock(&scif_info.fencelock); - atomic_inc(&ep->rma_info.fence_refcount); - list_add_tail(&fence->list, &scif_info.fence); - mutex_unlock(&scif_info.fencelock); - - schedule_work(&scif_info.misc_work); -} - -/** - * scif_recv_wait_resp: Handle SCIF_WAIT_(N)ACK messages. - * @scifdev: SCIF device - * @msg: Interrupt message - * - * The peer has responded to a SCIF_WAIT message. - */ -void scif_recv_wait_resp(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - struct scif_fence_info *fence_req = - (struct scif_fence_info *)msg->payload[1]; - - mutex_lock(&ep->rma_info.rma_lock); - if (msg->uop == SCIF_WAIT_ACK) - fence_req->state = OP_COMPLETED; - else - fence_req->state = OP_FAILED; - mutex_unlock(&ep->rma_info.rma_lock); - complete(&fence_req->comp); -} - -/** - * scif_recv_sig_local: Handle SCIF_SIG_LOCAL request - * @scifdev: SCIF device - * @msg: Interrupt message - * - * The peer has requested a signal on a local offset. - */ -void scif_recv_sig_local(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - int err; - - err = scif_prog_signal(ep, msg->payload[1], msg->payload[2], - SCIF_WINDOW_SELF); - if (err) - msg->uop = SCIF_SIG_NACK; - else - msg->uop = SCIF_SIG_ACK; - msg->payload[0] = ep->remote_ep; - scif_nodeqp_send(ep->remote_dev, msg); -} - -/** - * scif_recv_sig_remote: Handle SCIF_SIGNAL_REMOTE request - * @scifdev: SCIF device - * @msg: Interrupt message - * - * The peer has requested a signal on a remote offset. - */ -void scif_recv_sig_remote(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - int err; - - err = scif_prog_signal(ep, msg->payload[1], msg->payload[2], - SCIF_WINDOW_PEER); - if (err) - msg->uop = SCIF_SIG_NACK; - else - msg->uop = SCIF_SIG_ACK; - msg->payload[0] = ep->remote_ep; - scif_nodeqp_send(ep->remote_dev, msg); -} - -/** - * scif_recv_sig_resp: Handle SCIF_SIG_(N)ACK messages. - * @scifdev: SCIF device - * @msg: Interrupt message - * - * The peer has responded to a signal request. - */ -void scif_recv_sig_resp(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - struct scif_fence_info *fence_req = - (struct scif_fence_info *)msg->payload[3]; - - mutex_lock(&ep->rma_info.rma_lock); - if (msg->uop == SCIF_SIG_ACK) - fence_req->state = OP_COMPLETED; - else - fence_req->state = OP_FAILED; - mutex_unlock(&ep->rma_info.rma_lock); - complete(&fence_req->comp); -} - -static inline void *scif_get_local_va(off_t off, struct scif_window *window) -{ - struct page **pages = window->pinned_pages->pages; - int page_nr = (off - window->offset) >> PAGE_SHIFT; - off_t page_off = off & ~PAGE_MASK; - - return page_address(pages[page_nr]) + page_off; -} - -static void scif_prog_signal_cb(void *arg) -{ - struct scif_cb_arg *cb_arg = arg; - - dma_pool_free(cb_arg->ep->remote_dev->signal_pool, cb_arg->status, - cb_arg->src_dma_addr); - kfree(cb_arg); -} - -static int _scif_prog_signal(scif_epd_t epd, dma_addr_t dst, u64 val) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - struct dma_chan *chan = ep->rma_info.dma_chan; - struct dma_device *ddev = chan->device; - bool x100 = !is_dma_copy_aligned(chan->device, 1, 1, 1); - struct dma_async_tx_descriptor *tx; - struct scif_status *status = NULL; - struct scif_cb_arg *cb_arg = NULL; - dma_addr_t src; - dma_cookie_t cookie; - int err; - - tx = ddev->device_prep_dma_memcpy(chan, 0, 0, 0, DMA_PREP_FENCE); - if (!tx) { - err = -ENOMEM; - dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - goto alloc_fail; - } - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - err = (int)cookie; - dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - goto alloc_fail; - } - dma_async_issue_pending(chan); - if (x100) { - /* - * For X100 use the status descriptor to write the value to - * the destination. - */ - tx = ddev->device_prep_dma_imm_data(chan, dst, val, 0); - } else { - status = dma_pool_alloc(ep->remote_dev->signal_pool, GFP_KERNEL, - &src); - if (!status) { - err = -ENOMEM; - dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - goto alloc_fail; - } - status->val = val; - status->src_dma_addr = src; - status->ep = ep; - src += offsetof(struct scif_status, val); - tx = ddev->device_prep_dma_memcpy(chan, dst, src, sizeof(val), - DMA_PREP_INTERRUPT); - } - if (!tx) { - err = -ENOMEM; - dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - goto dma_fail; - } - if (!x100) { - cb_arg = kmalloc(sizeof(*cb_arg), GFP_KERNEL); - if (!cb_arg) { - err = -ENOMEM; - goto dma_fail; - } - cb_arg->src_dma_addr = src; - cb_arg->status = status; - cb_arg->ep = ep; - tx->callback = scif_prog_signal_cb; - tx->callback_param = cb_arg; - } - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - err = -EIO; - dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - goto dma_fail; - } - dma_async_issue_pending(chan); - return 0; -dma_fail: - if (!x100) { - dma_pool_free(ep->remote_dev->signal_pool, status, - src - offsetof(struct scif_status, val)); - kfree(cb_arg); - } -alloc_fail: - return err; -} - -/** - * scif_prog_signal: - * @epd: Endpoint Descriptor - * @offset: registered address to write @val to - * @val: Value to be written at @offset - * @type: Type of the window. - * - * Arrange to write a value to the registered offset after ensuring that the - * offset provided is indeed valid. - */ -int scif_prog_signal(scif_epd_t epd, off_t offset, u64 val, - enum scif_window_type type) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - struct scif_window *window = NULL; - struct scif_rma_req req; - dma_addr_t dst_dma_addr; - int err; - - mutex_lock(&ep->rma_info.rma_lock); - req.out_window = &window; - req.offset = offset; - req.nr_bytes = sizeof(u64); - req.prot = SCIF_PROT_WRITE; - req.type = SCIF_WINDOW_SINGLE; - if (type == SCIF_WINDOW_SELF) - req.head = &ep->rma_info.reg_list; - else - req.head = &ep->rma_info.remote_reg_list; - /* Does a valid window exist? */ - err = scif_query_window(&req); - if (err) { - dev_err(scif_info.mdev.this_device, - "%s %d err %d\n", __func__, __LINE__, err); - goto unlock_ret; - } - - if (scif_is_mgmt_node() && scifdev_self(ep->remote_dev)) { - u64 *dst_virt; - - if (type == SCIF_WINDOW_SELF) - dst_virt = scif_get_local_va(offset, window); - else - dst_virt = - scif_get_local_va(offset, (struct scif_window *) - window->peer_window); - *dst_virt = val; - } else { - dst_dma_addr = __scif_off_to_dma_addr(window, offset); - err = _scif_prog_signal(epd, dst_dma_addr, val); - } -unlock_ret: - mutex_unlock(&ep->rma_info.rma_lock); - return err; -} - -static int _scif_fence_wait(scif_epd_t epd, int mark) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - dma_cookie_t cookie = mark & ~SCIF_REMOTE_FENCE; - int err; - - /* Wait for DMA callback in scif_fence_mark_cb(..) */ - err = wait_event_interruptible_timeout(ep->rma_info.markwq, - dma_async_is_tx_complete( - ep->rma_info.dma_chan, - cookie, NULL, NULL) == - DMA_COMPLETE, - SCIF_NODE_ALIVE_TIMEOUT); - if (!err) - err = -ETIMEDOUT; - else if (err > 0) - err = 0; - return err; -} - -/** - * scif_rma_handle_remote_fences: - * - * This routine services remote fence requests. - */ -void scif_rma_handle_remote_fences(void) -{ - struct list_head *item, *tmp; - struct scif_remote_fence_info *fence; - struct scif_endpt *ep; - int mark, err; - - might_sleep(); - mutex_lock(&scif_info.fencelock); - list_for_each_safe(item, tmp, &scif_info.fence) { - fence = list_entry(item, struct scif_remote_fence_info, - list); - /* Remove fence from global list */ - list_del(&fence->list); - - /* Initiate the fence operation */ - ep = (struct scif_endpt *)fence->msg.payload[0]; - mark = fence->msg.payload[2]; - err = _scif_fence_wait(ep, mark); - if (err) - fence->msg.uop = SCIF_WAIT_NACK; - else - fence->msg.uop = SCIF_WAIT_ACK; - fence->msg.payload[0] = ep->remote_ep; - scif_nodeqp_send(ep->remote_dev, &fence->msg); - kfree(fence); - if (!atomic_sub_return(1, &ep->rma_info.fence_refcount)) - schedule_work(&scif_info.misc_work); - } - mutex_unlock(&scif_info.fencelock); -} - -static int _scif_send_fence(scif_epd_t epd, int uop, int mark, int *out_mark) -{ - int err; - struct scifmsg msg; - struct scif_fence_info *fence_req; - struct scif_endpt *ep = (struct scif_endpt *)epd; - - fence_req = kmalloc(sizeof(*fence_req), GFP_KERNEL); - if (!fence_req) { - err = -ENOMEM; - goto error; - } - - fence_req->state = OP_IN_PROGRESS; - init_completion(&fence_req->comp); - - msg.src = ep->port; - msg.uop = uop; - msg.payload[0] = ep->remote_ep; - msg.payload[1] = (u64)fence_req; - if (uop == SCIF_WAIT) - msg.payload[2] = mark; - spin_lock(&ep->lock); - if (ep->state == SCIFEP_CONNECTED) - err = scif_nodeqp_send(ep->remote_dev, &msg); - else - err = -ENOTCONN; - spin_unlock(&ep->lock); - if (err) - goto error_free; -retry: - /* Wait for a SCIF_WAIT_(N)ACK message */ - err = wait_for_completion_timeout(&fence_req->comp, - SCIF_NODE_ALIVE_TIMEOUT); - if (!err && scifdev_alive(ep)) - goto retry; - if (!err) - err = -ENODEV; - if (err > 0) - err = 0; - mutex_lock(&ep->rma_info.rma_lock); - if (err < 0) { - if (fence_req->state == OP_IN_PROGRESS) - fence_req->state = OP_FAILED; - } - if (fence_req->state == OP_FAILED && !err) - err = -ENOMEM; - if (uop == SCIF_MARK && fence_req->state == OP_COMPLETED) - *out_mark = SCIF_REMOTE_FENCE | fence_req->dma_mark; - mutex_unlock(&ep->rma_info.rma_lock); -error_free: - kfree(fence_req); -error: - return err; -} - -/** - * scif_send_fence_mark: - * @epd: end point descriptor. - * @out_mark: Output DMA mark reported by peer. - * - * Send a remote fence mark request. - */ -static int scif_send_fence_mark(scif_epd_t epd, int *out_mark) -{ - return _scif_send_fence(epd, SCIF_MARK, 0, out_mark); -} - -/** - * scif_send_fence_wait: - * @epd: end point descriptor. - * @mark: DMA mark to wait for. - * - * Send a remote fence wait request. - */ -static int scif_send_fence_wait(scif_epd_t epd, int mark) -{ - return _scif_send_fence(epd, SCIF_WAIT, mark, NULL); -} - -static int _scif_send_fence_signal_wait(struct scif_endpt *ep, - struct scif_fence_info *fence_req) -{ - int err; - -retry: - /* Wait for a SCIF_SIG_(N)ACK message */ - err = wait_for_completion_timeout(&fence_req->comp, - SCIF_NODE_ALIVE_TIMEOUT); - if (!err && scifdev_alive(ep)) - goto retry; - if (!err) - err = -ENODEV; - if (err > 0) - err = 0; - if (err < 0) { - mutex_lock(&ep->rma_info.rma_lock); - if (fence_req->state == OP_IN_PROGRESS) - fence_req->state = OP_FAILED; - mutex_unlock(&ep->rma_info.rma_lock); - } - if (fence_req->state == OP_FAILED && !err) - err = -ENXIO; - return err; -} - -/** - * scif_send_fence_signal: - * @epd: endpoint descriptor - * @loff: local offset - * @lval: local value to write to loffset - * @roff: remote offset - * @rval: remote value to write to roffset - * @flags: flags - * - * Sends a remote fence signal request - */ -static int scif_send_fence_signal(scif_epd_t epd, off_t roff, u64 rval, - off_t loff, u64 lval, int flags) -{ - int err = 0; - struct scifmsg msg; - struct scif_fence_info *fence_req; - struct scif_endpt *ep = (struct scif_endpt *)epd; - - fence_req = kmalloc(sizeof(*fence_req), GFP_KERNEL); - if (!fence_req) { - err = -ENOMEM; - goto error; - } - - fence_req->state = OP_IN_PROGRESS; - init_completion(&fence_req->comp); - msg.src = ep->port; - if (flags & SCIF_SIGNAL_LOCAL) { - msg.uop = SCIF_SIG_LOCAL; - msg.payload[0] = ep->remote_ep; - msg.payload[1] = roff; - msg.payload[2] = rval; - msg.payload[3] = (u64)fence_req; - spin_lock(&ep->lock); - if (ep->state == SCIFEP_CONNECTED) - err = scif_nodeqp_send(ep->remote_dev, &msg); - else - err = -ENOTCONN; - spin_unlock(&ep->lock); - if (err) - goto error_free; - err = _scif_send_fence_signal_wait(ep, fence_req); - if (err) - goto error_free; - } - fence_req->state = OP_IN_PROGRESS; - - if (flags & SCIF_SIGNAL_REMOTE) { - msg.uop = SCIF_SIG_REMOTE; - msg.payload[0] = ep->remote_ep; - msg.payload[1] = loff; - msg.payload[2] = lval; - msg.payload[3] = (u64)fence_req; - spin_lock(&ep->lock); - if (ep->state == SCIFEP_CONNECTED) - err = scif_nodeqp_send(ep->remote_dev, &msg); - else - err = -ENOTCONN; - spin_unlock(&ep->lock); - if (err) - goto error_free; - err = _scif_send_fence_signal_wait(ep, fence_req); - } -error_free: - kfree(fence_req); -error: - return err; -} - -static void scif_fence_mark_cb(void *arg) -{ - struct scif_endpt *ep = (struct scif_endpt *)arg; - - wake_up_interruptible(&ep->rma_info.markwq); - atomic_dec(&ep->rma_info.fence_refcount); -} - -/** - * _scif_fence_mark: - * @epd: endpoint descriptor - * @mark: DMA mark to set-up - * - * Set up a mark for this endpoint and return the value of the mark. - */ -int _scif_fence_mark(scif_epd_t epd, int *mark) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - struct dma_chan *chan = ep->rma_info.dma_chan; - struct dma_device *ddev = chan->device; - struct dma_async_tx_descriptor *tx; - dma_cookie_t cookie; - int err; - - tx = ddev->device_prep_dma_memcpy(chan, 0, 0, 0, DMA_PREP_FENCE); - if (!tx) { - err = -ENOMEM; - dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - return err; - } - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - err = (int)cookie; - dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - return err; - } - dma_async_issue_pending(chan); - tx = ddev->device_prep_dma_interrupt(chan, DMA_PREP_INTERRUPT); - if (!tx) { - err = -ENOMEM; - dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - return err; - } - tx->callback = scif_fence_mark_cb; - tx->callback_param = ep; - *mark = cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - err = (int)cookie; - dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - return err; - } - atomic_inc(&ep->rma_info.fence_refcount); - dma_async_issue_pending(chan); - return 0; -} - -#define SCIF_LOOPB_MAGIC_MARK 0xdead - -int scif_fence_mark(scif_epd_t epd, int flags, int *mark) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - int err = 0; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI fence_mark: ep %p flags 0x%x mark 0x%x\n", - ep, flags, *mark); - err = scif_verify_epd(ep); - if (err) - return err; - - /* Invalid flags? */ - if (flags & ~(SCIF_FENCE_INIT_SELF | SCIF_FENCE_INIT_PEER)) - return -EINVAL; - - /* At least one of init self or peer RMA should be set */ - if (!(flags & (SCIF_FENCE_INIT_SELF | SCIF_FENCE_INIT_PEER))) - return -EINVAL; - - /* Exactly one of init self or peer RMA should be set but not both */ - if ((flags & SCIF_FENCE_INIT_SELF) && (flags & SCIF_FENCE_INIT_PEER)) - return -EINVAL; - - /* - * Management node loopback does not need to use DMA. - * Return a valid mark to be symmetric. - */ - if (scifdev_self(ep->remote_dev) && scif_is_mgmt_node()) { - *mark = SCIF_LOOPB_MAGIC_MARK; - return 0; - } - - if (flags & SCIF_FENCE_INIT_SELF) - err = _scif_fence_mark(epd, mark); - else - err = scif_send_fence_mark(ep, mark); - - if (err) - dev_err(scif_info.mdev.this_device, - "%s %d err %d\n", __func__, __LINE__, err); - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI fence_mark: ep %p flags 0x%x mark 0x%x err %d\n", - ep, flags, *mark, err); - return err; -} -EXPORT_SYMBOL_GPL(scif_fence_mark); - -int scif_fence_wait(scif_epd_t epd, int mark) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - int err = 0; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI fence_wait: ep %p mark 0x%x\n", - ep, mark); - err = scif_verify_epd(ep); - if (err) - return err; - /* - * Management node loopback does not need to use DMA. - * The only valid mark provided is 0 so simply - * return success if the mark is valid. - */ - if (scifdev_self(ep->remote_dev) && scif_is_mgmt_node()) { - if (mark == SCIF_LOOPB_MAGIC_MARK) - return 0; - else - return -EINVAL; - } - if (mark & SCIF_REMOTE_FENCE) - err = scif_send_fence_wait(epd, mark); - else - err = _scif_fence_wait(epd, mark); - if (err < 0) - dev_err(scif_info.mdev.this_device, - "%s %d err %d\n", __func__, __LINE__, err); - return err; -} -EXPORT_SYMBOL_GPL(scif_fence_wait); - -int scif_fence_signal(scif_epd_t epd, off_t loff, u64 lval, - off_t roff, u64 rval, int flags) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - int err = 0; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI fence_signal: ep %p loff 0x%lx lval 0x%llx roff 0x%lx rval 0x%llx flags 0x%x\n", - ep, loff, lval, roff, rval, flags); - err = scif_verify_epd(ep); - if (err) - return err; - - /* Invalid flags? */ - if (flags & ~(SCIF_FENCE_INIT_SELF | SCIF_FENCE_INIT_PEER | - SCIF_SIGNAL_LOCAL | SCIF_SIGNAL_REMOTE)) - return -EINVAL; - - /* At least one of init self or peer RMA should be set */ - if (!(flags & (SCIF_FENCE_INIT_SELF | SCIF_FENCE_INIT_PEER))) - return -EINVAL; - - /* Exactly one of init self or peer RMA should be set but not both */ - if ((flags & SCIF_FENCE_INIT_SELF) && (flags & SCIF_FENCE_INIT_PEER)) - return -EINVAL; - - /* At least one of SCIF_SIGNAL_LOCAL or SCIF_SIGNAL_REMOTE required */ - if (!(flags & (SCIF_SIGNAL_LOCAL | SCIF_SIGNAL_REMOTE))) - return -EINVAL; - - /* Only Dword offsets allowed */ - if ((flags & SCIF_SIGNAL_LOCAL) && (loff & (sizeof(u32) - 1))) - return -EINVAL; - - /* Only Dword aligned offsets allowed */ - if ((flags & SCIF_SIGNAL_REMOTE) && (roff & (sizeof(u32) - 1))) - return -EINVAL; - - if (flags & SCIF_FENCE_INIT_PEER) { - err = scif_send_fence_signal(epd, roff, rval, loff, - lval, flags); - } else { - /* Local Signal in Local RAS */ - if (flags & SCIF_SIGNAL_LOCAL) { - err = scif_prog_signal(epd, loff, lval, - SCIF_WINDOW_SELF); - if (err) - goto error_ret; - } - - /* Signal in Remote RAS */ - if (flags & SCIF_SIGNAL_REMOTE) - err = scif_prog_signal(epd, roff, - rval, SCIF_WINDOW_PEER); - } -error_ret: - if (err) - dev_err(scif_info.mdev.this_device, - "%s %d err %d\n", __func__, __LINE__, err); - return err; -} -EXPORT_SYMBOL_GPL(scif_fence_signal); diff --git a/drivers/misc/mic/scif/scif_main.c b/drivers/misc/mic/scif/scif_main.c deleted file mode 100644 index e2278bf9f11d..000000000000 --- a/drivers/misc/mic/scif/scif_main.c +++ /dev/null @@ -1,351 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel SCIF driver. - */ -#include -#include - -#include -#include "../common/mic_dev.h" -#include "../bus/scif_bus.h" -#include "scif_peer_bus.h" -#include "scif_main.h" -#include "scif_map.h" - -struct scif_info scif_info = { - .mdev = { - .minor = MISC_DYNAMIC_MINOR, - .name = "scif", - .fops = &scif_fops, - } -}; - -struct scif_dev *scif_dev; -struct kmem_cache *unaligned_cache; -static atomic_t g_loopb_cnt; - -/* Runs in the context of intr_wq */ -static void scif_intr_bh_handler(struct work_struct *work) -{ - struct scif_dev *scifdev = - container_of(work, struct scif_dev, intr_bh); - - if (scifdev_self(scifdev)) - scif_loopb_msg_handler(scifdev, scifdev->qpairs); - else - scif_nodeqp_intrhandler(scifdev, scifdev->qpairs); -} - -int scif_setup_intr_wq(struct scif_dev *scifdev) -{ - if (!scifdev->intr_wq) { - snprintf(scifdev->intr_wqname, sizeof(scifdev->intr_wqname), - "SCIF INTR %d", scifdev->node); - scifdev->intr_wq = - alloc_ordered_workqueue(scifdev->intr_wqname, 0); - if (!scifdev->intr_wq) - return -ENOMEM; - INIT_WORK(&scifdev->intr_bh, scif_intr_bh_handler); - } - return 0; -} - -void scif_destroy_intr_wq(struct scif_dev *scifdev) -{ - if (scifdev->intr_wq) { - destroy_workqueue(scifdev->intr_wq); - scifdev->intr_wq = NULL; - } -} - -irqreturn_t scif_intr_handler(int irq, void *data) -{ - struct scif_dev *scifdev = data; - struct scif_hw_dev *sdev = scifdev->sdev; - - sdev->hw_ops->ack_interrupt(sdev, scifdev->db); - queue_work(scifdev->intr_wq, &scifdev->intr_bh); - return IRQ_HANDLED; -} - -static void scif_qp_setup_handler(struct work_struct *work) -{ - struct scif_dev *scifdev = container_of(work, struct scif_dev, - qp_dwork.work); - struct scif_hw_dev *sdev = scifdev->sdev; - dma_addr_t da = 0; - int err; - - if (scif_is_mgmt_node()) { - struct mic_bootparam *bp = sdev->dp; - - da = bp->scif_card_dma_addr; - scifdev->rdb = bp->h2c_scif_db; - } else { - struct mic_bootparam __iomem *bp = sdev->rdp; - - da = readq(&bp->scif_host_dma_addr); - scifdev->rdb = ioread8(&bp->c2h_scif_db); - } - if (da) { - err = scif_qp_response(da, scifdev); - if (err) - dev_err(&scifdev->sdev->dev, - "scif_qp_response err %d\n", err); - } else { - schedule_delayed_work(&scifdev->qp_dwork, - msecs_to_jiffies(1000)); - } -} - -static int scif_setup_scifdev(void) -{ - /* We support a maximum of 129 SCIF nodes including the mgmt node */ -#define MAX_SCIF_NODES 129 - int i; - u8 num_nodes = MAX_SCIF_NODES; - - scif_dev = kcalloc(num_nodes, sizeof(*scif_dev), GFP_KERNEL); - if (!scif_dev) - return -ENOMEM; - for (i = 0; i < num_nodes; i++) { - struct scif_dev *scifdev = &scif_dev[i]; - - scifdev->node = i; - scifdev->exit = OP_IDLE; - init_waitqueue_head(&scifdev->disconn_wq); - mutex_init(&scifdev->lock); - INIT_WORK(&scifdev->peer_add_work, scif_add_peer_device); - INIT_DELAYED_WORK(&scifdev->p2p_dwork, - scif_poll_qp_state); - INIT_DELAYED_WORK(&scifdev->qp_dwork, - scif_qp_setup_handler); - INIT_LIST_HEAD(&scifdev->p2p); - RCU_INIT_POINTER(scifdev->spdev, NULL); - } - return 0; -} - -static void scif_destroy_scifdev(void) -{ - kfree(scif_dev); - scif_dev = NULL; -} - -static int scif_probe(struct scif_hw_dev *sdev) -{ - struct scif_dev *scifdev = &scif_dev[sdev->dnode]; - int rc; - - dev_set_drvdata(&sdev->dev, sdev); - scifdev->sdev = sdev; - - if (1 == atomic_add_return(1, &g_loopb_cnt)) { - struct scif_dev *loopb_dev = &scif_dev[sdev->snode]; - - loopb_dev->sdev = sdev; - rc = scif_setup_loopback_qp(loopb_dev); - if (rc) - goto exit; - } - - rc = scif_setup_intr_wq(scifdev); - if (rc) - goto destroy_loopb; - rc = scif_setup_qp(scifdev); - if (rc) - goto destroy_intr; - scifdev->db = sdev->hw_ops->next_db(sdev); - scifdev->cookie = sdev->hw_ops->request_irq(sdev, scif_intr_handler, - "SCIF_INTR", scifdev, - scifdev->db); - if (IS_ERR(scifdev->cookie)) { - rc = PTR_ERR(scifdev->cookie); - goto free_qp; - } - if (scif_is_mgmt_node()) { - struct mic_bootparam *bp = sdev->dp; - - bp->c2h_scif_db = scifdev->db; - bp->scif_host_dma_addr = scifdev->qp_dma_addr; - } else { - struct mic_bootparam __iomem *bp = sdev->rdp; - - iowrite8(scifdev->db, &bp->h2c_scif_db); - writeq(scifdev->qp_dma_addr, &bp->scif_card_dma_addr); - } - schedule_delayed_work(&scifdev->qp_dwork, - msecs_to_jiffies(1000)); - return rc; -free_qp: - scif_free_qp(scifdev); -destroy_intr: - scif_destroy_intr_wq(scifdev); -destroy_loopb: - if (atomic_dec_and_test(&g_loopb_cnt)) - scif_destroy_loopback_qp(&scif_dev[sdev->snode]); -exit: - return rc; -} - -void scif_stop(struct scif_dev *scifdev) -{ - struct scif_dev *dev; - int i; - - for (i = scif_info.maxid; i >= 0; i--) { - dev = &scif_dev[i]; - if (scifdev_self(dev)) - continue; - scif_handle_remove_node(i); - } -} - -static void scif_remove(struct scif_hw_dev *sdev) -{ - struct scif_dev *scifdev = &scif_dev[sdev->dnode]; - - if (scif_is_mgmt_node()) { - struct mic_bootparam *bp = sdev->dp; - - bp->c2h_scif_db = -1; - bp->scif_host_dma_addr = 0x0; - } else { - struct mic_bootparam __iomem *bp = sdev->rdp; - - iowrite8(-1, &bp->h2c_scif_db); - writeq(0x0, &bp->scif_card_dma_addr); - } - if (scif_is_mgmt_node()) { - scif_disconnect_node(scifdev->node, true); - } else { - scif_info.card_initiated_exit = true; - scif_stop(scifdev); - } - if (atomic_dec_and_test(&g_loopb_cnt)) - scif_destroy_loopback_qp(&scif_dev[sdev->snode]); - if (scifdev->cookie) { - sdev->hw_ops->free_irq(sdev, scifdev->cookie, scifdev); - scifdev->cookie = NULL; - } - scif_destroy_intr_wq(scifdev); - cancel_delayed_work(&scifdev->qp_dwork); - scif_free_qp(scifdev); - scifdev->rdb = -1; - scifdev->sdev = NULL; -} - -static struct scif_hw_dev_id id_table[] = { - { MIC_SCIF_DEV, SCIF_DEV_ANY_ID }, - { 0 }, -}; - -static struct scif_driver scif_driver = { - .driver.name = KBUILD_MODNAME, - .driver.owner = THIS_MODULE, - .id_table = id_table, - .probe = scif_probe, - .remove = scif_remove, -}; - -static int _scif_init(void) -{ - int rc; - - mutex_init(&scif_info.eplock); - spin_lock_init(&scif_info.rmalock); - spin_lock_init(&scif_info.nb_connect_lock); - spin_lock_init(&scif_info.port_lock); - mutex_init(&scif_info.conflock); - mutex_init(&scif_info.connlock); - mutex_init(&scif_info.fencelock); - INIT_LIST_HEAD(&scif_info.uaccept); - INIT_LIST_HEAD(&scif_info.listen); - INIT_LIST_HEAD(&scif_info.zombie); - INIT_LIST_HEAD(&scif_info.connected); - INIT_LIST_HEAD(&scif_info.disconnected); - INIT_LIST_HEAD(&scif_info.rma); - INIT_LIST_HEAD(&scif_info.rma_tc); - INIT_LIST_HEAD(&scif_info.mmu_notif_cleanup); - INIT_LIST_HEAD(&scif_info.fence); - INIT_LIST_HEAD(&scif_info.nb_connect_list); - init_waitqueue_head(&scif_info.exitwq); - scif_info.rma_tc_limit = SCIF_RMA_TEMP_CACHE_LIMIT; - scif_info.en_msg_log = 0; - scif_info.p2p_enable = 1; - rc = scif_setup_scifdev(); - if (rc) - goto error; - unaligned_cache = kmem_cache_create("Unaligned_DMA", - SCIF_KMEM_UNALIGNED_BUF_SIZE, - 0, SLAB_HWCACHE_ALIGN, NULL); - if (!unaligned_cache) { - rc = -ENOMEM; - goto free_sdev; - } - INIT_WORK(&scif_info.misc_work, scif_misc_handler); - INIT_WORK(&scif_info.mmu_notif_work, scif_mmu_notif_handler); - INIT_WORK(&scif_info.conn_work, scif_conn_handler); - idr_init(&scif_ports); - return 0; -free_sdev: - scif_destroy_scifdev(); -error: - return rc; -} - -static void _scif_exit(void) -{ - idr_destroy(&scif_ports); - kmem_cache_destroy(unaligned_cache); - scif_destroy_scifdev(); -} - -static int __init scif_init(void) -{ - struct miscdevice *mdev = &scif_info.mdev; - int rc; - - _scif_init(); - iova_cache_get(); - rc = scif_peer_bus_init(); - if (rc) - goto exit; - rc = scif_register_driver(&scif_driver); - if (rc) - goto peer_bus_exit; - rc = misc_register(mdev); - if (rc) - goto unreg_scif; - scif_init_debugfs(); - return 0; -unreg_scif: - scif_unregister_driver(&scif_driver); -peer_bus_exit: - scif_peer_bus_exit(); -exit: - _scif_exit(); - return rc; -} - -static void __exit scif_exit(void) -{ - scif_exit_debugfs(); - misc_deregister(&scif_info.mdev); - scif_unregister_driver(&scif_driver); - scif_peer_bus_exit(); - iova_cache_put(); - _scif_exit(); -} - -module_init(scif_init); -module_exit(scif_exit); - -MODULE_DEVICE_TABLE(scif, id_table); -MODULE_AUTHOR("Intel Corporation"); -MODULE_DESCRIPTION("Intel(R) SCIF driver"); -MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/scif/scif_main.h b/drivers/misc/mic/scif/scif_main.h deleted file mode 100644 index bb3ab97d5b35..000000000000 --- a/drivers/misc/mic/scif/scif_main.h +++ /dev/null @@ -1,274 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel SCIF driver. - */ -#ifndef SCIF_MAIN_H -#define SCIF_MAIN_H - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "../common/mic_dev.h" - -#define SCIF_MGMT_NODE 0 -#define SCIF_DEFAULT_WATCHDOG_TO 30 -#define SCIF_NODE_ACCEPT_TIMEOUT (3 * HZ) -#define SCIF_NODE_ALIVE_TIMEOUT (SCIF_DEFAULT_WATCHDOG_TO * HZ) -#define SCIF_RMA_TEMP_CACHE_LIMIT 0x20000 - -/* - * Generic state used for certain node QP message exchanges - * like Unregister, Alloc etc. - */ -enum scif_msg_state { - OP_IDLE = 1, - OP_IN_PROGRESS, - OP_COMPLETED, - OP_FAILED -}; - -/* - * struct scif_info - Global SCIF information - * - * @nodeid: Node ID this node is to others - * @maxid: Max known node ID - * @total: Total number of SCIF nodes - * @nr_zombies: number of zombie endpoints - * @eplock: Lock to synchronize listening, zombie endpoint lists - * @connlock: Lock to synchronize connected and disconnected lists - * @nb_connect_lock: Synchronize non blocking connect operations - * @port_lock: Synchronize access to SCIF ports - * @uaccept: List of user acceptreq waiting for acceptreg - * @listen: List of listening end points - * @zombie: List of zombie end points with pending RMA's - * @connected: List of end points in connected state - * @disconnected: List of end points in disconnected state - * @nb_connect_list: List for non blocking connections - * @misc_work: miscellaneous SCIF tasks - * @conflock: Lock to synchronize SCIF node configuration changes - * @en_msg_log: Enable debug message logging - * @p2p_enable: Enable P2P SCIF network - * @mdev: The MISC device - * @conn_work: Work for workqueue handling all connections - * @exitwq: Wait queue for waiting for an EXIT node QP message response - * @loopb_dev: Dummy SCIF device used for loopback - * @loopb_wq: Workqueue used for handling loopback messages - * @loopb_wqname[16]: Name of loopback workqueue - * @loopb_work: Used for submitting work to loopb_wq - * @loopb_recv_q: List of messages received on the loopb_wq - * @card_initiated_exit: set when the card has initiated the exit - * @rmalock: Synchronize access to RMA operations - * @fencelock: Synchronize access to list of remote fences requested. - * @rma: List of temporary registered windows to be destroyed. - * @rma_tc: List of temporary registered & cached Windows to be destroyed - * @fence: List of remote fence requests - * @mmu_notif_work: Work for registration caching MMU notifier workqueue - * @mmu_notif_cleanup: List of temporary cached windows for reg cache - * @rma_tc_limit: RMA temporary cache limit - */ -struct scif_info { - u8 nodeid; - u8 maxid; - u8 total; - u32 nr_zombies; - struct mutex eplock; - struct mutex connlock; - spinlock_t nb_connect_lock; - spinlock_t port_lock; - struct list_head uaccept; - struct list_head listen; - struct list_head zombie; - struct list_head connected; - struct list_head disconnected; - struct list_head nb_connect_list; - struct work_struct misc_work; - struct mutex conflock; - u8 en_msg_log; - u8 p2p_enable; - struct miscdevice mdev; - struct work_struct conn_work; - wait_queue_head_t exitwq; - struct scif_dev *loopb_dev; - struct workqueue_struct *loopb_wq; - char loopb_wqname[16]; - struct work_struct loopb_work; - struct list_head loopb_recv_q; - bool card_initiated_exit; - spinlock_t rmalock; - struct mutex fencelock; - struct list_head rma; - struct list_head rma_tc; - struct list_head fence; - struct work_struct mmu_notif_work; - struct list_head mmu_notif_cleanup; - unsigned long rma_tc_limit; -}; - -/* - * struct scif_p2p_info - SCIF mapping information used for P2P - * - * @ppi_peer_id - SCIF peer node id - * @ppi_sg - Scatter list for bar information (One for mmio and one for aper) - * @sg_nentries - Number of entries in the scatterlist - * @ppi_da: DMA address for MMIO and APER bars - * @ppi_len: Length of MMIO and APER bars - * @ppi_list: Link in list of mapping information - */ -struct scif_p2p_info { - u8 ppi_peer_id; - struct scatterlist *ppi_sg[2]; - u64 sg_nentries[2]; - dma_addr_t ppi_da[2]; - u64 ppi_len[2]; -#define SCIF_PPI_MMIO 0 -#define SCIF_PPI_APER 1 - struct list_head ppi_list; -}; - -/* - * struct scif_dev - SCIF remote device specific fields - * - * @node: Node id - * @p2p: List of P2P mapping information - * @qpairs: The node queue pair for exchanging control messages - * @intr_wq: Workqueue for handling Node QP messages - * @intr_wqname: Name of node QP workqueue for handling interrupts - * @intr_bh: Used for submitting work to intr_wq - * @lock: Lock used for synchronizing access to the scif device - * @sdev: SCIF hardware device on the SCIF hardware bus - * @db: doorbell the peer will trigger to generate an interrupt on self - * @rdb: Doorbell to trigger on the peer to generate an interrupt on the peer - * @cookie: Cookie received while registering the interrupt handler - * @peer_add_work: Work for handling device_add for peer devices - * @p2p_dwork: Delayed work to enable polling for P2P state - * @qp_dwork: Delayed work for enabling polling for remote QP information - * @p2p_retry: Number of times to retry polling of P2P state - * @base_addr: P2P aperture bar base address - * @mic_mw mmio: The peer MMIO information used for P2P - * @spdev: SCIF peer device on the SCIF peer bus - * @node_remove_ack_pending: True if a node_remove_ack is pending - * @exit_ack_pending: true if an exit_ack is pending - * @disconn_wq: Used while waiting for a node remove response - * @disconn_rescnt: Keeps track of number of node remove requests sent - * @exit: Status of exit message - * @qp_dma_addr: Queue pair DMA address passed to the peer - * @dma_ch_idx: Round robin index for DMA channels - * @signal_pool: DMA pool used for scheduling scif_fence_signal DMA's -*/ -struct scif_dev { - u8 node; - struct list_head p2p; - struct scif_qp *qpairs; - struct workqueue_struct *intr_wq; - char intr_wqname[16]; - struct work_struct intr_bh; - struct mutex lock; - struct scif_hw_dev *sdev; - int db; - int rdb; - struct mic_irq *cookie; - struct work_struct peer_add_work; - struct delayed_work p2p_dwork; - struct delayed_work qp_dwork; - int p2p_retry; - dma_addr_t base_addr; - struct mic_mw mmio; - struct scif_peer_dev __rcu *spdev; - bool node_remove_ack_pending; - bool exit_ack_pending; - wait_queue_head_t disconn_wq; - atomic_t disconn_rescnt; - enum scif_msg_state exit; - dma_addr_t qp_dma_addr; - int dma_ch_idx; - struct dma_pool *signal_pool; -}; - -extern bool scif_reg_cache_enable; -extern bool scif_ulimit_check; -extern struct scif_info scif_info; -extern struct idr scif_ports; -extern struct bus_type scif_peer_bus; -extern struct scif_dev *scif_dev; -extern const struct file_operations scif_fops; -extern const struct file_operations scif_anon_fops; - -/* Size of the RB for the Node QP */ -#define SCIF_NODE_QP_SIZE 0x10000 - -#include "scif_nodeqp.h" -#include "scif_rma.h" -#include "scif_rma_list.h" - -/* - * scifdev_self: - * @dev: The remote SCIF Device - * - * Returns true if the SCIF Device passed is the self aka Loopback SCIF device. - */ -static inline int scifdev_self(struct scif_dev *dev) -{ - return dev->node == scif_info.nodeid; -} - -static inline bool scif_is_mgmt_node(void) -{ - return !scif_info.nodeid; -} - -/* - * scifdev_is_p2p: - * @dev: The remote SCIF Device - * - * Returns true if the SCIF Device is a MIC Peer to Peer SCIF device. - */ -static inline bool scifdev_is_p2p(struct scif_dev *dev) -{ - if (scif_is_mgmt_node()) - return false; - else - return dev != &scif_dev[SCIF_MGMT_NODE] && - !scifdev_self(dev); -} - -/* - * scifdev_alive: - * @scifdev: The remote SCIF Device - * - * Returns true if the remote SCIF Device is running or sleeping for - * this endpoint. - */ -static inline int _scifdev_alive(struct scif_dev *scifdev) -{ - struct scif_peer_dev *spdev; - - rcu_read_lock(); - spdev = rcu_dereference(scifdev->spdev); - rcu_read_unlock(); - return !!spdev; -} - -#include "scif_epd.h" - -void __init scif_init_debugfs(void); -void scif_exit_debugfs(void); -int scif_setup_intr_wq(struct scif_dev *scifdev); -void scif_destroy_intr_wq(struct scif_dev *scifdev); -void scif_cleanup_scifdev(struct scif_dev *dev); -void scif_handle_remove_node(int node); -void scif_disconnect_node(u32 node_id, bool mgmt_initiated); -void scif_free_qp(struct scif_dev *dev); -void scif_misc_handler(struct work_struct *work); -void scif_stop(struct scif_dev *scifdev); -irqreturn_t scif_intr_handler(int irq, void *data); -#endif /* SCIF_MAIN_H */ diff --git a/drivers/misc/mic/scif/scif_map.h b/drivers/misc/mic/scif/scif_map.h deleted file mode 100644 index 96b760819bfc..000000000000 --- a/drivers/misc/mic/scif/scif_map.h +++ /dev/null @@ -1,127 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel SCIF driver. - */ -#ifndef SCIF_MAP_H -#define SCIF_MAP_H - -#include "../bus/scif_bus.h" - -static __always_inline void * -scif_alloc_coherent(dma_addr_t *dma_handle, - struct scif_dev *scifdev, size_t size, - gfp_t gfp) -{ - void *va; - - if (scifdev_self(scifdev)) { - va = kmalloc(size, gfp); - if (va) - *dma_handle = virt_to_phys(va); - } else { - va = dma_alloc_coherent(&scifdev->sdev->dev, - size, dma_handle, gfp); - if (va && scifdev_is_p2p(scifdev)) - *dma_handle = *dma_handle + scifdev->base_addr; - } - return va; -} - -static __always_inline void -scif_free_coherent(void *va, dma_addr_t local, - struct scif_dev *scifdev, size_t size) -{ - if (scifdev_self(scifdev)) { - kfree(va); - } else { - if (scifdev_is_p2p(scifdev) && local > scifdev->base_addr) - local = local - scifdev->base_addr; - dma_free_coherent(&scifdev->sdev->dev, - size, va, local); - } -} - -static __always_inline int -scif_map_single(dma_addr_t *dma_handle, - void *local, struct scif_dev *scifdev, size_t size) -{ - int err = 0; - - if (scifdev_self(scifdev)) { - *dma_handle = virt_to_phys((local)); - } else { - *dma_handle = dma_map_single(&scifdev->sdev->dev, - local, size, DMA_BIDIRECTIONAL); - if (dma_mapping_error(&scifdev->sdev->dev, *dma_handle)) - err = -ENOMEM; - else if (scifdev_is_p2p(scifdev)) - *dma_handle = *dma_handle + scifdev->base_addr; - } - if (err) - *dma_handle = 0; - return err; -} - -static __always_inline void -scif_unmap_single(dma_addr_t local, struct scif_dev *scifdev, - size_t size) -{ - if (!scifdev_self(scifdev)) { - if (scifdev_is_p2p(scifdev)) - local = local - scifdev->base_addr; - dma_unmap_single(&scifdev->sdev->dev, local, - size, DMA_BIDIRECTIONAL); - } -} - -static __always_inline void * -scif_ioremap(dma_addr_t phys, size_t size, struct scif_dev *scifdev) -{ - void *out_virt; - struct scif_hw_dev *sdev = scifdev->sdev; - - if (scifdev_self(scifdev)) - out_virt = phys_to_virt(phys); - else - out_virt = (void __force *) - sdev->hw_ops->remap(sdev, phys, size); - return out_virt; -} - -static __always_inline void -scif_iounmap(void *virt, size_t len, struct scif_dev *scifdev) -{ - if (!scifdev_self(scifdev)) { - struct scif_hw_dev *sdev = scifdev->sdev; - - sdev->hw_ops->unmap(sdev, (void __force __iomem *)virt); - } -} - -static __always_inline int -scif_map_page(dma_addr_t *dma_handle, struct page *page, - struct scif_dev *scifdev) -{ - int err = 0; - - if (scifdev_self(scifdev)) { - *dma_handle = page_to_phys(page); - } else { - struct scif_hw_dev *sdev = scifdev->sdev; - *dma_handle = dma_map_page(&sdev->dev, - page, 0x0, PAGE_SIZE, - DMA_BIDIRECTIONAL); - if (dma_mapping_error(&sdev->dev, *dma_handle)) - err = -ENOMEM; - else if (scifdev_is_p2p(scifdev)) - *dma_handle = *dma_handle + scifdev->base_addr; - } - if (err) - *dma_handle = 0; - return err; -} -#endif /* SCIF_MAP_H */ diff --git a/drivers/misc/mic/scif/scif_mmap.c b/drivers/misc/mic/scif/scif_mmap.c deleted file mode 100644 index a151d416f39c..000000000000 --- a/drivers/misc/mic/scif/scif_mmap.c +++ /dev/null @@ -1,690 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2015 Intel Corporation. - * - * Intel SCIF driver. - */ -#include "scif_main.h" - -/* - * struct scif_vma_info - Information about a remote memory mapping - * created via scif_mmap(..) - * @vma: VM area struct - * @list: link to list of active vmas - */ -struct scif_vma_info { - struct vm_area_struct *vma; - struct list_head list; -}; - -void scif_recv_munmap(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_rma_req req; - struct scif_window *window = NULL; - struct scif_window *recv_window = - (struct scif_window *)msg->payload[0]; - struct scif_endpt *ep; - - ep = (struct scif_endpt *)recv_window->ep; - req.out_window = &window; - req.offset = recv_window->offset; - req.prot = recv_window->prot; - req.nr_bytes = recv_window->nr_pages << PAGE_SHIFT; - req.type = SCIF_WINDOW_FULL; - req.head = &ep->rma_info.reg_list; - msg->payload[0] = ep->remote_ep; - - mutex_lock(&ep->rma_info.rma_lock); - /* Does a valid window exist? */ - if (scif_query_window(&req)) { - dev_err(&scifdev->sdev->dev, - "%s %d -ENXIO\n", __func__, __LINE__); - msg->uop = SCIF_UNREGISTER_ACK; - goto error; - } - - scif_put_window(window, window->nr_pages); - - if (!window->ref_count) { - atomic_inc(&ep->rma_info.tw_refcount); - ep->rma_info.async_list_del = 1; - list_del_init(&window->list); - scif_free_window_offset(ep, window, window->offset); - } -error: - mutex_unlock(&ep->rma_info.rma_lock); - if (window && !window->ref_count) - scif_queue_for_cleanup(window, &scif_info.rma); -} - -/* - * Remove valid remote memory mappings created via scif_mmap(..) from the - * process address space since the remote node is lost - */ -static void __scif_zap_mmaps(struct scif_endpt *ep) -{ - struct list_head *item; - struct scif_vma_info *info; - struct vm_area_struct *vma; - unsigned long size; - - spin_lock(&ep->lock); - list_for_each(item, &ep->rma_info.vma_list) { - info = list_entry(item, struct scif_vma_info, list); - vma = info->vma; - size = vma->vm_end - vma->vm_start; - zap_vma_ptes(vma, vma->vm_start, size); - dev_dbg(scif_info.mdev.this_device, - "%s ep %p zap vma %p size 0x%lx\n", - __func__, ep, info->vma, size); - } - spin_unlock(&ep->lock); -} - -/* - * Traverse the list of endpoints for a particular remote node and - * zap valid remote memory mappings since the remote node is lost - */ -static void _scif_zap_mmaps(int node, struct list_head *head) -{ - struct scif_endpt *ep; - struct list_head *item; - - mutex_lock(&scif_info.connlock); - list_for_each(item, head) { - ep = list_entry(item, struct scif_endpt, list); - if (ep->remote_dev->node == node) - __scif_zap_mmaps(ep); - } - mutex_unlock(&scif_info.connlock); -} - -/* - * Wrapper for removing remote memory mappings for a particular node. This API - * is called by peer nodes as part of handling a lost node. - */ -void scif_zap_mmaps(int node) -{ - _scif_zap_mmaps(node, &scif_info.connected); - _scif_zap_mmaps(node, &scif_info.disconnected); -} - -/* - * This API is only called while handling a lost node: - * a) Remote node is dead. - * b) Remote memory mappings have been zapped - * So we can traverse the remote_reg_list without any locks. Since - * the window has not yet been unregistered we can drop the ref count - * and queue it to the cleanup thread. - */ -static void __scif_cleanup_rma_for_zombies(struct scif_endpt *ep) -{ - struct list_head *pos, *tmp; - struct scif_window *window; - - list_for_each_safe(pos, tmp, &ep->rma_info.remote_reg_list) { - window = list_entry(pos, struct scif_window, list); - if (window->ref_count) - scif_put_window(window, window->nr_pages); - else - dev_err(scif_info.mdev.this_device, - "%s %d unexpected\n", - __func__, __LINE__); - if (!window->ref_count) { - atomic_inc(&ep->rma_info.tw_refcount); - list_del_init(&window->list); - scif_queue_for_cleanup(window, &scif_info.rma); - } - } -} - -/* Cleanup remote registration lists for zombie endpoints */ -void scif_cleanup_rma_for_zombies(int node) -{ - struct scif_endpt *ep; - struct list_head *item; - - mutex_lock(&scif_info.eplock); - list_for_each(item, &scif_info.zombie) { - ep = list_entry(item, struct scif_endpt, list); - if (ep->remote_dev && ep->remote_dev->node == node) - __scif_cleanup_rma_for_zombies(ep); - } - mutex_unlock(&scif_info.eplock); - flush_work(&scif_info.misc_work); -} - -/* Insert the VMA into the per endpoint VMA list */ -static int scif_insert_vma(struct scif_endpt *ep, struct vm_area_struct *vma) -{ - struct scif_vma_info *info; - int err = 0; - - info = kzalloc(sizeof(*info), GFP_KERNEL); - if (!info) { - err = -ENOMEM; - goto done; - } - info->vma = vma; - spin_lock(&ep->lock); - list_add_tail(&info->list, &ep->rma_info.vma_list); - spin_unlock(&ep->lock); -done: - return err; -} - -/* Delete the VMA from the per endpoint VMA list */ -static void scif_delete_vma(struct scif_endpt *ep, struct vm_area_struct *vma) -{ - struct list_head *item; - struct scif_vma_info *info; - - spin_lock(&ep->lock); - list_for_each(item, &ep->rma_info.vma_list) { - info = list_entry(item, struct scif_vma_info, list); - if (info->vma == vma) { - list_del(&info->list); - kfree(info); - break; - } - } - spin_unlock(&ep->lock); -} - -static phys_addr_t scif_get_phys(phys_addr_t phys, struct scif_endpt *ep) -{ - struct scif_dev *scifdev = (struct scif_dev *)ep->remote_dev; - struct scif_hw_dev *sdev = scifdev->sdev; - phys_addr_t out_phys, apt_base = 0; - - /* - * If the DMA address is card relative then we need to add the - * aperture base for mmap to work correctly - */ - if (!scifdev_self(scifdev) && sdev->aper && sdev->card_rel_da) - apt_base = sdev->aper->pa; - out_phys = apt_base + phys; - return out_phys; -} - -int scif_get_pages(scif_epd_t epd, off_t offset, size_t len, - struct scif_range **pages) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - struct scif_rma_req req; - struct scif_window *window = NULL; - int nr_pages, err, i; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI get_pinned_pages: ep %p offset 0x%lx len 0x%lx\n", - ep, offset, len); - err = scif_verify_epd(ep); - if (err) - return err; - - if (!len || (offset < 0) || - (offset + len < offset) || - (ALIGN(offset, PAGE_SIZE) != offset) || - (ALIGN(len, PAGE_SIZE) != len)) - return -EINVAL; - - nr_pages = len >> PAGE_SHIFT; - - req.out_window = &window; - req.offset = offset; - req.prot = 0; - req.nr_bytes = len; - req.type = SCIF_WINDOW_SINGLE; - req.head = &ep->rma_info.remote_reg_list; - - mutex_lock(&ep->rma_info.rma_lock); - /* Does a valid window exist? */ - err = scif_query_window(&req); - if (err) { - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, err); - goto error; - } - - /* Allocate scif_range */ - *pages = kzalloc(sizeof(**pages), GFP_KERNEL); - if (!*pages) { - err = -ENOMEM; - goto error; - } - - /* Allocate phys addr array */ - (*pages)->phys_addr = scif_zalloc(nr_pages * sizeof(dma_addr_t)); - if (!((*pages)->phys_addr)) { - err = -ENOMEM; - goto error; - } - - if (scif_is_mgmt_node() && !scifdev_self(ep->remote_dev)) { - /* Allocate virtual address array */ - ((*pages)->va = scif_zalloc(nr_pages * sizeof(void *))); - if (!(*pages)->va) { - err = -ENOMEM; - goto error; - } - } - /* Populate the values */ - (*pages)->cookie = window; - (*pages)->nr_pages = nr_pages; - (*pages)->prot_flags = window->prot; - - for (i = 0; i < nr_pages; i++) { - (*pages)->phys_addr[i] = - __scif_off_to_dma_addr(window, offset + - (i * PAGE_SIZE)); - (*pages)->phys_addr[i] = scif_get_phys((*pages)->phys_addr[i], - ep); - if (scif_is_mgmt_node() && !scifdev_self(ep->remote_dev)) - (*pages)->va[i] = - ep->remote_dev->sdev->aper->va + - (*pages)->phys_addr[i] - - ep->remote_dev->sdev->aper->pa; - } - - scif_get_window(window, nr_pages); -error: - mutex_unlock(&ep->rma_info.rma_lock); - if (err) { - if (*pages) { - scif_free((*pages)->phys_addr, - nr_pages * sizeof(dma_addr_t)); - scif_free((*pages)->va, - nr_pages * sizeof(void *)); - kfree(*pages); - *pages = NULL; - } - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, err); - } - return err; -} -EXPORT_SYMBOL_GPL(scif_get_pages); - -int scif_put_pages(struct scif_range *pages) -{ - struct scif_endpt *ep; - struct scif_window *window; - struct scifmsg msg; - - if (!pages || !pages->cookie) - return -EINVAL; - - window = pages->cookie; - - if (!window || window->magic != SCIFEP_MAGIC) - return -EINVAL; - - ep = (struct scif_endpt *)window->ep; - /* - * If the state is SCIFEP_CONNECTED or SCIFEP_DISCONNECTED then the - * callee should be allowed to release references to the pages, - * else the endpoint was not connected in the first place, - * hence the ENOTCONN. - */ - if (ep->state != SCIFEP_CONNECTED && ep->state != SCIFEP_DISCONNECTED) - return -ENOTCONN; - - mutex_lock(&ep->rma_info.rma_lock); - - scif_put_window(window, pages->nr_pages); - - /* Initiate window destruction if ref count is zero */ - if (!window->ref_count) { - list_del(&window->list); - mutex_unlock(&ep->rma_info.rma_lock); - scif_drain_dma_intr(ep->remote_dev->sdev, - ep->rma_info.dma_chan); - /* Inform the peer about this window being destroyed. */ - msg.uop = SCIF_MUNMAP; - msg.src = ep->port; - msg.payload[0] = window->peer_window; - /* No error handling for notification messages */ - scif_nodeqp_send(ep->remote_dev, &msg); - /* Destroy this window from the peer's registered AS */ - scif_destroy_remote_window(window); - } else { - mutex_unlock(&ep->rma_info.rma_lock); - } - - scif_free(pages->phys_addr, pages->nr_pages * sizeof(dma_addr_t)); - scif_free(pages->va, pages->nr_pages * sizeof(void *)); - kfree(pages); - return 0; -} -EXPORT_SYMBOL_GPL(scif_put_pages); - -/* - * scif_rma_list_mmap: - * - * Traverse the remote registration list starting from start_window: - * 1) Create VtoP mappings via remap_pfn_range(..) - * 2) Once step 1) and 2) complete successfully then traverse the range of - * windows again and bump the reference count. - * RMA lock must be held. - */ -static int scif_rma_list_mmap(struct scif_window *start_window, s64 offset, - int nr_pages, struct vm_area_struct *vma) -{ - s64 end_offset, loop_offset = offset; - struct scif_window *window = start_window; - int loop_nr_pages, nr_pages_left = nr_pages; - struct scif_endpt *ep = (struct scif_endpt *)start_window->ep; - struct list_head *head = &ep->rma_info.remote_reg_list; - int i, err = 0; - dma_addr_t phys_addr; - struct scif_window_iter src_win_iter; - size_t contig_bytes = 0; - - might_sleep(); - list_for_each_entry_from(window, head, list) { - end_offset = window->offset + - (window->nr_pages << PAGE_SHIFT); - loop_nr_pages = min_t(int, - (end_offset - loop_offset) >> PAGE_SHIFT, - nr_pages_left); - scif_init_window_iter(window, &src_win_iter); - for (i = 0; i < loop_nr_pages; i++) { - phys_addr = scif_off_to_dma_addr(window, loop_offset, - &contig_bytes, - &src_win_iter); - phys_addr = scif_get_phys(phys_addr, ep); - err = remap_pfn_range(vma, - vma->vm_start + - loop_offset - offset, - phys_addr >> PAGE_SHIFT, - PAGE_SIZE, - vma->vm_page_prot); - if (err) - goto error; - loop_offset += PAGE_SIZE; - } - nr_pages_left -= loop_nr_pages; - if (!nr_pages_left) - break; - } - /* - * No more failures expected. Bump up the ref count for all - * the windows. Another traversal from start_window required - * for handling errors encountered across windows during - * remap_pfn_range(..). - */ - loop_offset = offset; - nr_pages_left = nr_pages; - window = start_window; - head = &ep->rma_info.remote_reg_list; - list_for_each_entry_from(window, head, list) { - end_offset = window->offset + - (window->nr_pages << PAGE_SHIFT); - loop_nr_pages = min_t(int, - (end_offset - loop_offset) >> PAGE_SHIFT, - nr_pages_left); - scif_get_window(window, loop_nr_pages); - nr_pages_left -= loop_nr_pages; - loop_offset += (loop_nr_pages << PAGE_SHIFT); - if (!nr_pages_left) - break; - } -error: - if (err) - dev_err(scif_info.mdev.this_device, - "%s %d err %d\n", __func__, __LINE__, err); - return err; -} - -/* - * scif_rma_list_munmap: - * - * Traverse the remote registration list starting from window: - * 1) Decrement ref count. - * 2) If the ref count drops to zero then send a SCIF_MUNMAP message to peer. - * RMA lock must be held. - */ -static void scif_rma_list_munmap(struct scif_window *start_window, - s64 offset, int nr_pages) -{ - struct scifmsg msg; - s64 loop_offset = offset, end_offset; - int loop_nr_pages, nr_pages_left = nr_pages; - struct scif_endpt *ep = (struct scif_endpt *)start_window->ep; - struct list_head *head = &ep->rma_info.remote_reg_list; - struct scif_window *window = start_window, *_window; - - msg.uop = SCIF_MUNMAP; - msg.src = ep->port; - loop_offset = offset; - nr_pages_left = nr_pages; - list_for_each_entry_safe_from(window, _window, head, list) { - end_offset = window->offset + - (window->nr_pages << PAGE_SHIFT); - loop_nr_pages = min_t(int, - (end_offset - loop_offset) >> PAGE_SHIFT, - nr_pages_left); - scif_put_window(window, loop_nr_pages); - if (!window->ref_count) { - struct scif_dev *rdev = ep->remote_dev; - - scif_drain_dma_intr(rdev->sdev, - ep->rma_info.dma_chan); - /* Inform the peer about this munmap */ - msg.payload[0] = window->peer_window; - /* No error handling for Notification messages. */ - scif_nodeqp_send(ep->remote_dev, &msg); - list_del(&window->list); - /* Destroy this window from the peer's registered AS */ - scif_destroy_remote_window(window); - } - nr_pages_left -= loop_nr_pages; - loop_offset += (loop_nr_pages << PAGE_SHIFT); - if (!nr_pages_left) - break; - } -} - -/* - * The private data field of each VMA used to mmap a remote window - * points to an instance of struct vma_pvt - */ -struct vma_pvt { - struct scif_endpt *ep; /* End point for remote window */ - s64 offset; /* offset within remote window */ - bool valid_offset; /* offset is valid only if the original - * mmap request was for a single page - * else the offset within the vma is - * the correct offset - */ - struct kref ref; -}; - -static void vma_pvt_release(struct kref *ref) -{ - struct vma_pvt *vmapvt = container_of(ref, struct vma_pvt, ref); - - kfree(vmapvt); -} - -/** - * scif_vma_open - VMA open driver callback - * @vma: VMM memory area. - * The open method is called by the kernel to allow the subsystem implementing - * the VMA to initialize the area. This method is invoked any time a new - * reference to the VMA is made (when a process forks, for example). - * The one exception happens when the VMA is first created by mmap; - * in this case, the driver's mmap method is called instead. - * This function is also invoked when an existing VMA is split by the kernel - * due to a call to munmap on a subset of the VMA resulting in two VMAs. - * The kernel invokes this function only on one of the two VMAs. - */ -static void scif_vma_open(struct vm_area_struct *vma) -{ - struct vma_pvt *vmapvt = vma->vm_private_data; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI vma open: vma_start 0x%lx vma_end 0x%lx\n", - vma->vm_start, vma->vm_end); - scif_insert_vma(vmapvt->ep, vma); - kref_get(&vmapvt->ref); -} - -/** - * scif_munmap - VMA close driver callback. - * @vma: VMM memory area. - * When an area is destroyed, the kernel calls its close operation. - * Note that there's no usage count associated with VMA's; the area - * is opened and closed exactly once by each process that uses it. - */ -static void scif_munmap(struct vm_area_struct *vma) -{ - struct scif_endpt *ep; - struct vma_pvt *vmapvt = vma->vm_private_data; - int nr_pages = vma_pages(vma); - s64 offset; - struct scif_rma_req req; - struct scif_window *window = NULL; - int err; - - might_sleep(); - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI munmap: vma_start 0x%lx vma_end 0x%lx\n", - vma->vm_start, vma->vm_end); - ep = vmapvt->ep; - offset = vmapvt->valid_offset ? vmapvt->offset : - (vma->vm_pgoff) << PAGE_SHIFT; - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI munmap: ep %p nr_pages 0x%x offset 0x%llx\n", - ep, nr_pages, offset); - req.out_window = &window; - req.offset = offset; - req.nr_bytes = vma->vm_end - vma->vm_start; - req.prot = vma->vm_flags & (VM_READ | VM_WRITE); - req.type = SCIF_WINDOW_PARTIAL; - req.head = &ep->rma_info.remote_reg_list; - - mutex_lock(&ep->rma_info.rma_lock); - - err = scif_query_window(&req); - if (err) - dev_err(scif_info.mdev.this_device, - "%s %d err %d\n", __func__, __LINE__, err); - else - scif_rma_list_munmap(window, offset, nr_pages); - - mutex_unlock(&ep->rma_info.rma_lock); - /* - * The kernel probably zeroes these out but we still want - * to clean up our own mess just in case. - */ - vma->vm_ops = NULL; - vma->vm_private_data = NULL; - kref_put(&vmapvt->ref, vma_pvt_release); - scif_delete_vma(ep, vma); -} - -static const struct vm_operations_struct scif_vm_ops = { - .open = scif_vma_open, - .close = scif_munmap, -}; - -/** - * scif_mmap - Map pages in virtual address space to a remote window. - * @vma: VMM memory area. - * @epd: endpoint descriptor - * - * Return: Upon successful completion, scif_mmap() returns zero - * else an apt error is returned as documented in scif.h - */ -int scif_mmap(struct vm_area_struct *vma, scif_epd_t epd) -{ - struct scif_rma_req req; - struct scif_window *window = NULL; - struct scif_endpt *ep = (struct scif_endpt *)epd; - s64 start_offset = vma->vm_pgoff << PAGE_SHIFT; - int nr_pages = vma_pages(vma); - int err; - struct vma_pvt *vmapvt; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI mmap: ep %p start_offset 0x%llx nr_pages 0x%x\n", - ep, start_offset, nr_pages); - err = scif_verify_epd(ep); - if (err) - return err; - - might_sleep(); - - err = scif_insert_vma(ep, vma); - if (err) - return err; - - vmapvt = kzalloc(sizeof(*vmapvt), GFP_KERNEL); - if (!vmapvt) { - scif_delete_vma(ep, vma); - return -ENOMEM; - } - - vmapvt->ep = ep; - kref_init(&vmapvt->ref); - - req.out_window = &window; - req.offset = start_offset; - req.nr_bytes = vma->vm_end - vma->vm_start; - req.prot = vma->vm_flags & (VM_READ | VM_WRITE); - req.type = SCIF_WINDOW_PARTIAL; - req.head = &ep->rma_info.remote_reg_list; - - mutex_lock(&ep->rma_info.rma_lock); - /* Does a valid window exist? */ - err = scif_query_window(&req); - if (err) { - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, err); - goto error_unlock; - } - - /* Default prot for loopback */ - if (!scifdev_self(ep->remote_dev)) - vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); - - /* - * VM_DONTCOPY - Do not copy this vma on fork - * VM_DONTEXPAND - Cannot expand with mremap() - * VM_RESERVED - Count as reserved_vm like IO - * VM_PFNMAP - Page-ranges managed without "struct page" - * VM_IO - Memory mapped I/O or similar - * - * We do not want to copy this VMA automatically on a fork(), - * expand this VMA due to mremap() or swap out these pages since - * the VMA is actually backed by physical pages in the remote - * node's physical memory and not via a struct page. - */ - vma->vm_flags |= VM_DONTCOPY | VM_DONTEXPAND | VM_DONTDUMP; - - if (!scifdev_self(ep->remote_dev)) - vma->vm_flags |= VM_IO | VM_PFNMAP; - - /* Map this range of windows */ - err = scif_rma_list_mmap(window, start_offset, nr_pages, vma); - if (err) { - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, err); - goto error_unlock; - } - /* Set up the driver call back */ - vma->vm_ops = &scif_vm_ops; - vma->vm_private_data = vmapvt; -error_unlock: - mutex_unlock(&ep->rma_info.rma_lock); - if (err) { - kfree(vmapvt); - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, err); - scif_delete_vma(ep, vma); - } - return err; -} diff --git a/drivers/misc/mic/scif/scif_nm.c b/drivers/misc/mic/scif/scif_nm.c deleted file mode 100644 index c4d9422082b7..000000000000 --- a/drivers/misc/mic/scif/scif_nm.c +++ /dev/null @@ -1,229 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel SCIF driver. - */ -#include "scif_peer_bus.h" - -#include "scif_main.h" -#include "scif_map.h" - -/** - * scif_invalidate_ep() - Set state for all connected endpoints - * to disconnected and wake up all send/recv waitqueues - * - * @node: Node to invalidate - */ -static void scif_invalidate_ep(int node) -{ - struct scif_endpt *ep; - struct list_head *pos, *tmpq; - - flush_work(&scif_info.conn_work); - mutex_lock(&scif_info.connlock); - list_for_each_safe(pos, tmpq, &scif_info.disconnected) { - ep = list_entry(pos, struct scif_endpt, list); - if (ep->remote_dev->node == node) { - scif_unmap_all_windows(ep); - spin_lock(&ep->lock); - scif_cleanup_ep_qp(ep); - spin_unlock(&ep->lock); - } - } - list_for_each_safe(pos, tmpq, &scif_info.connected) { - ep = list_entry(pos, struct scif_endpt, list); - if (ep->remote_dev->node == node) { - list_del(pos); - spin_lock(&ep->lock); - ep->state = SCIFEP_DISCONNECTED; - list_add_tail(&ep->list, &scif_info.disconnected); - scif_cleanup_ep_qp(ep); - wake_up_interruptible(&ep->sendwq); - wake_up_interruptible(&ep->recvwq); - spin_unlock(&ep->lock); - scif_unmap_all_windows(ep); - } - } - mutex_unlock(&scif_info.connlock); -} - -void scif_free_qp(struct scif_dev *scifdev) -{ - struct scif_qp *qp = scifdev->qpairs; - - if (!qp) - return; - scif_unmap_single(qp->local_buf, scifdev, qp->inbound_q.size); - kfree(qp->inbound_q.rb_base); - scif_unmap_single(qp->local_qp, scifdev, sizeof(struct scif_qp)); - kfree(scifdev->qpairs); - scifdev->qpairs = NULL; -} - -static void scif_cleanup_qp(struct scif_dev *dev) -{ - struct scif_qp *qp = &dev->qpairs[0]; - - if (!qp) - return; - scif_iounmap((void *)qp->remote_qp, sizeof(struct scif_qp), dev); - scif_iounmap((void *)qp->outbound_q.rb_base, - sizeof(struct scif_qp), dev); - qp->remote_qp = NULL; - qp->local_write = 0; - qp->inbound_q.current_write_offset = 0; - qp->inbound_q.current_read_offset = 0; - if (scifdev_is_p2p(dev)) - scif_free_qp(dev); -} - -void scif_send_acks(struct scif_dev *dev) -{ - struct scifmsg msg; - - if (dev->node_remove_ack_pending) { - msg.uop = SCIF_NODE_REMOVE_ACK; - msg.src.node = scif_info.nodeid; - msg.dst.node = SCIF_MGMT_NODE; - msg.payload[0] = dev->node; - scif_nodeqp_send(&scif_dev[SCIF_MGMT_NODE], &msg); - dev->node_remove_ack_pending = false; - } - if (dev->exit_ack_pending) { - msg.uop = SCIF_EXIT_ACK; - msg.src.node = scif_info.nodeid; - msg.dst.node = dev->node; - scif_nodeqp_send(dev, &msg); - dev->exit_ack_pending = false; - } -} - -/** - * scif_cleanup_scifdev - Uninitialize SCIF data structures for remote - * SCIF device. - * @dev: Remote SCIF device. - */ -void scif_cleanup_scifdev(struct scif_dev *dev) -{ - struct scif_hw_dev *sdev = dev->sdev; - - if (!dev->sdev) - return; - if (scifdev_is_p2p(dev)) { - if (dev->cookie) { - sdev->hw_ops->free_irq(sdev, dev->cookie, dev); - dev->cookie = NULL; - } - scif_destroy_intr_wq(dev); - } - flush_work(&scif_info.misc_work); - scif_destroy_p2p(dev); - scif_invalidate_ep(dev->node); - scif_zap_mmaps(dev->node); - scif_cleanup_rma_for_zombies(dev->node); - flush_work(&scif_info.misc_work); - scif_send_acks(dev); - if (!dev->node && scif_info.card_initiated_exit) { - /* - * Send an SCIF_EXIT message which is the last message from MIC - * to the Host and wait for a SCIF_EXIT_ACK - */ - scif_send_exit(dev); - scif_info.card_initiated_exit = false; - } - scif_cleanup_qp(dev); -} - -/** - * scif_remove_node - * - * @node: Node to remove - */ -void scif_handle_remove_node(int node) -{ - struct scif_dev *scifdev = &scif_dev[node]; - - if (scif_peer_unregister_device(scifdev)) - scif_send_acks(scifdev); -} - -static int scif_send_rmnode_msg(int node, int remove_node) -{ - struct scifmsg notif_msg; - struct scif_dev *dev = &scif_dev[node]; - - notif_msg.uop = SCIF_NODE_REMOVE; - notif_msg.src.node = scif_info.nodeid; - notif_msg.dst.node = node; - notif_msg.payload[0] = remove_node; - return scif_nodeqp_send(dev, ¬if_msg); -} - -/** - * scif_node_disconnect - * - * @node_id: source node id [in] - * @mgmt_initiated: Disconnection initiated from the mgmt node - * - * Disconnect a node from the scif network. - */ -void scif_disconnect_node(u32 node_id, bool mgmt_initiated) -{ - int ret; - int msg_cnt = 0; - u32 i = 0; - struct scif_dev *scifdev = &scif_dev[node_id]; - - if (!node_id) - return; - - atomic_set(&scifdev->disconn_rescnt, 0); - - /* Destroy p2p network */ - for (i = 1; i <= scif_info.maxid; i++) { - if (i == node_id) - continue; - ret = scif_send_rmnode_msg(i, node_id); - if (!ret) - msg_cnt++; - } - /* Wait for the remote nodes to respond with SCIF_NODE_REMOVE_ACK */ - ret = wait_event_timeout(scifdev->disconn_wq, - (atomic_read(&scifdev->disconn_rescnt) - == msg_cnt), SCIF_NODE_ALIVE_TIMEOUT); - /* Tell the card to clean up */ - if (mgmt_initiated && _scifdev_alive(scifdev)) - /* - * Send an SCIF_EXIT message which is the last message from Host - * to the MIC and wait for a SCIF_EXIT_ACK - */ - scif_send_exit(scifdev); - atomic_set(&scifdev->disconn_rescnt, 0); - /* Tell the mgmt node to clean up */ - ret = scif_send_rmnode_msg(SCIF_MGMT_NODE, node_id); - if (!ret) - /* Wait for mgmt node to respond with SCIF_NODE_REMOVE_ACK */ - wait_event_timeout(scifdev->disconn_wq, - (atomic_read(&scifdev->disconn_rescnt) == 1), - SCIF_NODE_ALIVE_TIMEOUT); -} - -void scif_get_node_info(void) -{ - struct scifmsg msg; - DECLARE_COMPLETION_ONSTACK(node_info); - - msg.uop = SCIF_GET_NODE_INFO; - msg.src.node = scif_info.nodeid; - msg.dst.node = SCIF_MGMT_NODE; - msg.payload[3] = (u64)&node_info; - - if ((scif_nodeqp_send(&scif_dev[SCIF_MGMT_NODE], &msg))) - return; - - /* Wait for a response with SCIF_GET_NODE_INFO */ - wait_for_completion(&node_info); -} diff --git a/drivers/misc/mic/scif/scif_nodeqp.c b/drivers/misc/mic/scif/scif_nodeqp.c deleted file mode 100644 index 384ce08fa98a..000000000000 --- a/drivers/misc/mic/scif/scif_nodeqp.c +++ /dev/null @@ -1,1349 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel SCIF driver. - */ -#include "../bus/scif_bus.h" -#include "scif_peer_bus.h" -#include "scif_main.h" -#include "scif_nodeqp.h" -#include "scif_map.h" - -/* - ************************************************************************ - * SCIF node Queue Pair (QP) setup flow: - * - * 1) SCIF driver gets probed with a scif_hw_dev via the scif_hw_bus - * 2) scif_setup_qp(..) allocates the local qp and calls - * scif_setup_qp_connect(..) which allocates and maps the local - * buffer for the inbound QP - * 3) The local node updates the device page with the DMA address of the QP - * 4) A delayed work is scheduled (qp_dwork) which periodically reads if - * the peer node has updated its QP DMA address - * 5) Once a valid non zero address is found in the QP DMA address field - * in the device page, the local node maps the remote node's QP, - * updates its outbound QP and sends a SCIF_INIT message to the peer - * 6) The SCIF_INIT message is received by the peer node QP interrupt bottom - * half handler by calling scif_init(..) - * 7) scif_init(..) registers a new SCIF peer node by calling - * scif_peer_register_device(..) which signifies the addition of a new - * SCIF node - * 8) On the mgmt node, P2P network setup/teardown is initiated if all the - * remote nodes are online via scif_p2p_setup(..) - * 9) For P2P setup, the host maps the remote nodes' aperture and memory - * bars and sends a SCIF_NODE_ADD message to both nodes - * 10) As part of scif_nodeadd, both nodes set up their local inbound - * QPs and send a SCIF_NODE_ADD_ACK to the mgmt node - * 11) As part of scif_node_add_ack(..) the mgmt node forwards the - * SCIF_NODE_ADD_ACK to the remote nodes - * 12) As part of scif_node_add_ack(..) the remote nodes update their - * outbound QPs, make sure they can access memory on the remote node - * and then add a new SCIF peer node by calling - * scif_peer_register_device(..) which signifies the addition of a new - * SCIF node. - * 13) The SCIF network is now established across all nodes. - * - ************************************************************************ - * SCIF node QP teardown flow (initiated by non mgmt node): - * - * 1) SCIF driver gets a remove callback with a scif_hw_dev via the scif_hw_bus - * 2) The device page QP DMA address field is updated with 0x0 - * 3) A non mgmt node now cleans up all local data structures and sends a - * SCIF_EXIT message to the peer and waits for a SCIF_EXIT_ACK - * 4) As part of scif_exit(..) handling scif_disconnect_node(..) is called - * 5) scif_disconnect_node(..) sends a SCIF_NODE_REMOVE message to all the - * peers and waits for a SCIF_NODE_REMOVE_ACK - * 6) As part of scif_node_remove(..) a remote node unregisters the peer - * node from the SCIF network and sends a SCIF_NODE_REMOVE_ACK - * 7) When the mgmt node has received all the SCIF_NODE_REMOVE_ACKs - * it sends itself a node remove message whose handling cleans up local - * data structures and unregisters the peer node from the SCIF network - * 8) The mgmt node sends a SCIF_EXIT_ACK - * 9) Upon receipt of the SCIF_EXIT_ACK the node initiating the teardown - * completes the SCIF remove routine - * 10) The SCIF network is now torn down for the node initiating the - * teardown sequence - * - ************************************************************************ - * SCIF node QP teardown flow (initiated by mgmt node): - * - * 1) SCIF driver gets a remove callback with a scif_hw_dev via the scif_hw_bus - * 2) The device page QP DMA address field is updated with 0x0 - * 3) The mgmt node calls scif_disconnect_node(..) - * 4) scif_disconnect_node(..) sends a SCIF_NODE_REMOVE message to all the peers - * and waits for a SCIF_NODE_REMOVE_ACK - * 5) As part of scif_node_remove(..) a remote node unregisters the peer - * node from the SCIF network and sends a SCIF_NODE_REMOVE_ACK - * 6) When the mgmt node has received all the SCIF_NODE_REMOVE_ACKs - * it unregisters the peer node from the SCIF network - * 7) The mgmt node sends a SCIF_EXIT message and waits for a SCIF_EXIT_ACK. - * 8) A non mgmt node upon receipt of a SCIF_EXIT message calls scif_stop(..) - * which would clean up local data structures for all SCIF nodes and - * then send a SCIF_EXIT_ACK back to the mgmt node - * 9) Upon receipt of the SCIF_EXIT_ACK the the mgmt node sends itself a node - * remove message whose handling cleans up local data structures and - * destroys any P2P mappings. - * 10) The SCIF hardware device for which a remove callback was received is now - * disconnected from the SCIF network. - */ -/* - * Initializes "local" data structures for the QP. Allocates the QP - * ring buffer (rb) and initializes the "in bound" queue. - */ -int scif_setup_qp_connect(struct scif_qp *qp, dma_addr_t *qp_offset, - int local_size, struct scif_dev *scifdev) -{ - void *local_q = qp->inbound_q.rb_base; - int err = 0; - u32 tmp_rd = 0; - - spin_lock_init(&qp->send_lock); - spin_lock_init(&qp->recv_lock); - - /* Allocate rb only if not already allocated */ - if (!local_q) { - local_q = kzalloc(local_size, GFP_KERNEL); - if (!local_q) { - err = -ENOMEM; - return err; - } - } - - err = scif_map_single(&qp->local_buf, local_q, scifdev, local_size); - if (err) - goto kfree; - /* - * To setup the inbound_q, the buffer lives locally, the read pointer - * is remote and the write pointer is local. - */ - scif_rb_init(&qp->inbound_q, - &tmp_rd, - &qp->local_write, - local_q, get_count_order(local_size)); - /* - * The read pointer is NULL initially and it is unsafe to use the ring - * buffer til this changes! - */ - qp->inbound_q.read_ptr = NULL; - err = scif_map_single(qp_offset, qp, - scifdev, sizeof(struct scif_qp)); - if (err) - goto unmap; - qp->local_qp = *qp_offset; - return err; -unmap: - scif_unmap_single(qp->local_buf, scifdev, local_size); - qp->local_buf = 0; -kfree: - kfree(local_q); - return err; -} - -/* When the other side has already done it's allocation, this is called */ -int scif_setup_qp_accept(struct scif_qp *qp, dma_addr_t *qp_offset, - dma_addr_t phys, int local_size, - struct scif_dev *scifdev) -{ - void *local_q; - void *remote_q; - struct scif_qp *remote_qp; - int remote_size; - int err = 0; - - spin_lock_init(&qp->send_lock); - spin_lock_init(&qp->recv_lock); - /* Start by figuring out where we need to point */ - remote_qp = scif_ioremap(phys, sizeof(struct scif_qp), scifdev); - if (!remote_qp) - return -EIO; - qp->remote_qp = remote_qp; - if (qp->remote_qp->magic != SCIFEP_MAGIC) { - err = -EIO; - goto iounmap; - } - qp->remote_buf = remote_qp->local_buf; - remote_size = qp->remote_qp->inbound_q.size; - remote_q = scif_ioremap(qp->remote_buf, remote_size, scifdev); - if (!remote_q) { - err = -EIO; - goto iounmap; - } - qp->remote_qp->local_write = 0; - /* - * To setup the outbound_q, the buffer lives in remote memory, - * the read pointer is local, the write pointer is remote - */ - scif_rb_init(&qp->outbound_q, - &qp->local_read, - &qp->remote_qp->local_write, - remote_q, - get_count_order(remote_size)); - local_q = kzalloc(local_size, GFP_KERNEL); - if (!local_q) { - err = -ENOMEM; - goto iounmap_1; - } - err = scif_map_single(&qp->local_buf, local_q, scifdev, local_size); - if (err) - goto kfree; - qp->remote_qp->local_read = 0; - /* - * To setup the inbound_q, the buffer lives locally, the read pointer - * is remote and the write pointer is local - */ - scif_rb_init(&qp->inbound_q, - &qp->remote_qp->local_read, - &qp->local_write, - local_q, get_count_order(local_size)); - err = scif_map_single(qp_offset, qp, scifdev, - sizeof(struct scif_qp)); - if (err) - goto unmap; - qp->local_qp = *qp_offset; - return err; -unmap: - scif_unmap_single(qp->local_buf, scifdev, local_size); - qp->local_buf = 0; -kfree: - kfree(local_q); -iounmap_1: - scif_iounmap(remote_q, remote_size, scifdev); - qp->outbound_q.rb_base = NULL; -iounmap: - scif_iounmap(qp->remote_qp, sizeof(struct scif_qp), scifdev); - qp->remote_qp = NULL; - return err; -} - -int scif_setup_qp_connect_response(struct scif_dev *scifdev, - struct scif_qp *qp, u64 payload) -{ - int err = 0; - void *r_buf; - int remote_size; - phys_addr_t tmp_phys; - - qp->remote_qp = scif_ioremap(payload, sizeof(struct scif_qp), scifdev); - - if (!qp->remote_qp) { - err = -ENOMEM; - goto error; - } - - if (qp->remote_qp->magic != SCIFEP_MAGIC) { - dev_err(&scifdev->sdev->dev, - "SCIFEP_MAGIC mismatch between self %d remote %d\n", - scif_dev[scif_info.nodeid].node, scifdev->node); - err = -ENODEV; - goto error; - } - - tmp_phys = qp->remote_qp->local_buf; - remote_size = qp->remote_qp->inbound_q.size; - r_buf = scif_ioremap(tmp_phys, remote_size, scifdev); - - if (!r_buf) - return -EIO; - - qp->local_read = 0; - scif_rb_init(&qp->outbound_q, - &qp->local_read, - &qp->remote_qp->local_write, - r_buf, - get_count_order(remote_size)); - /* - * Because the node QP may already be processing an INIT message, set - * the read pointer so the cached read offset isn't lost - */ - qp->remote_qp->local_read = qp->inbound_q.current_read_offset; - /* - * resetup the inbound_q now that we know where the - * inbound_read really is. - */ - scif_rb_init(&qp->inbound_q, - &qp->remote_qp->local_read, - &qp->local_write, - qp->inbound_q.rb_base, - get_count_order(qp->inbound_q.size)); -error: - return err; -} - -static __always_inline void -scif_send_msg_intr(struct scif_dev *scifdev) -{ - struct scif_hw_dev *sdev = scifdev->sdev; - - if (scifdev_is_p2p(scifdev)) - sdev->hw_ops->send_p2p_intr(sdev, scifdev->rdb, &scifdev->mmio); - else - sdev->hw_ops->send_intr(sdev, scifdev->rdb); -} - -int scif_qp_response(phys_addr_t phys, struct scif_dev *scifdev) -{ - int err = 0; - struct scifmsg msg; - - err = scif_setup_qp_connect_response(scifdev, scifdev->qpairs, phys); - if (!err) { - /* - * Now that everything is setup and mapped, we're ready - * to tell the peer about our queue's location - */ - msg.uop = SCIF_INIT; - msg.dst.node = scifdev->node; - err = scif_nodeqp_send(scifdev, &msg); - } - return err; -} - -void scif_send_exit(struct scif_dev *scifdev) -{ - struct scifmsg msg; - int ret; - - scifdev->exit = OP_IN_PROGRESS; - msg.uop = SCIF_EXIT; - msg.src.node = scif_info.nodeid; - msg.dst.node = scifdev->node; - ret = scif_nodeqp_send(scifdev, &msg); - if (ret) - goto done; - /* Wait for a SCIF_EXIT_ACK message */ - wait_event_timeout(scif_info.exitwq, scifdev->exit == OP_COMPLETED, - SCIF_NODE_ALIVE_TIMEOUT); -done: - scifdev->exit = OP_IDLE; -} - -int scif_setup_qp(struct scif_dev *scifdev) -{ - int err = 0; - int local_size; - struct scif_qp *qp; - - local_size = SCIF_NODE_QP_SIZE; - - qp = kzalloc(sizeof(*qp), GFP_KERNEL); - if (!qp) { - err = -ENOMEM; - return err; - } - qp->magic = SCIFEP_MAGIC; - scifdev->qpairs = qp; - err = scif_setup_qp_connect(qp, &scifdev->qp_dma_addr, - local_size, scifdev); - if (err) - goto free_qp; - /* - * We're as setup as we can be. The inbound_q is setup, w/o a usable - * outbound q. When we get a message, the read_ptr will be updated, - * and we will pull the message. - */ - return err; -free_qp: - kfree(scifdev->qpairs); - scifdev->qpairs = NULL; - return err; -} - -static void scif_p2p_freesg(struct scatterlist *sg) -{ - kfree(sg); -} - -static struct scatterlist * -scif_p2p_setsg(phys_addr_t pa, int page_size, int page_cnt) -{ - struct scatterlist *sg; - struct page *page; - int i; - - sg = kmalloc_array(page_cnt, sizeof(struct scatterlist), GFP_KERNEL); - if (!sg) - return NULL; - sg_init_table(sg, page_cnt); - for (i = 0; i < page_cnt; i++) { - page = pfn_to_page(pa >> PAGE_SHIFT); - sg_set_page(&sg[i], page, page_size, 0); - pa += page_size; - } - return sg; -} - -/* Init p2p mappings required to access peerdev from scifdev */ -static struct scif_p2p_info * -scif_init_p2p_info(struct scif_dev *scifdev, struct scif_dev *peerdev) -{ - struct scif_p2p_info *p2p; - int num_mmio_pages, num_aper_pages, sg_page_shift, err, num_aper_chunks; - struct scif_hw_dev *psdev = peerdev->sdev; - struct scif_hw_dev *sdev = scifdev->sdev; - - num_mmio_pages = psdev->mmio->len >> PAGE_SHIFT; - num_aper_pages = psdev->aper->len >> PAGE_SHIFT; - - p2p = kzalloc(sizeof(*p2p), GFP_KERNEL); - if (!p2p) - return NULL; - p2p->ppi_sg[SCIF_PPI_MMIO] = scif_p2p_setsg(psdev->mmio->pa, - PAGE_SIZE, num_mmio_pages); - if (!p2p->ppi_sg[SCIF_PPI_MMIO]) - goto free_p2p; - p2p->sg_nentries[SCIF_PPI_MMIO] = num_mmio_pages; - sg_page_shift = get_order(min(psdev->aper->len, (u64)(1 << 30))); - num_aper_chunks = num_aper_pages >> (sg_page_shift - PAGE_SHIFT); - p2p->ppi_sg[SCIF_PPI_APER] = scif_p2p_setsg(psdev->aper->pa, - 1 << sg_page_shift, - num_aper_chunks); - p2p->sg_nentries[SCIF_PPI_APER] = num_aper_chunks; - err = dma_map_sg(&sdev->dev, p2p->ppi_sg[SCIF_PPI_MMIO], - num_mmio_pages, PCI_DMA_BIDIRECTIONAL); - if (err != num_mmio_pages) - goto scif_p2p_free; - err = dma_map_sg(&sdev->dev, p2p->ppi_sg[SCIF_PPI_APER], - num_aper_chunks, PCI_DMA_BIDIRECTIONAL); - if (err != num_aper_chunks) - goto dma_unmap; - p2p->ppi_da[SCIF_PPI_MMIO] = sg_dma_address(p2p->ppi_sg[SCIF_PPI_MMIO]); - p2p->ppi_da[SCIF_PPI_APER] = sg_dma_address(p2p->ppi_sg[SCIF_PPI_APER]); - p2p->ppi_len[SCIF_PPI_MMIO] = num_mmio_pages; - p2p->ppi_len[SCIF_PPI_APER] = num_aper_pages; - p2p->ppi_peer_id = peerdev->node; - return p2p; -dma_unmap: - dma_unmap_sg(&sdev->dev, p2p->ppi_sg[SCIF_PPI_MMIO], - p2p->sg_nentries[SCIF_PPI_MMIO], DMA_BIDIRECTIONAL); -scif_p2p_free: - scif_p2p_freesg(p2p->ppi_sg[SCIF_PPI_MMIO]); - scif_p2p_freesg(p2p->ppi_sg[SCIF_PPI_APER]); -free_p2p: - kfree(p2p); - return NULL; -} - -/* Uninitialize and release resources from a p2p mapping */ -static void scif_deinit_p2p_info(struct scif_dev *scifdev, - struct scif_p2p_info *p2p) -{ - struct scif_hw_dev *sdev = scifdev->sdev; - - dma_unmap_sg(&sdev->dev, p2p->ppi_sg[SCIF_PPI_MMIO], - p2p->sg_nentries[SCIF_PPI_MMIO], DMA_BIDIRECTIONAL); - dma_unmap_sg(&sdev->dev, p2p->ppi_sg[SCIF_PPI_APER], - p2p->sg_nentries[SCIF_PPI_APER], DMA_BIDIRECTIONAL); - scif_p2p_freesg(p2p->ppi_sg[SCIF_PPI_MMIO]); - scif_p2p_freesg(p2p->ppi_sg[SCIF_PPI_APER]); - kfree(p2p); -} - -/** - * scif_node_connect: Respond to SCIF_NODE_CONNECT interrupt message - * @scifdev: SCIF device - * @dst: Destination node - * - * Connect the src and dst node by setting up the p2p connection - * between them. Management node here acts like a proxy. - */ -static void scif_node_connect(struct scif_dev *scifdev, int dst) -{ - struct scif_dev *dev_j = scifdev; - struct scif_dev *dev_i = NULL; - struct scif_p2p_info *p2p_ij = NULL; /* bus addr for j from i */ - struct scif_p2p_info *p2p_ji = NULL; /* bus addr for i from j */ - struct scif_p2p_info *p2p; - struct list_head *pos, *tmp; - struct scifmsg msg; - int err; - u64 tmppayload; - - if (dst < 1 || dst > scif_info.maxid) - return; - - dev_i = &scif_dev[dst]; - - if (!_scifdev_alive(dev_i)) - return; - /* - * If the p2p connection is already setup or in the process of setting - * up then just ignore this request. The requested node will get - * informed by SCIF_NODE_ADD_ACK or SCIF_NODE_ADD_NACK - */ - if (!list_empty(&dev_i->p2p)) { - list_for_each_safe(pos, tmp, &dev_i->p2p) { - p2p = list_entry(pos, struct scif_p2p_info, ppi_list); - if (p2p->ppi_peer_id == dev_j->node) - return; - } - } - p2p_ij = scif_init_p2p_info(dev_i, dev_j); - if (!p2p_ij) - return; - p2p_ji = scif_init_p2p_info(dev_j, dev_i); - if (!p2p_ji) { - scif_deinit_p2p_info(dev_i, p2p_ij); - return; - } - list_add_tail(&p2p_ij->ppi_list, &dev_i->p2p); - list_add_tail(&p2p_ji->ppi_list, &dev_j->p2p); - - /* - * Send a SCIF_NODE_ADD to dev_i, pass it its bus address - * as seen from dev_j - */ - msg.uop = SCIF_NODE_ADD; - msg.src.node = dev_j->node; - msg.dst.node = dev_i->node; - - msg.payload[0] = p2p_ji->ppi_da[SCIF_PPI_APER]; - msg.payload[1] = p2p_ij->ppi_da[SCIF_PPI_MMIO]; - msg.payload[2] = p2p_ij->ppi_da[SCIF_PPI_APER]; - msg.payload[3] = p2p_ij->ppi_len[SCIF_PPI_APER] << PAGE_SHIFT; - - err = scif_nodeqp_send(dev_i, &msg); - if (err) { - dev_err(&scifdev->sdev->dev, - "%s %d error %d\n", __func__, __LINE__, err); - return; - } - - /* Same as above but to dev_j */ - msg.uop = SCIF_NODE_ADD; - msg.src.node = dev_i->node; - msg.dst.node = dev_j->node; - - tmppayload = msg.payload[0]; - msg.payload[0] = msg.payload[2]; - msg.payload[2] = tmppayload; - msg.payload[1] = p2p_ji->ppi_da[SCIF_PPI_MMIO]; - msg.payload[3] = p2p_ji->ppi_len[SCIF_PPI_APER] << PAGE_SHIFT; - - scif_nodeqp_send(dev_j, &msg); -} - -static void scif_p2p_setup(void) -{ - int i, j; - - if (!scif_info.p2p_enable) - return; - - for (i = 1; i <= scif_info.maxid; i++) - if (!_scifdev_alive(&scif_dev[i])) - return; - - for (i = 1; i <= scif_info.maxid; i++) { - for (j = 1; j <= scif_info.maxid; j++) { - struct scif_dev *scifdev = &scif_dev[i]; - - if (i == j) - continue; - scif_node_connect(scifdev, j); - } - } -} - -static char *message_types[] = {"BAD", - "INIT", - "EXIT", - "SCIF_EXIT_ACK", - "SCIF_NODE_ADD", - "SCIF_NODE_ADD_ACK", - "SCIF_NODE_ADD_NACK", - "REMOVE_NODE", - "REMOVE_NODE_ACK", - "CNCT_REQ", - "CNCT_GNT", - "CNCT_GNTACK", - "CNCT_GNTNACK", - "CNCT_REJ", - "DISCNCT", - "DISCNT_ACK", - "CLIENT_SENT", - "CLIENT_RCVD", - "SCIF_GET_NODE_INFO", - "REGISTER", - "REGISTER_ACK", - "REGISTER_NACK", - "UNREGISTER", - "UNREGISTER_ACK", - "UNREGISTER_NACK", - "ALLOC_REQ", - "ALLOC_GNT", - "ALLOC_REJ", - "FREE_PHYS", - "FREE_VIRT", - "MUNMAP", - "MARK", - "MARK_ACK", - "MARK_NACK", - "WAIT", - "WAIT_ACK", - "WAIT_NACK", - "SIGNAL_LOCAL", - "SIGNAL_REMOTE", - "SIG_ACK", - "SIG_NACK"}; - -static void -scif_display_message(struct scif_dev *scifdev, struct scifmsg *msg, - const char *label) -{ - if (!scif_info.en_msg_log) - return; - if (msg->uop > SCIF_MAX_MSG) { - dev_err(&scifdev->sdev->dev, - "%s: unknown msg type %d\n", label, msg->uop); - return; - } - dev_info(&scifdev->sdev->dev, - "%s: msg type %s, src %d:%d, dest %d:%d payload 0x%llx:0x%llx:0x%llx:0x%llx\n", - label, message_types[msg->uop], msg->src.node, msg->src.port, - msg->dst.node, msg->dst.port, msg->payload[0], msg->payload[1], - msg->payload[2], msg->payload[3]); -} - -int _scif_nodeqp_send(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_qp *qp = scifdev->qpairs; - int err = -ENOMEM, loop_cnt = 0; - - scif_display_message(scifdev, msg, "Sent"); - if (!qp) { - err = -EINVAL; - goto error; - } - spin_lock(&qp->send_lock); - - while ((err = scif_rb_write(&qp->outbound_q, - msg, sizeof(struct scifmsg)))) { - mdelay(1); -#define SCIF_NODEQP_SEND_TO_MSEC (3 * 1000) - if (loop_cnt++ > (SCIF_NODEQP_SEND_TO_MSEC)) { - err = -ENODEV; - break; - } - } - if (!err) - scif_rb_commit(&qp->outbound_q); - spin_unlock(&qp->send_lock); - if (!err) { - if (scifdev_self(scifdev)) - /* - * For loopback we need to emulate an interrupt by - * queuing work for the queue handling real node - * Qp interrupts. - */ - queue_work(scifdev->intr_wq, &scifdev->intr_bh); - else - scif_send_msg_intr(scifdev); - } -error: - if (err) - dev_dbg(&scifdev->sdev->dev, - "%s %d error %d uop %d\n", - __func__, __LINE__, err, msg->uop); - return err; -} - -/** - * scif_nodeqp_send - Send a message on the node queue pair - * @scifdev: Scif Device. - * @msg: The message to be sent. - */ -int scif_nodeqp_send(struct scif_dev *scifdev, struct scifmsg *msg) -{ - int err; - struct device *spdev = NULL; - - if (msg->uop > SCIF_EXIT_ACK) { - /* Don't send messages once the exit flow has begun */ - if (OP_IDLE != scifdev->exit) - return -ENODEV; - spdev = scif_get_peer_dev(scifdev); - if (IS_ERR(spdev)) { - err = PTR_ERR(spdev); - return err; - } - } - err = _scif_nodeqp_send(scifdev, msg); - if (msg->uop > SCIF_EXIT_ACK) - scif_put_peer_dev(spdev); - return err; -} - -/* - * scif_misc_handler: - * - * Work queue handler for servicing miscellaneous SCIF tasks. - * Examples include: - * 1) Remote fence requests. - * 2) Destruction of temporary registered windows - * created during scif_vreadfrom()/scif_vwriteto(). - * 3) Cleanup of zombie endpoints. - */ -void scif_misc_handler(struct work_struct *work) -{ - scif_rma_handle_remote_fences(); - scif_rma_destroy_windows(); - scif_rma_destroy_tcw_invalid(); - scif_cleanup_zombie_epd(); -} - -/** - * scif_init() - Respond to SCIF_INIT interrupt message - * @scifdev: Remote SCIF device node - * @msg: Interrupt message - */ -static __always_inline void -scif_init(struct scif_dev *scifdev, struct scifmsg *msg) -{ - /* - * Allow the thread waiting for device page updates for the peer QP DMA - * address to complete initializing the inbound_q. - */ - flush_delayed_work(&scifdev->qp_dwork); - - scif_peer_register_device(scifdev); - - if (scif_is_mgmt_node()) { - mutex_lock(&scif_info.conflock); - scif_p2p_setup(); - mutex_unlock(&scif_info.conflock); - } -} - -/** - * scif_exit() - Respond to SCIF_EXIT interrupt message - * @scifdev: Remote SCIF device node - * @unused: Interrupt message (unused) - * - * This function stops the SCIF interface for the node which sent - * the SCIF_EXIT message and starts waiting for that node to - * resetup the queue pair again. - */ -static __always_inline void -scif_exit(struct scif_dev *scifdev, struct scifmsg *unused) -{ - scifdev->exit_ack_pending = true; - if (scif_is_mgmt_node()) - scif_disconnect_node(scifdev->node, false); - else - scif_stop(scifdev); - schedule_delayed_work(&scifdev->qp_dwork, - msecs_to_jiffies(1000)); -} - -/** - * scif_exitack() - Respond to SCIF_EXIT_ACK interrupt message - * @scifdev: Remote SCIF device node - * @unused: Interrupt message (unused) - * - */ -static __always_inline void -scif_exit_ack(struct scif_dev *scifdev, struct scifmsg *unused) -{ - scifdev->exit = OP_COMPLETED; - wake_up(&scif_info.exitwq); -} - -/** - * scif_node_add() - Respond to SCIF_NODE_ADD interrupt message - * @scifdev: Remote SCIF device node - * @msg: Interrupt message - * - * When the mgmt node driver has finished initializing a MIC node queue pair it - * marks the node as online. It then looks for all currently online MIC cards - * and send a SCIF_NODE_ADD message to identify the ID of the new card for - * peer to peer initialization - * - * The local node allocates its incoming queue and sends its address in the - * SCIF_NODE_ADD_ACK message back to the mgmt node, the mgmt node "reflects" - * this message to the new node - */ -static __always_inline void -scif_node_add(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_dev *newdev; - dma_addr_t qp_offset; - int qp_connect; - struct scif_hw_dev *sdev; - - dev_dbg(&scifdev->sdev->dev, - "Scifdev %d:%d received NODE_ADD msg for node %d\n", - scifdev->node, msg->dst.node, msg->src.node); - dev_dbg(&scifdev->sdev->dev, - "Remote address for this node's aperture %llx\n", - msg->payload[0]); - newdev = &scif_dev[msg->src.node]; - newdev->node = msg->src.node; - newdev->sdev = scif_dev[SCIF_MGMT_NODE].sdev; - sdev = newdev->sdev; - - if (scif_setup_intr_wq(newdev)) { - dev_err(&scifdev->sdev->dev, - "failed to setup interrupts for %d\n", msg->src.node); - goto interrupt_setup_error; - } - newdev->mmio.va = ioremap(msg->payload[1], sdev->mmio->len); - if (!newdev->mmio.va) { - dev_err(&scifdev->sdev->dev, - "failed to map mmio for %d\n", msg->src.node); - goto mmio_map_error; - } - newdev->qpairs = kzalloc(sizeof(*newdev->qpairs), GFP_KERNEL); - if (!newdev->qpairs) - goto qp_alloc_error; - /* - * Set the base address of the remote node's memory since it gets - * added to qp_offset - */ - newdev->base_addr = msg->payload[0]; - - qp_connect = scif_setup_qp_connect(newdev->qpairs, &qp_offset, - SCIF_NODE_QP_SIZE, newdev); - if (qp_connect) { - dev_err(&scifdev->sdev->dev, - "failed to setup qp_connect %d\n", qp_connect); - goto qp_connect_error; - } - - newdev->db = sdev->hw_ops->next_db(sdev); - newdev->cookie = sdev->hw_ops->request_irq(sdev, scif_intr_handler, - "SCIF_INTR", newdev, - newdev->db); - if (IS_ERR(newdev->cookie)) - goto qp_connect_error; - newdev->qpairs->magic = SCIFEP_MAGIC; - newdev->qpairs->qp_state = SCIF_QP_OFFLINE; - - msg->uop = SCIF_NODE_ADD_ACK; - msg->dst.node = msg->src.node; - msg->src.node = scif_info.nodeid; - msg->payload[0] = qp_offset; - msg->payload[2] = newdev->db; - scif_nodeqp_send(&scif_dev[SCIF_MGMT_NODE], msg); - return; -qp_connect_error: - kfree(newdev->qpairs); - newdev->qpairs = NULL; -qp_alloc_error: - iounmap(newdev->mmio.va); - newdev->mmio.va = NULL; -mmio_map_error: -interrupt_setup_error: - dev_err(&scifdev->sdev->dev, - "node add failed for node %d\n", msg->src.node); - msg->uop = SCIF_NODE_ADD_NACK; - msg->dst.node = msg->src.node; - msg->src.node = scif_info.nodeid; - scif_nodeqp_send(&scif_dev[SCIF_MGMT_NODE], msg); -} - -void scif_poll_qp_state(struct work_struct *work) -{ -#define SCIF_NODE_QP_RETRY 100 -#define SCIF_NODE_QP_TIMEOUT 100 - struct scif_dev *peerdev = container_of(work, struct scif_dev, - p2p_dwork.work); - struct scif_qp *qp = &peerdev->qpairs[0]; - - if (qp->qp_state != SCIF_QP_ONLINE || - qp->remote_qp->qp_state != SCIF_QP_ONLINE) { - if (peerdev->p2p_retry++ == SCIF_NODE_QP_RETRY) { - dev_err(&peerdev->sdev->dev, - "Warning: QP check timeout with state %d\n", - qp->qp_state); - goto timeout; - } - schedule_delayed_work(&peerdev->p2p_dwork, - msecs_to_jiffies(SCIF_NODE_QP_TIMEOUT)); - return; - } - return; -timeout: - dev_err(&peerdev->sdev->dev, - "%s %d remote node %d offline, state = 0x%x\n", - __func__, __LINE__, peerdev->node, qp->qp_state); - qp->remote_qp->qp_state = SCIF_QP_OFFLINE; - scif_peer_unregister_device(peerdev); - scif_cleanup_scifdev(peerdev); -} - -/** - * scif_node_add_ack() - Respond to SCIF_NODE_ADD_ACK interrupt message - * @scifdev: Remote SCIF device node - * @msg: Interrupt message - * - * After a MIC node receives the SCIF_NODE_ADD_ACK message it send this - * message to the mgmt node to confirm the sequence is finished. - * - */ -static __always_inline void -scif_node_add_ack(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_dev *peerdev; - struct scif_qp *qp; - struct scif_dev *dst_dev = &scif_dev[msg->dst.node]; - - dev_dbg(&scifdev->sdev->dev, - "Scifdev %d received SCIF_NODE_ADD_ACK msg src %d dst %d\n", - scifdev->node, msg->src.node, msg->dst.node); - dev_dbg(&scifdev->sdev->dev, - "payload %llx %llx %llx %llx\n", msg->payload[0], - msg->payload[1], msg->payload[2], msg->payload[3]); - if (scif_is_mgmt_node()) { - /* - * the lock serializes with scif_qp_response_ack. The mgmt node - * is forwarding the NODE_ADD_ACK message from src to dst we - * need to make sure that the dst has already received a - * NODE_ADD for src and setup its end of the qp to dst - */ - mutex_lock(&scif_info.conflock); - msg->payload[1] = scif_info.maxid; - scif_nodeqp_send(dst_dev, msg); - mutex_unlock(&scif_info.conflock); - return; - } - peerdev = &scif_dev[msg->src.node]; - peerdev->sdev = scif_dev[SCIF_MGMT_NODE].sdev; - peerdev->node = msg->src.node; - - qp = &peerdev->qpairs[0]; - - if ((scif_setup_qp_connect_response(peerdev, &peerdev->qpairs[0], - msg->payload[0]))) - goto local_error; - peerdev->rdb = msg->payload[2]; - qp->remote_qp->qp_state = SCIF_QP_ONLINE; - - scif_peer_register_device(peerdev); - - schedule_delayed_work(&peerdev->p2p_dwork, 0); - return; -local_error: - scif_cleanup_scifdev(peerdev); -} - -/** - * scif_node_add_nack: Respond to SCIF_NODE_ADD_NACK interrupt message - * @scifdev: Remote SCIF device node - * @msg: Interrupt message - * - * SCIF_NODE_ADD failed, so inform the waiting wq. - */ -static __always_inline void -scif_node_add_nack(struct scif_dev *scifdev, struct scifmsg *msg) -{ - if (scif_is_mgmt_node()) { - struct scif_dev *dst_dev = &scif_dev[msg->dst.node]; - - dev_dbg(&scifdev->sdev->dev, - "SCIF_NODE_ADD_NACK received from %d\n", scifdev->node); - scif_nodeqp_send(dst_dev, msg); - } -} - -/** - * scif_node_remove: Handle SCIF_NODE_REMOVE message - * @scifdev: Remote SCIF device node - * @msg: Interrupt message - * - * Handle node removal. - */ -static __always_inline void -scif_node_remove(struct scif_dev *scifdev, struct scifmsg *msg) -{ - int node = msg->payload[0]; - struct scif_dev *scdev = &scif_dev[node]; - - scdev->node_remove_ack_pending = true; - scif_handle_remove_node(node); -} - -/** - * scif_node_remove_ack: Handle SCIF_NODE_REMOVE_ACK message - * @scifdev: Remote SCIF device node - * @msg: Interrupt message - * - * The peer has acked a SCIF_NODE_REMOVE message. - */ -static __always_inline void -scif_node_remove_ack(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_dev *sdev = &scif_dev[msg->payload[0]]; - - atomic_inc(&sdev->disconn_rescnt); - wake_up(&sdev->disconn_wq); -} - -/** - * scif_get_node_info: Respond to SCIF_GET_NODE_INFO interrupt message - * @scifdev: Remote SCIF device node - * @msg: Interrupt message - * - * Retrieve node info i.e maxid and total from the mgmt node. - */ -static __always_inline void -scif_get_node_info_resp(struct scif_dev *scifdev, struct scifmsg *msg) -{ - if (scif_is_mgmt_node()) { - swap(msg->dst.node, msg->src.node); - mutex_lock(&scif_info.conflock); - msg->payload[1] = scif_info.maxid; - msg->payload[2] = scif_info.total; - mutex_unlock(&scif_info.conflock); - scif_nodeqp_send(scifdev, msg); - } else { - struct completion *node_info = - (struct completion *)msg->payload[3]; - - mutex_lock(&scif_info.conflock); - scif_info.maxid = msg->payload[1]; - scif_info.total = msg->payload[2]; - complete_all(node_info); - mutex_unlock(&scif_info.conflock); - } -} - -static void -scif_msg_unknown(struct scif_dev *scifdev, struct scifmsg *msg) -{ - /* Bogus Node Qp Message? */ - dev_err(&scifdev->sdev->dev, - "Unknown message 0x%xn scifdev->node 0x%x\n", - msg->uop, scifdev->node); -} - -static void (*scif_intr_func[SCIF_MAX_MSG + 1]) - (struct scif_dev *, struct scifmsg *msg) = { - scif_msg_unknown, /* Error */ - scif_init, /* SCIF_INIT */ - scif_exit, /* SCIF_EXIT */ - scif_exit_ack, /* SCIF_EXIT_ACK */ - scif_node_add, /* SCIF_NODE_ADD */ - scif_node_add_ack, /* SCIF_NODE_ADD_ACK */ - scif_node_add_nack, /* SCIF_NODE_ADD_NACK */ - scif_node_remove, /* SCIF_NODE_REMOVE */ - scif_node_remove_ack, /* SCIF_NODE_REMOVE_ACK */ - scif_cnctreq, /* SCIF_CNCT_REQ */ - scif_cnctgnt, /* SCIF_CNCT_GNT */ - scif_cnctgnt_ack, /* SCIF_CNCT_GNTACK */ - scif_cnctgnt_nack, /* SCIF_CNCT_GNTNACK */ - scif_cnctrej, /* SCIF_CNCT_REJ */ - scif_discnct, /* SCIF_DISCNCT */ - scif_discnt_ack, /* SCIF_DISCNT_ACK */ - scif_clientsend, /* SCIF_CLIENT_SENT */ - scif_clientrcvd, /* SCIF_CLIENT_RCVD */ - scif_get_node_info_resp,/* SCIF_GET_NODE_INFO */ - scif_recv_reg, /* SCIF_REGISTER */ - scif_recv_reg_ack, /* SCIF_REGISTER_ACK */ - scif_recv_reg_nack, /* SCIF_REGISTER_NACK */ - scif_recv_unreg, /* SCIF_UNREGISTER */ - scif_recv_unreg_ack, /* SCIF_UNREGISTER_ACK */ - scif_recv_unreg_nack, /* SCIF_UNREGISTER_NACK */ - scif_alloc_req, /* SCIF_ALLOC_REQ */ - scif_alloc_gnt_rej, /* SCIF_ALLOC_GNT */ - scif_alloc_gnt_rej, /* SCIF_ALLOC_REJ */ - scif_free_virt, /* SCIF_FREE_VIRT */ - scif_recv_munmap, /* SCIF_MUNMAP */ - scif_recv_mark, /* SCIF_MARK */ - scif_recv_mark_resp, /* SCIF_MARK_ACK */ - scif_recv_mark_resp, /* SCIF_MARK_NACK */ - scif_recv_wait, /* SCIF_WAIT */ - scif_recv_wait_resp, /* SCIF_WAIT_ACK */ - scif_recv_wait_resp, /* SCIF_WAIT_NACK */ - scif_recv_sig_local, /* SCIF_SIG_LOCAL */ - scif_recv_sig_remote, /* SCIF_SIG_REMOTE */ - scif_recv_sig_resp, /* SCIF_SIG_ACK */ - scif_recv_sig_resp, /* SCIF_SIG_NACK */ -}; - -static int scif_max_msg_id = SCIF_MAX_MSG; -/** - * scif_nodeqp_msg_handler() - Common handler for node messages - * @scifdev: Remote device to respond to - * @qp: Remote memory pointer - * @msg: The message to be handled. - * - * This routine calls the appropriate routine to handle a Node Qp - * message receipt - */ -static void -scif_nodeqp_msg_handler(struct scif_dev *scifdev, - struct scif_qp *qp, struct scifmsg *msg) -{ - scif_display_message(scifdev, msg, "Rcvd"); - - if (msg->uop > (u32)scif_max_msg_id) { - /* Bogus Node Qp Message? */ - dev_err(&scifdev->sdev->dev, - "Unknown message 0x%xn scifdev->node 0x%x\n", - msg->uop, scifdev->node); - return; - } - - scif_intr_func[msg->uop](scifdev, msg); -} - -/** - * scif_nodeqp_intrhandler() - Interrupt handler for node messages - * @scifdev: Remote device to respond to - * @qp: Remote memory pointer - * - * This routine is triggered by the interrupt mechanism. It reads - * messages from the node queue RB and calls the Node QP Message handling - * routine. - */ -void scif_nodeqp_intrhandler(struct scif_dev *scifdev, struct scif_qp *qp) -{ - struct scifmsg msg; - int read_size; - - do { - read_size = scif_rb_get_next(&qp->inbound_q, &msg, sizeof(msg)); - if (!read_size) - break; - scif_nodeqp_msg_handler(scifdev, qp, &msg); - /* - * The node queue pair is unmapped so skip the read pointer - * update after receipt of a SCIF_EXIT_ACK - */ - if (SCIF_EXIT_ACK == msg.uop) - break; - scif_rb_update_read_ptr(&qp->inbound_q); - } while (1); -} - -/** - * scif_loopb_wq_handler - Loopback Workqueue Handler. - * @unused: loop back work (unused) - * - * This work queue routine is invoked by the loopback work queue handler. - * It grabs the recv lock, dequeues any available messages from the head - * of the loopback message list, calls the node QP message handler, - * waits for it to return, then frees up this message and dequeues more - * elements of the list if available. - */ -static void scif_loopb_wq_handler(struct work_struct *unused) -{ - struct scif_dev *scifdev = scif_info.loopb_dev; - struct scif_qp *qp = scifdev->qpairs; - struct scif_loopb_msg *msg; - - do { - msg = NULL; - spin_lock(&qp->recv_lock); - if (!list_empty(&scif_info.loopb_recv_q)) { - msg = list_first_entry(&scif_info.loopb_recv_q, - struct scif_loopb_msg, - list); - list_del(&msg->list); - } - spin_unlock(&qp->recv_lock); - - if (msg) { - scif_nodeqp_msg_handler(scifdev, qp, &msg->msg); - kfree(msg); - } - } while (msg); -} - -/** - * scif_loopb_msg_handler() - Workqueue handler for loopback messages. - * @scifdev: SCIF device - * @qp: Queue pair. - * - * This work queue routine is triggered when a loopback message is received. - * - * We need special handling for receiving Node Qp messages on a loopback SCIF - * device via two workqueues for receiving messages. - * - * The reason we need the extra workqueue which is not required with *normal* - * non-loopback SCIF devices is the potential classic deadlock described below: - * - * Thread A tries to send a message on a loopback SCIF device and blocks since - * there is no space in the RB while it has the send_lock held or another - * lock called lock X for example. - * - * Thread B: The Loopback Node QP message receive workqueue receives the message - * and tries to send a message (eg an ACK) to the loopback SCIF device. It tries - * to grab the send lock again or lock X and deadlocks with Thread A. The RB - * cannot be drained any further due to this classic deadlock. - * - * In order to avoid deadlocks as mentioned above we have an extra level of - * indirection achieved by having two workqueues. - * 1) The first workqueue whose handler is scif_loopb_msg_handler reads - * messages from the Node QP RB, adds them to a list and queues work for the - * second workqueue. - * - * 2) The second workqueue whose handler is scif_loopb_wq_handler dequeues - * messages from the list, handles them, frees up the memory and dequeues - * more elements from the list if possible. - */ -int -scif_loopb_msg_handler(struct scif_dev *scifdev, struct scif_qp *qp) -{ - int read_size; - struct scif_loopb_msg *msg; - - do { - msg = kmalloc(sizeof(*msg), GFP_KERNEL); - if (!msg) - return -ENOMEM; - read_size = scif_rb_get_next(&qp->inbound_q, &msg->msg, - sizeof(struct scifmsg)); - if (read_size != sizeof(struct scifmsg)) { - kfree(msg); - scif_rb_update_read_ptr(&qp->inbound_q); - break; - } - spin_lock(&qp->recv_lock); - list_add_tail(&msg->list, &scif_info.loopb_recv_q); - spin_unlock(&qp->recv_lock); - queue_work(scif_info.loopb_wq, &scif_info.loopb_work); - scif_rb_update_read_ptr(&qp->inbound_q); - } while (read_size == sizeof(struct scifmsg)); - return read_size; -} - -/** - * scif_setup_loopback_qp - One time setup work for Loopback Node Qp. - * @scifdev: SCIF device - * - * Sets up the required loopback workqueues, queue pairs and ring buffers - */ -int scif_setup_loopback_qp(struct scif_dev *scifdev) -{ - int err = 0; - void *local_q; - struct scif_qp *qp; - - err = scif_setup_intr_wq(scifdev); - if (err) - goto exit; - INIT_LIST_HEAD(&scif_info.loopb_recv_q); - snprintf(scif_info.loopb_wqname, sizeof(scif_info.loopb_wqname), - "SCIF LOOPB %d", scifdev->node); - scif_info.loopb_wq = - alloc_ordered_workqueue(scif_info.loopb_wqname, 0); - if (!scif_info.loopb_wq) { - err = -ENOMEM; - goto destroy_intr; - } - INIT_WORK(&scif_info.loopb_work, scif_loopb_wq_handler); - /* Allocate Self Qpair */ - scifdev->qpairs = kzalloc(sizeof(*scifdev->qpairs), GFP_KERNEL); - if (!scifdev->qpairs) { - err = -ENOMEM; - goto destroy_loopb_wq; - } - - qp = scifdev->qpairs; - qp->magic = SCIFEP_MAGIC; - spin_lock_init(&qp->send_lock); - spin_lock_init(&qp->recv_lock); - - local_q = kzalloc(SCIF_NODE_QP_SIZE, GFP_KERNEL); - if (!local_q) { - err = -ENOMEM; - goto free_qpairs; - } - /* - * For loopback the inbound_q and outbound_q are essentially the same - * since the Node sends a message on the loopback interface to the - * outbound_q which is then received on the inbound_q. - */ - scif_rb_init(&qp->outbound_q, - &qp->local_read, - &qp->local_write, - local_q, get_count_order(SCIF_NODE_QP_SIZE)); - - scif_rb_init(&qp->inbound_q, - &qp->local_read, - &qp->local_write, - local_q, get_count_order(SCIF_NODE_QP_SIZE)); - scif_info.nodeid = scifdev->node; - - scif_peer_register_device(scifdev); - - scif_info.loopb_dev = scifdev; - return err; -free_qpairs: - kfree(scifdev->qpairs); -destroy_loopb_wq: - destroy_workqueue(scif_info.loopb_wq); -destroy_intr: - scif_destroy_intr_wq(scifdev); -exit: - return err; -} - -/** - * scif_destroy_loopback_qp - One time uninit work for Loopback Node Qp - * @scifdev: SCIF device - * - * Destroys the workqueues and frees up the Ring Buffer and Queue Pair memory. - */ -int scif_destroy_loopback_qp(struct scif_dev *scifdev) -{ - scif_peer_unregister_device(scifdev); - destroy_workqueue(scif_info.loopb_wq); - scif_destroy_intr_wq(scifdev); - kfree(scifdev->qpairs->outbound_q.rb_base); - kfree(scifdev->qpairs); - scifdev->sdev = NULL; - scif_info.loopb_dev = NULL; - return 0; -} - -void scif_destroy_p2p(struct scif_dev *scifdev) -{ - struct scif_dev *peer_dev; - struct scif_p2p_info *p2p; - struct list_head *pos, *tmp; - int bd; - - mutex_lock(&scif_info.conflock); - /* Free P2P mappings in the given node for all its peer nodes */ - list_for_each_safe(pos, tmp, &scifdev->p2p) { - p2p = list_entry(pos, struct scif_p2p_info, ppi_list); - dma_unmap_sg(&scifdev->sdev->dev, p2p->ppi_sg[SCIF_PPI_MMIO], - p2p->sg_nentries[SCIF_PPI_MMIO], - DMA_BIDIRECTIONAL); - dma_unmap_sg(&scifdev->sdev->dev, p2p->ppi_sg[SCIF_PPI_APER], - p2p->sg_nentries[SCIF_PPI_APER], - DMA_BIDIRECTIONAL); - scif_p2p_freesg(p2p->ppi_sg[SCIF_PPI_MMIO]); - scif_p2p_freesg(p2p->ppi_sg[SCIF_PPI_APER]); - list_del(pos); - kfree(p2p); - } - - /* Free P2P mapping created in the peer nodes for the given node */ - for (bd = SCIF_MGMT_NODE + 1; bd <= scif_info.maxid; bd++) { - peer_dev = &scif_dev[bd]; - list_for_each_safe(pos, tmp, &peer_dev->p2p) { - p2p = list_entry(pos, struct scif_p2p_info, ppi_list); - if (p2p->ppi_peer_id == scifdev->node) { - dma_unmap_sg(&peer_dev->sdev->dev, - p2p->ppi_sg[SCIF_PPI_MMIO], - p2p->sg_nentries[SCIF_PPI_MMIO], - DMA_BIDIRECTIONAL); - dma_unmap_sg(&peer_dev->sdev->dev, - p2p->ppi_sg[SCIF_PPI_APER], - p2p->sg_nentries[SCIF_PPI_APER], - DMA_BIDIRECTIONAL); - scif_p2p_freesg(p2p->ppi_sg[SCIF_PPI_MMIO]); - scif_p2p_freesg(p2p->ppi_sg[SCIF_PPI_APER]); - list_del(pos); - kfree(p2p); - } - } - } - mutex_unlock(&scif_info.conflock); -} diff --git a/drivers/misc/mic/scif/scif_nodeqp.h b/drivers/misc/mic/scif/scif_nodeqp.h deleted file mode 100644 index 95896273138e..000000000000 --- a/drivers/misc/mic/scif/scif_nodeqp.h +++ /dev/null @@ -1,221 +0,0 @@ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * This file is provided under a dual BSD/GPLv2 license. When using or - * redistributing this file, you may do so under either license. - * - * GPL LICENSE SUMMARY - * - * Copyright(c) 2014 Intel Corporation. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * BSD LICENSE - * - * Copyright(c) 2014 Intel Corporation. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * - * * Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * * Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in - * the documentation and/or other materials provided with the - * distribution. - * * Neither the name of Intel Corporation nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR - * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT - * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, - * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT - * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - * - * Intel SCIF driver. - * - */ -#ifndef SCIF_NODEQP -#define SCIF_NODEQP - -#include "scif_rb.h" -#include "scif_peer_bus.h" - -#define SCIF_INIT 1 /* First message sent to the peer node for discovery */ -#define SCIF_EXIT 2 /* Last message from the peer informing intent to exit */ -#define SCIF_EXIT_ACK 3 /* Response to SCIF_EXIT message */ -#define SCIF_NODE_ADD 4 /* Tell Online nodes a new node exits */ -#define SCIF_NODE_ADD_ACK 5 /* Confirm to mgmt node sequence is finished */ -#define SCIF_NODE_ADD_NACK 6 /* SCIF_NODE_ADD failed */ -#define SCIF_NODE_REMOVE 7 /* Request to deactivate a SCIF node */ -#define SCIF_NODE_REMOVE_ACK 8 /* Response to a SCIF_NODE_REMOVE message */ -#define SCIF_CNCT_REQ 9 /* Phys addr of Request connection to a port */ -#define SCIF_CNCT_GNT 10 /* Phys addr of new Grant connection request */ -#define SCIF_CNCT_GNTACK 11 /* Error type Reject a connection request */ -#define SCIF_CNCT_GNTNACK 12 /* Error type Reject a connection request */ -#define SCIF_CNCT_REJ 13 /* Error type Reject a connection request */ -#define SCIF_DISCNCT 14 /* Notify peer that connection is being terminated */ -#define SCIF_DISCNT_ACK 15 /* Notify peer that connection is being terminated */ -#define SCIF_CLIENT_SENT 16 /* Notify the peer that data has been written */ -#define SCIF_CLIENT_RCVD 17 /* Notify the peer that data has been read */ -#define SCIF_GET_NODE_INFO 18 /* Get current node mask from the mgmt node*/ -#define SCIF_REGISTER 19 /* Tell peer about a new registered window */ -#define SCIF_REGISTER_ACK 20 /* Notify peer about unregistration success */ -#define SCIF_REGISTER_NACK 21 /* Notify peer about registration success */ -#define SCIF_UNREGISTER 22 /* Tell peer about unregistering a window */ -#define SCIF_UNREGISTER_ACK 23 /* Notify peer about registration failure */ -#define SCIF_UNREGISTER_NACK 24 /* Notify peer about unregistration failure */ -#define SCIF_ALLOC_REQ 25 /* Request a mapped buffer */ -#define SCIF_ALLOC_GNT 26 /* Notify peer about allocation success */ -#define SCIF_ALLOC_REJ 27 /* Notify peer about allocation failure */ -#define SCIF_FREE_VIRT 28 /* Free previously allocated virtual memory */ -#define SCIF_MUNMAP 29 /* Acknowledgment for a SCIF_MMAP request */ -#define SCIF_MARK 30 /* SCIF Remote Fence Mark Request */ -#define SCIF_MARK_ACK 31 /* SCIF Remote Fence Mark Success */ -#define SCIF_MARK_NACK 32 /* SCIF Remote Fence Mark Failure */ -#define SCIF_WAIT 33 /* SCIF Remote Fence Wait Request */ -#define SCIF_WAIT_ACK 34 /* SCIF Remote Fence Wait Success */ -#define SCIF_WAIT_NACK 35 /* SCIF Remote Fence Wait Failure */ -#define SCIF_SIG_LOCAL 36 /* SCIF Remote Fence Local Signal Request */ -#define SCIF_SIG_REMOTE 37 /* SCIF Remote Fence Remote Signal Request */ -#define SCIF_SIG_ACK 38 /* SCIF Remote Fence Remote Signal Success */ -#define SCIF_SIG_NACK 39 /* SCIF Remote Fence Remote Signal Failure */ -#define SCIF_MAX_MSG SCIF_SIG_NACK - -/* - * struct scifmsg - Node QP message format - * - * @src: Source information - * @dst: Destination information - * @uop: The message opcode - * @payload: Unique payload format for each message - */ -struct scifmsg { - struct scif_port_id src; - struct scif_port_id dst; - u32 uop; - u64 payload[4]; -} __packed; - -/* - * struct scif_allocmsg - Used with SCIF_ALLOC_REQ to request - * the remote note to allocate memory - * - * phys_addr: Physical address of the buffer - * vaddr: Virtual address of the buffer - * size: Size of the buffer - * state: Current state - * allocwq: wait queue for status - */ -struct scif_allocmsg { - dma_addr_t phys_addr; - unsigned long vaddr; - size_t size; - enum scif_msg_state state; - wait_queue_head_t allocwq; -}; - -/* - * struct scif_qp - Node Queue Pair - * - * Interesting structure -- a little difficult because we can only - * write across the PCIe, so any r/w pointer we need to read is - * local. We only need to read the read pointer on the inbound_q - * and read the write pointer in the outbound_q - * - * @magic: Magic value to ensure the peer sees the QP correctly - * @outbound_q: The outbound ring buffer for sending messages - * @inbound_q: The inbound ring buffer for receiving messages - * @local_write: Local write index - * @local_read: Local read index - * @remote_qp: The remote queue pair - * @local_buf: DMA address of local ring buffer - * @local_qp: DMA address of the local queue pair data structure - * @remote_buf: DMA address of remote ring buffer - * @qp_state: QP state i.e. online or offline used for P2P - * @send_lock: synchronize access to outbound queue - * @recv_lock: Synchronize access to inbound queue - */ -struct scif_qp { - u64 magic; -#define SCIFEP_MAGIC 0x5c1f000000005c1fULL - struct scif_rb outbound_q; - struct scif_rb inbound_q; - - u32 local_write __aligned(64); - u32 local_read __aligned(64); - struct scif_qp *remote_qp; - dma_addr_t local_buf; - dma_addr_t local_qp; - dma_addr_t remote_buf; - u32 qp_state; -#define SCIF_QP_OFFLINE 0xdead -#define SCIF_QP_ONLINE 0xc0de - spinlock_t send_lock; - spinlock_t recv_lock; -}; - -/* - * struct scif_loopb_msg - An element in the loopback Node QP message list. - * - * @msg - The SCIF node QP message - * @list - link in the list of messages - */ -struct scif_loopb_msg { - struct scifmsg msg; - struct list_head list; -}; - -int scif_nodeqp_send(struct scif_dev *scifdev, struct scifmsg *msg); -int _scif_nodeqp_send(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_nodeqp_intrhandler(struct scif_dev *scifdev, struct scif_qp *qp); -int scif_loopb_msg_handler(struct scif_dev *scifdev, struct scif_qp *qp); -int scif_setup_qp(struct scif_dev *scifdev); -int scif_qp_response(phys_addr_t phys, struct scif_dev *dev); -int scif_setup_qp_connect(struct scif_qp *qp, dma_addr_t *qp_offset, - int local_size, struct scif_dev *scifdev); -int scif_setup_qp_accept(struct scif_qp *qp, dma_addr_t *qp_offset, - dma_addr_t phys, int local_size, - struct scif_dev *scifdev); -int scif_setup_qp_connect_response(struct scif_dev *scifdev, - struct scif_qp *qp, u64 payload); -int scif_setup_loopback_qp(struct scif_dev *scifdev); -int scif_destroy_loopback_qp(struct scif_dev *scifdev); -void scif_poll_qp_state(struct work_struct *work); -void scif_destroy_p2p(struct scif_dev *scifdev); -void scif_send_exit(struct scif_dev *scifdev); -static inline struct device *scif_get_peer_dev(struct scif_dev *scifdev) -{ - struct scif_peer_dev *spdev; - struct device *spdev_ret; - - rcu_read_lock(); - spdev = rcu_dereference(scifdev->spdev); - if (spdev) - spdev_ret = get_device(&spdev->dev); - else - spdev_ret = ERR_PTR(-ENODEV); - rcu_read_unlock(); - return spdev_ret; -} - -static inline void scif_put_peer_dev(struct device *dev) -{ - put_device(dev); -} -#endif /* SCIF_NODEQP */ diff --git a/drivers/misc/mic/scif/scif_peer_bus.c b/drivers/misc/mic/scif/scif_peer_bus.c deleted file mode 100644 index 6d608308bb60..000000000000 --- a/drivers/misc/mic/scif/scif_peer_bus.c +++ /dev/null @@ -1,175 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel SCIF driver. - */ -#include "scif_main.h" -#include "../bus/scif_bus.h" -#include "scif_peer_bus.h" - -static inline struct scif_peer_dev * -dev_to_scif_peer(struct device *dev) -{ - return container_of(dev, struct scif_peer_dev, dev); -} - -struct bus_type scif_peer_bus = { - .name = "scif_peer_bus", -}; - -static void scif_peer_release_dev(struct device *d) -{ - struct scif_peer_dev *sdev = dev_to_scif_peer(d); - struct scif_dev *scifdev = &scif_dev[sdev->dnode]; - - scif_cleanup_scifdev(scifdev); - kfree(sdev); -} - -static int scif_peer_initialize_device(struct scif_dev *scifdev) -{ - struct scif_peer_dev *spdev; - int ret; - - spdev = kzalloc(sizeof(*spdev), GFP_KERNEL); - if (!spdev) { - ret = -ENOMEM; - goto err; - } - - spdev->dev.parent = scifdev->sdev->dev.parent; - spdev->dev.release = scif_peer_release_dev; - spdev->dnode = scifdev->node; - spdev->dev.bus = &scif_peer_bus; - dev_set_name(&spdev->dev, "scif_peer-dev%u", spdev->dnode); - - device_initialize(&spdev->dev); - get_device(&spdev->dev); - rcu_assign_pointer(scifdev->spdev, spdev); - - mutex_lock(&scif_info.conflock); - scif_info.total++; - scif_info.maxid = max_t(u32, spdev->dnode, scif_info.maxid); - mutex_unlock(&scif_info.conflock); - return 0; -err: - dev_err(&scifdev->sdev->dev, - "dnode %d: initialize_device rc %d\n", scifdev->node, ret); - return ret; -} - -static int scif_peer_add_device(struct scif_dev *scifdev) -{ - struct scif_peer_dev *spdev = rcu_dereference(scifdev->spdev); - char pool_name[16]; - int ret; - - ret = device_add(&spdev->dev); - put_device(&spdev->dev); - if (ret) { - dev_err(&scifdev->sdev->dev, - "dnode %d: peer device_add failed\n", scifdev->node); - goto put_spdev; - } - - scnprintf(pool_name, sizeof(pool_name), "scif-%d", spdev->dnode); - scifdev->signal_pool = dmam_pool_create(pool_name, &scifdev->sdev->dev, - sizeof(struct scif_status), 1, - 0); - if (!scifdev->signal_pool) { - dev_err(&scifdev->sdev->dev, - "dnode %d: dmam_pool_create failed\n", scifdev->node); - ret = -ENOMEM; - goto del_spdev; - } - dev_dbg(&spdev->dev, "Added peer dnode %d\n", spdev->dnode); - return 0; -del_spdev: - device_del(&spdev->dev); -put_spdev: - RCU_INIT_POINTER(scifdev->spdev, NULL); - synchronize_rcu(); - put_device(&spdev->dev); - - mutex_lock(&scif_info.conflock); - scif_info.total--; - mutex_unlock(&scif_info.conflock); - return ret; -} - -void scif_add_peer_device(struct work_struct *work) -{ - struct scif_dev *scifdev = container_of(work, struct scif_dev, - peer_add_work); - - scif_peer_add_device(scifdev); -} - -/* - * Peer device registration is split into a device_initialize and a device_add. - * The reason for doing this is as follows: First, peer device registration - * itself cannot be done in the message processing thread and must be delegated - * to another workqueue, otherwise if SCIF client probe, called during peer - * device registration, calls scif_connect(..), it will block the message - * processing thread causing a deadlock. Next, device_initialize is done in the - * "top-half" message processing thread and device_add in the "bottom-half" - * workqueue. If this is not done, SCIF_CNCT_REQ message processing executing - * concurrently with SCIF_INIT message processing is unable to get a reference - * on the peer device, thereby failing the connect request. - */ -void scif_peer_register_device(struct scif_dev *scifdev) -{ - int ret; - - mutex_lock(&scifdev->lock); - ret = scif_peer_initialize_device(scifdev); - if (ret) - goto exit; - schedule_work(&scifdev->peer_add_work); -exit: - mutex_unlock(&scifdev->lock); -} - -int scif_peer_unregister_device(struct scif_dev *scifdev) -{ - struct scif_peer_dev *spdev; - - mutex_lock(&scifdev->lock); - /* Flush work to ensure device register is complete */ - flush_work(&scifdev->peer_add_work); - - /* - * Continue holding scifdev->lock since theoretically unregister_device - * can be called simultaneously from multiple threads - */ - spdev = rcu_dereference(scifdev->spdev); - if (!spdev) { - mutex_unlock(&scifdev->lock); - return -ENODEV; - } - - RCU_INIT_POINTER(scifdev->spdev, NULL); - synchronize_rcu(); - mutex_unlock(&scifdev->lock); - - dev_dbg(&spdev->dev, "Removing peer dnode %d\n", spdev->dnode); - device_unregister(&spdev->dev); - - mutex_lock(&scif_info.conflock); - scif_info.total--; - mutex_unlock(&scif_info.conflock); - return 0; -} - -int scif_peer_bus_init(void) -{ - return bus_register(&scif_peer_bus); -} - -void scif_peer_bus_exit(void) -{ - bus_unregister(&scif_peer_bus); -} diff --git a/drivers/misc/mic/scif/scif_peer_bus.h b/drivers/misc/mic/scif/scif_peer_bus.h deleted file mode 100644 index 2ea4c51c18c1..000000000000 --- a/drivers/misc/mic/scif/scif_peer_bus.h +++ /dev/null @@ -1,23 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel SCIF driver. - */ -#ifndef _SCIF_PEER_BUS_H_ -#define _SCIF_PEER_BUS_H_ - -#include -#include -#include - -struct scif_dev; - -void scif_add_peer_device(struct work_struct *work); -void scif_peer_register_device(struct scif_dev *sdev); -int scif_peer_unregister_device(struct scif_dev *scifdev); -int scif_peer_bus_init(void); -void scif_peer_bus_exit(void); -#endif /* _SCIF_PEER_BUS_H */ diff --git a/drivers/misc/mic/scif/scif_ports.c b/drivers/misc/mic/scif/scif_ports.c deleted file mode 100644 index 4bdb5ef9a139..000000000000 --- a/drivers/misc/mic/scif/scif_ports.c +++ /dev/null @@ -1,116 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel SCIF driver. - */ -#include - -#include "scif_main.h" - -#define SCIF_PORT_COUNT 0x10000 /* Ports available */ - -struct idr scif_ports; - -/** - * struct scif_port - SCIF port information - * - * @ref_cnt: Reference count since there can be multiple endpoints - * created via scif_accept(..) simultaneously using a port. - */ -struct scif_port { - int ref_cnt; -}; - -/** - * __scif_get_port - Reserve a specified port # for SCIF and add it - * to the global list. - * @start: lowest port # to be reserved (inclusive). - * @end: highest port # to be reserved (exclusive). - * - * @return : Allocated SCIF port #, or -ENOSPC if port unavailable. - * On memory allocation failure, returns -ENOMEM. - */ -static int __scif_get_port(int start, int end) -{ - int id; - struct scif_port *port = kzalloc(sizeof(*port), GFP_ATOMIC); - - if (!port) - return -ENOMEM; - spin_lock(&scif_info.port_lock); - id = idr_alloc(&scif_ports, port, start, end, GFP_ATOMIC); - if (id >= 0) - port->ref_cnt++; - spin_unlock(&scif_info.port_lock); - return id; -} - -/** - * scif_rsrv_port - Reserve a specified port # for SCIF. - * @port : port # to be reserved. - * - * @return : Allocated SCIF port #, or -ENOSPC if port unavailable. - * On memory allocation failure, returns -ENOMEM. - */ -int scif_rsrv_port(u16 port) -{ - return __scif_get_port(port, port + 1); -} - -/** - * scif_get_new_port - Get and reserve any port # for SCIF in the range - * SCIF_PORT_RSVD + 1 to SCIF_PORT_COUNT - 1. - * - * @return : Allocated SCIF port #, or -ENOSPC if no ports available. - * On memory allocation failure, returns -ENOMEM. - */ -int scif_get_new_port(void) -{ - return __scif_get_port(SCIF_PORT_RSVD + 1, SCIF_PORT_COUNT); -} - -/** - * scif_get_port - Increment the reference count for a SCIF port - * @id : SCIF port - * - * @return : None - */ -void scif_get_port(u16 id) -{ - struct scif_port *port; - - if (!id) - return; - spin_lock(&scif_info.port_lock); - port = idr_find(&scif_ports, id); - if (port) - port->ref_cnt++; - spin_unlock(&scif_info.port_lock); -} - -/** - * scif_put_port - Release a reserved SCIF port - * @id : SCIF port to be released. - * - * @return : None - */ -void scif_put_port(u16 id) -{ - struct scif_port *port; - - if (!id) - return; - spin_lock(&scif_info.port_lock); - port = idr_find(&scif_ports, id); - if (port) { - port->ref_cnt--; - if (!port->ref_cnt) { - idr_remove(&scif_ports, id); - kfree(port); - } - } - spin_unlock(&scif_info.port_lock); -} diff --git a/drivers/misc/mic/scif/scif_rb.c b/drivers/misc/mic/scif/scif_rb.c deleted file mode 100644 index e425882ae06d..000000000000 --- a/drivers/misc/mic/scif/scif_rb.c +++ /dev/null @@ -1,240 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel SCIF driver. - */ -#include -#include -#include -#include - -#include "scif_rb.h" - -#define scif_rb_ring_cnt(head, tail, size) CIRC_CNT(head, tail, size) -#define scif_rb_ring_space(head, tail, size) CIRC_SPACE(head, tail, size) - -/** - * scif_rb_init - Initializes the ring buffer - * @rb: ring buffer - * @read_ptr: A pointer to the read offset - * @write_ptr: A pointer to the write offset - * @rb_base: A pointer to the base of the ring buffer - * @size: The size of the ring buffer in powers of two - */ -void scif_rb_init(struct scif_rb *rb, u32 *read_ptr, u32 *write_ptr, - void *rb_base, u8 size) -{ - rb->rb_base = rb_base; - rb->size = (1 << size); - rb->read_ptr = read_ptr; - rb->write_ptr = write_ptr; - rb->current_read_offset = *read_ptr; - rb->current_write_offset = *write_ptr; -} - -/* Copies a message to the ring buffer -- handles the wrap around case */ -static void memcpy_torb(struct scif_rb *rb, void *header, - void *msg, u32 size) -{ - u32 size1, size2; - - if (header + size >= rb->rb_base + rb->size) { - /* Need to call two copies if it wraps around */ - size1 = (u32)(rb->rb_base + rb->size - header); - size2 = size - size1; - memcpy_toio((void __iomem __force *)header, msg, size1); - memcpy_toio((void __iomem __force *)rb->rb_base, - msg + size1, size2); - } else { - memcpy_toio((void __iomem __force *)header, msg, size); - } -} - -/* Copies a message from the ring buffer -- handles the wrap around case */ -static void memcpy_fromrb(struct scif_rb *rb, void *header, - void *msg, u32 size) -{ - u32 size1, size2; - - if (header + size >= rb->rb_base + rb->size) { - /* Need to call two copies if it wraps around */ - size1 = (u32)(rb->rb_base + rb->size - header); - size2 = size - size1; - memcpy_fromio(msg, (void __iomem __force *)header, size1); - memcpy_fromio(msg + size1, - (void __iomem __force *)rb->rb_base, size2); - } else { - memcpy_fromio(msg, (void __iomem __force *)header, size); - } -} - -/** - * scif_rb_space - Query space available for writing to the RB - * @rb: ring buffer - * - * Return: size available for writing to RB in bytes. - */ -u32 scif_rb_space(struct scif_rb *rb) -{ - rb->current_read_offset = *rb->read_ptr; - /* - * Update from the HW read pointer only once the peer has exposed the - * new empty slot. This barrier is paired with the memory barrier - * scif_rb_update_read_ptr() - */ - mb(); - return scif_rb_ring_space(rb->current_write_offset, - rb->current_read_offset, rb->size); -} - -/** - * scif_rb_write - Write a message to the RB - * @rb: ring buffer - * @msg: buffer to send the message. Must be at least size bytes long - * @size: the size (in bytes) to be copied to the RB - * - * This API does not block if there isn't enough space in the RB. - * Returns: 0 on success or -ENOMEM on failure - */ -int scif_rb_write(struct scif_rb *rb, void *msg, u32 size) -{ - void *header; - - if (scif_rb_space(rb) < size) - return -ENOMEM; - header = rb->rb_base + rb->current_write_offset; - memcpy_torb(rb, header, msg, size); - /* - * Wait until scif_rb_commit(). Update the local ring - * buffer data, not the shared data until commit. - */ - rb->current_write_offset = - (rb->current_write_offset + size) & (rb->size - 1); - return 0; -} - -/** - * scif_rb_commit - To submit the message to let the peer fetch it - * @rb: ring buffer - */ -void scif_rb_commit(struct scif_rb *rb) -{ - /* - * We must ensure ordering between the all the data committed - * previously before we expose the new message to the peer by - * updating the write_ptr. This write barrier is paired with - * the read barrier in scif_rb_count(..) - */ - wmb(); - WRITE_ONCE(*rb->write_ptr, rb->current_write_offset); -#ifdef CONFIG_INTEL_MIC_CARD - /* - * X100 Si bug: For the case where a Core is performing an EXT_WR - * followed by a Doorbell Write, the Core must perform two EXT_WR to the - * same address with the same data before it does the Doorbell Write. - * This way, if ordering is violated for the Interrupt Message, it will - * fall just behind the first Posted associated with the first EXT_WR. - */ - WRITE_ONCE(*rb->write_ptr, rb->current_write_offset); -#endif -} - -/** - * scif_rb_get - To get next message from the ring buffer - * @rb: ring buffer - * @size: Number of bytes to be read - * - * Return: NULL if no bytes to be read from the ring buffer, otherwise the - * pointer to the next byte - */ -static void *scif_rb_get(struct scif_rb *rb, u32 size) -{ - void *header = NULL; - - if (scif_rb_count(rb, size) >= size) - header = rb->rb_base + rb->current_read_offset; - return header; -} - -/* - * scif_rb_get_next - Read from ring buffer. - * @rb: ring buffer - * @msg: buffer to hold the message. Must be at least size bytes long - * @size: Number of bytes to be read - * - * Return: number of bytes read if available bytes are >= size, otherwise - * returns zero. - */ -u32 scif_rb_get_next(struct scif_rb *rb, void *msg, u32 size) -{ - void *header = NULL; - int read_size = 0; - - header = scif_rb_get(rb, size); - if (header) { - u32 next_cmd_offset = - (rb->current_read_offset + size) & (rb->size - 1); - - read_size = size; - rb->current_read_offset = next_cmd_offset; - memcpy_fromrb(rb, header, msg, size); - } - return read_size; -} - -/** - * scif_rb_update_read_ptr - * @rb: ring buffer - */ -void scif_rb_update_read_ptr(struct scif_rb *rb) -{ - u32 new_offset; - - new_offset = rb->current_read_offset; - /* - * We must ensure ordering between the all the data committed or read - * previously before we expose the empty slot to the peer by updating - * the read_ptr. This barrier is paired with the memory barrier in - * scif_rb_space(..) - */ - mb(); - WRITE_ONCE(*rb->read_ptr, new_offset); -#ifdef CONFIG_INTEL_MIC_CARD - /* - * X100 Si Bug: For the case where a Core is performing an EXT_WR - * followed by a Doorbell Write, the Core must perform two EXT_WR to the - * same address with the same data before it does the Doorbell Write. - * This way, if ordering is violated for the Interrupt Message, it will - * fall just behind the first Posted associated with the first EXT_WR. - */ - WRITE_ONCE(*rb->read_ptr, new_offset); -#endif -} - -/** - * scif_rb_count - * @rb: ring buffer - * @size: Number of bytes expected to be read - * - * Return: number of bytes that can be read from the RB - */ -u32 scif_rb_count(struct scif_rb *rb, u32 size) -{ - if (scif_rb_ring_cnt(rb->current_write_offset, - rb->current_read_offset, - rb->size) < size) { - rb->current_write_offset = *rb->write_ptr; - /* - * Update from the HW write pointer if empty only once the peer - * has exposed the new message. This read barrier is paired - * with the write barrier in scif_rb_commit(..) - */ - smp_rmb(); - } - return scif_rb_ring_cnt(rb->current_write_offset, - rb->current_read_offset, - rb->size); -} diff --git a/drivers/misc/mic/scif/scif_rb.h b/drivers/misc/mic/scif/scif_rb.h deleted file mode 100644 index 166dffe3093d..000000000000 --- a/drivers/misc/mic/scif/scif_rb.h +++ /dev/null @@ -1,100 +0,0 @@ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * This file is provided under a dual BSD/GPLv2 license. When using or - * redistributing this file, you may do so under either license. - * - * GPL LICENSE SUMMARY - * - * Copyright(c) 2014 Intel Corporation. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * BSD LICENSE - * - * Copyright(c) 2014 Intel Corporation. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * - * * Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * * Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in - * the documentation and/or other materials provided with the - * distribution. - * * Neither the name of Intel Corporation nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR - * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT - * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, - * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT - * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - * - * Intel SCIF driver. - */ -#ifndef SCIF_RB_H -#define SCIF_RB_H -/* - * This file describes a general purpose, byte based ring buffer. Writers to the - * ring buffer need to synchronize using a lock. The same is true for readers, - * although in practice, the ring buffer has a single reader. It is lockless - * between producer and consumer so it can handle being used across the PCIe - * bus. The ring buffer ensures that there are no reads across the PCIe bus for - * performance reasons. Two of these are used to form a single bidirectional - * queue-pair across PCIe. - */ -/* - * struct scif_rb - SCIF Ring Buffer - * - * @rb_base: The base of the memory used for storing RB messages - * @read_ptr: Pointer to the read offset - * @write_ptr: Pointer to the write offset - * @size: Size of the memory in rb_base - * @current_read_offset: Cached read offset for performance - * @current_write_offset: Cached write offset for performance - */ -struct scif_rb { - void *rb_base; - u32 *read_ptr; - u32 *write_ptr; - u32 size; - u32 current_read_offset; - u32 current_write_offset; -}; - -/* methods used by both */ -void scif_rb_init(struct scif_rb *rb, u32 *read_ptr, u32 *write_ptr, - void *rb_base, u8 size); -/* writer only methods */ -/* write a new command, then scif_rb_commit() */ -int scif_rb_write(struct scif_rb *rb, void *msg, u32 size); -/* after write(), then scif_rb_commit() */ -void scif_rb_commit(struct scif_rb *rb); -/* query space available for writing to a RB. */ -u32 scif_rb_space(struct scif_rb *rb); - -/* reader only methods */ -/* read a new message from the ring buffer of size bytes */ -u32 scif_rb_get_next(struct scif_rb *rb, void *msg, u32 size); -/* update the read pointer so that the space can be reused */ -void scif_rb_update_read_ptr(struct scif_rb *rb); -/* count the number of bytes that can be read */ -u32 scif_rb_count(struct scif_rb *rb, u32 size); -#endif diff --git a/drivers/misc/mic/scif/scif_rma.c b/drivers/misc/mic/scif/scif_rma.c deleted file mode 100644 index 18fb9d8b8a4b..000000000000 --- a/drivers/misc/mic/scif/scif_rma.c +++ /dev/null @@ -1,1760 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2015 Intel Corporation. - * - * Intel SCIF driver. - */ -#include -#include -#include -#include - -#include "scif_main.h" -#include "scif_map.h" - -/* Used to skip ulimit checks for registrations with SCIF_MAP_KERNEL flag */ -#define SCIF_MAP_ULIMIT 0x40 - -bool scif_ulimit_check = 1; - -/** - * scif_rma_ep_init: - * @ep: end point - * - * Initialize RMA per EP data structures. - */ -void scif_rma_ep_init(struct scif_endpt *ep) -{ - struct scif_endpt_rma_info *rma = &ep->rma_info; - - mutex_init(&rma->rma_lock); - init_iova_domain(&rma->iovad, PAGE_SIZE, SCIF_IOVA_START_PFN); - spin_lock_init(&rma->tc_lock); - mutex_init(&rma->mmn_lock); - INIT_LIST_HEAD(&rma->reg_list); - INIT_LIST_HEAD(&rma->remote_reg_list); - atomic_set(&rma->tw_refcount, 0); - atomic_set(&rma->tcw_refcount, 0); - atomic_set(&rma->tcw_total_pages, 0); - atomic_set(&rma->fence_refcount, 0); - - rma->async_list_del = 0; - rma->dma_chan = NULL; - INIT_LIST_HEAD(&rma->mmn_list); - INIT_LIST_HEAD(&rma->vma_list); - init_waitqueue_head(&rma->markwq); -} - -/** - * scif_rma_ep_can_uninit: - * @ep: end point - * - * Returns 1 if an endpoint can be uninitialized and 0 otherwise. - */ -int scif_rma_ep_can_uninit(struct scif_endpt *ep) -{ - int ret = 0; - - mutex_lock(&ep->rma_info.rma_lock); - /* Destroy RMA Info only if both lists are empty */ - if (list_empty(&ep->rma_info.reg_list) && - list_empty(&ep->rma_info.remote_reg_list) && - list_empty(&ep->rma_info.mmn_list) && - !atomic_read(&ep->rma_info.tw_refcount) && - !atomic_read(&ep->rma_info.tcw_refcount) && - !atomic_read(&ep->rma_info.fence_refcount)) - ret = 1; - mutex_unlock(&ep->rma_info.rma_lock); - return ret; -} - -/** - * scif_create_pinned_pages: - * @nr_pages: number of pages in window - * @prot: read/write protection - * - * Allocate and prepare a set of pinned pages. - */ -static struct scif_pinned_pages * -scif_create_pinned_pages(int nr_pages, int prot) -{ - struct scif_pinned_pages *pin; - - might_sleep(); - pin = scif_zalloc(sizeof(*pin)); - if (!pin) - goto error; - - pin->pages = scif_zalloc(nr_pages * sizeof(*pin->pages)); - if (!pin->pages) - goto error_free_pinned_pages; - - pin->prot = prot; - pin->magic = SCIFEP_MAGIC; - return pin; - -error_free_pinned_pages: - scif_free(pin, sizeof(*pin)); -error: - return NULL; -} - -/** - * scif_destroy_pinned_pages: - * @pin: A set of pinned pages. - * - * Deallocate resources for pinned pages. - */ -static int scif_destroy_pinned_pages(struct scif_pinned_pages *pin) -{ - int j; - int writeable = pin->prot & SCIF_PROT_WRITE; - int kernel = SCIF_MAP_KERNEL & pin->map_flags; - - if (kernel) { - for (j = 0; j < pin->nr_pages; j++) { - if (pin->pages[j] && !kernel) { - if (writeable) - set_page_dirty_lock(pin->pages[j]); - put_page(pin->pages[j]); - } - } - } else - unpin_user_pages_dirty_lock(pin->pages, pin->nr_pages, - writeable); - scif_free(pin->pages, - pin->nr_pages * sizeof(*pin->pages)); - scif_free(pin, sizeof(*pin)); - return 0; -} - -/* - * scif_create_window: - * @ep: end point - * @nr_pages: number of pages - * @offset: registration offset - * @temp: true if a temporary window is being created - * - * Allocate and prepare a self registration window. - */ -struct scif_window *scif_create_window(struct scif_endpt *ep, int nr_pages, - s64 offset, bool temp) -{ - struct scif_window *window; - - might_sleep(); - window = scif_zalloc(sizeof(*window)); - if (!window) - goto error; - - window->dma_addr = scif_zalloc(nr_pages * sizeof(*window->dma_addr)); - if (!window->dma_addr) - goto error_free_window; - - window->num_pages = scif_zalloc(nr_pages * sizeof(*window->num_pages)); - if (!window->num_pages) - goto error_free_window; - - window->offset = offset; - window->ep = (u64)ep; - window->magic = SCIFEP_MAGIC; - window->reg_state = OP_IDLE; - init_waitqueue_head(&window->regwq); - window->unreg_state = OP_IDLE; - init_waitqueue_head(&window->unregwq); - INIT_LIST_HEAD(&window->list); - window->type = SCIF_WINDOW_SELF; - window->temp = temp; - return window; - -error_free_window: - scif_free(window->dma_addr, - nr_pages * sizeof(*window->dma_addr)); - scif_free(window, sizeof(*window)); -error: - return NULL; -} - -/** - * scif_destroy_incomplete_window: - * @ep: end point - * @window: registration window - * - * Deallocate resources for self window. - */ -static void scif_destroy_incomplete_window(struct scif_endpt *ep, - struct scif_window *window) -{ - int err; - int nr_pages = window->nr_pages; - struct scif_allocmsg *alloc = &window->alloc_handle; - struct scifmsg msg; - -retry: - /* Wait for a SCIF_ALLOC_GNT/REJ message */ - err = wait_event_timeout(alloc->allocwq, - alloc->state != OP_IN_PROGRESS, - SCIF_NODE_ALIVE_TIMEOUT); - if (!err && scifdev_alive(ep)) - goto retry; - - mutex_lock(&ep->rma_info.rma_lock); - if (alloc->state == OP_COMPLETED) { - msg.uop = SCIF_FREE_VIRT; - msg.src = ep->port; - msg.payload[0] = ep->remote_ep; - msg.payload[1] = window->alloc_handle.vaddr; - msg.payload[2] = (u64)window; - msg.payload[3] = SCIF_REGISTER; - _scif_nodeqp_send(ep->remote_dev, &msg); - } - mutex_unlock(&ep->rma_info.rma_lock); - - scif_free_window_offset(ep, window, window->offset); - scif_free(window->dma_addr, nr_pages * sizeof(*window->dma_addr)); - scif_free(window->num_pages, nr_pages * sizeof(*window->num_pages)); - scif_free(window, sizeof(*window)); -} - -/** - * scif_unmap_window: - * @remote_dev: SCIF remote device - * @window: registration window - * - * Delete any DMA mappings created for a registered self window - */ -void scif_unmap_window(struct scif_dev *remote_dev, struct scif_window *window) -{ - int j; - - if (scif_is_iommu_enabled() && !scifdev_self(remote_dev)) { - if (window->st) { - dma_unmap_sg(&remote_dev->sdev->dev, - window->st->sgl, window->st->nents, - DMA_BIDIRECTIONAL); - sg_free_table(window->st); - kfree(window->st); - window->st = NULL; - } - } else { - for (j = 0; j < window->nr_contig_chunks; j++) { - if (window->dma_addr[j]) { - scif_unmap_single(window->dma_addr[j], - remote_dev, - window->num_pages[j] << - PAGE_SHIFT); - window->dma_addr[j] = 0x0; - } - } - } -} - -static inline struct mm_struct *__scif_acquire_mm(void) -{ - if (scif_ulimit_check) - return get_task_mm(current); - return NULL; -} - -static inline void __scif_release_mm(struct mm_struct *mm) -{ - if (mm) - mmput(mm); -} - -static inline int -__scif_dec_pinned_vm_lock(struct mm_struct *mm, - int nr_pages) -{ - if (!mm || !nr_pages || !scif_ulimit_check) - return 0; - - atomic64_sub(nr_pages, &mm->pinned_vm); - return 0; -} - -static inline int __scif_check_inc_pinned_vm(struct mm_struct *mm, - int nr_pages) -{ - unsigned long locked, lock_limit; - - if (!mm || !nr_pages || !scif_ulimit_check) - return 0; - - lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT; - locked = atomic64_add_return(nr_pages, &mm->pinned_vm); - - if ((locked > lock_limit) && !capable(CAP_IPC_LOCK)) { - atomic64_sub(nr_pages, &mm->pinned_vm); - dev_err(scif_info.mdev.this_device, - "locked(%lu) > lock_limit(%lu)\n", - locked, lock_limit); - return -ENOMEM; - } - return 0; -} - -/** - * scif_destroy_window: - * @ep: end point - * @window: registration window - * - * Deallocate resources for self window. - */ -int scif_destroy_window(struct scif_endpt *ep, struct scif_window *window) -{ - int j; - struct scif_pinned_pages *pinned_pages = window->pinned_pages; - int nr_pages = window->nr_pages; - - might_sleep(); - if (!window->temp && window->mm) { - __scif_dec_pinned_vm_lock(window->mm, window->nr_pages); - __scif_release_mm(window->mm); - window->mm = NULL; - } - - scif_free_window_offset(ep, window, window->offset); - scif_unmap_window(ep->remote_dev, window); - /* - * Decrement references for this set of pinned pages from - * this window. - */ - j = atomic_sub_return(1, &pinned_pages->ref_count); - if (j < 0) - dev_err(scif_info.mdev.this_device, - "%s %d incorrect ref count %d\n", - __func__, __LINE__, j); - /* - * If the ref count for pinned_pages is zero then someone - * has already called scif_unpin_pages() for it and we should - * destroy the page cache. - */ - if (!j) - scif_destroy_pinned_pages(window->pinned_pages); - scif_free(window->dma_addr, nr_pages * sizeof(*window->dma_addr)); - scif_free(window->num_pages, nr_pages * sizeof(*window->num_pages)); - window->magic = 0; - scif_free(window, sizeof(*window)); - return 0; -} - -/** - * scif_create_remote_lookup: - * @remote_dev: SCIF remote device - * @window: remote window - * - * Allocate and prepare lookup entries for the remote - * end to copy over the physical addresses. - * Returns 0 on success and appropriate errno on failure. - */ -static int scif_create_remote_lookup(struct scif_dev *remote_dev, - struct scif_window *window) -{ - int i, j, err = 0; - int nr_pages = window->nr_pages; - bool vmalloc_dma_phys, vmalloc_num_pages; - - might_sleep(); - /* Map window */ - err = scif_map_single(&window->mapped_offset, - window, remote_dev, sizeof(*window)); - if (err) - goto error_window; - - /* Compute the number of lookup entries. 21 == 2MB Shift */ - window->nr_lookup = ALIGN(nr_pages * PAGE_SIZE, - ((2) * 1024 * 1024)) >> 21; - - window->dma_addr_lookup.lookup = - scif_alloc_coherent(&window->dma_addr_lookup.offset, - remote_dev, window->nr_lookup * - sizeof(*window->dma_addr_lookup.lookup), - GFP_KERNEL | __GFP_ZERO); - if (!window->dma_addr_lookup.lookup) { - err = -ENOMEM; - goto error_window; - } - - window->num_pages_lookup.lookup = - scif_alloc_coherent(&window->num_pages_lookup.offset, - remote_dev, window->nr_lookup * - sizeof(*window->num_pages_lookup.lookup), - GFP_KERNEL | __GFP_ZERO); - if (!window->num_pages_lookup.lookup) { - err = -ENOMEM; - goto error_window; - } - - vmalloc_dma_phys = is_vmalloc_addr(&window->dma_addr[0]); - vmalloc_num_pages = is_vmalloc_addr(&window->num_pages[0]); - - /* Now map each of the pages containing physical addresses */ - for (i = 0, j = 0; i < nr_pages; i += SCIF_NR_ADDR_IN_PAGE, j++) { - err = scif_map_page(&window->dma_addr_lookup.lookup[j], - vmalloc_dma_phys ? - vmalloc_to_page(&window->dma_addr[i]) : - virt_to_page(&window->dma_addr[i]), - remote_dev); - if (err) - goto error_window; - err = scif_map_page(&window->num_pages_lookup.lookup[j], - vmalloc_num_pages ? - vmalloc_to_page(&window->num_pages[i]) : - virt_to_page(&window->num_pages[i]), - remote_dev); - if (err) - goto error_window; - } - return 0; -error_window: - return err; -} - -/** - * scif_destroy_remote_lookup: - * @remote_dev: SCIF remote device - * @window: remote window - * - * Destroy lookup entries used for the remote - * end to copy over the physical addresses. - */ -static void scif_destroy_remote_lookup(struct scif_dev *remote_dev, - struct scif_window *window) -{ - int i, j; - - if (window->nr_lookup) { - struct scif_rma_lookup *lup = &window->dma_addr_lookup; - struct scif_rma_lookup *npup = &window->num_pages_lookup; - - for (i = 0, j = 0; i < window->nr_pages; - i += SCIF_NR_ADDR_IN_PAGE, j++) { - if (lup->lookup && lup->lookup[j]) - scif_unmap_single(lup->lookup[j], - remote_dev, - PAGE_SIZE); - if (npup->lookup && npup->lookup[j]) - scif_unmap_single(npup->lookup[j], - remote_dev, - PAGE_SIZE); - } - if (lup->lookup) - scif_free_coherent(lup->lookup, lup->offset, - remote_dev, window->nr_lookup * - sizeof(*lup->lookup)); - if (npup->lookup) - scif_free_coherent(npup->lookup, npup->offset, - remote_dev, window->nr_lookup * - sizeof(*npup->lookup)); - if (window->mapped_offset) - scif_unmap_single(window->mapped_offset, - remote_dev, sizeof(*window)); - window->nr_lookup = 0; - } -} - -/** - * scif_create_remote_window: - * @scifdev: SCIF device - * @nr_pages: number of pages in window - * - * Allocate and prepare a remote registration window. - */ -static struct scif_window * -scif_create_remote_window(struct scif_dev *scifdev, int nr_pages) -{ - struct scif_window *window; - - might_sleep(); - window = scif_zalloc(sizeof(*window)); - if (!window) - goto error_ret; - - window->magic = SCIFEP_MAGIC; - window->nr_pages = nr_pages; - - window->dma_addr = scif_zalloc(nr_pages * sizeof(*window->dma_addr)); - if (!window->dma_addr) - goto error_window; - - window->num_pages = scif_zalloc(nr_pages * - sizeof(*window->num_pages)); - if (!window->num_pages) - goto error_window; - - if (scif_create_remote_lookup(scifdev, window)) - goto error_window; - - window->type = SCIF_WINDOW_PEER; - window->unreg_state = OP_IDLE; - INIT_LIST_HEAD(&window->list); - return window; -error_window: - scif_destroy_remote_window(window); -error_ret: - return NULL; -} - -/** - * scif_destroy_remote_window: - * @window: remote registration window - * - * Deallocate resources for remote window. - */ -void -scif_destroy_remote_window(struct scif_window *window) -{ - scif_free(window->dma_addr, window->nr_pages * - sizeof(*window->dma_addr)); - scif_free(window->num_pages, window->nr_pages * - sizeof(*window->num_pages)); - window->magic = 0; - scif_free(window, sizeof(*window)); -} - -/** - * scif_iommu_map: create DMA mappings if the IOMMU is enabled - * @remote_dev: SCIF remote device - * @window: remote registration window - * - * Map the physical pages using dma_map_sg(..) and then detect the number - * of contiguous DMA mappings allocated - */ -static int scif_iommu_map(struct scif_dev *remote_dev, - struct scif_window *window) -{ - struct scatterlist *sg; - int i, err; - scif_pinned_pages_t pin = window->pinned_pages; - - window->st = kzalloc(sizeof(*window->st), GFP_KERNEL); - if (!window->st) - return -ENOMEM; - - err = sg_alloc_table(window->st, window->nr_pages, GFP_KERNEL); - if (err) - return err; - - for_each_sg(window->st->sgl, sg, window->st->nents, i) - sg_set_page(sg, pin->pages[i], PAGE_SIZE, 0x0); - - err = dma_map_sg(&remote_dev->sdev->dev, window->st->sgl, - window->st->nents, DMA_BIDIRECTIONAL); - if (!err) - return -ENOMEM; - /* Detect contiguous ranges of DMA mappings */ - sg = window->st->sgl; - for (i = 0; sg; i++) { - dma_addr_t last_da; - - window->dma_addr[i] = sg_dma_address(sg); - window->num_pages[i] = sg_dma_len(sg) >> PAGE_SHIFT; - last_da = sg_dma_address(sg) + sg_dma_len(sg); - while ((sg = sg_next(sg)) && sg_dma_address(sg) == last_da) { - window->num_pages[i] += - (sg_dma_len(sg) >> PAGE_SHIFT); - last_da = window->dma_addr[i] + - sg_dma_len(sg); - } - window->nr_contig_chunks++; - } - return 0; -} - -/** - * scif_map_window: - * @remote_dev: SCIF remote device - * @window: self registration window - * - * Map pages of a window into the aperture/PCI. - * Also determine addresses required for DMA. - */ -int -scif_map_window(struct scif_dev *remote_dev, struct scif_window *window) -{ - int i, j, k, err = 0, nr_contig_pages; - scif_pinned_pages_t pin; - phys_addr_t phys_prev, phys_curr; - - might_sleep(); - - pin = window->pinned_pages; - - if (intel_iommu_enabled && !scifdev_self(remote_dev)) - return scif_iommu_map(remote_dev, window); - - for (i = 0, j = 0; i < window->nr_pages; i += nr_contig_pages, j++) { - phys_prev = page_to_phys(pin->pages[i]); - nr_contig_pages = 1; - - /* Detect physically contiguous chunks */ - for (k = i + 1; k < window->nr_pages; k++) { - phys_curr = page_to_phys(pin->pages[k]); - if (phys_curr != (phys_prev + PAGE_SIZE)) - break; - phys_prev = phys_curr; - nr_contig_pages++; - } - window->num_pages[j] = nr_contig_pages; - window->nr_contig_chunks++; - if (scif_is_mgmt_node()) { - /* - * Management node has to deal with SMPT on X100 and - * hence the DMA mapping is required - */ - err = scif_map_single(&window->dma_addr[j], - phys_to_virt(page_to_phys( - pin->pages[i])), - remote_dev, - nr_contig_pages << PAGE_SHIFT); - if (err) - return err; - } else { - window->dma_addr[j] = page_to_phys(pin->pages[i]); - } - } - return err; -} - -/** - * scif_send_scif_unregister: - * @ep: end point - * @window: self registration window - * - * Send a SCIF_UNREGISTER message. - */ -static int scif_send_scif_unregister(struct scif_endpt *ep, - struct scif_window *window) -{ - struct scifmsg msg; - - msg.uop = SCIF_UNREGISTER; - msg.src = ep->port; - msg.payload[0] = window->alloc_handle.vaddr; - msg.payload[1] = (u64)window; - return scif_nodeqp_send(ep->remote_dev, &msg); -} - -/** - * scif_unregister_window: - * @window: self registration window - * - * Send an unregistration request and wait for a response. - */ -int scif_unregister_window(struct scif_window *window) -{ - int err = 0; - struct scif_endpt *ep = (struct scif_endpt *)window->ep; - bool send_msg = false; - - might_sleep(); - switch (window->unreg_state) { - case OP_IDLE: - { - window->unreg_state = OP_IN_PROGRESS; - send_msg = true; - } - fallthrough; - case OP_IN_PROGRESS: - { - scif_get_window(window, 1); - mutex_unlock(&ep->rma_info.rma_lock); - if (send_msg) { - err = scif_send_scif_unregister(ep, window); - if (err) { - window->unreg_state = OP_COMPLETED; - goto done; - } - } else { - /* Return ENXIO since unregistration is in progress */ - mutex_lock(&ep->rma_info.rma_lock); - return -ENXIO; - } -retry: - /* Wait for a SCIF_UNREGISTER_(N)ACK message */ - err = wait_event_timeout(window->unregwq, - window->unreg_state != OP_IN_PROGRESS, - SCIF_NODE_ALIVE_TIMEOUT); - if (!err && scifdev_alive(ep)) - goto retry; - if (!err) { - err = -ENODEV; - window->unreg_state = OP_COMPLETED; - dev_err(scif_info.mdev.this_device, - "%s %d err %d\n", __func__, __LINE__, err); - } - if (err > 0) - err = 0; -done: - mutex_lock(&ep->rma_info.rma_lock); - scif_put_window(window, 1); - break; - } - case OP_FAILED: - { - if (!scifdev_alive(ep)) { - err = -ENODEV; - window->unreg_state = OP_COMPLETED; - } - break; - } - case OP_COMPLETED: - break; - default: - err = -ENODEV; - } - - if (window->unreg_state == OP_COMPLETED && window->ref_count) - scif_put_window(window, window->nr_pages); - - if (!window->ref_count) { - atomic_inc(&ep->rma_info.tw_refcount); - list_del_init(&window->list); - scif_free_window_offset(ep, window, window->offset); - mutex_unlock(&ep->rma_info.rma_lock); - if ((!!(window->pinned_pages->map_flags & SCIF_MAP_KERNEL)) && - scifdev_alive(ep)) { - scif_drain_dma_intr(ep->remote_dev->sdev, - ep->rma_info.dma_chan); - } else { - if (!__scif_dec_pinned_vm_lock(window->mm, - window->nr_pages)) { - __scif_release_mm(window->mm); - window->mm = NULL; - } - } - scif_queue_for_cleanup(window, &scif_info.rma); - mutex_lock(&ep->rma_info.rma_lock); - } - return err; -} - -/** - * scif_send_alloc_request: - * @ep: end point - * @window: self registration window - * - * Send a remote window allocation request - */ -static int scif_send_alloc_request(struct scif_endpt *ep, - struct scif_window *window) -{ - struct scifmsg msg; - struct scif_allocmsg *alloc = &window->alloc_handle; - - /* Set up the Alloc Handle */ - alloc->state = OP_IN_PROGRESS; - init_waitqueue_head(&alloc->allocwq); - - /* Send out an allocation request */ - msg.uop = SCIF_ALLOC_REQ; - msg.payload[1] = window->nr_pages; - msg.payload[2] = (u64)&window->alloc_handle; - return _scif_nodeqp_send(ep->remote_dev, &msg); -} - -/** - * scif_prep_remote_window: - * @ep: end point - * @window: self registration window - * - * Send a remote window allocation request, wait for an allocation response, - * and prepares the remote window by copying over the page lists - */ -static int scif_prep_remote_window(struct scif_endpt *ep, - struct scif_window *window) -{ - struct scifmsg msg; - struct scif_window *remote_window; - struct scif_allocmsg *alloc = &window->alloc_handle; - dma_addr_t *dma_phys_lookup, *tmp, *num_pages_lookup, *tmp1; - int i = 0, j = 0; - int nr_contig_chunks, loop_nr_contig_chunks; - int remaining_nr_contig_chunks, nr_lookup; - int err, map_err; - - map_err = scif_map_window(ep->remote_dev, window); - if (map_err) - dev_err(&ep->remote_dev->sdev->dev, - "%s %d map_err %d\n", __func__, __LINE__, map_err); - remaining_nr_contig_chunks = window->nr_contig_chunks; - nr_contig_chunks = window->nr_contig_chunks; -retry: - /* Wait for a SCIF_ALLOC_GNT/REJ message */ - err = wait_event_timeout(alloc->allocwq, - alloc->state != OP_IN_PROGRESS, - SCIF_NODE_ALIVE_TIMEOUT); - mutex_lock(&ep->rma_info.rma_lock); - /* Synchronize with the thread waking up allocwq */ - mutex_unlock(&ep->rma_info.rma_lock); - if (!err && scifdev_alive(ep)) - goto retry; - - if (!err) - err = -ENODEV; - - if (err > 0) - err = 0; - else - return err; - - /* Bail out. The remote end rejected this request */ - if (alloc->state == OP_FAILED) - return -ENOMEM; - - if (map_err) { - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, map_err); - msg.uop = SCIF_FREE_VIRT; - msg.src = ep->port; - msg.payload[0] = ep->remote_ep; - msg.payload[1] = window->alloc_handle.vaddr; - msg.payload[2] = (u64)window; - msg.payload[3] = SCIF_REGISTER; - spin_lock(&ep->lock); - if (ep->state == SCIFEP_CONNECTED) - err = _scif_nodeqp_send(ep->remote_dev, &msg); - else - err = -ENOTCONN; - spin_unlock(&ep->lock); - return err; - } - - remote_window = scif_ioremap(alloc->phys_addr, sizeof(*window), - ep->remote_dev); - - /* Compute the number of lookup entries. 21 == 2MB Shift */ - nr_lookup = ALIGN(nr_contig_chunks, SCIF_NR_ADDR_IN_PAGE) - >> ilog2(SCIF_NR_ADDR_IN_PAGE); - - dma_phys_lookup = - scif_ioremap(remote_window->dma_addr_lookup.offset, - nr_lookup * - sizeof(*remote_window->dma_addr_lookup.lookup), - ep->remote_dev); - num_pages_lookup = - scif_ioremap(remote_window->num_pages_lookup.offset, - nr_lookup * - sizeof(*remote_window->num_pages_lookup.lookup), - ep->remote_dev); - - while (remaining_nr_contig_chunks) { - loop_nr_contig_chunks = min_t(int, remaining_nr_contig_chunks, - (int)SCIF_NR_ADDR_IN_PAGE); - /* #1/2 - Copy physical addresses over to the remote side */ - - /* #2/2 - Copy DMA addresses (addresses that are fed into the - * DMA engine) We transfer bus addresses which are then - * converted into a MIC physical address on the remote - * side if it is a MIC, if the remote node is a mgmt node we - * transfer the MIC physical address - */ - tmp = scif_ioremap(dma_phys_lookup[j], - loop_nr_contig_chunks * - sizeof(*window->dma_addr), - ep->remote_dev); - tmp1 = scif_ioremap(num_pages_lookup[j], - loop_nr_contig_chunks * - sizeof(*window->num_pages), - ep->remote_dev); - if (scif_is_mgmt_node()) { - memcpy_toio((void __force __iomem *)tmp, - &window->dma_addr[i], loop_nr_contig_chunks - * sizeof(*window->dma_addr)); - memcpy_toio((void __force __iomem *)tmp1, - &window->num_pages[i], loop_nr_contig_chunks - * sizeof(*window->num_pages)); - } else { - if (scifdev_is_p2p(ep->remote_dev)) { - /* - * add remote node's base address for this node - * to convert it into a MIC address - */ - int m; - dma_addr_t dma_addr; - - for (m = 0; m < loop_nr_contig_chunks; m++) { - dma_addr = window->dma_addr[i + m] + - ep->remote_dev->base_addr; - writeq(dma_addr, - (void __force __iomem *)&tmp[m]); - } - memcpy_toio((void __force __iomem *)tmp1, - &window->num_pages[i], - loop_nr_contig_chunks - * sizeof(*window->num_pages)); - } else { - /* Mgmt node or loopback - transfer DMA - * addresses as is, this is the same as a - * MIC physical address (we use the dma_addr - * and not the phys_addr array since the - * phys_addr is only setup if there is a mmap() - * request from the mgmt node) - */ - memcpy_toio((void __force __iomem *)tmp, - &window->dma_addr[i], - loop_nr_contig_chunks * - sizeof(*window->dma_addr)); - memcpy_toio((void __force __iomem *)tmp1, - &window->num_pages[i], - loop_nr_contig_chunks * - sizeof(*window->num_pages)); - } - } - remaining_nr_contig_chunks -= loop_nr_contig_chunks; - i += loop_nr_contig_chunks; - j++; - scif_iounmap(tmp, loop_nr_contig_chunks * - sizeof(*window->dma_addr), ep->remote_dev); - scif_iounmap(tmp1, loop_nr_contig_chunks * - sizeof(*window->num_pages), ep->remote_dev); - } - - /* Prepare the remote window for the peer */ - remote_window->peer_window = (u64)window; - remote_window->offset = window->offset; - remote_window->prot = window->prot; - remote_window->nr_contig_chunks = nr_contig_chunks; - remote_window->ep = ep->remote_ep; - scif_iounmap(num_pages_lookup, - nr_lookup * - sizeof(*remote_window->num_pages_lookup.lookup), - ep->remote_dev); - scif_iounmap(dma_phys_lookup, - nr_lookup * - sizeof(*remote_window->dma_addr_lookup.lookup), - ep->remote_dev); - scif_iounmap(remote_window, sizeof(*remote_window), ep->remote_dev); - window->peer_window = alloc->vaddr; - return err; -} - -/** - * scif_send_scif_register: - * @ep: end point - * @window: self registration window - * - * Send a SCIF_REGISTER message if EP is connected and wait for a - * SCIF_REGISTER_(N)ACK message else send a SCIF_FREE_VIRT - * message so that the peer can free its remote window allocated earlier. - */ -static int scif_send_scif_register(struct scif_endpt *ep, - struct scif_window *window) -{ - int err = 0; - struct scifmsg msg; - - msg.src = ep->port; - msg.payload[0] = ep->remote_ep; - msg.payload[1] = window->alloc_handle.vaddr; - msg.payload[2] = (u64)window; - spin_lock(&ep->lock); - if (ep->state == SCIFEP_CONNECTED) { - msg.uop = SCIF_REGISTER; - window->reg_state = OP_IN_PROGRESS; - err = _scif_nodeqp_send(ep->remote_dev, &msg); - spin_unlock(&ep->lock); - if (!err) { -retry: - /* Wait for a SCIF_REGISTER_(N)ACK message */ - err = wait_event_timeout(window->regwq, - window->reg_state != - OP_IN_PROGRESS, - SCIF_NODE_ALIVE_TIMEOUT); - if (!err && scifdev_alive(ep)) - goto retry; - err = !err ? -ENODEV : 0; - if (window->reg_state == OP_FAILED) - err = -ENOTCONN; - } - } else { - msg.uop = SCIF_FREE_VIRT; - msg.payload[3] = SCIF_REGISTER; - err = _scif_nodeqp_send(ep->remote_dev, &msg); - spin_unlock(&ep->lock); - if (!err) - err = -ENOTCONN; - } - return err; -} - -/** - * scif_get_window_offset: - * @ep: end point descriptor - * @flags: flags - * @offset: offset hint - * @num_pages: number of pages - * @out_offset: computed offset returned by reference. - * - * Compute/Claim a new offset for this EP. - */ -int scif_get_window_offset(struct scif_endpt *ep, int flags, s64 offset, - int num_pages, s64 *out_offset) -{ - s64 page_index; - struct iova *iova_ptr; - int err = 0; - - if (flags & SCIF_MAP_FIXED) { - page_index = SCIF_IOVA_PFN(offset); - iova_ptr = reserve_iova(&ep->rma_info.iovad, page_index, - page_index + num_pages - 1); - if (!iova_ptr) - err = -EADDRINUSE; - } else { - iova_ptr = alloc_iova(&ep->rma_info.iovad, num_pages, - SCIF_DMA_63BIT_PFN - 1, 0); - if (!iova_ptr) - err = -ENOMEM; - } - if (!err) - *out_offset = (iova_ptr->pfn_lo) << PAGE_SHIFT; - return err; -} - -/** - * scif_free_window_offset: - * @ep: end point descriptor - * @window: registration window - * @offset: Offset to be freed - * - * Free offset for this EP. The callee is supposed to grab - * the RMA mutex before calling this API. - */ -void scif_free_window_offset(struct scif_endpt *ep, - struct scif_window *window, s64 offset) -{ - if ((window && !window->offset_freed) || !window) { - free_iova(&ep->rma_info.iovad, offset >> PAGE_SHIFT); - if (window) - window->offset_freed = true; - } -} - -/** - * scif_alloc_req: Respond to SCIF_ALLOC_REQ interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * Remote side is requesting a memory allocation. - */ -void scif_alloc_req(struct scif_dev *scifdev, struct scifmsg *msg) -{ - int err; - struct scif_window *window = NULL; - int nr_pages = msg->payload[1]; - - window = scif_create_remote_window(scifdev, nr_pages); - if (!window) { - err = -ENOMEM; - goto error; - } - - /* The peer's allocation request is granted */ - msg->uop = SCIF_ALLOC_GNT; - msg->payload[0] = (u64)window; - msg->payload[1] = window->mapped_offset; - err = scif_nodeqp_send(scifdev, msg); - if (err) - scif_destroy_remote_window(window); - return; -error: - /* The peer's allocation request is rejected */ - dev_err(&scifdev->sdev->dev, - "%s %d error %d alloc_ptr %p nr_pages 0x%x\n", - __func__, __LINE__, err, window, nr_pages); - msg->uop = SCIF_ALLOC_REJ; - scif_nodeqp_send(scifdev, msg); -} - -/** - * scif_alloc_gnt_rej: Respond to SCIF_ALLOC_GNT/REJ interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * Remote side responded to a memory allocation. - */ -void scif_alloc_gnt_rej(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_allocmsg *handle = (struct scif_allocmsg *)msg->payload[2]; - struct scif_window *window = container_of(handle, struct scif_window, - alloc_handle); - struct scif_endpt *ep = (struct scif_endpt *)window->ep; - - mutex_lock(&ep->rma_info.rma_lock); - handle->vaddr = msg->payload[0]; - handle->phys_addr = msg->payload[1]; - if (msg->uop == SCIF_ALLOC_GNT) - handle->state = OP_COMPLETED; - else - handle->state = OP_FAILED; - wake_up(&handle->allocwq); - mutex_unlock(&ep->rma_info.rma_lock); -} - -/** - * scif_free_virt: Respond to SCIF_FREE_VIRT interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * Free up memory kmalloc'd earlier. - */ -void scif_free_virt(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_window *window = (struct scif_window *)msg->payload[1]; - - scif_destroy_remote_window(window); -} - -static void -scif_fixup_aper_base(struct scif_dev *dev, struct scif_window *window) -{ - int j; - struct scif_hw_dev *sdev = dev->sdev; - phys_addr_t apt_base = 0; - - /* - * Add the aperture base if the DMA address is not card relative - * since the DMA addresses need to be an offset into the bar - */ - if (!scifdev_self(dev) && window->type == SCIF_WINDOW_PEER && - sdev->aper && !sdev->card_rel_da) - apt_base = sdev->aper->pa; - else - return; - - for (j = 0; j < window->nr_contig_chunks; j++) { - if (window->num_pages[j]) - window->dma_addr[j] += apt_base; - else - break; - } -} - -/** - * scif_recv_reg: Respond to SCIF_REGISTER interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * Update remote window list with a new registered window. - */ -void scif_recv_reg(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; - struct scif_window *window = - (struct scif_window *)msg->payload[1]; - - mutex_lock(&ep->rma_info.rma_lock); - spin_lock(&ep->lock); - if (ep->state == SCIFEP_CONNECTED) { - msg->uop = SCIF_REGISTER_ACK; - scif_nodeqp_send(ep->remote_dev, msg); - scif_fixup_aper_base(ep->remote_dev, window); - /* No further failures expected. Insert new window */ - scif_insert_window(window, &ep->rma_info.remote_reg_list); - } else { - msg->uop = SCIF_REGISTER_NACK; - scif_nodeqp_send(ep->remote_dev, msg); - } - spin_unlock(&ep->lock); - mutex_unlock(&ep->rma_info.rma_lock); - /* free up any lookup resources now that page lists are transferred */ - scif_destroy_remote_lookup(ep->remote_dev, window); - /* - * We could not insert the window but we need to - * destroy the window. - */ - if (msg->uop == SCIF_REGISTER_NACK) - scif_destroy_remote_window(window); -} - -/** - * scif_recv_unreg: Respond to SCIF_UNREGISTER interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * Remove window from remote registration list; - */ -void scif_recv_unreg(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_rma_req req; - struct scif_window *window = NULL; - struct scif_window *recv_window = - (struct scif_window *)msg->payload[0]; - struct scif_endpt *ep; - int del_window = 0; - - ep = (struct scif_endpt *)recv_window->ep; - req.out_window = &window; - req.offset = recv_window->offset; - req.prot = 0; - req.nr_bytes = recv_window->nr_pages << PAGE_SHIFT; - req.type = SCIF_WINDOW_FULL; - req.head = &ep->rma_info.remote_reg_list; - msg->payload[0] = ep->remote_ep; - - mutex_lock(&ep->rma_info.rma_lock); - /* Does a valid window exist? */ - if (scif_query_window(&req)) { - dev_err(&scifdev->sdev->dev, - "%s %d -ENXIO\n", __func__, __LINE__); - msg->uop = SCIF_UNREGISTER_ACK; - goto error; - } - if (window) { - if (window->ref_count) - scif_put_window(window, window->nr_pages); - else - dev_err(&scifdev->sdev->dev, - "%s %d ref count should be +ve\n", - __func__, __LINE__); - window->unreg_state = OP_COMPLETED; - if (!window->ref_count) { - msg->uop = SCIF_UNREGISTER_ACK; - atomic_inc(&ep->rma_info.tw_refcount); - ep->rma_info.async_list_del = 1; - list_del_init(&window->list); - del_window = 1; - } else { - /* NACK! There are valid references to this window */ - msg->uop = SCIF_UNREGISTER_NACK; - } - } else { - /* The window did not make its way to the list at all. ACK */ - msg->uop = SCIF_UNREGISTER_ACK; - scif_destroy_remote_window(recv_window); - } -error: - mutex_unlock(&ep->rma_info.rma_lock); - if (del_window) - scif_drain_dma_intr(ep->remote_dev->sdev, - ep->rma_info.dma_chan); - scif_nodeqp_send(ep->remote_dev, msg); - if (del_window) - scif_queue_for_cleanup(window, &scif_info.rma); -} - -/** - * scif_recv_reg_ack: Respond to SCIF_REGISTER_ACK interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * Wake up the window waiting to complete registration. - */ -void scif_recv_reg_ack(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_window *window = - (struct scif_window *)msg->payload[2]; - struct scif_endpt *ep = (struct scif_endpt *)window->ep; - - mutex_lock(&ep->rma_info.rma_lock); - window->reg_state = OP_COMPLETED; - wake_up(&window->regwq); - mutex_unlock(&ep->rma_info.rma_lock); -} - -/** - * scif_recv_reg_nack: Respond to SCIF_REGISTER_NACK interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * Wake up the window waiting to inform it that registration - * cannot be completed. - */ -void scif_recv_reg_nack(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_window *window = - (struct scif_window *)msg->payload[2]; - struct scif_endpt *ep = (struct scif_endpt *)window->ep; - - mutex_lock(&ep->rma_info.rma_lock); - window->reg_state = OP_FAILED; - wake_up(&window->regwq); - mutex_unlock(&ep->rma_info.rma_lock); -} - -/** - * scif_recv_unreg_ack: Respond to SCIF_UNREGISTER_ACK interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * Wake up the window waiting to complete unregistration. - */ -void scif_recv_unreg_ack(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_window *window = - (struct scif_window *)msg->payload[1]; - struct scif_endpt *ep = (struct scif_endpt *)window->ep; - - mutex_lock(&ep->rma_info.rma_lock); - window->unreg_state = OP_COMPLETED; - wake_up(&window->unregwq); - mutex_unlock(&ep->rma_info.rma_lock); -} - -/** - * scif_recv_unreg_nack: Respond to SCIF_UNREGISTER_NACK interrupt message - * @scifdev: SCIF device - * @msg: Interrupt message - * - * Wake up the window waiting to inform it that unregistration - * cannot be completed immediately. - */ -void scif_recv_unreg_nack(struct scif_dev *scifdev, struct scifmsg *msg) -{ - struct scif_window *window = - (struct scif_window *)msg->payload[1]; - struct scif_endpt *ep = (struct scif_endpt *)window->ep; - - mutex_lock(&ep->rma_info.rma_lock); - window->unreg_state = OP_FAILED; - wake_up(&window->unregwq); - mutex_unlock(&ep->rma_info.rma_lock); -} - -int __scif_pin_pages(void *addr, size_t len, int *out_prot, - int map_flags, scif_pinned_pages_t *pages) -{ - struct scif_pinned_pages *pinned_pages; - int nr_pages, err = 0, i; - bool vmalloc_addr = false; - bool try_upgrade = false; - int prot = *out_prot; - int ulimit = 0; - struct mm_struct *mm = NULL; - - /* Unsupported flags */ - if (map_flags & ~(SCIF_MAP_KERNEL | SCIF_MAP_ULIMIT)) - return -EINVAL; - ulimit = !!(map_flags & SCIF_MAP_ULIMIT); - - /* Unsupported protection requested */ - if (prot & ~(SCIF_PROT_READ | SCIF_PROT_WRITE)) - return -EINVAL; - - /* addr/len must be page aligned. len should be non zero */ - if (!len || - (ALIGN((u64)addr, PAGE_SIZE) != (u64)addr) || - (ALIGN((u64)len, PAGE_SIZE) != (u64)len)) - return -EINVAL; - - might_sleep(); - - nr_pages = len >> PAGE_SHIFT; - - /* Allocate a set of pinned pages */ - pinned_pages = scif_create_pinned_pages(nr_pages, prot); - if (!pinned_pages) - return -ENOMEM; - - if (map_flags & SCIF_MAP_KERNEL) { - if (is_vmalloc_addr(addr)) - vmalloc_addr = true; - - for (i = 0; i < nr_pages; i++) { - if (vmalloc_addr) - pinned_pages->pages[i] = - vmalloc_to_page(addr + (i * PAGE_SIZE)); - else - pinned_pages->pages[i] = - virt_to_page(addr + (i * PAGE_SIZE)); - } - pinned_pages->nr_pages = nr_pages; - pinned_pages->map_flags = SCIF_MAP_KERNEL; - } else { - /* - * SCIF supports registration caching. If a registration has - * been requested with read only permissions, then we try - * to pin the pages with RW permissions so that a subsequent - * transfer with RW permission can hit the cache instead of - * invalidating it. If the upgrade fails with RW then we - * revert back to R permission and retry - */ - if (prot == SCIF_PROT_READ) - try_upgrade = true; - prot |= SCIF_PROT_WRITE; -retry: - mm = current->mm; - if (ulimit) { - err = __scif_check_inc_pinned_vm(mm, nr_pages); - if (err) { - pinned_pages->nr_pages = 0; - goto error_unmap; - } - } - - pinned_pages->nr_pages = pin_user_pages_fast( - (u64)addr, - nr_pages, - (prot & SCIF_PROT_WRITE) ? FOLL_WRITE : 0, - pinned_pages->pages); - if (nr_pages != pinned_pages->nr_pages) { - if (pinned_pages->nr_pages < 0) - pinned_pages->nr_pages = 0; - if (try_upgrade) { - if (ulimit) - __scif_dec_pinned_vm_lock(mm, nr_pages); - /* Roll back any pinned pages */ - unpin_user_pages(pinned_pages->pages, - pinned_pages->nr_pages); - prot &= ~SCIF_PROT_WRITE; - try_upgrade = false; - goto retry; - } - } - pinned_pages->map_flags = 0; - } - - if (pinned_pages->nr_pages < nr_pages) { - err = -EFAULT; - goto dec_pinned; - } - - *out_prot = prot; - atomic_set(&pinned_pages->ref_count, 1); - *pages = pinned_pages; - return err; -dec_pinned: - if (ulimit) - __scif_dec_pinned_vm_lock(mm, nr_pages); - /* Something went wrong! Rollback */ -error_unmap: - scif_destroy_pinned_pages(pinned_pages); - *pages = NULL; - dev_dbg(scif_info.mdev.this_device, - "%s %d err %d len 0x%lx\n", __func__, __LINE__, err, len); - return err; -} - -int scif_pin_pages(void *addr, size_t len, int prot, - int map_flags, scif_pinned_pages_t *pages) -{ - return __scif_pin_pages(addr, len, &prot, map_flags, pages); -} -EXPORT_SYMBOL_GPL(scif_pin_pages); - -int scif_unpin_pages(scif_pinned_pages_t pinned_pages) -{ - int err = 0, ret; - - if (!pinned_pages || SCIFEP_MAGIC != pinned_pages->magic) - return -EINVAL; - - ret = atomic_sub_return(1, &pinned_pages->ref_count); - if (ret < 0) { - dev_err(scif_info.mdev.this_device, - "%s %d scif_unpin_pages called without pinning? rc %d\n", - __func__, __LINE__, ret); - return -EINVAL; - } - /* - * Destroy the window if the ref count for this set of pinned - * pages has dropped to zero. If it is positive then there is - * a valid registered window which is backed by these pages and - * it will be destroyed once all such windows are unregistered. - */ - if (!ret) - err = scif_destroy_pinned_pages(pinned_pages); - - return err; -} -EXPORT_SYMBOL_GPL(scif_unpin_pages); - -static inline void -scif_insert_local_window(struct scif_window *window, struct scif_endpt *ep) -{ - mutex_lock(&ep->rma_info.rma_lock); - scif_insert_window(window, &ep->rma_info.reg_list); - mutex_unlock(&ep->rma_info.rma_lock); -} - -off_t scif_register_pinned_pages(scif_epd_t epd, - scif_pinned_pages_t pinned_pages, - off_t offset, int map_flags) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - s64 computed_offset; - struct scif_window *window; - int err; - size_t len; - struct device *spdev; - - /* Unsupported flags */ - if (map_flags & ~SCIF_MAP_FIXED) - return -EINVAL; - - len = pinned_pages->nr_pages << PAGE_SHIFT; - - /* - * Offset is not page aligned/negative or offset+len - * wraps around with SCIF_MAP_FIXED. - */ - if ((map_flags & SCIF_MAP_FIXED) && - ((ALIGN(offset, PAGE_SIZE) != offset) || - (offset < 0) || - (len > LONG_MAX - offset))) - return -EINVAL; - - might_sleep(); - - err = scif_verify_epd(ep); - if (err) - return err; - /* - * It is an error to pass pinned_pages to scif_register_pinned_pages() - * after calling scif_unpin_pages(). - */ - if (!atomic_add_unless(&pinned_pages->ref_count, 1, 0)) - return -EINVAL; - - /* Compute the offset for this registration */ - err = scif_get_window_offset(ep, map_flags, offset, - len, &computed_offset); - if (err) { - atomic_sub(1, &pinned_pages->ref_count); - return err; - } - - /* Allocate and prepare self registration window */ - window = scif_create_window(ep, pinned_pages->nr_pages, - computed_offset, false); - if (!window) { - atomic_sub(1, &pinned_pages->ref_count); - scif_free_window_offset(ep, NULL, computed_offset); - return -ENOMEM; - } - - window->pinned_pages = pinned_pages; - window->nr_pages = pinned_pages->nr_pages; - window->prot = pinned_pages->prot; - - spdev = scif_get_peer_dev(ep->remote_dev); - if (IS_ERR(spdev)) { - err = PTR_ERR(spdev); - scif_destroy_window(ep, window); - return err; - } - err = scif_send_alloc_request(ep, window); - if (err) { - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, err); - goto error_unmap; - } - - /* Prepare the remote registration window */ - err = scif_prep_remote_window(ep, window); - if (err) { - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, err); - goto error_unmap; - } - - /* Tell the peer about the new window */ - err = scif_send_scif_register(ep, window); - if (err) { - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, err); - goto error_unmap; - } - - scif_put_peer_dev(spdev); - /* No further failures expected. Insert new window */ - scif_insert_local_window(window, ep); - return computed_offset; -error_unmap: - scif_destroy_window(ep, window); - scif_put_peer_dev(spdev); - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, err); - return err; -} -EXPORT_SYMBOL_GPL(scif_register_pinned_pages); - -off_t scif_register(scif_epd_t epd, void *addr, size_t len, off_t offset, - int prot, int map_flags) -{ - scif_pinned_pages_t pinned_pages; - off_t err; - struct scif_endpt *ep = (struct scif_endpt *)epd; - s64 computed_offset; - struct scif_window *window; - struct mm_struct *mm = NULL; - struct device *spdev; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI register: ep %p addr %p len 0x%lx offset 0x%lx prot 0x%x map_flags 0x%x\n", - epd, addr, len, offset, prot, map_flags); - /* Unsupported flags */ - if (map_flags & ~(SCIF_MAP_FIXED | SCIF_MAP_KERNEL)) - return -EINVAL; - - /* - * Offset is not page aligned/negative or offset+len - * wraps around with SCIF_MAP_FIXED. - */ - if ((map_flags & SCIF_MAP_FIXED) && - ((ALIGN(offset, PAGE_SIZE) != offset) || - (offset < 0) || - (len > LONG_MAX - offset))) - return -EINVAL; - - /* Unsupported protection requested */ - if (prot & ~(SCIF_PROT_READ | SCIF_PROT_WRITE)) - return -EINVAL; - - /* addr/len must be page aligned. len should be non zero */ - if (!len || (ALIGN((u64)addr, PAGE_SIZE) != (u64)addr) || - (ALIGN(len, PAGE_SIZE) != len)) - return -EINVAL; - - might_sleep(); - - err = scif_verify_epd(ep); - if (err) - return err; - - /* Compute the offset for this registration */ - err = scif_get_window_offset(ep, map_flags, offset, - len >> PAGE_SHIFT, &computed_offset); - if (err) - return err; - - spdev = scif_get_peer_dev(ep->remote_dev); - if (IS_ERR(spdev)) { - err = PTR_ERR(spdev); - scif_free_window_offset(ep, NULL, computed_offset); - return err; - } - /* Allocate and prepare self registration window */ - window = scif_create_window(ep, len >> PAGE_SHIFT, - computed_offset, false); - if (!window) { - scif_free_window_offset(ep, NULL, computed_offset); - scif_put_peer_dev(spdev); - return -ENOMEM; - } - - window->nr_pages = len >> PAGE_SHIFT; - - err = scif_send_alloc_request(ep, window); - if (err) { - scif_destroy_incomplete_window(ep, window); - scif_put_peer_dev(spdev); - return err; - } - - if (!(map_flags & SCIF_MAP_KERNEL)) { - mm = __scif_acquire_mm(); - map_flags |= SCIF_MAP_ULIMIT; - } - /* Pin down the pages */ - err = __scif_pin_pages(addr, len, &prot, - map_flags & (SCIF_MAP_KERNEL | SCIF_MAP_ULIMIT), - &pinned_pages); - if (err) { - scif_destroy_incomplete_window(ep, window); - __scif_release_mm(mm); - goto error; - } - - window->pinned_pages = pinned_pages; - window->prot = pinned_pages->prot; - window->mm = mm; - - /* Prepare the remote registration window */ - err = scif_prep_remote_window(ep, window); - if (err) { - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %ld\n", __func__, __LINE__, err); - goto error_unmap; - } - - /* Tell the peer about the new window */ - err = scif_send_scif_register(ep, window); - if (err) { - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %ld\n", __func__, __LINE__, err); - goto error_unmap; - } - - scif_put_peer_dev(spdev); - /* No further failures expected. Insert new window */ - scif_insert_local_window(window, ep); - dev_dbg(&ep->remote_dev->sdev->dev, - "SCIFAPI register: ep %p addr %p len 0x%lx computed_offset 0x%llx\n", - epd, addr, len, computed_offset); - return computed_offset; -error_unmap: - scif_destroy_window(ep, window); -error: - scif_put_peer_dev(spdev); - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %ld\n", __func__, __LINE__, err); - return err; -} -EXPORT_SYMBOL_GPL(scif_register); - -int -scif_unregister(scif_epd_t epd, off_t offset, size_t len) -{ - struct scif_endpt *ep = (struct scif_endpt *)epd; - struct scif_window *window = NULL; - struct scif_rma_req req; - int nr_pages, err; - struct device *spdev; - - dev_dbg(scif_info.mdev.this_device, - "SCIFAPI unregister: ep %p offset 0x%lx len 0x%lx\n", - ep, offset, len); - /* len must be page aligned. len should be non zero */ - if (!len || - (ALIGN((u64)len, PAGE_SIZE) != (u64)len)) - return -EINVAL; - - /* Offset is not page aligned or offset+len wraps around */ - if ((ALIGN(offset, PAGE_SIZE) != offset) || - (offset < 0) || - (len > LONG_MAX - offset)) - return -EINVAL; - - err = scif_verify_epd(ep); - if (err) - return err; - - might_sleep(); - nr_pages = len >> PAGE_SHIFT; - - req.out_window = &window; - req.offset = offset; - req.prot = 0; - req.nr_bytes = len; - req.type = SCIF_WINDOW_FULL; - req.head = &ep->rma_info.reg_list; - - spdev = scif_get_peer_dev(ep->remote_dev); - if (IS_ERR(spdev)) { - err = PTR_ERR(spdev); - return err; - } - mutex_lock(&ep->rma_info.rma_lock); - /* Does a valid window exist? */ - err = scif_query_window(&req); - if (err) { - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, err); - goto error; - } - /* Unregister all the windows in this range */ - err = scif_rma_list_unregister(window, offset, nr_pages); - if (err) - dev_err(&ep->remote_dev->sdev->dev, - "%s %d err %d\n", __func__, __LINE__, err); -error: - mutex_unlock(&ep->rma_info.rma_lock); - scif_put_peer_dev(spdev); - return err; -} -EXPORT_SYMBOL_GPL(scif_unregister); diff --git a/drivers/misc/mic/scif/scif_rma.h b/drivers/misc/mic/scif/scif_rma.h deleted file mode 100644 index 964dd0fc3657..000000000000 --- a/drivers/misc/mic/scif/scif_rma.h +++ /dev/null @@ -1,477 +0,0 @@ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * This file is provided under a dual BSD/GPLv2 license. When using or - * redistributing this file, you may do so under either license. - * - * GPL LICENSE SUMMARY - * - * Copyright(c) 2015 Intel Corporation. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * BSD LICENSE - * - * Copyright(c) 2015 Intel Corporation. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * - * * Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * * Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in - * the documentation and/or other materials provided with the - * distribution. - * * Neither the name of Intel Corporation nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR - * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT - * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, - * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT - * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - * - * Intel SCIF driver. - * - */ -#ifndef SCIF_RMA_H -#define SCIF_RMA_H - -#include -#include - -#include "../bus/scif_bus.h" - -/* If this bit is set then the mark is a remote fence mark */ -#define SCIF_REMOTE_FENCE_BIT 31 -/* Magic value used to indicate a remote fence request */ -#define SCIF_REMOTE_FENCE BIT_ULL(SCIF_REMOTE_FENCE_BIT) - -#define SCIF_MAX_UNALIGNED_BUF_SIZE (1024 * 1024ULL) -#define SCIF_KMEM_UNALIGNED_BUF_SIZE (SCIF_MAX_UNALIGNED_BUF_SIZE + \ - (L1_CACHE_BYTES << 1)) - -#define SCIF_IOVA_START_PFN (1) -#define SCIF_IOVA_PFN(addr) ((addr) >> PAGE_SHIFT) -#define SCIF_DMA_64BIT_PFN SCIF_IOVA_PFN(DMA_BIT_MASK(64)) -#define SCIF_DMA_63BIT_PFN SCIF_IOVA_PFN(DMA_BIT_MASK(63)) - -/* - * struct scif_endpt_rma_info - Per Endpoint Remote Memory Access Information - * - * @reg_list: List of registration windows for self - * @remote_reg_list: List of registration windows for peer - * @iovad: Offset generator - * @rma_lock: Synchronizes access to self/remote list and also protects the - * window from being destroyed while RMAs are in progress. - * @tc_lock: Synchronizes access to temporary cached windows list - * for SCIF Registration Caching. - * @mmn_lock: Synchronizes access to the list of MMU notifiers registered - * @tw_refcount: Keeps track of number of outstanding temporary registered - * windows created by scif_vreadfrom/scif_vwriteto which have - * not been destroyed. - * @tcw_refcount: Same as tw_refcount but for temporary cached windows - * @tcw_total_pages: Same as tcw_refcount but in terms of pages pinned - * @mmn_list: MMU notifier so that we can destroy the windows when required - * @fence_refcount: Keeps track of number of outstanding remote fence - * requests which have been received by the peer. - * @dma_chan: DMA channel used for all DMA transfers for this endpoint. - * @async_list_del: Detect asynchronous list entry deletion - * @vma_list: List of vmas with remote memory mappings - * @markwq: Wait queue used for scif_fence_mark/scif_fence_wait -*/ -struct scif_endpt_rma_info { - struct list_head reg_list; - struct list_head remote_reg_list; - struct iova_domain iovad; - struct mutex rma_lock; - spinlock_t tc_lock; - struct mutex mmn_lock; - atomic_t tw_refcount; - atomic_t tcw_refcount; - atomic_t tcw_total_pages; - struct list_head mmn_list; - atomic_t fence_refcount; - struct dma_chan *dma_chan; - int async_list_del; - struct list_head vma_list; - wait_queue_head_t markwq; -}; - -/* - * struct scif_fence_info - used for tracking fence requests - * - * @state: State of this transfer - * @wq: Fences wait on this queue - * @dma_mark: Used for storing the DMA mark - */ -struct scif_fence_info { - enum scif_msg_state state; - struct completion comp; - int dma_mark; -}; - -/* - * struct scif_remote_fence_info - used for tracking remote fence requests - * - * @msg: List of SCIF node QP fence messages - * @list: Link to list of remote fence requests - */ -struct scif_remote_fence_info { - struct scifmsg msg; - struct list_head list; -}; - -/* - * Specifies whether an RMA operation can span across partial windows, a single - * window or multiple contiguous windows. Mmaps can span across partial windows. - * Unregistration can span across complete windows. scif_get_pages() can span a - * single window. A window can also be of type self or peer. - */ -enum scif_window_type { - SCIF_WINDOW_PARTIAL, - SCIF_WINDOW_SINGLE, - SCIF_WINDOW_FULL, - SCIF_WINDOW_SELF, - SCIF_WINDOW_PEER -}; - -/* The number of physical addresses that can be stored in a PAGE. */ -#define SCIF_NR_ADDR_IN_PAGE (0x1000 >> 3) - -/* - * struct scif_rma_lookup - RMA lookup data structure for page list transfers - * - * Store an array of lookup offsets. Each offset in this array maps - * one 4K page containing 512 physical addresses i.e. 2MB. 512 such - * offsets in a 4K page will correspond to 1GB of registered address space. - - * @lookup: Array of offsets - * @offset: DMA offset of lookup array - */ -struct scif_rma_lookup { - dma_addr_t *lookup; - dma_addr_t offset; -}; - -/* - * struct scif_pinned_pages - A set of pinned pages obtained with - * scif_pin_pages() which could be part of multiple registered - * windows across different end points. - * - * @nr_pages: Number of pages which is defined as a s64 instead of an int - * to avoid sign extension with buffers >= 2GB - * @prot: read/write protections - * @map_flags: Flags specified during the pin operation - * @ref_count: Reference count bumped in terms of number of pages - * @magic: A magic value - * @pages: Array of pointers to struct pages populated with get_user_pages(..) - */ -struct scif_pinned_pages { - s64 nr_pages; - int prot; - int map_flags; - atomic_t ref_count; - u64 magic; - struct page **pages; -}; - -/* - * struct scif_status - Stores DMA status update information - * - * @src_dma_addr: Source buffer DMA address - * @val: src location for value to be written to the destination - * @ep: SCIF endpoint - */ -struct scif_status { - dma_addr_t src_dma_addr; - u64 val; - struct scif_endpt *ep; -}; - -/* - * struct scif_cb_arg - Stores the argument of the callback func - * - * @src_dma_addr: Source buffer DMA address - * @status: DMA status - * @ep: SCIF endpoint - */ -struct scif_cb_arg { - dma_addr_t src_dma_addr; - struct scif_status *status; - struct scif_endpt *ep; -}; - -/* - * struct scif_window - Registration Window for Self and Remote - * - * @nr_pages: Number of pages which is defined as a s64 instead of an int - * to avoid sign extension with buffers >= 2GB - * @nr_contig_chunks: Number of contiguous physical chunks - * @prot: read/write protections - * @ref_count: reference count in terms of number of pages - * @magic: Cookie to detect corruption - * @offset: registered offset - * @va_for_temp: va address that this window represents - * @dma_mark: Used to determine if all DMAs against the window are done - * @ep: Pointer to EP. Useful for passing EP around with messages to - avoid expensive list traversals. - * @list: link to list of windows for the endpoint - * @type: self or peer window - * @peer_window: Pointer to peer window. Useful for sending messages to peer - * without requiring an extra list traversal - * @unreg_state: unregistration state - * @offset_freed: True if the offset has been freed - * @temp: True for temporary windows created via scif_vreadfrom/scif_vwriteto - * @mm: memory descriptor for the task_struct which initiated the RMA - * @st: scatter gather table for DMA mappings with IOMMU enabled - * @pinned_pages: The set of pinned_pages backing this window - * @alloc_handle: Handle for sending ALLOC_REQ - * @regwq: Wait Queue for an registration (N)ACK - * @reg_state: Registration state - * @unregwq: Wait Queue for an unregistration (N)ACK - * @dma_addr_lookup: Lookup for physical addresses used for DMA - * @nr_lookup: Number of entries in lookup - * @mapped_offset: Offset used to map the window by the peer - * @dma_addr: Array of physical addresses used for Mgmt node & MIC initiated DMA - * @num_pages: Array specifying number of pages for each physical address - */ -struct scif_window { - s64 nr_pages; - int nr_contig_chunks; - int prot; - int ref_count; - u64 magic; - s64 offset; - unsigned long va_for_temp; - int dma_mark; - u64 ep; - struct list_head list; - enum scif_window_type type; - u64 peer_window; - enum scif_msg_state unreg_state; - bool offset_freed; - bool temp; - struct mm_struct *mm; - struct sg_table *st; - union { - struct { - struct scif_pinned_pages *pinned_pages; - struct scif_allocmsg alloc_handle; - wait_queue_head_t regwq; - enum scif_msg_state reg_state; - wait_queue_head_t unregwq; - }; - struct { - struct scif_rma_lookup dma_addr_lookup; - struct scif_rma_lookup num_pages_lookup; - int nr_lookup; - dma_addr_t mapped_offset; - }; - }; - dma_addr_t *dma_addr; - u64 *num_pages; -} __packed; - -/* - * scif_mmu_notif - SCIF mmu notifier information - * - * @mmu_notifier ep_mmu_notifier: MMU notifier operations - * @tc_reg_list: List of temp registration windows for self - * @mm: memory descriptor for the task_struct which initiated the RMA - * @ep: SCIF endpoint - * @list: link to list of MMU notifier information - */ -struct scif_mmu_notif { -#ifdef CONFIG_MMU_NOTIFIER - struct mmu_notifier ep_mmu_notifier; -#endif - struct list_head tc_reg_list; - struct mm_struct *mm; - struct scif_endpt *ep; - struct list_head list; -}; - -enum scif_rma_dir { - SCIF_LOCAL_TO_REMOTE, - SCIF_REMOTE_TO_LOCAL -}; - -extern struct kmem_cache *unaligned_cache; -/* Initialize RMA for this EP */ -void scif_rma_ep_init(struct scif_endpt *ep); -/* Check if epd can be uninitialized */ -int scif_rma_ep_can_uninit(struct scif_endpt *ep); -/* Obtain a new offset. Callee must grab RMA lock */ -int scif_get_window_offset(struct scif_endpt *ep, int flags, - s64 offset, int nr_pages, s64 *out_offset); -/* Free offset. Callee must grab RMA lock */ -void scif_free_window_offset(struct scif_endpt *ep, - struct scif_window *window, s64 offset); -/* Create self registration window */ -struct scif_window *scif_create_window(struct scif_endpt *ep, int nr_pages, - s64 offset, bool temp); -/* Destroy self registration window.*/ -int scif_destroy_window(struct scif_endpt *ep, struct scif_window *window); -void scif_unmap_window(struct scif_dev *remote_dev, struct scif_window *window); -/* Map pages of self window to Aperture/PCI */ -int scif_map_window(struct scif_dev *remote_dev, - struct scif_window *window); -/* Unregister a self window */ -int scif_unregister_window(struct scif_window *window); -/* Destroy remote registration window */ -void -scif_destroy_remote_window(struct scif_window *window); -/* remove valid remote memory mappings from process address space */ -void scif_zap_mmaps(int node); -/* Query if any applications have remote memory mappings */ -bool scif_rma_do_apps_have_mmaps(int node); -/* Cleanup remote registration lists for zombie endpoints */ -void scif_cleanup_rma_for_zombies(int node); -/* Reserve a DMA channel for a particular endpoint */ -int scif_reserve_dma_chan(struct scif_endpt *ep); -/* Setup a DMA mark for an endpoint */ -int _scif_fence_mark(scif_epd_t epd, int *mark); -int scif_prog_signal(scif_epd_t epd, off_t offset, u64 val, - enum scif_window_type type); -void scif_alloc_req(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_alloc_gnt_rej(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_free_virt(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_recv_reg(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_recv_unreg(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_recv_reg_ack(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_recv_reg_nack(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_recv_unreg_ack(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_recv_unreg_nack(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_recv_munmap(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_recv_mark(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_recv_mark_resp(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_recv_wait(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_recv_wait_resp(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_recv_sig_local(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_recv_sig_remote(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_recv_sig_resp(struct scif_dev *scifdev, struct scifmsg *msg); -void scif_mmu_notif_handler(struct work_struct *work); -void scif_rma_handle_remote_fences(void); -void scif_rma_destroy_windows(void); -void scif_rma_destroy_tcw_invalid(void); -int scif_drain_dma_intr(struct scif_hw_dev *sdev, struct dma_chan *chan); - -struct scif_window_iter { - s64 offset; - int index; -}; - -static inline void -scif_init_window_iter(struct scif_window *window, struct scif_window_iter *iter) -{ - iter->offset = window->offset; - iter->index = 0; -} - -dma_addr_t scif_off_to_dma_addr(struct scif_window *window, s64 off, - size_t *nr_bytes, - struct scif_window_iter *iter); -static inline -dma_addr_t __scif_off_to_dma_addr(struct scif_window *window, s64 off) -{ - return scif_off_to_dma_addr(window, off, NULL, NULL); -} - -static inline bool scif_unaligned(off_t src_offset, off_t dst_offset) -{ - src_offset = src_offset & (L1_CACHE_BYTES - 1); - dst_offset = dst_offset & (L1_CACHE_BYTES - 1); - return !(src_offset == dst_offset); -} - -/* - * scif_zalloc: - * @size: Size of the allocation request. - * - * Helper API which attempts to allocate zeroed pages via - * __get_free_pages(..) first and then falls back on - * vzalloc(..) if that fails. - */ -static inline void *scif_zalloc(size_t size) -{ - void *ret = NULL; - size_t align = ALIGN(size, PAGE_SIZE); - - if (align && get_order(align) < MAX_ORDER) - ret = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, - get_order(align)); - return ret ? ret : vzalloc(align); -} - -/* - * scif_free: - * @addr: Address to be freed. - * @size: Size of the allocation. - * Helper API which frees memory allocated via scif_zalloc(). - */ -static inline void scif_free(void *addr, size_t size) -{ - size_t align = ALIGN(size, PAGE_SIZE); - - if (is_vmalloc_addr(addr)) - vfree(addr); - else - free_pages((unsigned long)addr, get_order(align)); -} - -static inline void scif_get_window(struct scif_window *window, int nr_pages) -{ - window->ref_count += nr_pages; -} - -static inline void scif_put_window(struct scif_window *window, int nr_pages) -{ - window->ref_count -= nr_pages; -} - -static inline void scif_set_window_ref(struct scif_window *window, int nr_pages) -{ - window->ref_count = nr_pages; -} - -static inline void -scif_queue_for_cleanup(struct scif_window *window, struct list_head *list) -{ - spin_lock(&scif_info.rmalock); - list_add_tail(&window->list, list); - spin_unlock(&scif_info.rmalock); - schedule_work(&scif_info.misc_work); -} - -static inline void __scif_rma_destroy_tcw_helper(struct scif_window *window) -{ - list_del_init(&window->list); - scif_queue_for_cleanup(window, &scif_info.rma_tc); -} - -static inline bool scif_is_iommu_enabled(void) -{ -#ifdef CONFIG_INTEL_IOMMU - return intel_iommu_enabled; -#else - return false; -#endif -} -#endif /* SCIF_RMA_H */ diff --git a/drivers/misc/mic/scif/scif_rma_list.c b/drivers/misc/mic/scif/scif_rma_list.c deleted file mode 100644 index ef923ba134c8..000000000000 --- a/drivers/misc/mic/scif/scif_rma_list.c +++ /dev/null @@ -1,282 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2015 Intel Corporation. - * - * Intel SCIF driver. - */ -#include "scif_main.h" -#include -#include - -/* - * scif_insert_tcw: - * - * Insert a temp window to the temp registration list sorted by va_for_temp. - * RMA lock must be held. - */ -void scif_insert_tcw(struct scif_window *window, struct list_head *head) -{ - struct scif_window *curr = NULL; - struct scif_window *prev = list_entry(head, struct scif_window, list); - struct list_head *item; - - INIT_LIST_HEAD(&window->list); - /* Compare with tail and if the entry is new tail add it to the end */ - if (!list_empty(head)) { - curr = list_entry(head->prev, struct scif_window, list); - if (curr->va_for_temp < window->va_for_temp) { - list_add_tail(&window->list, head); - return; - } - } - list_for_each(item, head) { - curr = list_entry(item, struct scif_window, list); - if (curr->va_for_temp > window->va_for_temp) - break; - prev = curr; - } - list_add(&window->list, &prev->list); -} - -/* - * scif_insert_window: - * - * Insert a window to the self registration list sorted by offset. - * RMA lock must be held. - */ -void scif_insert_window(struct scif_window *window, struct list_head *head) -{ - struct scif_window *curr = NULL, *prev = NULL; - struct list_head *item; - - INIT_LIST_HEAD(&window->list); - list_for_each(item, head) { - curr = list_entry(item, struct scif_window, list); - if (curr->offset > window->offset) - break; - prev = curr; - } - if (!prev) - list_add(&window->list, head); - else - list_add(&window->list, &prev->list); - scif_set_window_ref(window, window->nr_pages); -} - -/* - * scif_query_tcw: - * - * Query the temp cached registration list of ep for an overlapping window - * in case of permission mismatch, destroy the previous window. if permissions - * match and overlap is partial, destroy the window but return the new range - * RMA lock must be held. - */ -int scif_query_tcw(struct scif_endpt *ep, struct scif_rma_req *req) -{ - struct list_head *item, *temp, *head = req->head; - struct scif_window *window; - u64 start_va_window, start_va_req = req->va_for_temp; - u64 end_va_window, end_va_req = start_va_req + req->nr_bytes; - - if (!req->nr_bytes) - return -EINVAL; - /* - * Avoid traversing the entire list to find out that there - * is no entry that matches - */ - if (!list_empty(head)) { - window = list_last_entry(head, struct scif_window, list); - end_va_window = window->va_for_temp + - (window->nr_pages << PAGE_SHIFT); - if (start_va_req > end_va_window) - return -ENXIO; - } - list_for_each_safe(item, temp, head) { - window = list_entry(item, struct scif_window, list); - start_va_window = window->va_for_temp; - end_va_window = window->va_for_temp + - (window->nr_pages << PAGE_SHIFT); - if (start_va_req < start_va_window && - end_va_req < start_va_window) - break; - if (start_va_req >= end_va_window) - continue; - if ((window->prot & req->prot) == req->prot) { - if (start_va_req >= start_va_window && - end_va_req <= end_va_window) { - *req->out_window = window; - return 0; - } - /* expand window */ - if (start_va_req < start_va_window) { - req->nr_bytes += - start_va_window - start_va_req; - req->va_for_temp = start_va_window; - } - if (end_va_req >= end_va_window) - req->nr_bytes += end_va_window - end_va_req; - } - /* Destroy the old window to create a new one */ - __scif_rma_destroy_tcw_helper(window); - break; - } - return -ENXIO; -} - -/* - * scif_query_window: - * - * Query the registration list and check if a valid contiguous - * range of windows exist. - * RMA lock must be held. - */ -int scif_query_window(struct scif_rma_req *req) -{ - struct list_head *item; - struct scif_window *window; - s64 end_offset, offset = req->offset; - u64 tmp_min, nr_bytes_left = req->nr_bytes; - - if (!req->nr_bytes) - return -EINVAL; - - list_for_each(item, req->head) { - window = list_entry(item, struct scif_window, list); - end_offset = window->offset + - (window->nr_pages << PAGE_SHIFT); - if (offset < window->offset) - /* Offset not found! */ - return -ENXIO; - if (offset >= end_offset) - continue; - /* Check read/write protections. */ - if ((window->prot & req->prot) != req->prot) - return -EPERM; - if (nr_bytes_left == req->nr_bytes) - /* Store the first window */ - *req->out_window = window; - tmp_min = min((u64)end_offset - offset, nr_bytes_left); - nr_bytes_left -= tmp_min; - offset += tmp_min; - /* - * Range requested encompasses - * multiple windows contiguously. - */ - if (!nr_bytes_left) { - /* Done for partial window */ - if (req->type == SCIF_WINDOW_PARTIAL || - req->type == SCIF_WINDOW_SINGLE) - return 0; - /* Extra logic for full windows */ - if (offset == end_offset) - /* Spanning multiple whole windows */ - return 0; - /* Not spanning multiple whole windows */ - return -ENXIO; - } - if (req->type == SCIF_WINDOW_SINGLE) - break; - } - dev_err(scif_info.mdev.this_device, - "%s %d ENXIO\n", __func__, __LINE__); - return -ENXIO; -} - -/* - * scif_rma_list_unregister: - * - * Traverse the self registration list starting from window: - * 1) Call scif_unregister_window(..) - * RMA lock must be held. - */ -int scif_rma_list_unregister(struct scif_window *window, - s64 offset, int nr_pages) -{ - struct scif_endpt *ep = (struct scif_endpt *)window->ep; - struct list_head *head = &ep->rma_info.reg_list; - s64 end_offset; - int err = 0; - int loop_nr_pages; - struct scif_window *_window; - - list_for_each_entry_safe_from(window, _window, head, list) { - end_offset = window->offset + (window->nr_pages << PAGE_SHIFT); - loop_nr_pages = min((int)((end_offset - offset) >> PAGE_SHIFT), - nr_pages); - err = scif_unregister_window(window); - if (err) - return err; - nr_pages -= loop_nr_pages; - offset += (loop_nr_pages << PAGE_SHIFT); - if (!nr_pages) - break; - } - return 0; -} - -/* - * scif_unmap_all_window: - * - * Traverse all the windows in the self registration list and: - * 1) Delete any DMA mappings created - */ -void scif_unmap_all_windows(scif_epd_t epd) -{ - struct list_head *item, *tmp; - struct scif_window *window; - struct scif_endpt *ep = (struct scif_endpt *)epd; - struct list_head *head = &ep->rma_info.reg_list; - - mutex_lock(&ep->rma_info.rma_lock); - list_for_each_safe(item, tmp, head) { - window = list_entry(item, struct scif_window, list); - scif_unmap_window(ep->remote_dev, window); - } - mutex_unlock(&ep->rma_info.rma_lock); -} - -/* - * scif_unregister_all_window: - * - * Traverse all the windows in the self registration list and: - * 1) Call scif_unregister_window(..) - * RMA lock must be held. - */ -int scif_unregister_all_windows(scif_epd_t epd) -{ - struct list_head *item, *tmp; - struct scif_window *window; - struct scif_endpt *ep = (struct scif_endpt *)epd; - struct list_head *head = &ep->rma_info.reg_list; - int err = 0; - - mutex_lock(&ep->rma_info.rma_lock); -retry: - item = NULL; - tmp = NULL; - list_for_each_safe(item, tmp, head) { - window = list_entry(item, struct scif_window, list); - ep->rma_info.async_list_del = 0; - err = scif_unregister_window(window); - if (err) - dev_err(scif_info.mdev.this_device, - "%s %d err %d\n", - __func__, __LINE__, err); - /* - * Need to restart list traversal if there has been - * an asynchronous list entry deletion. - */ - if (READ_ONCE(ep->rma_info.async_list_del)) - goto retry; - } - mutex_unlock(&ep->rma_info.rma_lock); - if (!list_empty(&ep->rma_info.mmn_list)) { - spin_lock(&scif_info.rmalock); - list_add_tail(&ep->mmu_list, &scif_info.mmu_notif_cleanup); - spin_unlock(&scif_info.rmalock); - schedule_work(&scif_info.mmu_notif_work); - } - return err; -} diff --git a/drivers/misc/mic/scif/scif_rma_list.h b/drivers/misc/mic/scif/scif_rma_list.h deleted file mode 100644 index 0f8e0ed65614..000000000000 --- a/drivers/misc/mic/scif/scif_rma_list.h +++ /dev/null @@ -1,48 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2015 Intel Corporation. - * - * Intel SCIF driver. - */ -#ifndef SCIF_RMA_LIST_H -#define SCIF_RMA_LIST_H - -/* - * struct scif_rma_req - Self Registration list RMA Request query - * - * @out_window - Returns the window if found - * @offset: Starting offset - * @nr_bytes: number of bytes - * @prot: protection requested i.e. read or write or both - * @type: Specify single, partial or multiple windows - * @head: Head of list on which to search - * @va_for_temp: VA for searching temporary cached windows - */ -struct scif_rma_req { - struct scif_window **out_window; - union { - s64 offset; - unsigned long va_for_temp; - }; - size_t nr_bytes; - int prot; - enum scif_window_type type; - struct list_head *head; -}; - -/* Insert */ -void scif_insert_window(struct scif_window *window, struct list_head *head); -void scif_insert_tcw(struct scif_window *window, - struct list_head *head); -/* Query */ -int scif_query_window(struct scif_rma_req *request); -int scif_query_tcw(struct scif_endpt *ep, struct scif_rma_req *request); -/* Called from close to unregister all self windows */ -int scif_unregister_all_windows(scif_epd_t epd); -void scif_unmap_all_windows(scif_epd_t epd); -/* Traverse list and unregister */ -int scif_rma_list_unregister(struct scif_window *window, s64 offset, - int nr_pages); -#endif /* SCIF_RMA_LIST_H */ diff --git a/drivers/misc/mic/vop/Makefile b/drivers/misc/mic/vop/Makefile deleted file mode 100644 index 51b9b0022786..000000000000 --- a/drivers/misc/mic/vop/Makefile +++ /dev/null @@ -1,10 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0-only -# -# Makefile - Intel MIC Linux driver. -# Copyright(c) 2016, Intel Corporation. -# -obj-$(CONFIG_VOP) := vop.o - -vop-objs += vop_main.o -vop-objs += vop_debugfs.o -vop-objs += vop_vringh.o diff --git a/drivers/misc/mic/vop/vop_debugfs.c b/drivers/misc/mic/vop/vop_debugfs.c deleted file mode 100644 index 9d4f175f4dd1..000000000000 --- a/drivers/misc/mic/vop/vop_debugfs.c +++ /dev/null @@ -1,184 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2016 Intel Corporation. - * - * Intel Virtio Over PCIe (VOP) driver. - */ -#include -#include - -#include "vop_main.h" - -static int vop_dp_show(struct seq_file *s, void *pos) -{ - struct mic_device_desc *d; - struct mic_device_ctrl *dc; - struct mic_vqconfig *vqconfig; - __u32 *features; - __u8 *config; - struct vop_info *vi = s->private; - struct vop_device *vpdev = vi->vpdev; - struct mic_bootparam *bootparam = vpdev->hw_ops->get_dp(vpdev); - int j, k; - - seq_printf(s, "Bootparam: magic 0x%x\n", - bootparam->magic); - seq_printf(s, "Bootparam: h2c_config_db %d\n", - bootparam->h2c_config_db); - seq_printf(s, "Bootparam: node_id %d\n", - bootparam->node_id); - seq_printf(s, "Bootparam: c2h_scif_db %d\n", - bootparam->c2h_scif_db); - seq_printf(s, "Bootparam: h2c_scif_db %d\n", - bootparam->h2c_scif_db); - seq_printf(s, "Bootparam: scif_host_dma_addr 0x%llx\n", - bootparam->scif_host_dma_addr); - seq_printf(s, "Bootparam: scif_card_dma_addr 0x%llx\n", - bootparam->scif_card_dma_addr); - - for (j = sizeof(*bootparam); - j < MIC_DP_SIZE; j += mic_total_desc_size(d)) { - d = (void *)bootparam + j; - dc = (void *)d + mic_aligned_desc_size(d); - - /* end of list */ - if (d->type == 0) - break; - - if (d->type == -1) - continue; - - seq_printf(s, "Type %d ", d->type); - seq_printf(s, "Num VQ %d ", d->num_vq); - seq_printf(s, "Feature Len %d\n", d->feature_len); - seq_printf(s, "Config Len %d ", d->config_len); - seq_printf(s, "Shutdown Status %d\n", d->status); - - for (k = 0; k < d->num_vq; k++) { - vqconfig = mic_vq_config(d) + k; - seq_printf(s, "vqconfig[%d]: ", k); - seq_printf(s, "address 0x%llx ", - vqconfig->address); - seq_printf(s, "num %d ", vqconfig->num); - seq_printf(s, "used address 0x%llx\n", - vqconfig->used_address); - } - - features = (__u32 *)mic_vq_features(d); - seq_printf(s, "Features: Host 0x%x ", features[0]); - seq_printf(s, "Guest 0x%x\n", features[1]); - - config = mic_vq_configspace(d); - for (k = 0; k < d->config_len; k++) - seq_printf(s, "config[%d]=%d\n", k, config[k]); - - seq_puts(s, "Device control:\n"); - seq_printf(s, "Config Change %d ", dc->config_change); - seq_printf(s, "Vdev reset %d\n", dc->vdev_reset); - seq_printf(s, "Guest Ack %d ", dc->guest_ack); - seq_printf(s, "Host ack %d\n", dc->host_ack); - seq_printf(s, "Used address updated %d ", - dc->used_address_updated); - seq_printf(s, "Vdev 0x%llx\n", dc->vdev); - seq_printf(s, "c2h doorbell %d ", dc->c2h_vdev_db); - seq_printf(s, "h2c doorbell %d\n", dc->h2c_vdev_db); - } - schedule_work(&vi->hotplug_work); - return 0; -} - -DEFINE_SHOW_ATTRIBUTE(vop_dp); - -static int vop_vdev_info_show(struct seq_file *s, void *unused) -{ - struct vop_info *vi = s->private; - struct list_head *pos, *tmp; - struct vop_vdev *vdev; - int i, j; - - mutex_lock(&vi->vop_mutex); - list_for_each_safe(pos, tmp, &vi->vdev_list) { - vdev = list_entry(pos, struct vop_vdev, list); - seq_printf(s, "VDEV type %d state %s in %ld out %ld in_dma %ld out_dma %ld\n", - vdev->virtio_id, - vop_vdevup(vdev) ? "UP" : "DOWN", - vdev->in_bytes, - vdev->out_bytes, - vdev->in_bytes_dma, - vdev->out_bytes_dma); - for (i = 0; i < MIC_MAX_VRINGS; i++) { - struct vring_desc *desc; - struct vring_avail *avail; - struct vring_used *used; - struct vop_vringh *vvr = &vdev->vvr[i]; - struct vringh *vrh = &vvr->vrh; - int num = vrh->vring.num; - - if (!num) - continue; - desc = vrh->vring.desc; - seq_printf(s, "vring i %d avail_idx %d", - i, vvr->vring.info->avail_idx & (num - 1)); - seq_printf(s, " vring i %d avail_idx %d\n", - i, vvr->vring.info->avail_idx); - seq_printf(s, "vrh i %d weak_barriers %d", - i, vrh->weak_barriers); - seq_printf(s, " last_avail_idx %d last_used_idx %d", - vrh->last_avail_idx, vrh->last_used_idx); - seq_printf(s, " completed %d\n", vrh->completed); - for (j = 0; j < num; j++) { - seq_printf(s, "desc[%d] addr 0x%llx len %d", - j, desc->addr, desc->len); - seq_printf(s, " flags 0x%x next %d\n", - desc->flags, desc->next); - desc++; - } - avail = vrh->vring.avail; - seq_printf(s, "avail flags 0x%x idx %d\n", - vringh16_to_cpu(vrh, avail->flags), - vringh16_to_cpu(vrh, - avail->idx) & (num - 1)); - seq_printf(s, "avail flags 0x%x idx %d\n", - vringh16_to_cpu(vrh, avail->flags), - vringh16_to_cpu(vrh, avail->idx)); - for (j = 0; j < num; j++) - seq_printf(s, "avail ring[%d] %d\n", - j, avail->ring[j]); - used = vrh->vring.used; - seq_printf(s, "used flags 0x%x idx %d\n", - vringh16_to_cpu(vrh, used->flags), - vringh16_to_cpu(vrh, used->idx) & (num - 1)); - seq_printf(s, "used flags 0x%x idx %d\n", - vringh16_to_cpu(vrh, used->flags), - vringh16_to_cpu(vrh, used->idx)); - for (j = 0; j < num; j++) - seq_printf(s, "used ring[%d] id %d len %d\n", - j, vringh32_to_cpu(vrh, - used->ring[j].id), - vringh32_to_cpu(vrh, - used->ring[j].len)); - } - } - mutex_unlock(&vi->vop_mutex); - - return 0; -} - -DEFINE_SHOW_ATTRIBUTE(vop_vdev_info); - -void vop_init_debugfs(struct vop_info *vi) -{ - char name[16]; - - snprintf(name, sizeof(name), "%s%d", KBUILD_MODNAME, vi->vpdev->dnode); - vi->dbg = debugfs_create_dir(name, NULL); - debugfs_create_file("dp", 0444, vi->dbg, vi, &vop_dp_fops); - debugfs_create_file("vdev_info", 0444, vi->dbg, vi, &vop_vdev_info_fops); -} - -void vop_exit_debugfs(struct vop_info *vi) -{ - debugfs_remove_recursive(vi->dbg); -} diff --git a/drivers/misc/mic/vop/vop_main.c b/drivers/misc/mic/vop/vop_main.c deleted file mode 100644 index 714b94f42d38..000000000000 --- a/drivers/misc/mic/vop/vop_main.c +++ /dev/null @@ -1,784 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2016 Intel Corporation. - * - * Adapted from: - * - * virtio for kvm on s390 - * - * Copyright IBM Corp. 2008 - * - * Author(s): Christian Borntraeger - * - * Intel Virtio Over PCIe (VOP) driver. - */ -#include -#include -#include -#include -#include - -#include "vop_main.h" - -#define VOP_MAX_VRINGS 4 - -/* - * _vop_vdev - Allocated per virtio device instance injected by the peer. - * - * @vdev: Virtio device - * @desc: Virtio device page descriptor - * @dc: Virtio device control - * @vpdev: VOP device which is the parent for this virtio device - * @vr: Buffer for accessing the VRING - * @used_virt: Virtual address of used ring - * @used: DMA address of used ring - * @used_size: Size of the used buffer - * @reset_done: Track whether VOP reset is complete - * @virtio_cookie: Cookie returned upon requesting a interrupt - * @c2h_vdev_db: The doorbell used by the guest to interrupt the host - * @h2c_vdev_db: The doorbell used by the host to interrupt the guest - * @dnode: The destination node - */ -struct _vop_vdev { - struct virtio_device vdev; - struct mic_device_desc __iomem *desc; - struct mic_device_ctrl __iomem *dc; - struct vop_device *vpdev; - void __iomem *vr[VOP_MAX_VRINGS]; - void *used_virt[VOP_MAX_VRINGS]; - dma_addr_t used[VOP_MAX_VRINGS]; - int used_size[VOP_MAX_VRINGS]; - struct completion reset_done; - struct mic_irq *virtio_cookie; - int c2h_vdev_db; - int h2c_vdev_db; - int dnode; -}; - -#define to_vopvdev(vd) container_of(vd, struct _vop_vdev, vdev) - -#define _vop_aligned_desc_size(d) __mic_align(_vop_desc_size(d), 8) - -/* Helper API to obtain the parent of the virtio device */ -static inline struct device *_vop_dev(struct _vop_vdev *vdev) -{ - return vdev->vdev.dev.parent; -} - -static inline unsigned _vop_desc_size(struct mic_device_desc __iomem *desc) -{ - return sizeof(*desc) - + ioread8(&desc->num_vq) * sizeof(struct mic_vqconfig) - + ioread8(&desc->feature_len) * 2 - + ioread8(&desc->config_len); -} - -static inline struct mic_vqconfig __iomem * -_vop_vq_config(struct mic_device_desc __iomem *desc) -{ - return (struct mic_vqconfig __iomem *)(desc + 1); -} - -static inline u8 __iomem * -_vop_vq_features(struct mic_device_desc __iomem *desc) -{ - return (u8 __iomem *)(_vop_vq_config(desc) + ioread8(&desc->num_vq)); -} - -static inline u8 __iomem * -_vop_vq_configspace(struct mic_device_desc __iomem *desc) -{ - return _vop_vq_features(desc) + ioread8(&desc->feature_len) * 2; -} - -static inline unsigned -_vop_total_desc_size(struct mic_device_desc __iomem *desc) -{ - return _vop_aligned_desc_size(desc) + sizeof(struct mic_device_ctrl); -} - -/* This gets the device's feature bits. */ -static u64 vop_get_features(struct virtio_device *vdev) -{ - unsigned int i, bits; - u64 features = 0; - struct mic_device_desc __iomem *desc = to_vopvdev(vdev)->desc; - u8 __iomem *in_features = _vop_vq_features(desc); - int feature_len = ioread8(&desc->feature_len); - - bits = min_t(unsigned, feature_len, sizeof(vdev->features)) * 8; - for (i = 0; i < bits; i++) - if (ioread8(&in_features[i / 8]) & (BIT(i % 8))) - features |= BIT_ULL(i); - - return features; -} - -static void vop_transport_features(struct virtio_device *vdev) -{ - /* - * Packed ring isn't enabled on virtio_vop for now, - * because virtio_vop uses vring_new_virtqueue() which - * creates virtio rings on preallocated memory. - */ - __virtio_clear_bit(vdev, VIRTIO_F_RING_PACKED); - __virtio_set_bit(vdev, VIRTIO_F_ACCESS_PLATFORM); -} - -static int vop_finalize_features(struct virtio_device *vdev) -{ - unsigned int i, bits; - struct mic_device_desc __iomem *desc = to_vopvdev(vdev)->desc; - u8 feature_len = ioread8(&desc->feature_len); - /* Second half of bitmap is features we accept. */ - u8 __iomem *out_features = - _vop_vq_features(desc) + feature_len; - - /* Give virtio_ring a chance to accept features. */ - vring_transport_features(vdev); - - /* Give virtio_vop a chance to accept features. */ - vop_transport_features(vdev); - - memset_io(out_features, 0, feature_len); - bits = min_t(unsigned, feature_len, - sizeof(vdev->features)) * 8; - for (i = 0; i < bits; i++) { - if (__virtio_test_bit(vdev, i)) - iowrite8(ioread8(&out_features[i / 8]) | (1 << (i % 8)), - &out_features[i / 8]); - } - return 0; -} - -/* - * Reading and writing elements in config space - */ -static void vop_get(struct virtio_device *vdev, unsigned int offset, - void *buf, unsigned len) -{ - struct mic_device_desc __iomem *desc = to_vopvdev(vdev)->desc; - - if (offset + len > ioread8(&desc->config_len)) - return; - memcpy_fromio(buf, _vop_vq_configspace(desc) + offset, len); -} - -static void vop_set(struct virtio_device *vdev, unsigned int offset, - const void *buf, unsigned len) -{ - struct mic_device_desc __iomem *desc = to_vopvdev(vdev)->desc; - - if (offset + len > ioread8(&desc->config_len)) - return; - memcpy_toio(_vop_vq_configspace(desc) + offset, buf, len); -} - -/* - * The operations to get and set the status word just access the status - * field of the device descriptor. set_status also interrupts the host - * to tell about status changes. - */ -static u8 vop_get_status(struct virtio_device *vdev) -{ - return ioread8(&to_vopvdev(vdev)->desc->status); -} - -static void vop_set_status(struct virtio_device *dev, u8 status) -{ - struct _vop_vdev *vdev = to_vopvdev(dev); - struct vop_device *vpdev = vdev->vpdev; - - if (!status) - return; - iowrite8(status, &vdev->desc->status); - vpdev->hw_ops->send_intr(vpdev, vdev->c2h_vdev_db); -} - -/* Inform host on a virtio device reset and wait for ack from host */ -static void vop_reset_inform_host(struct virtio_device *dev) -{ - struct _vop_vdev *vdev = to_vopvdev(dev); - struct mic_device_ctrl __iomem *dc = vdev->dc; - struct vop_device *vpdev = vdev->vpdev; - int retry; - - iowrite8(0, &dc->host_ack); - iowrite8(1, &dc->vdev_reset); - vpdev->hw_ops->send_intr(vpdev, vdev->c2h_vdev_db); - - /* Wait till host completes all card accesses and acks the reset */ - for (retry = 100; retry--;) { - if (ioread8(&dc->host_ack)) - break; - msleep(100); - } - - dev_dbg(_vop_dev(vdev), "%s: retry: %d\n", __func__, retry); - - /* Reset status to 0 in case we timed out */ - iowrite8(0, &vdev->desc->status); -} - -static void vop_reset(struct virtio_device *dev) -{ - struct _vop_vdev *vdev = to_vopvdev(dev); - - dev_dbg(_vop_dev(vdev), "%s: virtio id %d\n", - __func__, dev->id.device); - - vop_reset_inform_host(dev); - complete_all(&vdev->reset_done); -} - -/* - * The virtio_ring code calls this API when it wants to notify the Host. - */ -static bool vop_notify(struct virtqueue *vq) -{ - struct _vop_vdev *vdev = vq->priv; - struct vop_device *vpdev = vdev->vpdev; - - vpdev->hw_ops->send_intr(vpdev, vdev->c2h_vdev_db); - return true; -} - -static void vop_del_vq(struct virtqueue *vq, int n) -{ - struct _vop_vdev *vdev = to_vopvdev(vq->vdev); - struct vop_device *vpdev = vdev->vpdev; - - dma_unmap_single(&vpdev->dev, vdev->used[n], - vdev->used_size[n], DMA_BIDIRECTIONAL); - free_pages((unsigned long)vdev->used_virt[n], - get_order(vdev->used_size[n])); - vring_del_virtqueue(vq); - vpdev->hw_ops->unmap(vpdev, vdev->vr[n]); - vdev->vr[n] = NULL; -} - -static void vop_del_vqs(struct virtio_device *dev) -{ - struct _vop_vdev *vdev = to_vopvdev(dev); - struct virtqueue *vq, *n; - int idx = 0; - - dev_dbg(_vop_dev(vdev), "%s\n", __func__); - - list_for_each_entry_safe(vq, n, &dev->vqs, list) - vop_del_vq(vq, idx++); -} - -static struct virtqueue *vop_new_virtqueue(unsigned int index, - unsigned int num, - struct virtio_device *vdev, - bool context, - void *pages, - bool (*notify)(struct virtqueue *vq), - void (*callback)(struct virtqueue *vq), - const char *name, - void *used) -{ - bool weak_barriers = false; - struct vring vring; - - vring_init(&vring, num, pages, MIC_VIRTIO_RING_ALIGN); - vring.used = used; - - return __vring_new_virtqueue(index, vring, vdev, weak_barriers, context, - notify, callback, name); -} - -/* - * This routine will assign vring's allocated in host/io memory. Code in - * virtio_ring.c however continues to access this io memory as if it were local - * memory without io accessors. - */ -static struct virtqueue *vop_find_vq(struct virtio_device *dev, - unsigned index, - void (*callback)(struct virtqueue *vq), - const char *name, bool ctx) -{ - struct _vop_vdev *vdev = to_vopvdev(dev); - struct vop_device *vpdev = vdev->vpdev; - struct mic_vqconfig __iomem *vqconfig; - struct mic_vqconfig config; - struct virtqueue *vq; - void __iomem *va; - struct _mic_vring_info __iomem *info; - void *used; - int vr_size, _vr_size, err, magic; - u8 type = ioread8(&vdev->desc->type); - - if (index >= ioread8(&vdev->desc->num_vq)) - return ERR_PTR(-ENOENT); - - if (!name) - return ERR_PTR(-ENOENT); - - /* First assign the vring's allocated in host memory */ - vqconfig = _vop_vq_config(vdev->desc) + index; - memcpy_fromio(&config, vqconfig, sizeof(config)); - _vr_size = round_up(vring_size(le16_to_cpu(config.num), MIC_VIRTIO_RING_ALIGN), 4); - vr_size = PAGE_ALIGN(_vr_size + sizeof(struct _mic_vring_info)); - va = vpdev->hw_ops->remap(vpdev, le64_to_cpu(config.address), vr_size); - if (!va) - return ERR_PTR(-ENOMEM); - vdev->vr[index] = va; - memset_io(va, 0x0, _vr_size); - - info = va + _vr_size; - magic = ioread32(&info->magic); - - if (WARN(magic != MIC_MAGIC + type + index, "magic mismatch")) { - err = -EIO; - goto unmap; - } - - vdev->used_size[index] = PAGE_ALIGN(sizeof(__u16) * 3 + - sizeof(struct vring_used_elem) * - le16_to_cpu(config.num)); - used = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, - get_order(vdev->used_size[index])); - vdev->used_virt[index] = used; - if (!used) { - err = -ENOMEM; - dev_err(_vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, err); - goto unmap; - } - - vq = vop_new_virtqueue(index, le16_to_cpu(config.num), dev, ctx, - (void __force *)va, vop_notify, callback, - name, used); - if (!vq) { - err = -ENOMEM; - goto free_used; - } - - vdev->used[index] = dma_map_single(&vpdev->dev, used, - vdev->used_size[index], - DMA_BIDIRECTIONAL); - if (dma_mapping_error(&vpdev->dev, vdev->used[index])) { - err = -ENOMEM; - dev_err(_vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, err); - goto del_vq; - } - writeq(vdev->used[index], &vqconfig->used_address); - - vq->priv = vdev; - return vq; -del_vq: - vring_del_virtqueue(vq); -free_used: - free_pages((unsigned long)used, - get_order(vdev->used_size[index])); -unmap: - vpdev->hw_ops->unmap(vpdev, vdev->vr[index]); - return ERR_PTR(err); -} - -static int vop_find_vqs(struct virtio_device *dev, unsigned nvqs, - struct virtqueue *vqs[], - vq_callback_t *callbacks[], - const char * const names[], const bool *ctx, - struct irq_affinity *desc) -{ - struct _vop_vdev *vdev = to_vopvdev(dev); - struct vop_device *vpdev = vdev->vpdev; - struct mic_device_ctrl __iomem *dc = vdev->dc; - int i, err, retry, queue_idx = 0; - - /* We must have this many virtqueues. */ - if (nvqs > ioread8(&vdev->desc->num_vq)) - return -ENOENT; - - for (i = 0; i < nvqs; ++i) { - if (!names[i]) { - vqs[i] = NULL; - continue; - } - - dev_dbg(_vop_dev(vdev), "%s: %d: %s\n", - __func__, i, names[i]); - vqs[i] = vop_find_vq(dev, queue_idx++, callbacks[i], names[i], - ctx ? ctx[i] : false); - if (IS_ERR(vqs[i])) { - err = PTR_ERR(vqs[i]); - goto error; - } - } - - iowrite8(1, &dc->used_address_updated); - /* - * Send an interrupt to the host to inform it that used - * rings have been re-assigned. - */ - vpdev->hw_ops->send_intr(vpdev, vdev->c2h_vdev_db); - for (retry = 100; --retry;) { - if (!ioread8(&dc->used_address_updated)) - break; - msleep(100); - } - - dev_dbg(_vop_dev(vdev), "%s: retry: %d\n", __func__, retry); - if (!retry) { - err = -ENODEV; - goto error; - } - - return 0; -error: - vop_del_vqs(dev); - return err; -} - -/* - * The config ops structure as defined by virtio config - */ -static const struct virtio_config_ops vop_vq_config_ops = { - .get_features = vop_get_features, - .finalize_features = vop_finalize_features, - .get = vop_get, - .set = vop_set, - .get_status = vop_get_status, - .set_status = vop_set_status, - .reset = vop_reset, - .find_vqs = vop_find_vqs, - .del_vqs = vop_del_vqs, -}; - -static irqreturn_t vop_virtio_intr_handler(int irq, void *data) -{ - struct _vop_vdev *vdev = data; - struct vop_device *vpdev = vdev->vpdev; - struct virtqueue *vq; - - vpdev->hw_ops->ack_interrupt(vpdev, vdev->h2c_vdev_db); - list_for_each_entry(vq, &vdev->vdev.vqs, list) - vring_interrupt(0, vq); - - return IRQ_HANDLED; -} - -static void vop_virtio_release_dev(struct device *_d) -{ - struct virtio_device *vdev = - container_of(_d, struct virtio_device, dev); - struct _vop_vdev *vop_vdev = - container_of(vdev, struct _vop_vdev, vdev); - - kfree(vop_vdev); -} - -/* - * adds a new device and register it with virtio - * appropriate drivers are loaded by the device model - */ -static int _vop_add_device(struct mic_device_desc __iomem *d, - unsigned int offset, struct vop_device *vpdev, - int dnode) -{ - struct _vop_vdev *vdev, *reg_dev = NULL; - int ret; - u8 type = ioread8(&d->type); - - vdev = kzalloc(sizeof(*vdev), GFP_KERNEL); - if (!vdev) - return -ENOMEM; - - vdev->vpdev = vpdev; - vdev->vdev.dev.parent = &vpdev->dev; - vdev->vdev.dev.release = vop_virtio_release_dev; - vdev->vdev.id.device = type; - vdev->vdev.config = &vop_vq_config_ops; - vdev->desc = d; - vdev->dc = (void __iomem *)d + _vop_aligned_desc_size(d); - vdev->dnode = dnode; - vdev->vdev.priv = (void *)(unsigned long)dnode; - init_completion(&vdev->reset_done); - - vdev->h2c_vdev_db = vpdev->hw_ops->next_db(vpdev); - vdev->virtio_cookie = vpdev->hw_ops->request_irq(vpdev, - vop_virtio_intr_handler, "virtio intr", - vdev, vdev->h2c_vdev_db); - if (IS_ERR(vdev->virtio_cookie)) { - ret = PTR_ERR(vdev->virtio_cookie); - goto kfree; - } - iowrite8((u8)vdev->h2c_vdev_db, &vdev->dc->h2c_vdev_db); - vdev->c2h_vdev_db = ioread8(&vdev->dc->c2h_vdev_db); - - ret = register_virtio_device(&vdev->vdev); - reg_dev = vdev; - if (ret) { - dev_err(_vop_dev(vdev), - "Failed to register vop device %u type %u\n", - offset, type); - goto free_irq; - } - writeq((unsigned long)vdev, &vdev->dc->vdev); - dev_dbg(_vop_dev(vdev), "%s: registered vop device %u type %u vdev %p\n", - __func__, offset, type, vdev); - - return 0; - -free_irq: - vpdev->hw_ops->free_irq(vpdev, vdev->virtio_cookie, vdev); -kfree: - if (reg_dev) - put_device(&vdev->vdev.dev); - else - kfree(vdev); - return ret; -} - -/* - * match for a vop device with a specific desc pointer - */ -static int vop_match_desc(struct device *dev, void *data) -{ - struct virtio_device *_dev = dev_to_virtio(dev); - struct _vop_vdev *vdev = to_vopvdev(_dev); - - return vdev->desc == (void __iomem *)data; -} - -static struct _vop_vdev *vop_dc_to_vdev(struct mic_device_ctrl __iomem *dc) -{ - return (struct _vop_vdev *)(unsigned long)readq(&dc->vdev); -} - -static void _vop_handle_config_change(struct mic_device_desc __iomem *d, - unsigned int offset, - struct vop_device *vpdev) -{ - struct mic_device_ctrl __iomem *dc - = (void __iomem *)d + _vop_aligned_desc_size(d); - struct _vop_vdev *vdev = vop_dc_to_vdev(dc); - - if (ioread8(&dc->config_change) != MIC_VIRTIO_PARAM_CONFIG_CHANGED) - return; - - dev_dbg(&vpdev->dev, "%s %d\n", __func__, __LINE__); - virtio_config_changed(&vdev->vdev); - iowrite8(1, &dc->guest_ack); -} - -/* - * removes a virtio device if a hot remove event has been - * requested by the host. - */ -static int _vop_remove_device(struct mic_device_desc __iomem *d, - unsigned int offset, struct vop_device *vpdev) -{ - struct mic_device_ctrl __iomem *dc - = (void __iomem *)d + _vop_aligned_desc_size(d); - struct _vop_vdev *vdev = vop_dc_to_vdev(dc); - u8 status; - int ret = -1; - - if (ioread8(&dc->config_change) == MIC_VIRTIO_PARAM_DEV_REMOVE) { - struct device *dev = get_device(&vdev->vdev.dev); - - dev_dbg(&vpdev->dev, - "%s %d config_change %d type %d vdev %p\n", - __func__, __LINE__, - ioread8(&dc->config_change), ioread8(&d->type), vdev); - status = ioread8(&d->status); - reinit_completion(&vdev->reset_done); - unregister_virtio_device(&vdev->vdev); - vpdev->hw_ops->free_irq(vpdev, vdev->virtio_cookie, vdev); - iowrite8(-1, &dc->h2c_vdev_db); - if (status & VIRTIO_CONFIG_S_DRIVER_OK) - wait_for_completion(&vdev->reset_done); - put_device(dev); - iowrite8(1, &dc->guest_ack); - dev_dbg(&vpdev->dev, "%s %d guest_ack %d\n", - __func__, __LINE__, ioread8(&dc->guest_ack)); - iowrite8(-1, &d->type); - ret = 0; - } - return ret; -} - -#define REMOVE_DEVICES true - -static void _vop_scan_devices(void __iomem *dp, struct vop_device *vpdev, - bool remove, int dnode) -{ - s8 type; - unsigned int i; - struct mic_device_desc __iomem *d; - struct mic_device_ctrl __iomem *dc; - struct device *dev; - - for (i = sizeof(struct mic_bootparam); - i < MIC_DP_SIZE; i += _vop_total_desc_size(d)) { - d = dp + i; - dc = (void __iomem *)d + _vop_aligned_desc_size(d); - /* - * This read barrier is paired with the corresponding write - * barrier on the host which is inserted before adding or - * removing a virtio device descriptor, by updating the type. - */ - rmb(); - type = ioread8(&d->type); - - /* end of list */ - if (type == 0) - break; - - if (type == -1) - continue; - - /* device already exists */ - dev = device_find_child(&vpdev->dev, (void __force *)d, - vop_match_desc); - if (dev) { - if (remove) - iowrite8(MIC_VIRTIO_PARAM_DEV_REMOVE, - &dc->config_change); - put_device(dev); - _vop_handle_config_change(d, i, vpdev); - _vop_remove_device(d, i, vpdev); - if (remove) { - iowrite8(0, &dc->config_change); - iowrite8(0, &dc->guest_ack); - } - continue; - } - - /* new device */ - dev_dbg(&vpdev->dev, "%s %d Adding new virtio device %p\n", - __func__, __LINE__, d); - if (!remove) - _vop_add_device(d, i, vpdev, dnode); - } -} - -static void vop_scan_devices(struct vop_info *vi, - struct vop_device *vpdev, bool remove) -{ - void __iomem *dp = vpdev->hw_ops->get_remote_dp(vpdev); - - if (!dp) - return; - mutex_lock(&vi->vop_mutex); - _vop_scan_devices(dp, vpdev, remove, vpdev->dnode); - mutex_unlock(&vi->vop_mutex); -} - -/* - * vop_hotplug_device tries to find changes in the device page. - */ -static void vop_hotplug_devices(struct work_struct *work) -{ - struct vop_info *vi = container_of(work, struct vop_info, - hotplug_work); - - vop_scan_devices(vi, vi->vpdev, !REMOVE_DEVICES); -} - -/* - * Interrupt handler for hot plug/config changes etc. - */ -static irqreturn_t vop_extint_handler(int irq, void *data) -{ - struct vop_info *vi = data; - struct mic_bootparam __iomem *bp; - struct vop_device *vpdev = vi->vpdev; - - bp = vpdev->hw_ops->get_remote_dp(vpdev); - dev_dbg(&vpdev->dev, "%s %d hotplug work\n", - __func__, __LINE__); - vpdev->hw_ops->ack_interrupt(vpdev, ioread8(&bp->h2c_config_db)); - schedule_work(&vi->hotplug_work); - return IRQ_HANDLED; -} - -static int vop_driver_probe(struct vop_device *vpdev) -{ - struct vop_info *vi; - int rc; - - vi = kzalloc(sizeof(*vi), GFP_KERNEL); - if (!vi) { - rc = -ENOMEM; - goto exit; - } - dev_set_drvdata(&vpdev->dev, vi); - vi->vpdev = vpdev; - - mutex_init(&vi->vop_mutex); - INIT_WORK(&vi->hotplug_work, vop_hotplug_devices); - if (vpdev->dnode) { - rc = vop_host_init(vi); - if (rc < 0) - goto free; - } else { - struct mic_bootparam __iomem *bootparam; - - vop_scan_devices(vi, vpdev, !REMOVE_DEVICES); - - vi->h2c_config_db = vpdev->hw_ops->next_db(vpdev); - vi->cookie = vpdev->hw_ops->request_irq(vpdev, - vop_extint_handler, - "virtio_config_intr", - vi, vi->h2c_config_db); - if (IS_ERR(vi->cookie)) { - rc = PTR_ERR(vi->cookie); - goto free; - } - bootparam = vpdev->hw_ops->get_remote_dp(vpdev); - iowrite8(vi->h2c_config_db, &bootparam->h2c_config_db); - } - vop_init_debugfs(vi); - return 0; -free: - kfree(vi); -exit: - return rc; -} - -static void vop_driver_remove(struct vop_device *vpdev) -{ - struct vop_info *vi = dev_get_drvdata(&vpdev->dev); - - if (vpdev->dnode) { - vop_host_uninit(vi); - } else { - struct mic_bootparam __iomem *bootparam = - vpdev->hw_ops->get_remote_dp(vpdev); - if (bootparam) - iowrite8(-1, &bootparam->h2c_config_db); - vpdev->hw_ops->free_irq(vpdev, vi->cookie, vi); - flush_work(&vi->hotplug_work); - vop_scan_devices(vi, vpdev, REMOVE_DEVICES); - } - vop_exit_debugfs(vi); - kfree(vi); -} - -static const struct vop_device_id id_table[] = { - { VOP_DEV_TRNSP, VOP_DEV_ANY_ID }, - { 0 }, -}; - -static struct vop_driver vop_driver = { - .driver.name = KBUILD_MODNAME, - .driver.owner = THIS_MODULE, - .id_table = id_table, - .probe = vop_driver_probe, - .remove = vop_driver_remove, -}; - -module_vop_driver(vop_driver); - -MODULE_DEVICE_TABLE(mbus, id_table); -MODULE_AUTHOR("Intel Corporation"); -MODULE_DESCRIPTION("Intel(R) Virtio Over PCIe (VOP) driver"); -MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/vop/vop_main.h b/drivers/misc/mic/vop/vop_main.h deleted file mode 100644 index 2451d9218137..000000000000 --- a/drivers/misc/mic/vop/vop_main.h +++ /dev/null @@ -1,158 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2016 Intel Corporation. - * - * Intel Virtio Over PCIe (VOP) driver. - */ -#ifndef _VOP_MAIN_H_ -#define _VOP_MAIN_H_ - -#include -#include -#include -#include - -#include -#include "../common/mic_dev.h" - -#include "../bus/vop_bus.h" - -/* - * Note on endianness. - * 1. Host can be both BE or LE - * 2. Guest/card is LE. Host uses le_to_cpu to access desc/avail - * rings and ioreadXX/iowriteXX to access used ring. - * 3. Device page exposed by host to guest contains LE values. Guest - * accesses these using ioreadXX/iowriteXX etc. This way in general we - * obey the virtio spec according to which guest works with native - * endianness and host is aware of guest endianness and does all - * required endianness conversion. - * 4. Data provided from user space to guest (in ADD_DEVICE and - * CONFIG_CHANGE ioctl's) is not interpreted by the driver and should be - * in guest endianness. - */ - -/* - * vop_info - Allocated per invocation of VOP probe - * - * @vpdev: VOP device - * @hotplug_work: Handle virtio device creation, deletion and configuration - * @cookie: Cookie received upon requesting a virtio configuration interrupt - * @h2c_config_db: The doorbell used by the peer to indicate a config change - * @vdev_list: List of "active" virtio devices injected in the peer node - * @vop_mutex: Synchronize access to the device page as well as serialize - * creation/deletion of virtio devices on the peer node - * @dp: Peer device page information - * @dbg: Debugfs entry - * @dma_ch: The DMA channel used by this transport for data transfers. - * @name: Name for this transport used in misc device creation. - * @miscdev: The misc device registered. - */ -struct vop_info { - struct vop_device *vpdev; - struct work_struct hotplug_work; - struct mic_irq *cookie; - int h2c_config_db; - struct list_head vdev_list; - struct mutex vop_mutex; - void __iomem *dp; - struct dentry *dbg; - struct dma_chan *dma_ch; - char name[16]; - struct miscdevice miscdev; -}; - -/** - * struct vop_vringh - Virtio ring host information. - * - * @vring: The VOP vring used for setting up user space mappings. - * @vrh: The host VRINGH used for accessing the card vrings. - * @riov: The VRINGH read kernel IOV. - * @wiov: The VRINGH write kernel IOV. - * @head: The VRINGH head index address passed to vringh_getdesc_kern(..). - * @vr_mutex: Mutex for synchronizing access to the VRING. - * @buf: Temporary kernel buffer used to copy in/out data - * from/to the card via DMA. - * @buf_da: dma address of buf. - * @vdev: Back pointer to VOP virtio device for vringh_notify(..). - */ -struct vop_vringh { - struct mic_vring vring; - struct vringh vrh; - struct vringh_kiov riov; - struct vringh_kiov wiov; - u16 head; - struct mutex vr_mutex; - void *buf; - dma_addr_t buf_da; - struct vop_vdev *vdev; -}; - -/** - * struct vop_vdev - Host information for a card Virtio device. - * - * @virtio_id - Virtio device id. - * @waitq - Waitqueue to allow ring3 apps to poll. - * @vpdev - pointer to VOP bus device. - * @poll_wake - Used for waking up threads blocked in poll. - * @out_bytes - Debug stats for number of bytes copied from host to card. - * @in_bytes - Debug stats for number of bytes copied from card to host. - * @out_bytes_dma - Debug stats for number of bytes copied from host to card - * using DMA. - * @in_bytes_dma - Debug stats for number of bytes copied from card to host - * using DMA. - * @tx_len_unaligned - Debug stats for number of bytes copied to the card where - * the transfer length did not have the required DMA alignment. - * @tx_dst_unaligned - Debug stats for number of bytes copied where the - * destination address on the card did not have the required DMA alignment. - * @vvr - Store per VRING data structures. - * @virtio_bh_work - Work struct used to schedule virtio bottom half handling. - * @dd - Virtio device descriptor. - * @dc - Virtio device control fields. - * @list - List of Virtio devices. - * @virtio_db - The doorbell used by the card to interrupt the host. - * @virtio_cookie - The cookie returned while requesting interrupts. - * @vi: Transport information. - * @vdev_mutex: Mutex synchronizing virtio device injection, - * removal and data transfers. - * @destroy: Track if a virtio device is being destroyed. - * @deleted: The virtio device has been deleted. - */ -struct vop_vdev { - int virtio_id; - wait_queue_head_t waitq; - struct vop_device *vpdev; - int poll_wake; - unsigned long out_bytes; - unsigned long in_bytes; - unsigned long out_bytes_dma; - unsigned long in_bytes_dma; - unsigned long tx_len_unaligned; - unsigned long tx_dst_unaligned; - unsigned long rx_dst_unaligned; - struct vop_vringh vvr[MIC_MAX_VRINGS]; - struct work_struct virtio_bh_work; - struct mic_device_desc *dd; - struct mic_device_ctrl *dc; - struct list_head list; - int virtio_db; - struct mic_irq *virtio_cookie; - struct vop_info *vi; - struct mutex vdev_mutex; - struct completion destroy; - bool deleted; -}; - -/* Helper API to check if a virtio device is running */ -static inline bool vop_vdevup(struct vop_vdev *vdev) -{ - return !!vdev->dd->status; -} - -void vop_init_debugfs(struct vop_info *vi); -void vop_exit_debugfs(struct vop_info *vi); -int vop_host_init(struct vop_info *vi); -void vop_host_uninit(struct vop_info *vi); -#endif diff --git a/drivers/misc/mic/vop/vop_vringh.c b/drivers/misc/mic/vop/vop_vringh.c deleted file mode 100644 index 7014ffe88632..000000000000 --- a/drivers/misc/mic/vop/vop_vringh.c +++ /dev/null @@ -1,1166 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2016 Intel Corporation. - * - * Intel Virtio Over PCIe (VOP) driver. - */ -#include -#include -#include - -#include -#include "../common/mic_dev.h" - -#include -#include "vop_main.h" - -/* Helper API to obtain the VOP PCIe device */ -static inline struct device *vop_dev(struct vop_vdev *vdev) -{ - return vdev->vpdev->dev.parent; -} - -/* Helper API to check if a virtio device is initialized */ -static inline int vop_vdev_inited(struct vop_vdev *vdev) -{ - if (!vdev) - return -EINVAL; - /* Device has not been created yet */ - if (!vdev->dd || !vdev->dd->type) { - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, -EINVAL); - return -EINVAL; - } - /* Device has been removed/deleted */ - if (vdev->dd->type == -1) { - dev_dbg(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, -ENODEV); - return -ENODEV; - } - return 0; -} - -static void _vop_notify(struct vringh *vrh) -{ - struct vop_vringh *vvrh = container_of(vrh, struct vop_vringh, vrh); - struct vop_vdev *vdev = vvrh->vdev; - struct vop_device *vpdev = vdev->vpdev; - s8 db = vdev->dc->h2c_vdev_db; - - if (db != -1) - vpdev->hw_ops->send_intr(vpdev, db); -} - -static void vop_virtio_init_post(struct vop_vdev *vdev) -{ - struct mic_vqconfig *vqconfig = mic_vq_config(vdev->dd); - struct vop_device *vpdev = vdev->vpdev; - int i, used_size; - - for (i = 0; i < vdev->dd->num_vq; i++) { - used_size = PAGE_ALIGN(sizeof(u16) * 3 + - sizeof(struct vring_used_elem) * - le16_to_cpu(vqconfig->num)); - if (!le64_to_cpu(vqconfig[i].used_address)) { - dev_warn(vop_dev(vdev), "used_address zero??\n"); - continue; - } - vdev->vvr[i].vrh.vring.used = - (void __force *)vpdev->hw_ops->remap( - vpdev, - le64_to_cpu(vqconfig[i].used_address), - used_size); - } - - vdev->dc->used_address_updated = 0; - - dev_info(vop_dev(vdev), "%s: device type %d LINKUP\n", - __func__, vdev->virtio_id); -} - -static inline void vop_virtio_device_reset(struct vop_vdev *vdev) -{ - int i; - - dev_dbg(vop_dev(vdev), "%s: status %d device type %d RESET\n", - __func__, vdev->dd->status, vdev->virtio_id); - - for (i = 0; i < vdev->dd->num_vq; i++) - /* - * Avoid lockdep false positive. The + 1 is for the vop - * mutex which is held in the reset devices code path. - */ - mutex_lock_nested(&vdev->vvr[i].vr_mutex, i + 1); - - /* 0 status means "reset" */ - vdev->dd->status = 0; - vdev->dc->vdev_reset = 0; - vdev->dc->host_ack = 1; - - for (i = 0; i < vdev->dd->num_vq; i++) { - struct vringh *vrh = &vdev->vvr[i].vrh; - - vdev->vvr[i].vring.info->avail_idx = 0; - vrh->completed = 0; - vrh->last_avail_idx = 0; - vrh->last_used_idx = 0; - } - - for (i = 0; i < vdev->dd->num_vq; i++) - mutex_unlock(&vdev->vvr[i].vr_mutex); -} - -static void vop_virtio_reset_devices(struct vop_info *vi) -{ - struct list_head *pos, *tmp; - struct vop_vdev *vdev; - - list_for_each_safe(pos, tmp, &vi->vdev_list) { - vdev = list_entry(pos, struct vop_vdev, list); - vop_virtio_device_reset(vdev); - vdev->poll_wake = 1; - wake_up(&vdev->waitq); - } -} - -static void vop_bh_handler(struct work_struct *work) -{ - struct vop_vdev *vdev = container_of(work, struct vop_vdev, - virtio_bh_work); - - if (vdev->dc->used_address_updated) - vop_virtio_init_post(vdev); - - if (vdev->dc->vdev_reset) - vop_virtio_device_reset(vdev); - - vdev->poll_wake = 1; - wake_up(&vdev->waitq); -} - -static irqreturn_t _vop_virtio_intr_handler(int irq, void *data) -{ - struct vop_vdev *vdev = data; - struct vop_device *vpdev = vdev->vpdev; - - vpdev->hw_ops->ack_interrupt(vpdev, vdev->virtio_db); - schedule_work(&vdev->virtio_bh_work); - return IRQ_HANDLED; -} - -static int vop_virtio_config_change(struct vop_vdev *vdev, void *argp) -{ - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wake); - int ret = 0, retry, i; - struct vop_device *vpdev = vdev->vpdev; - struct vop_info *vi = dev_get_drvdata(&vpdev->dev); - struct mic_bootparam *bootparam = vpdev->hw_ops->get_dp(vpdev); - s8 db = bootparam->h2c_config_db; - - mutex_lock(&vi->vop_mutex); - for (i = 0; i < vdev->dd->num_vq; i++) - mutex_lock_nested(&vdev->vvr[i].vr_mutex, i + 1); - - if (db == -1 || vdev->dd->type == -1) { - ret = -EIO; - goto exit; - } - - memcpy(mic_vq_configspace(vdev->dd), argp, vdev->dd->config_len); - vdev->dc->config_change = MIC_VIRTIO_PARAM_CONFIG_CHANGED; - vpdev->hw_ops->send_intr(vpdev, db); - - for (retry = 100; retry--;) { - ret = wait_event_timeout(wake, vdev->dc->guest_ack, - msecs_to_jiffies(100)); - if (ret) - break; - } - - dev_dbg(vop_dev(vdev), - "%s %d retry: %d\n", __func__, __LINE__, retry); - vdev->dc->config_change = 0; - vdev->dc->guest_ack = 0; -exit: - for (i = 0; i < vdev->dd->num_vq; i++) - mutex_unlock(&vdev->vvr[i].vr_mutex); - mutex_unlock(&vi->vop_mutex); - return ret; -} - -static int vop_copy_dp_entry(struct vop_vdev *vdev, - struct mic_device_desc *argp, __u8 *type, - struct mic_device_desc **devpage) -{ - struct vop_device *vpdev = vdev->vpdev; - struct mic_device_desc *devp; - struct mic_vqconfig *vqconfig; - int ret = 0, i; - bool slot_found = false; - - vqconfig = mic_vq_config(argp); - for (i = 0; i < argp->num_vq; i++) { - if (le16_to_cpu(vqconfig[i].num) > MIC_MAX_VRING_ENTRIES) { - ret = -EINVAL; - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, ret); - goto exit; - } - } - - /* Find the first free device page entry */ - for (i = sizeof(struct mic_bootparam); - i < MIC_DP_SIZE - mic_total_desc_size(argp); - i += mic_total_desc_size(devp)) { - devp = vpdev->hw_ops->get_dp(vpdev) + i; - if (devp->type == 0 || devp->type == -1) { - slot_found = true; - break; - } - } - if (!slot_found) { - ret = -EINVAL; - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, ret); - goto exit; - } - /* - * Save off the type before doing the memcpy. Type will be set in the - * end after completing all initialization for the new device. - */ - *type = argp->type; - argp->type = 0; - memcpy(devp, argp, mic_desc_size(argp)); - - *devpage = devp; -exit: - return ret; -} - -static void vop_init_device_ctrl(struct vop_vdev *vdev, - struct mic_device_desc *devpage) -{ - struct mic_device_ctrl *dc; - - dc = (void *)devpage + mic_aligned_desc_size(devpage); - - dc->config_change = 0; - dc->guest_ack = 0; - dc->vdev_reset = 0; - dc->host_ack = 0; - dc->used_address_updated = 0; - dc->c2h_vdev_db = -1; - dc->h2c_vdev_db = -1; - vdev->dc = dc; -} - -static int vop_virtio_add_device(struct vop_vdev *vdev, - struct mic_device_desc *argp) -{ - struct vop_info *vi = vdev->vi; - struct vop_device *vpdev = vi->vpdev; - struct mic_device_desc *dd = NULL; - struct mic_vqconfig *vqconfig; - int vr_size, i, j, ret; - u8 type = 0; - s8 db = -1; - char irqname[16]; - struct mic_bootparam *bootparam; - u16 num; - dma_addr_t vr_addr; - - bootparam = vpdev->hw_ops->get_dp(vpdev); - init_waitqueue_head(&vdev->waitq); - INIT_LIST_HEAD(&vdev->list); - vdev->vpdev = vpdev; - - ret = vop_copy_dp_entry(vdev, argp, &type, &dd); - if (ret) { - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, ret); - return ret; - } - - vop_init_device_ctrl(vdev, dd); - - vdev->dd = dd; - vdev->virtio_id = type; - vqconfig = mic_vq_config(dd); - INIT_WORK(&vdev->virtio_bh_work, vop_bh_handler); - - for (i = 0; i < dd->num_vq; i++) { - struct vop_vringh *vvr = &vdev->vvr[i]; - struct mic_vring *vr = &vdev->vvr[i].vring; - - num = le16_to_cpu(vqconfig[i].num); - mutex_init(&vvr->vr_mutex); - vr_size = PAGE_ALIGN(round_up(vring_size(num, MIC_VIRTIO_RING_ALIGN), 4) + - sizeof(struct _mic_vring_info)); - vr->va = (void *) - __get_free_pages(GFP_KERNEL | __GFP_ZERO, - get_order(vr_size)); - if (!vr->va) { - ret = -ENOMEM; - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, ret); - goto err; - } - vr->len = vr_size; - vr->info = vr->va + round_up(vring_size(num, MIC_VIRTIO_RING_ALIGN), 4); - vr->info->magic = cpu_to_le32(MIC_MAGIC + vdev->virtio_id + i); - vr_addr = dma_map_single(&vpdev->dev, vr->va, vr_size, - DMA_BIDIRECTIONAL); - if (dma_mapping_error(&vpdev->dev, vr_addr)) { - free_pages((unsigned long)vr->va, get_order(vr_size)); - ret = -ENOMEM; - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, ret); - goto err; - } - vqconfig[i].address = cpu_to_le64(vr_addr); - - vring_init(&vr->vr, num, vr->va, MIC_VIRTIO_RING_ALIGN); - ret = vringh_init_kern(&vvr->vrh, - *(u32 *)mic_vq_features(vdev->dd), - num, false, vr->vr.desc, vr->vr.avail, - vr->vr.used); - if (ret) { - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, ret); - goto err; - } - vringh_kiov_init(&vvr->riov, NULL, 0); - vringh_kiov_init(&vvr->wiov, NULL, 0); - vvr->head = USHRT_MAX; - vvr->vdev = vdev; - vvr->vrh.notify = _vop_notify; - dev_dbg(&vpdev->dev, - "%s %d index %d va %p info %p vr_size 0x%x\n", - __func__, __LINE__, i, vr->va, vr->info, vr_size); - vvr->buf = (void *)__get_free_pages(GFP_KERNEL, - get_order(VOP_INT_DMA_BUF_SIZE)); - vvr->buf_da = dma_map_single(&vpdev->dev, - vvr->buf, VOP_INT_DMA_BUF_SIZE, - DMA_BIDIRECTIONAL); - } - - snprintf(irqname, sizeof(irqname), "vop%dvirtio%d", vpdev->index, - vdev->virtio_id); - vdev->virtio_db = vpdev->hw_ops->next_db(vpdev); - vdev->virtio_cookie = vpdev->hw_ops->request_irq(vpdev, - _vop_virtio_intr_handler, irqname, vdev, - vdev->virtio_db); - if (IS_ERR(vdev->virtio_cookie)) { - ret = PTR_ERR(vdev->virtio_cookie); - dev_dbg(&vpdev->dev, "request irq failed\n"); - goto err; - } - - vdev->dc->c2h_vdev_db = vdev->virtio_db; - - /* - * Order the type update with previous stores. This write barrier - * is paired with the corresponding read barrier before the uncached - * system memory read of the type, on the card while scanning the - * device page. - */ - smp_wmb(); - dd->type = type; - argp->type = type; - - if (bootparam) { - db = bootparam->h2c_config_db; - if (db != -1) - vpdev->hw_ops->send_intr(vpdev, db); - } - dev_dbg(&vpdev->dev, "Added virtio id %d db %d\n", dd->type, db); - return 0; -err: - vqconfig = mic_vq_config(dd); - for (j = 0; j < i; j++) { - struct vop_vringh *vvr = &vdev->vvr[j]; - - dma_unmap_single(&vpdev->dev, le64_to_cpu(vqconfig[j].address), - vvr->vring.len, DMA_BIDIRECTIONAL); - free_pages((unsigned long)vvr->vring.va, - get_order(vvr->vring.len)); - } - return ret; -} - -static void vop_dev_remove(struct vop_info *pvi, struct mic_device_ctrl *devp, - struct vop_device *vpdev) -{ - struct mic_bootparam *bootparam = vpdev->hw_ops->get_dp(vpdev); - s8 db; - int ret, retry; - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wake); - - devp->config_change = MIC_VIRTIO_PARAM_DEV_REMOVE; - db = bootparam->h2c_config_db; - if (db != -1) - vpdev->hw_ops->send_intr(vpdev, db); - else - goto done; - for (retry = 15; retry--;) { - ret = wait_event_timeout(wake, devp->guest_ack, - msecs_to_jiffies(1000)); - if (ret) - break; - } -done: - devp->config_change = 0; - devp->guest_ack = 0; -} - -static void vop_virtio_del_device(struct vop_vdev *vdev) -{ - struct vop_info *vi = vdev->vi; - struct vop_device *vpdev = vdev->vpdev; - int i; - struct mic_vqconfig *vqconfig; - struct mic_bootparam *bootparam = vpdev->hw_ops->get_dp(vpdev); - - if (!bootparam) - goto skip_hot_remove; - vop_dev_remove(vi, vdev->dc, vpdev); -skip_hot_remove: - vpdev->hw_ops->free_irq(vpdev, vdev->virtio_cookie, vdev); - flush_work(&vdev->virtio_bh_work); - vqconfig = mic_vq_config(vdev->dd); - for (i = 0; i < vdev->dd->num_vq; i++) { - struct vop_vringh *vvr = &vdev->vvr[i]; - - dma_unmap_single(&vpdev->dev, - vvr->buf_da, VOP_INT_DMA_BUF_SIZE, - DMA_BIDIRECTIONAL); - free_pages((unsigned long)vvr->buf, - get_order(VOP_INT_DMA_BUF_SIZE)); - vringh_kiov_cleanup(&vvr->riov); - vringh_kiov_cleanup(&vvr->wiov); - dma_unmap_single(&vpdev->dev, le64_to_cpu(vqconfig[i].address), - vvr->vring.len, DMA_BIDIRECTIONAL); - free_pages((unsigned long)vvr->vring.va, - get_order(vvr->vring.len)); - } - /* - * Order the type update with previous stores. This write barrier - * is paired with the corresponding read barrier before the uncached - * system memory read of the type, on the card while scanning the - * device page. - */ - smp_wmb(); - vdev->dd->type = -1; -} - -/* - * vop_sync_dma - Wrapper for synchronous DMAs. - * - * @dev - The address of the pointer to the device instance used - * for DMA registration. - * @dst - destination DMA address. - * @src - source DMA address. - * @len - size of the transfer. - * - * Return DMA_SUCCESS on success - */ -static int vop_sync_dma(struct vop_vdev *vdev, dma_addr_t dst, dma_addr_t src, - size_t len) -{ - int err = 0; - struct dma_device *ddev; - struct dma_async_tx_descriptor *tx; - struct vop_info *vi = dev_get_drvdata(&vdev->vpdev->dev); - struct dma_chan *vop_ch = vi->dma_ch; - - if (!vop_ch) { - err = -EBUSY; - goto error; - } - ddev = vop_ch->device; - tx = ddev->device_prep_dma_memcpy(vop_ch, dst, src, len, - DMA_PREP_FENCE); - if (!tx) { - err = -ENOMEM; - goto error; - } else { - dma_cookie_t cookie; - - cookie = tx->tx_submit(tx); - if (dma_submit_error(cookie)) { - err = -ENOMEM; - goto error; - } - dma_async_issue_pending(vop_ch); - err = dma_sync_wait(vop_ch, cookie); - } -error: - if (err) - dev_err(&vi->vpdev->dev, "%s %d err %d\n", - __func__, __LINE__, err); - return err; -} - -#define VOP_USE_DMA true - -/* - * Initiates the copies across the PCIe bus from card memory to a user - * space buffer. When transfers are done using DMA, source/destination - * addresses and transfer length must follow the alignment requirements of - * the MIC DMA engine. - */ -static int vop_virtio_copy_to_user(struct vop_vdev *vdev, void __user *ubuf, - size_t len, u64 daddr, size_t dlen, - int vr_idx) -{ - struct vop_device *vpdev = vdev->vpdev; - void __iomem *dbuf = vpdev->hw_ops->remap(vpdev, daddr, len); - struct vop_vringh *vvr = &vdev->vvr[vr_idx]; - struct vop_info *vi = dev_get_drvdata(&vpdev->dev); - size_t dma_alignment; - bool x200; - size_t dma_offset, partlen; - int err; - - if (!VOP_USE_DMA || !vi->dma_ch) { - if (copy_to_user(ubuf, (void __force *)dbuf, len)) { - err = -EFAULT; - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, err); - goto err; - } - vdev->in_bytes += len; - err = 0; - goto err; - } - - dma_alignment = 1 << vi->dma_ch->device->copy_align; - x200 = is_dma_copy_aligned(vi->dma_ch->device, 1, 1, 1); - - dma_offset = daddr - round_down(daddr, dma_alignment); - daddr -= dma_offset; - len += dma_offset; - /* - * X100 uses DMA addresses as seen by the card so adding - * the aperture base is not required for DMA. However x200 - * requires DMA addresses to be an offset into the bar so - * add the aperture base for x200. - */ - if (x200) - daddr += vpdev->aper->pa; - while (len) { - partlen = min_t(size_t, len, VOP_INT_DMA_BUF_SIZE); - err = vop_sync_dma(vdev, vvr->buf_da, daddr, - ALIGN(partlen, dma_alignment)); - if (err) { - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, err); - goto err; - } - if (copy_to_user(ubuf, vvr->buf + dma_offset, - partlen - dma_offset)) { - err = -EFAULT; - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, err); - goto err; - } - daddr += partlen; - ubuf += partlen; - dbuf += partlen; - vdev->in_bytes_dma += partlen; - vdev->in_bytes += partlen; - len -= partlen; - dma_offset = 0; - } - err = 0; -err: - vpdev->hw_ops->unmap(vpdev, dbuf); - dev_dbg(vop_dev(vdev), - "%s: ubuf %p dbuf %p len 0x%zx vr_idx 0x%x\n", - __func__, ubuf, dbuf, len, vr_idx); - return err; -} - -/* - * Initiates copies across the PCIe bus from a user space buffer to card - * memory. When transfers are done using DMA, source/destination addresses - * and transfer length must follow the alignment requirements of the MIC - * DMA engine. - */ -static int vop_virtio_copy_from_user(struct vop_vdev *vdev, void __user *ubuf, - size_t len, u64 daddr, size_t dlen, - int vr_idx) -{ - struct vop_device *vpdev = vdev->vpdev; - void __iomem *dbuf = vpdev->hw_ops->remap(vpdev, daddr, len); - struct vop_vringh *vvr = &vdev->vvr[vr_idx]; - struct vop_info *vi = dev_get_drvdata(&vdev->vpdev->dev); - size_t dma_alignment; - bool x200; - size_t partlen; - bool dma = VOP_USE_DMA && vi->dma_ch; - int err = 0; - size_t offset = 0; - - if (dma) { - dma_alignment = 1 << vi->dma_ch->device->copy_align; - x200 = is_dma_copy_aligned(vi->dma_ch->device, 1, 1, 1); - - if (daddr & (dma_alignment - 1)) { - vdev->tx_dst_unaligned += len; - dma = false; - } else if (ALIGN(len, dma_alignment) > dlen) { - vdev->tx_len_unaligned += len; - dma = false; - } - } - - if (!dma) - goto memcpy; - - /* - * X100 uses DMA addresses as seen by the card so adding - * the aperture base is not required for DMA. However x200 - * requires DMA addresses to be an offset into the bar so - * add the aperture base for x200. - */ - if (x200) - daddr += vpdev->aper->pa; - while (len) { - partlen = min_t(size_t, len, VOP_INT_DMA_BUF_SIZE); - - if (copy_from_user(vvr->buf, ubuf, partlen)) { - err = -EFAULT; - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, err); - goto err; - } - err = vop_sync_dma(vdev, daddr, vvr->buf_da, - ALIGN(partlen, dma_alignment)); - if (err) { - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, err); - goto err; - } - daddr += partlen; - ubuf += partlen; - dbuf += partlen; - vdev->out_bytes_dma += partlen; - vdev->out_bytes += partlen; - len -= partlen; - } -memcpy: - /* - * We are copying to IO below and should ideally use something - * like copy_from_user_toio(..) if it existed. - */ - while (len) { - partlen = min_t(size_t, len, VOP_INT_DMA_BUF_SIZE); - - if (copy_from_user(vvr->buf, ubuf + offset, partlen)) { - err = -EFAULT; - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, err); - goto err; - } - memcpy_toio(dbuf + offset, vvr->buf, partlen); - offset += partlen; - vdev->out_bytes += partlen; - len -= partlen; - } - err = 0; -err: - vpdev->hw_ops->unmap(vpdev, dbuf); - dev_dbg(vop_dev(vdev), - "%s: ubuf %p dbuf %p len 0x%zx vr_idx 0x%x\n", - __func__, ubuf, dbuf, len, vr_idx); - return err; -} - -#define MIC_VRINGH_READ true - -/* Determine the total number of bytes consumed in a VRINGH KIOV */ -static inline u32 vop_vringh_iov_consumed(struct vringh_kiov *iov) -{ - int i; - u32 total = iov->consumed; - - for (i = 0; i < iov->i; i++) - total += iov->iov[i].iov_len; - return total; -} - -/* - * Traverse the VRINGH KIOV and issue the APIs to trigger the copies. - * This API is heavily based on the vringh_iov_xfer(..) implementation - * in vringh.c. The reason we cannot reuse vringh_iov_pull_kern(..) - * and vringh_iov_push_kern(..) directly is because there is no - * way to override the VRINGH xfer(..) routines as of v3.10. - */ -static int vop_vringh_copy(struct vop_vdev *vdev, struct vringh_kiov *iov, - void __user *ubuf, size_t len, bool read, int vr_idx, - size_t *out_len) -{ - int ret = 0; - size_t partlen, tot_len = 0; - - while (len && iov->i < iov->used) { - struct kvec *kiov = &iov->iov[iov->i]; - unsigned long daddr = (unsigned long)kiov->iov_base; - - partlen = min(kiov->iov_len, len); - if (read) - ret = vop_virtio_copy_to_user(vdev, ubuf, partlen, - daddr, - kiov->iov_len, - vr_idx); - else - ret = vop_virtio_copy_from_user(vdev, ubuf, partlen, - daddr, - kiov->iov_len, - vr_idx); - if (ret) { - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, ret); - break; - } - len -= partlen; - ubuf += partlen; - tot_len += partlen; - iov->consumed += partlen; - kiov->iov_len -= partlen; - kiov->iov_base += partlen; - if (!kiov->iov_len) { - /* Fix up old iov element then increment. */ - kiov->iov_len = iov->consumed; - kiov->iov_base -= iov->consumed; - - iov->consumed = 0; - iov->i++; - } - } - *out_len = tot_len; - return ret; -} - -/* - * Use the standard VRINGH infrastructure in the kernel to fetch new - * descriptors, initiate the copies and update the used ring. - */ -static int _vop_virtio_copy(struct vop_vdev *vdev, struct mic_copy_desc *copy) -{ - int ret = 0; - u32 iovcnt = copy->iovcnt; - struct iovec iov; - struct iovec __user *u_iov = copy->iov; - void __user *ubuf = NULL; - struct vop_vringh *vvr = &vdev->vvr[copy->vr_idx]; - struct vringh_kiov *riov = &vvr->riov; - struct vringh_kiov *wiov = &vvr->wiov; - struct vringh *vrh = &vvr->vrh; - u16 *head = &vvr->head; - struct mic_vring *vr = &vvr->vring; - size_t len = 0, out_len; - - copy->out_len = 0; - /* Fetch a new IOVEC if all previous elements have been processed */ - if (riov->i == riov->used && wiov->i == wiov->used) { - ret = vringh_getdesc_kern(vrh, riov, wiov, - head, GFP_KERNEL); - /* Check if there are available descriptors */ - if (ret <= 0) - return ret; - } - while (iovcnt) { - if (!len) { - /* Copy over a new iovec from user space. */ - ret = copy_from_user(&iov, u_iov, sizeof(*u_iov)); - if (ret) { - ret = -EINVAL; - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, ret); - break; - } - len = iov.iov_len; - ubuf = iov.iov_base; - } - /* Issue all the read descriptors first */ - ret = vop_vringh_copy(vdev, riov, ubuf, len, - MIC_VRINGH_READ, copy->vr_idx, &out_len); - if (ret) { - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, ret); - break; - } - len -= out_len; - ubuf += out_len; - copy->out_len += out_len; - /* Issue the write descriptors next */ - ret = vop_vringh_copy(vdev, wiov, ubuf, len, - !MIC_VRINGH_READ, copy->vr_idx, &out_len); - if (ret) { - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, ret); - break; - } - len -= out_len; - ubuf += out_len; - copy->out_len += out_len; - if (!len) { - /* One user space iovec is now completed */ - iovcnt--; - u_iov++; - } - /* Exit loop if all elements in KIOVs have been processed. */ - if (riov->i == riov->used && wiov->i == wiov->used) - break; - } - /* - * Update the used ring if a descriptor was available and some data was - * copied in/out and the user asked for a used ring update. - */ - if (*head != USHRT_MAX && copy->out_len && copy->update_used) { - u32 total = 0; - - /* Determine the total data consumed */ - total += vop_vringh_iov_consumed(riov); - total += vop_vringh_iov_consumed(wiov); - vringh_complete_kern(vrh, *head, total); - *head = USHRT_MAX; - if (vringh_need_notify_kern(vrh) > 0) - vringh_notify(vrh); - vringh_kiov_cleanup(riov); - vringh_kiov_cleanup(wiov); - /* Update avail idx for user space */ - vr->info->avail_idx = vrh->last_avail_idx; - } - return ret; -} - -static inline int vop_verify_copy_args(struct vop_vdev *vdev, - struct mic_copy_desc *copy) -{ - if (!vdev || copy->vr_idx >= vdev->dd->num_vq) - return -EINVAL; - return 0; -} - -/* Copy a specified number of virtio descriptors in a chain */ -static int vop_virtio_copy_desc(struct vop_vdev *vdev, - struct mic_copy_desc *copy) -{ - int err; - struct vop_vringh *vvr; - - err = vop_verify_copy_args(vdev, copy); - if (err) - return err; - - vvr = &vdev->vvr[copy->vr_idx]; - mutex_lock(&vvr->vr_mutex); - if (!vop_vdevup(vdev)) { - err = -ENODEV; - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, err); - goto err; - } - err = _vop_virtio_copy(vdev, copy); - if (err) { - dev_err(vop_dev(vdev), "%s %d err %d\n", - __func__, __LINE__, err); - } -err: - mutex_unlock(&vvr->vr_mutex); - return err; -} - -static int vop_open(struct inode *inode, struct file *f) -{ - struct vop_vdev *vdev; - struct vop_info *vi = container_of(f->private_data, - struct vop_info, miscdev); - - vdev = kzalloc(sizeof(*vdev), GFP_KERNEL); - if (!vdev) - return -ENOMEM; - vdev->vi = vi; - mutex_init(&vdev->vdev_mutex); - f->private_data = vdev; - init_completion(&vdev->destroy); - complete(&vdev->destroy); - return 0; -} - -static int vop_release(struct inode *inode, struct file *f) -{ - struct vop_vdev *vdev = f->private_data, *vdev_tmp; - struct vop_info *vi = vdev->vi; - struct list_head *pos, *tmp; - bool found = false; - - mutex_lock(&vdev->vdev_mutex); - if (vdev->deleted) - goto unlock; - mutex_lock(&vi->vop_mutex); - list_for_each_safe(pos, tmp, &vi->vdev_list) { - vdev_tmp = list_entry(pos, struct vop_vdev, list); - if (vdev == vdev_tmp) { - vop_virtio_del_device(vdev); - list_del(pos); - found = true; - break; - } - } - mutex_unlock(&vi->vop_mutex); -unlock: - mutex_unlock(&vdev->vdev_mutex); - if (!found) - wait_for_completion(&vdev->destroy); - f->private_data = NULL; - kfree(vdev); - return 0; -} - -static long vop_ioctl(struct file *f, unsigned int cmd, unsigned long arg) -{ - struct vop_vdev *vdev = f->private_data; - struct vop_info *vi = vdev->vi; - void __user *argp = (void __user *)arg; - int ret; - - switch (cmd) { - case MIC_VIRTIO_ADD_DEVICE: - { - struct mic_device_desc dd, *dd_config; - - if (copy_from_user(&dd, argp, sizeof(dd))) - return -EFAULT; - - if (mic_aligned_desc_size(&dd) > MIC_MAX_DESC_BLK_SIZE || - dd.num_vq > MIC_MAX_VRINGS) - return -EINVAL; - - dd_config = memdup_user(argp, mic_desc_size(&dd)); - if (IS_ERR(dd_config)) - return PTR_ERR(dd_config); - - /* Ensure desc has not changed between the two reads */ - if (memcmp(&dd, dd_config, sizeof(dd))) { - ret = -EINVAL; - goto free_ret; - } - mutex_lock(&vdev->vdev_mutex); - mutex_lock(&vi->vop_mutex); - ret = vop_virtio_add_device(vdev, dd_config); - if (ret) - goto unlock_ret; - list_add_tail(&vdev->list, &vi->vdev_list); -unlock_ret: - mutex_unlock(&vi->vop_mutex); - mutex_unlock(&vdev->vdev_mutex); -free_ret: - kfree(dd_config); - return ret; - } - case MIC_VIRTIO_COPY_DESC: - { - struct mic_copy_desc copy; - - mutex_lock(&vdev->vdev_mutex); - ret = vop_vdev_inited(vdev); - if (ret) - goto _unlock_ret; - - if (copy_from_user(©, argp, sizeof(copy))) { - ret = -EFAULT; - goto _unlock_ret; - } - - ret = vop_virtio_copy_desc(vdev, ©); - if (ret < 0) - goto _unlock_ret; - if (copy_to_user( - &((struct mic_copy_desc __user *)argp)->out_len, - ©.out_len, sizeof(copy.out_len))) - ret = -EFAULT; -_unlock_ret: - mutex_unlock(&vdev->vdev_mutex); - return ret; - } - case MIC_VIRTIO_CONFIG_CHANGE: - { - void *buf; - - mutex_lock(&vdev->vdev_mutex); - ret = vop_vdev_inited(vdev); - if (ret) - goto __unlock_ret; - buf = memdup_user(argp, vdev->dd->config_len); - if (IS_ERR(buf)) { - ret = PTR_ERR(buf); - goto __unlock_ret; - } - ret = vop_virtio_config_change(vdev, buf); - kfree(buf); -__unlock_ret: - mutex_unlock(&vdev->vdev_mutex); - return ret; - } - default: - return -ENOIOCTLCMD; - }; - return 0; -} - -/* - * We return EPOLLIN | EPOLLOUT from poll when new buffers are enqueued, and - * not when previously enqueued buffers may be available. This means that - * in the card->host (TX) path, when userspace is unblocked by poll it - * must drain all available descriptors or it can stall. - */ -static __poll_t vop_poll(struct file *f, poll_table *wait) -{ - struct vop_vdev *vdev = f->private_data; - __poll_t mask = 0; - - mutex_lock(&vdev->vdev_mutex); - if (vop_vdev_inited(vdev)) { - mask = EPOLLERR; - goto done; - } - poll_wait(f, &vdev->waitq, wait); - if (vop_vdev_inited(vdev)) { - mask = EPOLLERR; - } else if (vdev->poll_wake) { - vdev->poll_wake = 0; - mask = EPOLLIN | EPOLLOUT; - } -done: - mutex_unlock(&vdev->vdev_mutex); - return mask; -} - -static inline int -vop_query_offset(struct vop_vdev *vdev, unsigned long offset, - unsigned long *size, unsigned long *pa) -{ - struct vop_device *vpdev = vdev->vpdev; - unsigned long start = MIC_DP_SIZE; - int i; - - /* - * MMAP interface is as follows: - * offset region - * 0x0 virtio device_page - * 0x1000 first vring - * 0x1000 + size of 1st vring second vring - * .... - */ - if (!offset) { - *pa = virt_to_phys(vpdev->hw_ops->get_dp(vpdev)); - *size = MIC_DP_SIZE; - return 0; - } - - for (i = 0; i < vdev->dd->num_vq; i++) { - struct vop_vringh *vvr = &vdev->vvr[i]; - - if (offset == start) { - *pa = virt_to_phys(vvr->vring.va); - *size = vvr->vring.len; - return 0; - } - start += vvr->vring.len; - } - return -1; -} - -/* - * Maps the device page and virtio rings to user space for readonly access. - */ -static int vop_mmap(struct file *f, struct vm_area_struct *vma) -{ - struct vop_vdev *vdev = f->private_data; - unsigned long offset = vma->vm_pgoff << PAGE_SHIFT; - unsigned long pa, size = vma->vm_end - vma->vm_start, size_rem = size; - int i, err; - - err = vop_vdev_inited(vdev); - if (err) - goto ret; - if (vma->vm_flags & VM_WRITE) { - err = -EACCES; - goto ret; - } - while (size_rem) { - i = vop_query_offset(vdev, offset, &size, &pa); - if (i < 0) { - err = -EINVAL; - goto ret; - } - err = remap_pfn_range(vma, vma->vm_start + offset, - pa >> PAGE_SHIFT, size, - vma->vm_page_prot); - if (err) - goto ret; - size_rem -= size; - offset += size; - } -ret: - return err; -} - -static const struct file_operations vop_fops = { - .open = vop_open, - .release = vop_release, - .unlocked_ioctl = vop_ioctl, - .poll = vop_poll, - .mmap = vop_mmap, - .owner = THIS_MODULE, -}; - -int vop_host_init(struct vop_info *vi) -{ - int rc; - struct miscdevice *mdev; - struct vop_device *vpdev = vi->vpdev; - - INIT_LIST_HEAD(&vi->vdev_list); - vi->dma_ch = vpdev->dma_ch; - mdev = &vi->miscdev; - mdev->minor = MISC_DYNAMIC_MINOR; - snprintf(vi->name, sizeof(vi->name), "vop_virtio%d", vpdev->index); - mdev->name = vi->name; - mdev->fops = &vop_fops; - mdev->parent = &vpdev->dev; - - rc = misc_register(mdev); - if (rc) - dev_err(&vpdev->dev, "%s failed rc %d\n", __func__, rc); - return rc; -} - -void vop_host_uninit(struct vop_info *vi) -{ - struct list_head *pos, *tmp; - struct vop_vdev *vdev; - - mutex_lock(&vi->vop_mutex); - vop_virtio_reset_devices(vi); - list_for_each_safe(pos, tmp, &vi->vdev_list) { - vdev = list_entry(pos, struct vop_vdev, list); - list_del(pos); - reinit_completion(&vdev->destroy); - mutex_unlock(&vi->vop_mutex); - mutex_lock(&vdev->vdev_mutex); - vop_virtio_del_device(vdev); - vdev->deleted = true; - mutex_unlock(&vdev->vdev_mutex); - complete(&vdev->destroy); - mutex_lock(&vi->vop_mutex); - } - mutex_unlock(&vi->vop_mutex); - misc_deregister(&vi->miscdev); -} diff --git a/include/linux/mic_bus.h b/include/linux/mic_bus.h deleted file mode 100644 index e99c789424e0..000000000000 --- a/include/linux/mic_bus.h +++ /dev/null @@ -1,100 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2014 Intel Corporation. - * - * Intel MIC Bus driver. - * - * This implementation is very similar to the virtio bus driver - * implementation @ include/linux/virtio.h. - */ -#ifndef _MIC_BUS_H_ -#define _MIC_BUS_H_ -/* - * Everything a mbus driver needs to work with any particular mbus - * implementation. - */ -#include -#include - -struct mbus_device_id { - __u32 device; - __u32 vendor; -}; - -#define MBUS_DEV_DMA_HOST 2 -#define MBUS_DEV_DMA_MIC 3 -#define MBUS_DEV_ANY_ID 0xffffffff - -/** - * mbus_device - representation of a device using mbus - * @mmio_va: virtual address of mmio space - * @hw_ops: the hardware ops supported by this device. - * @id: the device type identification (used to match it with a driver). - * @dev: underlying device. - * be used to communicate with. - * @index: unique position on the mbus bus - */ -struct mbus_device { - void __iomem *mmio_va; - struct mbus_hw_ops *hw_ops; - struct mbus_device_id id; - struct device dev; - int index; -}; - -/** - * mbus_driver - operations for a mbus I/O driver - * @driver: underlying device driver (populate name and owner). - * @id_table: the ids serviced by this driver. - * @probe: the function to call when a device is found. Returns 0 or -errno. - * @remove: the function to call when a device is removed. - */ -struct mbus_driver { - struct device_driver driver; - const struct mbus_device_id *id_table; - int (*probe)(struct mbus_device *dev); - void (*scan)(struct mbus_device *dev); - void (*remove)(struct mbus_device *dev); -}; - -/** - * struct mic_irq - opaque pointer used as cookie - */ -struct mic_irq; - -/** - * mbus_hw_ops - Hardware operations for accessing a MIC device on the MIC bus. - */ -struct mbus_hw_ops { - struct mic_irq* (*request_threaded_irq)(struct mbus_device *mbdev, - irq_handler_t handler, - irq_handler_t thread_fn, - const char *name, void *data, - int intr_src); - void (*free_irq)(struct mbus_device *mbdev, - struct mic_irq *cookie, void *data); - void (*ack_interrupt)(struct mbus_device *mbdev, int num); -}; - -struct mbus_device * -mbus_register_device(struct device *pdev, int id, const struct dma_map_ops *dma_ops, - struct mbus_hw_ops *hw_ops, int index, - void __iomem *mmio_va); -void mbus_unregister_device(struct mbus_device *mbdev); - -int mbus_register_driver(struct mbus_driver *drv); -void mbus_unregister_driver(struct mbus_driver *drv); - -static inline struct mbus_device *dev_to_mbus(struct device *_dev) -{ - return container_of(_dev, struct mbus_device, dev); -} - -static inline struct mbus_driver *drv_to_mbus(struct device_driver *drv) -{ - return container_of(drv, struct mbus_driver, driver); -} - -#endif /* _MIC_BUS_H */ diff --git a/include/linux/scif.h b/include/linux/scif.h deleted file mode 100644 index 329e695b8fe5..000000000000 --- a/include/linux/scif.h +++ /dev/null @@ -1,1339 +0,0 @@ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * This file is provided under a dual BSD/GPLv2 license. When using or - * redistributing this file, you may do so under either license. - * - * GPL LICENSE SUMMARY - * - * Copyright(c) 2014 Intel Corporation. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * BSD LICENSE - * - * Copyright(c) 2014 Intel Corporation. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * - * * Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * * Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in - * the documentation and/or other materials provided with the - * distribution. - * * Neither the name of Intel Corporation nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR - * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT - * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, - * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT - * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - * - * Intel SCIF driver. - * - */ -#ifndef __SCIF_H__ -#define __SCIF_H__ - -#include -#include -#include -#include - -#define SCIF_ACCEPT_SYNC 1 -#define SCIF_SEND_BLOCK 1 -#define SCIF_RECV_BLOCK 1 - -enum { - SCIF_PROT_READ = (1 << 0), - SCIF_PROT_WRITE = (1 << 1) -}; - -enum { - SCIF_MAP_FIXED = 0x10, - SCIF_MAP_KERNEL = 0x20, -}; - -enum { - SCIF_FENCE_INIT_SELF = (1 << 0), - SCIF_FENCE_INIT_PEER = (1 << 1), - SCIF_SIGNAL_LOCAL = (1 << 4), - SCIF_SIGNAL_REMOTE = (1 << 5) -}; - -enum { - SCIF_RMA_USECPU = (1 << 0), - SCIF_RMA_USECACHE = (1 << 1), - SCIF_RMA_SYNC = (1 << 2), - SCIF_RMA_ORDERED = (1 << 3) -}; - -/* End of SCIF Admin Reserved Ports */ -#define SCIF_ADMIN_PORT_END 1024 - -/* End of SCIF Reserved Ports */ -#define SCIF_PORT_RSVD 1088 - -typedef struct scif_endpt *scif_epd_t; -typedef struct scif_pinned_pages *scif_pinned_pages_t; - -/** - * struct scif_range - SCIF registered range used in kernel mode - * @cookie: cookie used internally by SCIF - * @nr_pages: number of pages of PAGE_SIZE - * @prot_flags: R/W protection - * @phys_addr: Array of bus addresses - * @va: Array of kernel virtual addresses backed by the pages in the phys_addr - * array. The va is populated only when called on the host for a remote - * SCIF connection on MIC. This is required to support the use case of DMA - * between MIC and another device which is not a SCIF node e.g., an IB or - * ethernet NIC. - */ -struct scif_range { - void *cookie; - int nr_pages; - int prot_flags; - dma_addr_t *phys_addr; - void __iomem **va; -}; - -/** - * struct scif_pollepd - SCIF endpoint to be monitored via scif_poll - * @epd: SCIF endpoint - * @events: requested events - * @revents: returned events - */ -struct scif_pollepd { - scif_epd_t epd; - __poll_t events; - __poll_t revents; -}; - -/** - * scif_peer_dev - representation of a peer SCIF device - * - * Peer devices show up as PCIe devices for the mgmt node but not the cards. - * The mgmt node discovers all the cards on the PCIe bus and informs the other - * cards about their peers. Upon notification of a peer a node adds a peer - * device to the peer bus to maintain symmetry in the way devices are - * discovered across all nodes in the SCIF network. - * - * @dev: underlying device - * @dnode - The destination node which this device will communicate with. - */ -struct scif_peer_dev { - struct device dev; - u8 dnode; -}; - -/** - * scif_client - representation of a SCIF client - * @name: client name - * @probe - client method called when a peer device is registered - * @remove - client method called when a peer device is unregistered - * @si - subsys_interface used internally for implementing SCIF clients - */ -struct scif_client { - const char *name; - void (*probe)(struct scif_peer_dev *spdev); - void (*remove)(struct scif_peer_dev *spdev); - struct subsys_interface si; -}; - -#define SCIF_OPEN_FAILED ((scif_epd_t)-1) -#define SCIF_REGISTER_FAILED ((off_t)-1) -#define SCIF_MMAP_FAILED ((void *)-1) - -/** - * scif_open() - Create an endpoint - * - * Return: - * Upon successful completion, scif_open() returns an endpoint descriptor to - * be used in subsequent SCIF functions calls to refer to that endpoint; - * otherwise in user mode SCIF_OPEN_FAILED (that is ((scif_epd_t)-1)) is - * returned and errno is set to indicate the error; in kernel mode a NULL - * scif_epd_t is returned. - * - * Errors: - * ENOMEM - Insufficient kernel memory was available - */ -scif_epd_t scif_open(void); - -/** - * scif_bind() - Bind an endpoint to a port - * @epd: endpoint descriptor - * @pn: port number - * - * scif_bind() binds endpoint epd to port pn, where pn is a port number on the - * local node. If pn is zero, a port number greater than or equal to - * SCIF_PORT_RSVD is assigned and returned. Each endpoint may be bound to - * exactly one local port. Ports less than 1024 when requested can only be bound - * by system (or root) processes or by processes executed by privileged users. - * - * Return: - * Upon successful completion, scif_bind() returns the port number to which epd - * is bound; otherwise in user mode -1 is returned and errno is set to - * indicate the error; in kernel mode the negative of one of the following - * errors is returned. - * - * Errors: - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * EINVAL - the endpoint or the port is already bound - * EISCONN - The endpoint is already connected - * ENOSPC - No port number available for assignment - * EACCES - The port requested is protected and the user is not the superuser - */ -int scif_bind(scif_epd_t epd, u16 pn); - -/** - * scif_listen() - Listen for connections on an endpoint - * @epd: endpoint descriptor - * @backlog: maximum pending connection requests - * - * scif_listen() marks the endpoint epd as a listening endpoint - that is, as - * an endpoint that will be used to accept incoming connection requests. Once - * so marked, the endpoint is said to be in the listening state and may not be - * used as the endpoint of a connection. - * - * The endpoint, epd, must have been bound to a port. - * - * The backlog argument defines the maximum length to which the queue of - * pending connections for epd may grow. If a connection request arrives when - * the queue is full, the client may receive an error with an indication that - * the connection was refused. - * - * Return: - * Upon successful completion, scif_listen() returns 0; otherwise in user mode - * -1 is returned and errno is set to indicate the error; in kernel mode the - * negative of one of the following errors is returned. - * - * Errors: - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * EINVAL - the endpoint is not bound to a port - * EISCONN - The endpoint is already connected or listening - */ -int scif_listen(scif_epd_t epd, int backlog); - -/** - * scif_connect() - Initiate a connection on a port - * @epd: endpoint descriptor - * @dst: global id of port to which to connect - * - * The scif_connect() function requests the connection of endpoint epd to remote - * port dst. If the connection is successful, a peer endpoint, bound to dst, is - * created on node dst.node. On successful return, the connection is complete. - * - * If the endpoint epd has not already been bound to a port, scif_connect() - * will bind it to an unused local port. - * - * A connection is terminated when an endpoint of the connection is closed, - * either explicitly by scif_close(), or when a process that owns one of the - * endpoints of the connection is terminated. - * - * In user space, scif_connect() supports an asynchronous connection mode - * if the application has set the O_NONBLOCK flag on the endpoint via the - * fcntl() system call. Setting this flag will result in the calling process - * not to wait during scif_connect(). - * - * Return: - * Upon successful completion, scif_connect() returns the port ID to which the - * endpoint, epd, is bound; otherwise in user mode -1 is returned and errno is - * set to indicate the error; in kernel mode the negative of one of the - * following errors is returned. - * - * Errors: - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * ECONNREFUSED - The destination was not listening for connections or refused - * the connection request - * EINVAL - dst.port is not a valid port ID - * EISCONN - The endpoint is already connected - * ENOMEM - No buffer space is available - * ENODEV - The destination node does not exist, or the node is lost or existed, - * but is not currently in the network since it may have crashed - * ENOSPC - No port number available for assignment - * EOPNOTSUPP - The endpoint is listening and cannot be connected - */ -int scif_connect(scif_epd_t epd, struct scif_port_id *dst); - -/** - * scif_accept() - Accept a connection on an endpoint - * @epd: endpoint descriptor - * @peer: global id of port to which connected - * @newepd: new connected endpoint descriptor - * @flags: flags - * - * The scif_accept() call extracts the first connection request from the queue - * of pending connections for the port on which epd is listening. scif_accept() - * creates a new endpoint, bound to the same port as epd, and allocates a new - * SCIF endpoint descriptor, returned in newepd, for the endpoint. The new - * endpoint is connected to the endpoint through which the connection was - * requested. epd is unaffected by this call, and remains in the listening - * state. - * - * On successful return, peer holds the global port identifier (node id and - * local port number) of the port which requested the connection. - * - * A connection is terminated when an endpoint of the connection is closed, - * either explicitly by scif_close(), or when a process that owns one of the - * endpoints of the connection is terminated. - * - * The number of connections that can (subsequently) be accepted on epd is only - * limited by system resources (memory). - * - * The flags argument is formed by OR'ing together zero or more of the - * following values. - * SCIF_ACCEPT_SYNC - block until a connection request is presented. If - * SCIF_ACCEPT_SYNC is not in flags, and no pending - * connections are present on the queue, scif_accept() - * fails with an EAGAIN error - * - * In user mode, the select() and poll() functions can be used to determine - * when there is a connection request. In kernel mode, the scif_poll() - * function may be used for this purpose. A readable event will be delivered - * when a connection is requested. - * - * Return: - * Upon successful completion, scif_accept() returns 0; otherwise in user mode - * -1 is returned and errno is set to indicate the error; in kernel mode the - * negative of one of the following errors is returned. - * - * Errors: - * EAGAIN - SCIF_ACCEPT_SYNC is not set and no connections are present to be - * accepted or SCIF_ACCEPT_SYNC is not set and remote node failed to complete - * its connection request - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * EINTR - Interrupted function - * EINVAL - epd is not a listening endpoint, or flags is invalid, or peer is - * NULL, or newepd is NULL - * ENODEV - The requesting node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOMEM - Not enough space - * ENOENT - Secondary part of epd registration failed - */ -int scif_accept(scif_epd_t epd, struct scif_port_id *peer, scif_epd_t - *newepd, int flags); - -/** - * scif_close() - Close an endpoint - * @epd: endpoint descriptor - * - * scif_close() closes an endpoint and performs necessary teardown of - * facilities associated with that endpoint. - * - * If epd is a listening endpoint then it will no longer accept connection - * requests on the port to which it is bound. Any pending connection requests - * are rejected. - * - * If epd is a connected endpoint, then its peer endpoint is also closed. RMAs - * which are in-process through epd or its peer endpoint will complete before - * scif_close() returns. Registered windows of the local and peer endpoints are - * released as if scif_unregister() was called against each window. - * - * Closing a SCIF endpoint does not affect local registered memory mapped by - * a SCIF endpoint on a remote node. The local memory remains mapped by the peer - * SCIF endpoint explicitly removed by calling munmap(..) by the peer. - * - * If the peer endpoint's receive queue is not empty at the time that epd is - * closed, then the peer endpoint can be passed as the endpoint parameter to - * scif_recv() until the receive queue is empty. - * - * epd is freed and may no longer be accessed. - * - * Return: - * Upon successful completion, scif_close() returns 0; otherwise in user mode - * -1 is returned and errno is set to indicate the error; in kernel mode the - * negative of one of the following errors is returned. - * - * Errors: - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - */ -int scif_close(scif_epd_t epd); - -/** - * scif_send() - Send a message - * @epd: endpoint descriptor - * @msg: message buffer address - * @len: message length - * @flags: blocking mode flags - * - * scif_send() sends data to the peer of endpoint epd. Up to len bytes of data - * are copied from memory starting at address msg. On successful execution the - * return value of scif_send() is the number of bytes that were sent, and is - * zero if no bytes were sent because len was zero. scif_send() may be called - * only when the endpoint is in a connected state. - * - * If a scif_send() call is non-blocking, then it sends only those bytes which - * can be sent without waiting, up to a maximum of len bytes. - * - * If a scif_send() call is blocking, then it normally returns after sending - * all len bytes. If a blocking call is interrupted or the connection is - * reset, the call is considered successful if some bytes were sent or len is - * zero, otherwise the call is considered unsuccessful. - * - * In user mode, the select() and poll() functions can be used to determine - * when the send queue is not full. In kernel mode, the scif_poll() function - * may be used for this purpose. - * - * It is recommended that scif_send()/scif_recv() only be used for short - * control-type message communication between SCIF endpoints. The SCIF RMA - * APIs are expected to provide better performance for transfer sizes of - * 1024 bytes or longer for the current MIC hardware and software - * implementation. - * - * scif_send() will block until the entire message is sent if SCIF_SEND_BLOCK - * is passed as the flags argument. - * - * Return: - * Upon successful completion, scif_send() returns the number of bytes sent; - * otherwise in user mode -1 is returned and errno is set to indicate the - * error; in kernel mode the negative of one of the following errors is - * returned. - * - * Errors: - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * ECONNRESET - Connection reset by peer - * EINVAL - flags is invalid, or len is negative - * ENODEV - The remote node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOMEM - Not enough space - * ENOTCONN - The endpoint is not connected - */ -int scif_send(scif_epd_t epd, void *msg, int len, int flags); - -/** - * scif_recv() - Receive a message - * @epd: endpoint descriptor - * @msg: message buffer address - * @len: message buffer length - * @flags: blocking mode flags - * - * scif_recv() receives data from the peer of endpoint epd. Up to len bytes of - * data are copied to memory starting at address msg. On successful execution - * the return value of scif_recv() is the number of bytes that were received, - * and is zero if no bytes were received because len was zero. scif_recv() may - * be called only when the endpoint is in a connected state. - * - * If a scif_recv() call is non-blocking, then it receives only those bytes - * which can be received without waiting, up to a maximum of len bytes. - * - * If a scif_recv() call is blocking, then it normally returns after receiving - * all len bytes. If the blocking call was interrupted due to a disconnection, - * subsequent calls to scif_recv() will copy all bytes received upto the point - * of disconnection. - * - * In user mode, the select() and poll() functions can be used to determine - * when data is available to be received. In kernel mode, the scif_poll() - * function may be used for this purpose. - * - * It is recommended that scif_send()/scif_recv() only be used for short - * control-type message communication between SCIF endpoints. The SCIF RMA - * APIs are expected to provide better performance for transfer sizes of - * 1024 bytes or longer for the current MIC hardware and software - * implementation. - * - * scif_recv() will block until the entire message is received if - * SCIF_RECV_BLOCK is passed as the flags argument. - * - * Return: - * Upon successful completion, scif_recv() returns the number of bytes - * received; otherwise in user mode -1 is returned and errno is set to - * indicate the error; in kernel mode the negative of one of the following - * errors is returned. - * - * Errors: - * EAGAIN - The destination node is returning from a low power state - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * ECONNRESET - Connection reset by peer - * EINVAL - flags is invalid, or len is negative - * ENODEV - The remote node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOMEM - Not enough space - * ENOTCONN - The endpoint is not connected - */ -int scif_recv(scif_epd_t epd, void *msg, int len, int flags); - -/** - * scif_register() - Mark a memory region for remote access. - * @epd: endpoint descriptor - * @addr: starting virtual address - * @len: length of range - * @offset: offset of window - * @prot_flags: read/write protection flags - * @map_flags: mapping flags - * - * The scif_register() function opens a window, a range of whole pages of the - * registered address space of the endpoint epd, starting at offset po and - * continuing for len bytes. The value of po, further described below, is a - * function of the parameters offset and len, and the value of map_flags. Each - * page of the window represents the physical memory page which backs the - * corresponding page of the range of virtual address pages starting at addr - * and continuing for len bytes. addr and len are constrained to be multiples - * of the page size. A successful scif_register() call returns po. - * - * When SCIF_MAP_FIXED is set in the map_flags argument, po will be offset - * exactly, and offset is constrained to be a multiple of the page size. The - * mapping established by scif_register() will not replace any existing - * registration; an error is returned if any page within the range [offset, - * offset + len - 1] intersects an existing window. - * - * When SCIF_MAP_FIXED is not set, the implementation uses offset in an - * implementation-defined manner to arrive at po. The po value so chosen will - * be an area of the registered address space that the implementation deems - * suitable for a mapping of len bytes. An offset value of 0 is interpreted as - * granting the implementation complete freedom in selecting po, subject to - * constraints described below. A non-zero value of offset is taken to be a - * suggestion of an offset near which the mapping should be placed. When the - * implementation selects a value for po, it does not replace any extant - * window. In all cases, po will be a multiple of the page size. - * - * The physical pages which are so represented by a window are available for - * access in calls to mmap(), scif_readfrom(), scif_writeto(), - * scif_vreadfrom(), and scif_vwriteto(). While a window is registered, the - * physical pages represented by the window will not be reused by the memory - * subsystem for any other purpose. Note that the same physical page may be - * represented by multiple windows. - * - * Subsequent operations which change the memory pages to which virtual - * addresses are mapped (such as mmap(), munmap()) have no effect on - * existing window. - * - * If the process will fork(), it is recommended that the registered - * virtual address range be marked with MADV_DONTFORK. Doing so will prevent - * problems due to copy-on-write semantics. - * - * The prot_flags argument is formed by OR'ing together one or more of the - * following values. - * SCIF_PROT_READ - allow read operations from the window - * SCIF_PROT_WRITE - allow write operations to the window - * - * Return: - * Upon successful completion, scif_register() returns the offset at which the - * mapping was placed (po); otherwise in user mode SCIF_REGISTER_FAILED (that - * is (off_t *)-1) is returned and errno is set to indicate the error; in - * kernel mode the negative of one of the following errors is returned. - * - * Errors: - * EADDRINUSE - SCIF_MAP_FIXED is set in map_flags, and pages in the range - * [offset, offset + len -1] are already registered - * EAGAIN - The mapping could not be performed due to lack of resources - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * ECONNRESET - Connection reset by peer - * EINVAL - map_flags is invalid, or prot_flags is invalid, or SCIF_MAP_FIXED is - * set in flags, and offset is not a multiple of the page size, or addr is not a - * multiple of the page size, or len is not a multiple of the page size, or is - * 0, or offset is negative - * ENODEV - The remote node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOMEM - Not enough space - * ENOTCONN -The endpoint is not connected - */ -off_t scif_register(scif_epd_t epd, void *addr, size_t len, off_t offset, - int prot_flags, int map_flags); - -/** - * scif_unregister() - Mark a memory region for remote access. - * @epd: endpoint descriptor - * @offset: start of range to unregister - * @len: length of range to unregister - * - * The scif_unregister() function closes those previously registered windows - * which are entirely within the range [offset, offset + len - 1]. It is an - * error to specify a range which intersects only a subrange of a window. - * - * On a successful return, pages within the window may no longer be specified - * in calls to mmap(), scif_readfrom(), scif_writeto(), scif_vreadfrom(), - * scif_vwriteto(), scif_get_pages, and scif_fence_signal(). The window, - * however, continues to exist until all previous references against it are - * removed. A window is referenced if there is a mapping to it created by - * mmap(), or if scif_get_pages() was called against the window - * (and the pages have not been returned via scif_put_pages()). A window is - * also referenced while an RMA, in which some range of the window is a source - * or destination, is in progress. Finally a window is referenced while some - * offset in that window was specified to scif_fence_signal(), and the RMAs - * marked by that call to scif_fence_signal() have not completed. While a - * window is in this state, its registered address space pages are not - * available for use in a new registered window. - * - * When all such references to the window have been removed, its references to - * all the physical pages which it represents are removed. Similarly, the - * registered address space pages of the window become available for - * registration in a new window. - * - * Return: - * Upon successful completion, scif_unregister() returns 0; otherwise in user - * mode -1 is returned and errno is set to indicate the error; in kernel mode - * the negative of one of the following errors is returned. In the event of an - * error, no windows are unregistered. - * - * Errors: - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * ECONNRESET - Connection reset by peer - * EINVAL - the range [offset, offset + len - 1] intersects a subrange of a - * window, or offset is negative - * ENODEV - The remote node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOTCONN - The endpoint is not connected - * ENXIO - Offsets in the range [offset, offset + len - 1] are invalid for the - * registered address space of epd - */ -int scif_unregister(scif_epd_t epd, off_t offset, size_t len); - -/** - * scif_readfrom() - Copy from a remote address space - * @epd: endpoint descriptor - * @loffset: offset in local registered address space to - * which to copy - * @len: length of range to copy - * @roffset: offset in remote registered address space - * from which to copy - * @rma_flags: transfer mode flags - * - * scif_readfrom() copies len bytes from the remote registered address space of - * the peer of endpoint epd, starting at the offset roffset to the local - * registered address space of epd, starting at the offset loffset. - * - * Each of the specified ranges [loffset, loffset + len - 1] and [roffset, - * roffset + len - 1] must be within some registered window or windows of the - * local and remote nodes. A range may intersect multiple registered windows, - * but only if those windows are contiguous in the registered address space. - * - * If rma_flags includes SCIF_RMA_USECPU, then the data is copied using - * programmed read/writes. Otherwise the data is copied using DMA. If rma_- - * flags includes SCIF_RMA_SYNC, then scif_readfrom() will return after the - * transfer is complete. Otherwise, the transfer may be performed asynchron- - * ously. The order in which any two asynchronous RMA operations complete - * is non-deterministic. The synchronization functions, scif_fence_mark()/ - * scif_fence_wait() and scif_fence_signal(), can be used to synchronize to - * the completion of asynchronous RMA operations on the same endpoint. - * - * The DMA transfer of individual bytes is not guaranteed to complete in - * address order. If rma_flags includes SCIF_RMA_ORDERED, then the last - * cacheline or partial cacheline of the source range will become visible on - * the destination node after all other transferred data in the source - * range has become visible on the destination node. - * - * The optimal DMA performance will likely be realized if both - * loffset and roffset are cacheline aligned (are a multiple of 64). Lower - * performance will likely be realized if loffset and roffset are not - * cacheline aligned but are separated by some multiple of 64. The lowest level - * of performance is likely if loffset and roffset are not separated by a - * multiple of 64. - * - * The rma_flags argument is formed by ORing together zero or more of the - * following values. - * SCIF_RMA_USECPU - perform the transfer using the CPU, otherwise use the DMA - * engine. - * SCIF_RMA_SYNC - perform the transfer synchronously, returning after the - * transfer has completed. Passing this flag results in the - * current implementation busy waiting and consuming CPU cycles - * while the DMA transfer is in progress for best performance by - * avoiding the interrupt latency. - * SCIF_RMA_ORDERED - ensure that the last cacheline or partial cacheline of - * the source range becomes visible on the destination node - * after all other transferred data in the source range has - * become visible on the destination - * - * Return: - * Upon successful completion, scif_readfrom() returns 0; otherwise in user - * mode -1 is returned and errno is set to indicate the error; in kernel mode - * the negative of one of the following errors is returned. - * - * Errors: - * EACCES - Attempt to write to a read-only range - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * ECONNRESET - Connection reset by peer - * EINVAL - rma_flags is invalid - * ENODEV - The remote node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOTCONN - The endpoint is not connected - * ENXIO - The range [loffset, loffset + len - 1] is invalid for the registered - * address space of epd, or, The range [roffset, roffset + len - 1] is invalid - * for the registered address space of the peer of epd, or loffset or roffset - * is negative - */ -int scif_readfrom(scif_epd_t epd, off_t loffset, size_t len, off_t - roffset, int rma_flags); - -/** - * scif_writeto() - Copy to a remote address space - * @epd: endpoint descriptor - * @loffset: offset in local registered address space - * from which to copy - * @len: length of range to copy - * @roffset: offset in remote registered address space to - * which to copy - * @rma_flags: transfer mode flags - * - * scif_writeto() copies len bytes from the local registered address space of - * epd, starting at the offset loffset to the remote registered address space - * of the peer of endpoint epd, starting at the offset roffset. - * - * Each of the specified ranges [loffset, loffset + len - 1] and [roffset, - * roffset + len - 1] must be within some registered window or windows of the - * local and remote nodes. A range may intersect multiple registered windows, - * but only if those windows are contiguous in the registered address space. - * - * If rma_flags includes SCIF_RMA_USECPU, then the data is copied using - * programmed read/writes. Otherwise the data is copied using DMA. If rma_- - * flags includes SCIF_RMA_SYNC, then scif_writeto() will return after the - * transfer is complete. Otherwise, the transfer may be performed asynchron- - * ously. The order in which any two asynchronous RMA operations complete - * is non-deterministic. The synchronization functions, scif_fence_mark()/ - * scif_fence_wait() and scif_fence_signal(), can be used to synchronize to - * the completion of asynchronous RMA operations on the same endpoint. - * - * The DMA transfer of individual bytes is not guaranteed to complete in - * address order. If rma_flags includes SCIF_RMA_ORDERED, then the last - * cacheline or partial cacheline of the source range will become visible on - * the destination node after all other transferred data in the source - * range has become visible on the destination node. - * - * The optimal DMA performance will likely be realized if both - * loffset and roffset are cacheline aligned (are a multiple of 64). Lower - * performance will likely be realized if loffset and roffset are not cacheline - * aligned but are separated by some multiple of 64. The lowest level of - * performance is likely if loffset and roffset are not separated by a multiple - * of 64. - * - * The rma_flags argument is formed by ORing together zero or more of the - * following values. - * SCIF_RMA_USECPU - perform the transfer using the CPU, otherwise use the DMA - * engine. - * SCIF_RMA_SYNC - perform the transfer synchronously, returning after the - * transfer has completed. Passing this flag results in the - * current implementation busy waiting and consuming CPU cycles - * while the DMA transfer is in progress for best performance by - * avoiding the interrupt latency. - * SCIF_RMA_ORDERED - ensure that the last cacheline or partial cacheline of - * the source range becomes visible on the destination node - * after all other transferred data in the source range has - * become visible on the destination - * - * Return: - * Upon successful completion, scif_readfrom() returns 0; otherwise in user - * mode -1 is returned and errno is set to indicate the error; in kernel mode - * the negative of one of the following errors is returned. - * - * Errors: - * EACCES - Attempt to write to a read-only range - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * ECONNRESET - Connection reset by peer - * EINVAL - rma_flags is invalid - * ENODEV - The remote node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOTCONN - The endpoint is not connected - * ENXIO - The range [loffset, loffset + len - 1] is invalid for the registered - * address space of epd, or, The range [roffset , roffset + len -1] is invalid - * for the registered address space of the peer of epd, or loffset or roffset - * is negative - */ -int scif_writeto(scif_epd_t epd, off_t loffset, size_t len, off_t - roffset, int rma_flags); - -/** - * scif_vreadfrom() - Copy from a remote address space - * @epd: endpoint descriptor - * @addr: address to which to copy - * @len: length of range to copy - * @roffset: offset in remote registered address space - * from which to copy - * @rma_flags: transfer mode flags - * - * scif_vreadfrom() copies len bytes from the remote registered address - * space of the peer of endpoint epd, starting at the offset roffset, to local - * memory, starting at addr. - * - * The specified range [roffset, roffset + len - 1] must be within some - * registered window or windows of the remote nodes. The range may - * intersect multiple registered windows, but only if those windows are - * contiguous in the registered address space. - * - * If rma_flags includes SCIF_RMA_USECPU, then the data is copied using - * programmed read/writes. Otherwise the data is copied using DMA. If rma_- - * flags includes SCIF_RMA_SYNC, then scif_vreadfrom() will return after the - * transfer is complete. Otherwise, the transfer may be performed asynchron- - * ously. The order in which any two asynchronous RMA operations complete - * is non-deterministic. The synchronization functions, scif_fence_mark()/ - * scif_fence_wait() and scif_fence_signal(), can be used to synchronize to - * the completion of asynchronous RMA operations on the same endpoint. - * - * The DMA transfer of individual bytes is not guaranteed to complete in - * address order. If rma_flags includes SCIF_RMA_ORDERED, then the last - * cacheline or partial cacheline of the source range will become visible on - * the destination node after all other transferred data in the source - * range has become visible on the destination node. - * - * If rma_flags includes SCIF_RMA_USECACHE, then the physical pages which back - * the specified local memory range may be remain in a pinned state even after - * the specified transfer completes. This may reduce overhead if some or all of - * the same virtual address range is referenced in a subsequent call of - * scif_vreadfrom() or scif_vwriteto(). - * - * The optimal DMA performance will likely be realized if both - * addr and roffset are cacheline aligned (are a multiple of 64). Lower - * performance will likely be realized if addr and roffset are not - * cacheline aligned but are separated by some multiple of 64. The lowest level - * of performance is likely if addr and roffset are not separated by a - * multiple of 64. - * - * The rma_flags argument is formed by ORing together zero or more of the - * following values. - * SCIF_RMA_USECPU - perform the transfer using the CPU, otherwise use the DMA - * engine. - * SCIF_RMA_USECACHE - enable registration caching - * SCIF_RMA_SYNC - perform the transfer synchronously, returning after the - * transfer has completed. Passing this flag results in the - * current implementation busy waiting and consuming CPU cycles - * while the DMA transfer is in progress for best performance by - * avoiding the interrupt latency. - * SCIF_RMA_ORDERED - ensure that the last cacheline or partial cacheline of - * the source range becomes visible on the destination node - * after all other transferred data in the source range has - * become visible on the destination - * - * Return: - * Upon successful completion, scif_vreadfrom() returns 0; otherwise in user - * mode -1 is returned and errno is set to indicate the error; in kernel mode - * the negative of one of the following errors is returned. - * - * Errors: - * EACCES - Attempt to write to a read-only range - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * ECONNRESET - Connection reset by peer - * EINVAL - rma_flags is invalid - * ENODEV - The remote node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOTCONN - The endpoint is not connected - * ENXIO - Offsets in the range [roffset, roffset + len - 1] are invalid for the - * registered address space of epd - */ -int scif_vreadfrom(scif_epd_t epd, void *addr, size_t len, off_t roffset, - int rma_flags); - -/** - * scif_vwriteto() - Copy to a remote address space - * @epd: endpoint descriptor - * @addr: address from which to copy - * @len: length of range to copy - * @roffset: offset in remote registered address space to - * which to copy - * @rma_flags: transfer mode flags - * - * scif_vwriteto() copies len bytes from the local memory, starting at addr, to - * the remote registered address space of the peer of endpoint epd, starting at - * the offset roffset. - * - * The specified range [roffset, roffset + len - 1] must be within some - * registered window or windows of the remote nodes. The range may intersect - * multiple registered windows, but only if those windows are contiguous in the - * registered address space. - * - * If rma_flags includes SCIF_RMA_USECPU, then the data is copied using - * programmed read/writes. Otherwise the data is copied using DMA. If rma_- - * flags includes SCIF_RMA_SYNC, then scif_vwriteto() will return after the - * transfer is complete. Otherwise, the transfer may be performed asynchron- - * ously. The order in which any two asynchronous RMA operations complete - * is non-deterministic. The synchronization functions, scif_fence_mark()/ - * scif_fence_wait() and scif_fence_signal(), can be used to synchronize to - * the completion of asynchronous RMA operations on the same endpoint. - * - * The DMA transfer of individual bytes is not guaranteed to complete in - * address order. If rma_flags includes SCIF_RMA_ORDERED, then the last - * cacheline or partial cacheline of the source range will become visible on - * the destination node after all other transferred data in the source - * range has become visible on the destination node. - * - * If rma_flags includes SCIF_RMA_USECACHE, then the physical pages which back - * the specified local memory range may be remain in a pinned state even after - * the specified transfer completes. This may reduce overhead if some or all of - * the same virtual address range is referenced in a subsequent call of - * scif_vreadfrom() or scif_vwriteto(). - * - * The optimal DMA performance will likely be realized if both - * addr and offset are cacheline aligned (are a multiple of 64). Lower - * performance will likely be realized if addr and offset are not cacheline - * aligned but are separated by some multiple of 64. The lowest level of - * performance is likely if addr and offset are not separated by a multiple of - * 64. - * - * The rma_flags argument is formed by ORing together zero or more of the - * following values. - * SCIF_RMA_USECPU - perform the transfer using the CPU, otherwise use the DMA - * engine. - * SCIF_RMA_USECACHE - allow registration caching - * SCIF_RMA_SYNC - perform the transfer synchronously, returning after the - * transfer has completed. Passing this flag results in the - * current implementation busy waiting and consuming CPU cycles - * while the DMA transfer is in progress for best performance by - * avoiding the interrupt latency. - * SCIF_RMA_ORDERED - ensure that the last cacheline or partial cacheline of - * the source range becomes visible on the destination node - * after all other transferred data in the source range has - * become visible on the destination - * - * Return: - * Upon successful completion, scif_vwriteto() returns 0; otherwise in user - * mode -1 is returned and errno is set to indicate the error; in kernel mode - * the negative of one of the following errors is returned. - * - * Errors: - * EACCES - Attempt to write to a read-only range - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * ECONNRESET - Connection reset by peer - * EINVAL - rma_flags is invalid - * ENODEV - The remote node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOTCONN - The endpoint is not connected - * ENXIO - Offsets in the range [roffset, roffset + len - 1] are invalid for the - * registered address space of epd - */ -int scif_vwriteto(scif_epd_t epd, void *addr, size_t len, off_t roffset, - int rma_flags); - -/** - * scif_fence_mark() - Mark previously issued RMAs - * @epd: endpoint descriptor - * @flags: control flags - * @mark: marked value returned as output. - * - * scif_fence_mark() returns after marking the current set of all uncompleted - * RMAs initiated through the endpoint epd or the current set of all - * uncompleted RMAs initiated through the peer of endpoint epd. The RMAs are - * marked with a value returned at mark. The application may subsequently call - * scif_fence_wait(), passing the value returned at mark, to await completion - * of all RMAs so marked. - * - * The flags argument has exactly one of the following values. - * SCIF_FENCE_INIT_SELF - RMA operations initiated through endpoint - * epd are marked - * SCIF_FENCE_INIT_PEER - RMA operations initiated through the peer - * of endpoint epd are marked - * - * Return: - * Upon successful completion, scif_fence_mark() returns 0; otherwise in user - * mode -1 is returned and errno is set to indicate the error; in kernel mode - * the negative of one of the following errors is returned. - * - * Errors: - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * ECONNRESET - Connection reset by peer - * EINVAL - flags is invalid - * ENODEV - The remote node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOTCONN - The endpoint is not connected - * ENOMEM - Insufficient kernel memory was available - */ -int scif_fence_mark(scif_epd_t epd, int flags, int *mark); - -/** - * scif_fence_wait() - Wait for completion of marked RMAs - * @epd: endpoint descriptor - * @mark: mark request - * - * scif_fence_wait() returns after all RMAs marked with mark have completed. - * The value passed in mark must have been obtained in a previous call to - * scif_fence_mark(). - * - * Return: - * Upon successful completion, scif_fence_wait() returns 0; otherwise in user - * mode -1 is returned and errno is set to indicate the error; in kernel mode - * the negative of one of the following errors is returned. - * - * Errors: - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * ECONNRESET - Connection reset by peer - * ENODEV - The remote node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOTCONN - The endpoint is not connected - * ENOMEM - Insufficient kernel memory was available - */ -int scif_fence_wait(scif_epd_t epd, int mark); - -/** - * scif_fence_signal() - Request a memory update on completion of RMAs - * @epd: endpoint descriptor - * @loff: local offset - * @lval: local value to write to loffset - * @roff: remote offset - * @rval: remote value to write to roffset - * @flags: flags - * - * scif_fence_signal() returns after marking the current set of all uncompleted - * RMAs initiated through the endpoint epd or marking the current set of all - * uncompleted RMAs initiated through the peer of endpoint epd. - * - * If flags includes SCIF_SIGNAL_LOCAL, then on completion of the RMAs in the - * marked set, lval is written to memory at the address corresponding to offset - * loff in the local registered address space of epd. loff must be within a - * registered window. If flags includes SCIF_SIGNAL_REMOTE, then on completion - * of the RMAs in the marked set, rval is written to memory at the address - * corresponding to offset roff in the remote registered address space of epd. - * roff must be within a remote registered window of the peer of epd. Note - * that any specified offset must be DWORD (4 byte / 32 bit) aligned. - * - * The flags argument is formed by OR'ing together the following. - * Exactly one of the following values. - * SCIF_FENCE_INIT_SELF - RMA operations initiated through endpoint - * epd are marked - * SCIF_FENCE_INIT_PEER - RMA operations initiated through the peer - * of endpoint epd are marked - * One or more of the following values. - * SCIF_SIGNAL_LOCAL - On completion of the marked set of RMAs, write lval to - * memory at the address corresponding to offset loff in the local - * registered address space of epd. - * SCIF_SIGNAL_REMOTE - On completion of the marked set of RMAs, write rval to - * memory at the address corresponding to offset roff in the remote - * registered address space of epd. - * - * Return: - * Upon successful completion, scif_fence_signal() returns 0; otherwise in - * user mode -1 is returned and errno is set to indicate the error; in kernel - * mode the negative of one of the following errors is returned. - * - * Errors: - * EBADF, ENOTTY - epd is not a valid endpoint descriptor - * ECONNRESET - Connection reset by peer - * EINVAL - flags is invalid, or loff or roff are not DWORD aligned - * ENODEV - The remote node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOTCONN - The endpoint is not connected - * ENXIO - loff is invalid for the registered address of epd, or roff is invalid - * for the registered address space, of the peer of epd - */ -int scif_fence_signal(scif_epd_t epd, off_t loff, u64 lval, off_t roff, - u64 rval, int flags); - -/** - * scif_get_node_ids() - Return information about online nodes - * @nodes: array in which to return online node IDs - * @len: number of entries in the nodes array - * @self: address to place the node ID of the local node - * - * scif_get_node_ids() fills in the nodes array with up to len node IDs of the - * nodes in the SCIF network. If there is not enough space in nodes, as - * indicated by the len parameter, only len node IDs are returned in nodes. The - * return value of scif_get_node_ids() is the total number of nodes currently in - * the SCIF network. By checking the return value against the len parameter, - * the user may determine if enough space for nodes was allocated. - * - * The node ID of the local node is returned at self. - * - * Return: - * Upon successful completion, scif_get_node_ids() returns the actual number of - * online nodes in the SCIF network including 'self'; otherwise in user mode - * -1 is returned and errno is set to indicate the error; in kernel mode no - * errors are returned. - */ -int scif_get_node_ids(u16 *nodes, int len, u16 *self); - -/** - * scif_pin_pages() - Pin a set of pages - * @addr: Virtual address of range to pin - * @len: Length of range to pin - * @prot_flags: Page protection flags - * @map_flags: Page classification flags - * @pinned_pages: Handle to pinned pages - * - * scif_pin_pages() pins (locks in physical memory) the physical pages which - * back the range of virtual address pages starting at addr and continuing for - * len bytes. addr and len are constrained to be multiples of the page size. A - * successful scif_pin_pages() call returns a handle to pinned_pages which may - * be used in subsequent calls to scif_register_pinned_pages(). - * - * The pages will remain pinned as long as there is a reference against the - * scif_pinned_pages_t value returned by scif_pin_pages() and until - * scif_unpin_pages() is called, passing the scif_pinned_pages_t value. A - * reference is added to a scif_pinned_pages_t value each time a window is - * created by calling scif_register_pinned_pages() and passing the - * scif_pinned_pages_t value. A reference is removed from a - * scif_pinned_pages_t value each time such a window is deleted. - * - * Subsequent operations which change the memory pages to which virtual - * addresses are mapped (such as mmap(), munmap()) have no effect on the - * scif_pinned_pages_t value or windows created against it. - * - * If the process will fork(), it is recommended that the registered - * virtual address range be marked with MADV_DONTFORK. Doing so will prevent - * problems due to copy-on-write semantics. - * - * The prot_flags argument is formed by OR'ing together one or more of the - * following values. - * SCIF_PROT_READ - allow read operations against the pages - * SCIF_PROT_WRITE - allow write operations against the pages - * The map_flags argument can be set as SCIF_MAP_KERNEL to interpret addr as a - * kernel space address. By default, addr is interpreted as a user space - * address. - * - * Return: - * Upon successful completion, scif_pin_pages() returns 0; otherwise the - * negative of one of the following errors is returned. - * - * Errors: - * EINVAL - prot_flags is invalid, map_flags is invalid, or offset is negative - * ENOMEM - Not enough space - */ -int scif_pin_pages(void *addr, size_t len, int prot_flags, int map_flags, - scif_pinned_pages_t *pinned_pages); - -/** - * scif_unpin_pages() - Unpin a set of pages - * @pinned_pages: Handle to pinned pages to be unpinned - * - * scif_unpin_pages() prevents scif_register_pinned_pages() from registering new - * windows against pinned_pages. The physical pages represented by pinned_pages - * will remain pinned until all windows previously registered against - * pinned_pages are deleted (the window is scif_unregister()'d and all - * references to the window are removed (see scif_unregister()). - * - * pinned_pages must have been obtain from a previous call to scif_pin_pages(). - * After calling scif_unpin_pages(), it is an error to pass pinned_pages to - * scif_register_pinned_pages(). - * - * Return: - * Upon successful completion, scif_unpin_pages() returns 0; otherwise the - * negative of one of the following errors is returned. - * - * Errors: - * EINVAL - pinned_pages is not valid - */ -int scif_unpin_pages(scif_pinned_pages_t pinned_pages); - -/** - * scif_register_pinned_pages() - Mark a memory region for remote access. - * @epd: endpoint descriptor - * @pinned_pages: Handle to pinned pages - * @offset: Registered address space offset - * @map_flags: Flags which control where pages are mapped - * - * The scif_register_pinned_pages() function opens a window, a range of whole - * pages of the registered address space of the endpoint epd, starting at - * offset po. The value of po, further described below, is a function of the - * parameters offset and pinned_pages, and the value of map_flags. Each page of - * the window represents a corresponding physical memory page of the range - * represented by pinned_pages; the length of the window is the same as the - * length of range represented by pinned_pages. A successful - * scif_register_pinned_pages() call returns po as the return value. - * - * When SCIF_MAP_FIXED is set in the map_flags argument, po will be offset - * exactly, and offset is constrained to be a multiple of the page size. The - * mapping established by scif_register_pinned_pages() will not replace any - * existing registration; an error is returned if any page of the new window - * would intersect an existing window. - * - * When SCIF_MAP_FIXED is not set, the implementation uses offset in an - * implementation-defined manner to arrive at po. The po so chosen will be an - * area of the registered address space that the implementation deems suitable - * for a mapping of the required size. An offset value of 0 is interpreted as - * granting the implementation complete freedom in selecting po, subject to - * constraints described below. A non-zero value of offset is taken to be a - * suggestion of an offset near which the mapping should be placed. When the - * implementation selects a value for po, it does not replace any extant - * window. In all cases, po will be a multiple of the page size. - * - * The physical pages which are so represented by a window are available for - * access in calls to scif_get_pages(), scif_readfrom(), scif_writeto(), - * scif_vreadfrom(), and scif_vwriteto(). While a window is registered, the - * physical pages represented by the window will not be reused by the memory - * subsystem for any other purpose. Note that the same physical page may be - * represented by multiple windows. - * - * Windows created by scif_register_pinned_pages() are unregistered by - * scif_unregister(). - * - * The map_flags argument can be set to SCIF_MAP_FIXED which interprets a - * fixed offset. - * - * Return: - * Upon successful completion, scif_register_pinned_pages() returns the offset - * at which the mapping was placed (po); otherwise the negative of one of the - * following errors is returned. - * - * Errors: - * EADDRINUSE - SCIF_MAP_FIXED is set in map_flags and pages in the new window - * would intersect an existing window - * EAGAIN - The mapping could not be performed due to lack of resources - * ECONNRESET - Connection reset by peer - * EINVAL - map_flags is invalid, or SCIF_MAP_FIXED is set in map_flags, and - * offset is not a multiple of the page size, or offset is negative - * ENODEV - The remote node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOMEM - Not enough space - * ENOTCONN - The endpoint is not connected - */ -off_t scif_register_pinned_pages(scif_epd_t epd, - scif_pinned_pages_t pinned_pages, - off_t offset, int map_flags); - -/** - * scif_get_pages() - Add references to remote registered pages - * @epd: endpoint descriptor - * @offset: remote registered offset - * @len: length of range of pages - * @pages: returned scif_range structure - * - * scif_get_pages() returns the addresses of the physical pages represented by - * those pages of the registered address space of the peer of epd, starting at - * offset and continuing for len bytes. offset and len are constrained to be - * multiples of the page size. - * - * All of the pages in the specified range [offset, offset + len - 1] must be - * within a single window of the registered address space of the peer of epd. - * - * The addresses are returned as a virtually contiguous array pointed to by the - * phys_addr component of the scif_range structure whose address is returned in - * pages. The nr_pages component of scif_range is the length of the array. The - * prot_flags component of scif_range holds the protection flag value passed - * when the pages were registered. - * - * Each physical page whose address is returned by scif_get_pages() remains - * available and will not be released for reuse until the scif_range structure - * is returned in a call to scif_put_pages(). The scif_range structure returned - * by scif_get_pages() must be unmodified. - * - * It is an error to call scif_close() on an endpoint on which a scif_range - * structure of that endpoint has not been returned to scif_put_pages(). - * - * Return: - * Upon successful completion, scif_get_pages() returns 0; otherwise the - * negative of one of the following errors is returned. - * Errors: - * ECONNRESET - Connection reset by peer. - * EINVAL - offset is not a multiple of the page size, or offset is negative, or - * len is not a multiple of the page size - * ENODEV - The remote node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOTCONN - The endpoint is not connected - * ENXIO - Offsets in the range [offset, offset + len - 1] are invalid - * for the registered address space of the peer epd - */ -int scif_get_pages(scif_epd_t epd, off_t offset, size_t len, - struct scif_range **pages); - -/** - * scif_put_pages() - Remove references from remote registered pages - * @pages: pages to be returned - * - * scif_put_pages() releases a scif_range structure previously obtained by - * calling scif_get_pages(). The physical pages represented by pages may - * be reused when the window which represented those pages is unregistered. - * Therefore, those pages must not be accessed after calling scif_put_pages(). - * - * Return: - * Upon successful completion, scif_put_pages() returns 0; otherwise the - * negative of one of the following errors is returned. - * Errors: - * EINVAL - pages does not point to a valid scif_range structure, or - * the scif_range structure pointed to by pages was already returned - * ENODEV - The remote node is lost or existed, but is not currently in the - * network since it may have crashed - * ENOTCONN - The endpoint is not connected - */ -int scif_put_pages(struct scif_range *pages); - -/** - * scif_poll() - Wait for some event on an endpoint - * @epds: Array of endpoint descriptors - * @nepds: Length of epds - * @timeout: Upper limit on time for which scif_poll() will block - * - * scif_poll() waits for one of a set of endpoints to become ready to perform - * an I/O operation. - * - * The epds argument specifies the endpoint descriptors to be examined and the - * events of interest for each endpoint descriptor. epds is a pointer to an - * array with one member for each open endpoint descriptor of interest. - * - * The number of items in the epds array is specified in nepds. The epd field - * of scif_pollepd is an endpoint descriptor of an open endpoint. The field - * events is a bitmask specifying the events which the application is - * interested in. The field revents is an output parameter, filled by the - * kernel with the events that actually occurred. The bits returned in revents - * can include any of those specified in events, or one of the values EPOLLERR, - * EPOLLHUP, or EPOLLNVAL. (These three bits are meaningless in the events - * field, and will be set in the revents field whenever the corresponding - * condition is true.) - * - * If none of the events requested (and no error) has occurred for any of the - * endpoint descriptors, then scif_poll() blocks until one of the events occurs. - * - * The timeout argument specifies an upper limit on the time for which - * scif_poll() will block, in milliseconds. Specifying a negative value in - * timeout means an infinite timeout. - * - * The following bits may be set in events and returned in revents. - * EPOLLIN - Data may be received without blocking. For a connected - * endpoint, this means that scif_recv() may be called without blocking. For a - * listening endpoint, this means that scif_accept() may be called without - * blocking. - * EPOLLOUT - Data may be sent without blocking. For a connected endpoint, this - * means that scif_send() may be called without blocking. EPOLLOUT may also be - * used to block waiting for a non-blocking connect to complete. This bit value - * has no meaning for a listening endpoint and is ignored if specified. - * - * The following bits are only returned in revents, and are ignored if set in - * events. - * EPOLLERR - An error occurred on the endpoint - * EPOLLHUP - The connection to the peer endpoint was disconnected - * EPOLLNVAL - The specified endpoint descriptor is invalid. - * - * Return: - * Upon successful completion, scif_poll() returns a non-negative value. A - * positive value indicates the total number of endpoint descriptors that have - * been selected (that is, endpoint descriptors for which the revents member is - * non-zero). A value of 0 indicates that the call timed out and no endpoint - * descriptors have been selected. Otherwise in user mode -1 is returned and - * errno is set to indicate the error; in kernel mode the negative of one of - * the following errors is returned. - * - * Errors: - * EINTR - A signal occurred before any requested event - * EINVAL - The nepds argument is greater than {OPEN_MAX} - * ENOMEM - There was no space to allocate file descriptor tables - */ -int scif_poll(struct scif_pollepd *epds, unsigned int nepds, long timeout); - -/** - * scif_client_register() - Register a SCIF client - * @client: client to be registered - * - * scif_client_register() registers a SCIF client. The probe() method - * of the client is called when SCIF peer devices come online and the - * remove() method is called when the peer devices disappear. - * - * Return: - * Upon successful completion, scif_client_register() returns a non-negative - * value. Otherwise the return value is the same as subsys_interface_register() - * in the kernel. - */ -int scif_client_register(struct scif_client *client); - -/** - * scif_client_unregister() - Unregister a SCIF client - * @client: client to be unregistered - * - * scif_client_unregister() unregisters a SCIF client. - * - * Return: - * None - */ -void scif_client_unregister(struct scif_client *client); - -#endif /* __SCIF_H__ */ diff --git a/include/uapi/linux/mic_common.h b/include/uapi/linux/mic_common.h deleted file mode 100644 index 504e523f702c..000000000000 --- a/include/uapi/linux/mic_common.h +++ /dev/null @@ -1,235 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License, version 2, as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * The full GNU General Public License is included in this distribution in - * the file called "COPYING". - * - * Intel MIC driver. - * - */ -#ifndef __MIC_COMMON_H_ -#define __MIC_COMMON_H_ - -#include - -#define __mic_align(a, x) (((a) + (x) - 1) & ~((x) - 1)) - -/** - * struct mic_device_desc: Virtio device information shared between the - * virtio driver and userspace backend - * - * @type: Device type: console/network/disk etc. Type 0/-1 terminates. - * @num_vq: Number of virtqueues. - * @feature_len: Number of bytes of feature bits. Multiply by 2: one for - host features and one for guest acknowledgements. - * @config_len: Number of bytes of the config array after virtqueues. - * @status: A status byte, written by the Guest. - * @config: Start of the following variable length config. - */ -struct mic_device_desc { - __s8 type; - __u8 num_vq; - __u8 feature_len; - __u8 config_len; - __u8 status; - __le64 config[0]; -} __attribute__ ((aligned(8))); - -/** - * struct mic_device_ctrl: Per virtio device information in the device page - * used internally by the host and card side drivers. - * - * @vdev: Used for storing MIC vdev information by the guest. - * @config_change: Set to 1 by host when a config change is requested. - * @vdev_reset: Set to 1 by guest to indicate virtio device has been reset. - * @guest_ack: Set to 1 by guest to ack a command. - * @host_ack: Set to 1 by host to ack a command. - * @used_address_updated: Set to 1 by guest when the used address should be - * updated. - * @c2h_vdev_db: The doorbell number to be used by guest. Set by host. - * @h2c_vdev_db: The doorbell number to be used by host. Set by guest. - */ -struct mic_device_ctrl { - __le64 vdev; - __u8 config_change; - __u8 vdev_reset; - __u8 guest_ack; - __u8 host_ack; - __u8 used_address_updated; - __s8 c2h_vdev_db; - __s8 h2c_vdev_db; -} __attribute__ ((aligned(8))); - -/** - * struct mic_bootparam: Virtio device independent information in device page - * - * @magic: A magic value used by the card to ensure it can see the host - * @h2c_config_db: Host to Card Virtio config doorbell set by card - * @node_id: Unique id of the node - * @h2c_scif_db - Host to card SCIF doorbell set by card - * @c2h_scif_db - Card to host SCIF doorbell set by host - * @scif_host_dma_addr - SCIF host queue pair DMA address - * @scif_card_dma_addr - SCIF card queue pair DMA address - */ -struct mic_bootparam { - __le32 magic; - __s8 h2c_config_db; - __u8 node_id; - __u8 h2c_scif_db; - __u8 c2h_scif_db; - __u64 scif_host_dma_addr; - __u64 scif_card_dma_addr; -} __attribute__ ((aligned(8))); - -/** - * struct mic_device_page: High level representation of the device page - * - * @bootparam: The bootparam structure is used for sharing information and - * status updates between MIC host and card drivers. - * @desc: Array of MIC virtio device descriptors. - */ -struct mic_device_page { - struct mic_bootparam bootparam; - struct mic_device_desc desc[0]; -}; -/** - * struct mic_vqconfig: This is how we expect the device configuration field - * for a virtqueue to be laid out in config space. - * - * @address: Guest/MIC physical address of the virtio ring - * (avail and desc rings) - * @used_address: Guest/MIC physical address of the used ring - * @num: The number of entries in the virtio_ring - */ -struct mic_vqconfig { - __le64 address; - __le64 used_address; - __le16 num; -} __attribute__ ((aligned(8))); - -/* - * The alignment to use between consumer and producer parts of vring. - * This is pagesize for historical reasons. - */ -#define MIC_VIRTIO_RING_ALIGN 4096 - -#define MIC_MAX_VRINGS 4 -#define MIC_VRING_ENTRIES 128 - -/* - * Max vring entries (power of 2) to ensure desc and avail rings - * fit in a single page - */ -#define MIC_MAX_VRING_ENTRIES 128 - -/** - * Max size of the desc block in bytes: includes: - * - struct mic_device_desc - * - struct mic_vqconfig (num_vq of these) - * - host and guest features - * - virtio device config space - */ -#define MIC_MAX_DESC_BLK_SIZE 256 - -/** - * struct _mic_vring_info - Host vring info exposed to userspace backend - * for the avail index and magic for the card. - * - * @avail_idx: host avail idx - * @magic: A magic debug cookie. - */ -struct _mic_vring_info { - __u16 avail_idx; - __le32 magic; -}; - -/** - * struct mic_vring - Vring information. - * - * @vr: The virtio ring. - * @info: Host vring information exposed to the userspace backend for the - * avail index and magic for the card. - * @va: The va for the buffer allocated for vr and info. - * @len: The length of the buffer required for allocating vr and info. - */ -struct mic_vring { - struct vring vr; - struct _mic_vring_info *info; - void *va; - int len; -}; - -#define mic_aligned_desc_size(d) __mic_align(mic_desc_size(d), 8) - -#ifndef INTEL_MIC_CARD -static inline unsigned mic_desc_size(const struct mic_device_desc *desc) -{ - return sizeof(*desc) + desc->num_vq * sizeof(struct mic_vqconfig) - + desc->feature_len * 2 + desc->config_len; -} - -static inline struct mic_vqconfig * -mic_vq_config(const struct mic_device_desc *desc) -{ - return (struct mic_vqconfig *)(desc + 1); -} - -static inline __u8 *mic_vq_features(const struct mic_device_desc *desc) -{ - return (__u8 *)(mic_vq_config(desc) + desc->num_vq); -} - -static inline __u8 *mic_vq_configspace(const struct mic_device_desc *desc) -{ - return mic_vq_features(desc) + desc->feature_len * 2; -} -static inline unsigned mic_total_desc_size(struct mic_device_desc *desc) -{ - return mic_aligned_desc_size(desc) + sizeof(struct mic_device_ctrl); -} -#endif - -/* Device page size */ -#define MIC_DP_SIZE 4096 - -#define MIC_MAGIC 0xc0ffee00 - -/** - * enum mic_states - MIC states. - */ -enum mic_states { - MIC_READY = 0, - MIC_BOOTING, - MIC_ONLINE, - MIC_SHUTTING_DOWN, - MIC_RESETTING, - MIC_RESET_FAILED, - MIC_LAST -}; - -/** - * enum mic_status - MIC status reported by card after - * a host or card initiated shutdown or a card crash. - */ -enum mic_status { - MIC_NOP = 0, - MIC_CRASHED, - MIC_HALTED, - MIC_POWER_OFF, - MIC_RESTART, - MIC_STATUS_LAST -}; - -#endif diff --git a/include/uapi/linux/mic_ioctl.h b/include/uapi/linux/mic_ioctl.h deleted file mode 100644 index 687b9cd9d3e2..000000000000 --- a/include/uapi/linux/mic_ioctl.h +++ /dev/null @@ -1,77 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License, version 2, as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * The full GNU General Public License is included in this distribution in - * the file called "COPYING". - * - * Intel MIC Host driver. - * - */ -#ifndef _MIC_IOCTL_H_ -#define _MIC_IOCTL_H_ - -#include - -/* - * mic_copy - MIC virtio descriptor copy. - * - * @iov: An array of IOVEC structures containing user space buffers. - * @iovcnt: Number of IOVEC structures in iov. - * @vr_idx: The vring index. - * @update_used: A non zero value results in used index being updated. - * @out_len: The aggregate of the total length written to or read from - * the virtio device. - */ -struct mic_copy_desc { -#ifdef __KERNEL__ - struct iovec __user *iov; -#else - struct iovec *iov; -#endif - __u32 iovcnt; - __u8 vr_idx; - __u8 update_used; - __u32 out_len; -}; - -/* - * Add a new virtio device - * The (struct mic_device_desc *) pointer points to a device page entry - * for the virtio device consisting of: - * - struct mic_device_desc - * - struct mic_vqconfig (num_vq of these) - * - host and guest features - * - virtio device config space - * The total size referenced by the pointer should equal the size returned - * by desc_size() in mic_common.h - */ -#define MIC_VIRTIO_ADD_DEVICE _IOWR('s', 1, struct mic_device_desc *) - -/* - * Copy the number of entries in the iovec and update the used index - * if requested by the user. - */ -#define MIC_VIRTIO_COPY_DESC _IOWR('s', 2, struct mic_copy_desc *) - -/* - * Notify virtio device of a config change - * The (__u8 *) pointer points to config space values for the device - * as they should be written into the device page. The total size - * referenced by the pointer should equal the config_len field of struct - * mic_device_desc. - */ -#define MIC_VIRTIO_CONFIG_CHANGE _IOWR('s', 5, __u8 *) - -#endif diff --git a/samples/mic/mpssd/.gitignore b/samples/mic/mpssd/.gitignore deleted file mode 100644 index aa03f1eb37a0..000000000000 --- a/samples/mic/mpssd/.gitignore +++ /dev/null @@ -1,2 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0-only -mpssd diff --git a/samples/mic/mpssd/Makefile b/samples/mic/mpssd/Makefile deleted file mode 100644 index a7a6e0c70424..000000000000 --- a/samples/mic/mpssd/Makefile +++ /dev/null @@ -1,28 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0 -ifndef CROSS_COMPILE -uname_M := $(shell uname -m 2>/dev/null || echo not) -ARCH ?= $(shell echo $(uname_M) | sed -e s/i.86/x86/ -e s/x86_64/x86/) - -ifeq ($(ARCH),x86) - -PROGS := mpssd -CC = $(CROSS_COMPILE)gcc -CFLAGS := -I../../../usr/include -I../../../tools/include - -ifdef DEBUG -CFLAGS += -DDEBUG=$(DEBUG) -endif - -all: $(PROGS) -mpssd: mpssd.c sysfs.c - $(CC) $(CFLAGS) mpssd.c sysfs.c -o mpssd -lpthread - -install: - install mpssd /usr/sbin/mpssd - install micctrl /usr/sbin/micctrl - -clean: - rm -fr $(PROGS) - -endif -endif diff --git a/samples/mic/mpssd/micctrl b/samples/mic/mpssd/micctrl deleted file mode 100755 index 030a60b04046..000000000000 --- a/samples/mic/mpssd/micctrl +++ /dev/null @@ -1,162 +0,0 @@ -#!/bin/bash -# SPDX-License-Identifier: GPL-2.0-only -# Intel MIC Platform Software Stack (MPSS) -# -# Copyright(c) 2013 Intel Corporation. -# -# Intel MIC User Space Tools. -# -# micctrl - Controls MIC boot/start/stop. -# -# chkconfig: 2345 95 05 -# description: start MPSS stack processing. -# -### BEGIN INIT INFO -# Provides: micctrl -### END INIT INFO - -# Source function library. -. /etc/init.d/functions - -sysfs="/sys/class/mic" - -_status() -{ - f=$sysfs/$1 - echo -e $1 state: "`cat $f/state`" shutdown_status: "`cat $f/shutdown_status`" -} - -status() -{ - if [ "`echo $1 | head -c3`" == "mic" ]; then - _status $1 - return $? - fi - for f in $sysfs/* - do - _status `basename $f` - RETVAL=$? - [ $RETVAL -ne 0 ] && return $RETVAL - done - return 0 -} - -_reset() -{ - f=$sysfs/$1 - echo reset > $f/state -} - -reset() -{ - if [ "`echo $1 | head -c3`" == "mic" ]; then - _reset $1 - return $? - fi - for f in $sysfs/* - do - _reset `basename $f` - RETVAL=$? - [ $RETVAL -ne 0 ] && return $RETVAL - done - return 0 -} - -_boot() -{ - f=$sysfs/$1 - echo "linux" > $f/bootmode - echo "mic/uos.img" > $f/firmware - echo "mic/$1.image" > $f/ramdisk - echo "boot" > $f/state -} - -boot() -{ - if [ "`echo $1 | head -c3`" == "mic" ]; then - _boot $1 - return $? - fi - for f in $sysfs/* - do - _boot `basename $f` - RETVAL=$? - [ $RETVAL -ne 0 ] && return $RETVAL - done - return 0 -} - -_shutdown() -{ - f=$sysfs/$1 - echo shutdown > $f/state -} - -shutdown() -{ - if [ "`echo $1 | head -c3`" == "mic" ]; then - _shutdown $1 - return $? - fi - for f in $sysfs/* - do - _shutdown `basename $f` - RETVAL=$? - [ $RETVAL -ne 0 ] && return $RETVAL - done - return 0 -} - -_wait() -{ - f=$sysfs/$1 - while [ "`cat $f/state`" != "offline" -a "`cat $f/state`" != "online" ] - do - sleep 1 - echo -e "Waiting for $1 to go offline" - done -} - -wait() -{ - if [ "`echo $1 | head -c3`" == "mic" ]; then - _wait $1 - return $? - fi - # Wait for the cards to go offline - for f in $sysfs/* - do - _wait `basename $f` - RETVAL=$? - [ $RETVAL -ne 0 ] && return $RETVAL - done - return 0 -} - -if [ ! -d "$sysfs" ]; then - echo -e $"Module unloaded " - exit 3 -fi - -case $1 in - -s) - status $2 - ;; - -r) - reset $2 - ;; - -b) - boot $2 - ;; - -S) - shutdown $2 - ;; - -w) - wait $2 - ;; - *) - echo $"Usage: $0 {-s (status) |-r (reset) |-b (boot) |-S (shutdown) |-w (wait)}" - exit 2 -esac - -exit $? diff --git a/samples/mic/mpssd/mpss b/samples/mic/mpssd/mpss deleted file mode 100755 index 248ac7313c71..000000000000 --- a/samples/mic/mpssd/mpss +++ /dev/null @@ -1,189 +0,0 @@ -#!/bin/bash -# SPDX-License-Identifier: GPL-2.0-only -# Intel MIC Platform Software Stack (MPSS) -# -# Copyright(c) 2013 Intel Corporation. -# -# Intel MIC User Space Tools. -# -# mpss Start mpssd. -# -# chkconfig: 2345 95 05 -# description: start MPSS stack processing. -# -### BEGIN INIT INFO -# Provides: mpss -# Required-Start: -# Required-Stop: -# Short-Description: MPSS stack control -# Description: MPSS stack control -### END INIT INFO - -# Source function library. -. /etc/init.d/functions - -exec=/usr/sbin/mpssd -sysfs="/sys/class/mic" -mic_modules="mic_host mic_x100_dma scif vop" - -start() -{ - [ -x $exec ] || exit 5 - - if [ "`ps -e | awk '{print $4}' | grep mpssd | head -1`" = "mpssd" ]; then - echo -e $"MPSSD already running! " - success - echo - return 0 - fi - - echo -e $"Starting MPSS Stack" - echo -e $"Loading MIC drivers:" $mic_modules - - modprobe -a $mic_modules - RETVAL=$? - if [ $RETVAL -ne 0 ]; then - failure - echo - return $RETVAL - fi - - # Start the daemon - echo -n $"Starting MPSSD " - $exec - RETVAL=$? - if [ $RETVAL -ne 0 ]; then - failure - echo - return $RETVAL - fi - success - echo - - sleep 5 - - # Boot the cards - micctrl -b - - # Wait till ping works - for f in $sysfs/* - do - count=100 - ipaddr=`cat $f/cmdline` - ipaddr=${ipaddr#*address,} - ipaddr=`echo $ipaddr | cut -d, -f1 | cut -d\; -f1` - while [ $count -ge 0 ] - do - echo -e "Pinging "`basename $f`" " - ping -c 1 $ipaddr &> /dev/null - RETVAL=$? - if [ $RETVAL -eq 0 ]; then - success - break - fi - sleep 1 - count=`expr $count - 1` - done - [ $RETVAL -ne 0 ] && failure || success - echo - done - return $RETVAL -} - -stop() -{ - echo -e $"Shutting down MPSS Stack: " - - # Bail out if module is unloaded - if [ ! -d "$sysfs" ]; then - echo -n $"Module unloaded " - success - echo - return 0 - fi - - # Shut down the cards. - micctrl -S - - # Wait for the cards to go offline - for f in $sysfs/* - do - while [ "`cat $f/state`" != "ready" ] - do - sleep 1 - echo -e "Waiting for "`basename $f`" to become ready" - done - done - - # Display the status of the cards - micctrl -s - - # Kill MPSSD now - echo -n $"Killing MPSSD" - killall -9 mpssd 2>/dev/null - RETVAL=$? - [ $RETVAL -ne 0 ] && failure || success - echo - return $RETVAL -} - -restart() -{ - stop - sleep 5 - start -} - -status() -{ - micctrl -s - if [ "`ps -e | awk '{print $4}' | grep mpssd | head -n 1`" = "mpssd" ]; then - echo "mpssd is running" - else - echo "mpssd is stopped" - fi - return 0 -} - -unload() -{ - if [ ! -d "$sysfs" ]; then - echo -n $"No MIC_HOST Module: " - success - echo - return - fi - - stop - - sleep 5 - echo -n $"Removing MIC drivers:" $mic_modules - modprobe -r $mic_modules - RETVAL=$? - [ $RETVAL -ne 0 ] && failure || success - echo - return $RETVAL -} - -case $1 in - start) - start - ;; - stop) - stop - ;; - restart) - restart - ;; - status) - status - ;; - unload) - unload - ;; - *) - echo $"Usage: $0 {start|stop|restart|status|unload}" - exit 2 -esac - -exit $? diff --git a/samples/mic/mpssd/mpssd.c b/samples/mic/mpssd/mpssd.c deleted file mode 100644 index c03a05d498f0..000000000000 --- a/samples/mic/mpssd/mpssd.c +++ /dev/null @@ -1,1815 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Intel MIC User Space Tools. - */ - -#define _GNU_SOURCE - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "mpssd.h" -#include -#include -#include - -static void *init_mic(void *arg); - -static FILE *logfp; -static struct mic_info mic_list; - -#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0])) - -#define min_t(type, x, y) ({ \ - type __min1 = (x); \ - type __min2 = (y); \ - __min1 < __min2 ? __min1 : __min2; }) - -/* align addr on a size boundary - adjust address up/down if needed */ -#define _ALIGN_DOWN(addr, size) ((addr)&(~((size)-1))) -#define _ALIGN_UP(addr, size) _ALIGN_DOWN(addr + size - 1, size) - -/* align addr on a size boundary - adjust address up if needed */ -#define _ALIGN(addr, size) _ALIGN_UP(addr, size) - -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) _ALIGN(addr, PAGE_SIZE) - -#define READ_ONCE(x) (*(volatile typeof(x) *)&(x)) - -#define GSO_ENABLED 1 -#define MAX_GSO_SIZE (64 * 1024) -#define ETH_H_LEN 14 -#define MAX_NET_PKT_SIZE (_ALIGN_UP(MAX_GSO_SIZE + ETH_H_LEN, 64)) -#define MIC_DEVICE_PAGE_END 0x1000 - -#ifndef VIRTIO_NET_HDR_F_DATA_VALID -#define VIRTIO_NET_HDR_F_DATA_VALID 2 /* Csum is valid */ -#endif - -static struct { - struct mic_device_desc dd; - struct mic_vqconfig vqconfig[2]; - __u32 host_features, guest_acknowledgements; - struct virtio_console_config cons_config; -} virtcons_dev_page = { - .dd = { - .type = VIRTIO_ID_CONSOLE, - .num_vq = ARRAY_SIZE(virtcons_dev_page.vqconfig), - .feature_len = sizeof(virtcons_dev_page.host_features), - .config_len = sizeof(virtcons_dev_page.cons_config), - }, - .vqconfig[0] = { - .num = htole16(MIC_VRING_ENTRIES), - }, - .vqconfig[1] = { - .num = htole16(MIC_VRING_ENTRIES), - }, -}; - -static struct { - struct mic_device_desc dd; - struct mic_vqconfig vqconfig[2]; - __u32 host_features, guest_acknowledgements; - struct virtio_net_config net_config; -} virtnet_dev_page = { - .dd = { - .type = VIRTIO_ID_NET, - .num_vq = ARRAY_SIZE(virtnet_dev_page.vqconfig), - .feature_len = sizeof(virtnet_dev_page.host_features), - .config_len = sizeof(virtnet_dev_page.net_config), - }, - .vqconfig[0] = { - .num = htole16(MIC_VRING_ENTRIES), - }, - .vqconfig[1] = { - .num = htole16(MIC_VRING_ENTRIES), - }, -#if GSO_ENABLED - .host_features = htole32( - 1 << VIRTIO_NET_F_CSUM | - 1 << VIRTIO_NET_F_GSO | - 1 << VIRTIO_NET_F_GUEST_TSO4 | - 1 << VIRTIO_NET_F_GUEST_TSO6 | - 1 << VIRTIO_NET_F_GUEST_ECN), -#else - .host_features = 0, -#endif -}; - -static const char *mic_config_dir = "/etc/mpss"; -static const char *virtblk_backend = "VIRTBLK_BACKEND"; -static struct { - struct mic_device_desc dd; - struct mic_vqconfig vqconfig[1]; - __u32 host_features, guest_acknowledgements; - struct virtio_blk_config blk_config; -} virtblk_dev_page = { - .dd = { - .type = VIRTIO_ID_BLOCK, - .num_vq = ARRAY_SIZE(virtblk_dev_page.vqconfig), - .feature_len = sizeof(virtblk_dev_page.host_features), - .config_len = sizeof(virtblk_dev_page.blk_config), - }, - .vqconfig[0] = { - .num = htole16(MIC_VRING_ENTRIES), - }, - .host_features = - htole32(1<name, strerror(errno)); - return ret; - } - } - if (pid < 0) { - mpsslog("%s fork failed errno %s\n", - mic->name, strerror(errno)); - return ret; - } - - ret = waitpid(pid, NULL, 0); - if (ret < 0) { - mpsslog("%s waitpid failed errno %s\n", - mic->name, strerror(errno)); - return ret; - } - - snprintf(ipaddr, IFNAMSIZ, "172.31.%d.254/24", mic->id + 1); - - pid = fork(); - if (pid == 0) { - ifargv[0] = "ip"; - ifargv[1] = "addr"; - ifargv[2] = "add"; - ifargv[3] = ipaddr; - ifargv[4] = "dev"; - ifargv[5] = dev; - ifargv[6] = NULL; - mpsslog("Configuring %s ipaddr %s\n", dev, ipaddr); - ret = execvp("ip", ifargv); - if (ret < 0) { - mpsslog("%s execvp failed errno %s\n", - mic->name, strerror(errno)); - return ret; - } - } - if (pid < 0) { - mpsslog("%s fork failed errno %s\n", - mic->name, strerror(errno)); - return ret; - } - - ret = waitpid(pid, NULL, 0); - if (ret < 0) { - mpsslog("%s waitpid failed errno %s\n", - mic->name, strerror(errno)); - return ret; - } - mpsslog("MIC name %s %s %d DONE!\n", - mic->name, __func__, __LINE__); - return 0; -} - -static int tun_alloc(struct mic_info *mic, char *dev) -{ - struct ifreq ifr; - int fd, err; -#if GSO_ENABLED - unsigned offload; -#endif - fd = open("/dev/net/tun", O_RDWR); - if (fd < 0) { - mpsslog("Could not open /dev/net/tun %s\n", strerror(errno)); - goto done; - } - - memset(&ifr, 0, sizeof(ifr)); - - ifr.ifr_flags = IFF_TAP | IFF_NO_PI | IFF_VNET_HDR; - if (*dev) - strncpy(ifr.ifr_name, dev, IFNAMSIZ); - - err = ioctl(fd, TUNSETIFF, (void *)&ifr); - if (err < 0) { - mpsslog("%s %s %d TUNSETIFF failed %s\n", - mic->name, __func__, __LINE__, strerror(errno)); - close(fd); - return err; - } -#if GSO_ENABLED - offload = TUN_F_CSUM | TUN_F_TSO4 | TUN_F_TSO6 | TUN_F_TSO_ECN; - - err = ioctl(fd, TUNSETOFFLOAD, offload); - if (err < 0) { - mpsslog("%s %s %d TUNSETOFFLOAD failed %s\n", - mic->name, __func__, __LINE__, strerror(errno)); - close(fd); - return err; - } -#endif - strcpy(dev, ifr.ifr_name); - mpsslog("Created TAP %s\n", dev); -done: - return fd; -} - -#define NET_FD_VIRTIO_NET 0 -#define NET_FD_TUN 1 -#define MAX_NET_FD 2 - -static void set_dp(struct mic_info *mic, int type, void *dp) -{ - switch (type) { - case VIRTIO_ID_CONSOLE: - mic->mic_console.console_dp = dp; - return; - case VIRTIO_ID_NET: - mic->mic_net.net_dp = dp; - return; - case VIRTIO_ID_BLOCK: - mic->mic_virtblk.block_dp = dp; - return; - } - mpsslog("%s %s %d not found\n", mic->name, __func__, type); - assert(0); -} - -static void *get_dp(struct mic_info *mic, int type) -{ - switch (type) { - case VIRTIO_ID_CONSOLE: - return mic->mic_console.console_dp; - case VIRTIO_ID_NET: - return mic->mic_net.net_dp; - case VIRTIO_ID_BLOCK: - return mic->mic_virtblk.block_dp; - } - mpsslog("%s %s %d not found\n", mic->name, __func__, type); - assert(0); - return NULL; -} - -static struct mic_device_desc *get_device_desc(struct mic_info *mic, int type) -{ - struct mic_device_desc *d; - int i; - void *dp = get_dp(mic, type); - - for (i = sizeof(struct mic_bootparam); i < PAGE_SIZE; - i += mic_total_desc_size(d)) { - d = dp + i; - - /* End of list */ - if (d->type == 0) - break; - - if (d->type == -1) - continue; - - mpsslog("%s %s d-> type %d d %p\n", - mic->name, __func__, d->type, d); - - if (d->type == (__u8)type) - return d; - } - mpsslog("%s %s %d not found\n", mic->name, __func__, type); - return NULL; -} - -/* See comments in vhost.c for explanation of next_desc() */ -static unsigned next_desc(struct vring_desc *desc) -{ - unsigned int next; - - if (!(le16toh(desc->flags) & VRING_DESC_F_NEXT)) - return -1U; - next = le16toh(desc->next); - return next; -} - -/* Sum up all the IOVEC length */ -static ssize_t -sum_iovec_len(struct mic_copy_desc *copy) -{ - ssize_t sum = 0; - unsigned int i; - - for (i = 0; i < copy->iovcnt; i++) - sum += copy->iov[i].iov_len; - return sum; -} - -static inline void verify_out_len(struct mic_info *mic, - struct mic_copy_desc *copy) -{ - if (copy->out_len != sum_iovec_len(copy)) { - mpsslog("%s %s %d BUG copy->out_len 0x%x len 0x%zx\n", - mic->name, __func__, __LINE__, - copy->out_len, sum_iovec_len(copy)); - assert(copy->out_len == sum_iovec_len(copy)); - } -} - -/* Display an iovec */ -static void -disp_iovec(struct mic_info *mic, struct mic_copy_desc *copy, - const char *s, int line) -{ - unsigned int i; - - for (i = 0; i < copy->iovcnt; i++) - mpsslog("%s %s %d copy->iov[%d] addr %p len 0x%zx\n", - mic->name, s, line, i, - copy->iov[i].iov_base, copy->iov[i].iov_len); -} - -static inline __u16 read_avail_idx(struct mic_vring *vr) -{ - return READ_ONCE(vr->info->avail_idx); -} - -static inline void txrx_prepare(int type, bool tx, struct mic_vring *vr, - struct mic_copy_desc *copy, ssize_t len) -{ - copy->vr_idx = tx ? 0 : 1; - copy->update_used = true; - if (type == VIRTIO_ID_NET) - copy->iov[1].iov_len = len - sizeof(struct virtio_net_hdr); - else - copy->iov[0].iov_len = len; -} - -/* Central API which triggers the copies */ -static int -mic_virtio_copy(struct mic_info *mic, int fd, - struct mic_vring *vr, struct mic_copy_desc *copy) -{ - int ret; - - ret = ioctl(fd, MIC_VIRTIO_COPY_DESC, copy); - if (ret) { - mpsslog("%s %s %d errno %s ret %d\n", - mic->name, __func__, __LINE__, - strerror(errno), ret); - } - return ret; -} - -static inline unsigned _vring_size(unsigned int num, unsigned long align) -{ - return _ALIGN_UP(((sizeof(struct vring_desc) * num + sizeof(__u16) * (3 + num) - + align - 1) & ~(align - 1)) - + sizeof(__u16) * 3 + sizeof(struct vring_used_elem) * num, 4); -} - -/* - * This initialization routine requires at least one - * vring i.e. vr0. vr1 is optional. - */ -static void * -init_vr(struct mic_info *mic, int fd, int type, - struct mic_vring *vr0, struct mic_vring *vr1, int num_vq) -{ - int vr_size; - char *va; - - vr_size = PAGE_ALIGN(_vring_size(MIC_VRING_ENTRIES, - MIC_VIRTIO_RING_ALIGN) + - sizeof(struct _mic_vring_info)); - va = mmap(NULL, MIC_DEVICE_PAGE_END + vr_size * num_vq, - PROT_READ, MAP_SHARED, fd, 0); - if (MAP_FAILED == va) { - mpsslog("%s %s %d mmap failed errno %s\n", - mic->name, __func__, __LINE__, - strerror(errno)); - goto done; - } - set_dp(mic, type, va); - vr0->va = (struct mic_vring *)&va[MIC_DEVICE_PAGE_END]; - vr0->info = vr0->va + - _vring_size(MIC_VRING_ENTRIES, MIC_VIRTIO_RING_ALIGN); - vring_init(&vr0->vr, - MIC_VRING_ENTRIES, vr0->va, MIC_VIRTIO_RING_ALIGN); - mpsslog("%s %s vr0 %p vr0->info %p vr_size 0x%x vring 0x%x ", - __func__, mic->name, vr0->va, vr0->info, vr_size, - _vring_size(MIC_VRING_ENTRIES, MIC_VIRTIO_RING_ALIGN)); - mpsslog("magic 0x%x expected 0x%x\n", - le32toh(vr0->info->magic), MIC_MAGIC + type); - assert(le32toh(vr0->info->magic) == MIC_MAGIC + type); - if (vr1) { - vr1->va = (struct mic_vring *) - &va[MIC_DEVICE_PAGE_END + vr_size]; - vr1->info = vr1->va + _vring_size(MIC_VRING_ENTRIES, - MIC_VIRTIO_RING_ALIGN); - vring_init(&vr1->vr, - MIC_VRING_ENTRIES, vr1->va, MIC_VIRTIO_RING_ALIGN); - mpsslog("%s %s vr1 %p vr1->info %p vr_size 0x%x vring 0x%x ", - __func__, mic->name, vr1->va, vr1->info, vr_size, - _vring_size(MIC_VRING_ENTRIES, MIC_VIRTIO_RING_ALIGN)); - mpsslog("magic 0x%x expected 0x%x\n", - le32toh(vr1->info->magic), MIC_MAGIC + type + 1); - assert(le32toh(vr1->info->magic) == MIC_MAGIC + type + 1); - } -done: - return va; -} - -static int -wait_for_card_driver(struct mic_info *mic, int fd, int type) -{ - struct pollfd pollfd; - int err; - struct mic_device_desc *desc = get_device_desc(mic, type); - __u8 prev_status; - - if (!desc) - return -ENODEV; - prev_status = desc->status; - pollfd.fd = fd; - mpsslog("%s %s Waiting .... desc-> type %d status 0x%x\n", - mic->name, __func__, type, desc->status); - - while (1) { - pollfd.events = POLLIN; - pollfd.revents = 0; - err = poll(&pollfd, 1, -1); - if (err < 0) { - mpsslog("%s %s poll failed %s\n", - mic->name, __func__, strerror(errno)); - continue; - } - - if (pollfd.revents) { - if (desc->status != prev_status) { - mpsslog("%s %s Waiting... desc-> type %d " - "status 0x%x\n", - mic->name, __func__, type, - desc->status); - prev_status = desc->status; - } - if (desc->status & VIRTIO_CONFIG_S_DRIVER_OK) { - mpsslog("%s %s poll.revents %d\n", - mic->name, __func__, pollfd.revents); - mpsslog("%s %s desc-> type %d status 0x%x\n", - mic->name, __func__, type, - desc->status); - break; - } - } - } - return 0; -} - -/* Spin till we have some descriptors */ -static void -spin_for_descriptors(struct mic_info *mic, struct mic_vring *vr) -{ - __u16 avail_idx = read_avail_idx(vr); - - while (avail_idx == le16toh(READ_ONCE(vr->vr.avail->idx))) { -#ifdef DEBUG - mpsslog("%s %s waiting for desc avail %d info_avail %d\n", - mic->name, __func__, - le16toh(vr->vr.avail->idx), vr->info->avail_idx); -#endif - sched_yield(); - } -} - -static void * -virtio_net(void *arg) -{ - static __u8 vnet_hdr[2][sizeof(struct virtio_net_hdr)]; - static __u8 vnet_buf[2][MAX_NET_PKT_SIZE] __attribute__ ((aligned(64))); - struct iovec vnet_iov[2][2] = { - { { .iov_base = vnet_hdr[0], .iov_len = sizeof(vnet_hdr[0]) }, - { .iov_base = vnet_buf[0], .iov_len = sizeof(vnet_buf[0]) } }, - { { .iov_base = vnet_hdr[1], .iov_len = sizeof(vnet_hdr[1]) }, - { .iov_base = vnet_buf[1], .iov_len = sizeof(vnet_buf[1]) } }, - }; - struct iovec *iov0 = vnet_iov[0], *iov1 = vnet_iov[1]; - struct mic_info *mic = (struct mic_info *)arg; - char if_name[IFNAMSIZ]; - struct pollfd net_poll[MAX_NET_FD]; - struct mic_vring tx_vr, rx_vr; - struct mic_copy_desc copy; - struct mic_device_desc *desc; - int err; - - snprintf(if_name, IFNAMSIZ, "mic%d", mic->id); - mic->mic_net.tap_fd = tun_alloc(mic, if_name); - if (mic->mic_net.tap_fd < 0) - goto done; - - if (tap_configure(mic, if_name)) - goto done; - mpsslog("MIC name %s id %d\n", mic->name, mic->id); - - net_poll[NET_FD_VIRTIO_NET].fd = mic->mic_net.virtio_net_fd; - net_poll[NET_FD_VIRTIO_NET].events = POLLIN; - net_poll[NET_FD_TUN].fd = mic->mic_net.tap_fd; - net_poll[NET_FD_TUN].events = POLLIN; - - if (MAP_FAILED == init_vr(mic, mic->mic_net.virtio_net_fd, - VIRTIO_ID_NET, &tx_vr, &rx_vr, - virtnet_dev_page.dd.num_vq)) { - mpsslog("%s init_vr failed %s\n", - mic->name, strerror(errno)); - goto done; - } - - copy.iovcnt = 2; - desc = get_device_desc(mic, VIRTIO_ID_NET); - - while (1) { - ssize_t len; - - net_poll[NET_FD_VIRTIO_NET].revents = 0; - net_poll[NET_FD_TUN].revents = 0; - - /* Start polling for data from tap and virtio net */ - err = poll(net_poll, 2, -1); - if (err < 0) { - mpsslog("%s poll failed %s\n", - __func__, strerror(errno)); - continue; - } - if (!(desc->status & VIRTIO_CONFIG_S_DRIVER_OK)) { - err = wait_for_card_driver(mic, - mic->mic_net.virtio_net_fd, - VIRTIO_ID_NET); - if (err) { - mpsslog("%s %s %d Exiting...\n", - mic->name, __func__, __LINE__); - break; - } - } - /* - * Check if there is data to be read from TUN and write to - * virtio net fd if there is. - */ - if (net_poll[NET_FD_TUN].revents & POLLIN) { - copy.iov = iov0; - len = readv(net_poll[NET_FD_TUN].fd, - copy.iov, copy.iovcnt); - if (len > 0) { - struct virtio_net_hdr *hdr - = (struct virtio_net_hdr *)vnet_hdr[0]; - - /* Disable checksums on the card since we are on - a reliable PCIe link */ - hdr->flags |= VIRTIO_NET_HDR_F_DATA_VALID; -#ifdef DEBUG - mpsslog("%s %s %d hdr->flags 0x%x ", mic->name, - __func__, __LINE__, hdr->flags); - mpsslog("copy.out_len %d hdr->gso_type 0x%x\n", - copy.out_len, hdr->gso_type); -#endif -#ifdef DEBUG - disp_iovec(mic, ©, __func__, __LINE__); - mpsslog("%s %s %d read from tap 0x%lx\n", - mic->name, __func__, __LINE__, - len); -#endif - spin_for_descriptors(mic, &tx_vr); - txrx_prepare(VIRTIO_ID_NET, 1, &tx_vr, ©, - len); - - err = mic_virtio_copy(mic, - mic->mic_net.virtio_net_fd, &tx_vr, - ©); - if (err < 0) { - mpsslog("%s %s %d mic_virtio_copy %s\n", - mic->name, __func__, __LINE__, - strerror(errno)); - } - if (!err) - verify_out_len(mic, ©); -#ifdef DEBUG - disp_iovec(mic, ©, __func__, __LINE__); - mpsslog("%s %s %d wrote to net 0x%lx\n", - mic->name, __func__, __LINE__, - sum_iovec_len(©)); -#endif - /* Reinitialize IOV for next run */ - iov0[1].iov_len = MAX_NET_PKT_SIZE; - } else if (len < 0) { - disp_iovec(mic, ©, __func__, __LINE__); - mpsslog("%s %s %d read failed %s ", mic->name, - __func__, __LINE__, strerror(errno)); - mpsslog("cnt %d sum %zd\n", - copy.iovcnt, sum_iovec_len(©)); - } - } - - /* - * Check if there is data to be read from virtio net and - * write to TUN if there is. - */ - if (net_poll[NET_FD_VIRTIO_NET].revents & POLLIN) { - while (rx_vr.info->avail_idx != - le16toh(rx_vr.vr.avail->idx)) { - copy.iov = iov1; - txrx_prepare(VIRTIO_ID_NET, 0, &rx_vr, ©, - MAX_NET_PKT_SIZE - + sizeof(struct virtio_net_hdr)); - - err = mic_virtio_copy(mic, - mic->mic_net.virtio_net_fd, &rx_vr, - ©); - if (!err) { -#ifdef DEBUG - struct virtio_net_hdr *hdr - = (struct virtio_net_hdr *) - vnet_hdr[1]; - - mpsslog("%s %s %d hdr->flags 0x%x, ", - mic->name, __func__, __LINE__, - hdr->flags); - mpsslog("out_len %d gso_type 0x%x\n", - copy.out_len, - hdr->gso_type); -#endif - /* Set the correct output iov_len */ - iov1[1].iov_len = copy.out_len - - sizeof(struct virtio_net_hdr); - verify_out_len(mic, ©); -#ifdef DEBUG - disp_iovec(mic, ©, __func__, - __LINE__); - mpsslog("%s %s %d ", - mic->name, __func__, __LINE__); - mpsslog("read from net 0x%lx\n", - sum_iovec_len(©)); -#endif - len = writev(net_poll[NET_FD_TUN].fd, - copy.iov, copy.iovcnt); - if (len != sum_iovec_len(©)) { - mpsslog("Tun write failed %s ", - strerror(errno)); - mpsslog("len 0x%zx ", len); - mpsslog("read_len 0x%zx\n", - sum_iovec_len(©)); - } else { -#ifdef DEBUG - disp_iovec(mic, ©, __func__, - __LINE__); - mpsslog("%s %s %d ", - mic->name, __func__, - __LINE__); - mpsslog("wrote to tap 0x%lx\n", - len); -#endif - } - } else { - mpsslog("%s %s %d mic_virtio_copy %s\n", - mic->name, __func__, __LINE__, - strerror(errno)); - break; - } - } - } - if (net_poll[NET_FD_VIRTIO_NET].revents & POLLERR) - mpsslog("%s: %s: POLLERR\n", __func__, mic->name); - } -done: - pthread_exit(NULL); -} - -/* virtio_console */ -#define VIRTIO_CONSOLE_FD 0 -#define MONITOR_FD (VIRTIO_CONSOLE_FD + 1) -#define MAX_CONSOLE_FD (MONITOR_FD + 1) /* must be the last one + 1 */ -#define MAX_BUFFER_SIZE PAGE_SIZE - -static void * -virtio_console(void *arg) -{ - static __u8 vcons_buf[2][PAGE_SIZE]; - struct iovec vcons_iov[2] = { - { .iov_base = vcons_buf[0], .iov_len = sizeof(vcons_buf[0]) }, - { .iov_base = vcons_buf[1], .iov_len = sizeof(vcons_buf[1]) }, - }; - struct iovec *iov0 = &vcons_iov[0], *iov1 = &vcons_iov[1]; - struct mic_info *mic = (struct mic_info *)arg; - int err; - struct pollfd console_poll[MAX_CONSOLE_FD]; - int pty_fd; - char *pts_name; - ssize_t len; - struct mic_vring tx_vr, rx_vr; - struct mic_copy_desc copy; - struct mic_device_desc *desc; - - pty_fd = posix_openpt(O_RDWR); - if (pty_fd < 0) { - mpsslog("can't open a pseudoterminal master device: %s\n", - strerror(errno)); - goto _return; - } - pts_name = ptsname(pty_fd); - if (pts_name == NULL) { - mpsslog("can't get pts name\n"); - goto _close_pty; - } - printf("%s console message goes to %s\n", mic->name, pts_name); - mpsslog("%s console message goes to %s\n", mic->name, pts_name); - err = grantpt(pty_fd); - if (err < 0) { - mpsslog("can't grant access: %s %s\n", - pts_name, strerror(errno)); - goto _close_pty; - } - err = unlockpt(pty_fd); - if (err < 0) { - mpsslog("can't unlock a pseudoterminal: %s %s\n", - pts_name, strerror(errno)); - goto _close_pty; - } - console_poll[MONITOR_FD].fd = pty_fd; - console_poll[MONITOR_FD].events = POLLIN; - - console_poll[VIRTIO_CONSOLE_FD].fd = mic->mic_console.virtio_console_fd; - console_poll[VIRTIO_CONSOLE_FD].events = POLLIN; - - if (MAP_FAILED == init_vr(mic, mic->mic_console.virtio_console_fd, - VIRTIO_ID_CONSOLE, &tx_vr, &rx_vr, - virtcons_dev_page.dd.num_vq)) { - mpsslog("%s init_vr failed %s\n", - mic->name, strerror(errno)); - goto _close_pty; - } - - copy.iovcnt = 1; - desc = get_device_desc(mic, VIRTIO_ID_CONSOLE); - - for (;;) { - console_poll[MONITOR_FD].revents = 0; - console_poll[VIRTIO_CONSOLE_FD].revents = 0; - err = poll(console_poll, MAX_CONSOLE_FD, -1); - if (err < 0) { - mpsslog("%s %d: poll failed: %s\n", __func__, __LINE__, - strerror(errno)); - continue; - } - if (!(desc->status & VIRTIO_CONFIG_S_DRIVER_OK)) { - err = wait_for_card_driver(mic, - mic->mic_console.virtio_console_fd, - VIRTIO_ID_CONSOLE); - if (err) { - mpsslog("%s %s %d Exiting...\n", - mic->name, __func__, __LINE__); - break; - } - } - - if (console_poll[MONITOR_FD].revents & POLLIN) { - copy.iov = iov0; - len = readv(pty_fd, copy.iov, copy.iovcnt); - if (len > 0) { -#ifdef DEBUG - disp_iovec(mic, ©, __func__, __LINE__); - mpsslog("%s %s %d read from tap 0x%lx\n", - mic->name, __func__, __LINE__, - len); -#endif - spin_for_descriptors(mic, &tx_vr); - txrx_prepare(VIRTIO_ID_CONSOLE, 1, &tx_vr, - ©, len); - - err = mic_virtio_copy(mic, - mic->mic_console.virtio_console_fd, - &tx_vr, ©); - if (err < 0) { - mpsslog("%s %s %d mic_virtio_copy %s\n", - mic->name, __func__, __LINE__, - strerror(errno)); - } - if (!err) - verify_out_len(mic, ©); -#ifdef DEBUG - disp_iovec(mic, ©, __func__, __LINE__); - mpsslog("%s %s %d wrote to net 0x%lx\n", - mic->name, __func__, __LINE__, - sum_iovec_len(©)); -#endif - /* Reinitialize IOV for next run */ - iov0->iov_len = PAGE_SIZE; - } else if (len < 0) { - disp_iovec(mic, ©, __func__, __LINE__); - mpsslog("%s %s %d read failed %s ", - mic->name, __func__, __LINE__, - strerror(errno)); - mpsslog("cnt %d sum %zd\n", - copy.iovcnt, sum_iovec_len(©)); - } - } - - if (console_poll[VIRTIO_CONSOLE_FD].revents & POLLIN) { - while (rx_vr.info->avail_idx != - le16toh(rx_vr.vr.avail->idx)) { - copy.iov = iov1; - txrx_prepare(VIRTIO_ID_CONSOLE, 0, &rx_vr, - ©, PAGE_SIZE); - - err = mic_virtio_copy(mic, - mic->mic_console.virtio_console_fd, - &rx_vr, ©); - if (!err) { - /* Set the correct output iov_len */ - iov1->iov_len = copy.out_len; - verify_out_len(mic, ©); -#ifdef DEBUG - disp_iovec(mic, ©, __func__, - __LINE__); - mpsslog("%s %s %d ", - mic->name, __func__, __LINE__); - mpsslog("read from net 0x%lx\n", - sum_iovec_len(©)); -#endif - len = writev(pty_fd, - copy.iov, copy.iovcnt); - if (len != sum_iovec_len(©)) { - mpsslog("Tun write failed %s ", - strerror(errno)); - mpsslog("len 0x%zx ", len); - mpsslog("read_len 0x%zx\n", - sum_iovec_len(©)); - } else { -#ifdef DEBUG - disp_iovec(mic, ©, __func__, - __LINE__); - mpsslog("%s %s %d ", - mic->name, __func__, - __LINE__); - mpsslog("wrote to tap 0x%lx\n", - len); -#endif - } - } else { - mpsslog("%s %s %d mic_virtio_copy %s\n", - mic->name, __func__, __LINE__, - strerror(errno)); - break; - } - } - } - if (console_poll[NET_FD_VIRTIO_NET].revents & POLLERR) - mpsslog("%s: %s: POLLERR\n", __func__, mic->name); - } -_close_pty: - close(pty_fd); -_return: - pthread_exit(NULL); -} - -static void -add_virtio_device(struct mic_info *mic, struct mic_device_desc *dd) -{ - char path[PATH_MAX]; - int fd, err; - - snprintf(path, PATH_MAX, "/dev/vop_virtio%d", mic->id); - fd = open(path, O_RDWR); - if (fd < 0) { - mpsslog("Could not open %s %s\n", path, strerror(errno)); - return; - } - - err = ioctl(fd, MIC_VIRTIO_ADD_DEVICE, dd); - if (err < 0) { - mpsslog("Could not add %d %s\n", dd->type, strerror(errno)); - close(fd); - return; - } - switch (dd->type) { - case VIRTIO_ID_NET: - mic->mic_net.virtio_net_fd = fd; - mpsslog("Added VIRTIO_ID_NET for %s\n", mic->name); - break; - case VIRTIO_ID_CONSOLE: - mic->mic_console.virtio_console_fd = fd; - mpsslog("Added VIRTIO_ID_CONSOLE for %s\n", mic->name); - break; - case VIRTIO_ID_BLOCK: - mic->mic_virtblk.virtio_block_fd = fd; - mpsslog("Added VIRTIO_ID_BLOCK for %s\n", mic->name); - break; - } -} - -static bool -set_backend_file(struct mic_info *mic) -{ - FILE *config; - char buff[PATH_MAX], *line, *evv, *p; - - snprintf(buff, PATH_MAX, "%s/mpssd%03d.conf", mic_config_dir, mic->id); - config = fopen(buff, "r"); - if (config == NULL) - return false; - do { /* look for "virtblk_backend=XXXX" */ - line = fgets(buff, PATH_MAX, config); - if (line == NULL) - break; - if (*line == '#') - continue; - p = strchr(line, '\n'); - if (p) - *p = '\0'; - } while (strncmp(line, virtblk_backend, strlen(virtblk_backend)) != 0); - fclose(config); - if (line == NULL) - return false; - evv = strchr(line, '='); - if (evv == NULL) - return false; - mic->mic_virtblk.backend_file = malloc(strlen(evv) + 1); - if (mic->mic_virtblk.backend_file == NULL) { - mpsslog("%s %d can't allocate memory\n", mic->name, mic->id); - return false; - } - strcpy(mic->mic_virtblk.backend_file, evv + 1); - return true; -} - -#define SECTOR_SIZE 512 -static bool -set_backend_size(struct mic_info *mic) -{ - mic->mic_virtblk.backend_size = lseek(mic->mic_virtblk.backend, 0, - SEEK_END); - if (mic->mic_virtblk.backend_size < 0) { - mpsslog("%s: can't seek: %s\n", - mic->name, mic->mic_virtblk.backend_file); - return false; - } - virtblk_dev_page.blk_config.capacity = - mic->mic_virtblk.backend_size / SECTOR_SIZE; - if ((mic->mic_virtblk.backend_size % SECTOR_SIZE) != 0) - virtblk_dev_page.blk_config.capacity++; - - virtblk_dev_page.blk_config.capacity = - htole64(virtblk_dev_page.blk_config.capacity); - - return true; -} - -static bool -open_backend(struct mic_info *mic) -{ - if (!set_backend_file(mic)) - goto _error_exit; - mic->mic_virtblk.backend = open(mic->mic_virtblk.backend_file, O_RDWR); - if (mic->mic_virtblk.backend < 0) { - mpsslog("%s: can't open: %s\n", mic->name, - mic->mic_virtblk.backend_file); - goto _error_free; - } - if (!set_backend_size(mic)) - goto _error_close; - mic->mic_virtblk.backend_addr = mmap(NULL, - mic->mic_virtblk.backend_size, - PROT_READ|PROT_WRITE, MAP_SHARED, - mic->mic_virtblk.backend, 0L); - if (mic->mic_virtblk.backend_addr == MAP_FAILED) { - mpsslog("%s: can't map: %s %s\n", - mic->name, mic->mic_virtblk.backend_file, - strerror(errno)); - goto _error_close; - } - return true; - - _error_close: - close(mic->mic_virtblk.backend); - _error_free: - free(mic->mic_virtblk.backend_file); - _error_exit: - return false; -} - -static void -close_backend(struct mic_info *mic) -{ - munmap(mic->mic_virtblk.backend_addr, mic->mic_virtblk.backend_size); - close(mic->mic_virtblk.backend); - free(mic->mic_virtblk.backend_file); -} - -static bool -start_virtblk(struct mic_info *mic, struct mic_vring *vring) -{ - if (((unsigned long)&virtblk_dev_page.blk_config % 8) != 0) { - mpsslog("%s: blk_config is not 8 byte aligned.\n", - mic->name); - return false; - } - add_virtio_device(mic, &virtblk_dev_page.dd); - if (MAP_FAILED == init_vr(mic, mic->mic_virtblk.virtio_block_fd, - VIRTIO_ID_BLOCK, vring, NULL, - virtblk_dev_page.dd.num_vq)) { - mpsslog("%s init_vr failed %s\n", - mic->name, strerror(errno)); - return false; - } - return true; -} - -static void -stop_virtblk(struct mic_info *mic) -{ - int vr_size, ret; - - vr_size = PAGE_ALIGN(_vring_size(MIC_VRING_ENTRIES, - MIC_VIRTIO_RING_ALIGN) + - sizeof(struct _mic_vring_info)); - ret = munmap(mic->mic_virtblk.block_dp, - MIC_DEVICE_PAGE_END + vr_size * virtblk_dev_page.dd.num_vq); - if (ret < 0) - mpsslog("%s munmap errno %d\n", mic->name, errno); - close(mic->mic_virtblk.virtio_block_fd); -} - -static __u8 -header_error_check(struct vring_desc *desc) -{ - if (le32toh(desc->len) != sizeof(struct virtio_blk_outhdr)) { - mpsslog("%s() %d: length is not sizeof(virtio_blk_outhd)\n", - __func__, __LINE__); - return -EIO; - } - if (!(le16toh(desc->flags) & VRING_DESC_F_NEXT)) { - mpsslog("%s() %d: alone\n", - __func__, __LINE__); - return -EIO; - } - if (le16toh(desc->flags) & VRING_DESC_F_WRITE) { - mpsslog("%s() %d: not read\n", - __func__, __LINE__); - return -EIO; - } - return 0; -} - -static int -read_header(int fd, struct virtio_blk_outhdr *hdr, __u32 desc_idx) -{ - struct iovec iovec; - struct mic_copy_desc copy; - - iovec.iov_len = sizeof(*hdr); - iovec.iov_base = hdr; - copy.iov = &iovec; - copy.iovcnt = 1; - copy.vr_idx = 0; /* only one vring on virtio_block */ - copy.update_used = false; /* do not update used index */ - return ioctl(fd, MIC_VIRTIO_COPY_DESC, ©); -} - -static int -transfer_blocks(int fd, struct iovec *iovec, __u32 iovcnt) -{ - struct mic_copy_desc copy; - - copy.iov = iovec; - copy.iovcnt = iovcnt; - copy.vr_idx = 0; /* only one vring on virtio_block */ - copy.update_used = false; /* do not update used index */ - return ioctl(fd, MIC_VIRTIO_COPY_DESC, ©); -} - -static __u8 -status_error_check(struct vring_desc *desc) -{ - if (le32toh(desc->len) != sizeof(__u8)) { - mpsslog("%s() %d: length is not sizeof(status)\n", - __func__, __LINE__); - return -EIO; - } - return 0; -} - -static int -write_status(int fd, __u8 *status) -{ - struct iovec iovec; - struct mic_copy_desc copy; - - iovec.iov_base = status; - iovec.iov_len = sizeof(*status); - copy.iov = &iovec; - copy.iovcnt = 1; - copy.vr_idx = 0; /* only one vring on virtio_block */ - copy.update_used = true; /* Update used index */ - return ioctl(fd, MIC_VIRTIO_COPY_DESC, ©); -} - -#ifndef VIRTIO_BLK_T_GET_ID -#define VIRTIO_BLK_T_GET_ID 8 -#endif - -static void * -virtio_block(void *arg) -{ - struct mic_info *mic = (struct mic_info *)arg; - int ret; - struct pollfd block_poll; - struct mic_vring vring; - __u16 avail_idx; - __u32 desc_idx; - struct vring_desc *desc; - struct iovec *iovec, *piov; - __u8 status; - __u32 buffer_desc_idx; - struct virtio_blk_outhdr hdr; - void *fos; - - for (;;) { /* forever */ - if (!open_backend(mic)) { /* No virtblk */ - for (mic->mic_virtblk.signaled = 0; - !mic->mic_virtblk.signaled;) - sleep(1); - continue; - } - - /* backend file is specified. */ - if (!start_virtblk(mic, &vring)) - goto _close_backend; - iovec = malloc(sizeof(*iovec) * - le32toh(virtblk_dev_page.blk_config.seg_max)); - if (!iovec) { - mpsslog("%s: can't alloc iovec: %s\n", - mic->name, strerror(ENOMEM)); - goto _stop_virtblk; - } - - block_poll.fd = mic->mic_virtblk.virtio_block_fd; - block_poll.events = POLLIN; - for (mic->mic_virtblk.signaled = 0; - !mic->mic_virtblk.signaled;) { - block_poll.revents = 0; - /* timeout in 1 sec to see signaled */ - ret = poll(&block_poll, 1, 1000); - if (ret < 0) { - mpsslog("%s %d: poll failed: %s\n", - __func__, __LINE__, - strerror(errno)); - continue; - } - - if (!(block_poll.revents & POLLIN)) { -#ifdef DEBUG - mpsslog("%s %d: block_poll.revents=0x%x\n", - __func__, __LINE__, block_poll.revents); -#endif - continue; - } - - /* POLLIN */ - while (vring.info->avail_idx != - le16toh(vring.vr.avail->idx)) { - /* read header element */ - avail_idx = - vring.info->avail_idx & - (vring.vr.num - 1); - desc_idx = le16toh( - vring.vr.avail->ring[avail_idx]); - desc = &vring.vr.desc[desc_idx]; -#ifdef DEBUG - mpsslog("%s() %d: avail_idx=%d ", - __func__, __LINE__, - vring.info->avail_idx); - mpsslog("vring.vr.num=%d desc=%p\n", - vring.vr.num, desc); -#endif - status = header_error_check(desc); - ret = read_header( - mic->mic_virtblk.virtio_block_fd, - &hdr, desc_idx); - if (ret < 0) { - mpsslog("%s() %d %s: ret=%d %s\n", - __func__, __LINE__, - mic->name, ret, - strerror(errno)); - break; - } - /* buffer element */ - piov = iovec; - status = 0; - fos = mic->mic_virtblk.backend_addr + - (hdr.sector * SECTOR_SIZE); - buffer_desc_idx = next_desc(desc); - desc_idx = buffer_desc_idx; - for (desc = &vring.vr.desc[buffer_desc_idx]; - desc->flags & VRING_DESC_F_NEXT; - desc_idx = next_desc(desc), - desc = &vring.vr.desc[desc_idx]) { - piov->iov_len = desc->len; - piov->iov_base = fos; - piov++; - fos += desc->len; - } - /* Returning NULLs for VIRTIO_BLK_T_GET_ID. */ - if (hdr.type & ~(VIRTIO_BLK_T_OUT | - VIRTIO_BLK_T_GET_ID)) { - /* - VIRTIO_BLK_T_IN - does not do - anything. Probably for documenting. - VIRTIO_BLK_T_SCSI_CMD - for - virtio_scsi. - VIRTIO_BLK_T_FLUSH - turned off in - config space. - VIRTIO_BLK_T_BARRIER - defined but not - used in anywhere. - */ - mpsslog("%s() %d: type %x ", - __func__, __LINE__, - hdr.type); - mpsslog("is not supported\n"); - status = -ENOTSUP; - - } else { - ret = transfer_blocks( - mic->mic_virtblk.virtio_block_fd, - iovec, - piov - iovec); - if (ret < 0 && - status != 0) - status = ret; - } - /* write status and update used pointer */ - if (status != 0) - status = status_error_check(desc); - ret = write_status( - mic->mic_virtblk.virtio_block_fd, - &status); -#ifdef DEBUG - mpsslog("%s() %d: write status=%d on desc=%p\n", - __func__, __LINE__, - status, desc); -#endif - } - } - free(iovec); -_stop_virtblk: - stop_virtblk(mic); -_close_backend: - close_backend(mic); - } /* forever */ - - pthread_exit(NULL); -} - -static void -reset(struct mic_info *mic) -{ -#define RESET_TIMEOUT 120 - int i = RESET_TIMEOUT; - setsysfs(mic->name, "state", "reset"); - while (i) { - char *state; - state = readsysfs(mic->name, "state"); - if (!state) - goto retry; - mpsslog("%s: %s %d state %s\n", - mic->name, __func__, __LINE__, state); - - if (!strcmp(state, "ready")) { - free(state); - break; - } - free(state); -retry: - sleep(1); - i--; - } -} - -static int -get_mic_shutdown_status(struct mic_info *mic, char *shutdown_status) -{ - if (!strcmp(shutdown_status, "nop")) - return MIC_NOP; - if (!strcmp(shutdown_status, "crashed")) - return MIC_CRASHED; - if (!strcmp(shutdown_status, "halted")) - return MIC_HALTED; - if (!strcmp(shutdown_status, "poweroff")) - return MIC_POWER_OFF; - if (!strcmp(shutdown_status, "restart")) - return MIC_RESTART; - mpsslog("%s: BUG invalid status %s\n", mic->name, shutdown_status); - /* Invalid state */ - assert(0); -}; - -static int get_mic_state(struct mic_info *mic) -{ - char *state = NULL; - enum mic_states mic_state; - - while (!state) { - state = readsysfs(mic->name, "state"); - sleep(1); - } - mpsslog("%s: %s %d state %s\n", - mic->name, __func__, __LINE__, state); - - if (!strcmp(state, "ready")) { - mic_state = MIC_READY; - } else if (!strcmp(state, "booting")) { - mic_state = MIC_BOOTING; - } else if (!strcmp(state, "online")) { - mic_state = MIC_ONLINE; - } else if (!strcmp(state, "shutting_down")) { - mic_state = MIC_SHUTTING_DOWN; - } else if (!strcmp(state, "reset_failed")) { - mic_state = MIC_RESET_FAILED; - } else if (!strcmp(state, "resetting")) { - mic_state = MIC_RESETTING; - } else { - mpsslog("%s: BUG invalid state %s\n", mic->name, state); - assert(0); - } - - free(state); - return mic_state; -}; - -static void mic_handle_shutdown(struct mic_info *mic) -{ -#define SHUTDOWN_TIMEOUT 60 - int i = SHUTDOWN_TIMEOUT; - char *shutdown_status; - while (i) { - shutdown_status = readsysfs(mic->name, "shutdown_status"); - if (!shutdown_status) { - sleep(1); - continue; - } - mpsslog("%s: %s %d shutdown_status %s\n", - mic->name, __func__, __LINE__, shutdown_status); - switch (get_mic_shutdown_status(mic, shutdown_status)) { - case MIC_RESTART: - mic->restart = 1; - case MIC_HALTED: - case MIC_POWER_OFF: - case MIC_CRASHED: - free(shutdown_status); - goto reset; - default: - break; - } - free(shutdown_status); - sleep(1); - i--; - } -reset: - if (!i) - mpsslog("%s: %s %d timing out waiting for shutdown_status %s\n", - mic->name, __func__, __LINE__, shutdown_status); - reset(mic); -} - -static int open_state_fd(struct mic_info *mic) -{ - char pathname[PATH_MAX]; - int fd; - - snprintf(pathname, PATH_MAX - 1, "%s/%s/%s", - MICSYSFSDIR, mic->name, "state"); - - fd = open(pathname, O_RDONLY); - if (fd < 0) - mpsslog("%s: opening file %s failed %s\n", - mic->name, pathname, strerror(errno)); - return fd; -} - -static int block_till_state_change(int fd, struct mic_info *mic) -{ - struct pollfd ufds[1]; - char value[PAGE_SIZE]; - int ret; - - ufds[0].fd = fd; - ufds[0].events = POLLERR | POLLPRI; - ret = poll(ufds, 1, -1); - if (ret < 0) { - mpsslog("%s: %s %d poll failed %s\n", - mic->name, __func__, __LINE__, strerror(errno)); - return ret; - } - - ret = lseek(fd, 0, SEEK_SET); - if (ret < 0) { - mpsslog("%s: %s %d Failed to seek to 0: %s\n", - mic->name, __func__, __LINE__, strerror(errno)); - return ret; - } - - ret = read(fd, value, sizeof(value)); - if (ret < 0) { - mpsslog("%s: %s %d Failed to read sysfs entry: %s\n", - mic->name, __func__, __LINE__, strerror(errno)); - return ret; - } - - return 0; -} - -static void * -mic_config(void *arg) -{ - struct mic_info *mic = (struct mic_info *)arg; - int fd, ret, stat = 0; - - fd = open_state_fd(mic); - if (fd < 0) { - mpsslog("%s: %s %d open state fd failed %s\n", - mic->name, __func__, __LINE__, strerror(errno)); - goto exit; - } - - do { - ret = block_till_state_change(fd, mic); - if (ret < 0) { - mpsslog("%s: %s %d block_till_state_change error %s\n", - mic->name, __func__, __LINE__, strerror(errno)); - goto close_exit; - } - - switch (get_mic_state(mic)) { - case MIC_SHUTTING_DOWN: - mic_handle_shutdown(mic); - break; - case MIC_READY: - case MIC_RESET_FAILED: - ret = kill(mic->pid, SIGTERM); - mpsslog("%s: %s %d kill pid %d ret %d\n", - mic->name, __func__, __LINE__, - mic->pid, ret); - if (!ret) { - ret = waitpid(mic->pid, &stat, - WIFSIGNALED(stat)); - mpsslog("%s: %s %d waitpid ret %d pid %d\n", - mic->name, __func__, __LINE__, - ret, mic->pid); - } - if (mic->boot_on_resume) { - setsysfs(mic->name, "state", "boot"); - mic->boot_on_resume = 0; - } - goto close_exit; - default: - break; - } - } while (1); - -close_exit: - close(fd); -exit: - init_mic(mic); - pthread_exit(NULL); -} - -static void -set_cmdline(struct mic_info *mic) -{ - char buffer[PATH_MAX]; - int len; - - len = snprintf(buffer, PATH_MAX, - "clocksource=tsc highres=off nohz=off "); - len += snprintf(buffer + len, PATH_MAX - len, - "cpufreq_on;corec6_off;pc3_off;pc6_off "); - len += snprintf(buffer + len, PATH_MAX - len, - "ifcfg=static;address,172.31.%d.1;netmask,255.255.255.0", - mic->id + 1); - - setsysfs(mic->name, "cmdline", buffer); - mpsslog("%s: Command line: \"%s\"\n", mic->name, buffer); - snprintf(buffer, PATH_MAX, "172.31.%d.1", mic->id + 1); - mpsslog("%s: IPADDR: \"%s\"\n", mic->name, buffer); -} - -static void -set_log_buf_info(struct mic_info *mic) -{ - int fd; - off_t len; - char system_map[] = "/lib/firmware/mic/System.map"; - char *map, *temp, log_buf[17] = {'\0'}; - - fd = open(system_map, O_RDONLY); - if (fd < 0) { - mpsslog("%s: Opening System.map failed: %d\n", - mic->name, errno); - return; - } - len = lseek(fd, 0, SEEK_END); - if (len < 0) { - mpsslog("%s: Reading System.map size failed: %d\n", - mic->name, errno); - close(fd); - return; - } - map = mmap(NULL, len, PROT_READ, MAP_PRIVATE, fd, 0); - if (map == MAP_FAILED) { - mpsslog("%s: mmap of System.map failed: %d\n", - mic->name, errno); - close(fd); - return; - } - temp = strstr(map, "__log_buf"); - if (!temp) { - mpsslog("%s: __log_buf not found: %d\n", mic->name, errno); - munmap(map, len); - close(fd); - return; - } - strncpy(log_buf, temp - 19, 16); - setsysfs(mic->name, "log_buf_addr", log_buf); - mpsslog("%s: log_buf_addr: %s\n", mic->name, log_buf); - temp = strstr(map, "log_buf_len"); - if (!temp) { - mpsslog("%s: log_buf_len not found: %d\n", mic->name, errno); - munmap(map, len); - close(fd); - return; - } - strncpy(log_buf, temp - 19, 16); - setsysfs(mic->name, "log_buf_len", log_buf); - mpsslog("%s: log_buf_len: %s\n", mic->name, log_buf); - munmap(map, len); - close(fd); -} - -static void -change_virtblk_backend(int x, siginfo_t *siginfo, void *p) -{ - struct mic_info *mic; - - for (mic = mic_list.next; mic != NULL; mic = mic->next) - mic->mic_virtblk.signaled = 1/* true */; -} - -static void -set_mic_boot_params(struct mic_info *mic) -{ - set_log_buf_info(mic); - set_cmdline(mic); -} - -static void * -init_mic(void *arg) -{ - struct mic_info *mic = (struct mic_info *)arg; - struct sigaction ignore = { - .sa_flags = 0, - .sa_handler = SIG_IGN - }; - struct sigaction act = { - .sa_flags = SA_SIGINFO, - .sa_sigaction = change_virtblk_backend, - }; - char buffer[PATH_MAX]; - int err, fd; - - /* - * Currently, one virtio block device is supported for each MIC card - * at a time. Any user (or test) can send a SIGUSR1 to the MIC daemon. - * The signal informs the virtio block backend about a change in the - * configuration file which specifies the virtio backend file name on - * the host. Virtio block backend then re-reads the configuration file - * and switches to the new block device. This signalling mechanism may - * not be required once multiple virtio block devices are supported by - * the MIC daemon. - */ - sigaction(SIGUSR1, &ignore, NULL); -retry: - fd = open_state_fd(mic); - if (fd < 0) { - mpsslog("%s: %s %d open state fd failed %s\n", - mic->name, __func__, __LINE__, strerror(errno)); - sleep(2); - goto retry; - } - - if (mic->restart) { - snprintf(buffer, PATH_MAX, "boot"); - setsysfs(mic->name, "state", buffer); - mpsslog("%s restarting mic %d\n", - mic->name, mic->restart); - mic->restart = 0; - } - - while (1) { - while (block_till_state_change(fd, mic)) { - mpsslog("%s: %s %d block_till_state_change error %s\n", - mic->name, __func__, __LINE__, strerror(errno)); - sleep(2); - continue; - } - - if (get_mic_state(mic) == MIC_BOOTING) - break; - } - - mic->pid = fork(); - switch (mic->pid) { - case 0: - add_virtio_device(mic, &virtcons_dev_page.dd); - add_virtio_device(mic, &virtnet_dev_page.dd); - err = pthread_create(&mic->mic_console.console_thread, NULL, - virtio_console, mic); - if (err) - mpsslog("%s virtcons pthread_create failed %s\n", - mic->name, strerror(err)); - err = pthread_create(&mic->mic_net.net_thread, NULL, - virtio_net, mic); - if (err) - mpsslog("%s virtnet pthread_create failed %s\n", - mic->name, strerror(err)); - err = pthread_create(&mic->mic_virtblk.block_thread, NULL, - virtio_block, mic); - if (err) - mpsslog("%s virtblk pthread_create failed %s\n", - mic->name, strerror(err)); - sigemptyset(&act.sa_mask); - err = sigaction(SIGUSR1, &act, NULL); - if (err) - mpsslog("%s sigaction SIGUSR1 failed %s\n", - mic->name, strerror(errno)); - while (1) - sleep(60); - case -1: - mpsslog("fork failed MIC name %s id %d errno %d\n", - mic->name, mic->id, errno); - break; - default: - err = pthread_create(&mic->config_thread, NULL, - mic_config, mic); - if (err) - mpsslog("%s mic_config pthread_create failed %s\n", - mic->name, strerror(err)); - } - - return NULL; -} - -static void -start_daemon(void) -{ - struct mic_info *mic; - int err; - - for (mic = mic_list.next; mic; mic = mic->next) { - set_mic_boot_params(mic); - err = pthread_create(&mic->init_thread, NULL, init_mic, mic); - if (err) - mpsslog("%s init_mic pthread_create failed %s\n", - mic->name, strerror(err)); - } - - while (1) - sleep(60); -} - -static int -init_mic_list(void) -{ - struct mic_info *mic = &mic_list; - struct dirent *file; - DIR *dp; - int cnt = 0; - - dp = opendir(MICSYSFSDIR); - if (!dp) - return 0; - - while ((file = readdir(dp)) != NULL) { - if (!strncmp(file->d_name, "mic", 3)) { - mic->next = calloc(1, sizeof(struct mic_info)); - if (mic->next) { - mic = mic->next; - mic->id = atoi(&file->d_name[3]); - mic->name = malloc(strlen(file->d_name) + 16); - if (mic->name) - strcpy(mic->name, file->d_name); - mpsslog("MIC name %s id %d\n", mic->name, - mic->id); - cnt++; - } - } - } - - closedir(dp); - return cnt; -} - -void -mpsslog(char *format, ...) -{ - va_list args; - char buffer[4096]; - char ts[52], *ts1; - time_t t; - - if (logfp == NULL) - return; - - va_start(args, format); - vsprintf(buffer, format, args); - va_end(args); - - time(&t); - ts1 = ctime_r(&t, ts); - ts1[strlen(ts1) - 1] = '\0'; - fprintf(logfp, "%s: %s", ts1, buffer); - - fflush(logfp); -} - -int -main(int argc, char *argv[]) -{ - int cnt; - pid_t pid; - - myname = argv[0]; - - logfp = fopen(LOGFILE_NAME, "a+"); - if (!logfp) { - fprintf(stderr, "cannot open logfile '%s'\n", LOGFILE_NAME); - exit(1); - } - pid = fork(); - switch (pid) { - case 0: - break; - case -1: - exit(2); - default: - exit(0); - } - - mpsslog("MIC Daemon start\n"); - - cnt = init_mic_list(); - if (cnt == 0) { - mpsslog("MIC module not loaded\n"); - exit(3); - } - mpsslog("MIC found %d devices\n", cnt); - - start_daemon(); - - exit(0); -} diff --git a/samples/mic/mpssd/mpssd.h b/samples/mic/mpssd/mpssd.h deleted file mode 100644 index 5f98bdafe653..000000000000 --- a/samples/mic/mpssd/mpssd.h +++ /dev/null @@ -1,89 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Intel MIC User Space Tools. - */ -#ifndef _MPSSD_H_ -#define _MPSSD_H_ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#define MICSYSFSDIR "/sys/class/mic" -#define LOGFILE_NAME "/var/log/mpssd" -#define PAGE_SIZE 4096 - -struct mic_console_info { - pthread_t console_thread; - int virtio_console_fd; - void *console_dp; -}; - -struct mic_net_info { - pthread_t net_thread; - int virtio_net_fd; - int tap_fd; - void *net_dp; -}; - -struct mic_virtblk_info { - pthread_t block_thread; - int virtio_block_fd; - void *block_dp; - volatile sig_atomic_t signaled; - char *backend_file; - int backend; - void *backend_addr; - long backend_size; -}; - -struct mic_info { - int id; - char *name; - pthread_t config_thread; - pthread_t init_thread; - pid_t pid; - struct mic_console_info mic_console; - struct mic_net_info mic_net; - struct mic_virtblk_info mic_virtblk; - int restart; - int boot_on_resume; - struct mic_info *next; -}; - -__attribute__((format(printf, 1, 2))) -void mpsslog(char *format, ...); -char *readsysfs(char *dir, char *entry); -int setsysfs(char *dir, char *entry, char *value); -#endif diff --git a/samples/mic/mpssd/sysfs.c b/samples/mic/mpssd/sysfs.c deleted file mode 100644 index 3fb08eb7ed9d..000000000000 --- a/samples/mic/mpssd/sysfs.c +++ /dev/null @@ -1,91 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * Intel MIC User Space Tools. - */ - -#include "mpssd.h" - -#define PAGE_SIZE 4096 - -char * -readsysfs(char *dir, char *entry) -{ - char filename[PATH_MAX]; - char value[PAGE_SIZE]; - char *string = NULL; - int fd; - int len; - - if (dir == NULL) - snprintf(filename, PATH_MAX, "%s/%s", MICSYSFSDIR, entry); - else - snprintf(filename, PATH_MAX, - "%s/%s/%s", MICSYSFSDIR, dir, entry); - - fd = open(filename, O_RDONLY); - if (fd < 0) { - mpsslog("Failed to open sysfs entry '%s': %s\n", - filename, strerror(errno)); - return NULL; - } - - len = read(fd, value, sizeof(value)); - if (len < 0) { - mpsslog("Failed to read sysfs entry '%s': %s\n", - filename, strerror(errno)); - goto readsys_ret; - } - if (len == 0) - goto readsys_ret; - - value[len - 1] = '\0'; - - string = malloc(strlen(value) + 1); - if (string) - strcpy(string, value); - -readsys_ret: - close(fd); - return string; -} - -int -setsysfs(char *dir, char *entry, char *value) -{ - char filename[PATH_MAX]; - char *oldvalue; - int fd, ret = 0; - - if (dir == NULL) - snprintf(filename, PATH_MAX, "%s/%s", MICSYSFSDIR, entry); - else - snprintf(filename, PATH_MAX, "%s/%s/%s", - MICSYSFSDIR, dir, entry); - - oldvalue = readsysfs(dir, entry); - - fd = open(filename, O_RDWR); - if (fd < 0) { - ret = errno; - mpsslog("Failed to open sysfs entry '%s': %s\n", - filename, strerror(errno)); - goto done; - } - - if (!oldvalue || strcmp(value, oldvalue)) { - if (write(fd, value, strlen(value)) < 0) { - ret = errno; - mpsslog("Failed to write new sysfs entry '%s': %s\n", - filename, strerror(errno)); - } - } - close(fd); -done: - if (oldvalue) - free(oldvalue); - return ret; -} -- cgit v1.2.3 From f27891abb7930d7a92bee2a47648a4a8708652d3 Mon Sep 17 00:00:00 2001 From: Amelie Delaunay Date: Wed, 28 Oct 2020 16:17:03 +0100 Subject: usb: typec: add missing MODULE_DEVICE_TABLE() to stusb160x When stusb160x driver is built as a module, no modalias information is available, and it prevents the module to be loaded by udev. Add MODULE_DEVICE_TABLE() to fix this issue. Fixes: da0cb6310094 ("usb: typec: add support for STUSB160x Type-C controller family") Signed-off-by: Amelie Delaunay Link: https://lore.kernel.org/r/20201028151703.31195-1-amelie.delaunay@st.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/stusb160x.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/typec/stusb160x.c b/drivers/usb/typec/stusb160x.c index da7f1957bcb3..42076697e26c 100644 --- a/drivers/usb/typec/stusb160x.c +++ b/drivers/usb/typec/stusb160x.c @@ -633,6 +633,7 @@ static const struct of_device_id stusb160x_of_match[] = { { .compatible = "st,stusb1600", .data = &stusb1600_regmap_config}, {}, }; +MODULE_DEVICE_TABLE(of, stusb160x_of_match); static int stusb160x_probe(struct i2c_client *client) { -- cgit v1.2.3 From befa1c2d65c639a7aefa793d36d7229732f14642 Mon Sep 17 00:00:00 2001 From: Amelie Delaunay Date: Wed, 28 Oct 2020 17:33:09 +0100 Subject: usb: typec: stusb160x: fix signedness comparison issue with enum variables chip->port_type and chip->pwr_opmode are enums and when GCC considers them as unsigned, the conditions are never met. This patch takes advantage of the ret variable and fixes the following warnings: drivers/usb/typec/stusb160x.c:548 stusb160x_get_fw_caps() warn: unsigned 'chip->port_type' is never less than zero. drivers/usb/typec/stusb160x.c:570 stusb160x_get_fw_caps() warn: unsigned 'chip->pwr_opmode' is never less than zero. Fixes: da0cb6310094 ("usb: typec: add support for STUSB160x Type-C controller family") Reported-by: kernel test robot Signed-off-by: Amelie Delaunay Link: https://lore.kernel.org/r/20201028163309.12878-1-amelie.delaunay@st.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/stusb160x.c | 20 ++++++++------------ 1 file changed, 8 insertions(+), 12 deletions(-) diff --git a/drivers/usb/typec/stusb160x.c b/drivers/usb/typec/stusb160x.c index 42076697e26c..2a618f02f4f1 100644 --- a/drivers/usb/typec/stusb160x.c +++ b/drivers/usb/typec/stusb160x.c @@ -544,11 +544,10 @@ static int stusb160x_get_fw_caps(struct stusb160x *chip, */ ret = fwnode_property_read_string(fwnode, "power-role", &cap_str); if (!ret) { - chip->port_type = typec_find_port_power_role(cap_str); - if ((int)chip->port_type < 0) { - ret = chip->port_type; + ret = typec_find_port_power_role(cap_str); + if (ret < 0) return ret; - } + chip->port_type = ret; } chip->capability.type = chip->port_type; @@ -565,16 +564,13 @@ static int stusb160x_get_fw_caps(struct stusb160x *chip, */ ret = fwnode_property_read_string(fwnode, "power-opmode", &cap_str); if (!ret) { - chip->pwr_opmode = typec_find_pwr_opmode(cap_str); + ret = typec_find_pwr_opmode(cap_str); /* Power delivery not yet supported */ - if ((int)chip->pwr_opmode < 0 || - chip->pwr_opmode == TYPEC_PWR_MODE_PD) { - ret = (int)chip->pwr_opmode < 0 ? chip->pwr_opmode : - -EINVAL; - dev_err(chip->dev, "bad power operation mode: %d\n", - chip->pwr_opmode); - return ret; + if (ret < 0 || ret == TYPEC_PWR_MODE_PD) { + dev_err(chip->dev, "bad power operation mode: %d\n", ret); + return -EINVAL; } + chip->pwr_opmode = ret; } return 0; -- cgit v1.2.3 From a75e2d604da8219d928be010bde37c3ab0ebd466 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Wed, 28 Oct 2020 22:31:22 +0200 Subject: xhci: Fix sizeof() mismatch An incorrect sizeof() is being used, sizeof(rhub->ports) is not correct, it should be sizeof(*rhub->ports). This bug did not cause any issues because it just so happens the sizes are the same. Fixes: bcaa9d5c5900 ("xhci: Create new structures to store xhci port information") Signed-off-by: Colin Ian King Signed-off-by: Mathias Nyman Link: https://lore.kernel.org/r/20201028203124.375344-2-mathias.nyman@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci-mem.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c index fe405cd38dbc..138ba4528dd3 100644 --- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -2252,8 +2252,8 @@ static void xhci_create_rhub_port_array(struct xhci_hcd *xhci, if (!rhub->num_ports) return; - rhub->ports = kcalloc_node(rhub->num_ports, sizeof(rhub->ports), flags, - dev_to_node(dev)); + rhub->ports = kcalloc_node(rhub->num_ports, sizeof(*rhub->ports), + flags, dev_to_node(dev)); for (i = 0; i < HCS_MAX_PORTS(xhci->hcs_params1); i++) { if (xhci->hw_ports[i].rhub != rhub || xhci->hw_ports[i].hcd_portnum == DUPLICATE_ENTRY) -- cgit v1.2.3 From 2a632815683d2d34df52b701a36fe5ac6654e719 Mon Sep 17 00:00:00 2001 From: Sandeep Singh Date: Wed, 28 Oct 2020 22:31:23 +0200 Subject: usb: xhci: Workaround for S3 issue on AMD SNPS 3.0 xHC On some platform of AMD, S3 fails with HCE and SRE errors. To fix this, need to disable a bit which is enable in sparse controller. Cc: stable@vger.kernel.org #v4.19+ Signed-off-by: Sanket Goswami Signed-off-by: Sandeep Singh Signed-off-by: Mathias Nyman Link: https://lore.kernel.org/r/20201028203124.375344-3-mathias.nyman@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci-pci.c | 17 +++++++++++++++++ drivers/usb/host/xhci.h | 1 + 2 files changed, 18 insertions(+) diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c index c26c06e5c88c..bf89172c43ca 100644 --- a/drivers/usb/host/xhci-pci.c +++ b/drivers/usb/host/xhci-pci.c @@ -23,6 +23,8 @@ #define SSIC_PORT_CFG2_OFFSET 0x30 #define PROG_DONE (1 << 30) #define SSIC_PORT_UNUSED (1 << 31) +#define SPARSE_DISABLE_BIT 17 +#define SPARSE_CNTL_ENABLE 0xC12C /* Device for a quirk */ #define PCI_VENDOR_ID_FRESCO_LOGIC 0x1b73 @@ -161,6 +163,9 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci) (pdev->device == 0x15e0 || pdev->device == 0x15e1)) xhci->quirks |= XHCI_SNPS_BROKEN_SUSPEND; + if (pdev->vendor == PCI_VENDOR_ID_AMD && pdev->device == 0x15e5) + xhci->quirks |= XHCI_DISABLE_SPARSE; + if (pdev->vendor == PCI_VENDOR_ID_AMD) xhci->quirks |= XHCI_TRUST_TX_LENGTH; @@ -498,6 +503,15 @@ static void xhci_pme_quirk(struct usb_hcd *hcd) readl(reg); } +static void xhci_sparse_control_quirk(struct usb_hcd *hcd) +{ + u32 reg; + + reg = readl(hcd->regs + SPARSE_CNTL_ENABLE); + reg &= ~BIT(SPARSE_DISABLE_BIT); + writel(reg, hcd->regs + SPARSE_CNTL_ENABLE); +} + static int xhci_pci_suspend(struct usb_hcd *hcd, bool do_wakeup) { struct xhci_hcd *xhci = hcd_to_xhci(hcd); @@ -517,6 +531,9 @@ static int xhci_pci_suspend(struct usb_hcd *hcd, bool do_wakeup) if (xhci->quirks & XHCI_SSIC_PORT_UNUSED) xhci_ssic_port_unused_quirk(hcd, true); + if (xhci->quirks & XHCI_DISABLE_SPARSE) + xhci_sparse_control_quirk(hcd); + ret = xhci_suspend(xhci, do_wakeup); if (ret && (xhci->quirks & XHCI_SSIC_PORT_UNUSED)) xhci_ssic_port_unused_quirk(hcd, false); diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h index 8be88379c0fb..ebb359ebb261 100644 --- a/drivers/usb/host/xhci.h +++ b/drivers/usb/host/xhci.h @@ -1877,6 +1877,7 @@ struct xhci_hcd { #define XHCI_SNPS_BROKEN_SUSPEND BIT_ULL(35) #define XHCI_RENESAS_FW_QUIRK BIT_ULL(36) #define XHCI_SKIP_PHY_INIT BIT_ULL(37) +#define XHCI_DISABLE_SPARSE BIT_ULL(38) unsigned int num_active_eps; unsigned int limit_active_eps; -- cgit v1.2.3 From 712da5fc4a66fbb1f144742e1b808498e932009b Mon Sep 17 00:00:00 2001 From: Mathias Nyman Date: Wed, 28 Oct 2020 22:31:24 +0200 Subject: xhci: Don't create stream debugfs files with spinlock held. Creating debugfs files while loding the spin_lock_irqsave(xhci->lock) creates a lock dependecy that could possibly deadlock. Lockdep warns: ===================================================== WARNING: HARDIRQ-safe -> HARDIRQ-unsafe lock order detected 5.10.0-rc1pdx86+ #8 Not tainted ----------------------------------------------------- systemd-udevd/386 [HC0[0]:SC0[0]:HE0:SE1] is trying to acquire: ffffffffb1a94038 (pin_fs_lock){+.+.}-{2:2}, at: simple_pin_fs+0x22/0xa0 and this task is already holding: ffff9e7b87fbc430 (&xhci->lock){-.-.}-{2:2}, at: xhci_alloc_streams+0x5f9/0x810 which would create a new lock dependency: (&xhci->lock){-.-.}-{2:2} -> (pin_fs_lock){+.+.}-{2:2} Create the files a bit later after lock is released. Fixes: 673d74683627 ("usb: xhci: add debugfs support for ep with stream") CC: Li Jun Reported-by: Hans de Goede Reported-by: Mike Galbraith Tested-by: Hans de Goede Signed-off-by: Mathias Nyman Link: https://lore.kernel.org/r/20201028203124.375344-4-mathias.nyman@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index 482fe8c5e3b4..d4a8d0efbbc4 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -3533,11 +3533,14 @@ static int xhci_alloc_streams(struct usb_hcd *hcd, struct usb_device *udev, xhci_dbg(xhci, "Slot %u ep ctx %u now has streams.\n", udev->slot_id, ep_index); vdev->eps[ep_index].ep_state |= EP_HAS_STREAMS; - xhci_debugfs_create_stream_files(xhci, vdev, ep_index); } xhci_free_command(xhci, config_cmd); spin_unlock_irqrestore(&xhci->lock, flags); + for (i = 0; i < num_eps; i++) { + ep_index = xhci_get_endpoint_index(&eps[i]->desc); + xhci_debugfs_create_stream_files(xhci, vdev, ep_index); + } /* Subtract 1 for stream 0, which drivers can't use */ return num_streams - 1; -- cgit v1.2.3 From 0f5cef2b2904b8d41d34ecb6c5d5bf2f00562b1b Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 26 Oct 2020 17:01:54 +0100 Subject: coresight: add module license When built as a loadable module, coresight now causes a warning about missing license information. WARNING: modpost: missing MODULE_LICENSE() in drivers/hwtracing/coresight/coresight.o Fixes: 8e264c52e1da ("coresight: core: Allow the coresight core driver to be built as a module") Signed-off-by: Arnd Bergmann Cc: Mathieu Poirier Link: https://lore.kernel.org/r/20201026160205.3704789-1-arnd@kernel.org Signed-off-by: Greg Kroah-Hartman --- drivers/hwtracing/coresight/coresight-core.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/hwtracing/coresight/coresight-core.c b/drivers/hwtracing/coresight/coresight-core.c index 6994c1309b2b..cc9e8025c533 100644 --- a/drivers/hwtracing/coresight/coresight-core.c +++ b/drivers/hwtracing/coresight/coresight-core.c @@ -1689,6 +1689,7 @@ static void __exit coresight_exit(void) module_init(coresight_init); module_exit(coresight_exit); +MODULE_LICENSE("GPL v2"); MODULE_AUTHOR("Pratik Patel "); MODULE_AUTHOR("Mathieu Poirier "); MODULE_DESCRIPTION("Arm CoreSight tracer driver"); -- cgit v1.2.3 From a3a0ded3ed38f4eafa86cef61799d50250da4dbe Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Wed, 8 Jul 2020 16:45:55 +0200 Subject: drm/vc4: hdmi: Add a name to the codec DAI component Since the components for a given device in ASoC are identified by their name, it makes sense to add one even though it's not strictly necessary. Signed-off-by: Maxime Ripard Reviewed-by: Dave Stevenson Link: https://patchwork.freedesktop.org/patch/msgid/20200708144555.718404-1-maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_hdmi.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c index 73ac52714e54..95779d50cca0 100644 --- a/drivers/gpu/drm/vc4/vc4_hdmi.c +++ b/drivers/gpu/drm/vc4/vc4_hdmi.c @@ -1077,6 +1077,7 @@ static const struct snd_soc_dapm_route vc4_hdmi_audio_routes[] = { }; static const struct snd_soc_component_driver vc4_hdmi_audio_component_drv = { + .name = "vc4-hdmi-codec-dai-component", .controls = vc4_hdmi_audio_controls, .num_controls = ARRAY_SIZE(vc4_hdmi_audio_controls), .dapm_widgets = vc4_hdmi_audio_widgets, -- cgit v1.2.3 From 5066f42c7d3c7766c16ba6e73da514af04d43ff7 Mon Sep 17 00:00:00 2001 From: Maxime Ripard Date: Wed, 28 Oct 2020 13:37:52 +0100 Subject: drm/vc4: Rework the structure conversion functions Most of the helpers to retrieve vc4 structures from the DRM base structures rely on the fact that the first member of the vc4 structure is the DRM one and just cast the pointers between them. However, this is pretty fragile especially since there's no check to make sure that the DRM structure is indeed at the offset 0 in the structure, so let's use container_of to make it more robust. Signed-off-by: Maxime Ripard Reviewed-by: Dave Stevenson Link: https://patchwork.freedesktop.org/patch/msgid/20201028123752.1733242-1-maxime@cerno.tech --- drivers/gpu/drm/vc4/vc4_drv.h | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h index 90b911fd2a7f..66d4fb16db8f 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.h +++ b/drivers/gpu/drm/vc4/vc4_drv.h @@ -287,7 +287,7 @@ struct vc4_bo { static inline struct vc4_bo * to_vc4_bo(struct drm_gem_object *bo) { - return (struct vc4_bo *)bo; + return container_of(to_drm_gem_cma_obj(bo), struct vc4_bo, base); } struct vc4_fence { @@ -300,7 +300,7 @@ struct vc4_fence { static inline struct vc4_fence * to_vc4_fence(struct dma_fence *fence) { - return (struct vc4_fence *)fence; + return container_of(fence, struct vc4_fence, base); } struct vc4_seqno_cb { @@ -347,7 +347,7 @@ struct vc4_plane { static inline struct vc4_plane * to_vc4_plane(struct drm_plane *plane) { - return (struct vc4_plane *)plane; + return container_of(plane, struct vc4_plane, base); } enum vc4_scaling_mode { @@ -423,7 +423,7 @@ struct vc4_plane_state { static inline struct vc4_plane_state * to_vc4_plane_state(struct drm_plane_state *state) { - return (struct vc4_plane_state *)state; + return container_of(state, struct vc4_plane_state, base); } enum vc4_encoder_type { @@ -499,7 +499,7 @@ struct vc4_crtc { static inline struct vc4_crtc * to_vc4_crtc(struct drm_crtc *crtc) { - return (struct vc4_crtc *)crtc; + return container_of(crtc, struct vc4_crtc, base); } static inline const struct vc4_crtc_data * @@ -537,7 +537,7 @@ struct vc4_crtc_state { static inline struct vc4_crtc_state * to_vc4_crtc_state(struct drm_crtc_state *crtc_state) { - return (struct vc4_crtc_state *)crtc_state; + return container_of(crtc_state, struct vc4_crtc_state, base); } #define V3D_READ(offset) readl(vc4->v3d->regs + offset) -- cgit v1.2.3 From 5fca3f062879f8e5214c56f3e3e2be6727900f5d Mon Sep 17 00:00:00 2001 From: Peter Chen Date: Thu, 22 Oct 2020 08:55:03 +0800 Subject: usb: cdns3: gadget: suspicious implicit sign extension The code: trb->length = cpu_to_le32(TRB_BURST_LEN(priv_ep->trb_burst_size) | TRB_LEN(length)); TRB_BURST_LEN(priv_ep->trb_burst_size) may be overflow for int 32 if priv_ep->trb_burst_size is equal or larger than 0x80; Below is the Coverity warning: sign_extension: Suspicious implicit sign extension: priv_ep->trb_burst_size with type u8 (8 bits, unsigned) is promoted in priv_ep->trb_burst_size << 24 to type int (32 bits, signed), then sign-extended to type unsigned long (64 bits, unsigned). If priv_ep->trb_burst_size << 24 is greater than 0x7FFFFFFF, the upper bits of the result will all be 1. To fix it, it needs to add an explicit cast to unsigned int type for ((p) << 24). Reviewed-by: Jun Li Signed-off-by: Peter Chen --- drivers/usb/cdns3/gadget.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/cdns3/gadget.h b/drivers/usb/cdns3/gadget.h index 1ccecd237530..737377913788 100644 --- a/drivers/usb/cdns3/gadget.h +++ b/drivers/usb/cdns3/gadget.h @@ -1072,7 +1072,7 @@ struct cdns3_trb { #define TRB_TDL_SS_SIZE_GET(p) (((p) & GENMASK(23, 17)) >> 17) /* transfer_len bitmasks - bits 31:24 */ -#define TRB_BURST_LEN(p) (((p) << 24) & GENMASK(31, 24)) +#define TRB_BURST_LEN(p) ((unsigned int)((p) << 24) & GENMASK(31, 24)) #define TRB_BURST_LEN_GET(p) (((p) & GENMASK(31, 24)) >> 24) /* Data buffer pointer bitmasks*/ -- cgit v1.2.3 From 8a6b88e66233f5f1779b0a1342aa9dc030dddcd5 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Wed, 28 Oct 2020 13:28:38 -0500 Subject: arm64: Add part number for Arm Cortex-A77 Add the MIDR part number info for the Arm Cortex-A77. Signed-off-by: Rob Herring Acked-by: Catalin Marinas Cc: Catalin Marinas Cc: Will Deacon Link: https://lore.kernel.org/r/20201028182839.166037-1-robh@kernel.org Signed-off-by: Will Deacon --- arch/arm64/include/asm/cputype.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm64/include/asm/cputype.h b/arch/arm64/include/asm/cputype.h index 7219cddeba66..9e2e9a63c7b6 100644 --- a/arch/arm64/include/asm/cputype.h +++ b/arch/arm64/include/asm/cputype.h @@ -71,6 +71,7 @@ #define ARM_CPU_PART_CORTEX_A55 0xD05 #define ARM_CPU_PART_CORTEX_A76 0xD0B #define ARM_CPU_PART_NEOVERSE_N1 0xD0C +#define ARM_CPU_PART_CORTEX_A77 0xD0D #define APM_CPU_PART_POTENZA 0x000 @@ -105,6 +106,7 @@ #define MIDR_CORTEX_A55 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A55) #define MIDR_CORTEX_A76 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A76) #define MIDR_NEOVERSE_N1 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_N1) +#define MIDR_CORTEX_A77 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A77) #define MIDR_THUNDERX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX) #define MIDR_THUNDERX_81XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_81XX) #define MIDR_THUNDERX_83XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_83XX) -- cgit v1.2.3 From 96d389ca10110d7eefb46feb6af9a0c6832f78f5 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Wed, 28 Oct 2020 13:28:39 -0500 Subject: arm64: Add workaround for Arm Cortex-A77 erratum 1508412 On Cortex-A77 r0p0 and r1p0, a sequence of a non-cacheable or device load and a store exclusive or PAR_EL1 read can cause a deadlock. The workaround requires a DMB SY before and after a PAR_EL1 register read. In addition, it's possible an interrupt (doing a device read) or KVM guest exit could be taken between the DMB and PAR read, so we also need a DMB before returning from interrupt and before returning to a guest. A deadlock is still possible with the workaround as KVM guests must also have the workaround. IOW, a malicious guest can deadlock an affected systems. This workaround also depends on a firmware counterpart to enable the h/w to insert DMB SY after load and store exclusive instructions. See the errata document SDEN-1152370 v10 [1] for more information. [1] https://static.docs.arm.com/101992/0010/Arm_Cortex_A77_MP074_Software_Developer_Errata_Notice_v10.pdf Signed-off-by: Rob Herring Reviewed-by: Catalin Marinas Acked-by: Marc Zyngier Cc: Catalin Marinas Cc: James Morse Cc: Suzuki K Poulose Cc: Will Deacon Cc: Julien Thierry Cc: kvmarm@lists.cs.columbia.edu Link: https://lore.kernel.org/r/20201028182839.166037-2-robh@kernel.org Signed-off-by: Will Deacon --- Documentation/arm64/silicon-errata.rst | 2 ++ arch/arm64/Kconfig | 20 ++++++++++++++++++++ arch/arm64/include/asm/cpucaps.h | 3 ++- arch/arm64/include/asm/sysreg.h | 9 +++++++++ arch/arm64/kernel/cpu_errata.c | 10 ++++++++++ arch/arm64/kernel/entry.S | 3 +++ arch/arm64/kvm/arm.c | 3 ++- arch/arm64/kvm/hyp/include/hyp/switch.h | 21 +++++++++++++-------- arch/arm64/kvm/hyp/include/hyp/sysreg-sr.h | 2 +- arch/arm64/kvm/hyp/nvhe/switch.c | 2 +- arch/arm64/kvm/hyp/vhe/switch.c | 2 +- arch/arm64/kvm/sys_regs.c | 2 +- arch/arm64/mm/fault.c | 2 +- 13 files changed, 66 insertions(+), 15 deletions(-) diff --git a/Documentation/arm64/silicon-errata.rst b/Documentation/arm64/silicon-errata.rst index d3587805de64..719510247292 100644 --- a/Documentation/arm64/silicon-errata.rst +++ b/Documentation/arm64/silicon-errata.rst @@ -90,6 +90,8 @@ stable kernels. +----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-A76 | #1463225 | ARM64_ERRATUM_1463225 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A77 | #1508412 | ARM64_ERRATUM_1508412 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-N1 | #1188873,1418040| ARM64_ERRATUM_1418040 | +----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-N1 | #1349291 | N/A | diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index f858c352f72a..1d466addb078 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -636,6 +636,26 @@ config ARM64_ERRATUM_1542419 If unsure, say Y. +config ARM64_ERRATUM_1508412 + bool "Cortex-A77: 1508412: workaround deadlock on sequence of NC/Device load and store exclusive or PAR read" + default y + help + This option adds a workaround for Arm Cortex-A77 erratum 1508412. + + Affected Cortex-A77 cores (r0p0, r1p0) could deadlock on a sequence + of a store-exclusive or read of PAR_EL1 and a load with device or + non-cacheable memory attributes. The workaround depends on a firmware + counterpart. + + KVM guests must also have the workaround implemented or they can + deadlock the system. + + Work around the issue by inserting DMB SY barriers around PAR_EL1 + register reads and warning KVM users. The DMB barrier is sufficient + to prevent a speculative PAR_EL1 read. + + If unsure, say Y. + config CAVIUM_ERRATUM_22375 bool "Cavium erratum 22375, 24313" default y diff --git a/arch/arm64/include/asm/cpucaps.h b/arch/arm64/include/asm/cpucaps.h index 42868dbd29fd..e7d98997c09c 100644 --- a/arch/arm64/include/asm/cpucaps.h +++ b/arch/arm64/include/asm/cpucaps.h @@ -65,7 +65,8 @@ #define ARM64_HAS_ARMv8_4_TTL 55 #define ARM64_HAS_TLB_RANGE 56 #define ARM64_MTE 57 +#define ARM64_WORKAROUND_1508412 58 -#define ARM64_NCAPS 58 +#define ARM64_NCAPS 59 #endif /* __ASM_CPUCAPS_H */ diff --git a/arch/arm64/include/asm/sysreg.h b/arch/arm64/include/asm/sysreg.h index d52c1b3ce589..174817ba119c 100644 --- a/arch/arm64/include/asm/sysreg.h +++ b/arch/arm64/include/asm/sysreg.h @@ -1007,6 +1007,7 @@ #include #include +#include #define __DEFINE_MRS_MSR_S_REGNUM \ " .irp num,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30\n" \ @@ -1095,6 +1096,14 @@ write_sysreg_s(__scs_new, sysreg); \ } while (0) +#define read_sysreg_par() ({ \ + u64 par; \ + asm(ALTERNATIVE("nop", "dmb sy", ARM64_WORKAROUND_1508412)); \ + par = read_sysreg(par_el1); \ + asm(ALTERNATIVE("nop", "dmb sy", ARM64_WORKAROUND_1508412)); \ + par; \ +}) + #endif #endif /* __ASM_SYSREG_H */ diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c index 24d75af344b1..61314fd70f13 100644 --- a/arch/arm64/kernel/cpu_errata.c +++ b/arch/arm64/kernel/cpu_errata.c @@ -522,6 +522,16 @@ const struct arm64_cpu_capabilities arm64_errata[] = { .matches = has_neoverse_n1_erratum_1542419, .cpu_enable = cpu_enable_trap_ctr_access, }, +#endif +#ifdef CONFIG_ARM64_ERRATUM_1508412 + { + /* we depend on the firmware portion for correctness */ + .desc = "ARM erratum 1508412 (kernel portion)", + .capability = ARM64_WORKAROUND_1508412, + ERRATA_MIDR_RANGE(MIDR_CORTEX_A77, + 0, 0, + 1, 0), + }, #endif { } diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S index f30007dff35f..b295fb912b12 100644 --- a/arch/arm64/kernel/entry.S +++ b/arch/arm64/kernel/entry.S @@ -365,6 +365,9 @@ alternative_insn eret, nop, ARM64_UNMAP_KERNEL_AT_EL0 br x30 #endif .else + /* Ensure any device/NC reads complete */ + alternative_insn nop, "dmb sy", ARM64_WORKAROUND_1508412 + eret .endif sb diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c index f56122eedffc..8f8fca47abfc 100644 --- a/arch/arm64/kvm/arm.c +++ b/arch/arm64/kvm/arm.c @@ -1719,7 +1719,8 @@ int kvm_arch_init(void *opaque) return -ENODEV; } - if (cpus_have_final_cap(ARM64_WORKAROUND_DEVICE_LOAD_ACQUIRE)) + if (cpus_have_final_cap(ARM64_WORKAROUND_DEVICE_LOAD_ACQUIRE) || + cpus_have_final_cap(ARM64_WORKAROUND_1508412)) kvm_info("Guests without required CPU erratum workarounds can deadlock system!\n" \ "Only trusted guests should be used on this system.\n"); diff --git a/arch/arm64/kvm/hyp/include/hyp/switch.h b/arch/arm64/kvm/hyp/include/hyp/switch.h index 313a8fa3c721..1f875a8f20c4 100644 --- a/arch/arm64/kvm/hyp/include/hyp/switch.h +++ b/arch/arm64/kvm/hyp/include/hyp/switch.h @@ -140,9 +140,9 @@ static inline bool __translate_far_to_hpfar(u64 far, u64 *hpfar) * We do need to save/restore PAR_EL1 though, as we haven't * saved the guest context yet, and we may return early... */ - par = read_sysreg(par_el1); + par = read_sysreg_par(); if (!__kvm_at("s1e1r", far)) - tmp = read_sysreg(par_el1); + tmp = read_sysreg_par(); else tmp = SYS_PAR_EL1_F; /* back to the guest */ write_sysreg(par, par_el1); @@ -421,7 +421,7 @@ static inline bool fixup_guest_exit(struct kvm_vcpu *vcpu, u64 *exit_code) if (cpus_have_final_cap(ARM64_WORKAROUND_CAVIUM_TX2_219_TVM) && kvm_vcpu_trap_get_class(vcpu) == ESR_ELx_EC_SYS64 && handle_tx2_tvm(vcpu)) - return true; + goto guest; /* * We trap the first access to the FP/SIMD to save the host context @@ -431,13 +431,13 @@ static inline bool fixup_guest_exit(struct kvm_vcpu *vcpu, u64 *exit_code) * Similarly for trapped SVE accesses. */ if (__hyp_handle_fpsimd(vcpu)) - return true; + goto guest; if (__hyp_handle_ptrauth(vcpu)) - return true; + goto guest; if (!__populate_fault_info(vcpu)) - return true; + goto guest; if (static_branch_unlikely(&vgic_v2_cpuif_trap)) { bool valid; @@ -452,7 +452,7 @@ static inline bool fixup_guest_exit(struct kvm_vcpu *vcpu, u64 *exit_code) int ret = __vgic_v2_perform_cpuif_access(vcpu); if (ret == 1) - return true; + goto guest; /* Promote an illegal access to an SError.*/ if (ret == -1) @@ -468,12 +468,17 @@ static inline bool fixup_guest_exit(struct kvm_vcpu *vcpu, u64 *exit_code) int ret = __vgic_v3_perform_cpuif_access(vcpu); if (ret == 1) - return true; + goto guest; } exit: /* Return to the host kernel and handle the exit */ return false; + +guest: + /* Re-enter the guest */ + asm(ALTERNATIVE("nop", "dmb sy", ARM64_WORKAROUND_1508412)); + return true; } static inline void __kvm_unexpected_el2_exception(void) diff --git a/arch/arm64/kvm/hyp/include/hyp/sysreg-sr.h b/arch/arm64/kvm/hyp/include/hyp/sysreg-sr.h index 7a986030145f..cce43bfe158f 100644 --- a/arch/arm64/kvm/hyp/include/hyp/sysreg-sr.h +++ b/arch/arm64/kvm/hyp/include/hyp/sysreg-sr.h @@ -43,7 +43,7 @@ static inline void __sysreg_save_el1_state(struct kvm_cpu_context *ctxt) ctxt_sys_reg(ctxt, CONTEXTIDR_EL1) = read_sysreg_el1(SYS_CONTEXTIDR); ctxt_sys_reg(ctxt, AMAIR_EL1) = read_sysreg_el1(SYS_AMAIR); ctxt_sys_reg(ctxt, CNTKCTL_EL1) = read_sysreg_el1(SYS_CNTKCTL); - ctxt_sys_reg(ctxt, PAR_EL1) = read_sysreg(par_el1); + ctxt_sys_reg(ctxt, PAR_EL1) = read_sysreg_par(); ctxt_sys_reg(ctxt, TPIDR_EL1) = read_sysreg(tpidr_el1); ctxt_sys_reg(ctxt, SP_EL1) = read_sysreg(sp_el1); diff --git a/arch/arm64/kvm/hyp/nvhe/switch.c b/arch/arm64/kvm/hyp/nvhe/switch.c index a457a0306e03..8ae8160bc93a 100644 --- a/arch/arm64/kvm/hyp/nvhe/switch.c +++ b/arch/arm64/kvm/hyp/nvhe/switch.c @@ -250,7 +250,7 @@ void __noreturn hyp_panic(void) { u64 spsr = read_sysreg_el2(SYS_SPSR); u64 elr = read_sysreg_el2(SYS_ELR); - u64 par = read_sysreg(par_el1); + u64 par = read_sysreg_par(); bool restore_host = true; struct kvm_cpu_context *host_ctxt; struct kvm_vcpu *vcpu; diff --git a/arch/arm64/kvm/hyp/vhe/switch.c b/arch/arm64/kvm/hyp/vhe/switch.c index fe69de16dadc..62546e20b251 100644 --- a/arch/arm64/kvm/hyp/vhe/switch.c +++ b/arch/arm64/kvm/hyp/vhe/switch.c @@ -215,7 +215,7 @@ void __noreturn hyp_panic(void) { u64 spsr = read_sysreg_el2(SYS_SPSR); u64 elr = read_sysreg_el2(SYS_ELR); - u64 par = read_sysreg(par_el1); + u64 par = read_sysreg_par(); __hyp_call_panic(spsr, elr, par); unreachable(); diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c index d9117bc56237..41348a7781d9 100644 --- a/arch/arm64/kvm/sys_regs.c +++ b/arch/arm64/kvm/sys_regs.c @@ -95,7 +95,7 @@ static bool __vcpu_read_sys_reg_from_cpu(int reg, u64 *val) case AMAIR_EL1: *val = read_sysreg_s(SYS_AMAIR_EL12); break; case CNTKCTL_EL1: *val = read_sysreg_s(SYS_CNTKCTL_EL12); break; case ELR_EL1: *val = read_sysreg_s(SYS_ELR_EL12); break; - case PAR_EL1: *val = read_sysreg_s(SYS_PAR_EL1); break; + case PAR_EL1: *val = read_sysreg_par(); break; case DACR32_EL2: *val = read_sysreg_s(SYS_DACR32_EL2); break; case IFSR32_EL2: *val = read_sysreg_s(SYS_IFSR32_EL2); break; case DBGVCR32_EL2: *val = read_sysreg_s(SYS_DBGVCR32_EL2); break; diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c index 94c99c1c19e3..1ee94002801f 100644 --- a/arch/arm64/mm/fault.c +++ b/arch/arm64/mm/fault.c @@ -262,7 +262,7 @@ static bool __kprobes is_spurious_el1_translation_fault(unsigned long addr, local_irq_save(flags); asm volatile("at s1e1r, %0" :: "r" (addr)); isb(); - par = read_sysreg(par_el1); + par = read_sysreg_par(); local_irq_restore(flags); /* -- cgit v1.2.3 From a62f68f5ca53ab61cba2f0a410d0add7a6d54a52 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Fri, 23 Oct 2020 17:35:46 +0200 Subject: cpufreq: Introduce cpufreq_driver_test_flags() Add a helper function to test the flags of the cpufreq driver in use againt a given flags mask. In particular, this will be needed to test the CPUFREQ_NEED_UPDATE_LIMITS cpufreq driver flag in the schedutil governor. Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/cpufreq.c | 12 ++++++++++++ include/linux/cpufreq.h | 1 + 2 files changed, 13 insertions(+) diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index ea58337fb65f..336b5e94cbc8 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1907,6 +1907,18 @@ void cpufreq_resume(void) } } +/** + * cpufreq_driver_test_flags - Test cpufreq driver's flags against given ones. + * @flags: Flags to test against the current cpufreq driver's flags. + * + * Assumes that the driver is there, so callers must ensure that this is the + * case. + */ +bool cpufreq_driver_test_flags(u16 flags) +{ + return !!(cpufreq_driver->flags & flags); +} + /** * cpufreq_get_current_driver - return current driver's name * diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index 038ed83aab41..1eaa04f1bae6 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -433,6 +433,7 @@ struct cpufreq_driver { int cpufreq_register_driver(struct cpufreq_driver *driver_data); int cpufreq_unregister_driver(struct cpufreq_driver *driver_data); +bool cpufreq_driver_test_flags(u16 flags); const char *cpufreq_get_current_driver(void); void *cpufreq_get_driver_data(void); -- cgit v1.2.3 From d1e7c2996e988866e7ceceb4641a0886885b7889 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 29 Oct 2020 12:12:46 +0100 Subject: cpufreq: schedutil: Always call driver if CPUFREQ_NEED_UPDATE_LIMITS is set Because sugov_update_next_freq() may skip a frequency update even if the need_freq_update flag has been set for the policy at hand, policy limits updates may not take effect as expected. For example, if the intel_pstate driver operates in the passive mode with HWP enabled, it needs to update the HWP min and max limits when the policy min and max limits change, respectively, but that may not happen if the target frequency does not change along with the limit at hand. In particular, if the policy min is changed first, causing the target frequency to be adjusted to it, and the policy max limit is changed later to the same value, the HWP max limit will not be updated to follow it as expected, because the target frequency is still equal to the policy min limit and it will not change until that limit is updated. To address this issue, modify get_next_freq() to let the driver callback run if the CPUFREQ_NEED_UPDATE_LIMITS cpufreq driver flag is set regardless of whether or not the new frequency to set is equal to the previous one. Fixes: f6ebbcf08f37 ("cpufreq: intel_pstate: Implement passive mode with HWP enabled") Reported-by: Zhang Rui Tested-by: Zhang Rui Cc: 5.9+ # 5.9+: 1c534352f47f cpufreq: Introduce CPUFREQ_NEED_UPDATE_LIMITS ... Cc: 5.9+ # 5.9+: a62f68f5ca53 cpufreq: Introduce cpufreq_driver_test_flags() Signed-off-by: Rafael J. Wysocki Acked-by: Viresh Kumar Signed-off-by: Rafael J. Wysocki --- kernel/sched/cpufreq_schedutil.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c index e254745a82cb..c03a5775d019 100644 --- a/kernel/sched/cpufreq_schedutil.c +++ b/kernel/sched/cpufreq_schedutil.c @@ -102,7 +102,8 @@ static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 time) static bool sugov_update_next_freq(struct sugov_policy *sg_policy, u64 time, unsigned int next_freq) { - if (sg_policy->next_freq == next_freq) + if (sg_policy->next_freq == next_freq && + !cpufreq_driver_test_flags(CPUFREQ_NEED_UPDATE_LIMITS)) return false; sg_policy->next_freq = next_freq; @@ -161,7 +162,8 @@ static unsigned int get_next_freq(struct sugov_policy *sg_policy, freq = map_util_freq(util, freq, max); - if (freq == sg_policy->cached_raw_freq && !sg_policy->need_freq_update) + if (freq == sg_policy->cached_raw_freq && !sg_policy->need_freq_update && + !cpufreq_driver_test_flags(CPUFREQ_NEED_UPDATE_LIMITS)) return sg_policy->next_freq; sg_policy->need_freq_update = false; -- cgit v1.2.3 From b40813ddcd6bf9f01d020804e4cb8febc480b9e4 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Wed, 28 Oct 2020 15:24:34 +0800 Subject: nbd: don't update block size after device is started Mounted NBD device can be resized, one use case is rbd-nbd. Fix the issue by setting up default block size, then not touch it in nbd_size_update() any more. This kind of usage is aligned with loop which has same use case too. Cc: stable@vger.kernel.org Fixes: c8a83a6b54d0 ("nbd: Use set_blocksize() to set device blocksize") Reported-by: lining Signed-off-by: Ming Lei Cc: Josef Bacik Cc: Jan Kara Tested-by: lining Signed-off-by: Jens Axboe --- drivers/block/nbd.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 0bed21c0c81b..c4f9ccf5cc2a 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -296,7 +296,7 @@ static void nbd_size_clear(struct nbd_device *nbd) } } -static void nbd_size_update(struct nbd_device *nbd) +static void nbd_size_update(struct nbd_device *nbd, bool start) { struct nbd_config *config = nbd->config; struct block_device *bdev = bdget_disk(nbd->disk, 0); @@ -313,7 +313,8 @@ static void nbd_size_update(struct nbd_device *nbd) if (bdev) { if (bdev->bd_disk) { bd_set_nr_sectors(bdev, nr_sectors); - set_blocksize(bdev, config->blksize); + if (start) + set_blocksize(bdev, config->blksize); } else set_bit(GD_NEED_PART_SCAN, &nbd->disk->state); bdput(bdev); @@ -328,7 +329,7 @@ static void nbd_size_set(struct nbd_device *nbd, loff_t blocksize, config->blksize = blocksize; config->bytesize = blocksize * nr_blocks; if (nbd->task_recv != NULL) - nbd_size_update(nbd); + nbd_size_update(nbd, false); } static void nbd_complete_rq(struct request *req) @@ -1308,7 +1309,7 @@ static int nbd_start_device(struct nbd_device *nbd) args->index = i; queue_work(nbd->recv_workq, &args->work); } - nbd_size_update(nbd); + nbd_size_update(nbd, true); return error; } -- cgit v1.2.3 From f9c9104288da543cd64f186f9e2fba389f415630 Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Thu, 29 Oct 2020 20:04:59 +0900 Subject: null_blk: Fix zone reset all tracing In the cae of the REQ_OP_ZONE_RESET_ALL operation, the command sector is ignored and the operation is applied to all sequential zones. For these commands, tracing the effect of the command using the command sector to determine the target zone is thus incorrect. Fix null_zone_mgmt() zone condition tracing in the case of REQ_OP_ZONE_RESET_ALL to apply tracing to all sequential zones that are not already empty. Fixes: 766c3297d7e1 ("null_blk: add trace in null_blk_zoned.c") Signed-off-by: Damien Le Moal Cc: stable@vger.kernel.org Signed-off-by: Jens Axboe --- drivers/block/null_blk_zoned.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/block/null_blk_zoned.c b/drivers/block/null_blk_zoned.c index 98056c88926b..b637b16a5f54 100644 --- a/drivers/block/null_blk_zoned.c +++ b/drivers/block/null_blk_zoned.c @@ -475,9 +475,14 @@ static blk_status_t null_zone_mgmt(struct nullb_cmd *cmd, enum req_opf op, switch (op) { case REQ_OP_ZONE_RESET_ALL: - for (i = dev->zone_nr_conv; i < dev->nr_zones; i++) - null_reset_zone(dev, &dev->zones[i]); - break; + for (i = dev->zone_nr_conv; i < dev->nr_zones; i++) { + zone = &dev->zones[i]; + if (zone->cond != BLK_ZONE_COND_EMPTY) { + null_reset_zone(dev, zone); + trace_nullb_zone_op(cmd, i, zone->cond); + } + } + return BLK_STS_OK; case REQ_OP_ZONE_RESET: ret = null_reset_zone(dev, zone); break; -- cgit v1.2.3 From aa1c09cb65e2ed17cb8e652bc7ec84e0af1229eb Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Thu, 29 Oct 2020 20:05:00 +0900 Subject: null_blk: Fix locking in zoned mode When the zoned mode is enabled in null_blk, Serializing read, write and zone management operations for each zone is necessary to protect device level information for managing zone resources (zone open and closed counters) as well as each zone condition and write pointer position. Commit 35bc10b2eafb ("null_blk: synchronization fix for zoned device") introduced a spinlock to implement this serialization. However, when memory backing is also enabled, GFP_NOIO memory allocations are executed under the spinlock, resulting in might_sleep() warnings. Furthermore, the zone_lock spinlock is locked/unlocked using spin_lock_irq/spin_unlock_irq, similarly to the memory backing code with the nullb->lock spinlock. This nested use of irq locks wrecks the irq enabled/disabled state. Fix all this by introducing a bitmap for per-zone lock, with locking implemented using wait_on_bit_lock_io() and clear_and_wake_up_bit(). This locking mechanism allows keeping a zone locked while executing null_process_cmd(), serializing all operations to the zone while allowing to sleep during memory backing allocation with GFP_NOIO. Device level zone resource management information is protected using a spinlock which is not held while executing null_process_cmd(); Fixes: 35bc10b2eafb ("null_blk: synchronization fix for zoned device") Signed-off-by: Damien Le Moal Signed-off-by: Jens Axboe --- drivers/block/null_blk.h | 3 +- drivers/block/null_blk_zoned.c | 104 +++++++++++++++++++++++++++++++---------- 2 files changed, 82 insertions(+), 25 deletions(-) diff --git a/drivers/block/null_blk.h b/drivers/block/null_blk.h index 3176b269b822..cfd00ad40355 100644 --- a/drivers/block/null_blk.h +++ b/drivers/block/null_blk.h @@ -47,7 +47,8 @@ struct nullb_device { unsigned int nr_zones_closed; struct blk_zone *zones; sector_t zone_size_sects; - spinlock_t zone_lock; + spinlock_t zone_dev_lock; + unsigned long *zone_locks; unsigned long size; /* device size in MB */ unsigned long completion_nsec; /* time in ns to complete a request */ diff --git a/drivers/block/null_blk_zoned.c b/drivers/block/null_blk_zoned.c index b637b16a5f54..8775acbb4f8f 100644 --- a/drivers/block/null_blk_zoned.c +++ b/drivers/block/null_blk_zoned.c @@ -1,5 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 #include +#include #include "null_blk.h" #define CREATE_TRACE_POINTS @@ -45,7 +46,13 @@ int null_init_zoned_dev(struct nullb_device *dev, struct request_queue *q) if (!dev->zones) return -ENOMEM; - spin_lock_init(&dev->zone_lock); + spin_lock_init(&dev->zone_dev_lock); + dev->zone_locks = bitmap_zalloc(dev->nr_zones, GFP_KERNEL); + if (!dev->zone_locks) { + kvfree(dev->zones); + return -ENOMEM; + } + if (dev->zone_nr_conv >= dev->nr_zones) { dev->zone_nr_conv = dev->nr_zones - 1; pr_info("changed the number of conventional zones to %u", @@ -124,15 +131,26 @@ int null_register_zoned_dev(struct nullb *nullb) void null_free_zoned_dev(struct nullb_device *dev) { + bitmap_free(dev->zone_locks); kvfree(dev->zones); } +static inline void null_lock_zone(struct nullb_device *dev, unsigned int zno) +{ + wait_on_bit_lock_io(dev->zone_locks, zno, TASK_UNINTERRUPTIBLE); +} + +static inline void null_unlock_zone(struct nullb_device *dev, unsigned int zno) +{ + clear_and_wake_up_bit(zno, dev->zone_locks); +} + int null_report_zones(struct gendisk *disk, sector_t sector, unsigned int nr_zones, report_zones_cb cb, void *data) { struct nullb *nullb = disk->private_data; struct nullb_device *dev = nullb->dev; - unsigned int first_zone, i; + unsigned int first_zone, i, zno; struct blk_zone zone; int error; @@ -143,17 +161,17 @@ int null_report_zones(struct gendisk *disk, sector_t sector, nr_zones = min(nr_zones, dev->nr_zones - first_zone); trace_nullb_report_zones(nullb, nr_zones); - for (i = 0; i < nr_zones; i++) { + zno = first_zone; + for (i = 0; i < nr_zones; i++, zno++) { /* * Stacked DM target drivers will remap the zone information by * modifying the zone information passed to the report callback. * So use a local copy to avoid corruption of the device zone * array. */ - spin_lock_irq(&dev->zone_lock); - memcpy(&zone, &dev->zones[first_zone + i], - sizeof(struct blk_zone)); - spin_unlock_irq(&dev->zone_lock); + null_lock_zone(dev, zno); + memcpy(&zone, &dev->zones[zno], sizeof(struct blk_zone)); + null_unlock_zone(dev, zno); error = cb(&zone, i, data); if (error) @@ -163,6 +181,10 @@ int null_report_zones(struct gendisk *disk, sector_t sector, return nr_zones; } +/* + * This is called in the case of memory backing from null_process_cmd() + * with the target zone already locked. + */ size_t null_zone_valid_read_len(struct nullb *nullb, sector_t sector, unsigned int len) { @@ -299,22 +321,27 @@ static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector, if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL) return null_process_cmd(cmd, REQ_OP_WRITE, sector, nr_sectors); + null_lock_zone(dev, zno); + spin_lock(&dev->zone_dev_lock); + switch (zone->cond) { case BLK_ZONE_COND_FULL: /* Cannot write to a full zone */ - return BLK_STS_IOERR; + ret = BLK_STS_IOERR; + goto unlock; case BLK_ZONE_COND_EMPTY: case BLK_ZONE_COND_CLOSED: ret = null_check_zone_resources(dev, zone); if (ret != BLK_STS_OK) - return ret; + goto unlock; break; case BLK_ZONE_COND_IMP_OPEN: case BLK_ZONE_COND_EXP_OPEN: break; default: /* Invalid zone condition */ - return BLK_STS_IOERR; + ret = BLK_STS_IOERR; + goto unlock; } /* @@ -330,11 +357,14 @@ static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector, else cmd->rq->__sector = sector; } else if (sector != zone->wp) { - return BLK_STS_IOERR; + ret = BLK_STS_IOERR; + goto unlock; } - if (zone->wp + nr_sectors > zone->start + zone->capacity) - return BLK_STS_IOERR; + if (zone->wp + nr_sectors > zone->start + zone->capacity) { + ret = BLK_STS_IOERR; + goto unlock; + } if (zone->cond == BLK_ZONE_COND_CLOSED) { dev->nr_zones_closed--; @@ -345,9 +375,11 @@ static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector, if (zone->cond != BLK_ZONE_COND_EXP_OPEN) zone->cond = BLK_ZONE_COND_IMP_OPEN; + spin_unlock(&dev->zone_dev_lock); ret = null_process_cmd(cmd, REQ_OP_WRITE, sector, nr_sectors); + spin_lock(&dev->zone_dev_lock); if (ret != BLK_STS_OK) - return ret; + goto unlock; zone->wp += nr_sectors; if (zone->wp == zone->start + zone->capacity) { @@ -357,7 +389,13 @@ static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector, dev->nr_zones_imp_open--; zone->cond = BLK_ZONE_COND_FULL; } - return BLK_STS_OK; + ret = BLK_STS_OK; + +unlock: + spin_unlock(&dev->zone_dev_lock); + null_unlock_zone(dev, zno); + + return ret; } static blk_status_t null_open_zone(struct nullb_device *dev, struct blk_zone *zone) @@ -468,21 +506,33 @@ static blk_status_t null_zone_mgmt(struct nullb_cmd *cmd, enum req_opf op, sector_t sector) { struct nullb_device *dev = cmd->nq->dev; - unsigned int zone_no = null_zone_no(dev, sector); - struct blk_zone *zone = &dev->zones[zone_no]; - blk_status_t ret = BLK_STS_OK; + unsigned int zone_no; + struct blk_zone *zone; + blk_status_t ret; size_t i; - switch (op) { - case REQ_OP_ZONE_RESET_ALL: + if (op == REQ_OP_ZONE_RESET_ALL) { for (i = dev->zone_nr_conv; i < dev->nr_zones; i++) { + null_lock_zone(dev, i); zone = &dev->zones[i]; if (zone->cond != BLK_ZONE_COND_EMPTY) { + spin_lock(&dev->zone_dev_lock); null_reset_zone(dev, zone); + spin_unlock(&dev->zone_dev_lock); trace_nullb_zone_op(cmd, i, zone->cond); } + null_unlock_zone(dev, i); } return BLK_STS_OK; + } + + zone_no = null_zone_no(dev, sector); + zone = &dev->zones[zone_no]; + + null_lock_zone(dev, zone_no); + spin_lock(&dev->zone_dev_lock); + + switch (op) { case REQ_OP_ZONE_RESET: ret = null_reset_zone(dev, zone); break; @@ -496,22 +546,27 @@ static blk_status_t null_zone_mgmt(struct nullb_cmd *cmd, enum req_opf op, ret = null_finish_zone(dev, zone); break; default: - return BLK_STS_NOTSUPP; + ret = BLK_STS_NOTSUPP; + break; } + spin_unlock(&dev->zone_dev_lock); + if (ret == BLK_STS_OK) trace_nullb_zone_op(cmd, zone_no, zone->cond); + null_unlock_zone(dev, zone_no); + return ret; } blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd, enum req_opf op, sector_t sector, sector_t nr_sectors) { - blk_status_t sts; struct nullb_device *dev = cmd->nq->dev; + unsigned int zno = null_zone_no(dev, sector); + blk_status_t sts; - spin_lock_irq(&dev->zone_lock); switch (op) { case REQ_OP_WRITE: sts = null_zone_write(cmd, sector, nr_sectors, false); @@ -527,9 +582,10 @@ blk_status_t null_process_zoned_cmd(struct nullb_cmd *cmd, enum req_opf op, sts = null_zone_mgmt(cmd, op, sector); break; default: + null_lock_zone(dev, zno); sts = null_process_cmd(cmd, op, sector, nr_sectors); + null_unlock_zone(dev, zno); } - spin_unlock_irq(&dev->zone_lock); return sts; } -- cgit v1.2.3 From fa04a40b169fcee615afbae97f71a09332993f64 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 21 Oct 2020 13:22:19 +0100 Subject: afs: Fix to take ref on page when PG_private is set Fix afs to take a ref on a page when it sets PG_private on it and to drop the ref when removing the flag. Note that in afs_write_begin(), a lot of the time, PG_private is already set on a page to which we're going to add some data. In such a case, we leave the bit set and mustn't increment the page count. As suggested by Matthew Wilcox, use attach/detach_page_private() where possible. Fixes: 31143d5d515e ("AFS: implement basic file write support") Reported-by: Matthew Wilcox (Oracle) Signed-off-by: David Howells Reviewed-by: Matthew Wilcox (Oracle) --- fs/afs/dir.c | 12 ++++-------- fs/afs/dir_edit.c | 6 ++---- fs/afs/file.c | 8 ++------ fs/afs/write.c | 18 ++++++++++-------- 4 files changed, 18 insertions(+), 26 deletions(-) diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 1d2e61e0ab04..1bb5b9d7f0a2 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -281,8 +281,7 @@ retry: if (ret < 0) goto error; - set_page_private(req->pages[i], 1); - SetPagePrivate(req->pages[i]); + attach_page_private(req->pages[i], (void *)1); unlock_page(req->pages[i]); i++; } else { @@ -1975,8 +1974,7 @@ static int afs_dir_releasepage(struct page *page, gfp_t gfp_flags) _enter("{{%llx:%llu}[%lu]}", dvnode->fid.vid, dvnode->fid.vnode, page->index); - set_page_private(page, 0); - ClearPagePrivate(page); + detach_page_private(page); /* The directory will need reloading. */ if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) @@ -2003,8 +2001,6 @@ static void afs_dir_invalidatepage(struct page *page, unsigned int offset, afs_stat_v(dvnode, n_inval); /* we clean up only if the entire page is being invalidated */ - if (offset == 0 && length == PAGE_SIZE) { - set_page_private(page, 0); - ClearPagePrivate(page); - } + if (offset == 0 && length == PAGE_SIZE) + detach_page_private(page); } diff --git a/fs/afs/dir_edit.c b/fs/afs/dir_edit.c index b108528bf010..2ffe09abae7f 100644 --- a/fs/afs/dir_edit.c +++ b/fs/afs/dir_edit.c @@ -243,10 +243,8 @@ void afs_edit_dir_add(struct afs_vnode *vnode, index, gfp); if (!page) goto error; - if (!PagePrivate(page)) { - set_page_private(page, 1); - SetPagePrivate(page); - } + if (!PagePrivate(page)) + attach_page_private(page, (void *)1); dir_page = kmap(page); } diff --git a/fs/afs/file.c b/fs/afs/file.c index 91225421ad37..322973d12614 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -627,11 +627,9 @@ static void afs_invalidatepage(struct page *page, unsigned int offset, #endif if (PagePrivate(page)) { - priv = page_private(page); + priv = (unsigned long)detach_page_private(page); trace_afs_page_dirty(vnode, tracepoint_string("inval"), page->index, priv); - set_page_private(page, 0); - ClearPagePrivate(page); } } @@ -661,11 +659,9 @@ static int afs_releasepage(struct page *page, gfp_t gfp_flags) #endif if (PagePrivate(page)) { - priv = page_private(page); + priv = (unsigned long)detach_page_private(page); trace_afs_page_dirty(vnode, tracepoint_string("rel"), page->index, priv); - set_page_private(page, 0); - ClearPagePrivate(page); } /* indicate that the page can be released */ diff --git a/fs/afs/write.c b/fs/afs/write.c index b937ec047ec9..02facb19a0f1 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -151,8 +151,10 @@ try_again: priv |= f; trace_afs_page_dirty(vnode, tracepoint_string("begin"), page->index, priv); - SetPagePrivate(page); - set_page_private(page, priv); + if (PagePrivate(page)) + set_page_private(page, priv); + else + attach_page_private(page, (void *)priv); _leave(" = 0"); return 0; @@ -334,10 +336,9 @@ static void afs_pages_written_back(struct afs_vnode *vnode, ASSERTCMP(pv.nr, ==, count); for (loop = 0; loop < count; loop++) { - priv = page_private(pv.pages[loop]); + priv = (unsigned long)detach_page_private(pv.pages[loop]); trace_afs_page_dirty(vnode, tracepoint_string("clear"), pv.pages[loop]->index, priv); - set_page_private(pv.pages[loop], 0); end_page_writeback(pv.pages[loop]); } first += count; @@ -863,8 +864,10 @@ vm_fault_t afs_page_mkwrite(struct vm_fault *vmf) priv |= 0; /* From */ trace_afs_page_dirty(vnode, tracepoint_string("mkwrite"), vmf->page->index, priv); - SetPagePrivate(vmf->page); - set_page_private(vmf->page, priv); + if (PagePrivate(vmf->page)) + set_page_private(vmf->page, priv); + else + attach_page_private(vmf->page, (void *)priv); file_update_time(file); sb_end_pagefault(inode->i_sb); @@ -926,10 +929,9 @@ int afs_launder_page(struct page *page) ret = afs_store_data(mapping, page->index, page->index, t, f, true); } + priv = (unsigned long)detach_page_private(page); trace_afs_page_dirty(vnode, tracepoint_string("laundered"), page->index, priv); - set_page_private(page, 0); - ClearPagePrivate(page); #ifdef CONFIG_AFS_FSCACHE if (PageFsCache(page)) { -- cgit v1.2.3 From 21db2cdc667f744691a407105b7712bc18d74023 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 22 Oct 2020 14:03:03 +0100 Subject: afs: Fix page leak on afs_write_begin() failure Fix the leak of the target page in afs_write_begin() when it fails. Fixes: 15b4650e55e0 ("afs: convert to new aops") Signed-off-by: David Howells cc: Nick Piggin --- fs/afs/write.c | 23 +++++++++++------------ 1 file changed, 11 insertions(+), 12 deletions(-) diff --git a/fs/afs/write.c b/fs/afs/write.c index 02facb19a0f1..7fae9f8b38eb 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -76,7 +76,7 @@ static int afs_fill_page(struct afs_vnode *vnode, struct key *key, */ int afs_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, - struct page **pagep, void **fsdata) + struct page **_page, void **fsdata) { struct afs_vnode *vnode = AFS_FS_I(file_inode(file)); struct page *page; @@ -110,9 +110,6 @@ int afs_write_begin(struct file *file, struct address_space *mapping, SetPageUptodate(page); } - /* page won't leak in error case: it eventually gets cleaned off LRU */ - *pagep = page; - try_again: /* See if this page is already partially written in a way that we can * merge the new write with. @@ -155,6 +152,7 @@ try_again: set_page_private(page, priv); else attach_page_private(page, (void *)priv); + *_page = page; _leave(" = 0"); return 0; @@ -164,17 +162,18 @@ try_again: flush_conflicting_write: _debug("flush conflict"); ret = write_one_page(page); - if (ret < 0) { - _leave(" = %d", ret); - return ret; - } + if (ret < 0) + goto error; ret = lock_page_killable(page); - if (ret < 0) { - _leave(" = %d", ret); - return ret; - } + if (ret < 0) + goto error; goto try_again; + +error: + put_page(page); + _leave(" = %d", ret); + return ret; } /* -- cgit v1.2.3 From f792e3ac82fe2c6c863e93187eb7ddfccab68fa7 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 26 Oct 2020 14:05:33 +0000 Subject: afs: Fix where page->private is set during write In afs, page->private is set to indicate the dirty region of a page. This is done in afs_write_begin(), but that can't take account of whether the copy into the page actually worked. Fix this by moving the change of page->private into afs_write_end(). Fixes: 4343d00872e1 ("afs: Get rid of the afs_writeback record") Signed-off-by: David Howells --- fs/afs/write.c | 41 ++++++++++++++++++++++++++--------------- 1 file changed, 26 insertions(+), 15 deletions(-) diff --git a/fs/afs/write.c b/fs/afs/write.c index 7fae9f8b38eb..f28d85c38cd8 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -135,23 +135,8 @@ try_again: if (!test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags) && (to < f || from > t)) goto flush_conflicting_write; - if (from < f) - f = from; - if (to > t) - t = to; - } else { - f = from; - t = to; } - priv = (unsigned long)t << AFS_PRIV_SHIFT; - priv |= f; - trace_afs_page_dirty(vnode, tracepoint_string("begin"), - page->index, priv); - if (PagePrivate(page)) - set_page_private(page, priv); - else - attach_page_private(page, (void *)priv); *_page = page; _leave(" = 0"); return 0; @@ -185,6 +170,9 @@ int afs_write_end(struct file *file, struct address_space *mapping, { struct afs_vnode *vnode = AFS_FS_I(file_inode(file)); struct key *key = afs_file_key(file); + unsigned long priv; + unsigned int f, from = pos & (PAGE_SIZE - 1); + unsigned int t, to = from + copied; loff_t i_size, maybe_i_size; int ret; @@ -216,6 +204,29 @@ int afs_write_end(struct file *file, struct address_space *mapping, SetPageUptodate(page); } + if (PagePrivate(page)) { + priv = page_private(page); + f = priv & AFS_PRIV_MAX; + t = priv >> AFS_PRIV_SHIFT; + if (from < f) + f = from; + if (to > t) + t = to; + priv = (unsigned long)t << AFS_PRIV_SHIFT; + priv |= f; + set_page_private(page, priv); + trace_afs_page_dirty(vnode, tracepoint_string("dirty+"), + page->index, priv); + } else { + f = from; + t = to; + priv = (unsigned long)t << AFS_PRIV_SHIFT; + priv |= f; + attach_page_private(page, (void *)priv); + trace_afs_page_dirty(vnode, tracepoint_string("dirty"), + page->index, priv); + } + set_page_dirty(page); if (PageDirty(page)) _debug("dirtied"); -- cgit v1.2.3 From 185f0c7073bd5c78f86265f703f5daf1306ab5a7 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 26 Oct 2020 13:22:47 +0000 Subject: afs: Wrap page->private manipulations in inline functions The afs filesystem uses page->private to store the dirty range within a page such that in the event of a conflicting 3rd-party write to the server, we write back just the bits that got changed locally. However, there are a couple of problems with this: (1) I need a bit to note if the page might be mapped so that partial invalidation doesn't shrink the range. (2) There aren't necessarily sufficient bits to store the entire range of data altered (say it's a 32-bit system with 64KiB pages or transparent huge pages are in use). So wrap the accesses in inline functions so that future commits can change how this works. Also move them out of the tracing header into the in-directory header. There's not really any need for them to be in the tracing header. Signed-off-by: David Howells --- fs/afs/internal.h | 28 ++++++++++++++++++++++++++++ fs/afs/write.c | 31 +++++++++++++------------------ include/trace/events/afs.h | 19 +++---------------- 3 files changed, 44 insertions(+), 34 deletions(-) diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 289f5dffa46f..edaccd07e18e 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -858,6 +858,34 @@ struct afs_vnode_cache_aux { u64 data_version; } __packed; +/* + * We use page->private to hold the amount of the page that we've written to, + * splitting the field into two parts. However, we need to represent a range + * 0...PAGE_SIZE inclusive, so we can't support 64K pages on a 32-bit system. + */ +#if PAGE_SIZE > 32768 +#define __AFS_PAGE_PRIV_MASK 0xffffffffUL +#define __AFS_PAGE_PRIV_SHIFT 32 +#else +#define __AFS_PAGE_PRIV_MASK 0xffffUL +#define __AFS_PAGE_PRIV_SHIFT 16 +#endif + +static inline size_t afs_page_dirty_from(unsigned long priv) +{ + return priv & __AFS_PAGE_PRIV_MASK; +} + +static inline size_t afs_page_dirty_to(unsigned long priv) +{ + return (priv >> __AFS_PAGE_PRIV_SHIFT) & __AFS_PAGE_PRIV_MASK; +} + +static inline unsigned long afs_page_dirty(size_t from, size_t to) +{ + return ((unsigned long)to << __AFS_PAGE_PRIV_SHIFT) | from; +} + #include /*****************************************************************************/ diff --git a/fs/afs/write.c b/fs/afs/write.c index f28d85c38cd8..ea1768b3c0b5 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -117,8 +117,8 @@ try_again: t = f = 0; if (PagePrivate(page)) { priv = page_private(page); - f = priv & AFS_PRIV_MAX; - t = priv >> AFS_PRIV_SHIFT; + f = afs_page_dirty_from(priv); + t = afs_page_dirty_to(priv); ASSERTCMP(f, <=, t); } @@ -206,22 +206,18 @@ int afs_write_end(struct file *file, struct address_space *mapping, if (PagePrivate(page)) { priv = page_private(page); - f = priv & AFS_PRIV_MAX; - t = priv >> AFS_PRIV_SHIFT; + f = afs_page_dirty_from(priv); + t = afs_page_dirty_to(priv); if (from < f) f = from; if (to > t) t = to; - priv = (unsigned long)t << AFS_PRIV_SHIFT; - priv |= f; + priv = afs_page_dirty(f, t); set_page_private(page, priv); trace_afs_page_dirty(vnode, tracepoint_string("dirty+"), page->index, priv); } else { - f = from; - t = to; - priv = (unsigned long)t << AFS_PRIV_SHIFT; - priv |= f; + priv = afs_page_dirty(from, to); attach_page_private(page, (void *)priv); trace_afs_page_dirty(vnode, tracepoint_string("dirty"), page->index, priv); @@ -522,8 +518,8 @@ static int afs_write_back_from_locked_page(struct address_space *mapping, */ start = primary_page->index; priv = page_private(primary_page); - offset = priv & AFS_PRIV_MAX; - to = priv >> AFS_PRIV_SHIFT; + offset = afs_page_dirty_from(priv); + to = afs_page_dirty_to(priv); trace_afs_page_dirty(vnode, tracepoint_string("store"), primary_page->index, priv); @@ -568,8 +564,8 @@ static int afs_write_back_from_locked_page(struct address_space *mapping, } priv = page_private(page); - f = priv & AFS_PRIV_MAX; - t = priv >> AFS_PRIV_SHIFT; + f = afs_page_dirty_from(priv); + t = afs_page_dirty_to(priv); if (f != 0 && !test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags)) { unlock_page(page); @@ -870,8 +866,7 @@ vm_fault_t afs_page_mkwrite(struct vm_fault *vmf) */ wait_on_page_writeback(vmf->page); - priv = (unsigned long)PAGE_SIZE << AFS_PRIV_SHIFT; /* To */ - priv |= 0; /* From */ + priv = afs_page_dirty(0, PAGE_SIZE); trace_afs_page_dirty(vnode, tracepoint_string("mkwrite"), vmf->page->index, priv); if (PagePrivate(vmf->page)) @@ -930,8 +925,8 @@ int afs_launder_page(struct page *page) f = 0; t = PAGE_SIZE; if (PagePrivate(page)) { - f = priv & AFS_PRIV_MAX; - t = priv >> AFS_PRIV_SHIFT; + f = afs_page_dirty_from(priv); + t = afs_page_dirty_to(priv); } trace_afs_page_dirty(vnode, tracepoint_string("launder"), diff --git a/include/trace/events/afs.h b/include/trace/events/afs.h index 8eb49231c6bb..866fc67d5aa5 100644 --- a/include/trace/events/afs.h +++ b/include/trace/events/afs.h @@ -966,19 +966,6 @@ TRACE_EVENT(afs_dir_check_failed, __entry->vnode, __entry->off, __entry->i_size) ); -/* - * We use page->private to hold the amount of the page that we've written to, - * splitting the field into two parts. However, we need to represent a range - * 0...PAGE_SIZE inclusive, so we can't support 64K pages on a 32-bit system. - */ -#if PAGE_SIZE > 32768 -#define AFS_PRIV_MAX 0xffffffff -#define AFS_PRIV_SHIFT 32 -#else -#define AFS_PRIV_MAX 0xffff -#define AFS_PRIV_SHIFT 16 -#endif - TRACE_EVENT(afs_page_dirty, TP_PROTO(struct afs_vnode *vnode, const char *where, pgoff_t page, unsigned long priv), @@ -999,10 +986,10 @@ TRACE_EVENT(afs_page_dirty, __entry->priv = priv; ), - TP_printk("vn=%p %lx %s %lu-%lu", + TP_printk("vn=%p %lx %s %zx-%zx", __entry->vnode, __entry->page, __entry->where, - __entry->priv & AFS_PRIV_MAX, - __entry->priv >> AFS_PRIV_SHIFT) + afs_page_dirty_from(__entry->priv), + afs_page_dirty_to(__entry->priv)) ); TRACE_EVENT(afs_call_state, -- cgit v1.2.3 From 65dd2d6072d393a3aa14ded8afa9a12f27d9c8ad Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 26 Oct 2020 13:57:44 +0000 Subject: afs: Alter dirty range encoding in page->private Currently, page->private on an afs page is used to store the range of dirtied data within the page, where the range includes the lower bound, but excludes the upper bound (e.g. 0-1 is a range covering a single byte). This, however, requires a superfluous bit for the last-byte bound so that on a 4KiB page, it can say 0-4096 to indicate the whole page, the idea being that having both numbers the same would indicate an empty range. This is unnecessary as the PG_private bit is clear if it's an empty range (as is PG_dirty). Alter the way the dirty range is encoded in page->private such that the upper bound is reduced by 1 (e.g. 0-0 is then specified the same single byte range mentioned above). Applying this to both bounds frees up two bits, one of which can be used in a future commit. This allows the afs filesystem to be compiled on ppc32 with 64K pages; without this, the following warnings are seen: ../fs/afs/internal.h: In function 'afs_page_dirty_to': ../fs/afs/internal.h:881:15: warning: right shift count >= width of type [-Wshift-count-overflow] 881 | return (priv >> __AFS_PAGE_PRIV_SHIFT) & __AFS_PAGE_PRIV_MASK; | ^~ ../fs/afs/internal.h: In function 'afs_page_dirty': ../fs/afs/internal.h:886:28: warning: left shift count >= width of type [-Wshift-count-overflow] 886 | return ((unsigned long)to << __AFS_PAGE_PRIV_SHIFT) | from; | ^~ Fixes: 4343d00872e1 ("afs: Get rid of the afs_writeback record") Signed-off-by: David Howells --- fs/afs/internal.h | 6 +++--- fs/afs/write.c | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/fs/afs/internal.h b/fs/afs/internal.h index edaccd07e18e..344c545f934c 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -863,7 +863,7 @@ struct afs_vnode_cache_aux { * splitting the field into two parts. However, we need to represent a range * 0...PAGE_SIZE inclusive, so we can't support 64K pages on a 32-bit system. */ -#if PAGE_SIZE > 32768 +#ifdef CONFIG_64BIT #define __AFS_PAGE_PRIV_MASK 0xffffffffUL #define __AFS_PAGE_PRIV_SHIFT 32 #else @@ -878,12 +878,12 @@ static inline size_t afs_page_dirty_from(unsigned long priv) static inline size_t afs_page_dirty_to(unsigned long priv) { - return (priv >> __AFS_PAGE_PRIV_SHIFT) & __AFS_PAGE_PRIV_MASK; + return ((priv >> __AFS_PAGE_PRIV_SHIFT) & __AFS_PAGE_PRIV_MASK) + 1; } static inline unsigned long afs_page_dirty(size_t from, size_t to) { - return ((unsigned long)to << __AFS_PAGE_PRIV_SHIFT) | from; + return ((unsigned long)(to - 1) << __AFS_PAGE_PRIV_SHIFT) | from; } #include diff --git a/fs/afs/write.c b/fs/afs/write.c index ea1768b3c0b5..1a49f5c89342 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -93,7 +93,7 @@ int afs_write_begin(struct file *file, struct address_space *mapping, /* We want to store information about how much of a page is altered in * page->private. */ - BUILD_BUG_ON(PAGE_SIZE > 32768 && sizeof(page->private) < 8); + BUILD_BUG_ON(PAGE_SIZE - 1 > __AFS_PAGE_PRIV_MASK && sizeof(page->private) < 8); page = grab_cache_page_write_begin(mapping, index, flags); if (!page) -- cgit v1.2.3 From f86726a69dec5df6ba051baf9265584419478b64 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 22 Oct 2020 14:08:23 +0100 Subject: afs: Fix afs_invalidatepage to adjust the dirty region Fix afs_invalidatepage() to adjust the dirty region recorded in page->private when truncating a page. If the dirty region is entirely removed, then the private data is cleared and the page dirty state is cleared. Without this, if the page is truncated and then expanded again by truncate, zeros from the expanded, but no-longer dirty region may get written back to the server if the page gets laundered due to a conflicting 3rd-party write. It mustn't, however, shorten the dirty region of the page if that page is still mmapped and has been marked dirty by afs_page_mkwrite(), so a flag is stored in page->private to record this. Fixes: 4343d00872e1 ("afs: Get rid of the afs_writeback record") Signed-off-by: David Howells --- fs/afs/file.c | 71 +++++++++++++++++++++++++++++++++++++++------- fs/afs/internal.h | 16 +++++++++-- fs/afs/write.c | 1 + include/trace/events/afs.h | 5 ++-- 4 files changed, 79 insertions(+), 14 deletions(-) diff --git a/fs/afs/file.c b/fs/afs/file.c index 322973d12614..85f5adf21aa0 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -601,6 +601,63 @@ static int afs_readpages(struct file *file, struct address_space *mapping, return ret; } +/* + * Adjust the dirty region of the page on truncation or full invalidation, + * getting rid of the markers altogether if the region is entirely invalidated. + */ +static void afs_invalidate_dirty(struct page *page, unsigned int offset, + unsigned int length) +{ + struct afs_vnode *vnode = AFS_FS_I(page->mapping->host); + unsigned long priv; + unsigned int f, t, end = offset + length; + + priv = page_private(page); + + /* we clean up only if the entire page is being invalidated */ + if (offset == 0 && length == thp_size(page)) + goto full_invalidate; + + /* If the page was dirtied by page_mkwrite(), the PTE stays writable + * and we don't get another notification to tell us to expand it + * again. + */ + if (afs_is_page_dirty_mmapped(priv)) + return; + + /* We may need to shorten the dirty region */ + f = afs_page_dirty_from(priv); + t = afs_page_dirty_to(priv); + + if (t <= offset || f >= end) + return; /* Doesn't overlap */ + + if (f < offset && t > end) + return; /* Splits the dirty region - just absorb it */ + + if (f >= offset && t <= end) + goto undirty; + + if (f < offset) + t = offset; + else + f = end; + if (f == t) + goto undirty; + + priv = afs_page_dirty(f, t); + set_page_private(page, priv); + trace_afs_page_dirty(vnode, tracepoint_string("trunc"), page->index, priv); + return; + +undirty: + trace_afs_page_dirty(vnode, tracepoint_string("undirty"), page->index, priv); + clear_page_dirty_for_io(page); +full_invalidate: + priv = (unsigned long)detach_page_private(page); + trace_afs_page_dirty(vnode, tracepoint_string("inval"), page->index, priv); +} + /* * invalidate part or all of a page * - release a page and clean up its private data if offset is 0 (indicating @@ -609,29 +666,23 @@ static int afs_readpages(struct file *file, struct address_space *mapping, static void afs_invalidatepage(struct page *page, unsigned int offset, unsigned int length) { - struct afs_vnode *vnode = AFS_FS_I(page->mapping->host); - unsigned long priv; - _enter("{%lu},%u,%u", page->index, offset, length); BUG_ON(!PageLocked(page)); +#ifdef CONFIG_AFS_FSCACHE /* we clean up only if the entire page is being invalidated */ if (offset == 0 && length == PAGE_SIZE) { -#ifdef CONFIG_AFS_FSCACHE if (PageFsCache(page)) { struct afs_vnode *vnode = AFS_FS_I(page->mapping->host); fscache_wait_on_page_write(vnode->cache, page); fscache_uncache_page(vnode->cache, page); } + } #endif - if (PagePrivate(page)) { - priv = (unsigned long)detach_page_private(page); - trace_afs_page_dirty(vnode, tracepoint_string("inval"), - page->index, priv); - } - } + if (PagePrivate(page)) + afs_invalidate_dirty(page, offset, length); _leave(""); } diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 344c545f934c..b0fce1f75397 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -864,11 +864,13 @@ struct afs_vnode_cache_aux { * 0...PAGE_SIZE inclusive, so we can't support 64K pages on a 32-bit system. */ #ifdef CONFIG_64BIT -#define __AFS_PAGE_PRIV_MASK 0xffffffffUL +#define __AFS_PAGE_PRIV_MASK 0x7fffffffUL #define __AFS_PAGE_PRIV_SHIFT 32 +#define __AFS_PAGE_PRIV_MMAPPED 0x80000000UL #else -#define __AFS_PAGE_PRIV_MASK 0xffffUL +#define __AFS_PAGE_PRIV_MASK 0x7fffUL #define __AFS_PAGE_PRIV_SHIFT 16 +#define __AFS_PAGE_PRIV_MMAPPED 0x8000UL #endif static inline size_t afs_page_dirty_from(unsigned long priv) @@ -886,6 +888,16 @@ static inline unsigned long afs_page_dirty(size_t from, size_t to) return ((unsigned long)(to - 1) << __AFS_PAGE_PRIV_SHIFT) | from; } +static inline unsigned long afs_page_dirty_mmapped(unsigned long priv) +{ + return priv | __AFS_PAGE_PRIV_MMAPPED; +} + +static inline bool afs_is_page_dirty_mmapped(unsigned long priv) +{ + return priv & __AFS_PAGE_PRIV_MMAPPED; +} + #include /*****************************************************************************/ diff --git a/fs/afs/write.c b/fs/afs/write.c index 1a49f5c89342..a2511e3ad2cc 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -867,6 +867,7 @@ vm_fault_t afs_page_mkwrite(struct vm_fault *vmf) wait_on_page_writeback(vmf->page); priv = afs_page_dirty(0, PAGE_SIZE); + priv = afs_page_dirty_mmapped(priv); trace_afs_page_dirty(vnode, tracepoint_string("mkwrite"), vmf->page->index, priv); if (PagePrivate(vmf->page)) diff --git a/include/trace/events/afs.h b/include/trace/events/afs.h index 866fc67d5aa5..4eef374d4413 100644 --- a/include/trace/events/afs.h +++ b/include/trace/events/afs.h @@ -986,10 +986,11 @@ TRACE_EVENT(afs_page_dirty, __entry->priv = priv; ), - TP_printk("vn=%p %lx %s %zx-%zx", + TP_printk("vn=%p %lx %s %zx-%zx%s", __entry->vnode, __entry->page, __entry->where, afs_page_dirty_from(__entry->priv), - afs_page_dirty_to(__entry->priv)) + afs_page_dirty_to(__entry->priv), + afs_is_page_dirty_mmapped(__entry->priv) ? " M" : "") ); TRACE_EVENT(afs_call_state, -- cgit v1.2.3 From 2d9900f26ad61e63a34f239bc76c80d2f8a6ff41 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 28 Oct 2020 12:08:39 +0000 Subject: afs: Fix dirty-region encoding on ppc32 with 64K pages The dirty region bounds stored in page->private on an afs page are 15 bits on a 32-bit box and can, at most, represent a range of up to 32K within a 32K page with a resolution of 1 byte. This is a problem for powerpc32 with 64K pages enabled. Further, transparent huge pages may get up to 2M, which will be a problem for the afs filesystem on all 32-bit arches in the future. Fix this by decreasing the resolution. For the moment, a 64K page will have a resolution determined from PAGE_SIZE. In the future, the page will need to be passed in to the helper functions so that the page size can be assessed and the resolution determined dynamically. Note that this might not be the ideal way to handle this, since it may allow some leakage of undirtied zero bytes to the server's copy in the case of a 3rd-party conflict. Fixing that would require a separately allocated record and is a more complicated fix. Fixes: 4343d00872e1 ("afs: Get rid of the afs_writeback record") Reported-by: kernel test robot Signed-off-by: David Howells Reviewed-by: Matthew Wilcox (Oracle) --- fs/afs/internal.h | 24 ++++++++++++++++++++---- fs/afs/write.c | 5 ----- 2 files changed, 20 insertions(+), 9 deletions(-) diff --git a/fs/afs/internal.h b/fs/afs/internal.h index b0fce1f75397..14d5d75f4b6e 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -861,7 +861,8 @@ struct afs_vnode_cache_aux { /* * We use page->private to hold the amount of the page that we've written to, * splitting the field into two parts. However, we need to represent a range - * 0...PAGE_SIZE inclusive, so we can't support 64K pages on a 32-bit system. + * 0...PAGE_SIZE, so we reduce the resolution if the size of the page + * exceeds what we can encode. */ #ifdef CONFIG_64BIT #define __AFS_PAGE_PRIV_MASK 0x7fffffffUL @@ -873,19 +874,34 @@ struct afs_vnode_cache_aux { #define __AFS_PAGE_PRIV_MMAPPED 0x8000UL #endif +static inline unsigned int afs_page_dirty_resolution(void) +{ + int shift = PAGE_SHIFT - (__AFS_PAGE_PRIV_SHIFT - 1); + return (shift > 0) ? shift : 0; +} + static inline size_t afs_page_dirty_from(unsigned long priv) { - return priv & __AFS_PAGE_PRIV_MASK; + unsigned long x = priv & __AFS_PAGE_PRIV_MASK; + + /* The lower bound is inclusive */ + return x << afs_page_dirty_resolution(); } static inline size_t afs_page_dirty_to(unsigned long priv) { - return ((priv >> __AFS_PAGE_PRIV_SHIFT) & __AFS_PAGE_PRIV_MASK) + 1; + unsigned long x = (priv >> __AFS_PAGE_PRIV_SHIFT) & __AFS_PAGE_PRIV_MASK; + + /* The upper bound is immediately beyond the region */ + return (x + 1) << afs_page_dirty_resolution(); } static inline unsigned long afs_page_dirty(size_t from, size_t to) { - return ((unsigned long)(to - 1) << __AFS_PAGE_PRIV_SHIFT) | from; + unsigned int res = afs_page_dirty_resolution(); + from >>= res; + to = (to - 1) >> res; + return (to << __AFS_PAGE_PRIV_SHIFT) | from; } static inline unsigned long afs_page_dirty_mmapped(unsigned long priv) diff --git a/fs/afs/write.c b/fs/afs/write.c index a2511e3ad2cc..50371207f327 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -90,11 +90,6 @@ int afs_write_begin(struct file *file, struct address_space *mapping, _enter("{%llx:%llu},{%lx},%u,%u", vnode->fid.vid, vnode->fid.vnode, index, from, to); - /* We want to store information about how much of a page is altered in - * page->private. - */ - BUILD_BUG_ON(PAGE_SIZE - 1 > __AFS_PAGE_PRIV_MASK && sizeof(page->private) < 8); - page = grab_cache_page_write_begin(mapping, index, flags); if (!page) return -ENOMEM; -- cgit v1.2.3 From 7cb6e22ba011c2dc1e94cc0dc94d9ef199715ed9 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Tue, 27 Oct 2020 19:11:30 +0200 Subject: xsysace: use platform_get_resource() and platform_get_irq_optional() Use platform_get_resource() to fetch the memory resource and platform_get_irq_optional() to get optional IRQ instead of open-coded variants. IRQ is not supposed to be changed at runtime, so there is no functional change in ace_fsm_yieldirq(). On the other hand we now take first resources instead of last ones to proceed. I can't imagine how broken should be firmware to have a garbage in the first resource slots. But if it the case, it needs to be documented. Signed-off-by: Andy Shevchenko Acked-by: Michal Simek Signed-off-by: Jens Axboe --- drivers/block/xsysace.c | 49 ++++++++++++++++++++++++++----------------------- 1 file changed, 26 insertions(+), 23 deletions(-) diff --git a/drivers/block/xsysace.c b/drivers/block/xsysace.c index 8d581c7536fb..eb8ef65778c3 100644 --- a/drivers/block/xsysace.c +++ b/drivers/block/xsysace.c @@ -443,22 +443,27 @@ static void ace_fix_driveid(u16 *id) #define ACE_FSM_NUM_STATES 11 /* Set flag to exit FSM loop and reschedule tasklet */ -static inline void ace_fsm_yield(struct ace_device *ace) +static inline void ace_fsm_yieldpoll(struct ace_device *ace) { - dev_dbg(ace->dev, "ace_fsm_yield()\n"); tasklet_schedule(&ace->fsm_tasklet); ace->fsm_continue_flag = 0; } +static inline void ace_fsm_yield(struct ace_device *ace) +{ + dev_dbg(ace->dev, "%s()\n", __func__); + ace_fsm_yieldpoll(ace); +} + /* Set flag to exit FSM loop and wait for IRQ to reschedule tasklet */ static inline void ace_fsm_yieldirq(struct ace_device *ace) { dev_dbg(ace->dev, "ace_fsm_yieldirq()\n"); - if (!ace->irq) - /* No IRQ assigned, so need to poll */ - tasklet_schedule(&ace->fsm_tasklet); - ace->fsm_continue_flag = 0; + if (ace->irq > 0) + ace->fsm_continue_flag = 0; + else + ace_fsm_yieldpoll(ace); } static bool ace_has_next_request(struct request_queue *q) @@ -1053,12 +1058,12 @@ static int ace_setup(struct ace_device *ace) ACE_CTRL_DATABUFRDYIRQ | ACE_CTRL_ERRORIRQ); /* Now we can hook up the irq handler */ - if (ace->irq) { + if (ace->irq > 0) { rc = request_irq(ace->irq, ace_interrupt, 0, "systemace", ace); if (rc) { /* Failure - fall back to polled mode */ dev_err(ace->dev, "request_irq failed\n"); - ace->irq = 0; + ace->irq = rc; } } @@ -1110,7 +1115,7 @@ static void ace_teardown(struct ace_device *ace) tasklet_kill(&ace->fsm_tasklet); - if (ace->irq) + if (ace->irq > 0) free_irq(ace->irq, ace); iounmap(ace->baseaddr); @@ -1123,11 +1128,6 @@ static int ace_alloc(struct device *dev, int id, resource_size_t physaddr, int rc; dev_dbg(dev, "ace_alloc(%p)\n", dev); - if (!physaddr) { - rc = -ENODEV; - goto err_noreg; - } - /* Allocate and initialize the ace device structure */ ace = kzalloc(sizeof(struct ace_device), GFP_KERNEL); if (!ace) { @@ -1153,7 +1153,6 @@ err_setup: dev_set_drvdata(dev, NULL); kfree(ace); err_alloc: -err_noreg: dev_err(dev, "could not initialize device, err=%i\n", rc); return rc; } @@ -1176,10 +1175,11 @@ static void ace_free(struct device *dev) static int ace_probe(struct platform_device *dev) { - resource_size_t physaddr = 0; int bus_width = ACE_BUS_WIDTH_16; /* FIXME: should not be hard coded */ + resource_size_t physaddr; + struct resource *res; u32 id = dev->id; - int irq = 0; + int irq; int i; dev_dbg(&dev->dev, "ace_probe(%p)\n", dev); @@ -1190,12 +1190,15 @@ static int ace_probe(struct platform_device *dev) if (of_find_property(dev->dev.of_node, "8-bit", NULL)) bus_width = ACE_BUS_WIDTH_8; - for (i = 0; i < dev->num_resources; i++) { - if (dev->resource[i].flags & IORESOURCE_MEM) - physaddr = dev->resource[i].start; - if (dev->resource[i].flags & IORESOURCE_IRQ) - irq = dev->resource[i].start; - } + res = platform_get_resource(dev, IORESOURCE_MEM, 0); + if (!res) + return -EINVAL; + + physaddr = res->start; + if (!physaddr) + return -ENODEV; + + irq = platform_get_irq_optional(dev, 0); /* Call the bus-independent setup code */ return ace_alloc(&dev->dev, id, physaddr, irq, bus_width); -- cgit v1.2.3 From 1f41be7d4e90e36084037cecba5978b3d7f849db Mon Sep 17 00:00:00 2001 From: David Disseldorp Date: Mon, 26 Oct 2020 22:03:10 +0100 Subject: lib/scatterlist: use consistent sg_copy_buffer() return type sg_copy_buffer() returns a size_t with the number of bytes copied. Return 0 instead of false if the copy is skipped. Signed-off-by: David Disseldorp Reviewed-by: Douglas Gilbert Reviewed-by: Johannes Thumshirn Signed-off-by: Jens Axboe --- lib/scatterlist.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/scatterlist.c b/lib/scatterlist.c index 0a482ef988e5..a59778946404 100644 --- a/lib/scatterlist.c +++ b/lib/scatterlist.c @@ -933,7 +933,7 @@ size_t sg_copy_buffer(struct scatterlist *sgl, unsigned int nents, void *buf, sg_miter_start(&miter, sgl, nents, sg_flags); if (!sg_miter_skip(&miter, skip)) - return false; + return 0; while ((offset < buflen) && sg_miter_next(&miter)) { unsigned int len; -- cgit v1.2.3 From 48ab6d5d1f096d6fac5b59f94af0aa394115a001 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 26 Oct 2020 16:27:55 +0100 Subject: dma-mapping: fix 32-bit overflow with CONFIG_ARM_LPAE=n On r8a7791/koelsch and shmobile_defconfig, PCIe probing fails with: rcar-pcie fe000000.pcie: Adjusted size 0x0 invalid rcar-pcie: probe of fe000000.pcie failed with error -22 of_dma_get_range() returns the following map: cpu_start 0x40000000 dma_start 0x40000000 size 0x080000000 offset 0 cpu_start 0x00000000 dma_start 0x00000000 size 0x100000000 offset 0 If CONFIG_ARM_LPAE=n, dma_addr_t is 32-bit. Hence when assigning r->dma_start + r->size to dma_end, this value will be truncated to 32-bit, yielding zero when processing the second table entry. Consequently, both dma_start and dma_end will be zero, leading to a zero size. Fix this by changing the dma_start and dma_end variables from dma_addr_t to u64. Fixes: e0d072782c734d27 ("dma-mapping: introduce DMA range map, supplanting dma_pfn_offset") Signed-off-by: Geert Uytterhoeven Signed-off-by: Christoph Hellwig --- drivers/of/device.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/of/device.c b/drivers/of/device.c index 655dee422563..3a469c79e6b0 100644 --- a/drivers/of/device.c +++ b/drivers/of/device.c @@ -93,7 +93,7 @@ int of_dma_configure_id(struct device *dev, struct device_node *np, { const struct iommu_ops *iommu; const struct bus_dma_region *map = NULL; - dma_addr_t dma_start = 0; + u64 dma_start = 0; u64 mask, end, size = 0; bool coherent; int ret; @@ -109,10 +109,10 @@ int of_dma_configure_id(struct device *dev, struct device_node *np, return ret == -ENODEV ? 0 : ret; } else { const struct bus_dma_region *r = map; - dma_addr_t dma_end = 0; + u64 dma_end = 0; /* Determine the overall bounds of all DMA regions */ - for (dma_start = ~(dma_addr_t)0; r->size; r++) { + for (dma_start = ~0ULL; r->size; r++) { /* Take lower and upper limits */ if (r->dma_start < dma_start) dma_start = r->dma_start; -- cgit v1.2.3 From 51467431200b91682b89d31317e35dcbca1469ce Mon Sep 17 00:00:00 2001 From: Masahiro Fujiwara Date: Tue, 27 Oct 2020 20:48:46 +0900 Subject: gtp: fix an use-before-init in gtp_newlink() *_pdp_find() from gtp_encap_recv() would trigger a crash when a peer sends GTP packets while creating new GTP device. RIP: 0010:gtp1_pdp_find.isra.0+0x68/0x90 [gtp] Call Trace: gtp_encap_recv+0xc2/0x2e0 [gtp] ? gtp1_pdp_find.isra.0+0x90/0x90 [gtp] udp_queue_rcv_one_skb+0x1fe/0x530 udp_queue_rcv_skb+0x40/0x1b0 udp_unicast_rcv_skb.isra.0+0x78/0x90 __udp4_lib_rcv+0x5af/0xc70 udp_rcv+0x1a/0x20 ip_protocol_deliver_rcu+0xc5/0x1b0 ip_local_deliver_finish+0x48/0x50 ip_local_deliver+0xe5/0xf0 ? ip_protocol_deliver_rcu+0x1b0/0x1b0 gtp_encap_enable() should be called after gtp_hastable_new() otherwise *_pdp_find() will access the uninitialized hash table. Fixes: 1e3a3abd8b28 ("gtp: make GTP sockets in gtp_newlink optional") Signed-off-by: Masahiro Fujiwara Link: https://lore.kernel.org/r/20201027114846.3924-1-fujiwara.masahiro@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/gtp.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/net/gtp.c b/drivers/net/gtp.c index 030a1a5afe05..dc668ed280b9 100644 --- a/drivers/net/gtp.c +++ b/drivers/net/gtp.c @@ -657,10 +657,6 @@ static int gtp_newlink(struct net *src_net, struct net_device *dev, gtp = netdev_priv(dev); - err = gtp_encap_enable(gtp, data); - if (err < 0) - return err; - if (!data[IFLA_GTP_PDP_HASHSIZE]) { hashsize = 1024; } else { @@ -671,12 +667,16 @@ static int gtp_newlink(struct net *src_net, struct net_device *dev, err = gtp_hashtable_new(gtp, hashsize); if (err < 0) - goto out_encap; + return err; + + err = gtp_encap_enable(gtp, data); + if (err < 0) + goto out_hashtable; err = register_netdevice(dev); if (err < 0) { netdev_dbg(dev, "failed to register new netdev %d\n", err); - goto out_hashtable; + goto out_encap; } gn = net_generic(dev_net(dev), gtp_net_id); @@ -687,11 +687,11 @@ static int gtp_newlink(struct net *src_net, struct net_device *dev, return 0; +out_encap: + gtp_encap_disable(gtp); out_hashtable: kfree(gtp->addr_hash); kfree(gtp->tid_hash); -out_encap: - gtp_encap_disable(gtp); return err; } -- cgit v1.2.3 From ceb1eb2fb609c88363e06618b8d4bbf7815a4e03 Mon Sep 17 00:00:00 2001 From: Tung Nguyen Date: Tue, 27 Oct 2020 10:24:03 +0700 Subject: tipc: fix memory leak caused by tipc_buf_append() Commit ed42989eab57 ("tipc: fix the skb_unshare() in tipc_buf_append()") replaced skb_unshare() with skb_copy() to not reduce the data reference counter of the original skb intentionally. This is not the correct way to handle the cloned skb because it causes memory leak in 2 following cases: 1/ Sending multicast messages via broadcast link The original skb list is cloned to the local skb list for local destination. After that, the data reference counter of each skb in the original list has the value of 2. This causes each skb not to be freed after receiving ACK: tipc_link_advance_transmq() { ... /* release skb */ __skb_unlink(skb, &l->transmq); kfree_skb(skb); <-- memory exists after being freed } 2/ Sending multicast messages via replicast link Similar to the above case, each skb cannot be freed after purging the skb list: tipc_mcast_xmit() { ... __skb_queue_purge(pkts); <-- memory exists after being freed } This commit fixes this issue by using skb_unshare() instead. Besides, to avoid use-after-free error reported by KASAN, the pointer to the fragment is set to NULL before calling skb_unshare() to make sure that the original skb is not freed after freeing the fragment 2 times in case skb_unshare() returns NULL. Fixes: ed42989eab57 ("tipc: fix the skb_unshare() in tipc_buf_append()") Acked-by: Jon Maloy Reported-by: Thang Hoang Ngo Signed-off-by: Tung Nguyen Reviewed-by: Xin Long Acked-by: Cong Wang Link: https://lore.kernel.org/r/20201027032403.1823-1-tung.q.nguyen@dektech.com.au Signed-off-by: Jakub Kicinski --- net/tipc/msg.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/net/tipc/msg.c b/net/tipc/msg.c index 2a78aa701572..32c79c59052b 100644 --- a/net/tipc/msg.c +++ b/net/tipc/msg.c @@ -150,12 +150,11 @@ int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf) if (fragid == FIRST_FRAGMENT) { if (unlikely(head)) goto err; - if (skb_cloned(frag)) - frag = skb_copy(frag, GFP_ATOMIC); + *buf = NULL; + frag = skb_unshare(frag, GFP_ATOMIC); if (unlikely(!frag)) goto err; head = *headbuf = frag; - *buf = NULL; TIPC_SKB_CB(head)->tail = NULL; if (skb_is_nonlinear(head)) { skb_walk_frags(head, tail) { -- cgit v1.2.3 From 4169e889e5889405d54cec27d6e9f7f0ce3c7096 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Wed, 2 Sep 2020 23:25:55 -0500 Subject: include: jhash/signal: Fix fall-through warnings for Clang In preparation to enable -Wimplicit-fallthrough for Clang, explicitly add break statements instead of letting the code fall through to the next case. This patch adds four break statements that, together, fix almost 40,000 warnings when building Linux 5.10-rc1 with Clang 12.0.0 and this[1] change reverted. Notice that in order to enable -Wimplicit-fallthrough for Clang, such change[1] is meant to be reverted at some point. So, this patch helps to move in that direction. Something important to mention is that there is currently a discrepancy between GCC and Clang when dealing with switch fall-through to empty case statements or to cases that only contain a break/continue/return statement[2][3][4]. Now that the -Wimplicit-fallthrough option has been globally enabled[5], any compiler should really warn on missing either a fallthrough annotation or any of the other case-terminating statements (break/continue/return/ goto) when falling through to the next case statement. Making exceptions to this introduces variation in case handling which may continue to lead to bugs, misunderstandings, and a general lack of robustness. The point of enabling options like -Wimplicit-fallthrough is to prevent human error and aid developers in spotting bugs before their code is even built/ submitted/committed, therefore eliminating classes of bugs. So, in order to really accomplish this, we should, and can, move in the direction of addressing any error-prone scenarios and get rid of the unintentional fallthrough bug-class in the kernel, entirely, even if there is some minor redundancy. Better to have explicit case-ending statements than continue to have exceptions where one must guess as to the right result. The compiler will eliminate any actual redundancy. [1] commit e2079e93f562c ("kbuild: Do not enable -Wimplicit-fallthrough for clang for now") [2] https://github.com/ClangBuiltLinux/linux/issues/636 [3] https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91432 [4] https://godbolt.org/z/xgkvIh [5] commit a035d552a93b ("Makefile: Globally enable fall-through warning") Co-developed-by: Kees Cook Signed-off-by: Kees Cook Signed-off-by: Gustavo A. R. Silva --- include/linux/jhash.h | 2 ++ include/linux/signal.h | 2 ++ 2 files changed, 4 insertions(+) diff --git a/include/linux/jhash.h b/include/linux/jhash.h index cfb62e9f37be..ab7f8c152b89 100644 --- a/include/linux/jhash.h +++ b/include/linux/jhash.h @@ -99,6 +99,7 @@ static inline u32 jhash(const void *key, u32 length, u32 initval) case 2: a += (u32)k[1]<<8; fallthrough; case 1: a += k[0]; __jhash_final(a, b, c); + break; case 0: /* Nothing left to add */ break; } @@ -136,6 +137,7 @@ static inline u32 jhash2(const u32 *k, u32 length, u32 initval) case 2: b += k[1]; fallthrough; case 1: a += k[0]; __jhash_final(a, b, c); + break; case 0: /* Nothing left to add */ break; } diff --git a/include/linux/signal.h b/include/linux/signal.h index 7bbc0e9cf084..b256f9c65661 100644 --- a/include/linux/signal.h +++ b/include/linux/signal.h @@ -238,6 +238,7 @@ static inline void siginitset(sigset_t *set, unsigned long mask) memset(&set->sig[1], 0, sizeof(long)*(_NSIG_WORDS-1)); break; case 2: set->sig[1] = 0; + break; case 1: ; } } @@ -250,6 +251,7 @@ static inline void siginitsetinv(sigset_t *set, unsigned long mask) memset(&set->sig[1], -1, sizeof(long)*(_NSIG_WORDS-1)); break; case 2: set->sig[1] = -1; + break; case 1: ; } } -- cgit v1.2.3 From 09a729b178be1cb20ffce620f7e5bfc04c4a13c0 Mon Sep 17 00:00:00 2001 From: Matthew Auld Date: Wed, 21 Oct 2020 11:36:06 +0100 Subject: drm/i915/region: fix max size calculation We are incorrectly limiting the max allocation size as per the mm max_order, which is effectively the largest power-of-two that we can fit in the region size. However, it's normal to setup the region or allocator with a non-power-of-two size(for example 3G), which we should already handle correctly, except it seems for the early too-big-check. v2: make sure we also exercise the I915_BO_ALLOC_CONTIGUOUS path, which is quite different, since for that we are actually limited by the largest power-of-two that we can fit within the region size. (Chris) Fixes: b908be543e44 ("drm/i915: support creating LMEM objects") Signed-off-by: Matthew Auld Cc: Chris Wilson Cc: CQ Tang Reviewed-by: Chris Wilson Signed-off-by: Chris Wilson Link: https://patchwork.freedesktop.org/patch/msgid/20201021103606.241395-1-matthew.auld@intel.com (cherry picked from commit 83ebef47f8ebe320d5c5673db82f9903a4f40a69) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/i915/intel_memory_region.c | 2 +- .../gpu/drm/i915/selftests/intel_memory_region.c | 77 ++++++++++++++++++++++ drivers/gpu/drm/i915/selftests/mock_region.c | 2 +- 3 files changed, 79 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_memory_region.c b/drivers/gpu/drm/i915/intel_memory_region.c index 6b5e9d88646d..180e1078ef7c 100644 --- a/drivers/gpu/drm/i915/intel_memory_region.c +++ b/drivers/gpu/drm/i915/intel_memory_region.c @@ -87,7 +87,7 @@ __intel_memory_region_get_pages_buddy(struct intel_memory_region *mem, min_order = ilog2(size) - ilog2(mem->mm.chunk_size); } - if (size > BIT(mem->mm.max_order) * mem->mm.chunk_size) + if (size > mem->mm.size) return -E2BIG; n_pages = size >> ilog2(mem->mm.chunk_size); diff --git a/drivers/gpu/drm/i915/selftests/intel_memory_region.c b/drivers/gpu/drm/i915/selftests/intel_memory_region.c index 334b0648e253..0aeba8e3af28 100644 --- a/drivers/gpu/drm/i915/selftests/intel_memory_region.c +++ b/drivers/gpu/drm/i915/selftests/intel_memory_region.c @@ -261,6 +261,82 @@ err_close_objects: return err; } +static int igt_mock_splintered_region(void *arg) +{ + struct intel_memory_region *mem = arg; + struct drm_i915_private *i915 = mem->i915; + struct drm_i915_gem_object *obj; + unsigned int expected_order; + LIST_HEAD(objects); + u64 size; + int err = 0; + + /* + * Sanity check we can still allocate everything even if the + * mm.max_order != mm.size. i.e our starting address space size is not a + * power-of-two. + */ + + size = (SZ_4G - 1) & PAGE_MASK; + mem = mock_region_create(i915, 0, size, PAGE_SIZE, 0); + if (IS_ERR(mem)) + return PTR_ERR(mem); + + if (mem->mm.size != size) { + pr_err("%s size mismatch(%llu != %llu)\n", + __func__, mem->mm.size, size); + err = -EINVAL; + goto out_put; + } + + expected_order = get_order(rounddown_pow_of_two(size)); + if (mem->mm.max_order != expected_order) { + pr_err("%s order mismatch(%u != %u)\n", + __func__, mem->mm.max_order, expected_order); + err = -EINVAL; + goto out_put; + } + + obj = igt_object_create(mem, &objects, size, 0); + if (IS_ERR(obj)) { + err = PTR_ERR(obj); + goto out_close; + } + + close_objects(mem, &objects); + + /* + * While we should be able allocate everything without any flag + * restrictions, if we consider I915_BO_ALLOC_CONTIGUOUS then we are + * actually limited to the largest power-of-two for the region size i.e + * max_order, due to the inner workings of the buddy allocator. So make + * sure that does indeed hold true. + */ + + obj = igt_object_create(mem, &objects, size, I915_BO_ALLOC_CONTIGUOUS); + if (!IS_ERR(obj)) { + pr_err("%s too large contiguous allocation was not rejected\n", + __func__); + err = -EINVAL; + goto out_close; + } + + obj = igt_object_create(mem, &objects, rounddown_pow_of_two(size), + I915_BO_ALLOC_CONTIGUOUS); + if (IS_ERR(obj)) { + pr_err("%s largest possible contiguous allocation failed\n", + __func__); + err = PTR_ERR(obj); + goto out_close; + } + +out_close: + close_objects(mem, &objects); +out_put: + intel_memory_region_put(mem); + return err; +} + static int igt_gpu_write_dw(struct intel_context *ce, struct i915_vma *vma, u32 dword, @@ -771,6 +847,7 @@ int intel_memory_region_mock_selftests(void) static const struct i915_subtest tests[] = { SUBTEST(igt_mock_fill), SUBTEST(igt_mock_contiguous), + SUBTEST(igt_mock_splintered_region), }; struct intel_memory_region *mem; struct drm_i915_private *i915; diff --git a/drivers/gpu/drm/i915/selftests/mock_region.c b/drivers/gpu/drm/i915/selftests/mock_region.c index 09660f5a0a4c..979d96f27c43 100644 --- a/drivers/gpu/drm/i915/selftests/mock_region.c +++ b/drivers/gpu/drm/i915/selftests/mock_region.c @@ -24,7 +24,7 @@ mock_object_create(struct intel_memory_region *mem, struct drm_i915_private *i915 = mem->i915; struct drm_i915_gem_object *obj; - if (size > BIT(mem->mm.max_order) * mem->mm.chunk_size) + if (size > mem->mm.size) return ERR_PTR(-E2BIG); obj = i915_gem_object_alloc(); -- cgit v1.2.3 From 5cbd7685b22823ebf432ec71eac1691b71c41037 Mon Sep 17 00:00:00 2001 From: Ville Syrjälä Date: Wed, 21 Oct 2020 16:14:39 +0300 Subject: drm/i915: Restore ILK-M RPS support MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Restore RPS for ILK-M. We lost it when an extra HAS_RPS() check appeared in intel_rps_enable(). Unfortunaltey this just makes the performance worse on my ILK because intel_ips insists on limiting the GPU freq to the minimum. If we don't do the RPS init then intel_ips will not limit the frequency for whatever reason. Either it can't get at some required information and thus makes wrong decisions, or we mess up some weights/etc. and cause it to make the wrong decisions when RPS init has been done, or the entire thing is just wrong. Would require a bunch of reverse engineering to figure out what's going on. Cc: stable@vger.kernel.org Cc: Chris Wilson Fixes: 9c878557b1eb ("drm/i915/gt: Use the RPM config register to determine clk frequencies") Signed-off-by: Ville Syrjälä Link: https://patchwork.freedesktop.org/patch/msgid/20201021131443.25616-1-ville.syrjala@linux.intel.com Reviewed-by: Chris Wilson (cherry picked from commit 2bf06370bcfb0dea5655e9a5ad460c7f7dca7739) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/i915/i915_pci.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c index 366ddfc8df6b..fb5e30de78c2 100644 --- a/drivers/gpu/drm/i915/i915_pci.c +++ b/drivers/gpu/drm/i915/i915_pci.c @@ -389,6 +389,7 @@ static const struct intel_device_info ilk_m_info = { GEN5_FEATURES, PLATFORM(INTEL_IRONLAKE), .is_mobile = 1, + .has_rps = true, .display.has_fbc = 1, }; -- cgit v1.2.3 From 61334ed227a5852100115180f5535b1396ed5227 Mon Sep 17 00:00:00 2001 From: Ville Syrjälä Date: Tue, 20 Oct 2020 22:43:29 +0300 Subject: drm/i915: Reject 90/270 degree rotated initial fbs MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We don't currently handle the initial fb readout correctly for 90/270 degree rotated scanout. Reject it. Cc: stable@vger.kernel.org Signed-off-by: Ville Syrjälä Link: https://patchwork.freedesktop.org/patch/msgid/20201020194330.28568-1-ville.syrjala@linux.intel.com Reviewed-by: Chris Wilson (cherry picked from commit a40a8305a732f4ecc2186ac7ca132ba062ed770d) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/i915/display/intel_display.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index 829b2a40a482..31337d2a2cde 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -10636,6 +10636,10 @@ skl_get_initial_plane_config(struct intel_crtc *crtc, val & PLANE_CTL_FLIP_HORIZONTAL) plane_config->rotation |= DRM_MODE_REFLECT_X; + /* 90/270 degree rotation would require extra work */ + if (drm_rotation_90_or_270(plane_config->rotation)) + goto error; + base = intel_de_read(dev_priv, PLANE_SURF(pipe, plane_id)) & 0xfffff000; plane_config->base = base; -- cgit v1.2.3 From 9c3f94e1681bb0ebd93390f014082042d8bc067a Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 27 Oct 2020 15:59:14 +0100 Subject: mptcp: add missing memory scheduling in the rx path When moving the skbs from the subflow into the msk receive queue, we must schedule there the required amount of memory. Try to borrow the required memory from the subflow, if needed, so that we leverage the existing TCP heuristic. Fixes: 6771bfd9ee24 ("mptcp: update mptcp ack sequence from work queue") Signed-off-by: Paolo Abeni Reviewed-by: Mat Martineau Link: https://lore.kernel.org/r/f6143a6193a083574f11b00dbf7b5ad151bc4ff4.1603810630.git.pabeni@redhat.com Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 185dacb39781..e7419fd15d84 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -274,6 +274,15 @@ static bool __mptcp_move_skb(struct mptcp_sock *msk, struct sock *ssk, skb_ext_reset(skb); skb_orphan(skb); + /* try to fetch required memory from subflow */ + if (!sk_rmem_schedule(sk, skb, skb->truesize)) { + if (ssk->sk_forward_alloc < skb->truesize) + goto drop; + __sk_mem_reclaim(ssk, skb->truesize); + if (!sk_rmem_schedule(sk, skb, skb->truesize)) + goto drop; + } + /* the skb map_seq accounts for the skb offset: * mptcp_subflow_get_mapped_dsn() is based on the current tp->copied_seq * value @@ -301,6 +310,7 @@ static bool __mptcp_move_skb(struct mptcp_sock *msk, struct sock *ssk, * will retransmit as needed, if needed. */ MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_DUPDATA); +drop: mptcp_drop(sk, skb); return false; } -- cgit v1.2.3 From 8fc3672a8ad3e782bac80e979bc2a2c10960cbe9 Mon Sep 17 00:00:00 2001 From: Lijun Pan Date: Tue, 27 Oct 2020 17:04:56 -0500 Subject: ibmvnic: fix ibmvnic_set_mac Jakub Kicinski brought up a concern in ibmvnic_set_mac(). ibmvnic_set_mac() does this: ether_addr_copy(adapter->mac_addr, addr->sa_data); if (adapter->state != VNIC_PROBED) rc = __ibmvnic_set_mac(netdev, addr->sa_data); So if state == VNIC_PROBED, the user can assign an invalid address to adapter->mac_addr, and ibmvnic_set_mac() will still return 0. The fix is to validate ethernet address at the beginning of ibmvnic_set_mac(), and move the ether_addr_copy to the case of "adapter->state != VNIC_PROBED". Fixes: c26eba03e407 ("ibmvnic: Update reset infrastructure to support tunable parameters") Signed-off-by: Lijun Pan Link: https://lore.kernel.org/r/20201027220456.71450-1-ljp@linux.ibm.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ibm/ibmvnic.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index 8148f796a807..af4dfbe28d56 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -1815,9 +1815,13 @@ static int ibmvnic_set_mac(struct net_device *netdev, void *p) int rc; rc = 0; - ether_addr_copy(adapter->mac_addr, addr->sa_data); - if (adapter->state != VNIC_PROBED) + if (!is_valid_ether_addr(addr->sa_data)) + return -EADDRNOTAVAIL; + + if (adapter->state != VNIC_PROBED) { + ether_addr_copy(adapter->mac_addr, addr->sa_data); rc = __ibmvnic_set_mac(netdev, addr->sa_data); + } return rc; } -- cgit v1.2.3 From eadd1befdd778a1eca57fad058782bd22b4db804 Mon Sep 17 00:00:00 2001 From: Aleksandr Nogikh Date: Wed, 28 Oct 2020 17:07:31 +0000 Subject: netem: fix zero division in tabledist Currently it is possible to craft a special netlink RTM_NEWQDISC command that can result in jitter being equal to 0x80000000. It is enough to set the 32 bit jitter to 0x02000000 (it will later be multiplied by 2^6) or just set the 64 bit jitter via TCA_NETEM_JITTER64. This causes an overflow during the generation of uniformly distributed numbers in tabledist(), which in turn leads to division by zero (sigma != 0, but sigma * 2 is 0). The related fragment of code needs 32-bit division - see commit 9b0ed89 ("netem: remove unnecessary 64 bit modulus"), so switching to 64 bit is not an option. Fix the issue by keeping the value of jitter within the range that can be adequately handled by tabledist() - [0;INT_MAX]. As negative std deviation makes no sense, take the absolute value of the passed value and cap it at INT_MAX. Inside tabledist(), switch to unsigned 32 bit arithmetic in order to prevent overflows. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Aleksandr Nogikh Reported-by: syzbot+ec762a6342ad0d3c0d8f@syzkaller.appspotmail.com Acked-by: Stephen Hemminger Link: https://lore.kernel.org/r/20201028170731.1383332-1-aleksandrnogikh@gmail.com Signed-off-by: Jakub Kicinski --- net/sched/sch_netem.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index 84f82771cdf5..0c345e43a09a 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -330,7 +330,7 @@ static s64 tabledist(s64 mu, s32 sigma, /* default uniform distribution */ if (dist == NULL) - return ((rnd % (2 * sigma)) + mu) - sigma; + return ((rnd % (2 * (u32)sigma)) + mu) - sigma; t = dist->table[rnd % dist->size]; x = (sigma % NETEM_DIST_SCALE) * t; @@ -812,6 +812,10 @@ static void get_slot(struct netem_sched_data *q, const struct nlattr *attr) q->slot_config.max_packets = INT_MAX; if (q->slot_config.max_bytes == 0) q->slot_config.max_bytes = INT_MAX; + + /* capping dist_jitter to the range acceptable by tabledist() */ + q->slot_config.dist_jitter = min_t(__s64, INT_MAX, abs(q->slot_config.dist_jitter)); + q->slot.packets_left = q->slot_config.max_packets; q->slot.bytes_left = q->slot_config.max_bytes; if (q->slot_config.min_delay | q->slot_config.max_delay | @@ -1037,6 +1041,9 @@ static int netem_change(struct Qdisc *sch, struct nlattr *opt, if (tb[TCA_NETEM_SLOT]) get_slot(q, tb[TCA_NETEM_SLOT]); + /* capping jitter to the range acceptable by tabledist() */ + q->jitter = min_t(s64, abs(q->jitter), INT_MAX); + return ret; get_table_failure: -- cgit v1.2.3 From 2734a24e6e5d18522fbf599135c59b82ec9b2c9e Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Thu, 29 Oct 2020 10:18:53 +0100 Subject: r8169: fix issue with forced threading in combination with shared interrupts As reported by Serge flag IRQF_NO_THREAD causes an error if the interrupt is actually shared and the other driver(s) don't have this flag set. This situation can occur if a PCI(e) legacy interrupt is used in combination with forced threading. There's no good way to deal with this properly, therefore we have to remove flag IRQF_NO_THREAD. For fixing the original forced threading issue switch to napi_schedule(). Fixes: 424a646e072a ("r8169: fix operation under forced interrupt threading") Link: https://www.spinics.net/lists/netdev/msg694960.html Reported-by: Serge Belyshev Signed-off-by: Heiner Kallweit Tested-by: Serge Belyshev Link: https://lore.kernel.org/r/b5b53bfe-35ac-3768-85bf-74d1290cf394@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/realtek/r8169_main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 3b6ddc706e92..00f13805c6f7 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -4573,7 +4573,7 @@ static irqreturn_t rtl8169_interrupt(int irq, void *dev_instance) } rtl_irq_disable(tp); - napi_schedule_irqoff(&tp->napi); + napi_schedule(&tp->napi); out: rtl_ack_events(tp, status); @@ -4746,7 +4746,7 @@ static int rtl_open(struct net_device *dev) rtl_request_firmware(tp); retval = request_irq(pci_irq_vector(pdev, 0), rtl8169_interrupt, - IRQF_NO_THREAD | IRQF_SHARED, dev->name, tp); + IRQF_SHARED, dev->name, tp); if (retval < 0) goto err_release_fw_2; -- cgit v1.2.3 From 39a7661dcf655c8198fd5d72412f5030a8e58444 Mon Sep 17 00:00:00 2001 From: Mike Leach Date: Thu, 29 Oct 2020 10:45:59 -0600 Subject: coresight: Fix uninitialised pointer bug in etm_setup_aux() Commit [bb1860efc817] changed the sink handling code introducing an uninitialised pointer bug. This results in the default sink selection failing. Prior to commit: static void etm_setup_aux(...) struct coresight_device *sink; /* First get the selected sink from user space. */ if (event->attr.config2) { id = (u32)event->attr.config2; sink = coresight_get_sink_by_id(id); } else { sink = coresight_get_enabled_sink(true); } *sink always initialised - possibly to NULL which triggers the automatic sink selection. After commit: static void etm_setup_aux(...) struct coresight_device *sink; /* First get the selected sink from user space. */ if (event->attr.config2) { id = (u32)event->attr.config2; sink = coresight_get_sink_by_id(id); } *sink pointer uninitialised when not providing a sink on the perf command line. This breaks later checks to enable automatic sink selection. Fixes: bb1860efc817 ("coresight: etm: perf: Sink selection using sysfs is deprecated") Signed-off-by: Mike Leach Signed-off-by: Mathieu Poirier Link: https://lore.kernel.org/r/20201029164559.1268531-3-mathieu.poirier@linaro.org Signed-off-by: Greg Kroah-Hartman --- drivers/hwtracing/coresight/coresight-etm-perf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/hwtracing/coresight/coresight-etm-perf.c b/drivers/hwtracing/coresight/coresight-etm-perf.c index c2c9b127d074..bdc34ca449f7 100644 --- a/drivers/hwtracing/coresight/coresight-etm-perf.c +++ b/drivers/hwtracing/coresight/coresight-etm-perf.c @@ -210,7 +210,7 @@ static void *etm_setup_aux(struct perf_event *event, void **pages, u32 id; int cpu = event->cpu; cpumask_t *mask; - struct coresight_device *sink; + struct coresight_device *sink = NULL; struct etm_event_data *event_data = NULL; event_data = alloc_event_data(cpu); -- cgit v1.2.3 From 80624263fa289b3416f7ca309491f1b75e579477 Mon Sep 17 00:00:00 2001 From: Suzuki K Poulose Date: Thu, 29 Oct 2020 10:45:58 -0600 Subject: coresight: cti: Initialize dynamic sysfs attributes With LOCKDEP enabled, CTI driver triggers the following splat due to uninitialized lock class for dynamically allocated attribute objects. [ 5.372901] coresight etm0: CPU0: ETM v4.0 initialized [ 5.376694] coresight etm1: CPU1: ETM v4.0 initialized [ 5.380785] coresight etm2: CPU2: ETM v4.0 initialized [ 5.385851] coresight etm3: CPU3: ETM v4.0 initialized [ 5.389808] BUG: key ffff00000564a798 has not been registered! [ 5.392456] ------------[ cut here ]------------ [ 5.398195] DEBUG_LOCKS_WARN_ON(1) [ 5.398233] WARNING: CPU: 1 PID: 32 at kernel/locking/lockdep.c:4623 lockdep_init_map_waits+0x14c/0x260 [ 5.406149] Modules linked in: [ 5.415411] CPU: 1 PID: 32 Comm: kworker/1:1 Not tainted 5.9.0-12034-gbbe85027ce80 #51 [ 5.418553] Hardware name: Qualcomm Technologies, Inc. APQ 8016 SBC (DT) [ 5.426453] Workqueue: events amba_deferred_retry_func [ 5.433299] pstate: 40000005 (nZcv daif -PAN -UAO -TCO BTYPE=--) [ 5.438252] pc : lockdep_init_map_waits+0x14c/0x260 [ 5.444410] lr : lockdep_init_map_waits+0x14c/0x260 [ 5.449007] sp : ffff800012bbb720 ... [ 5.531561] Call trace: [ 5.536847] lockdep_init_map_waits+0x14c/0x260 [ 5.539027] __kernfs_create_file+0xa8/0x1c8 [ 5.543539] sysfs_add_file_mode_ns+0xd0/0x208 [ 5.548054] internal_create_group+0x118/0x3c8 [ 5.552307] internal_create_groups+0x58/0xb8 [ 5.556733] sysfs_create_groups+0x2c/0x38 [ 5.561160] device_add+0x2d8/0x768 [ 5.565148] device_register+0x28/0x38 [ 5.568537] coresight_register+0xf8/0x320 [ 5.572358] cti_probe+0x1b0/0x3f0 ... Fix this by initializing the attributes when they are allocated. Fixes: 3c5597e39812 ("coresight: cti: Add connection information to sysfs") Reported-by: Leo Yan Tested-by: Leo Yan Cc: Mike Leach Cc: Mathieu Poirier Signed-off-by: Suzuki K Poulose Cc: stable Signed-off-by: Mathieu Poirier Link: https://lore.kernel.org/r/20201029164559.1268531-2-mathieu.poirier@linaro.org Signed-off-by: Greg Kroah-Hartman --- drivers/hwtracing/coresight/coresight-cti-sysfs.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/hwtracing/coresight/coresight-cti-sysfs.c b/drivers/hwtracing/coresight/coresight-cti-sysfs.c index 392757f3a019..7ff7e7780bbf 100644 --- a/drivers/hwtracing/coresight/coresight-cti-sysfs.c +++ b/drivers/hwtracing/coresight/coresight-cti-sysfs.c @@ -1065,6 +1065,13 @@ static int cti_create_con_sysfs_attr(struct device *dev, } eattr->var = con; con->con_attrs[attr_idx] = &eattr->attr.attr; + /* + * Initialize the dynamically allocated attribute + * to avoid LOCKDEP splat. See include/linux/sysfs.h + * for more details. + */ + sysfs_attr_init(con->con_attrs[attr_idx]); + return 0; } -- cgit v1.2.3 From 28e81c6270b3d0e9faadd565f314ad9ac8256620 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Mon, 26 Oct 2020 09:51:09 +0000 Subject: KVM: arm64: Don't corrupt tpidr_el2 on failed HVC call The hyp-init code starts by stashing a register in TPIDR_EL2 in in order to free a register. This happens no matter if the HVC call is legal or not. Although nothing wrong seems to come out of it, it feels odd to alter the EL2 state for something that eventually returns an error. Instead, use the fact that we know exactly which bits of the __kvm_hyp_init call are non-zero to perform the check with a series of EOR/ROR instructions, combined with a build-time check that the value is the one we expect. Signed-off-by: Marc Zyngier Link: https://lore.kernel.org/r/20201026095116.72051-2-maz@kernel.org --- arch/arm64/kvm/hyp/nvhe/hyp-init.S | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-) diff --git a/arch/arm64/kvm/hyp/nvhe/hyp-init.S b/arch/arm64/kvm/hyp/nvhe/hyp-init.S index 47224dc62c51..b11a9d7db677 100644 --- a/arch/arm64/kvm/hyp/nvhe/hyp-init.S +++ b/arch/arm64/kvm/hyp/nvhe/hyp-init.S @@ -57,16 +57,25 @@ __do_hyp_init: cmp x0, #HVC_STUB_HCALL_NR b.lo __kvm_handle_stub_hvc - /* Set tpidr_el2 for use by HYP to free a register */ - msr tpidr_el2, x2 - - mov x2, #KVM_HOST_SMCCC_FUNC(__kvm_hyp_init) - cmp x0, x2 - b.eq 1f + // We only actively check bits [24:31], and everything + // else has to be zero, which we check at build time. +#if (KVM_HOST_SMCCC_FUNC(__kvm_hyp_init) & 0xFFFFFFFF00FFFFFF) +#error Unexpected __KVM_HOST_SMCCC_FUNC___kvm_hyp_init value +#endif + + ror x0, x0, #24 + eor x0, x0, #((KVM_HOST_SMCCC_FUNC(__kvm_hyp_init) >> 24) & 0xF) + ror x0, x0, #4 + eor x0, x0, #((KVM_HOST_SMCCC_FUNC(__kvm_hyp_init) >> 28) & 0xF) + cbz x0, 1f mov x0, #SMCCC_RET_NOT_SUPPORTED eret -1: phys_to_ttbr x0, x1 +1: + /* Set tpidr_el2 for use by HYP to free a register */ + msr tpidr_el2, x2 + + phys_to_ttbr x0, x1 alternative_if ARM64_HAS_CNP orr x0, x0, #TTBR_CNP_BIT alternative_else_nop_endif -- cgit v1.2.3 From b6d6db4de86f5b1be9bbe59c73302538301915c7 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Mon, 26 Oct 2020 09:51:10 +0000 Subject: KVM: arm64: Remove leftover kern_hyp_va() in nVHE TLB invalidation The new calling convention says that pointers coming from the SMCCC interface are turned into their HYP version in the host HVC handler. However, there is still a stray kern_hyp_va() in the TLB invalidation code, which could result in a corrupted pointer. Drop the spurious conversion. Fixes: a071261d9318 ("KVM: arm64: nVHE: Fix pointers during SMCCC convertion") Signed-off-by: Marc Zyngier Link: https://lore.kernel.org/r/20201026095116.72051-3-maz@kernel.org --- arch/arm64/kvm/hyp/nvhe/tlb.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/arm64/kvm/hyp/nvhe/tlb.c b/arch/arm64/kvm/hyp/nvhe/tlb.c index 544bca3072b7..ad212d8fa417 100644 --- a/arch/arm64/kvm/hyp/nvhe/tlb.c +++ b/arch/arm64/kvm/hyp/nvhe/tlb.c @@ -121,7 +121,6 @@ void __kvm_tlb_flush_local_vmid(struct kvm_s2_mmu *mmu) struct tlb_inv_context cxt; /* Switch to requested VMID */ - mmu = kern_hyp_va(mmu); __tlb_switch_to_guest(mmu, &cxt); __tlbi(vmalle1); -- cgit v1.2.3 From d2782505fbe3055f983c772a448ac5cb419f9df7 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Mon, 26 Oct 2020 09:51:11 +0000 Subject: KVM: arm64: Drop useless PAN setting on host EL1 to EL2 transition Setting PSTATE.PAN when entering EL2 on nVHE doesn't make much sense as this bit only means something for translation regimes that include EL0. This obviously isn't the case in the nVHE case, so let's drop this setting. Signed-off-by: Marc Zyngier Reviewed-by: Vladimir Murzin Link: https://lore.kernel.org/r/20201026095116.72051-4-maz@kernel.org --- arch/arm64/kvm/hyp/nvhe/host.S | 2 -- 1 file changed, 2 deletions(-) diff --git a/arch/arm64/kvm/hyp/nvhe/host.S b/arch/arm64/kvm/hyp/nvhe/host.S index ff9a0f547b9f..ed27f06a31ba 100644 --- a/arch/arm64/kvm/hyp/nvhe/host.S +++ b/arch/arm64/kvm/hyp/nvhe/host.S @@ -17,8 +17,6 @@ SYM_FUNC_START(__host_exit) get_host_ctxt x0, x1 - ALTERNATIVE(nop, SET_PSTATE_PAN(1), ARM64_HAS_PAN, CONFIG_ARM64_PAN) - /* Store the host regs x2 and x3 */ stp x2, x3, [x0, #CPU_XREG_OFFSET(2)] -- cgit v1.2.3 From 7efe8ef274024ef1d5c495c79dfcbbff38c5f366 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Mon, 26 Oct 2020 14:44:23 +0000 Subject: KVM: arm64: Allocate stage-2 pgd pages with GFP_KERNEL_ACCOUNT For consistency with the rest of the stage-2 page-table page allocations (performing using a kvm_mmu_memory_cache), ensure that __GFP_ACCOUNT is included in the GFP flags for the PGD pages. Signed-off-by: Will Deacon Signed-off-by: Marc Zyngier Reviewed-by: Gavin Shan Cc: Marc Zyngier Cc: Quentin Perret Link: https://lore.kernel.org/r/20201026144423.24683-1-will@kernel.org --- arch/arm64/kvm/hyp/pgtable.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c index 0cdf6e461cbd..95141b0d6088 100644 --- a/arch/arm64/kvm/hyp/pgtable.c +++ b/arch/arm64/kvm/hyp/pgtable.c @@ -846,7 +846,7 @@ int kvm_pgtable_stage2_init(struct kvm_pgtable *pgt, struct kvm *kvm) u32 start_level = VTCR_EL2_TGRAN_SL0_BASE - sl0; pgd_sz = kvm_pgd_pages(ia_bits, start_level) * PAGE_SIZE; - pgt->pgd = alloc_pages_exact(pgd_sz, GFP_KERNEL | __GFP_ZERO); + pgt->pgd = alloc_pages_exact(pgd_sz, GFP_KERNEL_ACCOUNT | __GFP_ZERO); if (!pgt->pgd) return -ENOMEM; -- cgit v1.2.3 From 4a1c2c7f63c52ccb11770b5ae25920a6b79d3548 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Thu, 29 Oct 2020 17:24:09 +0000 Subject: KVM: arm64: Fix AArch32 handling of DBGD{CCINT,SCRext} and DBGVCR The DBGD{CCINT,SCRext} and DBGVCR register entries in the cp14 array are missing their target register, resulting in all accesses being targetted at the guard sysreg (indexed by __INVALID_SYSREG__). Point the emulation code at the actual register entries. Fixes: bdfb4b389c8d ("arm64: KVM: add trap handlers for AArch32 debug registers") Signed-off-by: Marc Zyngier Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20201029172409.2768336-1-maz@kernel.org --- arch/arm64/include/asm/kvm_host.h | 1 + arch/arm64/kvm/sys_regs.c | 6 +++--- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h index 0aecbab6a7fb..781d029b8aa8 100644 --- a/arch/arm64/include/asm/kvm_host.h +++ b/arch/arm64/include/asm/kvm_host.h @@ -239,6 +239,7 @@ enum vcpu_sysreg { #define cp14_DBGWCR0 (DBGWCR0_EL1 * 2) #define cp14_DBGWVR0 (DBGWVR0_EL1 * 2) #define cp14_DBGDCCINT (MDCCINT_EL1 * 2) +#define cp14_DBGVCR (DBGVCR32_EL2 * 2) #define NR_COPRO_REGS (NR_SYS_REGS * 2) diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c index 3c203cb8c103..983994f01a63 100644 --- a/arch/arm64/kvm/sys_regs.c +++ b/arch/arm64/kvm/sys_regs.c @@ -1881,9 +1881,9 @@ static const struct sys_reg_desc cp14_regs[] = { { Op1( 0), CRn( 0), CRm( 1), Op2( 0), trap_raz_wi }, DBG_BCR_BVR_WCR_WVR(1), /* DBGDCCINT */ - { Op1( 0), CRn( 0), CRm( 2), Op2( 0), trap_debug32 }, + { Op1( 0), CRn( 0), CRm( 2), Op2( 0), trap_debug32, NULL, cp14_DBGDCCINT }, /* DBGDSCRext */ - { Op1( 0), CRn( 0), CRm( 2), Op2( 2), trap_debug32 }, + { Op1( 0), CRn( 0), CRm( 2), Op2( 2), trap_debug32, NULL, cp14_DBGDSCRext }, DBG_BCR_BVR_WCR_WVR(2), /* DBGDTR[RT]Xint */ { Op1( 0), CRn( 0), CRm( 3), Op2( 0), trap_raz_wi }, @@ -1898,7 +1898,7 @@ static const struct sys_reg_desc cp14_regs[] = { { Op1( 0), CRn( 0), CRm( 6), Op2( 2), trap_raz_wi }, DBG_BCR_BVR_WCR_WVR(6), /* DBGVCR */ - { Op1( 0), CRn( 0), CRm( 7), Op2( 0), trap_debug32 }, + { Op1( 0), CRn( 0), CRm( 7), Op2( 0), trap_debug32, NULL, cp14_DBGVCR }, DBG_BCR_BVR_WCR_WVR(7), DBG_BCR_BVR_WCR_WVR(8), DBG_BCR_BVR_WCR_WVR(9), -- cgit v1.2.3 From e2fc6a9f686d037cbd9b08b9fb657685b4a722d3 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Thu, 29 Oct 2020 14:47:16 +0000 Subject: KVM: arm64: Fix masks in stage2_pte_cacheable() stage2_pte_cacheable() tries to figure out whether the mapping installed in its 'pte' parameter is cacheable or not. Unfortunately, it fails miserably because it extracts the memory attributes from the entry using FIELD_GET(), which returns the attributes shifted down to bit 0, but then compares this with the unshifted value generated by the PAGE_S2_MEMATTR() macro. A direct consequence of this bug is that cache maintenance is silently skipped, which in turn causes 32-bit guests to crash early on when their set/way maintenance is trapped but not emulated correctly. Fix the broken masks by avoiding the use of FIELD_GET() altogether. Fixes: 6d9d2115c480 ("KVM: arm64: Add support for stage-2 map()/unmap() in generic page-table") Reported-by: Marc Zyngier Signed-off-by: Will Deacon Signed-off-by: Marc Zyngier Cc: Quentin Perret Link: https://lore.kernel.org/r/20201029144716.30476-1-will@kernel.org --- arch/arm64/kvm/hyp/pgtable.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c index 95141b0d6088..0271b4a3b9fe 100644 --- a/arch/arm64/kvm/hyp/pgtable.c +++ b/arch/arm64/kvm/hyp/pgtable.c @@ -635,7 +635,7 @@ static void stage2_flush_dcache(void *addr, u64 size) static bool stage2_pte_cacheable(kvm_pte_t pte) { - u64 memattr = FIELD_GET(KVM_PTE_LEAF_ATTR_LO_S2_MEMATTR, pte); + u64 memattr = pte & KVM_PTE_LEAF_ATTR_LO_S2_MEMATTR; return memattr == PAGE_S2_MEMATTR(NORMAL); } -- cgit v1.2.3 From 2f40c46021bbb3ecd5c5f05764ecccbc276bc690 Mon Sep 17 00:00:00 2001 From: Gavin Shan Date: Mon, 26 Oct 2020 10:06:26 +1100 Subject: KVM: arm64: Use fallback mapping sizes for contiguous huge page sizes Although huge pages can be created out of multiple contiguous PMDs or PTEs, the corresponding sizes are not supported at Stage-2 yet. Instead of failing the mapping, fall back to the nearer supported mapping size (CONT_PMD to PMD and CONT_PTE to PTE respectively). Suggested-by: Marc Zyngier Signed-off-by: Gavin Shan [maz: rewritten commit message] Signed-off-by: Marc Zyngier Link: https://lore.kernel.org/r/20201025230626.18501-1-gshan@redhat.com --- arch/arm64/kvm/mmu.c | 26 +++++++++++++++++++------- 1 file changed, 19 insertions(+), 7 deletions(-) diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c index a816cb8e619b..e431d2d8e368 100644 --- a/arch/arm64/kvm/mmu.c +++ b/arch/arm64/kvm/mmu.c @@ -787,14 +787,26 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa, vma_shift = PAGE_SHIFT; } - if (vma_shift == PUD_SHIFT && - !fault_supports_stage2_huge_mapping(memslot, hva, PUD_SIZE)) - vma_shift = PMD_SHIFT; - - if (vma_shift == PMD_SHIFT && - !fault_supports_stage2_huge_mapping(memslot, hva, PMD_SIZE)) { - force_pte = true; + switch (vma_shift) { + case PUD_SHIFT: + if (fault_supports_stage2_huge_mapping(memslot, hva, PUD_SIZE)) + break; + fallthrough; + case CONT_PMD_SHIFT: + vma_shift = PMD_SHIFT; + fallthrough; + case PMD_SHIFT: + if (fault_supports_stage2_huge_mapping(memslot, hva, PMD_SIZE)) + break; + fallthrough; + case CONT_PTE_SHIFT: vma_shift = PAGE_SHIFT; + force_pte = true; + fallthrough; + case PAGE_SHIFT: + break; + default: + WARN_ONCE(1, "Unknown vma_shift %d", vma_shift); } vma_pagesize = 1UL << vma_shift; -- cgit v1.2.3 From 91a2c34b7d6fadc9c5d9433c620ea4c32ee7cae8 Mon Sep 17 00:00:00 2001 From: Santosh Shukla Date: Mon, 26 Oct 2020 16:54:07 +0530 Subject: KVM: arm64: Force PTE mapping on fault resulting in a device mapping VFIO allows a device driver to resolve a fault by mapping a MMIO range. This can be subsequently result in user_mem_abort() to try and compute a huge mapping based on the MMIO pfn, which is a sure recipe for things to go wrong. Instead, force a PTE mapping when the pfn faulted in has a device mapping. Fixes: 6d674e28f642 ("KVM: arm/arm64: Properly handle faulting of device mappings") Suggested-by: Marc Zyngier Signed-off-by: Santosh Shukla [maz: rewritten commit message] Signed-off-by: Marc Zyngier Reviewed-by: Gavin Shan Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/1603711447-11998-2-git-send-email-sashukla@nvidia.com --- arch/arm64/kvm/mmu.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c index e431d2d8e368..c7c6df6309d5 100644 --- a/arch/arm64/kvm/mmu.c +++ b/arch/arm64/kvm/mmu.c @@ -851,6 +851,7 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa, if (kvm_is_device_pfn(pfn)) { device = true; + force_pte = true; } else if (logging_active && !write_fault) { /* * Only actually map the page as writable if this was a write -- cgit v1.2.3 From 3d696f42c7f4bfcc54bde545898d4b89a74f9cc0 Mon Sep 17 00:00:00 2001 From: Tero Kristo Date: Thu, 29 Oct 2020 11:33:37 +0200 Subject: soc: ti: ti_sci_pm_domains: check for proper args count in xlate K2G devices still only use single parameter for power-domains property, so check for this properly in the driver. Without this, every peripheral fails to probe resulting in boot failure. Link: https://lore.kernel.org/r/20201029093337.21170-1-t-kristo@ti.com Fixes: efa5c01cd7ee ("soc: ti: ti_sci_pm_domains: switch to use multiple genpds instead of one") Reported-by: Nishanth Menon Signed-off-by: Tero Kristo Acked-by: Nishanth Menon Acked-by: Santosh Shilimkar Signed-off-by: Arnd Bergmann --- drivers/soc/ti/ti_sci_pm_domains.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/soc/ti/ti_sci_pm_domains.c b/drivers/soc/ti/ti_sci_pm_domains.c index af2126d2b2ff..8afb3f45d263 100644 --- a/drivers/soc/ti/ti_sci_pm_domains.c +++ b/drivers/soc/ti/ti_sci_pm_domains.c @@ -91,7 +91,7 @@ static struct generic_pm_domain *ti_sci_pd_xlate( struct genpd_onecell_data *genpd_data = data; unsigned int idx = genpdspec->args[0]; - if (genpdspec->args_count < 2) + if (genpdspec->args_count != 1 && genpdspec->args_count != 2) return ERR_PTR(-EINVAL); if (idx >= genpd_data->num_domains) { -- cgit v1.2.3 From a4147d855f50a676ebe61833a681f7c71945f343 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 31 Aug 2020 10:18:04 -0500 Subject: dmaengine: ti-cppi5: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9-rc1/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- include/linux/dma/ti-cppi5.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/linux/dma/ti-cppi5.h b/include/linux/dma/ti-cppi5.h index 5896441ee604..efa2f0309f00 100644 --- a/include/linux/dma/ti-cppi5.h +++ b/include/linux/dma/ti-cppi5.h @@ -47,7 +47,7 @@ struct cppi5_host_desc_t { u32 buf_info1; u32 org_buf_len; u64 org_buf_ptr; - u32 epib[0]; + u32 epib[]; } __packed; #define CPPI5_DESC_MIN_ALIGN (16U) @@ -139,7 +139,7 @@ struct cppi5_desc_epib_t { */ struct cppi5_monolithic_desc_t { struct cppi5_desc_hdr_t hdr; - u32 epib[0]; + u32 epib[]; }; #define CPPI5_INFO2_MDESC_DATA_OFFSET_SHIFT (18U) -- cgit v1.2.3 From 277ffd6c1ec0aa60856a03e18455fcca7d2a1186 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 31 Aug 2020 10:19:18 -0500 Subject: mailbox: zynqmp-ipi-message: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9-rc1/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- include/linux/mailbox/zynqmp-ipi-message.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/mailbox/zynqmp-ipi-message.h b/include/linux/mailbox/zynqmp-ipi-message.h index 9542b41eacfd..35ce84c8ca02 100644 --- a/include/linux/mailbox/zynqmp-ipi-message.h +++ b/include/linux/mailbox/zynqmp-ipi-message.h @@ -14,7 +14,7 @@ */ struct zynqmp_ipi_message { size_t len; - u8 data[0]; + u8 data[]; }; #endif /* _LINUX_ZYNQMP_IPI_MESSAGE_H_ */ -- cgit v1.2.3 From 883541051567a62add043a9f4ca5a31f2970bffd Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 31 Aug 2020 10:21:14 -0500 Subject: platform/chrome: cros_ec_commands: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9-rc1/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- include/linux/platform_data/cros_ec_commands.h | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/include/linux/platform_data/cros_ec_commands.h b/include/linux/platform_data/cros_ec_commands.h index 1fcfe9e63cb9..a3a9a878415f 100644 --- a/include/linux/platform_data/cros_ec_commands.h +++ b/include/linux/platform_data/cros_ec_commands.h @@ -1419,7 +1419,7 @@ struct ec_response_flash_info_2 { uint16_t num_banks_total; /* Number of banks described in banks array. */ uint16_t num_banks_desc; - struct ec_flash_bank banks[0]; + struct ec_flash_bank banks[]; } __ec_align4; /* @@ -2420,12 +2420,12 @@ struct ec_response_motion_sense_fifo_info { /* Total amount of vector lost */ uint16_t total_lost; /* Lost events since the last fifo_info, per sensors */ - uint16_t lost[0]; + uint16_t lost[]; } __ec_todo_packed; struct ec_response_motion_sense_fifo_data { uint32_t number_data; - struct ec_response_motion_sensor_data data[0]; + struct ec_response_motion_sensor_data data[]; } __ec_todo_packed; /* List supported activity recognition */ @@ -3093,7 +3093,7 @@ struct ec_response_tmp006_get_calibration_v1 { uint8_t algorithm; uint8_t num_params; uint8_t reserved[2]; - float val[0]; + float val[]; } __ec_align4; struct ec_params_tmp006_set_calibration_v1 { @@ -3101,7 +3101,7 @@ struct ec_params_tmp006_set_calibration_v1 { uint8_t algorithm; uint8_t num_params; uint8_t reserved; - float val[0]; + float val[]; } __ec_align4; @@ -5076,7 +5076,7 @@ struct ec_response_pd_log { uint8_t type; /* event type : see PD_EVENT_xx below */ uint8_t size_port; /* [7:5] port number [4:0] payload size in bytes */ uint16_t data; /* type-defined data payload */ - uint8_t payload[0]; /* optional additional data payload: 0..16 bytes */ + uint8_t payload[]; /* optional additional data payload: 0..16 bytes */ } __ec_align4; /* The timestamp is the microsecond counter shifted to get about a ms. */ @@ -5789,7 +5789,7 @@ struct ec_response_fp_encryption_status { struct ec_response_tp_frame_info { uint32_t n_frames; - uint32_t frame_sizes[0]; + uint32_t frame_sizes[]; } __ec_align4; /* Create a snapshot of current frame readings */ -- cgit v1.2.3 From 120088832042e6dc9866160ff267f8c347bf53e6 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 31 Aug 2020 10:21:55 -0500 Subject: platform/chrome: cros_ec_proto: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9-rc1/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- include/linux/platform_data/cros_ec_proto.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/platform_data/cros_ec_proto.h b/include/linux/platform_data/cros_ec_proto.h index 4a415ae851ef..02599687770c 100644 --- a/include/linux/platform_data/cros_ec_proto.h +++ b/include/linux/platform_data/cros_ec_proto.h @@ -69,7 +69,7 @@ struct cros_ec_command { uint32_t outsize; uint32_t insize; uint32_t result; - uint8_t data[0]; + uint8_t data[]; }; /** -- cgit v1.2.3 From 9d0a49c7023c0905ea19116cf74beb7d9611d8ac Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 31 Aug 2020 10:22:41 -0500 Subject: tracepoint: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9-rc1/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- kernel/tracepoint.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index 26efd22f0633..3f659f855074 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -50,7 +50,7 @@ static bool ok_to_free_tracepoints; */ struct tp_probes { struct rcu_head rcu; - struct tracepoint_func probes[0]; + struct tracepoint_func probes[]; }; static inline void *allocate_probes(int count) -- cgit v1.2.3 From fa29c9c11d4e2ba514421758991e5b3095642844 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 31 Aug 2020 10:23:32 -0500 Subject: params: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9-rc1/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- kernel/params.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/params.c b/kernel/params.c index 3835fb82c64b..164d79330849 100644 --- a/kernel/params.c +++ b/kernel/params.c @@ -530,7 +530,7 @@ struct module_param_attrs { unsigned int num; struct attribute_group grp; - struct param_attribute attrs[0]; + struct param_attribute attrs[]; }; #ifdef CONFIG_SYSFS -- cgit v1.2.3 From b08eadd2726356eb4867397fafd046d2b8aad4bf Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 31 Aug 2020 08:25:42 -0500 Subject: Bluetooth: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9-rc1/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- net/bluetooth/msft.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/bluetooth/msft.c b/net/bluetooth/msft.c index 8579bfeb2836..4b39534a14a1 100644 --- a/net/bluetooth/msft.c +++ b/net/bluetooth/msft.c @@ -12,12 +12,13 @@ struct msft_cp_read_supported_features { __u8 sub_opcode; } __packed; + struct msft_rp_read_supported_features { __u8 status; __u8 sub_opcode; __le64 features; __u8 evt_prefix_len; - __u8 evt_prefix[0]; + __u8 evt_prefix[]; } __packed; struct msft_data { -- cgit v1.2.3 From 5e01fdff04b7f7c3b8d456c11c8a9f978b4ddf65 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 31 Aug 2020 08:25:42 -0500 Subject: fs: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9-rc1/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- fs/binfmt_elf.c | 2 +- fs/hfs/btree.h | 2 +- fs/hfsplus/hfsplus_fs.h | 2 +- fs/isofs/rock.h | 8 ++++---- fs/select.c | 4 ++-- include/linux/fs.h | 2 +- 6 files changed, 10 insertions(+), 10 deletions(-) diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index b6b3d052ca86..fa50e8936f5f 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1690,7 +1690,7 @@ struct elf_thread_core_info { struct elf_thread_core_info *next; struct task_struct *task; struct elf_prstatus prstatus; - struct memelfnote notes[0]; + struct memelfnote notes[]; }; struct elf_note_info { diff --git a/fs/hfs/btree.h b/fs/hfs/btree.h index dcc2aab1b2c4..4ba45caf5939 100644 --- a/fs/hfs/btree.h +++ b/fs/hfs/btree.h @@ -60,7 +60,7 @@ struct hfs_bnode { wait_queue_head_t lock_wq; atomic_t refcnt; unsigned int page_offset; - struct page *page[0]; + struct page *page[]; }; #define HFS_BNODE_ERROR 0 diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h index 3b03fff68543..a92de5199ec3 100644 --- a/fs/hfsplus/hfsplus_fs.h +++ b/fs/hfsplus/hfsplus_fs.h @@ -117,7 +117,7 @@ struct hfs_bnode { wait_queue_head_t lock_wq; atomic_t refcnt; unsigned int page_offset; - struct page *page[0]; + struct page *page[]; }; #define HFS_BNODE_LOCK 0 diff --git a/fs/isofs/rock.h b/fs/isofs/rock.h index 1558cf22ef8a..ee9660e9671c 100644 --- a/fs/isofs/rock.h +++ b/fs/isofs/rock.h @@ -22,7 +22,7 @@ struct SU_ER_s { __u8 len_des; __u8 len_src; __u8 ext_ver; - __u8 data[0]; + __u8 data[]; } __attribute__ ((packed)); struct RR_RR_s { @@ -44,7 +44,7 @@ struct RR_PN_s { struct SL_component { __u8 flags; __u8 len; - __u8 text[0]; + __u8 text[]; } __attribute__ ((packed)); struct RR_SL_s { @@ -54,7 +54,7 @@ struct RR_SL_s { struct RR_NM_s { __u8 flags; - char name[0]; + char name[]; } __attribute__ ((packed)); struct RR_CL_s { @@ -71,7 +71,7 @@ struct stamp { struct RR_TF_s { __u8 flags; - struct stamp times[0]; /* Variable number of these beasts */ + struct stamp times[]; /* Variable number of these beasts */ } __attribute__ ((packed)); /* Linux-specific extension for transparent decompression */ diff --git a/fs/select.c b/fs/select.c index 7aef49552d4c..ebfebdfe5c69 100644 --- a/fs/select.c +++ b/fs/select.c @@ -97,7 +97,7 @@ u64 select_estimate_accuracy(struct timespec64 *tv) struct poll_table_page { struct poll_table_page * next; struct poll_table_entry * entry; - struct poll_table_entry entries[0]; + struct poll_table_entry entries[]; }; #define POLL_TABLE_FULL(table) \ @@ -836,7 +836,7 @@ SYSCALL_DEFINE1(old_select, struct sel_arg_struct __user *, arg) struct poll_list { struct poll_list *next; int len; - struct pollfd entries[0]; + struct pollfd entries[]; }; #define POLLFD_PER_PAGE ((PAGE_SIZE-sizeof(struct poll_list)) / sizeof(struct pollfd)) diff --git a/include/linux/fs.h b/include/linux/fs.h index 0bd126418bb6..21cc971fd960 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -3285,7 +3285,7 @@ static inline ino_t parent_ino(struct dentry *dentry) */ struct simple_transaction_argresp { ssize_t size; - char data[0]; + char data[]; }; #define SIMPLE_TRANSACTION_LIMIT (PAGE_SIZE - sizeof(struct simple_transaction_argresp)) -- cgit v1.2.3 From bfe124d197f61af632b9f3f4c2c0579bd7848448 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 31 Aug 2020 08:25:42 -0500 Subject: enetc: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9-rc1/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- drivers/net/ethernet/freescale/enetc/enetc_qos.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/freescale/enetc/enetc_qos.c b/drivers/net/ethernet/freescale/enetc/enetc_qos.c index c81be32bcedf..827f74e86d34 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_qos.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_qos.c @@ -402,7 +402,7 @@ struct enetc_psfp_gate { u32 num_entries; refcount_t refcount; struct hlist_node node; - struct action_gate_entry entries[0]; + struct action_gate_entry entries[]; }; /* Only enable the green color frame now -- cgit v1.2.3 From 4739eeafb9f0c45795407b3eb477dfcb2119f75b Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 31 Aug 2020 08:25:42 -0500 Subject: ima: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9-rc1/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- security/integrity/ima/ima.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/security/integrity/ima/ima.h b/security/integrity/ima/ima.h index 38043074ce5e..6ebefec616e4 100644 --- a/security/integrity/ima/ima.h +++ b/security/integrity/ima/ima.h @@ -101,7 +101,7 @@ struct ima_template_entry { struct tpm_digest *digests; struct ima_template_desc *template_desc; /* template descriptor */ u32 template_data_len; - struct ima_field_data template_data[0]; /* template related data */ + struct ima_field_data template_data[]; /* template related data */ }; struct ima_queue_entry { -- cgit v1.2.3 From 8fdaabe1c9b3226172ba2e9e525627219be6d29a Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Mon, 31 Aug 2020 08:25:42 -0500 Subject: scsi: target: tcmu: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9-rc1/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- drivers/target/target_core_user.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/target/target_core_user.c b/drivers/target/target_core_user.c index ea84d08747df..590e6d072228 100644 --- a/drivers/target/target_core_user.c +++ b/drivers/target/target_core_user.c @@ -194,7 +194,7 @@ struct tcmu_tmr { uint8_t tmr_type; uint32_t tmr_cmd_cnt; - int16_t tmr_cmd_ids[0]; + int16_t tmr_cmd_ids[]; }; /* -- cgit v1.2.3 From cfa736f5a6f31ca8a05459b5720aac030247ad1b Mon Sep 17 00:00:00 2001 From: Ralph Campbell Date: Mon, 31 Aug 2020 13:31:11 -0700 Subject: drm/nouveau/nouveau: fix the start/end range for migration The user level OpenCL code shouldn't have to align start and end addresses to a page boundary. That is better handled in the nouveau driver. The npages field is also redundant since it can be computed from the start and end addresses. Signed-off-by: Ralph Campbell Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nouveau_svm.c | 14 +++----------- 1 file changed, 3 insertions(+), 11 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c b/drivers/gpu/drm/nouveau/nouveau_svm.c index 2df1c0460559..4f69e4c3dafd 100644 --- a/drivers/gpu/drm/nouveau/nouveau_svm.c +++ b/drivers/gpu/drm/nouveau/nouveau_svm.c @@ -105,11 +105,11 @@ nouveau_svmm_bind(struct drm_device *dev, void *data, struct nouveau_cli *cli = nouveau_cli(file_priv); struct drm_nouveau_svm_bind *args = data; unsigned target, cmd, priority; - unsigned long addr, end, size; + unsigned long addr, end; struct mm_struct *mm; args->va_start &= PAGE_MASK; - args->va_end &= PAGE_MASK; + args->va_end = ALIGN(args->va_end, PAGE_SIZE); /* Sanity check arguments */ if (args->reserved0 || args->reserved1) @@ -118,8 +118,6 @@ nouveau_svmm_bind(struct drm_device *dev, void *data, return -EINVAL; if (args->va_start >= args->va_end) return -EINVAL; - if (!args->npages) - return -EINVAL; cmd = args->header >> NOUVEAU_SVM_BIND_COMMAND_SHIFT; cmd &= NOUVEAU_SVM_BIND_COMMAND_MASK; @@ -151,12 +149,6 @@ nouveau_svmm_bind(struct drm_device *dev, void *data, if (args->stride) return -EINVAL; - size = ((unsigned long)args->npages) << PAGE_SHIFT; - if ((args->va_start + size) <= args->va_start) - return -EINVAL; - if ((args->va_start + size) > args->va_end) - return -EINVAL; - /* * Ok we are ask to do something sane, for now we only support migrate * commands but we will add things like memory policy (what to do on @@ -171,7 +163,7 @@ nouveau_svmm_bind(struct drm_device *dev, void *data, return -EINVAL; } - for (addr = args->va_start, end = args->va_start + size; addr < end;) { + for (addr = args->va_start, end = args->va_end; addr < end;) { struct vm_area_struct *vma; unsigned long next; -- cgit v1.2.3 From 24d9422e26ea75118acf00172f83417c296f5b5f Mon Sep 17 00:00:00 2001 From: Lyude Paul Date: Fri, 4 Sep 2020 16:27:58 -0400 Subject: drm/nouveau/kms/nv50-: Program notifier offset before requesting disp caps Not entirely sure why this never came up when I originally tested this (maybe some BIOSes already have this setup?) but the ->caps_init vfunc appears to cause the display engine to throw an exception on driver init, at least on my ThinkPad P72: nouveau 0000:01:00.0: disp: chid 0 mthd 008c data 00000000 0000508c 0000102b This is magic nvidia speak for "You need to have the DMA notifier offset programmed before you can call NV507D_GET_CAPABILITIES." So, let's fix this by doing that, and also perform an update afterwards to prevent racing with the GPU when reading capabilities. v2: * Don't just program the DMA notifier offset, make sure to actually perform an update v3: * Don't call UPDATE() * Actually read the correct notifier fields, as apparently the CAPABILITIES_DONE field lives in a different location than the main NV_DISP_CORE_NOTIFIER_1 field. As well, 907d+ use a different CAPABILITIES_DONE field then pre-907d cards. v4: * Don't forget to check the return value of core507d_read_caps() v5: * Get rid of NV50_DISP_CAPS_NTFY[14], use NV50_DISP_CORE_NTFY * Disable notifier after calling GetCapabilities() Signed-off-by: Lyude Paul Fixes: 4a2cb4181b07 ("drm/nouveau/kms/nv50-: Probe SOR and PIOR caps for DP interlacing support") Cc: # v5.8+ Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/dispnv50/core.h | 2 ++ drivers/gpu/drm/nouveau/dispnv50/core507d.c | 41 ++++++++++++++++++++-- drivers/gpu/drm/nouveau/dispnv50/core907d.c | 36 ++++++++++++++++++- drivers/gpu/drm/nouveau/dispnv50/core917d.c | 2 +- .../gpu/drm/nouveau/include/nvhw/class/cl507d.h | 5 ++- .../gpu/drm/nouveau/include/nvhw/class/cl907d.h | 4 +++ 6 files changed, 85 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/nouveau/dispnv50/core.h b/drivers/gpu/drm/nouveau/dispnv50/core.h index 498622c0c670..f75088186fba 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/core.h +++ b/drivers/gpu/drm/nouveau/dispnv50/core.h @@ -44,6 +44,7 @@ int core507d_new_(const struct nv50_core_func *, struct nouveau_drm *, s32, struct nv50_core **); int core507d_init(struct nv50_core *); void core507d_ntfy_init(struct nouveau_bo *, u32); +int core507d_read_caps(struct nv50_disp *disp); int core507d_caps_init(struct nouveau_drm *, struct nv50_disp *); int core507d_ntfy_wait_done(struct nouveau_bo *, u32, struct nvif_device *); int core507d_update(struct nv50_core *, u32 *, bool); @@ -55,6 +56,7 @@ extern const struct nv50_outp_func pior507d; int core827d_new(struct nouveau_drm *, s32, struct nv50_core **); int core907d_new(struct nouveau_drm *, s32, struct nv50_core **); +int core907d_caps_init(struct nouveau_drm *drm, struct nv50_disp *disp); extern const struct nv50_outp_func dac907d; extern const struct nv50_outp_func sor907d; diff --git a/drivers/gpu/drm/nouveau/dispnv50/core507d.c b/drivers/gpu/drm/nouveau/dispnv50/core507d.c index 248edf69e168..e6f16a7750f0 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/core507d.c +++ b/drivers/gpu/drm/nouveau/dispnv50/core507d.c @@ -78,18 +78,55 @@ core507d_ntfy_init(struct nouveau_bo *bo, u32 offset) } int -core507d_caps_init(struct nouveau_drm *drm, struct nv50_disp *disp) +core507d_read_caps(struct nv50_disp *disp) { struct nvif_push *push = disp->core->chan.push; int ret; - if ((ret = PUSH_WAIT(push, 2))) + ret = PUSH_WAIT(push, 6); + if (ret) return ret; + PUSH_MTHD(push, NV507D, SET_NOTIFIER_CONTROL, + NVDEF(NV507D, SET_NOTIFIER_CONTROL, MODE, WRITE) | + NVVAL(NV507D, SET_NOTIFIER_CONTROL, OFFSET, NV50_DISP_CORE_NTFY >> 2) | + NVDEF(NV507D, SET_NOTIFIER_CONTROL, NOTIFY, ENABLE)); + PUSH_MTHD(push, NV507D, GET_CAPABILITIES, 0x00000000); + + PUSH_MTHD(push, NV507D, SET_NOTIFIER_CONTROL, + NVDEF(NV507D, SET_NOTIFIER_CONTROL, NOTIFY, DISABLE)); + return PUSH_KICK(push); } +int +core507d_caps_init(struct nouveau_drm *drm, struct nv50_disp *disp) +{ + struct nv50_core *core = disp->core; + struct nouveau_bo *bo = disp->sync; + s64 time; + int ret; + + NVBO_WR32(bo, NV50_DISP_CORE_NTFY, NV_DISP_CORE_NOTIFIER_1, CAPABILITIES_1, + NVDEF(NV_DISP_CORE_NOTIFIER_1, CAPABILITIES_1, DONE, FALSE)); + + ret = core507d_read_caps(disp); + if (ret < 0) + return ret; + + time = nvif_msec(core->chan.base.device, 2000ULL, + if (NVBO_TD32(bo, NV50_DISP_CORE_NTFY, + NV_DISP_CORE_NOTIFIER_1, CAPABILITIES_1, DONE, ==, TRUE)) + break; + usleep_range(1, 2); + ); + if (time < 0) + NV_ERROR(drm, "core caps notifier timeout\n"); + + return 0; +} + int core507d_init(struct nv50_core *core) { diff --git a/drivers/gpu/drm/nouveau/dispnv50/core907d.c b/drivers/gpu/drm/nouveau/dispnv50/core907d.c index b17c03529c78..8564d4dffaff 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/core907d.c +++ b/drivers/gpu/drm/nouveau/dispnv50/core907d.c @@ -22,11 +22,45 @@ #include "core.h" #include "head.h" +#include +#include + +#include + +#include "nouveau_bo.h" + +int +core907d_caps_init(struct nouveau_drm *drm, struct nv50_disp *disp) +{ + struct nv50_core *core = disp->core; + struct nouveau_bo *bo = disp->sync; + s64 time; + int ret; + + NVBO_WR32(bo, NV50_DISP_CORE_NTFY, NV907D_CORE_NOTIFIER_3, CAPABILITIES_4, + NVDEF(NV907D_CORE_NOTIFIER_3, CAPABILITIES_4, DONE, FALSE)); + + ret = core507d_read_caps(disp); + if (ret < 0) + return ret; + + time = nvif_msec(core->chan.base.device, 2000ULL, + if (NVBO_TD32(bo, NV50_DISP_CORE_NTFY, + NV907D_CORE_NOTIFIER_3, CAPABILITIES_4, DONE, ==, TRUE)) + break; + usleep_range(1, 2); + ); + if (time < 0) + NV_ERROR(drm, "core caps notifier timeout\n"); + + return 0; +} + static const struct nv50_core_func core907d = { .init = core507d_init, .ntfy_init = core507d_ntfy_init, - .caps_init = core507d_caps_init, + .caps_init = core907d_caps_init, .ntfy_wait_done = core507d_ntfy_wait_done, .update = core507d_update, .head = &head907d, diff --git a/drivers/gpu/drm/nouveau/dispnv50/core917d.c b/drivers/gpu/drm/nouveau/dispnv50/core917d.c index 66846f372080..1cd3a2a35dfb 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/core917d.c +++ b/drivers/gpu/drm/nouveau/dispnv50/core917d.c @@ -26,7 +26,7 @@ static const struct nv50_core_func core917d = { .init = core507d_init, .ntfy_init = core507d_ntfy_init, - .caps_init = core507d_caps_init, + .caps_init = core907d_caps_init, .ntfy_wait_done = core507d_ntfy_wait_done, .update = core507d_update, .head = &head917d, diff --git a/drivers/gpu/drm/nouveau/include/nvhw/class/cl507d.h b/drivers/gpu/drm/nouveau/include/nvhw/class/cl507d.h index 2e444bac701d..6a463f308b64 100644 --- a/drivers/gpu/drm/nouveau/include/nvhw/class/cl507d.h +++ b/drivers/gpu/drm/nouveau/include/nvhw/class/cl507d.h @@ -32,7 +32,10 @@ #define NV_DISP_CORE_NOTIFIER_1_COMPLETION_0_DONE_TRUE 0x00000001 #define NV_DISP_CORE_NOTIFIER_1_COMPLETION_0_R0 15:1 #define NV_DISP_CORE_NOTIFIER_1_COMPLETION_0_TIMESTAMP 29:16 - +#define NV_DISP_CORE_NOTIFIER_1_CAPABILITIES_1 0x00000001 +#define NV_DISP_CORE_NOTIFIER_1_CAPABILITIES_1_DONE 0:0 +#define NV_DISP_CORE_NOTIFIER_1_CAPABILITIES_1_DONE_FALSE 0x00000000 +#define NV_DISP_CORE_NOTIFIER_1_CAPABILITIES_1_DONE_TRUE 0x00000001 // class methods #define NV507D_UPDATE (0x00000080) diff --git a/drivers/gpu/drm/nouveau/include/nvhw/class/cl907d.h b/drivers/gpu/drm/nouveau/include/nvhw/class/cl907d.h index 34bc3eafac7d..79aff6ff3138 100644 --- a/drivers/gpu/drm/nouveau/include/nvhw/class/cl907d.h +++ b/drivers/gpu/drm/nouveau/include/nvhw/class/cl907d.h @@ -24,6 +24,10 @@ #ifndef _cl907d_h_ #define _cl907d_h_ +#define NV907D_CORE_NOTIFIER_3_CAPABILITIES_4 0x00000004 +#define NV907D_CORE_NOTIFIER_3_CAPABILITIES_4_DONE 0:0 +#define NV907D_CORE_NOTIFIER_3_CAPABILITIES_4_DONE_FALSE 0x00000000 +#define NV907D_CORE_NOTIFIER_3_CAPABILITIES_4_DONE_TRUE 0x00000001 #define NV907D_CORE_NOTIFIER_3_CAPABILITIES_CAP_SOR0_20 0x00000014 #define NV907D_CORE_NOTIFIER_3_CAPABILITIES_CAP_SOR0_20_SINGLE_LVDS18 0:0 #define NV907D_CORE_NOTIFIER_3_CAPABILITIES_CAP_SOR0_20_SINGLE_LVDS18_FALSE 0x00000000 -- cgit v1.2.3 From 925681454d7b557d404b5d28ef4469fac1b2e105 Mon Sep 17 00:00:00 2001 From: Karol Herbst Date: Wed, 7 Oct 2020 00:08:09 +0200 Subject: drm/nouveau/gem: fix "refcount_t: underflow; use-after-free" we can't use nouveau_bo_ref here as no ttm object was allocated and nouveau_bo_ref mainly deals with that. Simply deallocate the object. Signed-off-by: Karol Herbst Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nouveau_gem.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c index 89adadf4706b..549bc67feabb 100644 --- a/drivers/gpu/drm/nouveau/nouveau_gem.c +++ b/drivers/gpu/drm/nouveau/nouveau_gem.c @@ -190,7 +190,8 @@ nouveau_gem_new(struct nouveau_cli *cli, u64 size, int align, uint32_t domain, * to the caller, instead of a normal nouveau_bo ttm reference. */ ret = drm_gem_object_init(drm->dev, &nvbo->bo.base, size); if (ret) { - nouveau_bo_ref(NULL, &nvbo); + drm_gem_object_release(&nvbo->bo.base); + kfree(nvbo); return ret; } -- cgit v1.2.3 From dcd292c172493067a72672b245a3dd1bcf7268dd Mon Sep 17 00:00:00 2001 From: Karol Herbst Date: Tue, 13 Oct 2020 14:01:26 +0200 Subject: drm/nouveau/device: fix changing endianess code to work on older GPUs With this we try to detect if the endianess switch works and assume LE if not. Suggested by Ben. Fixes: 51c05340e407 ("drm/nouveau/device: detect if changing endianness failed") Signed-off-by: Karol Herbst Cc: # v5.8+ Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nvkm/engine/device/base.c | 39 +++++++++++++++-------- 1 file changed, 26 insertions(+), 13 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c index dcb70677d0ac..7851bec5f0e5 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c @@ -2924,17 +2924,34 @@ nvkm_device_del(struct nvkm_device **pdevice) } } +/* returns true if the GPU is in the CPU native byte order */ static inline bool nvkm_device_endianness(struct nvkm_device *device) { - u32 boot1 = nvkm_rd32(device, 0x000004) & 0x01000001; #ifdef __BIG_ENDIAN - if (!boot1) - return false; + const bool big_endian = true; #else - if (boot1) - return false; + const bool big_endian = false; #endif + + /* Read NV_PMC_BOOT_1, and assume non-functional endian switch if it + * doesn't contain the expected values. + */ + u32 pmc_boot_1 = nvkm_rd32(device, 0x000004); + if (pmc_boot_1 && pmc_boot_1 != 0x01000001) + return !big_endian; /* Assume GPU is LE in this case. */ + + /* 0 means LE and 0x01000001 means BE GPU. Condition is true when + * GPU/CPU endianness don't match. + */ + if (big_endian == !pmc_boot_1) { + nvkm_wr32(device, 0x000004, 0x01000001); + nvkm_rd32(device, 0x000000); + if (nvkm_rd32(device, 0x000004) != (big_endian ? 0x01000001 : 0x00000000)) + return !big_endian; /* Assume GPU is LE on any unexpected read-back. */ + } + + /* CPU/GPU endianness should (hopefully) match. */ return true; } @@ -2987,14 +3004,10 @@ nvkm_device_ctor(const struct nvkm_device_func *func, if (detect) { /* switch mmio to cpu's native endianness */ if (!nvkm_device_endianness(device)) { - nvkm_wr32(device, 0x000004, 0x01000001); - nvkm_rd32(device, 0x000000); - if (!nvkm_device_endianness(device)) { - nvdev_error(device, - "GPU not supported on big-endian\n"); - ret = -ENOSYS; - goto done; - } + nvdev_error(device, + "Couldn't switch GPU to CPUs endianess\n"); + ret = -ENOSYS; + goto done; } boot0 = nvkm_rd32(device, 0x000000); -- cgit v1.2.3 From 2d831155cf0607566e43d8465da33774b2dc7221 Mon Sep 17 00:00:00 2001 From: Lyude Paul Date: Tue, 29 Sep 2020 18:31:31 -0400 Subject: drm/nouveau/kms/nv50-: Get rid of bogus nouveau_conn_mode_valid() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Ville also pointed out that I got a lot of the logic here wrong as well, whoops. While I don't think anyone's likely using 3D output with nouveau, the next patch will make nouveau_conn_mode_valid() make a lot less sense. So, let's just get rid of it and open-code it like before, while taking care to move the 3D frame packing calculations on the dot clock into the right place. Signed-off-by: Lyude Paul Fixes: d6a9efece724 ("drm/nouveau/kms/nv50-: Share DP SST mode_valid() handling with MST") Cc: Ville Syrjälä Cc: # v5.8+ Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nouveau_connector.c | 36 ++++++++--------------------- drivers/gpu/drm/nouveau/nouveau_dp.c | 16 +++++++++---- 2 files changed, 21 insertions(+), 31 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index 49dd0cbc332f..6f21f36719fc 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++ b/drivers/gpu/drm/nouveau/nouveau_connector.c @@ -1023,29 +1023,6 @@ get_tmds_link_bandwidth(struct drm_connector *connector) return 112000 * duallink_scale; } -enum drm_mode_status -nouveau_conn_mode_clock_valid(const struct drm_display_mode *mode, - const unsigned min_clock, - const unsigned max_clock, - unsigned int *clock_out) -{ - unsigned int clock = mode->clock; - - if ((mode->flags & DRM_MODE_FLAG_3D_MASK) == - DRM_MODE_FLAG_3D_FRAME_PACKING) - clock *= 2; - - if (clock < min_clock) - return MODE_CLOCK_LOW; - if (clock > max_clock) - return MODE_CLOCK_HIGH; - - if (clock_out) - *clock_out = clock; - - return MODE_OK; -} - static enum drm_mode_status nouveau_connector_mode_valid(struct drm_connector *connector, struct drm_display_mode *mode) @@ -1053,7 +1030,7 @@ nouveau_connector_mode_valid(struct drm_connector *connector, struct nouveau_connector *nv_connector = nouveau_connector(connector); struct nouveau_encoder *nv_encoder = nv_connector->detected_encoder; struct drm_encoder *encoder = to_drm_encoder(nv_encoder); - unsigned min_clock = 25000, max_clock = min_clock; + unsigned int min_clock = 25000, max_clock = min_clock, clock = mode->clock; switch (nv_encoder->dcb->type) { case DCB_OUTPUT_LVDS: @@ -1082,8 +1059,15 @@ nouveau_connector_mode_valid(struct drm_connector *connector, return MODE_BAD; } - return nouveau_conn_mode_clock_valid(mode, min_clock, max_clock, - NULL); + if ((mode->flags & DRM_MODE_FLAG_3D_MASK) == DRM_MODE_FLAG_3D_FRAME_PACKING) + clock *= 2; + + if (clock < min_clock) + return MODE_CLOCK_LOW; + if (clock > max_clock) + return MODE_CLOCK_HIGH; + + return MODE_OK; } static struct drm_encoder * diff --git a/drivers/gpu/drm/nouveau/nouveau_dp.c b/drivers/gpu/drm/nouveau/nouveau_dp.c index 7b640e05bd4c..93e3751ad7f1 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dp.c +++ b/drivers/gpu/drm/nouveau/nouveau_dp.c @@ -232,12 +232,14 @@ nv50_dp_mode_valid(struct drm_connector *connector, unsigned *out_clock) { const unsigned min_clock = 25000; - unsigned max_clock, ds_clock, clock; - enum drm_mode_status ret; + unsigned max_clock, ds_clock, clock = mode->clock; if (mode->flags & DRM_MODE_FLAG_INTERLACE && !outp->caps.dp_interlace) return MODE_NO_INTERLACE; + if ((mode->flags & DRM_MODE_FLAG_3D_MASK) == DRM_MODE_FLAG_3D_FRAME_PACKING) + clock *= 2; + max_clock = outp->dp.link_nr * outp->dp.link_bw; ds_clock = drm_dp_downstream_max_dotclock(outp->dp.dpcd, outp->dp.downstream_ports); @@ -245,9 +247,13 @@ nv50_dp_mode_valid(struct drm_connector *connector, max_clock = min(max_clock, ds_clock); clock = mode->clock * (connector->display_info.bpc * 3) / 10; - ret = nouveau_conn_mode_clock_valid(mode, min_clock, max_clock, - &clock); + if (clock < min_clock) + return MODE_CLOCK_LOW; + if (clock > max_clock) + return MODE_CLOCK_HIGH; + if (out_clock) *out_clock = clock; - return ret; + + return MODE_OK; } -- cgit v1.2.3 From d7787cc04e0a1f2043264d1550465081096bd065 Mon Sep 17 00:00:00 2001 From: Lyude Paul Date: Tue, 29 Sep 2020 18:31:32 -0400 Subject: drm/nouveau/kms/nv50-: Fix clock checking algorithm in nv50_dp_mode_valid() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit While I thought I had this correct (since it actually did reject modes like I expected during testing), Ville Syrjala from Intel pointed out that the logic here isn't correct. max_clock refers to the max data rate supported by the DP encoder. So, limiting it to the output of ds_clock (which refers to the maximum dotclock of the downstream DP device) doesn't make any sense. Additionally, since we're using the connector's bpc as the canonical BPC we should use this in mode_valid until we support dynamically setting the bpp based on bandwidth constraints. https://lists.freedesktop.org/archives/dri-devel/2020-September/280276.html For more info. So, let's rewrite this using Ville's advice. v2: * Ville pointed out I mixed up the dotclock and the link rate. So fix that... * ...and also rename all the variables in this function to be more appropriately labeled so I stop mixing them up. * Reuse the bpp from the connector for now until we have dynamic bpp selection. * Use use DIV_ROUND_UP for calculating the mode rate like i915 does, which we should also have been doing from the start Signed-off-by: Lyude Paul Fixes: 409d38139b42 ("drm/nouveau/kms/nv50-: Use downstream DP clock limits for mode validation") Cc: Ville Syrjälä Cc: Lyude Paul Cc: Ben Skeggs Signed-off-by: Ben Skeggs --- drivers/gpu/drm/nouveau/nouveau_dp.c | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_dp.c b/drivers/gpu/drm/nouveau/nouveau_dp.c index 93e3751ad7f1..040ed88d362d 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dp.c +++ b/drivers/gpu/drm/nouveau/nouveau_dp.c @@ -231,8 +231,9 @@ nv50_dp_mode_valid(struct drm_connector *connector, const struct drm_display_mode *mode, unsigned *out_clock) { - const unsigned min_clock = 25000; - unsigned max_clock, ds_clock, clock = mode->clock; + const unsigned int min_clock = 25000; + unsigned int max_rate, mode_rate, ds_max_dotclock, clock = mode->clock; + const u8 bpp = connector->display_info.bpc * 3; if (mode->flags & DRM_MODE_FLAG_INTERLACE && !outp->caps.dp_interlace) return MODE_NO_INTERLACE; @@ -240,17 +241,17 @@ nv50_dp_mode_valid(struct drm_connector *connector, if ((mode->flags & DRM_MODE_FLAG_3D_MASK) == DRM_MODE_FLAG_3D_FRAME_PACKING) clock *= 2; - max_clock = outp->dp.link_nr * outp->dp.link_bw; - ds_clock = drm_dp_downstream_max_dotclock(outp->dp.dpcd, - outp->dp.downstream_ports); - if (ds_clock) - max_clock = min(max_clock, ds_clock); + max_rate = outp->dp.link_nr * outp->dp.link_bw; + mode_rate = DIV_ROUND_UP(clock * bpp, 8); + if (mode_rate > max_rate) + return MODE_CLOCK_HIGH; + + ds_max_dotclock = drm_dp_downstream_max_dotclock(outp->dp.dpcd, outp->dp.downstream_ports); + if (ds_max_dotclock && clock > ds_max_dotclock) + return MODE_CLOCK_HIGH; - clock = mode->clock * (connector->display_info.bpc * 3) / 10; if (clock < min_clock) return MODE_CLOCK_LOW; - if (clock > max_clock) - return MODE_CLOCK_HIGH; if (out_clock) *out_clock = clock; -- cgit v1.2.3 From 52d3967704aea6cb316d419a33a5e1d56d33a3c1 Mon Sep 17 00:00:00 2001 From: Pawel Laszczak Date: Thu, 22 Oct 2020 08:55:05 +0800 Subject: usb: cdns3: Fix on-chip memory overflow issue Patch fixes issue caused setting On-chip memory overflow bit in usb_sts register. The issue occurred because EP_CFG register was set twice before USB_STS.CFGSTS was set. Every write operation on EP_CFG.BUFFERING causes that controller increases internal counter holding the number of reserved on-chip buffers. First time this register was updated in function cdns3_ep_config before delegating SET_CONFIGURATION request to class driver and again it was updated when class wanted to enable endpoint. This patch fixes this issue by configuring endpoints enabled by class driver in cdns3_gadget_ep_enable and others just before status stage. Cc: stable@vger.kernel.org#v5.8+ Fixes: 7733f6c32e36 ("usb: cdns3: Add Cadence USB3 DRD Driver") Reported-and-tested-by: Peter Chen Signed-off-by: Pawel Laszczak Signed-off-by: Peter Chen --- drivers/usb/cdns3/ep0.c | 65 ++++++++++++++++------------- drivers/usb/cdns3/gadget.c | 102 +++++++++++++++++++++++++-------------------- drivers/usb/cdns3/gadget.h | 3 +- 3 files changed, 94 insertions(+), 76 deletions(-) diff --git a/drivers/usb/cdns3/ep0.c b/drivers/usb/cdns3/ep0.c index 4761c852d9c4..d3121a32cc68 100644 --- a/drivers/usb/cdns3/ep0.c +++ b/drivers/usb/cdns3/ep0.c @@ -137,48 +137,36 @@ static int cdns3_req_ep0_set_configuration(struct cdns3_device *priv_dev, struct usb_ctrlrequest *ctrl_req) { enum usb_device_state device_state = priv_dev->gadget.state; - struct cdns3_endpoint *priv_ep; u32 config = le16_to_cpu(ctrl_req->wValue); int result = 0; - int i; switch (device_state) { case USB_STATE_ADDRESS: - /* Configure non-control EPs */ - for (i = 0; i < CDNS3_ENDPOINTS_MAX_COUNT; i++) { - priv_ep = priv_dev->eps[i]; - if (!priv_ep) - continue; - - if (priv_ep->flags & EP_CLAIMED) - cdns3_ep_config(priv_ep); - } - result = cdns3_ep0_delegate_req(priv_dev, ctrl_req); - if (result) - return result; - - if (!config) { - cdns3_hw_reset_eps_config(priv_dev); - usb_gadget_set_state(&priv_dev->gadget, - USB_STATE_ADDRESS); - } + if (result || !config) + goto reset_config; break; case USB_STATE_CONFIGURED: result = cdns3_ep0_delegate_req(priv_dev, ctrl_req); + if (!config && !result) + goto reset_config; - if (!config && !result) { - cdns3_hw_reset_eps_config(priv_dev); - usb_gadget_set_state(&priv_dev->gadget, - USB_STATE_ADDRESS); - } break; default: - result = -EINVAL; + return -EINVAL; } + return 0; + +reset_config: + if (result != USB_GADGET_DELAYED_STATUS) + cdns3_hw_reset_eps_config(priv_dev); + + usb_gadget_set_state(&priv_dev->gadget, + USB_STATE_ADDRESS); + return result; } @@ -705,6 +693,7 @@ static int cdns3_gadget_ep0_queue(struct usb_ep *ep, unsigned long flags; int ret = 0; u8 zlp = 0; + int i; spin_lock_irqsave(&priv_dev->lock, flags); trace_cdns3_ep0_queue(priv_dev, request); @@ -720,6 +709,17 @@ static int cdns3_gadget_ep0_queue(struct usb_ep *ep, u32 val; cdns3_select_ep(priv_dev, 0x00); + + /* + * Configure all non-control EPs which are not enabled by class driver + */ + for (i = 0; i < CDNS3_ENDPOINTS_MAX_COUNT; i++) { + priv_ep = priv_dev->eps[i]; + if (priv_ep && priv_ep->flags & EP_CLAIMED && + !(priv_ep->flags & EP_ENABLED)) + cdns3_ep_config(priv_ep, 0); + } + cdns3_set_hw_configuration(priv_dev); cdns3_ep0_complete_setup(priv_dev, 0, 1); /* wait until configuration set */ @@ -811,6 +811,7 @@ void cdns3_ep0_config(struct cdns3_device *priv_dev) struct cdns3_usb_regs __iomem *regs; struct cdns3_endpoint *priv_ep; u32 max_packet_size = 64; + u32 ep_cfg; regs = priv_dev->regs; @@ -842,8 +843,10 @@ void cdns3_ep0_config(struct cdns3_device *priv_dev) BIT(0) | BIT(16)); } - writel(EP_CFG_ENABLE | EP_CFG_MAXPKTSIZE(max_packet_size), - ®s->ep_cfg); + ep_cfg = EP_CFG_ENABLE | EP_CFG_MAXPKTSIZE(max_packet_size); + + if (!(priv_ep->flags & EP_CONFIGURED)) + writel(ep_cfg, ®s->ep_cfg); writel(EP_STS_EN_SETUPEN | EP_STS_EN_DESCMISEN | EP_STS_EN_TRBERREN, ®s->ep_sts_en); @@ -851,8 +854,10 @@ void cdns3_ep0_config(struct cdns3_device *priv_dev) /* init ep in */ cdns3_select_ep(priv_dev, USB_DIR_IN); - writel(EP_CFG_ENABLE | EP_CFG_MAXPKTSIZE(max_packet_size), - ®s->ep_cfg); + if (!(priv_ep->flags & EP_CONFIGURED)) + writel(ep_cfg, ®s->ep_cfg); + + priv_ep->flags |= EP_CONFIGURED; writel(EP_STS_EN_SETUPEN | EP_STS_EN_TRBERREN, ®s->ep_sts_en); diff --git a/drivers/usb/cdns3/gadget.c b/drivers/usb/cdns3/gadget.c index 692acf7b9b14..36c8c1aede5e 100644 --- a/drivers/usb/cdns3/gadget.c +++ b/drivers/usb/cdns3/gadget.c @@ -296,6 +296,8 @@ static void cdns3_ep_stall_flush(struct cdns3_endpoint *priv_ep) */ void cdns3_hw_reset_eps_config(struct cdns3_device *priv_dev) { + int i; + writel(USB_CONF_CFGRST, &priv_dev->regs->usb_conf); cdns3_allow_enable_l1(priv_dev, 0); @@ -304,6 +306,10 @@ void cdns3_hw_reset_eps_config(struct cdns3_device *priv_dev) priv_dev->out_mem_is_allocated = 0; priv_dev->wait_for_setup = 0; priv_dev->using_streams = 0; + + for (i = 0; i < CDNS3_ENDPOINTS_MAX_COUNT; i++) + if (priv_dev->eps[i]) + priv_dev->eps[i]->flags &= ~EP_CONFIGURED; } /** @@ -1977,27 +1983,6 @@ static int cdns3_ep_onchip_buffer_reserve(struct cdns3_device *priv_dev, return 0; } -static void cdns3_stream_ep_reconfig(struct cdns3_device *priv_dev, - struct cdns3_endpoint *priv_ep) -{ - if (!priv_ep->use_streams || priv_dev->gadget.speed < USB_SPEED_SUPER) - return; - - if (priv_dev->dev_ver >= DEV_VER_V3) { - u32 mask = BIT(priv_ep->num + (priv_ep->dir ? 16 : 0)); - - /* - * Stream capable endpoints are handled by using ep_tdl - * register. Other endpoints use TDL from TRB feature. - */ - cdns3_clear_register_bit(&priv_dev->regs->tdl_from_trb, mask); - } - - /* Enable Stream Bit TDL chk and SID chk */ - cdns3_set_register_bit(&priv_dev->regs->ep_cfg, EP_CFG_STREAM_EN | - EP_CFG_TDL_CHK | EP_CFG_SID_CHK); -} - static void cdns3_configure_dmult(struct cdns3_device *priv_dev, struct cdns3_endpoint *priv_ep) { @@ -2035,8 +2020,9 @@ static void cdns3_configure_dmult(struct cdns3_device *priv_dev, /** * cdns3_ep_config Configure hardware endpoint * @priv_ep: extended endpoint object + * @enable: set EP_CFG_ENABLE bit in ep_cfg register. */ -void cdns3_ep_config(struct cdns3_endpoint *priv_ep) +int cdns3_ep_config(struct cdns3_endpoint *priv_ep, bool enable) { bool is_iso_ep = (priv_ep->type == USB_ENDPOINT_XFER_ISOC); struct cdns3_device *priv_dev = priv_ep->cdns3_dev; @@ -2097,7 +2083,7 @@ void cdns3_ep_config(struct cdns3_endpoint *priv_ep) break; default: /* all other speed are not supported */ - return; + return -EINVAL; } if (max_packet_size == 1024) @@ -2107,11 +2093,33 @@ void cdns3_ep_config(struct cdns3_endpoint *priv_ep) else priv_ep->trb_burst_size = 16; - ret = cdns3_ep_onchip_buffer_reserve(priv_dev, buffering + 1, - !!priv_ep->dir); - if (ret) { - dev_err(priv_dev->dev, "onchip mem is full, ep is invalid\n"); - return; + /* onchip buffer is only allocated before configuration */ + if (!priv_dev->hw_configured_flag) { + ret = cdns3_ep_onchip_buffer_reserve(priv_dev, buffering + 1, + !!priv_ep->dir); + if (ret) { + dev_err(priv_dev->dev, "onchip mem is full, ep is invalid\n"); + return ret; + } + } + + if (enable) + ep_cfg |= EP_CFG_ENABLE; + + if (priv_ep->use_streams && priv_dev->gadget.speed >= USB_SPEED_SUPER) { + if (priv_dev->dev_ver >= DEV_VER_V3) { + u32 mask = BIT(priv_ep->num + (priv_ep->dir ? 16 : 0)); + + /* + * Stream capable endpoints are handled by using ep_tdl + * register. Other endpoints use TDL from TRB feature. + */ + cdns3_clear_register_bit(&priv_dev->regs->tdl_from_trb, + mask); + } + + /* Enable Stream Bit TDL chk and SID chk */ + ep_cfg |= EP_CFG_STREAM_EN | EP_CFG_TDL_CHK | EP_CFG_SID_CHK; } ep_cfg |= EP_CFG_MAXPKTSIZE(max_packet_size) | @@ -2121,9 +2129,12 @@ void cdns3_ep_config(struct cdns3_endpoint *priv_ep) cdns3_select_ep(priv_dev, bEndpointAddress); writel(ep_cfg, &priv_dev->regs->ep_cfg); + priv_ep->flags |= EP_CONFIGURED; dev_dbg(priv_dev->dev, "Configure %s: with val %08x\n", priv_ep->name, ep_cfg); + + return 0; } /* Find correct direction for HW endpoint according to description */ @@ -2264,7 +2275,7 @@ static int cdns3_gadget_ep_enable(struct usb_ep *ep, u32 bEndpointAddress; unsigned long flags; int enable = 1; - int ret; + int ret = 0; int val; priv_ep = ep_to_cdns3_ep(ep); @@ -2303,6 +2314,17 @@ static int cdns3_gadget_ep_enable(struct usb_ep *ep, bEndpointAddress = priv_ep->num | priv_ep->dir; cdns3_select_ep(priv_dev, bEndpointAddress); + /* + * For some versions of controller at some point during ISO OUT traffic + * DMA reads Transfer Ring for the EP which has never got doorbell. + * This issue was detected only on simulation, but to avoid this issue + * driver add protection against it. To fix it driver enable ISO OUT + * endpoint before setting DRBL. This special treatment of ISO OUT + * endpoints are recommended by controller specification. + */ + if (priv_ep->type == USB_ENDPOINT_XFER_ISOC && !priv_ep->dir) + enable = 0; + if (usb_ss_max_streams(comp_desc) && usb_endpoint_xfer_bulk(desc)) { /* * Enable stream support (SS mode) related interrupts @@ -2313,13 +2335,17 @@ static int cdns3_gadget_ep_enable(struct usb_ep *ep, EP_STS_EN_SIDERREN | EP_STS_EN_MD_EXITEN | EP_STS_EN_STREAMREN; priv_ep->use_streams = true; - cdns3_stream_ep_reconfig(priv_dev, priv_ep); + ret = cdns3_ep_config(priv_ep, enable); priv_dev->using_streams |= true; } + } else { + ret = cdns3_ep_config(priv_ep, enable); } - ret = cdns3_allocate_trb_pool(priv_ep); + if (ret) + goto exit; + ret = cdns3_allocate_trb_pool(priv_ep); if (ret) goto exit; @@ -2349,20 +2375,6 @@ static int cdns3_gadget_ep_enable(struct usb_ep *ep, writel(reg, &priv_dev->regs->ep_sts_en); - /* - * For some versions of controller at some point during ISO OUT traffic - * DMA reads Transfer Ring for the EP which has never got doorbell. - * This issue was detected only on simulation, but to avoid this issue - * driver add protection against it. To fix it driver enable ISO OUT - * endpoint before setting DRBL. This special treatment of ISO OUT - * endpoints are recommended by controller specification. - */ - if (priv_ep->type == USB_ENDPOINT_XFER_ISOC && !priv_ep->dir) - enable = 0; - - if (enable) - cdns3_set_register_bit(&priv_dev->regs->ep_cfg, EP_CFG_ENABLE); - ep->desc = desc; priv_ep->flags &= ~(EP_PENDING_REQUEST | EP_STALLED | EP_STALL_PENDING | EP_QUIRK_ISO_OUT_EN | EP_QUIRK_EXTRA_BUF_EN); diff --git a/drivers/usb/cdns3/gadget.h b/drivers/usb/cdns3/gadget.h index 737377913788..21fa461c518e 100644 --- a/drivers/usb/cdns3/gadget.h +++ b/drivers/usb/cdns3/gadget.h @@ -1159,6 +1159,7 @@ struct cdns3_endpoint { #define EP_QUIRK_EXTRA_BUF_DET BIT(12) #define EP_QUIRK_EXTRA_BUF_EN BIT(13) #define EP_TDLCHK_EN BIT(15) +#define EP_CONFIGURED BIT(16) u32 flags; struct cdns3_request *descmis_req; @@ -1360,7 +1361,7 @@ void cdns3_gadget_giveback(struct cdns3_endpoint *priv_ep, int cdns3_init_ep0(struct cdns3_device *priv_dev, struct cdns3_endpoint *priv_ep); void cdns3_ep0_config(struct cdns3_device *priv_dev); -void cdns3_ep_config(struct cdns3_endpoint *priv_ep); +int cdns3_ep_config(struct cdns3_endpoint *priv_ep, bool enable); void cdns3_check_ep0_interrupt_proceed(struct cdns3_device *priv_dev, int dir); int __cdns3_gadget_wakeup(struct cdns3_device *priv_dev); -- cgit v1.2.3 From e11d2bf271589e70ea80a2ee3e116c40fcac62c2 Mon Sep 17 00:00:00 2001 From: Peter Chen Date: Thu, 29 Oct 2020 17:55:18 +0800 Subject: usb: cdns3: gadget: own the lock wrongly at the suspend routine When the system goes to suspend, if the controller is at device mode with cable connecting to host, the call stack is: cdns3_suspend-> cdns3_gadget_suspend -> cdns3_disconnect_gadget, after cdns3_disconnect_gadget is called, it owns lock wrongly, it causes the system being deadlock after resume due to at cdns3_device_thread_irq_handler, it tries to get the lock, but can't get it forever. To fix it, we delete the unlock-lock operations at cdns3_disconnect_gadget, and do it at the caller. Fixes: b1234e3b3b26 ("usb: cdns3: add runtime PM support") Acked-by: Pawel Laszczak Signed-off-by: Peter Chen --- drivers/usb/cdns3/gadget.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/usb/cdns3/gadget.c b/drivers/usb/cdns3/gadget.c index 36c8c1aede5e..66c1e6723eb1 100644 --- a/drivers/usb/cdns3/gadget.c +++ b/drivers/usb/cdns3/gadget.c @@ -1750,11 +1750,8 @@ static int cdns3_check_ep_interrupt_proceed(struct cdns3_endpoint *priv_ep) static void cdns3_disconnect_gadget(struct cdns3_device *priv_dev) { - if (priv_dev->gadget_driver && priv_dev->gadget_driver->disconnect) { - spin_unlock(&priv_dev->lock); + if (priv_dev->gadget_driver && priv_dev->gadget_driver->disconnect) priv_dev->gadget_driver->disconnect(&priv_dev->gadget); - spin_lock(&priv_dev->lock); - } } /** @@ -1765,6 +1762,7 @@ static void cdns3_disconnect_gadget(struct cdns3_device *priv_dev) */ static void cdns3_check_usb_interrupt_proceed(struct cdns3_device *priv_dev, u32 usb_ists) +__must_hold(&priv_dev->lock) { int speed = 0; @@ -1789,7 +1787,9 @@ static void cdns3_check_usb_interrupt_proceed(struct cdns3_device *priv_dev, /* Disconnection detected */ if (usb_ists & (USB_ISTS_DIS2I | USB_ISTS_DISI)) { + spin_unlock(&priv_dev->lock); cdns3_disconnect_gadget(priv_dev); + spin_lock(&priv_dev->lock); priv_dev->gadget.speed = USB_SPEED_UNKNOWN; usb_gadget_set_state(&priv_dev->gadget, USB_STATE_NOTATTACHED); cdns3_hw_reset_eps_config(priv_dev); @@ -3275,10 +3275,13 @@ err0: } static int cdns3_gadget_suspend(struct cdns3 *cdns, bool do_wakeup) +__must_hold(&cdns->lock) { struct cdns3_device *priv_dev = cdns->gadget_dev; + spin_unlock(&cdns->lock); cdns3_disconnect_gadget(priv_dev); + spin_lock(&cdns->lock); priv_dev->gadget.speed = USB_SPEED_UNKNOWN; usb_gadget_set_state(&priv_dev->gadget, USB_STATE_NOTATTACHED); -- cgit v1.2.3 From 0d519cbf38eed4f895aed197d4b135fa7f60f7c2 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Fri, 23 Oct 2020 15:10:37 +0200 Subject: debugfs: remove return value of debugfs_create_devm_seqfile() No one checks the return value of debugfs_create_devm_seqfile(), as it's not needed, so make the return value void, so that no one tries to do so in the future. Link: https://lore.kernel.org/r/20201023131037.2500765-1-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman --- Documentation/filesystems/debugfs.rst | 2 +- fs/debugfs/file.c | 15 +++++++-------- include/linux/debugfs.h | 18 ++++++++---------- 3 files changed, 16 insertions(+), 19 deletions(-) diff --git a/Documentation/filesystems/debugfs.rst b/Documentation/filesystems/debugfs.rst index 728ab57a611a..0f2292e367e6 100644 --- a/Documentation/filesystems/debugfs.rst +++ b/Documentation/filesystems/debugfs.rst @@ -199,7 +199,7 @@ of its elements. Note: Once array is created its size can not be changed. There is a helper function to create device related seq_file:: - struct dentry *debugfs_create_devm_seqfile(struct device *dev, + void debugfs_create_devm_seqfile(struct device *dev, const char *name, struct dentry *parent, int (*read_fn)(struct seq_file *s, diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index a768a09430c3..686e0ad28788 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -1127,24 +1127,23 @@ static const struct file_operations debugfs_devm_entry_ops = { * file will be created in the root of the debugfs filesystem. * @read_fn: function pointer called to print the seq_file content. */ -struct dentry *debugfs_create_devm_seqfile(struct device *dev, const char *name, - struct dentry *parent, - int (*read_fn)(struct seq_file *s, - void *data)) +void debugfs_create_devm_seqfile(struct device *dev, const char *name, + struct dentry *parent, + int (*read_fn)(struct seq_file *s, void *data)) { struct debugfs_devm_entry *entry; if (IS_ERR(parent)) - return ERR_PTR(-ENOENT); + return; entry = devm_kzalloc(dev, sizeof(*entry), GFP_KERNEL); if (!entry) - return ERR_PTR(-ENOMEM); + return; entry->read = read_fn; entry->dev = dev; - return debugfs_create_file(name, S_IRUGO, parent, entry, - &debugfs_devm_entry_ops); + debugfs_create_file(name, S_IRUGO, parent, entry, + &debugfs_devm_entry_ops); } EXPORT_SYMBOL_GPL(debugfs_create_devm_seqfile); diff --git a/include/linux/debugfs.h b/include/linux/debugfs.h index 851dd1f9a8a5..d6c4cc9ecc77 100644 --- a/include/linux/debugfs.h +++ b/include/linux/debugfs.h @@ -144,10 +144,9 @@ void debugfs_create_u32_array(const char *name, umode_t mode, struct dentry *parent, struct debugfs_u32_array *array); -struct dentry *debugfs_create_devm_seqfile(struct device *dev, const char *name, - struct dentry *parent, - int (*read_fn)(struct seq_file *s, - void *data)); +void debugfs_create_devm_seqfile(struct device *dev, const char *name, + struct dentry *parent, + int (*read_fn)(struct seq_file *s, void *data)); bool debugfs_initialized(void); @@ -327,13 +326,12 @@ static inline void debugfs_create_u32_array(const char *name, umode_t mode, { } -static inline struct dentry *debugfs_create_devm_seqfile(struct device *dev, - const char *name, - struct dentry *parent, - int (*read_fn)(struct seq_file *s, - void *data)) +static inline void debugfs_create_devm_seqfile(struct device *dev, + const char *name, + struct dentry *parent, + int (*read_fn)(struct seq_file *s, + void *data)) { - return ERR_PTR(-ENODEV); } static inline ssize_t debugfs_read_file_bool(struct file *file, -- cgit v1.2.3 From 7922460e33c81f41e0d2421417228b32e6fdbe94 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 23 Oct 2020 15:08:53 +0300 Subject: vhost_vdpa: Return -EFAULT if copy_from_user() fails The copy_to/from_user() functions return the number of bytes which we weren't able to copy but the ioctl should return -EFAULT if they fail. Fixes: a127c5bbb6a8 ("vhost-vdpa: fix backend feature ioctls") Signed-off-by: Dan Carpenter Link: https://lore.kernel.org/r/20201023120853.GI282278@mwanda Signed-off-by: Michael S. Tsirkin Cc: stable@vger.kernel.org Acked-by: Jason Wang --- drivers/vhost/vdpa.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c index 846de69d9c01..37f461cff892 100644 --- a/drivers/vhost/vdpa.c +++ b/drivers/vhost/vdpa.c @@ -432,12 +432,11 @@ static long vhost_vdpa_unlocked_ioctl(struct file *filep, void __user *argp = (void __user *)arg; u64 __user *featurep = argp; u64 features; - long r; + long r = 0; if (cmd == VHOST_SET_BACKEND_FEATURES) { - r = copy_from_user(&features, featurep, sizeof(features)); - if (r) - return r; + if (copy_from_user(&features, featurep, sizeof(features))) + return -EFAULT; if (features & ~VHOST_VDPA_BACKEND_FEATURES) return -EOPNOTSUPP; vhost_set_backend_features(&v->vdev, features); @@ -480,7 +479,8 @@ static long vhost_vdpa_unlocked_ioctl(struct file *filep, break; case VHOST_GET_BACKEND_FEATURES: features = VHOST_VDPA_BACKEND_FEATURES; - r = copy_to_user(featurep, &features, sizeof(features)); + if (copy_to_user(featurep, &features, sizeof(features))) + r = -EFAULT; break; case VHOST_VDPA_GET_IOVA_RANGE: r = vhost_vdpa_get_iova_range(v, argp); -- cgit v1.2.3 From 7ba08e81cb4aec9724ab7674a5de49e7a341062c Mon Sep 17 00:00:00 2001 From: Jing Xiangfeng Date: Mon, 26 Oct 2020 15:06:37 +0800 Subject: vdpa/mlx5: Fix error return in map_direct_mr() Fix to return the variable "err" from the error handling case instead of "ret". Fixes: 94abbccdf291 ("vdpa/mlx5: Add shared memory registration code") Signed-off-by: Jing Xiangfeng Link: https://lore.kernel.org/r/20201026070637.164321-1-jingxiangfeng@huawei.com Signed-off-by: Michael S. Tsirkin Acked-by: Eli Cohen Cc: stable@vger.kernel.org Acked-by: Jason Wang --- drivers/vdpa/mlx5/core/mr.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/vdpa/mlx5/core/mr.c b/drivers/vdpa/mlx5/core/mr.c index ef1c550f8266..4b6195666c58 100644 --- a/drivers/vdpa/mlx5/core/mr.c +++ b/drivers/vdpa/mlx5/core/mr.c @@ -239,7 +239,6 @@ static int map_direct_mr(struct mlx5_vdpa_dev *mvdev, struct mlx5_vdpa_direct_mr u64 paend; struct scatterlist *sg; struct device *dma = mvdev->mdev->device; - int ret; for (map = vhost_iotlb_itree_first(iotlb, mr->start, mr->end - 1); map; map = vhost_iotlb_itree_next(map, start, mr->end - 1)) { @@ -277,8 +276,8 @@ static int map_direct_mr(struct mlx5_vdpa_dev *mvdev, struct mlx5_vdpa_direct_mr done: mr->log_size = log_entity_size; mr->nsg = nsg; - ret = dma_map_sg_attrs(dma, mr->sg_head.sgl, mr->nsg, DMA_BIDIRECTIONAL, 0); - if (!ret) + err = dma_map_sg_attrs(dma, mr->sg_head.sgl, mr->nsg, DMA_BIDIRECTIONAL, 0); + if (!err) goto err_map; err = create_direct_mr(mvdev, mr); -- cgit v1.2.3 From 5e1a3149eec8675c2767cc465903f5e4829de5b0 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Thu, 29 Oct 2020 17:53:36 -0400 Subject: Revert "vhost-vdpa: fix page pinning leakage in error path" This reverts commit 7ed9e3d97c32d969caded2dfb6e67c1a2cc5a0b1. The patch creates a DoS risk since it can result in a high order memory allocation. Fixes: 7ed9e3d97c32d ("vhost-vdpa: fix page pinning leakage in error path") Cc: stable@vger.kernel.org Signed-off-by: Michael S. Tsirkin --- drivers/vhost/vdpa.c | 119 +++++++++++++++++++++------------------------------ 1 file changed, 48 insertions(+), 71 deletions(-) diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c index 37f461cff892..ec223da70a73 100644 --- a/drivers/vhost/vdpa.c +++ b/drivers/vhost/vdpa.c @@ -602,13 +602,11 @@ static int vhost_vdpa_process_iotlb_update(struct vhost_vdpa *v, struct vhost_dev *dev = &v->vdev; struct vhost_iotlb *iotlb = dev->iotlb; struct page **page_list; - struct vm_area_struct **vmas; + unsigned long list_size = PAGE_SIZE / sizeof(struct page *); unsigned int gup_flags = FOLL_LONGTERM; - unsigned long map_pfn, last_pfn = 0; - unsigned long npages, lock_limit; - unsigned long i, nmap = 0; + unsigned long npages, cur_base, map_pfn, last_pfn = 0; + unsigned long locked, lock_limit, pinned, i; u64 iova = msg->iova; - long pinned; int ret = 0; if (msg->iova < v->range.first || @@ -619,6 +617,10 @@ static int vhost_vdpa_process_iotlb_update(struct vhost_vdpa *v, msg->iova + msg->size - 1)) return -EEXIST; + page_list = (struct page **) __get_free_page(GFP_KERNEL); + if (!page_list) + return -ENOMEM; + if (msg->perm & VHOST_ACCESS_WO) gup_flags |= FOLL_WRITE; @@ -626,86 +628,61 @@ static int vhost_vdpa_process_iotlb_update(struct vhost_vdpa *v, if (!npages) return -EINVAL; - page_list = kvmalloc_array(npages, sizeof(struct page *), GFP_KERNEL); - vmas = kvmalloc_array(npages, sizeof(struct vm_area_struct *), - GFP_KERNEL); - if (!page_list || !vmas) { - ret = -ENOMEM; - goto free; - } - mmap_read_lock(dev->mm); + locked = atomic64_add_return(npages, &dev->mm->pinned_vm); lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT; - if (npages + atomic64_read(&dev->mm->pinned_vm) > lock_limit) { - ret = -ENOMEM; - goto unlock; - } - pinned = pin_user_pages(msg->uaddr & PAGE_MASK, npages, gup_flags, - page_list, vmas); - if (npages != pinned) { - if (pinned < 0) { - ret = pinned; - } else { - unpin_user_pages(page_list, pinned); - ret = -ENOMEM; - } - goto unlock; + if (locked > lock_limit) { + ret = -ENOMEM; + goto out; } + cur_base = msg->uaddr & PAGE_MASK; iova &= PAGE_MASK; - map_pfn = page_to_pfn(page_list[0]); - - /* One more iteration to avoid extra vdpa_map() call out of loop. */ - for (i = 0; i <= npages; i++) { - unsigned long this_pfn; - u64 csize; - - /* The last chunk may have no valid PFN next to it */ - this_pfn = i < npages ? page_to_pfn(page_list[i]) : -1UL; - - if (last_pfn && (this_pfn == -1UL || - this_pfn != last_pfn + 1)) { - /* Pin a contiguous chunk of memory */ - csize = last_pfn - map_pfn + 1; - ret = vhost_vdpa_map(v, iova, csize << PAGE_SHIFT, - map_pfn << PAGE_SHIFT, - msg->perm); - if (ret) { - /* - * Unpin the rest chunks of memory on the - * flight with no corresponding vdpa_map() - * calls having been made yet. On the other - * hand, vdpa_unmap() in the failure path - * is in charge of accounting the number of - * pinned pages for its own. - * This asymmetrical pattern of accounting - * is for efficiency to pin all pages at - * once, while there is no other callsite - * of vdpa_map() than here above. - */ - unpin_user_pages(&page_list[nmap], - npages - nmap); - goto out; + + while (npages) { + pinned = min_t(unsigned long, npages, list_size); + ret = pin_user_pages(cur_base, pinned, + gup_flags, page_list, NULL); + if (ret != pinned) + goto out; + + if (!last_pfn) + map_pfn = page_to_pfn(page_list[0]); + + for (i = 0; i < ret; i++) { + unsigned long this_pfn = page_to_pfn(page_list[i]); + u64 csize; + + if (last_pfn && (this_pfn != last_pfn + 1)) { + /* Pin a contiguous chunk of memory */ + csize = (last_pfn - map_pfn + 1) << PAGE_SHIFT; + if (vhost_vdpa_map(v, iova, csize, + map_pfn << PAGE_SHIFT, + msg->perm)) + goto out; + map_pfn = this_pfn; + iova += csize; } - atomic64_add(csize, &dev->mm->pinned_vm); - nmap += csize; - iova += csize << PAGE_SHIFT; - map_pfn = this_pfn; + + last_pfn = this_pfn; } - last_pfn = this_pfn; + + cur_base += ret << PAGE_SHIFT; + npages -= ret; } - WARN_ON(nmap != npages); + /* Pin the rest chunk */ + ret = vhost_vdpa_map(v, iova, (last_pfn - map_pfn + 1) << PAGE_SHIFT, + map_pfn << PAGE_SHIFT, msg->perm); out: - if (ret) + if (ret) { vhost_vdpa_unmap(v, msg->iova, msg->size); -unlock: + atomic64_sub(npages, &dev->mm->pinned_vm); + } mmap_read_unlock(dev->mm); -free: - kvfree(vmas); - kvfree(page_list); + free_page((unsigned long)page_list); return ret; } -- cgit v1.2.3 From 1eca16b231570c8ae57fb91fdfbc48eb52c6a93b Mon Sep 17 00:00:00 2001 From: Laurent Vivier Date: Tue, 27 Oct 2020 18:59:14 +0100 Subject: vdpa_sim: Fix DMA mask Since commit f959dcd6ddfd ("dma-direct: Fix potential NULL pointer dereference") an error is reported when we load vdpa_sim and virtio-vdpa: [ 129.351207] net eth0: Unexpected TXQ (0) queue failure: -12 It seems that dma_mask is not initialized. This patch initializes dma_mask() and calls dma_set_mask_and_coherent() to fix the problem. Full log: [ 128.548628] ------------[ cut here ]------------ [ 128.553268] WARNING: CPU: 23 PID: 1105 at kernel/dma/mapping.c:149 dma_map_page_attrs+0x14c/0x1d0 [ 128.562139] Modules linked in: virtio_net net_failover failover virtio_vdpa vdpa_sim vringh vhost_iotlb vdpa xt_CHECKSUM xt_MASQUERADE xt_conntrack ipt_REJECT nf_reject_ipv4 nft_compat nft_counter nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 nf_tables nfnetlink tun bridge stp llc iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi rfkill intel_rapl_msr intel_rapl_common isst_if_common sunrpc skx_edac nfit libnvdimm x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel ipmi_ssif kvm mgag200 i2c_algo_bit irqbypass drm_kms_helper crct10dif_pclmul crc32_pclmul syscopyarea ghash_clmulni_intel iTCO_wdt sysfillrect iTCO_vendor_support sysimgblt rapl fb_sys_fops dcdbas intel_cstate drm acpi_ipmi ipmi_si mei_me dell_smbios intel_uncore ipmi_devintf mei i2c_i801 dell_wmi_descriptor wmi_bmof pcspkr lpc_ich i2c_smbus ipmi_msghandler acpi_power_meter ip_tables xfs libcrc32c sd_mod t10_pi sg ahci libahci libata megaraid_sas tg3 crc32c_intel wmi dm_mirror dm_region_hash dm_log [ 128.562188] dm_mod [ 128.651334] CPU: 23 PID: 1105 Comm: NetworkManager Tainted: G S I 5.10.0-rc1+ #59 [ 128.659939] Hardware name: Dell Inc. PowerEdge R440/04JN2K, BIOS 2.8.1 06/30/2020 [ 128.667419] RIP: 0010:dma_map_page_attrs+0x14c/0x1d0 [ 128.672384] Code: 1c 25 28 00 00 00 0f 85 97 00 00 00 48 83 c4 10 5b 5d 41 5c 41 5d c3 4c 89 da eb d7 48 89 f2 48 2b 50 18 48 89 d0 eb 8d 0f 0b <0f> 0b 48 c7 c0 ff ff ff ff eb c3 48 89 d9 48 8b 40 40 e8 2d a0 aa [ 128.691131] RSP: 0018:ffffae0f0151f3c8 EFLAGS: 00010246 [ 128.696357] RAX: ffffffffc06b7400 RBX: 00000000000005fa RCX: 0000000000000000 [ 128.703488] RDX: 0000000000000040 RSI: ffffcee3c7861200 RDI: ffff9e2bc16cd000 [ 128.710620] RBP: 0000000000000000 R08: 0000000000000002 R09: 0000000000000000 [ 128.717754] R10: 0000000000000002 R11: 0000000000000000 R12: ffff9e472cb291f8 [ 128.724886] R13: ffff9e2bc14da780 R14: ffff9e472bc20000 R15: ffff9e2bc1b14940 [ 128.732020] FS: 00007f887bae23c0(0000) GS:ffff9e4ac01c0000(0000) knlGS:0000000000000000 [ 128.740105] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 128.745852] CR2: 0000562bc09de998 CR3: 00000003c156c006 CR4: 00000000007706e0 [ 128.752982] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 128.760114] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 128.767247] PKRU: 55555554 [ 128.769961] Call Trace: [ 128.772418] virtqueue_add+0x81e/0xb00 [ 128.776176] virtqueue_add_inbuf_ctx+0x26/0x30 [ 128.780625] try_fill_recv+0x3a2/0x6e0 [virtio_net] [ 128.785509] virtnet_open+0xf9/0x180 [virtio_net] [ 128.790217] __dev_open+0xe8/0x180 [ 128.793620] __dev_change_flags+0x1a7/0x210 [ 128.797808] dev_change_flags+0x21/0x60 [ 128.801646] do_setlink+0x328/0x10e0 [ 128.805227] ? __nla_validate_parse+0x121/0x180 [ 128.809757] ? __nla_parse+0x21/0x30 [ 128.813338] ? inet6_validate_link_af+0x5c/0xf0 [ 128.817871] ? cpumask_next+0x17/0x20 [ 128.821535] ? __snmp6_fill_stats64.isra.54+0x6b/0x110 [ 128.826676] ? __nla_validate_parse+0x47/0x180 [ 128.831120] __rtnl_newlink+0x541/0x8e0 [ 128.834962] ? __nla_reserve+0x38/0x50 [ 128.838713] ? security_sock_rcv_skb+0x2a/0x40 [ 128.843158] ? netlink_deliver_tap+0x2c/0x1e0 [ 128.847518] ? netlink_attachskb+0x1d8/0x220 [ 128.851793] ? skb_queue_tail+0x1b/0x50 [ 128.855641] ? fib6_clean_node+0x43/0x170 [ 128.859652] ? _cond_resched+0x15/0x30 [ 128.863406] ? kmem_cache_alloc_trace+0x3a3/0x420 [ 128.868110] rtnl_newlink+0x43/0x60 [ 128.871602] rtnetlink_rcv_msg+0x12c/0x380 [ 128.875701] ? rtnl_calcit.isra.39+0x110/0x110 [ 128.880147] netlink_rcv_skb+0x50/0x100 [ 128.883987] netlink_unicast+0x1a5/0x280 [ 128.887913] netlink_sendmsg+0x23d/0x470 [ 128.891839] sock_sendmsg+0x5b/0x60 [ 128.895331] ____sys_sendmsg+0x1ef/0x260 [ 128.899255] ? copy_msghdr_from_user+0x5c/0x90 [ 128.903702] ___sys_sendmsg+0x7c/0xc0 [ 128.907369] ? dev_forward_change+0x130/0x130 [ 128.911731] ? sysctl_head_finish.part.29+0x24/0x40 [ 128.916616] ? new_sync_write+0x11f/0x1b0 [ 128.920628] ? mntput_no_expire+0x47/0x240 [ 128.924727] __sys_sendmsg+0x57/0xa0 [ 128.928309] do_syscall_64+0x33/0x40 [ 128.931887] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 128.936937] RIP: 0033:0x7f88792e3857 [ 128.940518] Code: c3 66 90 41 54 41 89 d4 55 48 89 f5 53 89 fb 48 83 ec 10 e8 0b ed ff ff 44 89 e2 48 89 ee 89 df 41 89 c0 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 35 44 89 c7 48 89 44 24 08 e8 44 ed ff ff 48 [ 128.959263] RSP: 002b:00007ffdca60dea0 EFLAGS: 00000293 ORIG_RAX: 000000000000002e [ 128.966827] RAX: ffffffffffffffda RBX: 000000000000000c RCX: 00007f88792e3857 [ 128.973960] RDX: 0000000000000000 RSI: 00007ffdca60def0 RDI: 000000000000000c [ 128.981095] RBP: 00007ffdca60def0 R08: 0000000000000000 R09: 0000000000000000 [ 128.988224] R10: 0000000000000001 R11: 0000000000000293 R12: 0000000000000000 [ 128.995357] R13: 0000000000000000 R14: 00007ffdca60e0a8 R15: 00007ffdca60e09c [ 129.002492] CPU: 23 PID: 1105 Comm: NetworkManager Tainted: G S I 5.10.0-rc1+ #59 [ 129.011093] Hardware name: Dell Inc. PowerEdge R440/04JN2K, BIOS 2.8.1 06/30/2020 [ 129.018571] Call Trace: [ 129.021027] dump_stack+0x57/0x6a [ 129.024346] __warn.cold.14+0xe/0x3d [ 129.027925] ? dma_map_page_attrs+0x14c/0x1d0 [ 129.032283] report_bug+0xbd/0xf0 [ 129.035602] handle_bug+0x44/0x80 [ 129.038922] exc_invalid_op+0x13/0x60 [ 129.042589] asm_exc_invalid_op+0x12/0x20 [ 129.046602] RIP: 0010:dma_map_page_attrs+0x14c/0x1d0 [ 129.051566] Code: 1c 25 28 00 00 00 0f 85 97 00 00 00 48 83 c4 10 5b 5d 41 5c 41 5d c3 4c 89 da eb d7 48 89 f2 48 2b 50 18 48 89 d0 eb 8d 0f 0b <0f> 0b 48 c7 c0 ff ff ff ff eb c3 48 89 d9 48 8b 40 40 e8 2d a0 aa [ 129.070311] RSP: 0018:ffffae0f0151f3c8 EFLAGS: 00010246 [ 129.075536] RAX: ffffffffc06b7400 RBX: 00000000000005fa RCX: 0000000000000000 [ 129.082669] RDX: 0000000000000040 RSI: ffffcee3c7861200 RDI: ffff9e2bc16cd000 [ 129.089803] RBP: 0000000000000000 R08: 0000000000000002 R09: 0000000000000000 [ 129.096936] R10: 0000000000000002 R11: 0000000000000000 R12: ffff9e472cb291f8 [ 129.104068] R13: ffff9e2bc14da780 R14: ffff9e472bc20000 R15: ffff9e2bc1b14940 [ 129.111200] virtqueue_add+0x81e/0xb00 [ 129.114952] virtqueue_add_inbuf_ctx+0x26/0x30 [ 129.119399] try_fill_recv+0x3a2/0x6e0 [virtio_net] [ 129.124280] virtnet_open+0xf9/0x180 [virtio_net] [ 129.128984] __dev_open+0xe8/0x180 [ 129.132390] __dev_change_flags+0x1a7/0x210 [ 129.136575] dev_change_flags+0x21/0x60 [ 129.140415] do_setlink+0x328/0x10e0 [ 129.143994] ? __nla_validate_parse+0x121/0x180 [ 129.148528] ? __nla_parse+0x21/0x30 [ 129.152107] ? inet6_validate_link_af+0x5c/0xf0 [ 129.156639] ? cpumask_next+0x17/0x20 [ 129.160306] ? __snmp6_fill_stats64.isra.54+0x6b/0x110 [ 129.165443] ? __nla_validate_parse+0x47/0x180 [ 129.169890] __rtnl_newlink+0x541/0x8e0 [ 129.173731] ? __nla_reserve+0x38/0x50 [ 129.177483] ? security_sock_rcv_skb+0x2a/0x40 [ 129.181928] ? netlink_deliver_tap+0x2c/0x1e0 [ 129.186286] ? netlink_attachskb+0x1d8/0x220 [ 129.190560] ? skb_queue_tail+0x1b/0x50 [ 129.194401] ? fib6_clean_node+0x43/0x170 [ 129.198411] ? _cond_resched+0x15/0x30 [ 129.202163] ? kmem_cache_alloc_trace+0x3a3/0x420 [ 129.206869] rtnl_newlink+0x43/0x60 [ 129.210361] rtnetlink_rcv_msg+0x12c/0x380 [ 129.214462] ? rtnl_calcit.isra.39+0x110/0x110 [ 129.218908] netlink_rcv_skb+0x50/0x100 [ 129.222747] netlink_unicast+0x1a5/0x280 [ 129.226672] netlink_sendmsg+0x23d/0x470 [ 129.230599] sock_sendmsg+0x5b/0x60 [ 129.234090] ____sys_sendmsg+0x1ef/0x260 [ 129.238015] ? copy_msghdr_from_user+0x5c/0x90 [ 129.242461] ___sys_sendmsg+0x7c/0xc0 [ 129.246128] ? dev_forward_change+0x130/0x130 [ 129.250487] ? sysctl_head_finish.part.29+0x24/0x40 [ 129.255368] ? new_sync_write+0x11f/0x1b0 [ 129.259381] ? mntput_no_expire+0x47/0x240 [ 129.263478] __sys_sendmsg+0x57/0xa0 [ 129.267058] do_syscall_64+0x33/0x40 [ 129.270639] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 129.275689] RIP: 0033:0x7f88792e3857 [ 129.279268] Code: c3 66 90 41 54 41 89 d4 55 48 89 f5 53 89 fb 48 83 ec 10 e8 0b ed ff ff 44 89 e2 48 89 ee 89 df 41 89 c0 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 35 44 89 c7 48 89 44 24 08 e8 44 ed ff ff 48 [ 129.298015] RSP: 002b:00007ffdca60dea0 EFLAGS: 00000293 ORIG_RAX: 000000000000002e [ 129.305581] RAX: ffffffffffffffda RBX: 000000000000000c RCX: 00007f88792e3857 [ 129.312712] RDX: 0000000000000000 RSI: 00007ffdca60def0 RDI: 000000000000000c [ 129.319846] RBP: 00007ffdca60def0 R08: 0000000000000000 R09: 0000000000000000 [ 129.326978] R10: 0000000000000001 R11: 0000000000000293 R12: 0000000000000000 [ 129.334109] R13: 0000000000000000 R14: 00007ffdca60e0a8 R15: 00007ffdca60e09c [ 129.341249] ---[ end trace c551e8028fbaf59d ]--- [ 129.351207] net eth0: Unexpected TXQ (0) queue failure: -12 [ 129.360445] net eth0: Unexpected TXQ (0) queue failure: -12 [ 129.824428] net eth0: Unexpected TXQ (0) queue failure: -12 Fixes: 2c53d0f64c06 ("vdpasim: vDPA device simulator") Signed-off-by: Laurent Vivier Link: https://lore.kernel.org/r/20201027175914.689278-1-lvivier@redhat.com Signed-off-by: Michael S. Tsirkin Cc: stable@vger.kernel.org Acked-by: Jason Wang --- drivers/vdpa/vdpa_sim/vdpa_sim.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c index ff6c9fd8d879..c76f7277cea4 100644 --- a/drivers/vdpa/vdpa_sim/vdpa_sim.c +++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c @@ -361,7 +361,9 @@ static struct vdpasim *vdpasim_create(void) spin_lock_init(&vdpasim->iommu_lock); dev = &vdpasim->vdpa.dev; - dev->coherent_dma_mask = DMA_BIT_MASK(64); + dev->dma_mask = &dev->coherent_dma_mask; + if (dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64))) + goto err_iommu; set_dma_ops(dev, &vdpasim_dma_ops); vdpasim->iommu = vhost_iotlb_alloc(2048, 0); -- cgit v1.2.3 From e01afe36df1a9e42cacdb9c5b99512ec69a6f14b Mon Sep 17 00:00:00 2001 From: Zhu Lingshan Date: Fri, 23 Oct 2020 18:40:46 +0800 Subject: vdpa: handle irq bypass register failure case LKP considered variable 'ret' in vhost_vdpa_setup_vq_irq() as a unused variable, so suggest we remove it. Actually it stores return value of irq_bypass_register_producer(), but we did not check it, we should handle the failure case. This commit will print a message if irq bypass register producer fail, in this case, vqs still remain functional. Signed-off-by: Zhu Lingshan Reported-by: kernel test robot Link: https://lore.kernel.org/r/20201023104046.404794-1-lingshan.zhu@intel.com Signed-off-by: Michael S. Tsirkin Acked-by: Jason Wang --- drivers/vhost/vdpa.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c index ec223da70a73..2754f3069738 100644 --- a/drivers/vhost/vdpa.c +++ b/drivers/vhost/vdpa.c @@ -104,6 +104,9 @@ static void vhost_vdpa_setup_vq_irq(struct vhost_vdpa *v, u16 qid) vq->call_ctx.producer.token = vq->call_ctx.ctx; vq->call_ctx.producer.irq = irq; ret = irq_bypass_register_producer(&vq->call_ctx.producer); + if (unlikely(ret)) + dev_info(&v->dev, "vq %u, irq bypass producer (token %p) registration fails, ret = %d\n", + qid, vq->call_ctx.producer.token, ret); } static void vhost_vdpa_unsetup_vq_irq(struct vhost_vdpa *v, u16 qid) -- cgit v1.2.3 From 4a6a42db53aae049a8a64d4b273761bc80c46ebf Mon Sep 17 00:00:00 2001 From: Laurent Vivier Date: Thu, 29 Oct 2020 13:20:49 +0100 Subject: vdpasim: fix MAC address configuration vdpa_sim generates a ramdom MAC address but it is never used by upper layers because the VIRTIO_NET_F_MAC bit is not set in the features list. Because of that, virtio-net always regenerates a random MAC address each time it is loaded whereas the address should only change on vdpa_sim load/unload. Fix that by adding VIRTIO_NET_F_MAC in the features list of vdpa_sim. Fixes: 2c53d0f64c06 ("vdpasim: vDPA device simulator") Cc: jasowang@redhat.com Signed-off-by: Laurent Vivier Link: https://lore.kernel.org/r/20201029122050.776445-2-lvivier@redhat.com Signed-off-by: Michael S. Tsirkin Acked-by: Jason Wang --- drivers/vdpa/vdpa_sim/vdpa_sim.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c index c76f7277cea4..3192ce47e029 100644 --- a/drivers/vdpa/vdpa_sim/vdpa_sim.c +++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c @@ -60,7 +60,8 @@ struct vdpasim_virtqueue { static u64 vdpasim_features = (1ULL << VIRTIO_F_ANY_LAYOUT) | (1ULL << VIRTIO_F_VERSION_1) | - (1ULL << VIRTIO_F_ACCESS_PLATFORM); + (1ULL << VIRTIO_F_ACCESS_PLATFORM) | + (1ULL << VIRTIO_NET_F_MAC); /* State of each vdpasim device */ struct vdpasim { -- cgit v1.2.3 From 0c86d774883fa17e7c81b0c8838b88d06c2c911e Mon Sep 17 00:00:00 2001 From: Laurent Vivier Date: Thu, 29 Oct 2020 13:20:50 +0100 Subject: vdpasim: allow to assign a MAC address Add macaddr parameter to the module to set the MAC address to use Signed-off-by: Laurent Vivier Link: https://lore.kernel.org/r/20201029122050.776445-3-lvivier@redhat.com Signed-off-by: Michael S. Tsirkin Acked-by: Jason Wang --- drivers/vdpa/vdpa_sim/vdpa_sim.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c index 3192ce47e029..9e9ead83c19b 100644 --- a/drivers/vdpa/vdpa_sim/vdpa_sim.c +++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c @@ -38,6 +38,10 @@ static int batch_mapping = 1; module_param(batch_mapping, int, 0444); MODULE_PARM_DESC(batch_mapping, "Batched mapping 1 -Enable; 0 - Disable"); +static char *macaddr; +module_param(macaddr, charp, 0); +MODULE_PARM_DESC(macaddr, "Ethernet MAC address"); + struct vdpasim_virtqueue { struct vringh vring; struct vringh_kiov iov; @@ -375,7 +379,15 @@ static struct vdpasim *vdpasim_create(void) if (!vdpasim->buffer) goto err_iommu; - eth_random_addr(vdpasim->config.mac); + if (macaddr) { + mac_pton(macaddr, vdpasim->config.mac); + if (!is_valid_ether_addr(vdpasim->config.mac)) { + ret = -EADDRNOTAVAIL; + goto err_iommu; + } + } else { + eth_random_addr(vdpasim->config.mac); + } vringh_set_iotlb(&vdpasim->vqs[0].vring, vdpasim->iommu); vringh_set_iotlb(&vdpasim->vqs[1].vring, vdpasim->iommu); -- cgit v1.2.3 From ce3d31ad3cac765484463b4f5a0b6b1f8f1a963e Mon Sep 17 00:00:00 2001 From: Qian Cai Date: Wed, 28 Oct 2020 14:26:14 -0400 Subject: arm64/smp: Move rcu_cpu_starting() earlier The call to rcu_cpu_starting() in secondary_start_kernel() is not early enough in the CPU-hotplug onlining process, which results in lockdep splats as follows: WARNING: suspicious RCU usage ----------------------------- kernel/locking/lockdep.c:3497 RCU-list traversed in non-reader section!! other info that might help us debug this: RCU used illegally from offline CPU! rcu_scheduler_active = 1, debug_locks = 1 no locks held by swapper/1/0. Call trace: dump_backtrace+0x0/0x3c8 show_stack+0x14/0x60 dump_stack+0x14c/0x1c4 lockdep_rcu_suspicious+0x134/0x14c __lock_acquire+0x1c30/0x2600 lock_acquire+0x274/0xc48 _raw_spin_lock+0xc8/0x140 vprintk_emit+0x90/0x3d0 vprintk_default+0x34/0x40 vprintk_func+0x378/0x590 printk+0xa8/0xd4 __cpuinfo_store_cpu+0x71c/0x868 cpuinfo_store_cpu+0x2c/0xc8 secondary_start_kernel+0x244/0x318 This is avoided by moving the call to rcu_cpu_starting up near the beginning of the secondary_start_kernel() function. Signed-off-by: Qian Cai Acked-by: Paul E. McKenney Link: https://lore.kernel.org/lkml/160223032121.7002.1269740091547117869.tip-bot2@tip-bot2/ Link: https://lore.kernel.org/r/20201028182614.13655-1-cai@redhat.com Signed-off-by: Will Deacon --- arch/arm64/kernel/smp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm64/kernel/smp.c b/arch/arm64/kernel/smp.c index 82e75fc2c903..09c96f57818c 100644 --- a/arch/arm64/kernel/smp.c +++ b/arch/arm64/kernel/smp.c @@ -222,6 +222,7 @@ asmlinkage notrace void secondary_start_kernel(void) if (system_uses_irq_prio_masking()) init_gic_priority_masking(); + rcu_cpu_starting(cpu); preempt_disable(); trace_hardirqs_off(); -- cgit v1.2.3 From ec9d78070de986ecf581ea204fd322af4d2477ec Mon Sep 17 00:00:00 2001 From: Fangrui Song Date: Thu, 29 Oct 2020 11:19:51 -0700 Subject: arm64: Change .weak to SYM_FUNC_START_WEAK_PI for arch/arm64/lib/mem*.S Commit 39d114ddc682 ("arm64: add KASAN support") added .weak directives to arch/arm64/lib/mem*.S instead of changing the existing SYM_FUNC_START_PI macros. This can lead to the assembly snippet `.weak memcpy ... .globl memcpy` which will produce a STB_WEAK memcpy with GNU as but STB_GLOBAL memcpy with LLVM's integrated assembler before LLVM 12. LLVM 12 (since https://reviews.llvm.org/D90108) will error on such an overridden symbol binding. Use the appropriate SYM_FUNC_START_WEAK_PI instead. Fixes: 39d114ddc682 ("arm64: add KASAN support") Reported-by: Sami Tolvanen Signed-off-by: Fangrui Song Tested-by: Sami Tolvanen Tested-by: Nick Desaulniers Reviewed-by: Nick Desaulniers Cc: Link: https://lore.kernel.org/r/20201029181951.1866093-1-maskray@google.com Signed-off-by: Will Deacon --- arch/arm64/lib/memcpy.S | 3 +-- arch/arm64/lib/memmove.S | 3 +-- arch/arm64/lib/memset.S | 3 +-- 3 files changed, 3 insertions(+), 6 deletions(-) diff --git a/arch/arm64/lib/memcpy.S b/arch/arm64/lib/memcpy.S index e0bf83d556f2..dc8d2a216a6e 100644 --- a/arch/arm64/lib/memcpy.S +++ b/arch/arm64/lib/memcpy.S @@ -56,9 +56,8 @@ stp \reg1, \reg2, [\ptr], \val .endm - .weak memcpy SYM_FUNC_START_ALIAS(__memcpy) -SYM_FUNC_START_PI(memcpy) +SYM_FUNC_START_WEAK_PI(memcpy) #include "copy_template.S" ret SYM_FUNC_END_PI(memcpy) diff --git a/arch/arm64/lib/memmove.S b/arch/arm64/lib/memmove.S index 02cda2e33bde..1035dce4bdaf 100644 --- a/arch/arm64/lib/memmove.S +++ b/arch/arm64/lib/memmove.S @@ -45,9 +45,8 @@ C_h .req x12 D_l .req x13 D_h .req x14 - .weak memmove SYM_FUNC_START_ALIAS(__memmove) -SYM_FUNC_START_PI(memmove) +SYM_FUNC_START_WEAK_PI(memmove) cmp dstin, src b.lo __memcpy add tmp1, src, count diff --git a/arch/arm64/lib/memset.S b/arch/arm64/lib/memset.S index 77c3c7ba0084..a9c1c9a01ea9 100644 --- a/arch/arm64/lib/memset.S +++ b/arch/arm64/lib/memset.S @@ -42,9 +42,8 @@ dst .req x8 tmp3w .req w9 tmp3 .req x9 - .weak memset SYM_FUNC_START_ALIAS(__memset) -SYM_FUNC_START_PI(memset) +SYM_FUNC_START_WEAK_PI(memset) mov dst, dstin /* Preserve return value. */ and A_lw, val, #255 orr A_lw, A_lw, A_lw, lsl #8 -- cgit v1.2.3 From e9a33caec90e05673e2f7fb7c80f172031964d25 Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Mon, 26 Oct 2020 13:49:29 +0000 Subject: KVM: arm64: Factor out is_{vhe,nvhe}_hyp_code() Currently has_vhe() detects whether it is being compiled for VHE/NVHE hyp code based on preprocessor definitions, and uses this knowledge to avoid redundant runtime checks. There are other cases where we'd like to use this knowledge, so let's factor the preprocessor checks out into separate helpers. There should be no functional change as a result of this patch. Signed-off-by: Mark Rutland Signed-off-by: Marc Zyngier Acked-by: Will Deacon Cc: David Brazdil Cc: Marc Zyngier Cc: Will Deacon Link: https://lore.kernel.org/r/20201026134931.28246-2-mark.rutland@arm.com --- arch/arm64/include/asm/virt.h | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-) diff --git a/arch/arm64/include/asm/virt.h b/arch/arm64/include/asm/virt.h index 09977acc007d..300be14ba77b 100644 --- a/arch/arm64/include/asm/virt.h +++ b/arch/arm64/include/asm/virt.h @@ -83,16 +83,27 @@ static inline bool is_kernel_in_hyp_mode(void) return read_sysreg(CurrentEL) == CurrentEL_EL2; } +static __always_inline bool is_vhe_hyp_code(void) +{ + /* Only defined for code run in VHE hyp context */ + return __is_defined(__KVM_VHE_HYPERVISOR__); +} + +static __always_inline bool is_nvhe_hyp_code(void) +{ + /* Only defined for code run in NVHE hyp context */ + return __is_defined(__KVM_NVHE_HYPERVISOR__); +} + static __always_inline bool has_vhe(void) { /* - * The following macros are defined for code specic to VHE/nVHE. - * If has_vhe() is inlined into those compilation units, it can - * be determined statically. Otherwise fall back to caps. + * Code only run in VHE/NVHE hyp context can assume VHE is present or + * absent. Otherwise fall back to caps. */ - if (__is_defined(__KVM_VHE_HYPERVISOR__)) + if (is_vhe_hyp_code()) return true; - else if (__is_defined(__KVM_NVHE_HYPERVISOR__)) + else if (is_nvhe_hyp_code()) return false; else return cpus_have_final_cap(ARM64_HAS_VIRT_HOST_EXTN); -- cgit v1.2.3 From dfc4e3f08903ed8fe0b66cc25b64524a82654166 Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Mon, 26 Oct 2020 13:49:30 +0000 Subject: arm64: cpufeature: reorder cpus_have_{const, final}_cap() In a subsequent patch we'll modify cpus_have_const_cap() to call cpus_have_final_cap(), and hence we need to define cpus_have_final_cap() first. To make subsequent changes easier to follow, this patch reorders the two without making any other changes. There should be no functional change as a result of this patch. Signed-off-by: Mark Rutland Signed-off-by: Marc Zyngier Acked-by: Will Deacon Cc: David Brazdil Cc: Marc Zyngier Cc: Will Deacon Link: https://lore.kernel.org/r/20201026134931.28246-3-mark.rutland@arm.com --- arch/arm64/include/asm/cpufeature.h | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/arch/arm64/include/asm/cpufeature.h b/arch/arm64/include/asm/cpufeature.h index fba6700b457b..9f671aa0419b 100644 --- a/arch/arm64/include/asm/cpufeature.h +++ b/arch/arm64/include/asm/cpufeature.h @@ -428,35 +428,35 @@ static __always_inline bool __cpus_have_const_cap(int num) } /* - * Test for a capability, possibly with a runtime check. + * Test for a capability without a runtime check. * - * Before capabilities are finalized, this behaves as cpus_have_cap(). + * Before capabilities are finalized, this will BUG(). * After capabilities are finalized, this is patched to avoid a runtime check. * * @num must be a compile-time constant. */ -static __always_inline bool cpus_have_const_cap(int num) +static __always_inline bool cpus_have_final_cap(int num) { if (system_capabilities_finalized()) return __cpus_have_const_cap(num); else - return cpus_have_cap(num); + BUG(); } /* - * Test for a capability without a runtime check. + * Test for a capability, possibly with a runtime check. * - * Before capabilities are finalized, this will BUG(). + * Before capabilities are finalized, this behaves as cpus_have_cap(). * After capabilities are finalized, this is patched to avoid a runtime check. * * @num must be a compile-time constant. */ -static __always_inline bool cpus_have_final_cap(int num) +static __always_inline bool cpus_have_const_cap(int num) { if (system_capabilities_finalized()) return __cpus_have_const_cap(num); else - BUG(); + return cpus_have_cap(num); } static inline void cpus_set_cap(unsigned int num) -- cgit v1.2.3 From d86de40decaa14e6613af1b2783bf4d589d0f38b Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Mon, 26 Oct 2020 13:49:31 +0000 Subject: arm64: cpufeature: upgrade hyp caps to final We finalize caps before initializing kvm hyp code, and any use of cpus_have_const_cap() in kvm hyp code generates redundant and potentially unsound code to read the cpu_hwcaps array. A number of helper functions used in both hyp context and regular kernel context use cpus_have_const_cap(), as some regular kernel code runs before the capabilities are finalized. It's tedious and error-prone to write separate copies of these for hyp and non-hyp code. So that we can avoid the redundant code, let's automatically upgrade cpus_have_const_cap() to cpus_have_final_cap() when used in hyp context. With this change, there's never a reason to access to cpu_hwcaps array from hyp code, and we don't need to create an NVHE alias for this. This should have no effect on non-hyp code. Signed-off-by: Mark Rutland Signed-off-by: Marc Zyngier Acked-by: Will Deacon Cc: David Brazdil Cc: Marc Zyngier Cc: Will Deacon Link: https://lore.kernel.org/r/20201026134931.28246-4-mark.rutland@arm.com --- arch/arm64/include/asm/cpufeature.h | 26 ++++++++++++++++++++++++-- arch/arm64/include/asm/virt.h | 12 ------------ arch/arm64/kernel/image-vars.h | 1 - 3 files changed, 24 insertions(+), 15 deletions(-) diff --git a/arch/arm64/include/asm/cpufeature.h b/arch/arm64/include/asm/cpufeature.h index 9f671aa0419b..79d6a0371c78 100644 --- a/arch/arm64/include/asm/cpufeature.h +++ b/arch/arm64/include/asm/cpufeature.h @@ -375,6 +375,23 @@ cpucap_multi_entry_cap_matches(const struct arm64_cpu_capabilities *entry, return false; } +static __always_inline bool is_vhe_hyp_code(void) +{ + /* Only defined for code run in VHE hyp context */ + return __is_defined(__KVM_VHE_HYPERVISOR__); +} + +static __always_inline bool is_nvhe_hyp_code(void) +{ + /* Only defined for code run in NVHE hyp context */ + return __is_defined(__KVM_NVHE_HYPERVISOR__); +} + +static __always_inline bool is_hyp_code(void) +{ + return is_vhe_hyp_code() || is_nvhe_hyp_code(); +} + extern DECLARE_BITMAP(cpu_hwcaps, ARM64_NCAPS); extern struct static_key_false cpu_hwcap_keys[ARM64_NCAPS]; extern struct static_key_false arm64_const_caps_ready; @@ -444,8 +461,11 @@ static __always_inline bool cpus_have_final_cap(int num) } /* - * Test for a capability, possibly with a runtime check. + * Test for a capability, possibly with a runtime check for non-hyp code. * + * For hyp code, this behaves the same as cpus_have_final_cap(). + * + * For non-hyp code: * Before capabilities are finalized, this behaves as cpus_have_cap(). * After capabilities are finalized, this is patched to avoid a runtime check. * @@ -453,7 +473,9 @@ static __always_inline bool cpus_have_final_cap(int num) */ static __always_inline bool cpus_have_const_cap(int num) { - if (system_capabilities_finalized()) + if (is_hyp_code()) + return cpus_have_final_cap(num); + else if (system_capabilities_finalized()) return __cpus_have_const_cap(num); else return cpus_have_cap(num); diff --git a/arch/arm64/include/asm/virt.h b/arch/arm64/include/asm/virt.h index 300be14ba77b..6069be50baf9 100644 --- a/arch/arm64/include/asm/virt.h +++ b/arch/arm64/include/asm/virt.h @@ -83,18 +83,6 @@ static inline bool is_kernel_in_hyp_mode(void) return read_sysreg(CurrentEL) == CurrentEL_EL2; } -static __always_inline bool is_vhe_hyp_code(void) -{ - /* Only defined for code run in VHE hyp context */ - return __is_defined(__KVM_VHE_HYPERVISOR__); -} - -static __always_inline bool is_nvhe_hyp_code(void) -{ - /* Only defined for code run in NVHE hyp context */ - return __is_defined(__KVM_NVHE_HYPERVISOR__); -} - static __always_inline bool has_vhe(void) { /* diff --git a/arch/arm64/kernel/image-vars.h b/arch/arm64/kernel/image-vars.h index fbd4b6b1fde5..ad8432251733 100644 --- a/arch/arm64/kernel/image-vars.h +++ b/arch/arm64/kernel/image-vars.h @@ -87,7 +87,6 @@ KVM_NVHE_ALIAS(__icache_flags); /* Kernel symbols needed for cpus_have_final/const_caps checks. */ KVM_NVHE_ALIAS(arm64_const_caps_ready); KVM_NVHE_ALIAS(cpu_hwcap_keys); -KVM_NVHE_ALIAS(cpu_hwcaps); /* Static keys which are set if a vGIC trap should be handled in hyp. */ KVM_NVHE_ALIAS(vgic_v2_cpuif_trap); -- cgit v1.2.3 From 11ce90a45d380f09213f128b76d68bb525186ad8 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:20 +0100 Subject: scripts: get_abi.pl: change script to allow parsing in ReST mode Right now, several ABI files won't parse as ReST, as they contain severe violations to the spec, with makes the script to crash. So, the code has a sanity logic with escapes bad code and cleans tags that can cause Sphinx to crash. Add support for disabling this mode. Right now, as enabling rst-mode causes crash, it is disabled by default. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/34b691e3002e8987c24d851fe37640f95e506a92.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- scripts/get_abi.pl | 72 +++++++++++++++++++++++++++++++++++------------------- 1 file changed, 47 insertions(+), 25 deletions(-) diff --git a/scripts/get_abi.pl b/scripts/get_abi.pl index c738cb795514..107672cdacb3 100755 --- a/scripts/get_abi.pl +++ b/scripts/get_abi.pl @@ -12,8 +12,14 @@ my $man; my $debug; my $prefix="Documentation/ABI"; +# +# If true, assumes that the description is formatted with ReST +# +my $description_is_rst = 0; + GetOptions( "debug|d+" => \$debug, + "rst-source!" => \$description_is_rst, "dir=s" => \$prefix, 'help|?' => \$help, man => \$man @@ -137,14 +143,15 @@ sub parse_abi { next; } if ($tag eq "description") { - next if ($content =~ m/^\s*$/); - if ($content =~ m/^(\s*)(.*)/) { - my $new_content = $2; - $space = $new_tag . $sep . $1; - while ($space =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {} - $space =~ s/./ /g; - $data{$what}->{$tag} .= "$new_content\n"; + # Preserve initial spaces for the first line + $content = ' ' x length($new_tag) . $sep . $content; + $content =~ s,^(\s*):,$1 ,; + if ($content =~ m/^(\s*)(.*)$/) { + $space = $1; + $content = $2; } + while ($space =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {} + $data{$what}->{$tag} .= $content; } else { $data{$what}->{$tag} = $content; } @@ -160,11 +167,15 @@ sub parse_abi { if ($tag eq "description") { if (!$data{$what}->{description}) { - next if (m/^\s*\n/); + s/^($space)//; if (m/^(\s*)(.*)/) { - $space = $1; - while ($space =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {} - $data{$what}->{$tag} .= "$2\n"; + my $sp = $1; + while ($sp =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {} + my $content = "$sp$2"; + + $content =~ s/^($space)//; + + $data{$what}->{$tag} .= "$content"; } } else { my $content = $_; @@ -274,23 +285,27 @@ sub output_rest { print "Defined on file :ref:`$file <$fileref>`\n\n" if ($type ne "File"); my $desc = $data{$what}->{description}; - $desc =~ s/^\s+//; - - # Remove title markups from the description, as they won't work - $desc =~ s/\n[\-\*\=\^\~]+\n/\n/g; if (!($desc =~ /^\s*$/)) { - if ($desc =~ m/\:\n/ || $desc =~ m/\n[\t ]+/ || $desc =~ m/[\x00-\x08\x0b-\x1f\x7b-\xff]/) { - # put everything inside a code block - $desc =~ s/\n/\n /g; - - print "::\n\n"; - print " $desc\n\n"; + if ($description_is_rst) { + print "$desc\n\n"; } else { - # Escape any special chars from description - $desc =~s/([\x00-\x08\x0b-\x1f\x21-\x2a\x2d\x2f\x3c-\x40\x5c\x5e-\x60\x7b-\xff])/\\$1/g; + $desc =~ s/^\s+//; - print "$desc\n\n"; + # Remove title markups from the description, as they won't work + $desc =~ s/\n[\-\*\=\^\~]+\n/\n\n/g; + + if ($desc =~ m/\:\n/ || $desc =~ m/\n[\t ]+/ || $desc =~ m/[\x00-\x08\x0b-\x1f\x7b-\xff]/) { + # put everything inside a code block + $desc =~ s/\n/\n /g; + + print "::\n\n"; + print " $desc\n\n"; + } else { + # Escape any special chars from description + $desc =~s/([\x00-\x08\x0b-\x1f\x21-\x2a\x2d\x2f\x3c-\x40\x5c\x5e-\x60\x7b-\xff])/\\$1/g; + print "$desc\n\n"; + } } } else { print "DESCRIPTION MISSING for $what\n\n" if (!$data{$what}->{is_file}); @@ -382,7 +397,7 @@ abi_book.pl - parse the Linux ABI files and produce a ReST book. =head1 SYNOPSIS -B [--debug] [--man] [--help] [--dir=] [] +B [--debug] [--man] [--help] --[(no-)rst-source] [--dir=] [] Where can be: @@ -405,6 +420,13 @@ B - validate the ABI contents Changes the location of the ABI search. By default, it uses the Documentation/ABI directory. +=item B<--rst-source> and B<--no-rst-source> + +The input file may be using ReST syntax or not. Those two options allow +selecting between a rst-compliant source ABI (--rst-source), or a +plain text that may be violating ReST spec, so it requres some escaping +logic (--no-rst-source). + =item B<--debug> Put the script in verbose mode, useful for debugging. Can be called multiple -- cgit v1.2.3 From e9bca8918e3d409f3b575d8db5591e422506c1ea Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:21 +0100 Subject: scripts: get_abi.pl: fix parsing on ReST mode When the source ABI file is using ReST notation, the script should handle whitespaces and lines with care, as otherwise the file won't be properly recognized. Address the bugs that are on such part of the script. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/5c22c54fbd0cda797b691d52c568be6d0d1079d8.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- scripts/get_abi.pl | 30 ++++++++++++++---------------- 1 file changed, 14 insertions(+), 16 deletions(-) diff --git a/scripts/get_abi.pl b/scripts/get_abi.pl index 107672cdacb3..0c403af86fd5 100755 --- a/scripts/get_abi.pl +++ b/scripts/get_abi.pl @@ -151,7 +151,8 @@ sub parse_abi { $content = $2; } while ($space =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {} - $data{$what}->{$tag} .= $content; + + $data{$what}->{$tag} .= "$content\n" if ($content); } else { $data{$what}->{$tag} = $content; } @@ -166,31 +167,28 @@ sub parse_abi { } if ($tag eq "description") { + my $content = $_; + while ($content =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {} if (!$data{$what}->{description}) { - s/^($space)//; - if (m/^(\s*)(.*)/) { - my $sp = $1; - while ($sp =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {} - my $content = "$sp$2"; - - $content =~ s/^($space)//; - - $data{$what}->{$tag} .= "$content"; + # Preserve initial spaces for the first line + if ($content =~ m/^(\s*)(.*)$/) { + $space = $1; + $content = $2; } + + $data{$what}->{$tag} .= "$content\n" if ($content); } else { - my $content = $_; if (m/^\s*\n/) { $data{$what}->{$tag} .= $content; next; } - while ($content =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {} $space = "" if (!($content =~ s/^($space)//)); - # Compress spaces with tabs - $content =~ s<^ {8}> <\t>; - $content =~ s<^ {1,7}\t> <\t>; - $content =~ s< {1,7}\t> <\t>; +# # Compress spaces with tabs +# $content =~ s<^ {8}> <\t>; +# $content =~ s<^ {1,7}\t> <\t>; +# $content =~ s< {1,7}\t> <\t>; $data{$what}->{$tag} .= $content; } next; -- cgit v1.2.3 From 61439c4ada9fd0e4ce73aca81a71344380a3053b Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:22 +0100 Subject: scripts: get_abi.pl: Allow optionally record from where a line came from The get_abi.pl reads a lot of files and can join them on a single output file. Store where each "What:" output came from, in order to be able to optionally display it. This is useful for the Sphinx extension, with can now be able to blame what ABI file has issues, and on what line the What: description with problems begin. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/befc387011c5e3c6febd285b7f27610e41c90260.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- scripts/get_abi.pl | 22 +++++++++++++++++++++- 1 file changed, 21 insertions(+), 1 deletion(-) diff --git a/scripts/get_abi.pl b/scripts/get_abi.pl index 0c403af86fd5..6a4d387ebf3b 100755 --- a/scripts/get_abi.pl +++ b/scripts/get_abi.pl @@ -10,6 +10,7 @@ use Fcntl ':mode'; my $help; my $man; my $debug; +my $enable_lineno; my $prefix="Documentation/ABI"; # @@ -19,6 +20,7 @@ my $description_is_rst = 0; GetOptions( "debug|d+" => \$debug, + "enable-lineno" => \$enable_lineno, "rst-source!" => \$description_is_rst, "dir=s" => \$prefix, 'help|?' => \$help, @@ -67,6 +69,7 @@ sub parse_abi { $data{$nametag}->{file} = $name; $data{$nametag}->{filepath} = $file; $data{$nametag}->{is_file} = 1; + $data{$nametag}->{line_no} = 1; my $type = $file; $type =~ s,.*/(.*)/.*,$1,; @@ -126,6 +129,8 @@ sub parse_abi { if ($tag ne "" && $new_tag) { $tag = $new_tag; + $data{$what}->{line_no} = $ln; + if ($new_what) { @{$data{$what}->{label}} = @labels if ($data{$nametag}->{what}); @labels = (); @@ -221,6 +226,12 @@ sub output_rest { my $file = $data{$what}->{file}; my $filepath = $data{$what}->{filepath}; + if ($enable_lineno) { + printf "#define LINENO %s%s#%s\n\n", + $prefix, $data{$what}->{file}, + $data{$what}->{line_no}; + } + my $w = $what; $w =~ s/([\(\)\_\-\*\=\^\~\\])/\\$1/g; @@ -369,6 +380,10 @@ sub search_symbols { } } +# Ensure that the prefix will always end with a slash +# While this is not needed for find, it makes the patch nicer +# with --enable-lineno +$prefix =~ s,/?$,/,; # # Parses all ABI files located at $prefix dir @@ -395,7 +410,8 @@ abi_book.pl - parse the Linux ABI files and produce a ReST book. =head1 SYNOPSIS -B [--debug] [--man] [--help] --[(no-)rst-source] [--dir=] [] +B [--debug] [--enable-lineno] [--man] [--help] + [--(no-)rst-source] [--dir=] [] Where can be: @@ -425,6 +441,10 @@ selecting between a rst-compliant source ABI (--rst-source), or a plain text that may be violating ReST spec, so it requres some escaping logic (--no-rst-source). +=item B<--enable-lineno> + +Enable output of #define LINENO lines. + =item B<--debug> Put the script in verbose mode, useful for debugging. Can be called multiple -- cgit v1.2.3 From f82a8a74ea81df22b1641ee56cf7195cc2a1bd0b Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:23 +0100 Subject: scripts: get_abi.pl: improve its parser to better catch up indentation The original parser for indentation were relying on having just one description for each "what". However, that's not the case: there are a number of ABI symbols that got defined multiple times. Improve the parser for it to better handle descriptions if entries are duplicated. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/eb458bb30be0e5a89192d6057b2e8a7e910dbcb8.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- scripts/get_abi.pl | 42 +++++++++++++++++++----------------------- 1 file changed, 19 insertions(+), 23 deletions(-) diff --git a/scripts/get_abi.pl b/scripts/get_abi.pl index 6a4d387ebf3b..bd018eb3815b 100755 --- a/scripts/get_abi.pl +++ b/scripts/get_abi.pl @@ -147,17 +147,19 @@ sub parse_abi { parse_error($file, $ln, "'What:' should come first:", $_); next; } - if ($tag eq "description") { - # Preserve initial spaces for the first line + if ($new_tag eq "description") { + $sep =~ s,:, ,; $content = ' ' x length($new_tag) . $sep . $content; - $content =~ s,^(\s*):,$1 ,; - if ($content =~ m/^(\s*)(.*)$/) { + while ($content =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {} + if ($content =~ m/^(\s*)(\S.*)$/) { + # Preserve initial spaces for the first line $space = $1; - $content = $2; + $content = "$2\n"; + $data{$what}->{$tag} .= $content; + } else { + undef($space); } - while ($space =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {} - $data{$what}->{$tag} .= "$content\n" if ($content); } else { $data{$what}->{$tag} = $content; } @@ -174,28 +176,22 @@ sub parse_abi { if ($tag eq "description") { my $content = $_; while ($content =~ s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e) {} - if (!$data{$what}->{description}) { + if (m/^\s*\n/) { + $data{$what}->{$tag} .= "\n"; + next; + } + + if (!defined($space)) { # Preserve initial spaces for the first line - if ($content =~ m/^(\s*)(.*)$/) { + if ($content =~ m/^(\s*)(\S.*)$/) { $space = $1; - $content = $2; + $content = "$2\n"; } - - $data{$what}->{$tag} .= "$content\n" if ($content); } else { - if (m/^\s*\n/) { - $data{$what}->{$tag} .= $content; - next; - } - $space = "" if (!($content =~ s/^($space)//)); - -# # Compress spaces with tabs -# $content =~ s<^ {8}> <\t>; -# $content =~ s<^ {1,7}\t> <\t>; -# $content =~ s< {1,7}\t> <\t>; - $data{$what}->{$tag} .= $content; } + $data{$what}->{$tag} .= $content; + next; } if (m/^\s*(.*)/) { -- cgit v1.2.3 From 234948bf3d27e473ac8043d20dacda146beeb0cb Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:24 +0100 Subject: scripts: get_abi.pl: cleanup ABI cross-reference logic Right now, the cross-references are generated on a single step, when doing ReST output. While this is nice optimization, it prevents auto-creating cross-references for ABI symbols. So, split it into a separate logic. While here, turn on Perl warnings, as it helps to debug problems inside the script. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/dbc97c8c2dfd877921f058134c35b2a8b1f8414b.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- scripts/get_abi.pl | 147 ++++++++++++++++++++++++++++++----------------------- 1 file changed, 83 insertions(+), 64 deletions(-) diff --git a/scripts/get_abi.pl b/scripts/get_abi.pl index bd018eb3815b..d134cc1692ee 100755 --- a/scripts/get_abi.pl +++ b/scripts/get_abi.pl @@ -2,15 +2,16 @@ # SPDX-License-Identifier: GPL-2.0 use strict; +use warnings; use Pod::Usage; use Getopt::Long; use File::Find; use Fcntl ':mode'; -my $help; -my $man; -my $debug; -my $enable_lineno; +my $help = 0; +my $man = 0; +my $debug = 0; +my $enable_lineno = 0; my $prefix="Documentation/ABI"; # @@ -40,6 +41,7 @@ pod2usage(2) if ($cmd eq "search" && !$arg); require Data::Dumper if ($debug); my %data; +my %symbols; # # Displays an error message, printing file name and line @@ -76,12 +78,12 @@ sub parse_abi { my $what; my $new_what; - my $tag; + my $tag = ""; my $ln; my $xrefs; my $space; my @labels; - my $label; + my $label = ""; print STDERR "Opening $file\n" if ($debug > 1); open IN, $file; @@ -110,10 +112,18 @@ sub parse_abi { if ($new_tag =~ m/what/) { $space = ""; + $content =~ s/[,.;]$//; + if ($tag =~ m/what/) { $what .= ", " . $content; } else { - parse_error($file, $ln, "What '$what' doesn't have a description", "") if ($what && !$data{$what}->{description}); + if ($what) { + parse_error($file, $ln, "What '$what' doesn't have a description", "") if (!$data{$what}->{description}); + + foreach my $w(split /, /, $what) { + $symbols{$w} = $what; + }; + } $what = $content; $label = $content; @@ -122,7 +132,7 @@ sub parse_abi { push @labels, [($content, $label)]; $tag = $new_tag; - push @{$data{$nametag}->{xrefs}}, [($content, $label)] if ($data{$nametag}->{what}); + push @{$data{$nametag}->{symbols}}, $content if ($data{$nametag}->{what}); next; } @@ -132,7 +142,7 @@ sub parse_abi { $data{$what}->{line_no} = $ln; if ($new_what) { - @{$data{$what}->{label}} = @labels if ($data{$nametag}->{what}); + @{$data{$what}->{label_list}} = @labels if ($data{$nametag}->{what}); @labels = (); $label = ""; $new_what = 0; @@ -203,36 +213,24 @@ sub parse_abi { # Everything else is error parse_error($file, $ln, "Unexpected line:", $_); } - $data{$nametag}->{description} =~ s/^\n+//; + $data{$nametag}->{description} =~ s/^\n+// if ($data{$nametag}->{description}); + if ($what) { + parse_error($file, $ln, "What '$what' doesn't have a description", "") if (!$data{$what}->{description}); + + foreach my $w(split /, /,$what) { + $symbols{$w} = $what; + }; + } close IN; } -# -# Outputs the book on ReST format -# - -my %labels; - -sub output_rest { - foreach my $what (sort { - ($data{$a}->{type} eq "File") cmp ($data{$b}->{type} eq "File") || - $a cmp $b - } keys %data) { - my $type = $data{$what}->{type}; - my $file = $data{$what}->{file}; - my $filepath = $data{$what}->{filepath}; - - if ($enable_lineno) { - printf "#define LINENO %s%s#%s\n\n", - $prefix, $data{$what}->{file}, - $data{$what}->{line_no}; - } - - my $w = $what; - $w =~ s/([\(\)\_\-\*\=\^\~\\])/\\$1/g; +sub create_labels { + my %labels; + foreach my $what (keys %data) { + next if ($data{$what}->{file} eq "File"); - foreach my $p (@{$data{$what}->{label}}) { + foreach my $p (@{$data{$what}->{label_list}}) { my ($content, $label) = @{$p}; $label = "abi_" . $label . " "; $label =~ tr/A-Z/a-z/; @@ -249,16 +247,39 @@ sub output_rest { } $labels{$label} = 1; - $data{$what}->{label} .= $label; - - printf ".. _%s:\n\n", $label; + $data{$what}->{label} = $label; # only one label is enough last; } + } +} + +# +# Outputs the book on ReST format +# + +sub output_rest { + create_labels(); + + foreach my $what (sort { + ($data{$a}->{type} eq "File") cmp ($data{$b}->{type} eq "File") || + $a cmp $b + } keys %data) { + my $type = $data{$what}->{type}; + my $file = $data{$what}->{file}; + my $filepath = $data{$what}->{filepath}; + + if ($enable_lineno) { + printf "#define LINENO %s%s#%s\n\n", + $prefix, $data{$what}->{file}, + $data{$what}->{line_no}; + } + my $w = $what; + $w =~ s/([\(\)\_\-\*\=\^\~\\])/\\$1/g; - $filepath =~ s,.*/(.*/.*),\1,;; + $filepath =~ s,.*/(.*/.*),$1,;; $filepath =~ s,[/\-],_,g;; my $fileref = "abi_file_".$filepath; @@ -269,8 +290,9 @@ sub output_rest { print ".. _$fileref:\n\n"; print "$w\n$bar\n\n"; } else { - my @names = split /\s*,\s*/,$w; + printf ".. _%s:\n\n", $data{$what}->{label}; + my @names = split /, /,$w; my $len = 0; foreach my $name (@names) { @@ -284,12 +306,13 @@ sub output_rest { printf "| %s", $name . " " x ($len - length($name)) . " |\n"; print "+-" . "-" x $len . "-+\n"; } - print "\n"; - } - print "Defined on file :ref:`$file <$fileref>`\n\n" if ($type ne "File"); + print "\nDefined on file :ref:`$file <$fileref>`\n\n"; + } - my $desc = $data{$what}->{description}; + my $desc = ""; + $desc = $data{$what}->{description} if (defined($data{$what}->{description})); + $desc =~ s/\s+$/\n/; if (!($desc =~ /^\s*$/)) { if ($description_is_rst) { @@ -316,18 +339,11 @@ sub output_rest { print "DESCRIPTION MISSING for $what\n\n" if (!$data{$what}->{is_file}); } - if ($data{$what}->{xrefs}) { + if ($data{$what}->{symbols}) { printf "Has the following ABI:\n\n"; - foreach my $p(@{$data{$what}->{xrefs}}) { - my ($content, $label) = @{$p}; - $label = "abi_" . $label . " "; - $label =~ tr/A-Z/a-z/; - - # Convert special chars to "_" - $label =~s/([\x00-\x2f\x3a-\x40\x5b-\x60\x7b-\xff])/_/g; - $label =~ s,_+,_,g; - $label =~ s,_$,,; + foreach my $content(@{$data{$what}->{symbols}}) { + my $label = $data{$symbols{$content}}->{label}; # Escape special chars from content $content =~s/([\x00-\x1f\x21-\x2f\x3a-\x40\x7b-\xff])/\\$1/g; @@ -355,17 +371,20 @@ sub search_symbols { print "\n$what\n$bar\n\n"; - my $kernelversion = $data{$what}->{kernelversion}; - my $contact = $data{$what}->{contact}; - my $users = $data{$what}->{users}; - my $date = $data{$what}->{date}; - my $desc = $data{$what}->{description}; - $kernelversion =~ s/^\s+//; - $contact =~ s/^\s+//; - $users =~ s/^\s+//; - $users =~ s/\n//g; - $date =~ s/^\s+//; - $desc =~ s/^\s+//; + my $kernelversion = $data{$what}->{kernelversion} if (defined($data{$what}->{kernelversion})); + my $contact = $data{$what}->{contact} if (defined($data{$what}->{contact})); + my $users = $data{$what}->{users} if (defined($data{$what}->{users})); + my $date = $data{$what}->{date} if (defined($data{$what}->{date})); + my $desc = $data{$what}->{description} if (defined($data{$what}->{description})); + + $kernelversion =~ s/^\s+// if ($kernelversion); + $contact =~ s/^\s+// if ($contact); + if ($users) { + $users =~ s/^\s+//; + $users =~ s/\n//g; + } + $date =~ s/^\s+// if ($date); + $desc =~ s/^\s+// if ($desc); printf "Kernel version:\t\t%s\n", $kernelversion if ($kernelversion); printf "Date:\t\t\t%s\n", $date if ($date); -- cgit v1.2.3 From c7ba333426bcf4d7330d3a0a768e8048a82e4cc7 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:25 +0100 Subject: scripts: get_abi.pl: detect duplicated ABI definitions The ABI should define only once each What. The current script logic assumes that. However, that's not the case, currently: there are several symbols with a generic definition, and per-driver ones. Better handle such cases, by preserving the cross-references with the files that define them, but also track such cases, producing warnings, as they should be fixed. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/d7a73b8b3aae5b2bff9279996ff9ca4cdfc89196.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- scripts/get_abi.pl | 85 +++++++++++++++++++++++++++++++++++++----------------- 1 file changed, 58 insertions(+), 27 deletions(-) diff --git a/scripts/get_abi.pl b/scripts/get_abi.pl index d134cc1692ee..1d1408fad130 100755 --- a/scripts/get_abi.pl +++ b/scripts/get_abi.pl @@ -114,6 +114,8 @@ sub parse_abi { $space = ""; $content =~ s/[,.;]$//; + push @{$symbols{$content}->{file}}, " $file:" . ($ln - 1); + if ($tag =~ m/what/) { $what .= ", " . $content; } else { @@ -121,7 +123,7 @@ sub parse_abi { parse_error($file, $ln, "What '$what' doesn't have a description", "") if (!$data{$what}->{description}); foreach my $w(split /, /, $what) { - $symbols{$w} = $what; + $symbols{$w}->{xref} = $what; }; } @@ -139,8 +141,6 @@ sub parse_abi { if ($tag ne "" && $new_tag) { $tag = $new_tag; - $data{$what}->{line_no} = $ln; - if ($new_what) { @{$data{$what}->{label_list}} = @labels if ($data{$nametag}->{what}); @labels = (); @@ -148,9 +148,19 @@ sub parse_abi { $new_what = 0; $data{$what}->{type} = $type; - $data{$what}->{file} = $name; - $data{$what}->{filepath} = $file; + if (!defined($data{$what}->{file})) { + $data{$what}->{file} = $name; + $data{$what}->{filepath} = $file; + } else { + if ($name ne $data{$what}->{file}) { + $data{$what}->{file} .= " " . $name; + $data{$what}->{filepath} .= " " . $file; + } + } print STDERR "\twhat: $what\n" if ($debug > 1); + $data{$what}->{line_no} = $ln; + } else { + $data{$what}->{line_no} = $ln if (!defined($data{$what}->{line_no})); } if (!$what) { @@ -218,7 +228,7 @@ sub parse_abi { parse_error($file, $ln, "What '$what' doesn't have a description", "") if (!$data{$what}->{description}); foreach my $w(split /, /,$what) { - $symbols{$w} = $what; + $symbols{$w}->{xref} = $what; }; } close IN; @@ -267,29 +277,20 @@ sub output_rest { $a cmp $b } keys %data) { my $type = $data{$what}->{type}; - my $file = $data{$what}->{file}; - my $filepath = $data{$what}->{filepath}; + + my @file = split / /, $data{$what}->{file}; + my @filepath = split / /, $data{$what}->{filepath}; if ($enable_lineno) { printf "#define LINENO %s%s#%s\n\n", - $prefix, $data{$what}->{file}, + $prefix, $file[0], $data{$what}->{line_no}; } my $w = $what; $w =~ s/([\(\)\_\-\*\=\^\~\\])/\\$1/g; - $filepath =~ s,.*/(.*/.*),$1,;; - $filepath =~ s,[/\-],_,g;; - my $fileref = "abi_file_".$filepath; - - if ($type eq "File") { - my $bar = $w; - $bar =~ s/./-/g; - - print ".. _$fileref:\n\n"; - print "$w\n$bar\n\n"; - } else { + if ($type ne "File") { printf ".. _%s:\n\n", $data{$what}->{label}; my @names = split /, /,$w; @@ -307,7 +308,26 @@ sub output_rest { print "+-" . "-" x $len . "-+\n"; } - print "\nDefined on file :ref:`$file <$fileref>`\n\n"; + print "\n"; + } + + for (my $i = 0; $i < scalar(@filepath); $i++) { + my $path = $filepath[$i]; + my $f = $file[$i]; + + $path =~ s,.*/(.*/.*),$1,;; + $path =~ s,[/\-],_,g;; + my $fileref = "abi_file_".$path; + + if ($type eq "File") { + my $bar = $w; + $bar =~ s/./-/g; + + print ".. _$fileref:\n\n"; + print "$w\n$bar\n\n"; + } else { + print "Defined on file :ref:`$f <$fileref>`\n\n"; + } } my $desc = ""; @@ -343,7 +363,7 @@ sub output_rest { printf "Has the following ABI:\n\n"; foreach my $content(@{$data{$what}->{symbols}}) { - my $label = $data{$symbols{$content}}->{label}; + my $label = $data{$symbols{$content}->{xref}}->{label}; # Escape special chars from content $content =~s/([\x00-\x1f\x21-\x2f\x3a-\x40\x7b-\xff])/\\$1/g; @@ -390,7 +410,7 @@ sub search_symbols { printf "Date:\t\t\t%s\n", $date if ($date); printf "Contact:\t\t%s\n", $contact if ($contact); printf "Users:\t\t\t%s\n", $users if ($users); - print "Defined on file:\t$file\n\n"; + print "Defined on file(s):\t$file\n\n"; print "Description:\n\n$desc"; } } @@ -410,12 +430,23 @@ print STDERR Data::Dumper->Dump([\%data], [qw(*data)]) if ($debug); # # Handles the command # -if ($cmd eq "rest") { - output_rest; -} elsif ($cmd eq "search") { +if ($cmd eq "search") { search_symbols; -} +} else { + if ($cmd eq "rest") { + output_rest; + } + + # Warn about duplicated ABI entries + foreach my $what(sort keys %symbols) { + my @files = @{$symbols{$what}->{file}}; + + next if (scalar(@files) == 1); + printf STDERR "Warning: $what is defined %d times: @files\n", + scalar(@files); + } +} __END__ -- cgit v1.2.3 From a16ab14eb7a92bfebf71a8a9185c749de86851f4 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:26 +0100 Subject: scripts: get_abi.pl: output users in ReST format Right now, the script only outputs Users on search. Print it also in ReST format. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/83def2a692c4c678f21ee6e76b66a54d2be1e796.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- scripts/get_abi.pl | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/scripts/get_abi.pl b/scripts/get_abi.pl index 1d1408fad130..e5a5588a8639 100755 --- a/scripts/get_abi.pl +++ b/scripts/get_abi.pl @@ -371,6 +371,14 @@ sub output_rest { print "- :ref:`$content <$label>`\n\n"; } } + + if (defined($data{$what}->{users})) { + my $users = $data{$what}->{users}; + + $users =~ s/\n/\n\t/g; + printf "Users:\n\t%s\n\n", $users if ($users ne ""); + } + } } -- cgit v1.2.3 From a4ea67bc65a15dadcda747c7f9d7b0e76e7c842a Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:27 +0100 Subject: scripts: get_abi.pl: prevent duplicated file names The same filename may exist on multiple directories within ABI. Create separate entries at the internal database for each of them. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/ef760f68f54e132c4be52f0027189b4ba31554ec.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- scripts/get_abi.pl | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/scripts/get_abi.pl b/scripts/get_abi.pl index e5a5588a8639..5612f019fae0 100755 --- a/scripts/get_abi.pl +++ b/scripts/get_abi.pl @@ -65,7 +65,10 @@ sub parse_abi { my $name = $file; $name =~ s,.*/,,; - my $nametag = "File $name"; + my $fn = $file; + $fn =~ s,Documentation/ABI/,,; + + my $nametag = "File $fn"; $data{$nametag}->{what} = "File $name"; $data{$nametag}->{type} = "File"; $data{$nametag}->{file} = $name; @@ -320,16 +323,18 @@ sub output_rest { my $fileref = "abi_file_".$path; if ($type eq "File") { - my $bar = $w; - $bar =~ s/./-/g; - print ".. _$fileref:\n\n"; - print "$w\n$bar\n\n"; } else { print "Defined on file :ref:`$f <$fileref>`\n\n"; } } + if ($type eq "File") { + my $bar = $w; + $bar =~ s/./-/g; + print "$w\n$bar\n\n"; + } + my $desc = ""; $desc = $data{$what}->{description} if (defined($data{$what}->{description})); $desc =~ s/\s+$/\n/; -- cgit v1.2.3 From c01d62d3fecb27f4c76a0e14b7e5afc15497a74b Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:28 +0100 Subject: scripts: get_abi.pl: use bold font for ABI definitions As we're using tables, let's distinguish the ones used by ABI by using a bold font. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/9ed645c2519169817379c1df5b8cf03828e92520.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- scripts/get_abi.pl | 1 + 1 file changed, 1 insertion(+) diff --git a/scripts/get_abi.pl b/scripts/get_abi.pl index 5612f019fae0..12a23f9a2f86 100755 --- a/scripts/get_abi.pl +++ b/scripts/get_abi.pl @@ -300,6 +300,7 @@ sub output_rest { my $len = 0; foreach my $name (@names) { + $name = "**$name**"; $len = length($name) if (length($name) > $len); } -- cgit v1.2.3 From 55e5414f2f8e31a8abbd95c43c24b69e83560a20 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:29 +0100 Subject: scripts: get_abi.pl: auto-generate cross references There are several cross-references that can be automatically generated: - References to .rst files inside Documentation/ - References to other ABI files; - References to ABI symbols at /sys/*. Add a logic to automatically parse them and convert into cross references. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/abe756d4f94fb6ffcc3dd3902a766c7c3990ea89.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- scripts/get_abi.pl | 31 +++++++++++++++++++++++++++++++ 1 file changed, 31 insertions(+) diff --git a/scripts/get_abi.pl b/scripts/get_abi.pl index 12a23f9a2f86..ff4f9f82ecad 100755 --- a/scripts/get_abi.pl +++ b/scripts/get_abi.pl @@ -3,6 +3,7 @@ use strict; use warnings; +use utf8; use Pod::Usage; use Getopt::Long; use File::Find; @@ -272,6 +273,9 @@ sub create_labels { # Outputs the book on ReST format # +# \b doesn't work well with paths. So, we need to define something else +my $bondary = qr { (?"; + $desc =~ s,\bDocumentation/ABI/$f\b,:ref:`$xref`,g; + } + + @matches = $desc =~ m,$bondary(/sys/[^\s\.\,\;\:\*\s\`\'\(\)]+)$bondary,; + + foreach my $s (@matches) { + if (defined($data{$s}) && defined($data{$s}->{label})) { + my $xref = $s; + + $xref =~ s/([\x00-\x1f\x21-\x2f\x3a-\x40\x7b-\xff])/\\$1/g; + $xref = ":ref:`$xref <" . $data{$s}->{label} . ">`"; + + $desc =~ s,$bondary$s$bondary,$xref,g; + } + } + print "$desc\n\n"; } else { $desc =~ s/^\s+//; -- cgit v1.2.3 From 678a63f9888c28150e08cf6cb5fec9fbaf2bf81c Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:30 +0100 Subject: docs: kernellog.py: add support for info() An extension may want to just inform about something. So, add support for it. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/0ddebd8677605d789d53433c8a5344c68da82a73.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/sphinx/kernellog.py | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/Documentation/sphinx/kernellog.py b/Documentation/sphinx/kernellog.py index af924f51a7dc..8ac7d274f542 100644 --- a/Documentation/sphinx/kernellog.py +++ b/Documentation/sphinx/kernellog.py @@ -25,4 +25,8 @@ def verbose(app, message): else: app.verbose(message) - +def info(app, message): + if UseLogging: + logger.info(message) + else: + app.info(message) -- cgit v1.2.3 From 9ca876f96323934b8ace04b123531f05e3bed029 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:31 +0100 Subject: docs: kernel_abi.py: add a script to parse ABI documentation The ABI documentation is special: it is not plain text files, but, instead, files with an strict format, as specified by Documentation/ABI/README. Add a parser for it. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/48abf1a410237e63f85354a8cd7027fdf25657bf.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/sphinx/kernel_abi.py | 157 +++++++++++++++++++++++++++++++++++++ 1 file changed, 157 insertions(+) create mode 100644 Documentation/sphinx/kernel_abi.py diff --git a/Documentation/sphinx/kernel_abi.py b/Documentation/sphinx/kernel_abi.py new file mode 100644 index 000000000000..f8e7a02066c2 --- /dev/null +++ b/Documentation/sphinx/kernel_abi.py @@ -0,0 +1,157 @@ +# -*- coding: utf-8; mode: python -*- +# SPDX-License-Identifier: GPL-2.0 +u""" + kernel-abi + ~~~~~~~~~~ + + Implementation of the ``kernel-abi`` reST-directive. + + :copyright: Copyright (C) 2016 Markus Heiser + :copyright: Copyright (C) 2016-2020 Mauro Carvalho Chehab + :maintained-by: Mauro Carvalho Chehab + :license: GPL Version 2, June 1991 see Linux/COPYING for details. + + The ``kernel-abi`` (:py:class:`KernelCmd`) directive calls the + scripts/get_abi.pl script to parse the Kernel ABI files. + + Overview of directive's argument and options. + + .. code-block:: rst + + .. kernel-abi:: + :debug: + + The argument ```` is required. It contains the + location of the ABI files to be parsed. + + ``debug`` + Inserts a code-block with the *raw* reST. Sometimes it is helpful to see + what reST is generated. + +""" + +import sys +import os +from os import path +import subprocess + +from sphinx.ext.autodoc import AutodocReporter + +from docutils import nodes +from docutils.parsers.rst import Directive, directives +from docutils.statemachine import ViewList +from docutils.utils.error_reporting import ErrorString + + +__version__ = '1.0' + +# We can't assume that six is installed +PY3 = sys.version_info[0] == 3 +PY2 = sys.version_info[0] == 2 +if PY3: + # pylint: disable=C0103, W0622 + unicode = str + basestring = str + +def setup(app): + + app.add_directive("kernel-abi", KernelCmd) + return dict( + version = __version__ + , parallel_read_safe = True + , parallel_write_safe = True + ) + +class KernelCmd(Directive): + + u"""KernelABI (``kernel-abi``) directive""" + + required_arguments = 1 + optional_arguments = 0 + has_content = False + final_argument_whitespace = True + + option_spec = { + "debug" : directives.flag + } + + def warn(self, message, **replace): + replace["fname"] = self.state.document.current_source + replace["line_no"] = replace.get("line_no", self.lineno) + message = ("%(fname)s:%(line_no)s: [kernel-abi WARN] : " + message) % replace + self.state.document.settings.env.app.warn(message, prefix="") + + def run(self): + + doc = self.state.document + if not doc.settings.file_insertion_enabled: + raise self.warning("docutils: file insertion disabled") + + env = doc.settings.env + cwd = path.dirname(doc.current_source) + cmd = "get_abi.pl rest --dir " + cmd += self.arguments[0] + + srctree = path.abspath(os.environ["srctree"]) + + fname = cmd + + # extend PATH with $(srctree)/scripts + path_env = os.pathsep.join([ + srctree + os.sep + "scripts", + os.environ["PATH"] + ]) + shell_env = os.environ.copy() + shell_env["PATH"] = path_env + shell_env["srctree"] = srctree + + lines = self.runCmd(cmd, shell=True, cwd=cwd, env=shell_env) + nodeList = self.nestedParse(lines, fname) + return nodeList + + def runCmd(self, cmd, **kwargs): + u"""Run command ``cmd`` and return it's stdout as unicode.""" + + try: + proc = subprocess.Popen( + cmd + , stdout = subprocess.PIPE + , stderr = subprocess.PIPE + , universal_newlines = True + , **kwargs + ) + out, err = proc.communicate() + if err: + self.warn(err) + if proc.returncode != 0: + raise self.severe( + u"command '%s' failed with return code %d" + % (cmd, proc.returncode) + ) + except OSError as exc: + raise self.severe(u"problems with '%s' directive: %s." + % (self.name, ErrorString(exc))) + return unicode(out) + + def nestedParse(self, lines, fname): + content = ViewList() + node = nodes.section() + + if "debug" in self.options: + code_block = "\n\n.. code-block:: rst\n :linenos:\n" + for l in lines.split("\n"): + code_block += "\n " + l + lines = code_block + "\n\n" + + for c, l in enumerate(lines.split("\n")): + content.append(l, fname, c) + + buf = self.state.memo.title_styles, self.state.memo.section_level, self.state.memo.reporter + self.state.memo.title_styles = [] + self.state.memo.section_level = 0 + self.state.memo.reporter = AutodocReporter(content, self.state.memo.reporter) + try: + self.state.nested_parse(content, 0, node, match_titles=1) + finally: + self.state.memo.title_styles, self.state.memo.section_level, self.state.memo.reporter = buf + return node.children -- cgit v1.2.3 From 823830d4065bfb115b948c9f21e713a15d8d95ef Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:32 +0100 Subject: docs: kernel_abi.py: fix UTF-8 support The parser breaks with UTF-8 characters with Sphinx 1.4. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/9e7c8e3b0efaa1ae0536da6493ab438bd3f9fe58.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/sphinx/kernel_abi.py | 19 +++++++------------ 1 file changed, 7 insertions(+), 12 deletions(-) diff --git a/Documentation/sphinx/kernel_abi.py b/Documentation/sphinx/kernel_abi.py index f8e7a02066c2..fe69c213716d 100644 --- a/Documentation/sphinx/kernel_abi.py +++ b/Documentation/sphinx/kernel_abi.py @@ -1,5 +1,7 @@ # -*- coding: utf-8; mode: python -*- +# coding=utf-8 # SPDX-License-Identifier: GPL-2.0 +# u""" kernel-abi ~~~~~~~~~~ @@ -30,6 +32,7 @@ u""" """ +import codecs import sys import os from os import path @@ -45,14 +48,6 @@ from docutils.utils.error_reporting import ErrorString __version__ = '1.0' -# We can't assume that six is installed -PY3 = sys.version_info[0] == 3 -PY2 = sys.version_info[0] == 2 -if PY3: - # pylint: disable=C0103, W0622 - unicode = str - basestring = str - def setup(app): app.add_directive("kernel-abi", KernelCmd) @@ -117,12 +112,12 @@ class KernelCmd(Directive): cmd , stdout = subprocess.PIPE , stderr = subprocess.PIPE - , universal_newlines = True , **kwargs ) out, err = proc.communicate() - if err: - self.warn(err) + + out, err = codecs.decode(out, 'utf-8'), codecs.decode(err, 'utf-8') + if proc.returncode != 0: raise self.severe( u"command '%s' failed with return code %d" @@ -131,7 +126,7 @@ class KernelCmd(Directive): except OSError as exc: raise self.severe(u"problems with '%s' directive: %s." % (self.name, ErrorString(exc))) - return unicode(out) + return out def nestedParse(self, lines, fname): content = ViewList() -- cgit v1.2.3 From c830fa9a4d363caacc4ba3f63c06b9e967ad13a4 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:33 +0100 Subject: docs: kernel_abi.py: make it compatible with Sphinx 1.7+ The same way kerneldoc.py needed changes to work with newer Sphinx, this script needs the same changes. While here, reorganize the include order to match kerneldoc.py. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/f2b25caef5db7738629773a03463908d3b39b83a.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/sphinx/kernel_abi.py | 39 ++++++++++++++++++++++++++------------ 1 file changed, 27 insertions(+), 12 deletions(-) diff --git a/Documentation/sphinx/kernel_abi.py b/Documentation/sphinx/kernel_abi.py index fe69c213716d..8601a3b75a28 100644 --- a/Documentation/sphinx/kernel_abi.py +++ b/Documentation/sphinx/kernel_abi.py @@ -33,18 +33,27 @@ u""" """ import codecs -import sys import os -from os import path import subprocess +import sys -from sphinx.ext.autodoc import AutodocReporter +from os import path -from docutils import nodes -from docutils.parsers.rst import Directive, directives +from docutils import nodes, statemachine from docutils.statemachine import ViewList +from docutils.parsers.rst import directives, Directive from docutils.utils.error_reporting import ErrorString +# +# AutodocReporter is only good up to Sphinx 1.7 +# +import sphinx + +Use_SSI = sphinx.__version__[:3] >= '1.7' +if Use_SSI: + from sphinx.util.docutils import switch_source_input +else: + from sphinx.ext.autodoc import AutodocReporter __version__ = '1.0' @@ -142,11 +151,17 @@ class KernelCmd(Directive): content.append(l, fname, c) buf = self.state.memo.title_styles, self.state.memo.section_level, self.state.memo.reporter - self.state.memo.title_styles = [] - self.state.memo.section_level = 0 - self.state.memo.reporter = AutodocReporter(content, self.state.memo.reporter) - try: - self.state.nested_parse(content, 0, node, match_titles=1) - finally: - self.state.memo.title_styles, self.state.memo.section_level, self.state.memo.reporter = buf + + if Use_SSI: + with switch_source_input(self.state, content): + self.state.nested_parse(content, 0, node, match_titles=1) + else: + self.state.memo.title_styles = [] + self.state.memo.section_level = 0 + self.state.memo.reporter = AutodocReporter(content, self.state.memo.reporter) + try: + self.state.nested_parse(content, 0, node, match_titles=1) + finally: + self.state.memo.title_styles, self.state.memo.section_level, self.state.memo.reporter = buf + return node.children -- cgit v1.2.3 From 997b7c8b4a95681dc75e6b720ad6f0ea705368c3 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:34 +0100 Subject: docs: kernel_abi.py: use --enable-lineno for get_abi.pl Just like kernel-doc extension, we need to be able to identify what part of an imported document has issues, as reporting them as: get_abi.pl rest --dir $srctree/Documentation/ABI/obsolete --rst-source:1689: ERROR: Unexpected indentation. Makes a lot harder for someone to fix. It should be noticed that it the line which will be reported is the line where the "What:" definition is, and not the line with actually has an error. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/d6155ab16fb7631f2fa8e7a770eae72f24bf7cc5.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/sphinx/kernel_abi.py | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) diff --git a/Documentation/sphinx/kernel_abi.py b/Documentation/sphinx/kernel_abi.py index 8601a3b75a28..096dec482e96 100644 --- a/Documentation/sphinx/kernel_abi.py +++ b/Documentation/sphinx/kernel_abi.py @@ -36,6 +36,7 @@ import codecs import os import subprocess import sys +import re from os import path @@ -93,7 +94,7 @@ class KernelCmd(Directive): env = doc.settings.env cwd = path.dirname(doc.current_source) - cmd = "get_abi.pl rest --dir " + cmd = "get_abi.pl rest --enable-lineno --dir " cmd += self.arguments[0] srctree = path.abspath(os.environ["srctree"]) @@ -137,7 +138,7 @@ class KernelCmd(Directive): % (self.name, ErrorString(exc))) return out - def nestedParse(self, lines, fname): + def nestedParse(self, lines, f): content = ViewList() node = nodes.section() @@ -147,8 +148,17 @@ class KernelCmd(Directive): code_block += "\n " + l lines = code_block + "\n\n" - for c, l in enumerate(lines.split("\n")): - content.append(l, fname, c) + line_regex = re.compile("^#define LINENO (\S+)\#([0-9]+)$") + ln = 0 + + for line in lines.split("\n"): + match = line_regex.search(line) + if match: + f = match.group(1) + # sphinx counts lines from 0 + ln = int(match.group(2)) - 1 + else: + content.append(line, f, ln) buf = self.state.memo.title_styles, self.state.memo.section_level, self.state.memo.reporter -- cgit v1.2.3 From 3c543d29891a42b23c1f9a3a4788ae37fb94cce2 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:35 +0100 Subject: docs: kernel_abi.py: Handle with a lazy Sphinx parser The Sphinx docutils parser is lazy: if the content is bigger than a certain number of lines, it silenlty stops parsing it, producing an incomplete content. This seems to be worse on newer Sphinx versions, like 2.0. So, change the logic to parse the contents per input file. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/4659b60795739308e34d2d00c57ee0742a9cd2ab.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/sphinx/kernel_abi.py | 39 +++++++++++++++++++++++++------------- 1 file changed, 26 insertions(+), 13 deletions(-) diff --git a/Documentation/sphinx/kernel_abi.py b/Documentation/sphinx/kernel_abi.py index 096dec482e96..ce5f3b0ae811 100644 --- a/Documentation/sphinx/kernel_abi.py +++ b/Documentation/sphinx/kernel_abi.py @@ -37,6 +37,7 @@ import os import subprocess import sys import re +import kernellog from os import path @@ -80,12 +81,6 @@ class KernelCmd(Directive): "debug" : directives.flag } - def warn(self, message, **replace): - replace["fname"] = self.state.document.current_source - replace["line_no"] = replace.get("line_no", self.lineno) - message = ("%(fname)s:%(line_no)s: [kernel-abi WARN] : " + message) % replace - self.state.document.settings.env.app.warn(message, prefix="") - def run(self): doc = self.state.document @@ -111,7 +106,7 @@ class KernelCmd(Directive): shell_env["srctree"] = srctree lines = self.runCmd(cmd, shell=True, cwd=cwd, env=shell_env) - nodeList = self.nestedParse(lines, fname) + nodeList = self.nestedParse(lines, self.arguments[0]) return nodeList def runCmd(self, cmd, **kwargs): @@ -138,9 +133,9 @@ class KernelCmd(Directive): % (self.name, ErrorString(exc))) return out - def nestedParse(self, lines, f): + def nestedParse(self, lines, fname): content = ViewList() - node = nodes.section() + node = nodes.section() if "debug" in self.options: code_block = "\n\n.. code-block:: rst\n :linenos:\n" @@ -150,22 +145,42 @@ class KernelCmd(Directive): line_regex = re.compile("^#define LINENO (\S+)\#([0-9]+)$") ln = 0 + n = 0 + f = fname for line in lines.split("\n"): + n = n + 1 match = line_regex.search(line) if match: - f = match.group(1) + new_f = match.group(1) + + # Sphinx parser is lazy: it stops parsing contents in the + # middle, if it is too big. So, handle it per input file + if new_f != f and content: + self.do_parse(content, node) + content = ViewList() + + f = new_f + # sphinx counts lines from 0 ln = int(match.group(2)) - 1 else: content.append(line, f, ln) - buf = self.state.memo.title_styles, self.state.memo.section_level, self.state.memo.reporter + kernellog.info(self.state.document.settings.env.app, "%s: parsed %i lines" % (fname, n)) + + if content: + self.do_parse(content, node) + return node.children + + def do_parse(self, content, node): if Use_SSI: with switch_source_input(self.state, content): self.state.nested_parse(content, 0, node, match_titles=1) else: + buf = self.state.memo.title_styles, self.state.memo.section_level, self.state.memo.reporter + self.state.memo.title_styles = [] self.state.memo.section_level = 0 self.state.memo.reporter = AutodocReporter(content, self.state.memo.reporter) @@ -173,5 +188,3 @@ class KernelCmd(Directive): self.state.nested_parse(content, 0, node, match_titles=1) finally: self.state.memo.title_styles, self.state.memo.section_level, self.state.memo.reporter = buf - - return node.children -- cgit v1.2.3 From a7ee04b3ef0d2b2514730928a04a112f04ffe7df Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:36 +0100 Subject: docs: add ABI documentation to the admin-guide book As we don't want a generic Sphinx extension to execute commands, change the one proposed to Markus to call the abi_book.pl script. Use a script to parse the Documentation/ABI directory and output it at the admin-guide. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/5651482b06500e69a1acdf92152f90a203e6521d.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/admin-guide/abi-obsolete.rst | 10 ++++++++++ Documentation/admin-guide/abi-removed.rst | 4 ++++ Documentation/admin-guide/abi-stable.rst | 13 +++++++++++++ Documentation/admin-guide/abi-testing.rst | 19 +++++++++++++++++++ Documentation/admin-guide/abi.rst | 11 +++++++++++ Documentation/admin-guide/index.rst | 2 ++ Documentation/conf.py | 3 ++- 7 files changed, 61 insertions(+), 1 deletion(-) create mode 100644 Documentation/admin-guide/abi-obsolete.rst create mode 100644 Documentation/admin-guide/abi-removed.rst create mode 100644 Documentation/admin-guide/abi-stable.rst create mode 100644 Documentation/admin-guide/abi-testing.rst create mode 100644 Documentation/admin-guide/abi.rst diff --git a/Documentation/admin-guide/abi-obsolete.rst b/Documentation/admin-guide/abi-obsolete.rst new file mode 100644 index 000000000000..cda9168445a5 --- /dev/null +++ b/Documentation/admin-guide/abi-obsolete.rst @@ -0,0 +1,10 @@ +ABI obsolete symbols +==================== + +Documents interfaces that are still remaining in the kernel, but are +marked to be removed at some later point in time. + +The description of the interface will document the reason why it is +obsolete and when it can be expected to be removed. + +.. kernel-abi:: $srctree/Documentation/ABI/obsolete diff --git a/Documentation/admin-guide/abi-removed.rst b/Documentation/admin-guide/abi-removed.rst new file mode 100644 index 000000000000..497978fc9632 --- /dev/null +++ b/Documentation/admin-guide/abi-removed.rst @@ -0,0 +1,4 @@ +ABI removed symbols +=================== + +.. kernel-abi:: $srctree/Documentation/ABI/removed diff --git a/Documentation/admin-guide/abi-stable.rst b/Documentation/admin-guide/abi-stable.rst new file mode 100644 index 000000000000..7495d7a35048 --- /dev/null +++ b/Documentation/admin-guide/abi-stable.rst @@ -0,0 +1,13 @@ +ABI stable symbols +================== + +Documents the interfaces that the developer has defined to be stable. + +Userspace programs are free to use these interfaces with no +restrictions, and backward compatibility for them will be guaranteed +for at least 2 years. + +Most interfaces (like syscalls) are expected to never change and always +be available. + +.. kernel-abi:: $srctree/Documentation/ABI/stable diff --git a/Documentation/admin-guide/abi-testing.rst b/Documentation/admin-guide/abi-testing.rst new file mode 100644 index 000000000000..5c886fc50b9e --- /dev/null +++ b/Documentation/admin-guide/abi-testing.rst @@ -0,0 +1,19 @@ +ABI testing symbols +=================== + +Documents interfaces that are felt to be stable, +as the main development of this interface has been completed. + +The interface can be changed to add new features, but the +current interface will not break by doing this, unless grave +errors or security problems are found in them. + +Userspace programs can start to rely on these interfaces, but they must +be aware of changes that can occur before these interfaces move to +be marked stable. + +Programs that use these interfaces are strongly encouraged to add their +name to the description of these interfaces, so that the kernel +developers can easily notify them if any changes occur. + +.. kernel-abi:: $srctree/Documentation/ABI/testing diff --git a/Documentation/admin-guide/abi.rst b/Documentation/admin-guide/abi.rst new file mode 100644 index 000000000000..3b9645c77469 --- /dev/null +++ b/Documentation/admin-guide/abi.rst @@ -0,0 +1,11 @@ +===================== +Linux ABI description +===================== + +.. toctree:: + :maxdepth: 1 + + abi-stable + abi-testing + abi-obsolete + abi-removed diff --git a/Documentation/admin-guide/index.rst b/Documentation/admin-guide/index.rst index ed1cf94ea50c..4e0c4ae44acd 100644 --- a/Documentation/admin-guide/index.rst +++ b/Documentation/admin-guide/index.rst @@ -18,6 +18,8 @@ etc. devices sysctl/index + abi + This section describes CPU vulnerabilities and their mitigations. .. toctree:: diff --git a/Documentation/conf.py b/Documentation/conf.py index 376dd0ddf39c..1e44983853a2 100644 --- a/Documentation/conf.py +++ b/Documentation/conf.py @@ -38,7 +38,8 @@ needs_sphinx = '1.3' # ones. extensions = ['kerneldoc', 'rstFlatTable', 'kernel_include', 'kfigure', 'sphinx.ext.ifconfig', 'automarkup', - 'maintainers_include', 'sphinx.ext.autosectionlabel' ] + 'maintainers_include', 'sphinx.ext.autosectionlabel', + 'kernel_abi'] # # cdomain is badly broken in Sphinx 3+. Leaving it out generates *most* -- cgit v1.2.3 From c7e45ea429c93139e85ddf3e0a8500a997f0f745 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:37 +0100 Subject: docs: ABI: README: specify that files should be ReST compatible As we plan to remove the escaping code from the scripts/get_abi.pl, specify at the ABI README file that the content of the file should be ReST compatible. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/178a997070debd1953ba7d302c375948501d6193.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/README | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/Documentation/ABI/README b/Documentation/ABI/README index 3121029dce21..8bac9cb09a6d 100644 --- a/Documentation/ABI/README +++ b/Documentation/ABI/README @@ -32,7 +32,7 @@ The different levels of stability are: layout of the files below for details on how to do this.) obsolete/ - This directory documents interfaces that are still remaining in + This directory documents interfaces that are still remaining in the kernel, but are marked to be removed at some later point in time. The description of the interface will document the reason why it is obsolete and when it can be expected to be removed. @@ -58,6 +58,14 @@ Users: All users of this interface who wish to be notified when be changed further. +Note: + The fields should be use a simple notation, compatible with ReST markup. + Also, the file **should not** have a top-level index, like:: + + === + foo + === + How things move between levels: Interfaces in stable may move to obsolete, as long as the proper -- cgit v1.2.3 From 98913408c5465ac477f80da7affe347b413edaa4 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:38 +0100 Subject: docs: ABI: stable: make files ReST compatible Several entries at the stable ABI files won't parse if we pass them directly to the ReST output. Adjust them, in order to allow adding their contents as-is at the stable ABI book. Acked-by: Srinivas Kandagatla Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/467a0dfbcdf00db710a629d3fe4a2563750339d8.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/stable/firewire-cdev | 4 + Documentation/ABI/stable/sysfs-acpi-pmprofile | 22 +++--- Documentation/ABI/stable/sysfs-bus-firewire | 3 + Documentation/ABI/stable/sysfs-bus-nvmem | 19 ++--- Documentation/ABI/stable/sysfs-bus-usb | 6 +- Documentation/ABI/stable/sysfs-class-backlight | 1 + Documentation/ABI/stable/sysfs-class-infiniband | 93 +++++++++++++++------- Documentation/ABI/stable/sysfs-class-rfkill | 13 ++- Documentation/ABI/stable/sysfs-class-tpm | 96 +++++++++++------------ Documentation/ABI/stable/sysfs-devices | 5 +- Documentation/ABI/stable/sysfs-driver-ib_srp | 1 + Documentation/ABI/stable/sysfs-firmware-efi-vars | 4 + Documentation/ABI/stable/sysfs-firmware-opal-dump | 5 ++ Documentation/ABI/stable/sysfs-firmware-opal-elog | 2 + Documentation/ABI/stable/sysfs-hypervisor-xen | 3 + Documentation/ABI/stable/vdso | 5 +- 16 files changed, 179 insertions(+), 103 deletions(-) diff --git a/Documentation/ABI/stable/firewire-cdev b/Documentation/ABI/stable/firewire-cdev index f72ed653878a..c9e8ff026154 100644 --- a/Documentation/ABI/stable/firewire-cdev +++ b/Documentation/ABI/stable/firewire-cdev @@ -14,12 +14,14 @@ Description: Each /dev/fw* is associated with one IEEE 1394 node, which can be remote or local nodes. Operations on a /dev/fw* file have different scope: + - The 1394 node which is associated with the file: - Asynchronous request transmission - Get the Configuration ROM - Query node ID - Query maximum speed of the path between this node and local node + - The 1394 bus (i.e. "card") to which the node is attached to: - Isochronous stream transmission and reception - Asynchronous stream transmission and reception @@ -31,6 +33,7 @@ Description: manager - Query cycle time - Bus reset initiation, bus reset event reception + - All 1394 buses: - Allocation of IEEE 1212 address ranges on the local link layers, reception of inbound requests to such @@ -43,6 +46,7 @@ Description: userland implement different access permission models, some operations are restricted to /dev/fw* files that are associated with a local node: + - Addition of descriptors or directories to the local nodes' Configuration ROM - PHY packet transmission and reception diff --git a/Documentation/ABI/stable/sysfs-acpi-pmprofile b/Documentation/ABI/stable/sysfs-acpi-pmprofile index 964c7a8afb26..fd97d22b677f 100644 --- a/Documentation/ABI/stable/sysfs-acpi-pmprofile +++ b/Documentation/ABI/stable/sysfs-acpi-pmprofile @@ -6,17 +6,21 @@ Description: The ACPI pm_profile sysfs interface exports the platform power management (and performance) requirement expectations as provided by BIOS. The integer value is directly passed as retrieved from the FADT ACPI table. -Values: For possible values see ACPI specification: + +Values: For possible values see ACPI specification: 5.2.9 Fixed ACPI Description Table (FADT) Field: Preferred_PM_Profile Currently these values are defined by spec: - 0 Unspecified - 1 Desktop - 2 Mobile - 3 Workstation - 4 Enterprise Server - 5 SOHO Server - 6 Appliance PC - 7 Performance Server + + == ================= + 0 Unspecified + 1 Desktop + 2 Mobile + 3 Workstation + 4 Enterprise Server + 5 SOHO Server + 6 Appliance PC + 7 Performance Server >7 Reserved + == ================= diff --git a/Documentation/ABI/stable/sysfs-bus-firewire b/Documentation/ABI/stable/sysfs-bus-firewire index 41e5a0cd1e3e..9ac9eddb82ef 100644 --- a/Documentation/ABI/stable/sysfs-bus-firewire +++ b/Documentation/ABI/stable/sysfs-bus-firewire @@ -47,6 +47,7 @@ Description: IEEE 1394 node device attribute. Read-only and immutable. Values: 1: The sysfs entry represents a local node (a controller card). + 0: The sysfs entry represents a remote node. @@ -125,7 +126,9 @@ Description: Read-only attribute, immutable during the target's lifetime. Format, as exposed by firewire-sbp2 since 2.6.22, May 2007: Colon-separated hexadecimal string representations of + u64 EUI-64 : u24 directory_ID : u16 LUN + without 0x prefixes, without whitespace. The former sbp2 driver (removed in 2.6.37 after being superseded by firewire-sbp2) used a somewhat shorter format which was not as close to SAM. diff --git a/Documentation/ABI/stable/sysfs-bus-nvmem b/Documentation/ABI/stable/sysfs-bus-nvmem index 9ffba8576f7b..c399323f37de 100644 --- a/Documentation/ABI/stable/sysfs-bus-nvmem +++ b/Documentation/ABI/stable/sysfs-bus-nvmem @@ -9,13 +9,14 @@ Description: Note: This file is only present if CONFIG_NVMEM_SYSFS is enabled - ex: - hexdump /sys/bus/nvmem/devices/qfprom0/nvmem + ex:: - 0000000 0000 0000 0000 0000 0000 0000 0000 0000 - * - 00000a0 db10 2240 0000 e000 0c00 0c00 0000 0c00 - 0000000 0000 0000 0000 0000 0000 0000 0000 0000 - ... - * - 0001000 + hexdump /sys/bus/nvmem/devices/qfprom0/nvmem + + 0000000 0000 0000 0000 0000 0000 0000 0000 0000 + * + 00000a0 db10 2240 0000 e000 0c00 0c00 0000 0c00 + 0000000 0000 0000 0000 0000 0000 0000 0000 0000 + ... + * + 0001000 diff --git a/Documentation/ABI/stable/sysfs-bus-usb b/Documentation/ABI/stable/sysfs-bus-usb index b832eeff9999..cad4bc232520 100644 --- a/Documentation/ABI/stable/sysfs-bus-usb +++ b/Documentation/ABI/stable/sysfs-bus-usb @@ -50,8 +50,10 @@ Description: Tools can use this file and the connected_duration file to compute the percentage of time that a device has been active. - For example, - echo $((100 * `cat active_duration` / `cat connected_duration`)) + For example:: + + echo $((100 * `cat active_duration` / `cat connected_duration`)) + will give an integer percentage. Note that this does not account for counter wrap. Users: diff --git a/Documentation/ABI/stable/sysfs-class-backlight b/Documentation/ABI/stable/sysfs-class-backlight index 70302f370e7e..023fb52645f8 100644 --- a/Documentation/ABI/stable/sysfs-class-backlight +++ b/Documentation/ABI/stable/sysfs-class-backlight @@ -4,6 +4,7 @@ KernelVersion: 2.6.12 Contact: Richard Purdie Description: Control BACKLIGHT power, values are FB_BLANK_* from fb.h + - FB_BLANK_UNBLANK (0) : power on. - FB_BLANK_POWERDOWN (4) : power off Users: HAL diff --git a/Documentation/ABI/stable/sysfs-class-infiniband b/Documentation/ABI/stable/sysfs-class-infiniband index 87b11f91b425..348c4ac803ad 100644 --- a/Documentation/ABI/stable/sysfs-class-infiniband +++ b/Documentation/ABI/stable/sysfs-class-infiniband @@ -8,12 +8,14 @@ Date: Apr, 2005 KernelVersion: v2.6.12 Contact: linux-rdma@vger.kernel.org Description: + =============== =========================================== node_type: (RO) Node type (CA, RNIC, usNIC, usNIC UDP, switch or router) node_guid: (RO) Node GUID sys_image_guid: (RO) System image GUID + =============== =========================================== What: /sys/class/infiniband//node_desc @@ -47,6 +49,7 @@ KernelVersion: v2.6.12 Contact: linux-rdma@vger.kernel.org Description: + =============== =============================================== lid: (RO) Port LID rate: (RO) Port data rate (active width * active @@ -66,8 +69,9 @@ Description: cap_mask: (RO) Port capability mask. 2 bits here are settable- IsCommunicationManagementSupported - (set when CM module is loaded) and IsSM (set via - open of issmN file). + (set when CM module is loaded) and IsSM (set + via open of issmN file). + =============== =============================================== What: /sys/class/infiniband//ports//link_layer @@ -103,8 +107,7 @@ Date: Apr, 2005 KernelVersion: v2.6.12 Contact: linux-rdma@vger.kernel.org Description: - Errors info: - ----------- + **Errors info**: symbol_error: (RO) Total number of minor link errors detected on one or more physical lanes. @@ -142,8 +145,7 @@ Description: intervention. It can also indicate hardware issues or extremely poor link signal integrity - Data info: - --------- + **Data info**: port_xmit_data: (RO) Total number of data octets, divided by 4 (lanes), transmitted on all VLs. This is 64 bit counter @@ -176,8 +178,7 @@ Description: transmitted on all VLs from the port. This may include multicast packets with errors. - Misc info: - --------- + **Misc info**: port_xmit_discards: (RO) Total number of outbound packets discarded by the port because the port is down or congested. @@ -244,9 +245,11 @@ Description: two umad devices and two issm devices, while a switch will have one device of each type (for switch port 0). + ======= ===================================== ibdev: (RO) Show Infiniband (IB) device name port: (RO) Display port number + ======= ===================================== What: /sys/class/infiniband_mad/abi_version @@ -264,10 +267,12 @@ Date: Sept, 2005 KernelVersion: v2.6.14 Contact: linux-rdma@vger.kernel.org Description: + =============== =========================================== ibdev: (RO) Display Infiniband (IB) device name abi_version: (RO) Show ABI version of IB device specific interfaces. + =============== =========================================== What: /sys/class/infiniband_verbs/abi_version @@ -289,12 +294,14 @@ Date: Apr, 2005 KernelVersion: v2.6.12 Contact: linux-rdma@vger.kernel.org Description: + =============== ================================================ hw_rev: (RO) Hardware revision number hca_type: (RO) Host Channel Adapter type: MT23108, MT25208 (MT23108 compat mode), MT25208 or MT25204 board_id: (RO) Manufacturing board ID + =============== ================================================ sysfs interface for Mellanox ConnectX HCA IB driver (mlx4) @@ -307,11 +314,13 @@ Date: Sep, 2007 KernelVersion: v2.6.24 Contact: linux-rdma@vger.kernel.org Description: + =============== =============================== hw_rev: (RO) Hardware revision number hca_type: (RO) Host channel adapter type board_id: (RO) Manufacturing board ID + =============== =============================== What: /sys/class/infiniband/mlx4_X/iov/ports//gids/ @@ -337,6 +346,7 @@ Description: example, ports/1/pkeys/10 contains the value at index 10 in port 1's P_Key table. + ======================= ========================================== gids/: (RO) The physical port gids n = 0..127 admin_guids/: (RW) Allows examining or changing the @@ -365,6 +375,7 @@ Description: guest, whenever it uses its pkey index 1, will actually be using the real pkey index 10. + ======================= ========================================== What: /sys/class/infiniband/mlx4_X/iov//ports//smi_enabled @@ -376,12 +387,14 @@ Description: Enabling QP0 on VFs for selected VF/port. By default, no VFs are enabled for QP0 operation. - smi_enabled: (RO) Indicates whether smi is currently enabled - for the indicated VF/port + ================= ==== =========================================== + smi_enabled: (RO) Indicates whether smi is currently enabled + for the indicated VF/port - enable_smi_admin:(RW) Used by the admin to request that smi - capability be enabled or disabled for the - indicated VF/port. 0 = disable, 1 = enable. + enable_smi_admin: (RW) Used by the admin to request that smi + capability be enabled or disabled for the + indicated VF/port. 0 = disable, 1 = enable. + ================= ==== =========================================== The requested enablement will occur at the next reset of the VF (e.g. driver restart on the VM which owns the VF). @@ -398,6 +411,7 @@ KernelVersion: v2.6.35 Contact: linux-rdma@vger.kernel.org Description: + =============== ============================================= hw_rev: (RO) Hardware revision number hca_type: (RO) Driver short name. Should normally match @@ -406,6 +420,7 @@ Description: board_id: (RO) Manufacturing board id. (Vendor + device information) + =============== ============================================= sysfs interface for Intel IB driver qib @@ -426,6 +441,7 @@ Date: May, 2010 KernelVersion: v2.6.35 Contact: linux-rdma@vger.kernel.org Description: + =============== ====================================================== version: (RO) Display version information of installed software and drivers. @@ -452,6 +468,7 @@ Description: chip_reset: (WO) Reset the chip if possible by writing "reset" to this file. Only allowed if no user contexts are open that use chip resources. + =============== ====================================================== What: /sys/class/infiniband/qibX/ports/N/sl2vl/[0-15] @@ -471,14 +488,16 @@ Contact: linux-rdma@vger.kernel.org Description: Per-port congestion control. Both are binary attributes. - cc_table_bin: (RO) Congestion control table size followed by + =============== ================================================ + cc_table_bin (RO) Congestion control table size followed by table entries. - cc_settings_bin:(RO) Congestion settings: port control, control + cc_settings_bin (RO) Congestion settings: port control, control map and an array of 16 entries for the congestion entries - increase, timer, event log trigger threshold and the minimum injection rate delay. + =============== ================================================ What: /sys/class/infiniband/qibX/ports/N/linkstate/loopback What: /sys/class/infiniband/qibX/ports/N/linkstate/led_override @@ -491,6 +510,7 @@ Contact: linux-rdma@vger.kernel.org Description: [to be documented] + =============== =============================================== loopback: (WO) led_override: (WO) hrtbt_enable: (RW) @@ -501,6 +521,7 @@ Description: errors. Possible states are- "Initted", "Present", "IB_link_up", "IB_configured" or "Fatal_Hardware_Error". + =============== =============================================== What: /sys/class/infiniband/qibX/ports/N/diag_counters/rc_resends What: /sys/class/infiniband/qibX/ports/N/diag_counters/seq_naks @@ -549,6 +570,7 @@ Contact: Christian Benvenuti , linux-rdma@vger.kernel.org Description: + =============== =============================================== board_id: (RO) Manufacturing board id config: (RO) Report the configuration for this PF @@ -561,6 +583,7 @@ Description: iface: (RO) Shows which network interface this usNIC entry is associated to (visible with ifconfig). + =============== =============================================== What: /sys/class/infiniband/usnic_X/qpn/summary What: /sys/class/infiniband/usnic_X/qpn/context @@ -605,6 +628,7 @@ Date: May, 2016 KernelVersion: v4.6 Contact: linux-rdma@vger.kernel.org Description: + =============== ============================================= hw_rev: (RO) Hardware revision number board_id: (RO) Manufacturing board id @@ -623,6 +647,7 @@ Description: available. tempsense: (RO) Thermal sense information + =============== ============================================= What: /sys/class/infiniband/hfi1_X/ports/N/CCMgtA/cc_settings_bin @@ -634,19 +659,21 @@ Contact: linux-rdma@vger.kernel.org Description: Per-port congestion control. - cc_table_bin: (RO) CCA tables used by PSM2 Congestion control + =============== ================================================ + cc_table_bin (RO) CCA tables used by PSM2 Congestion control table size followed by table entries. Binary attribute. - cc_settings_bin:(RO) Congestion settings: port control, control + cc_settings_bin (RO) Congestion settings: port control, control map and an array of 16 entries for the congestion entries - increase, timer, event log trigger threshold and the minimum injection rate delay. Binary attribute. - cc_prescan: (RW) enable prescanning for faster BECN + cc_prescan (RW) enable prescanning for faster BECN response. Write "on" to enable and "off" to disable. + =============== ================================================ What: /sys/class/infiniband/hfi1_X/ports/N/sc2vl/[0-31] What: /sys/class/infiniband/hfi1_X/ports/N/sl2sc/[0-31] @@ -655,11 +682,13 @@ Date: May, 2016 KernelVersion: v4.6 Contact: linux-rdma@vger.kernel.org Description: + =============== =================================================== sc2vl/: (RO) 32 files (0 - 31) used to translate sl->vl sl2sc/: (RO) 32 files (0 - 31) used to translate sl->sc vl2mtu/: (RO) 16 files (0 - 15) used to determine MTU for vl + =============== =================================================== What: /sys/class/infiniband/hfi1_X/sdma_N/cpu_list @@ -670,26 +699,28 @@ Contact: linux-rdma@vger.kernel.org Description: sdma/ contains one directory per sdma engine (0 - 15) + =============== ============================================== cpu_list: (RW) List of cpus for user-process to sdma engine assignment. vl: (RO) Displays the virtual lane (vl) the sdma engine maps to. + =============== ============================================== This interface gives the user control on the affinity settings for the device. As an example, to set an sdma engine irq affinity and thread affinity of a user processes to use the sdma engine, which is "near" in terms of NUMA configuration, or - physical cpu location, the user will do: + physical cpu location, the user will do:: - echo "3" > /proc/irq//smp_affinity_list - echo "4-7" > /sys/devices/.../sdma3/cpu_list - cat /sys/devices/.../sdma3/vl - 0 - echo "8" > /proc/irq//smp_affinity_list - echo "9-12" > /sys/devices/.../sdma4/cpu_list - cat /sys/devices/.../sdma4/vl - 1 + echo "3" > /proc/irq//smp_affinity_list + echo "4-7" > /sys/devices/.../sdma3/cpu_list + cat /sys/devices/.../sdma3/vl + 0 + echo "8" > /proc/irq//smp_affinity_list + echo "9-12" > /sys/devices/.../sdma4/cpu_list + cat /sys/devices/.../sdma4/vl + 1 to make sure that when a process runs on cpus 4,5,6, or 7, and uses vl=0, then sdma engine 3 is selected by the driver, and @@ -711,11 +742,13 @@ Date: Jan, 2016 KernelVersion: v4.10 Contact: linux-rdma@vger.kernel.org Description: + =============== ==== ======================== hw_rev: (RO) Hardware revision number hca_type: (RO) Show HCA type (I40IW) board_id: (RO) I40IW board ID + =============== ==== ======================== sysfs interface for QLogic qedr NIC Driver @@ -728,9 +761,11 @@ KernelVersion: v4.10 Contact: linux-rdma@vger.kernel.org Description: + =============== ==== ======================== hw_rev: (RO) Hardware revision number hca_type: (RO) Display HCA type + =============== ==== ======================== sysfs interface for VMware Paravirtual RDMA driver @@ -744,11 +779,13 @@ KernelVersion: v4.10 Contact: linux-rdma@vger.kernel.org Description: + =============== ==== ===================================== hw_rev: (RO) Hardware revision number hca_type: (RO) Host channel adapter type board_id: (RO) Display PVRDMA manufacturing board ID + =============== ==== ===================================== sysfs interface for Broadcom NetXtreme-E RoCE driver @@ -760,6 +797,8 @@ Date: Feb, 2017 KernelVersion: v4.11 Contact: linux-rdma@vger.kernel.org Description: + =============== ==== ========================= hw_rev: (RO) Hardware revision number hca_type: (RO) Host channel adapter type + =============== ==== ========================= diff --git a/Documentation/ABI/stable/sysfs-class-rfkill b/Documentation/ABI/stable/sysfs-class-rfkill index 5b154f922643..037979f7dc4b 100644 --- a/Documentation/ABI/stable/sysfs-class-rfkill +++ b/Documentation/ABI/stable/sysfs-class-rfkill @@ -2,7 +2,7 @@ rfkill - radio frequency (RF) connector kill switch support For details to this subsystem look at Documentation/driver-api/rfkill.rst. -For the deprecated /sys/class/rfkill/*/claim knobs of this interface look in +For the deprecated ``/sys/class/rfkill/*/claim`` knobs of this interface look in Documentation/ABI/removed/sysfs-class-rfkill. What: /sys/class/rfkill @@ -36,9 +36,10 @@ KernelVersion v2.6.22 Contact: linux-wireless@vger.kernel.org Description: Whether the soft blocked state is initialised from non-volatile storage at startup. -Values: A numeric value. - 0: false - 1: true +Values: A numeric value: + + - 0: false + - 1: true What: /sys/class/rfkill/rfkill[0-9]+/state @@ -54,6 +55,7 @@ Description: Current state of the transmitter. through this interface. There will likely be another attempt to remove it in the future. Values: A numeric value. + 0: RFKILL_STATE_SOFT_BLOCKED transmitter is turned off by software 1: RFKILL_STATE_UNBLOCKED @@ -69,6 +71,7 @@ KernelVersion v2.6.34 Contact: linux-wireless@vger.kernel.org Description: Current hardblock state. This file is read only. Values: A numeric value. + 0: inactive The transmitter is (potentially) active. 1: active @@ -82,7 +85,9 @@ KernelVersion v2.6.34 Contact: linux-wireless@vger.kernel.org Description: Current softblock state. This file is read and write. Values: A numeric value. + 0: inactive The transmitter is (potentially) active. + 1: active The transmitter is turned off by software. diff --git a/Documentation/ABI/stable/sysfs-class-tpm b/Documentation/ABI/stable/sysfs-class-tpm index 58e94e7d55be..ec464cf7861a 100644 --- a/Documentation/ABI/stable/sysfs-class-tpm +++ b/Documentation/ABI/stable/sysfs-class-tpm @@ -32,11 +32,11 @@ KernelVersion: 2.6.12 Contact: linux-integrity@vger.kernel.org Description: The "caps" property contains TPM manufacturer and version info. - Example output: + Example output:: - Manufacturer: 0x53544d20 - TCG version: 1.2 - Firmware version: 8.16 + Manufacturer: 0x53544d20 + TCG version: 1.2 + Firmware version: 8.16 Manufacturer is a hex dump of the 4 byte manufacturer info space in a TPM. TCG version shows the TCG TPM spec level that @@ -54,9 +54,9 @@ Description: The "durations" property shows the 3 vendor-specific values any longer than necessary before starting to poll for a result. - Example output: + Example output:: - 3015000 4508000 180995000 [original] + 3015000 4508000 180995000 [original] Here the short, medium and long durations are displayed in usecs. "[original]" indicates that the values are displayed @@ -92,14 +92,14 @@ Description: The "pcrs" property will dump the current value of all Platform values may be constantly changing, the output is only valid for a snapshot in time. - Example output: + Example output:: - PCR-00: 3A 3F 78 0F 11 A4 B4 99 69 FC AA 80 CD 6E 39 57 C3 3B 22 75 - PCR-01: 3A 3F 78 0F 11 A4 B4 99 69 FC AA 80 CD 6E 39 57 C3 3B 22 75 - PCR-02: 3A 3F 78 0F 11 A4 B4 99 69 FC AA 80 CD 6E 39 57 C3 3B 22 75 - PCR-03: 3A 3F 78 0F 11 A4 B4 99 69 FC AA 80 CD 6E 39 57 C3 3B 22 75 - PCR-04: 3A 3F 78 0F 11 A4 B4 99 69 FC AA 80 CD 6E 39 57 C3 3B 22 75 - ... + PCR-00: 3A 3F 78 0F 11 A4 B4 99 69 FC AA 80 CD 6E 39 57 C3 3B 22 75 + PCR-01: 3A 3F 78 0F 11 A4 B4 99 69 FC AA 80 CD 6E 39 57 C3 3B 22 75 + PCR-02: 3A 3F 78 0F 11 A4 B4 99 69 FC AA 80 CD 6E 39 57 C3 3B 22 75 + PCR-03: 3A 3F 78 0F 11 A4 B4 99 69 FC AA 80 CD 6E 39 57 C3 3B 22 75 + PCR-04: 3A 3F 78 0F 11 A4 B4 99 69 FC AA 80 CD 6E 39 57 C3 3B 22 75 + ... The number of PCRs and hex bytes needed to represent a PCR value will vary depending on TPM chip version. For TPM 1.1 and @@ -119,44 +119,44 @@ Description: The "pubek" property will return the TPM's public endorsement ated at TPM manufacture time and exists for the life of the chip. - Example output: - - Algorithm: 00 00 00 01 - Encscheme: 00 03 - Sigscheme: 00 01 - Parameters: 00 00 08 00 00 00 00 02 00 00 00 00 - Modulus length: 256 - Modulus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ossible values: - - Algorithm: TPM_ALG_RSA (1) - Encscheme: TPM_ES_RSAESPKCSv15 (2) + Example output:: + + Algorithm: 00 00 00 01 + Encscheme: 00 03 + Sigscheme: 00 01 + Parameters: 00 00 08 00 00 00 00 02 00 00 00 00 + Modulus length: 256 + Modulus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ossible values:: + + Algorithm: TPM_ALG_RSA (1) + Encscheme: TPM_ES_RSAESPKCSv15 (2) TPM_ES_RSAESOAEP_SHA1_MGF1 (3) - Sigscheme: TPM_SS_NONE (1) - Parameters, a byte string of 3 u32 values: + Sigscheme: TPM_SS_NONE (1) + Parameters, a byte string of 3 u32 values: Key Length (bits): 00 00 08 00 (2048) Num primes: 00 00 00 02 (2) Exponent Size: 00 00 00 00 (0 means the default exp) - Modulus Length: 256 (bytes) - Modulus: The 256 byte Endorsement Key modulus + Modulus Length: 256 (bytes) + Modulus: The 256 byte Endorsement Key modulus What: /sys/class/tpm/tpmX/device/temp_deactivated Date: April 2006 @@ -176,9 +176,9 @@ Description: The "timeouts" property shows the 4 vendor-specific values timeouts is defined by the TPM interface spec that the chip conforms to. - Example output: + Example output:: - 750000 750000 750000 750000 [original] + 750000 750000 750000 750000 [original] The four timeout values are shown in usecs, with a trailing "[original]" or "[adjusted]" depending on whether the values diff --git a/Documentation/ABI/stable/sysfs-devices b/Documentation/ABI/stable/sysfs-devices index 4404bd9b96c1..42bf1eab5677 100644 --- a/Documentation/ABI/stable/sysfs-devices +++ b/Documentation/ABI/stable/sysfs-devices @@ -1,5 +1,6 @@ -# Note: This documents additional properties of any device beyond what -# is documented in Documentation/admin-guide/sysfs-rules.rst +Note: + This documents additional properties of any device beyond what + is documented in Documentation/admin-guide/sysfs-rules.rst What: /sys/devices/*/of_node Date: February 2015 diff --git a/Documentation/ABI/stable/sysfs-driver-ib_srp b/Documentation/ABI/stable/sysfs-driver-ib_srp index 84972a57caae..bada15a329f7 100644 --- a/Documentation/ABI/stable/sysfs-driver-ib_srp +++ b/Documentation/ABI/stable/sysfs-driver-ib_srp @@ -6,6 +6,7 @@ Description: Interface for making ib_srp connect to a new target. One can request ib_srp to connect to a new target by writing a comma-separated list of login parameters to this sysfs attribute. The supported parameters are: + * id_ext, a 16-digit hexadecimal number specifying the eight byte identifier extension in the 16-byte SRP target port identifier. The target port identifier is sent by ib_srp diff --git a/Documentation/ABI/stable/sysfs-firmware-efi-vars b/Documentation/ABI/stable/sysfs-firmware-efi-vars index 5def20b9019e..46ccd233e359 100644 --- a/Documentation/ABI/stable/sysfs-firmware-efi-vars +++ b/Documentation/ABI/stable/sysfs-firmware-efi-vars @@ -17,6 +17,7 @@ Description: directory has a name of the form "-" and contains the following files: + =============== ======================================== attributes: A read-only text file enumerating the EFI variable flags. Potential values include: @@ -59,12 +60,14 @@ Description: size: As ASCII representation of the size of the variable's value. + =============== ======================================== In addition, two other magic binary files are provided in the top-level directory and are used for adding and removing variables: + =============== ======================================== new_var: Takes a "struct efi_variable" and instructs the EFI firmware to create a new variable. @@ -73,3 +76,4 @@ Description: instructs the EFI firmware to remove any variable that has a matching vendor GUID and variable key name. + =============== ======================================== diff --git a/Documentation/ABI/stable/sysfs-firmware-opal-dump b/Documentation/ABI/stable/sysfs-firmware-opal-dump index 32fe7f5c4880..1f74f45327ba 100644 --- a/Documentation/ABI/stable/sysfs-firmware-opal-dump +++ b/Documentation/ABI/stable/sysfs-firmware-opal-dump @@ -7,6 +7,7 @@ Description: This is only for the powerpc/powernv platform. + =============== =============================================== initiate_dump: When '1' is written to it, we will initiate a dump. Read this file for supported commands. @@ -19,8 +20,11 @@ Description: and ID of the dump, use the id and type files. Do not rely on any particular size of dump type or dump id. + =============== =============================================== Each dump has the following files: + + =============== =============================================== id: An ASCII representation of the dump ID in hex (e.g. '0x01') type: An ASCII representation of the type of @@ -39,3 +43,4 @@ Description: inaccessible. Reading this file will get a list of supported actions. + =============== =============================================== diff --git a/Documentation/ABI/stable/sysfs-firmware-opal-elog b/Documentation/ABI/stable/sysfs-firmware-opal-elog index 2536434d49d0..7c8a61a2d005 100644 --- a/Documentation/ABI/stable/sysfs-firmware-opal-elog +++ b/Documentation/ABI/stable/sysfs-firmware-opal-elog @@ -38,6 +38,7 @@ Description: For each log entry (directory), there are the following files: + ============== ================================================ id: An ASCII representation of the ID of the error log, in hex - e.g. "0x01". @@ -58,3 +59,4 @@ Description: entry will be removed from sysfs. Reading this file will list the supported operations (currently just acknowledge). + ============== ================================================ diff --git a/Documentation/ABI/stable/sysfs-hypervisor-xen b/Documentation/ABI/stable/sysfs-hypervisor-xen index 3cf5cdfcd9a8..748593c64568 100644 --- a/Documentation/ABI/stable/sysfs-hypervisor-xen +++ b/Documentation/ABI/stable/sysfs-hypervisor-xen @@ -33,6 +33,8 @@ Description: If running under Xen: Space separated list of supported guest system types. Each type is in the format: -.- With: + + ======== ============================================ : "xen" -- x86: paravirtualized, arm: standard "hvm" -- x86 only: fully virtualized : major guest interface version @@ -43,6 +45,7 @@ Description: If running under Xen: "x86_64": 64 bit x86 guest "armv7l": 32 bit arm guest "aarch64": 64 bit arm guest + ======== ============================================ What: /sys/hypervisor/properties/changeset Date: March 2009 diff --git a/Documentation/ABI/stable/vdso b/Documentation/ABI/stable/vdso index 55406ec8a35a..73ed1240a5c0 100644 --- a/Documentation/ABI/stable/vdso +++ b/Documentation/ABI/stable/vdso @@ -23,6 +23,7 @@ Unless otherwise noted, the set of symbols with any given version and the ABI of those symbols is considered stable. It may vary across architectures, though. -(As of this writing, this ABI documentation as been confirmed for x86_64. +Note: + As of this writing, this ABI documentation as been confirmed for x86_64. The maintainers of the other vDSO-using architectures should confirm - that it is correct for their architecture.) + that it is correct for their architecture. -- cgit v1.2.3 From 34433332841de2787f903fcf7de8dc3e06780f4a Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:39 +0100 Subject: docs: ABI: testing: make the files compatible with ReST output Some files over there won't parse well by Sphinx. Fix them. Acked-by: Jonathan Cameron # for IIO Acked-by: Fabrice Gasnier Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/58cf3c2d611e0197fb215652719ebd82ca2658db.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- .../ABI/testing/configfs-spear-pcie-gadget | 36 +-- Documentation/ABI/testing/configfs-usb-gadget | 87 +++++--- Documentation/ABI/testing/configfs-usb-gadget-hid | 10 +- .../ABI/testing/configfs-usb-gadget-rndis | 16 +- Documentation/ABI/testing/configfs-usb-gadget-uac1 | 18 +- Documentation/ABI/testing/configfs-usb-gadget-uvc | 220 ++++++++++-------- Documentation/ABI/testing/debugfs-ec | 11 +- Documentation/ABI/testing/debugfs-pktcdvd | 11 +- Documentation/ABI/testing/dev-kmsg | 27 ++- Documentation/ABI/testing/evm | 17 +- Documentation/ABI/testing/ima_policy | 30 ++- Documentation/ABI/testing/procfs-diskstats | 40 ++-- Documentation/ABI/testing/sysfs-block | 38 ++-- Documentation/ABI/testing/sysfs-block-device | 2 + Documentation/ABI/testing/sysfs-bus-acpi | 18 +- .../testing/sysfs-bus-event_source-devices-format | 3 +- .../ABI/testing/sysfs-bus-i2c-devices-pca954x | 27 ++- Documentation/ABI/testing/sysfs-bus-iio | 11 + .../testing/sysfs-bus-iio-adc-envelope-detector | 5 +- Documentation/ABI/testing/sysfs-bus-iio-cros-ec | 2 +- .../ABI/testing/sysfs-bus-iio-dfsdm-adc-stm32 | 8 +- .../ABI/testing/sysfs-bus-iio-lptimer-stm32 | 29 ++- .../ABI/testing/sysfs-bus-iio-magnetometer-hmc5843 | 19 +- .../ABI/testing/sysfs-bus-iio-temperature-max31856 | 19 +- .../ABI/testing/sysfs-bus-iio-timer-stm32 | 143 +++++++----- .../ABI/testing/sysfs-bus-intel_th-devices-msc | 4 + Documentation/ABI/testing/sysfs-bus-nfit | 2 +- .../ABI/testing/sysfs-bus-pci-devices-aer_stats | 119 +++++----- Documentation/ABI/testing/sysfs-bus-rapidio | 23 +- Documentation/ABI/testing/sysfs-bus-thunderbolt | 40 ++-- Documentation/ABI/testing/sysfs-bus-usb | 30 +-- .../ABI/testing/sysfs-bus-usb-devices-usbsevseg | 7 +- Documentation/ABI/testing/sysfs-bus-vfio-mdev | 10 +- Documentation/ABI/testing/sysfs-class-cxl | 15 +- Documentation/ABI/testing/sysfs-class-led | 2 +- .../ABI/testing/sysfs-class-led-driver-el15203000 | 247 +++++++++++---------- .../ABI/testing/sysfs-class-led-driver-sc27xx | 4 +- Documentation/ABI/testing/sysfs-class-mic | 52 +++-- Documentation/ABI/testing/sysfs-class-ocxl | 3 + Documentation/ABI/testing/sysfs-class-power | 73 +++++- .../ABI/testing/sysfs-class-power-twl4030 | 33 +-- Documentation/ABI/testing/sysfs-class-rc | 30 ++- Documentation/ABI/testing/sysfs-class-scsi_host | 7 +- Documentation/ABI/testing/sysfs-class-typec | 12 +- .../ABI/testing/sysfs-devices-platform-ACPI-TAD | 4 + .../ABI/testing/sysfs-devices-platform-docg3 | 10 +- .../sysfs-devices-platform-sh_mobile_lcdc_fb | 8 +- Documentation/ABI/testing/sysfs-devices-system-cpu | 99 ++++++--- .../ABI/testing/sysfs-devices-system-ibm-rtl | 6 +- .../ABI/testing/sysfs-driver-bd9571mwv-regulator | 4 + Documentation/ABI/testing/sysfs-driver-genwqe | 11 +- .../ABI/testing/sysfs-driver-hid-logitech-lg4ff | 18 +- Documentation/ABI/testing/sysfs-driver-hid-wiimote | 11 +- .../ABI/testing/sysfs-driver-samsung-laptop | 13 +- .../ABI/testing/sysfs-driver-toshiba_acpi | 26 +++ .../ABI/testing/sysfs-driver-toshiba_haps | 2 + Documentation/ABI/testing/sysfs-driver-wacom | 4 +- Documentation/ABI/testing/sysfs-firmware-acpi | 237 ++++++++++---------- .../ABI/testing/sysfs-firmware-dmi-entries | 50 +++-- Documentation/ABI/testing/sysfs-firmware-gsmi | 2 +- Documentation/ABI/testing/sysfs-firmware-memmap | 16 +- Documentation/ABI/testing/sysfs-fs-ext4 | 4 +- Documentation/ABI/testing/sysfs-hypervisor-xen | 13 +- Documentation/ABI/testing/sysfs-kernel-boot_params | 23 +- .../ABI/testing/sysfs-kernel-mm-hugepages | 12 +- .../ABI/testing/sysfs-platform-asus-laptop | 21 +- Documentation/ABI/testing/sysfs-platform-asus-wmi | 1 + Documentation/ABI/testing/sysfs-platform-at91 | 10 +- .../ABI/testing/sysfs-platform-eeepc-laptop | 14 +- .../ABI/testing/sysfs-platform-ideapad-laptop | 9 +- .../testing/sysfs-platform-intel-wmi-thunderbolt | 1 + Documentation/ABI/testing/sysfs-platform-sst-atom | 13 +- .../ABI/testing/sysfs-platform-usbip-vudc | 11 +- Documentation/ABI/testing/sysfs-ptp | 2 +- 74 files changed, 1336 insertions(+), 879 deletions(-) diff --git a/Documentation/ABI/testing/configfs-spear-pcie-gadget b/Documentation/ABI/testing/configfs-spear-pcie-gadget index 840c324ef34d..cf877bd341df 100644 --- a/Documentation/ABI/testing/configfs-spear-pcie-gadget +++ b/Documentation/ABI/testing/configfs-spear-pcie-gadget @@ -10,22 +10,24 @@ Description: This interfaces can be used to show spear's PCIe device capability. Nodes are only visible when configfs is mounted. To mount configfs - in /config directory use: - # mount -t configfs none /config/ + in /config directory use:: - For nth PCIe Device Controller - /config/pcie-gadget.n/ - link ... used to enable ltssm and read its status. - int_type ...used to configure and read type of supported - interrupt - no_of_msi ... used to configure number of MSI vector needed and + # mount -t configfs none /config/ + + For nth PCIe Device Controller /config/pcie-gadget.n/: + + =============== ====================================================== + link used to enable ltssm and read its status. + int_type used to configure and read type of supported interrupt + no_of_msi used to configure number of MSI vector needed and to read no of MSI granted. - inta ... write 1 to assert INTA and 0 to de-assert. - send_msi ... write MSI vector to be sent. - vendor_id ... used to write and read vendor id (hex) - device_id ... used to write and read device id (hex) - bar0_size ... used to write and read bar0_size - bar0_address ... used to write and read bar0 mapped area in hex. - bar0_rw_offset ... used to write and read offset of bar0 where - bar0_data will be written or read. - bar0_data ... used to write and read data at bar0_rw_offset. + inta write 1 to assert INTA and 0 to de-assert. + send_msi write MSI vector to be sent. + vendor_id used to write and read vendor id (hex) + device_id used to write and read device id (hex) + bar0_size used to write and read bar0_size + bar0_address used to write and read bar0 mapped area in hex. + bar0_rw_offset used to write and read offset of bar0 where bar0_data + will be written or read. + bar0_data used to write and read data at bar0_rw_offset. + =============== ====================================================== diff --git a/Documentation/ABI/testing/configfs-usb-gadget b/Documentation/ABI/testing/configfs-usb-gadget index 4594cc2435e8..dc351e9af80a 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget +++ b/Documentation/ABI/testing/configfs-usb-gadget @@ -12,22 +12,24 @@ Description: The attributes of a gadget: - UDC - bind a gadget to UDC/unbind a gadget; - write UDC's name found in /sys/class/udc/* - to bind a gadget, empty string "" to unbind. - - max_speed - maximum speed the driver supports. Valid - names are super-speed-plus, super-speed, - high-speed, full-speed, and low-speed. - - bDeviceClass - USB device class code - bDeviceSubClass - USB device subclass code - bDeviceProtocol - USB device protocol code - bMaxPacketSize0 - maximum endpoint 0 packet size - bcdDevice - bcd device release number - bcdUSB - bcd USB specification version number - idProduct - product ID - idVendor - vendor ID + ================ ============================================ + UDC bind a gadget to UDC/unbind a gadget; + write UDC's name found in /sys/class/udc/* + to bind a gadget, empty string "" to unbind. + + max_speed maximum speed the driver supports. Valid + names are super-speed-plus, super-speed, + high-speed, full-speed, and low-speed. + + bDeviceClass USB device class code + bDeviceSubClass USB device subclass code + bDeviceProtocol USB device protocol code + bMaxPacketSize0 maximum endpoint 0 packet size + bcdDevice bcd device release number + bcdUSB bcd USB specification version number + idProduct product ID + idVendor vendor ID + ================ ============================================ What: /config/usb-gadget/gadget/configs Date: Jun 2013 @@ -41,8 +43,10 @@ KernelVersion: 3.11 Description: The attributes of a configuration: - bmAttributes - configuration characteristics - MaxPower - maximum power consumption from the bus + ================ ====================================== + bmAttributes configuration characteristics + MaxPower maximum power consumption from the bus + ================ ====================================== What: /config/usb-gadget/gadget/configs/config/strings Date: Jun 2013 @@ -57,7 +61,9 @@ KernelVersion: 3.11 Description: The attributes: - configuration - configuration description + ================ ========================= + configuration configuration description + ================ ========================= What: /config/usb-gadget/gadget/functions @@ -76,8 +82,10 @@ Description: The attributes: - compatible_id - 8-byte string for "Compatible ID" - sub_compatible_id - 8-byte string for "Sub Compatible ID" + ================= ===================================== + compatible_id 8-byte string for "Compatible ID" + sub_compatible_id 8-byte string for "Sub Compatible ID" + ================= ===================================== What: /config/usb-gadget/gadget/functions/./interface./ Date: May 2014 @@ -89,16 +97,19 @@ Description: The attributes: - type - value 1..7 for interpreting the data - 1: unicode string - 2: unicode string with environment variable - 3: binary - 4: little-endian 32-bit - 5: big-endian 32-bit - 6: unicode string with a symbolic link - 7: multiple unicode strings - data - blob of data to be interpreted depending on + ===== =============================================== + type value 1..7 for interpreting the data + + - 1: unicode string + - 2: unicode string with environment variable + - 3: binary + - 4: little-endian 32-bit + - 5: big-endian 32-bit + - 6: unicode string with a symbolic link + - 7: multiple unicode strings + data blob of data to be interpreted depending on type + ===== =============================================== What: /config/usb-gadget/gadget/strings Date: Jun 2013 @@ -113,9 +124,11 @@ KernelVersion: 3.11 Description: The attributes: - serialnumber - gadget's serial number (string) - product - gadget's product description - manufacturer - gadget's manufacturer description + ============ ================================= + serialnumber gadget's serial number (string) + product gadget's product description + manufacturer gadget's manufacturer description + ============ ================================= What: /config/usb-gadget/gadget/os_desc Date: May 2014 @@ -123,8 +136,10 @@ KernelVersion: 3.16 Description: This group contains "OS String" extension handling attributes. - use - flag turning "OS Desctiptors" support on/off - b_vendor_code - one-byte value used for custom per-device and + ============= =============================================== + use flag turning "OS Desctiptors" support on/off + b_vendor_code one-byte value used for custom per-device and per-interface requests - qw_sign - an identifier to be reported as "OS String" + qw_sign an identifier to be reported as "OS String" proper + ============= =============================================== diff --git a/Documentation/ABI/testing/configfs-usb-gadget-hid b/Documentation/ABI/testing/configfs-usb-gadget-hid index f12e00e6baa3..748705c4cb58 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget-hid +++ b/Documentation/ABI/testing/configfs-usb-gadget-hid @@ -4,8 +4,10 @@ KernelVersion: 3.19 Description: The attributes: - protocol - HID protocol to use - report_desc - blob corresponding to HID report descriptors + ============= ============================================ + protocol HID protocol to use + report_desc blob corresponding to HID report descriptors except the data passed through /dev/hidg - report_length - HID report length - subclass - HID device subclass to use + report_length HID report length + subclass HID device subclass to use + ============= ============================================ diff --git a/Documentation/ABI/testing/configfs-usb-gadget-rndis b/Documentation/ABI/testing/configfs-usb-gadget-rndis index 137399095d74..9416eda7fe93 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget-rndis +++ b/Documentation/ABI/testing/configfs-usb-gadget-rndis @@ -4,14 +4,16 @@ KernelVersion: 3.11 Description: The attributes: - ifname - network device interface name associated with + ========= ============================================= + ifname network device interface name associated with this function instance - qmult - queue length multiplier for high and + qmult queue length multiplier for high and super speed - host_addr - MAC address of host's end of this + host_addr MAC address of host's end of this Ethernet over USB link - dev_addr - MAC address of device's end of this + dev_addr MAC address of device's end of this Ethernet over USB link - class - USB interface class, default is 02 (hex) - subclass - USB interface subclass, default is 06 (hex) - protocol - USB interface protocol, default is 00 (hex) + class USB interface class, default is 02 (hex) + subclass USB interface subclass, default is 06 (hex) + protocol USB interface protocol, default is 00 (hex) + ========= ============================================= diff --git a/Documentation/ABI/testing/configfs-usb-gadget-uac1 b/Documentation/ABI/testing/configfs-usb-gadget-uac1 index abfe447c848f..dc23fd776943 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget-uac1 +++ b/Documentation/ABI/testing/configfs-usb-gadget-uac1 @@ -4,11 +4,13 @@ KernelVersion: 4.14 Description: The attributes: - c_chmask - capture channel mask - c_srate - capture sampling rate - c_ssize - capture sample size (bytes) - p_chmask - playback channel mask - p_srate - playback sampling rate - p_ssize - playback sample size (bytes) - req_number - the number of pre-allocated request - for both capture and playback + ========== =================================== + c_chmask capture channel mask + c_srate capture sampling rate + c_ssize capture sample size (bytes) + p_chmask playback channel mask + p_srate playback sampling rate + p_ssize playback sample size (bytes) + req_number the number of pre-allocated request + for both capture and playback + ========== =================================== diff --git a/Documentation/ABI/testing/configfs-usb-gadget-uvc b/Documentation/ABI/testing/configfs-usb-gadget-uvc index 809765bd9573..cee81b0347bb 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget-uvc +++ b/Documentation/ABI/testing/configfs-usb-gadget-uvc @@ -3,9 +3,11 @@ Date: Dec 2014 KernelVersion: 4.0 Description: UVC function directory - streaming_maxburst - 0..15 (ss only) - streaming_maxpacket - 1..1023 (fs), 1..3072 (hs/ss) - streaming_interval - 1..16 + =================== ============================= + streaming_maxburst 0..15 (ss only) + streaming_maxpacket 1..1023 (fs), 1..3072 (hs/ss) + streaming_interval 1..16 + =================== ============================= What: /config/usb-gadget/gadget/functions/uvc.name/control Date: Dec 2014 @@ -13,8 +15,11 @@ KernelVersion: 4.0 Description: Control descriptors All attributes read only: - bInterfaceNumber - USB interface number for this - streaming interface + + ================ ============================= + bInterfaceNumber USB interface number for this + streaming interface + ================ ============================= What: /config/usb-gadget/gadget/functions/uvc.name/control/class Date: Dec 2014 @@ -47,13 +52,16 @@ KernelVersion: 4.0 Description: Default output terminal descriptors All attributes read only: - iTerminal - index of string descriptor - bSourceID - id of the terminal to which this terminal + + ============== ============================================= + iTerminal index of string descriptor + bSourceID id of the terminal to which this terminal is connected - bAssocTerminal - id of the input terminal to which this output + bAssocTerminal id of the input terminal to which this output terminal is associated - wTerminalType - terminal type - bTerminalID - a non-zero id of this terminal + wTerminalType terminal type + bTerminalID a non-zero id of this terminal + ============== ============================================= What: /config/usb-gadget/gadget/functions/uvc.name/control/terminal/camera Date: Dec 2014 @@ -66,16 +74,19 @@ KernelVersion: 4.0 Description: Default camera terminal descriptors All attributes read only: - bmControls - bitmap specifying which controls are - supported for the video stream - wOcularFocalLength - the value of Locular - wObjectiveFocalLengthMax- the value of Lmin - wObjectiveFocalLengthMin- the value of Lmax - iTerminal - index of string descriptor - bAssocTerminal - id of the output terminal to which - this terminal is connected - wTerminalType - terminal type - bTerminalID - a non-zero id of this terminal + + ======================== ==================================== + bmControls bitmap specifying which controls are + supported for the video stream + wOcularFocalLength the value of Locular + wObjectiveFocalLengthMax the value of Lmin + wObjectiveFocalLengthMin the value of Lmax + iTerminal index of string descriptor + bAssocTerminal id of the output terminal to which + this terminal is connected + wTerminalType terminal type + bTerminalID a non-zero id of this terminal + ======================== ==================================== What: /config/usb-gadget/gadget/functions/uvc.name/control/processing Date: Dec 2014 @@ -88,13 +99,16 @@ KernelVersion: 4.0 Description: Default processing unit descriptors All attributes read only: - iProcessing - index of string descriptor - bmControls - bitmap specifying which controls are + + =============== ======================================== + iProcessing index of string descriptor + bmControls bitmap specifying which controls are supported for the video stream - wMaxMultiplier - maximum digital magnification x100 - bSourceID - id of the terminal to which this unit is + wMaxMultiplier maximum digital magnification x100 + bSourceID id of the terminal to which this unit is connected - bUnitID - a non-zero id of this unit + bUnitID a non-zero id of this unit + =============== ======================================== What: /config/usb-gadget/gadget/functions/uvc.name/control/header Date: Dec 2014 @@ -114,8 +128,11 @@ KernelVersion: 4.0 Description: Streaming descriptors All attributes read only: - bInterfaceNumber - USB interface number for this - streaming interface + + ================ ============================= + bInterfaceNumber USB interface number for this + streaming interface + ================ ============================= What: /config/usb-gadget/gadget/functions/uvc.name/streaming/class Date: Dec 2014 @@ -148,13 +165,16 @@ KernelVersion: 4.0 Description: Default color matching descriptors All attributes read only: - bMatrixCoefficients - matrix used to compute luma and - chroma values from the color primaries - bTransferCharacteristics- optoelectronic transfer - characteristic of the source picutre, - also called the gamma function - bColorPrimaries - color primaries and the reference - white + + ======================== ====================================== + bMatrixCoefficients matrix used to compute luma and + chroma values from the color primaries + bTransferCharacteristics optoelectronic transfer + characteristic of the source picutre, + also called the gamma function + bColorPrimaries color primaries and the reference + white + ======================== ====================================== What: /config/usb-gadget/gadget/functions/uvc.name/streaming/mjpeg Date: Dec 2014 @@ -168,47 +188,52 @@ Description: Specific MJPEG format descriptors All attributes read only, except bmaControls and bDefaultFrameIndex: - bFormatIndex - unique id for this format descriptor; + + =================== ===================================== + bFormatIndex unique id for this format descriptor; only defined after parent header is linked into the streaming class; read-only - bmaControls - this format's data for bmaControls in + bmaControls this format's data for bmaControls in the streaming header - bmInterfaceFlags - specifies interlace information, + bmInterfaceFlags specifies interlace information, read-only - bAspectRatioY - the X dimension of the picture aspect + bAspectRatioY the X dimension of the picture aspect ratio, read-only - bAspectRatioX - the Y dimension of the picture aspect + bAspectRatioX the Y dimension of the picture aspect ratio, read-only - bmFlags - characteristics of this format, + bmFlags characteristics of this format, read-only - bDefaultFrameIndex - optimum frame index for this stream + bDefaultFrameIndex optimum frame index for this stream + =================== ===================================== What: /config/usb-gadget/gadget/functions/uvc.name/streaming/mjpeg/name/name Date: Dec 2014 KernelVersion: 4.0 Description: Specific MJPEG frame descriptors - bFrameIndex - unique id for this framedescriptor; - only defined after parent format is - linked into the streaming header; - read-only - dwFrameInterval - indicates how frame interval can be - programmed; a number of values - separated by newline can be specified - dwDefaultFrameInterval - the frame interval the device would - like to use as default - dwMaxVideoFrameBufferSize- the maximum number of bytes the - compressor will produce for a video - frame or still image - dwMaxBitRate - the maximum bit rate at the shortest - frame interval in bps - dwMinBitRate - the minimum bit rate at the longest - frame interval in bps - wHeight - height of decoded bitmap frame in px - wWidth - width of decoded bitmam frame in px - bmCapabilities - still image support, fixed frame-rate - support + ========================= ===================================== + bFrameIndex unique id for this framedescriptor; + only defined after parent format is + linked into the streaming header; + read-only + dwFrameInterval indicates how frame interval can be + programmed; a number of values + separated by newline can be specified + dwDefaultFrameInterval the frame interval the device would + like to use as default + dwMaxVideoFrameBufferSize the maximum number of bytes the + compressor will produce for a video + frame or still image + dwMaxBitRate the maximum bit rate at the shortest + frame interval in bps + dwMinBitRate the minimum bit rate at the longest + frame interval in bps + wHeight height of decoded bitmap frame in px + wWidth width of decoded bitmam frame in px + bmCapabilities still image support, fixed frame-rate + support + ========================= ===================================== What: /config/usb-gadget/gadget/functions/uvc.name/streaming/uncompressed Date: Dec 2014 @@ -220,50 +245,54 @@ Date: Dec 2014 KernelVersion: 4.0 Description: Specific uncompressed format descriptors - bFormatIndex - unique id for this format descriptor; + ================== ======================================= + bFormatIndex unique id for this format descriptor; only defined after parent header is linked into the streaming class; read-only - bmaControls - this format's data for bmaControls in + bmaControls this format's data for bmaControls in the streaming header - bmInterfaceFlags - specifies interlace information, + bmInterfaceFlags specifies interlace information, read-only - bAspectRatioY - the X dimension of the picture aspect + bAspectRatioY the X dimension of the picture aspect ratio, read-only - bAspectRatioX - the Y dimension of the picture aspect + bAspectRatioX the Y dimension of the picture aspect ratio, read-only - bDefaultFrameIndex - optimum frame index for this stream - bBitsPerPixel - number of bits per pixel used to + bDefaultFrameIndex optimum frame index for this stream + bBitsPerPixel number of bits per pixel used to specify color in the decoded video frame - guidFormat - globally unique id used to identify + guidFormat globally unique id used to identify stream-encoding format + ================== ======================================= What: /config/usb-gadget/gadget/functions/uvc.name/streaming/uncompressed/name/name Date: Dec 2014 KernelVersion: 4.0 Description: Specific uncompressed frame descriptors - bFrameIndex - unique id for this framedescriptor; - only defined after parent format is - linked into the streaming header; - read-only - dwFrameInterval - indicates how frame interval can be - programmed; a number of values - separated by newline can be specified - dwDefaultFrameInterval - the frame interval the device would - like to use as default - dwMaxVideoFrameBufferSize- the maximum number of bytes the - compressor will produce for a video - frame or still image - dwMaxBitRate - the maximum bit rate at the shortest - frame interval in bps - dwMinBitRate - the minimum bit rate at the longest - frame interval in bps - wHeight - height of decoded bitmap frame in px - wWidth - width of decoded bitmam frame in px - bmCapabilities - still image support, fixed frame-rate - support + ========================= ===================================== + bFrameIndex unique id for this framedescriptor; + only defined after parent format is + linked into the streaming header; + read-only + dwFrameInterval indicates how frame interval can be + programmed; a number of values + separated by newline can be specified + dwDefaultFrameInterval the frame interval the device would + like to use as default + dwMaxVideoFrameBufferSize the maximum number of bytes the + compressor will produce for a video + frame or still image + dwMaxBitRate the maximum bit rate at the shortest + frame interval in bps + dwMinBitRate the minimum bit rate at the longest + frame interval in bps + wHeight height of decoded bitmap frame in px + wWidth width of decoded bitmam frame in px + bmCapabilities still image support, fixed frame-rate + support + ========================= ===================================== What: /config/usb-gadget/gadget/functions/uvc.name/streaming/header Date: Dec 2014 @@ -276,17 +305,20 @@ KernelVersion: 4.0 Description: Specific streaming header descriptors All attributes read only: - bTriggerUsage - how the host software will respond to + + ==================== ===================================== + bTriggerUsage how the host software will respond to a hardware trigger interrupt event - bTriggerSupport - flag specifying if hardware + bTriggerSupport flag specifying if hardware triggering is supported - bStillCaptureMethod - method of still image caputre + bStillCaptureMethod method of still image caputre supported - bTerminalLink - id of the output terminal to which + bTerminalLink id of the output terminal to which the video endpoint of this interface is connected - bmInfo - capabilities of this video streaming + bmInfo capabilities of this video streaming interface + ==================== ===================================== What: /sys/class/udc/udc.name/device/gadget/video4linux/video.name/function_name Date: May 2018 diff --git a/Documentation/ABI/testing/debugfs-ec b/Documentation/ABI/testing/debugfs-ec index 6546115a94da..ab6099daa8f5 100644 --- a/Documentation/ABI/testing/debugfs-ec +++ b/Documentation/ABI/testing/debugfs-ec @@ -6,7 +6,7 @@ Description: General information like which GPE is assigned to the EC and whether the global lock should get used. Knowing the EC GPE one can watch the amount of HW events related to -the EC here (XY -> GPE number from /sys/kernel/debug/ec/*/gpe): +the EC here (XY -> GPE number from `/sys/kernel/debug/ec/*/gpe`): /sys/firmware/acpi/interrupts/gpeXY The io file is binary and a userspace tool located here: @@ -14,7 +14,8 @@ ftp://ftp.suse.com/pub/people/trenn/sources/ec/ should get used to read out the 256 Embedded Controller registers or writing to them. -CAUTION: Do not write to the Embedded Controller if you don't know -what you are doing! Rebooting afterwards also is a good idea. -This can influence the way your machine is cooled and fans may -not get switched on again after you did a wrong write. +CAUTION: + Do not write to the Embedded Controller if you don't know + what you are doing! Rebooting afterwards also is a good idea. + This can influence the way your machine is cooled and fans may + not get switched on again after you did a wrong write. diff --git a/Documentation/ABI/testing/debugfs-pktcdvd b/Documentation/ABI/testing/debugfs-pktcdvd index cf11736acb76..787907d70462 100644 --- a/Documentation/ABI/testing/debugfs-pktcdvd +++ b/Documentation/ABI/testing/debugfs-pktcdvd @@ -4,16 +4,15 @@ KernelVersion: 2.6.20 Contact: Thomas Maier Description: -debugfs interface ------------------ - The pktcdvd module (packet writing driver) creates these files in debugfs: /sys/kernel/debug/pktcdvd/pktcdvd[0-7]/ + + ==== ====== ==================================== info (0444) Lots of driver statistics and infos. + ==== ====== ==================================== -Example: -------- +Example:: -cat /sys/kernel/debug/pktcdvd/pktcdvd0/info + cat /sys/kernel/debug/pktcdvd/pktcdvd0/info diff --git a/Documentation/ABI/testing/dev-kmsg b/Documentation/ABI/testing/dev-kmsg index 3c0bb76e3417..a377b6c093c9 100644 --- a/Documentation/ABI/testing/dev-kmsg +++ b/Documentation/ABI/testing/dev-kmsg @@ -6,6 +6,7 @@ Description: The /dev/kmsg character device node provides userspace access to the kernel's printk buffer. Injecting messages: + Every write() to the opened device node places a log entry in the kernel's printk buffer. @@ -21,6 +22,7 @@ Description: The /dev/kmsg character device node provides userspace access the messages can always be reliably determined. Accessing the buffer: + Every read() from the opened device node receives one record of the kernel's printk buffer. @@ -48,6 +50,7 @@ Description: The /dev/kmsg character device node provides userspace access if needed, without limiting the interface to a single reader. The device supports seek with the following parameters: + SEEK_SET, 0 seek to the first entry in the buffer SEEK_END, 0 @@ -87,18 +90,22 @@ Description: The /dev/kmsg character device node provides userspace access readable context of the message, for reliable processing in userspace. - Example: - 7,160,424069,-;pci_root PNP0A03:00: host bridge window [io 0x0000-0x0cf7] (ignored) - SUBSYSTEM=acpi - DEVICE=+acpi:PNP0A03:00 - 6,339,5140900,-;NET: Registered protocol family 10 - 30,340,5690716,-;udevd[80]: starting version 181 + Example:: + + 7,160,424069,-;pci_root PNP0A03:00: host bridge window [io 0x0000-0x0cf7] (ignored) + SUBSYSTEM=acpi + DEVICE=+acpi:PNP0A03:00 + 6,339,5140900,-;NET: Registered protocol family 10 + 30,340,5690716,-;udevd[80]: starting version 181 The DEVICE= key uniquely identifies devices the following way: - b12:8 - block dev_t - c127:3 - char dev_t - n8 - netdev ifindex - +sound:card0 - subsystem:devname + + ============ ================= + b12:8 block dev_t + c127:3 char dev_t + n8 netdev ifindex + +sound:card0 subsystem:devname + ============ ================= The flags field carries '-' by default. A 'c' indicates a fragment of a line. Note, that these hints about continuation diff --git a/Documentation/ABI/testing/evm b/Documentation/ABI/testing/evm index 201d10319fa1..3c477ba48a31 100644 --- a/Documentation/ABI/testing/evm +++ b/Documentation/ABI/testing/evm @@ -17,26 +17,33 @@ Description: echoing a value to /evm made up of the following bits: + === ================================================== Bit Effect + === ================================================== 0 Enable HMAC validation and creation 1 Enable digital signature validation 2 Permit modification of EVM-protected metadata at runtime. Not supported if HMAC validation and creation is enabled. 31 Disable further runtime modification of EVM policy + === ================================================== - For example: + For example:: - echo 1 >/evm + echo 1 >/evm will enable HMAC validation and creation - echo 0x80000003 >/evm + :: + + echo 0x80000003 >/evm will enable HMAC and digital signature validation and HMAC creation and disable all further modification of policy. - echo 0x80000006 >/evm + :: + + echo 0x80000006 >/evm will enable digital signature validation, permit modification of EVM-protected metadata and @@ -65,7 +72,7 @@ Description: Shows the set of extended attributes used to calculate or validate the EVM signature, and allows additional attributes to be added at runtime. Any signatures generated after - additional attributes are added (and on files posessing those + additional attributes are added (and on files possessing those additional attributes) will only be valid if the same additional attributes are configured on system boot. Writing a single period (.) will lock the xattr list from any further diff --git a/Documentation/ABI/testing/ima_policy b/Documentation/ABI/testing/ima_policy index cd572912c593..e35263f97fc1 100644 --- a/Documentation/ABI/testing/ima_policy +++ b/Documentation/ABI/testing/ima_policy @@ -15,19 +15,22 @@ Description: IMA appraisal, if configured, uses these file measurements for local measurement appraisal. - rule format: action [condition ...] + :: - action: measure | dont_measure | appraise | dont_appraise | - audit | hash | dont_hash - condition:= base | lsm [option] + rule format: action [condition ...] + + action: measure | dont_measure | appraise | dont_appraise | + audit | hash | dont_hash + condition:= base | lsm [option] base: [[func=] [mask=] [fsmagic=] [fsuuid=] [uid=] [euid=] [fowner=] [fsname=]] lsm: [[subj_user=] [subj_role=] [subj_type=] [obj_user=] [obj_role=] [obj_type=]] option: [[appraise_type=]] [template=] [permit_directio] [appraise_flag=] [keyrings=] - base: func:= [BPRM_CHECK][MMAP_CHECK][CREDS_CHECK][FILE_CHECK][MODULE_CHECK] - [FIRMWARE_CHECK] + base: + func:= [BPRM_CHECK][MMAP_CHECK][CREDS_CHECK][FILE_CHECK]MODULE_CHECK] + [FIRMWARE_CHECK] [KEXEC_KERNEL_CHECK] [KEXEC_INITRAMFS_CHECK] [KEXEC_CMDLINE] [KEY_CHECK] mask:= [[^]MAY_READ] [[^]MAY_WRITE] [[^]MAY_APPEND] @@ -37,8 +40,9 @@ Description: uid:= decimal value euid:= decimal value fowner:= decimal value - lsm: are LSM specific - option: appraise_type:= [imasig] [imasig|modsig] + lsm: are LSM specific + option: + appraise_type:= [imasig] [imasig|modsig] appraise_flag:= [check_blacklist] Currently, blacklist check is only for files signed with appended signature. @@ -49,7 +53,7 @@ Description: (eg, ima-ng). Only valid when action is "measure". pcr:= decimal value - default policy: + default policy: # PROC_SUPER_MAGIC dont_measure fsmagic=0x9fa0 dont_appraise fsmagic=0x9fa0 @@ -97,7 +101,8 @@ Description: Examples of LSM specific definitions: - SELinux: + SELinux:: + dont_measure obj_type=var_log_t dont_appraise obj_type=var_log_t dont_measure obj_type=auditd_log_t @@ -105,10 +110,11 @@ Description: measure subj_user=system_u func=FILE_CHECK mask=MAY_READ measure subj_role=system_r func=FILE_CHECK mask=MAY_READ - Smack: + Smack:: + measure subj_user=_ func=FILE_CHECK mask=MAY_READ - Example of measure rules using alternate PCRs: + Example of measure rules using alternate PCRs:: measure func=KEXEC_KERNEL_CHECK pcr=4 measure func=KEXEC_INITRAMFS_CHECK pcr=5 diff --git a/Documentation/ABI/testing/procfs-diskstats b/Documentation/ABI/testing/procfs-diskstats index 70dcaf2481f4..df5a3a8c1edf 100644 --- a/Documentation/ABI/testing/procfs-diskstats +++ b/Documentation/ABI/testing/procfs-diskstats @@ -6,28 +6,32 @@ Description: of block devices. Each line contains the following 14 fields: - 1 - major number - 2 - minor mumber - 3 - device name - 4 - reads completed successfully - 5 - reads merged - 6 - sectors read - 7 - time spent reading (ms) - 8 - writes completed - 9 - writes merged - 10 - sectors written - 11 - time spent writing (ms) - 12 - I/Os currently in progress - 13 - time spent doing I/Os (ms) - 14 - weighted time spent doing I/Os (ms) + == =================================== + 1 major number + 2 minor mumber + 3 device name + 4 reads completed successfully + 5 reads merged + 6 sectors read + 7 time spent reading (ms) + 8 writes completed + 9 writes merged + 10 sectors written + 11 time spent writing (ms) + 12 I/Os currently in progress + 13 time spent doing I/Os (ms) + 14 weighted time spent doing I/Os (ms) + == =================================== Kernel 4.18+ appends four more fields for discard tracking putting the total at 18: - 15 - discards completed successfully - 16 - discards merged - 17 - sectors discarded - 18 - time spent discarding + == =================================== + 15 discards completed successfully + 16 discards merged + 17 sectors discarded + 18 time spent discarding + == =================================== Kernel 5.5+ appends two more fields for flush requests: diff --git a/Documentation/ABI/testing/sysfs-block b/Documentation/ABI/testing/sysfs-block index 2322eb748b38..e34cdeeeb9d4 100644 --- a/Documentation/ABI/testing/sysfs-block +++ b/Documentation/ABI/testing/sysfs-block @@ -4,23 +4,27 @@ Contact: Jerome Marchand Description: The /sys/block//stat files displays the I/O statistics of disk . They contain 11 fields: - 1 - reads completed successfully - 2 - reads merged - 3 - sectors read - 4 - time spent reading (ms) - 5 - writes completed - 6 - writes merged - 7 - sectors written - 8 - time spent writing (ms) - 9 - I/Os currently in progress - 10 - time spent doing I/Os (ms) - 11 - weighted time spent doing I/Os (ms) - 12 - discards completed - 13 - discards merged - 14 - sectors discarded - 15 - time spent discarding (ms) - 16 - flush requests completed - 17 - time spent flushing (ms) + + == ============================================== + 1 reads completed successfully + 2 reads merged + 3 sectors read + 4 time spent reading (ms) + 5 writes completed + 6 writes merged + 7 sectors written + 8 time spent writing (ms) + 9 I/Os currently in progress + 10 time spent doing I/Os (ms) + 11 weighted time spent doing I/Os (ms) + 12 discards completed + 13 discards merged + 14 sectors discarded + 15 time spent discarding (ms) + 16 flush requests completed + 17 time spent flushing (ms) + == ============================================== + For more details refer Documentation/admin-guide/iostats.rst diff --git a/Documentation/ABI/testing/sysfs-block-device b/Documentation/ABI/testing/sysfs-block-device index 17f2bc7dd261..aa0fb500e3c9 100644 --- a/Documentation/ABI/testing/sysfs-block-device +++ b/Documentation/ABI/testing/sysfs-block-device @@ -8,11 +8,13 @@ Description: It has the following valid values: + == ======================================================== 0 OFF - the LED is not activated on activity 1 BLINK_ON - the LED blinks on every 10ms when activity is detected. 2 BLINK_OFF - the LED is on when idle, and blinks off every 10ms when activity is detected. + == ======================================================== Note that the user must turn sw_activity OFF it they wish to control the activity LED via the em_message file. diff --git a/Documentation/ABI/testing/sysfs-bus-acpi b/Documentation/ABI/testing/sysfs-bus-acpi index e7898cfe5fb1..c78603497b97 100644 --- a/Documentation/ABI/testing/sysfs-bus-acpi +++ b/Documentation/ABI/testing/sysfs-bus-acpi @@ -67,14 +67,16 @@ Description: The return value is a decimal integer representing the device's status bitmap: - Bit [0] – Set if the device is present. - Bit [1] – Set if the device is enabled and decoding its - resources. - Bit [2] – Set if the device should be shown in the UI. - Bit [3] – Set if the device is functioning properly (cleared if - device failed its diagnostics). - Bit [4] – Set if the battery is present. - Bits [31:5] – Reserved (must be cleared) + =========== ================================================== + Bit [0] Set if the device is present. + Bit [1] Set if the device is enabled and decoding its + resources. + Bit [2] Set if the device should be shown in the UI. + Bit [3] Set if the device is functioning properly (cleared + if device failed its diagnostics). + Bit [4] Set if the battery is present. + Bits [31:5] Reserved (must be cleared) + =========== ================================================== If bit [0] is clear, then bit 1 must also be clear (a device that is not present cannot be enabled). diff --git a/Documentation/ABI/testing/sysfs-bus-event_source-devices-format b/Documentation/ABI/testing/sysfs-bus-event_source-devices-format index 5bb793ec926c..df7ccc1b2fba 100644 --- a/Documentation/ABI/testing/sysfs-bus-event_source-devices-format +++ b/Documentation/ABI/testing/sysfs-bus-event_source-devices-format @@ -10,7 +10,8 @@ Description: name/value pairs. Userspace must be prepared for the possibility that attributes - define overlapping bit ranges. For example: + define overlapping bit ranges. For example:: + attr1 = 'config:0-23' attr2 = 'config:0-7' attr3 = 'config:12-35' diff --git a/Documentation/ABI/testing/sysfs-bus-i2c-devices-pca954x b/Documentation/ABI/testing/sysfs-bus-i2c-devices-pca954x index 0b0de8cd0d13..b6c69eb80ca4 100644 --- a/Documentation/ABI/testing/sysfs-bus-i2c-devices-pca954x +++ b/Documentation/ABI/testing/sysfs-bus-i2c-devices-pca954x @@ -6,15 +6,18 @@ Description: Value that exists only for mux devices that can be written to control the behaviour of the multiplexer on idle. Possible values: - -2 - disconnect on idle, i.e. deselect the last used - channel, which is useful when there is a device - with an address that conflicts with another - device on another mux on the same parent bus. - -1 - leave the mux as-is, which is the most optimal - setting in terms of I2C operations and is the - default mode. - 0.. - set the mux to a predetermined channel, - which is useful if there is one channel that is - used almost always, and you want to reduce the - latency for normal operations after rare - transactions on other channels + + =========== =============================================== + -2 disconnect on idle, i.e. deselect the last used + channel, which is useful when there is a device + with an address that conflicts with another + device on another mux on the same parent bus. + -1 leave the mux as-is, which is the most optimal + setting in terms of I2C operations and is the + default mode. + 0.. set the mux to a predetermined channel, + which is useful if there is one channel that is + used almost always, and you want to reduce the + latency for normal operations after rare + transactions on other channels + =========== =============================================== diff --git a/Documentation/ABI/testing/sysfs-bus-iio b/Documentation/ABI/testing/sysfs-bus-iio index a9d51810a3ba..e3df71987eff 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio +++ b/Documentation/ABI/testing/sysfs-bus-iio @@ -65,6 +65,7 @@ Contact: linux-iio@vger.kernel.org Description: When the internal sampling clock can only take a specific set of frequencies, we can specify the available values with: + - a small discrete set of values like "0 2 4 6 8" - a range with minimum, step and maximum frequencies like "[min step max]" @@ -1627,17 +1628,21 @@ Description: Mounting matrix for IIO sensors. This is a rotation matrix which informs userspace about sensor chip's placement relative to the main hardware it is mounted on. + Main hardware placement is defined according to the local reference frame related to the physical quantity the sensor measures. + Given that the rotation matrix is defined in a board specific way (platform data and / or device-tree), the main hardware reference frame definition is left to the implementor's choice (see below for a magnetometer example). + Applications should apply this rotation matrix to samples so that when main hardware reference frame is aligned onto local reference frame, then sensor chip reference frame is also perfectly aligned with it. + Matrix is a 3x3 unitary matrix and typically looks like [0, 1, 0; 1, 0, 0; 0, 0, -1]. Identity matrix [1, 0, 0; 0, 1, 0; 0, 0, 1] means sensor chip and main hardware @@ -1646,8 +1651,10 @@ Description: For example, a mounting matrix for a magnetometer sensor informs userspace about sensor chip's ORIENTATION relative to the main hardware. + More specifically, main hardware orientation is defined with respect to the LOCAL EARTH GEOMAGNETIC REFERENCE FRAME where : + * Y is in the ground plane and positive towards magnetic North ; * X is in the ground plane, perpendicular to the North axis and positive towards the East ; @@ -1656,13 +1663,16 @@ Description: An implementor might consider that for a hand-held device, a 'natural' orientation would be 'front facing camera at the top'. The main hardware reference frame could then be described as : + * Y is in the plane of the screen and is positive towards the top of the screen ; * X is in the plane of the screen, perpendicular to Y axis, and positive towards the right hand side of the screen ; * Z is perpendicular to the screen plane and positive out of the screen. + Another example for a quadrotor UAV might be : + * Y is in the plane of the propellers and positive towards the front-view camera; * X is in the plane of the propellers, perpendicular to Y axis, @@ -1704,6 +1714,7 @@ Description: This interface is deprecated; please use the Counter subsystem. A list of possible counting directions which are: + - "up" : counter device is increasing. - "down": counter device is decreasing. diff --git a/Documentation/ABI/testing/sysfs-bus-iio-adc-envelope-detector b/Documentation/ABI/testing/sysfs-bus-iio-adc-envelope-detector index 2071f9bcfaa5..1c2a07f7a75e 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio-adc-envelope-detector +++ b/Documentation/ABI/testing/sysfs-bus-iio-adc-envelope-detector @@ -5,7 +5,8 @@ Contact: Peter Rosin Description: The DAC is used to find the peak level of an alternating voltage input signal by a binary search using the output - of a comparator wired to an interrupt pin. Like so: + of a comparator wired to an interrupt pin. Like so:: + _ | \ input +------>-------|+ \ @@ -19,10 +20,12 @@ Description: | irq|------<-------' | | '-------' + The boolean invert attribute (0/1) should be set when the input signal is centered around the maximum value of the dac instead of zero. The envelope detector will search from below in this case and will also invert the result. + The edge/level of the interrupt is also switched to its opposite value. diff --git a/Documentation/ABI/testing/sysfs-bus-iio-cros-ec b/Documentation/ABI/testing/sysfs-bus-iio-cros-ec index 6158f831c761..adf24c40126f 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio-cros-ec +++ b/Documentation/ABI/testing/sysfs-bus-iio-cros-ec @@ -4,7 +4,7 @@ KernelVersion: 4.7 Contact: linux-iio@vger.kernel.org Description: Writing '1' will perform a FOC (Fast Online Calibration). The - corresponding calibration offsets can be read from *_calibbias + corresponding calibration offsets can be read from `*_calibbias` entries. What: /sys/bus/iio/devices/iio:deviceX/location diff --git a/Documentation/ABI/testing/sysfs-bus-iio-dfsdm-adc-stm32 b/Documentation/ABI/testing/sysfs-bus-iio-dfsdm-adc-stm32 index 0e66ae9b0071..91439d6d60b5 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio-dfsdm-adc-stm32 +++ b/Documentation/ABI/testing/sysfs-bus-iio-dfsdm-adc-stm32 @@ -3,14 +3,20 @@ KernelVersion: 4.14 Contact: arnaud.pouliquen@st.com Description: For audio purpose only. + Used by audio driver to set/get the spi input frequency. + This is mandatory if DFSDM is slave on SPI bus, to provide information on the SPI clock frequency during runtime Notice that the SPI frequency should be a multiple of sample frequency to ensure the precision. - if DFSDM input is SPI master + + if DFSDM input is SPI master: + Reading SPI clkout frequency, error on writing + If DFSDM input is SPI Slave: + Reading returns value previously set. Writing value before starting conversions. diff --git a/Documentation/ABI/testing/sysfs-bus-iio-lptimer-stm32 b/Documentation/ABI/testing/sysfs-bus-iio-lptimer-stm32 index ad2cc63e4bf8..73498ff666bd 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio-lptimer-stm32 +++ b/Documentation/ABI/testing/sysfs-bus-iio-lptimer-stm32 @@ -17,9 +17,11 @@ KernelVersion: 4.13 Contact: fabrice.gasnier@st.com Description: Configure the device counter quadrature modes: + - non-quadrature: Encoder IN1 input servers as the count input (up direction). + - quadrature: Encoder IN1 and IN2 inputs are mixed to get direction and count. @@ -35,23 +37,26 @@ KernelVersion: 4.13 Contact: fabrice.gasnier@st.com Description: Configure the device encoder/counter active edge: + - rising-edge - falling-edge - both-edges In non-quadrature mode, device counts up on active edge. + In quadrature mode, encoder counting scenarios are as follows: - ---------------------------------------------------------------- + + +---------+----------+--------------------+--------------------+ | Active | Level on | IN1 signal | IN2 signal | - | edge | opposite |------------------------------------------ + | edge | opposite +----------+---------+----------+---------+ | | signal | Rising | Falling | Rising | Falling | - ---------------------------------------------------------------- - | Rising | High -> | Down | - | Up | - | - | edge | Low -> | Up | - | Down | - | - ---------------------------------------------------------------- - | Falling | High -> | - | Up | - | Down | - | edge | Low -> | - | Down | - | Up | - ---------------------------------------------------------------- - | Both | High -> | Down | Up | Up | Down | - | edges | Low -> | Up | Down | Down | Up | - ---------------------------------------------------------------- + +---------+----------+----------+---------+----------+---------+ + | Rising | High -> | Down | - | Up | - | + | edge | Low -> | Up | - | Down | - | + +---------+----------+----------+---------+----------+---------+ + | Falling | High -> | - | Up | - | Down | + | edge | Low -> | - | Down | - | Up | + +---------+----------+----------+---------+----------+---------+ + | Both | High -> | Down | Up | Up | Down | + | edges | Low -> | Up | Down | Down | Up | + +---------+----------+----------+---------+----------+---------+ diff --git a/Documentation/ABI/testing/sysfs-bus-iio-magnetometer-hmc5843 b/Documentation/ABI/testing/sysfs-bus-iio-magnetometer-hmc5843 index 6275e9f56e6c..13f099ef6a95 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio-magnetometer-hmc5843 +++ b/Documentation/ABI/testing/sysfs-bus-iio-magnetometer-hmc5843 @@ -5,11 +5,16 @@ Contact: linux-iio@vger.kernel.org Description: Current configuration and available configurations for the bias current. - normal - Normal measurement configurations (default) - positivebias - Positive bias configuration - negativebias - Negative bias configuration - disabled - Only available on HMC5983. Disables magnetic + + ============ ============================================ + normal Normal measurement configurations (default) + positivebias Positive bias configuration + negativebias Negative bias configuration + disabled Only available on HMC5983. Disables magnetic sensor and enables temperature sensor. - Note: The effect of this configuration may vary - according to the device. For exact documentation - check the device's datasheet. + ============ ============================================ + + Note: + The effect of this configuration may vary + according to the device. For exact documentation + check the device's datasheet. diff --git a/Documentation/ABI/testing/sysfs-bus-iio-temperature-max31856 b/Documentation/ABI/testing/sysfs-bus-iio-temperature-max31856 index 3b3509a3ef2f..e5ef6d8e5da1 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio-temperature-max31856 +++ b/Documentation/ABI/testing/sysfs-bus-iio-temperature-max31856 @@ -5,9 +5,12 @@ Description: Open-circuit fault. The detection of open-circuit faults, such as those caused by broken thermocouple wires. Reading returns either '1' or '0'. - '1' = An open circuit such as broken thermocouple wires - has been detected. - '0' = No open circuit or broken thermocouple wires are detected + + === ======================================================= + '1' An open circuit such as broken thermocouple wires + has been detected. + '0' No open circuit or broken thermocouple wires are detected + === ======================================================= What: /sys/bus/iio/devices/iio:deviceX/fault_ovuv KernelVersion: 5.1 @@ -18,7 +21,11 @@ Description: cables by integrated MOSFETs at the T+ and T- inputs, and the BIAS output. These MOSFETs turn off when the input voltage is negative or greater than VDD. + Reading returns either '1' or '0'. - '1' = The input voltage is negative or greater than VDD. - '0' = The input voltage is positive and less than VDD (normal - state). + + === ======================================================= + '1' The input voltage is negative or greater than VDD. + '0' The input voltage is positive and less than VDD (normal + state). + === ======================================================= diff --git a/Documentation/ABI/testing/sysfs-bus-iio-timer-stm32 b/Documentation/ABI/testing/sysfs-bus-iio-timer-stm32 index b7259234ad70..a10a4de3e5fe 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio-timer-stm32 +++ b/Documentation/ABI/testing/sysfs-bus-iio-timer-stm32 @@ -3,67 +3,85 @@ KernelVersion: 4.11 Contact: benjamin.gaignard@st.com Description: Reading returns the list possible master modes which are: - - "reset" : The UG bit from the TIMx_EGR register is + + + - "reset" + The UG bit from the TIMx_EGR register is used as trigger output (TRGO). - - "enable" : The Counter Enable signal CNT_EN is used + - "enable" + The Counter Enable signal CNT_EN is used as trigger output. - - "update" : The update event is selected as trigger output. + - "update" + The update event is selected as trigger output. For instance a master timer can then be used as a prescaler for a slave timer. - - "compare_pulse" : The trigger output send a positive pulse - when the CC1IF flag is to be set. - - "OC1REF" : OC1REF signal is used as trigger output. - - "OC2REF" : OC2REF signal is used as trigger output. - - "OC3REF" : OC3REF signal is used as trigger output. - - "OC4REF" : OC4REF signal is used as trigger output. + - "compare_pulse" + The trigger output send a positive pulse + when the CC1IF flag is to be set. + - "OC1REF" + OC1REF signal is used as trigger output. + - "OC2REF" + OC2REF signal is used as trigger output. + - "OC3REF" + OC3REF signal is used as trigger output. + - "OC4REF" + OC4REF signal is used as trigger output. + Additional modes (on TRGO2 only): - - "OC5REF" : OC5REF signal is used as trigger output. - - "OC6REF" : OC6REF signal is used as trigger output. + + - "OC5REF" + OC5REF signal is used as trigger output. + - "OC6REF" + OC6REF signal is used as trigger output. - "compare_pulse_OC4REF": - OC4REF rising or falling edges generate pulses. + OC4REF rising or falling edges generate pulses. - "compare_pulse_OC6REF": - OC6REF rising or falling edges generate pulses. + OC6REF rising or falling edges generate pulses. - "compare_pulse_OC4REF_r_or_OC6REF_r": - OC4REF or OC6REF rising edges generate pulses. + OC4REF or OC6REF rising edges generate pulses. - "compare_pulse_OC4REF_r_or_OC6REF_f": - OC4REF rising or OC6REF falling edges generate pulses. + OC4REF rising or OC6REF falling edges generate + pulses. - "compare_pulse_OC5REF_r_or_OC6REF_r": - OC5REF or OC6REF rising edges generate pulses. + OC5REF or OC6REF rising edges generate pulses. - "compare_pulse_OC5REF_r_or_OC6REF_f": - OC5REF rising or OC6REF falling edges generate pulses. - - +-----------+ +-------------+ +---------+ - | Prescaler +-> | Counter | +-> | Master | TRGO(2) - +-----------+ +--+--------+-+ |-> | Control +--> - | | || +---------+ - +--v--------+-+ OCxREF || +---------+ - | Chx compare +----------> | Output | ChX - +-----------+-+ | | Control +--> - . | | +---------+ - . | | . - +-----------v-+ OC6REF | . - | Ch6 compare +---------+> - +-------------+ - - Example with: "compare_pulse_OC4REF_r_or_OC6REF_r": - - X - X X - X . . X - X . . X - X . . X - count X . . . . X - . . . . - . . . . - +---------------+ - OC4REF | . . | - +-+ . . +-+ - . +---+ . - OC6REF . | | . - +-------+ +-------+ - +-+ +-+ - TRGO2 | | | | - +-+ +---+ +---------+ + OC5REF rising or OC6REF falling edges generate + pulses. + + :: + + +-----------+ +-------------+ +---------+ + | Prescaler +-> | Counter | +-> | Master | TRGO(2) + +-----------+ +--+--------+-+ |-> | Control +--> + | | || +---------+ + +--v--------+-+ OCxREF || +---------+ + | Chx compare +----------> | Output | ChX + +-----------+-+ | | Control +--> + . | | +---------+ + . | | . + +-----------v-+ OC6REF | . + | Ch6 compare +---------+> + +-------------+ + + Example with: "compare_pulse_OC4REF_r_or_OC6REF_r":: + + X + X X + X . . X + X . . X + X . . X + count X . . . . X + . . . . + . . . . + +---------------+ + OC4REF | . . | + +-+ . . +-+ + . +---+ . + OC6REF . | | . + +-------+ +-------+ + +-+ +-+ + TRGO2 | | | | + +-+ +---+ +---------+ What: /sys/bus/iio/devices/triggerX/master_mode KernelVersion: 4.11 @@ -91,6 +109,30 @@ Description: When counting down the counter start from preset value and fire event when reach 0. +What: /sys/bus/iio/devices/iio:deviceX/in_count_quadrature_mode_available +KernelVersion: 4.12 +Contact: benjamin.gaignard@st.com +Description: + Reading returns the list possible quadrature modes. + +What: /sys/bus/iio/devices/iio:deviceX/in_count0_quadrature_mode +KernelVersion: 4.12 +Contact: benjamin.gaignard@st.com +Description: + Configure the device counter quadrature modes: + + channel_A: + Encoder A input servers as the count input and B as + the UP/DOWN direction control input. + + channel_B: + Encoder B input serves as the count input and A as + the UP/DOWN direction control input. + + quadrature: + Encoder A and B inputs are mixed to get direction + and count with a scale of 0.25. + What: /sys/bus/iio/devices/iio:deviceX/in_count_enable_mode_available KernelVersion: 4.12 Contact: benjamin.gaignard@st.com @@ -104,6 +146,7 @@ Description: Configure the device counter enable modes, in all case counting direction is set by in_count0_count_direction attribute and the counter is clocked by the internal clock. + always: Counter is always ON. diff --git a/Documentation/ABI/testing/sysfs-bus-intel_th-devices-msc b/Documentation/ABI/testing/sysfs-bus-intel_th-devices-msc index 7fd2601c2831..a74252e580a5 100644 --- a/Documentation/ABI/testing/sysfs-bus-intel_th-devices-msc +++ b/Documentation/ABI/testing/sysfs-bus-intel_th-devices-msc @@ -9,11 +9,13 @@ Date: June 2015 KernelVersion: 4.3 Contact: Alexander Shishkin Description: (RW) Configure MSC operating mode: + - "single", for contiguous buffer mode (high-order alloc); - "multi", for multiblock mode; - "ExI", for DCI handler mode; - "debug", for debug mode; - any of the currently loaded buffer sinks. + If operating mode changes, existing buffer is deallocated, provided there are no active users and tracing is not enabled, otherwise the write will fail. @@ -23,10 +25,12 @@ Date: June 2015 KernelVersion: 4.3 Contact: Alexander Shishkin Description: (RW) Configure MSC buffer size for "single" or "multi" modes. + In single mode, this is a single number of pages, has to be power of 2. In multiblock mode, this is a comma-separated list of numbers of pages for each window to be allocated. Number of windows is not limited. + Writing to this file deallocates existing buffer (provided there are no active users and tracing is not enabled) and then allocates a new one. diff --git a/Documentation/ABI/testing/sysfs-bus-nfit b/Documentation/ABI/testing/sysfs-bus-nfit index e4f76e7eab93..63ef0b9ecce7 100644 --- a/Documentation/ABI/testing/sysfs-bus-nfit +++ b/Documentation/ABI/testing/sysfs-bus-nfit @@ -1,4 +1,4 @@ -For all of the nmem device attributes under nfit/*, see the 'NVDIMM Firmware +For all of the nmem device attributes under ``nfit/*``, see the 'NVDIMM Firmware Interface Table (NFIT)' section in the ACPI specification (http://www.uefi.org/specifications) for more details. diff --git a/Documentation/ABI/testing/sysfs-bus-pci-devices-aer_stats b/Documentation/ABI/testing/sysfs-bus-pci-devices-aer_stats index 3c9a8c4a25eb..860db53037a5 100644 --- a/Documentation/ABI/testing/sysfs-bus-pci-devices-aer_stats +++ b/Documentation/ABI/testing/sysfs-bus-pci-devices-aer_stats @@ -1,6 +1,6 @@ -========================== PCIe Device AER statistics -========================== +-------------------------- + These attributes show up under all the devices that are AER capable. These statistical counters indicate the errors "as seen/reported by the device". Note that this may mean that if an endpoint is causing problems, the AER @@ -17,19 +17,18 @@ Description: List of correctable errors seen and reported by this PCI device using ERR_COR. Note that since multiple errors may be reported using a single ERR_COR message, thus TOTAL_ERR_COR at the end of the file may not match the actual - total of all the errors in the file. Sample output: -------------------------------------------------------------------------- -localhost /sys/devices/pci0000:00/0000:00:1c.0 # cat aer_dev_correctable -Receiver Error 2 -Bad TLP 0 -Bad DLLP 0 -RELAY_NUM Rollover 0 -Replay Timer Timeout 0 -Advisory Non-Fatal 0 -Corrected Internal Error 0 -Header Log Overflow 0 -TOTAL_ERR_COR 2 -------------------------------------------------------------------------- + total of all the errors in the file. Sample output:: + + localhost /sys/devices/pci0000:00/0000:00:1c.0 # cat aer_dev_correctable + Receiver Error 2 + Bad TLP 0 + Bad DLLP 0 + RELAY_NUM Rollover 0 + Replay Timer Timeout 0 + Advisory Non-Fatal 0 + Corrected Internal Error 0 + Header Log Overflow 0 + TOTAL_ERR_COR 2 What: /sys/bus/pci/devices//aer_dev_fatal Date: July 2018 @@ -39,28 +38,27 @@ Description: List of uncorrectable fatal errors seen and reported by this PCI device using ERR_FATAL. Note that since multiple errors may be reported using a single ERR_FATAL message, thus TOTAL_ERR_FATAL at the end of the file may not match the actual - total of all the errors in the file. Sample output: -------------------------------------------------------------------------- -localhost /sys/devices/pci0000:00/0000:00:1c.0 # cat aer_dev_fatal -Undefined 0 -Data Link Protocol 0 -Surprise Down Error 0 -Poisoned TLP 0 -Flow Control Protocol 0 -Completion Timeout 0 -Completer Abort 0 -Unexpected Completion 0 -Receiver Overflow 0 -Malformed TLP 0 -ECRC 0 -Unsupported Request 0 -ACS Violation 0 -Uncorrectable Internal Error 0 -MC Blocked TLP 0 -AtomicOp Egress Blocked 0 -TLP Prefix Blocked Error 0 -TOTAL_ERR_FATAL 0 -------------------------------------------------------------------------- + total of all the errors in the file. Sample output:: + + localhost /sys/devices/pci0000:00/0000:00:1c.0 # cat aer_dev_fatal + Undefined 0 + Data Link Protocol 0 + Surprise Down Error 0 + Poisoned TLP 0 + Flow Control Protocol 0 + Completion Timeout 0 + Completer Abort 0 + Unexpected Completion 0 + Receiver Overflow 0 + Malformed TLP 0 + ECRC 0 + Unsupported Request 0 + ACS Violation 0 + Uncorrectable Internal Error 0 + MC Blocked TLP 0 + AtomicOp Egress Blocked 0 + TLP Prefix Blocked Error 0 + TOTAL_ERR_FATAL 0 What: /sys/bus/pci/devices//aer_dev_nonfatal Date: July 2018 @@ -70,32 +68,31 @@ Description: List of uncorrectable nonfatal errors seen and reported by this PCI device using ERR_NONFATAL. Note that since multiple errors may be reported using a single ERR_FATAL message, thus TOTAL_ERR_NONFATAL at the end of the file may not match the - actual total of all the errors in the file. Sample output: -------------------------------------------------------------------------- -localhost /sys/devices/pci0000:00/0000:00:1c.0 # cat aer_dev_nonfatal -Undefined 0 -Data Link Protocol 0 -Surprise Down Error 0 -Poisoned TLP 0 -Flow Control Protocol 0 -Completion Timeout 0 -Completer Abort 0 -Unexpected Completion 0 -Receiver Overflow 0 -Malformed TLP 0 -ECRC 0 -Unsupported Request 0 -ACS Violation 0 -Uncorrectable Internal Error 0 -MC Blocked TLP 0 -AtomicOp Egress Blocked 0 -TLP Prefix Blocked Error 0 -TOTAL_ERR_NONFATAL 0 -------------------------------------------------------------------------- + actual total of all the errors in the file. Sample output:: + + localhost /sys/devices/pci0000:00/0000:00:1c.0 # cat aer_dev_nonfatal + Undefined 0 + Data Link Protocol 0 + Surprise Down Error 0 + Poisoned TLP 0 + Flow Control Protocol 0 + Completion Timeout 0 + Completer Abort 0 + Unexpected Completion 0 + Receiver Overflow 0 + Malformed TLP 0 + ECRC 0 + Unsupported Request 0 + ACS Violation 0 + Uncorrectable Internal Error 0 + MC Blocked TLP 0 + AtomicOp Egress Blocked 0 + TLP Prefix Blocked Error 0 + TOTAL_ERR_NONFATAL 0 -============================ PCIe Rootport AER statistics -============================ +---------------------------- + These attributes show up under only the rootports (or root complex event collectors) that are AER capable. These indicate the number of error messages as "reported to" the rootport. Please note that the rootports also transmit diff --git a/Documentation/ABI/testing/sysfs-bus-rapidio b/Documentation/ABI/testing/sysfs-bus-rapidio index 13208b27dd87..634ea207a50a 100644 --- a/Documentation/ABI/testing/sysfs-bus-rapidio +++ b/Documentation/ABI/testing/sysfs-bus-rapidio @@ -4,24 +4,27 @@ Description: an individual subdirectory with the following name format of device_name "nn:d:iiii", where: - nn - two-digit hexadecimal ID of RapidIO network where the + ==== ======================================================== + nn two-digit hexadecimal ID of RapidIO network where the device resides - d - device type: 'e' - for endpoint or 's' - for switch - iiii - four-digit device destID for endpoints, or switchID for + d device type: 'e' - for endpoint or 's' - for switch + iiii four-digit device destID for endpoints, or switchID for switches + ==== ======================================================== For example, below is a list of device directories that represents a typical RapidIO network with one switch, one host, and two agent endpoints, as it is seen by the enumerating host - (with destID = 1): + (with destID = 1):: - /sys/bus/rapidio/devices/00:e:0000 - /sys/bus/rapidio/devices/00:e:0002 - /sys/bus/rapidio/devices/00:s:0001 + /sys/bus/rapidio/devices/00:e:0000 + /sys/bus/rapidio/devices/00:e:0002 + /sys/bus/rapidio/devices/00:s:0001 - NOTE: An enumerating or discovering endpoint does not create a - sysfs entry for itself, this is why an endpoint with destID=1 is - not shown in the list. + NOTE: + An enumerating or discovering endpoint does not create a + sysfs entry for itself, this is why an endpoint with destID=1 + is not shown in the list. Attributes Common for All RapidIO Devices ----------------------------------------- diff --git a/Documentation/ABI/testing/sysfs-bus-thunderbolt b/Documentation/ABI/testing/sysfs-bus-thunderbolt index dd565c378b40..171127294674 100644 --- a/Documentation/ABI/testing/sysfs-bus-thunderbolt +++ b/Documentation/ABI/testing/sysfs-bus-thunderbolt @@ -37,16 +37,18 @@ Contact: thunderbolt-software@lists.01.org Description: This attribute holds current Thunderbolt security level set by the system BIOS. Possible values are: - none: All devices are automatically authorized - user: Devices are only authorized based on writing - appropriate value to the authorized attribute - secure: Require devices that support secure connect at - minimum. User needs to authorize each device. - dponly: Automatically tunnel Display port (and USB). No - PCIe tunnels are created. - usbonly: Automatically tunnel USB controller of the + ======= ================================================== + none All devices are automatically authorized + user Devices are only authorized based on writing + appropriate value to the authorized attribute + secure Require devices that support secure connect at + minimum. User needs to authorize each device. + dponly Automatically tunnel Display port (and USB). No + PCIe tunnels are created. + usbonly Automatically tunnel USB controller of the connected Thunderbolt dock (and Display Port). All PCIe links downstream of the dock are removed. + ======= ================================================== What: /sys/bus/thunderbolt/devices/.../authorized Date: Sep 2017 @@ -61,17 +63,23 @@ Description: This attribute is used to authorize Thunderbolt devices yet authorized. Possible values are supported: - 1: The device will be authorized and connected + + == =========================================== + 1 The device will be authorized and connected + == =========================================== When key attribute contains 32 byte hex string the possible values are: - 1: The 32 byte hex string is added to the device NVM and - the device is authorized. - 2: Send a challenge based on the 32 byte hex string. If the - challenge response from device is valid, the device is - authorized. In case of failure errno will be ENOKEY if - the device did not contain a key at all, and - EKEYREJECTED if the challenge response did not match. + + == ======================================================== + 1 The 32 byte hex string is added to the device NVM and + the device is authorized. + 2 Send a challenge based on the 32 byte hex string. If the + challenge response from device is valid, the device is + authorized. In case of failure errno will be ENOKEY if + the device did not contain a key at all, and + EKEYREJECTED if the challenge response did not match. + == ======================================================== What: /sys/bus/thunderbolt/devices/.../boot Date: Jun 2018 diff --git a/Documentation/ABI/testing/sysfs-bus-usb b/Documentation/ABI/testing/sysfs-bus-usb index 614d216dff1d..e449b8374f6a 100644 --- a/Documentation/ABI/testing/sysfs-bus-usb +++ b/Documentation/ABI/testing/sysfs-bus-usb @@ -72,24 +72,27 @@ Description: table at compile time. The format for the device ID is: idVendor idProduct bInterfaceClass RefIdVendor RefIdProduct The vendor ID and device ID fields are required, the - rest is optional. The Ref* tuple can be used to tell the + rest is optional. The `Ref*` tuple can be used to tell the driver to use the same driver_data for the new device as it is used for the reference device. Upon successfully adding an ID, the driver will probe - for the device and attempt to bind to it. For example: - # echo "8086 10f5" > /sys/bus/usb/drivers/foo/new_id + for the device and attempt to bind to it. For example:: + + # echo "8086 10f5" > /sys/bus/usb/drivers/foo/new_id Here add a new device (0458:7045) using driver_data from - an already supported device (0458:704c): - # echo "0458 7045 0 0458 704c" > /sys/bus/usb/drivers/foo/new_id + an already supported device (0458:704c):: + + # echo "0458 7045 0 0458 704c" > /sys/bus/usb/drivers/foo/new_id Reading from this file will list all dynamically added device IDs in the same format, with one entry per - line. For example: - # cat /sys/bus/usb/drivers/foo/new_id - 8086 10f5 - dead beef 06 - f00d cafe + line. For example:: + + # cat /sys/bus/usb/drivers/foo/new_id + 8086 10f5 + dead beef 06 + f00d cafe The list will be truncated at PAGE_SIZE bytes due to sysfs restrictions. @@ -209,6 +212,7 @@ Description: advance, and behaves well according to the specification. This attribute is a bit-field that controls the behavior of a specific port: + - Bit 0 of this field selects the "old" enumeration scheme, as it is considerably faster (it only causes one USB reset instead of 2). @@ -233,10 +237,10 @@ Description: poll() for monitoring changes to this value in user space. Any time this value changes the corresponding hub device will send a - udev event with the following attributes: + udev event with the following attributes:: - OVER_CURRENT_PORT=/sys/bus/usb/devices/.../(hub interface)/portX - OVER_CURRENT_COUNT=[current value of this sysfs attribute] + OVER_CURRENT_PORT=/sys/bus/usb/devices/.../(hub interface)/portX + OVER_CURRENT_COUNT=[current value of this sysfs attribute] What: /sys/bus/usb/devices/.../(hub interface)/portX/usb3_lpm_permit Date: November 2015 diff --git a/Documentation/ABI/testing/sysfs-bus-usb-devices-usbsevseg b/Documentation/ABI/testing/sysfs-bus-usb-devices-usbsevseg index 9ade80f81f96..2f86e4223bfc 100644 --- a/Documentation/ABI/testing/sysfs-bus-usb-devices-usbsevseg +++ b/Documentation/ABI/testing/sysfs-bus-usb-devices-usbsevseg @@ -12,8 +12,11 @@ KernelVersion: 2.6.26 Contact: Harrison Metzger Description: Controls the devices display mode. For a 6 character display the values are + MSB 0x06; LSB 0x3F, and + for an 8 character display the values are + MSB 0x08; LSB 0xFF. What: /sys/bus/usb/.../textmode @@ -37,7 +40,7 @@ KernelVersion: 2.6.26 Contact: Harrison Metzger Description: Controls the decimal places on the device. To set the nth decimal place, give this field - the value of 10 ** n. Assume this field has + the value of ``10 ** n``. Assume this field has the value k and has 1 or more decimal places set, to set the mth place (where m is not already set), - change this fields value to k + 10 ** m. + change this fields value to ``k + 10 ** m``. diff --git a/Documentation/ABI/testing/sysfs-bus-vfio-mdev b/Documentation/ABI/testing/sysfs-bus-vfio-mdev index 452dbe39270e..59fc804265db 100644 --- a/Documentation/ABI/testing/sysfs-bus-vfio-mdev +++ b/Documentation/ABI/testing/sysfs-bus-vfio-mdev @@ -28,8 +28,9 @@ Description: Writing UUID to this file will create mediated device of type for parent device . This is a write-only file. - For example: - # echo "83b8f4f2-509f-382f-3c1e-e6bfe0fa1001" > \ + For example:: + + # echo "83b8f4f2-509f-382f-3c1e-e6bfe0fa1001" > \ /sys/devices/foo/mdev_supported_types/foo-1/create What: /sys/.../mdev_supported_types//devices/ @@ -107,5 +108,6 @@ Description: Writing '1' to this file destroys the mediated device. The vendor driver can fail the remove() callback if that device is active and the vendor driver doesn't support hot unplug. - Example: - # echo 1 > /sys/bus/mdev/devices//remove + Example:: + + # echo 1 > /sys/bus/mdev/devices//remove diff --git a/Documentation/ABI/testing/sysfs-class-cxl b/Documentation/ABI/testing/sysfs-class-cxl index 7970e3713e70..a6f51a104c44 100644 --- a/Documentation/ABI/testing/sysfs-class-cxl +++ b/Documentation/ABI/testing/sysfs-class-cxl @@ -72,11 +72,16 @@ Description: read/write when performing the START_WORK ioctl. Only applicable when running under hashed page table mmu. Possible values: - none: No prefaulting (default) - work_element_descriptor: Treat the work element - descriptor as an effective address and - prefault what it points to. - all: all segments process calling START_WORK maps. + + ======================= ====================================== + none No prefaulting (default) + work_element_descriptor Treat the work element + descriptor as an effective address and + prefault what it points to. + all all segments process calling + START_WORK maps. + ======================= ====================================== + Users: https://github.com/ibm-capi/libcxl What: /sys/class/cxl//reset diff --git a/Documentation/ABI/testing/sysfs-class-led b/Documentation/ABI/testing/sysfs-class-led index 5f67f7ab277b..65e040978f73 100644 --- a/Documentation/ABI/testing/sysfs-class-led +++ b/Documentation/ABI/testing/sysfs-class-led @@ -50,7 +50,7 @@ Description: You can change triggers in a similar manner to the way an IO scheduler is chosen. Trigger specific parameters can appear in /sys/class/leds/ once a given trigger is selected. For - their documentation see sysfs-class-led-trigger-*. + their documentation see `sysfs-class-led-trigger-*`. What: /sys/class/leds//inverted Date: January 2011 diff --git a/Documentation/ABI/testing/sysfs-class-led-driver-el15203000 b/Documentation/ABI/testing/sysfs-class-led-driver-el15203000 index f520ece9b64c..69befe947d7e 100644 --- a/Documentation/ABI/testing/sysfs-class-led-driver-el15203000 +++ b/Documentation/ABI/testing/sysfs-class-led-driver-el15203000 @@ -6,127 +6,132 @@ Description: The LEDs board supports only predefined patterns by firmware for specific LEDs. - Breathing mode for Screen frame light tube: - "0 4000 1 4000" - - ^ - | - Max-| --- - | / \ - | / \ - | / \ / - | / \ / - Min-|- --- - | - 0------4------8--> time (sec) - - Cascade mode for Pipe LED: - "1 800 2 800 4 800 8 800 16 800" - - ^ - | - 0 On -|----+ +----+ +--- - | | | | | - Off-| +-------------------+ +-------------------+ - | - 1 On -| +----+ +----+ - | | | | | - Off |----+ +-------------------+ +------------------ - | - 2 On -| +----+ +----+ - | | | | | - Off-|---------+ +-------------------+ +------------- - | - 3 On -| +----+ +----+ - | | | | | - Off-|--------------+ +-------------------+ +-------- - | - 4 On -| +----+ +----+ - | | | | | - Off-|-------------------+ +-------------------+ +--- - | - 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) - - Inverted cascade mode for Pipe LED: - "30 800 29 800 27 800 23 800 15 800" - - ^ - | - 0 On -| +-------------------+ +-------------------+ - | | | | | - Off-|----+ +----+ +--- - | - 1 On -|----+ +-------------------+ +------------------ - | | | | | - Off | +----+ +----+ - | - 2 On -|---------+ +-------------------+ +------------- - | | | | | - Off-| +----+ +----+ - | - 3 On -|--------------+ +-------------------+ +-------- - | | | | | - Off-| +----+ +----+ - | - 4 On -|-------------------+ +-------------------+ +--- - | | | | | - Off-| +----+ +----+ - | - 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) - - Bounce mode for Pipe LED: - "1 800 2 800 4 800 8 800 16 800 16 800 8 800 4 800 2 800 1 800" - - ^ - | - 0 On -|----+ +-------- - | | | - Off-| +---------------------------------------+ - | - 1 On -| +----+ +----+ - | | | | | - Off |----+ +-----------------------------+ +-------- - | - 2 On -| +----+ +----+ - | | | | | - Off-|---------+ +-------------------+ +------------- - | - 3 On -| +----+ +----+ - | | | | | - Off-|--------------+ +---------+ +------------------ - | - 4 On -| +---------+ - | | | - Off-|-------------------+ +----------------------- - | - 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) - - Inverted bounce mode for Pipe LED: - "30 800 29 800 27 800 23 800 15 800 15 800 23 800 27 800 29 800 30 800" - - ^ - | - 0 On -| +---------------------------------------+ - | | | - Off-|----+ +-------- - | - 1 On -|----+ +-----------------------------+ +-------- - | | | | | - Off | +----+ +----+ - | - 2 On -|---------+ +-------------------+ +------------- - | | | | | - Off-| +----+ +----+ - | - 3 On -|--------------+ +---------+ +------------------ - | | | | | - Off-| +----+ +----+ - | - 4 On -|-------------------+ +----------------------- - | | | - Off-| +---------+ - | - 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) + Breathing mode for Screen frame light tube:: + + "0 4000 1 4000" + + ^ + | + Max-| --- + | / \ + | / \ + | / \ / + | / \ / + Min-|- --- + | + 0------4------8--> time (sec) + + Cascade mode for Pipe LED:: + + "1 800 2 800 4 800 8 800 16 800" + + ^ + | + 0 On -|----+ +----+ +--- + | | | | | + Off-| +-------------------+ +-------------------+ + | + 1 On -| +----+ +----+ + | | | | | + Off |----+ +-------------------+ +------------------ + | + 2 On -| +----+ +----+ + | | | | | + Off-|---------+ +-------------------+ +------------- + | + 3 On -| +----+ +----+ + | | | | | + Off-|--------------+ +-------------------+ +-------- + | + 4 On -| +----+ +----+ + | | | | | + Off-|-------------------+ +-------------------+ +--- + | + 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) + + Inverted cascade mode for Pipe LED:: + + "30 800 29 800 27 800 23 800 15 800" + + ^ + | + 0 On -| +-------------------+ +-------------------+ + | | | | | + Off-|----+ +----+ +--- + | + 1 On -|----+ +-------------------+ +------------------ + | | | | | + Off | +----+ +----+ + | + 2 On -|---------+ +-------------------+ +------------- + | | | | | + Off-| +----+ +----+ + | + 3 On -|--------------+ +-------------------+ +-------- + | | | | | + Off-| +----+ +----+ + | + 4 On -|-------------------+ +-------------------+ +--- + | | | | | + Off-| +----+ +----+ + | + 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) + + Bounce mode for Pipe LED:: + + "1 800 2 800 4 800 8 800 16 800 16 800 8 800 4 800 2 800 1 800" + + ^ + | + 0 On -|----+ +-------- + | | | + Off-| +---------------------------------------+ + | + 1 On -| +----+ +----+ + | | | | | + Off |----+ +-----------------------------+ +-------- + | + 2 On -| +----+ +----+ + | | | | | + Off-|---------+ +-------------------+ +------------- + | + 3 On -| +----+ +----+ + | | | | | + Off-|--------------+ +---------+ +------------------ + | + 4 On -| +---------+ + | | | + Off-|-------------------+ +----------------------- + | + 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) + + Inverted bounce mode for Pipe LED:: + + "30 800 29 800 27 800 23 800 15 800 15 800 23 800 27 800 29 800 30 800" + + ^ + | + 0 On -| +---------------------------------------+ + | | | + Off-|----+ +-------- + | + 1 On -|----+ +-----------------------------+ +-------- + | | | | | + Off | +----+ +----+ + | + 2 On -|---------+ +-------------------+ +------------- + | | | | | + Off-| +----+ +----+ + | + 3 On -|--------------+ +---------+ +------------------ + | | | | | + Off-| +----+ +----+ + | + 4 On -|-------------------+ +----------------------- + | | | + Off-| +---------+ + | + 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) What: /sys/class/leds//repeat Date: September 2019 diff --git a/Documentation/ABI/testing/sysfs-class-led-driver-sc27xx b/Documentation/ABI/testing/sysfs-class-led-driver-sc27xx index 45b1e605d355..215482379580 100644 --- a/Documentation/ABI/testing/sysfs-class-led-driver-sc27xx +++ b/Documentation/ABI/testing/sysfs-class-led-driver-sc27xx @@ -12,8 +12,8 @@ Description: format, we should set brightness as 0 for rise stage, fall stage and low stage. - Min stage duration: 125 ms - Max stage duration: 31875 ms + - Min stage duration: 125 ms + - Max stage duration: 31875 ms Since the stage duration step is 125 ms, the duration should be a multiplier of 125, like 125ms, 250ms, 375ms, 500ms ... 31875ms. diff --git a/Documentation/ABI/testing/sysfs-class-mic b/Documentation/ABI/testing/sysfs-class-mic index 6ef682603179..bd0e780c3760 100644 --- a/Documentation/ABI/testing/sysfs-class-mic +++ b/Documentation/ABI/testing/sysfs-class-mic @@ -41,24 +41,33 @@ Description: When read, this entry provides the current state of an Intel MIC device in the context of the card OS. Possible values that will be read are: - "ready" - The MIC device is ready to boot the card OS. On - reading this entry after an OSPM resume, a "boot" has to be - written to this entry if the card was previously shutdown - during OSPM suspend. - "booting" - The MIC device has initiated booting a card OS. - "online" - The MIC device has completed boot and is online - "shutting_down" - The card OS is shutting down. - "resetting" - A reset has been initiated for the MIC device - "reset_failed" - The MIC device has failed to reset. + + + =============== =============================================== + "ready" The MIC device is ready to boot the card OS. + On reading this entry after an OSPM resume, + a "boot" has to be written to this entry if + the card was previously shutdown during OSPM + suspend. + "booting" The MIC device has initiated booting a card OS. + "online" The MIC device has completed boot and is online + "shutting_down" The card OS is shutting down. + "resetting" A reset has been initiated for the MIC device + "reset_failed" The MIC device has failed to reset. + =============== =============================================== When written, this sysfs entry triggers different state change operations depending upon the current state of the card OS. Acceptable values are: - "boot" - Boot the card OS image specified by the combination - of firmware, ramdisk, cmdline and bootmode - sysfs entries. - "reset" - Initiates device reset. - "shutdown" - Initiates card OS shutdown. + + + ========== =================================================== + "boot" Boot the card OS image specified by the combination + of firmware, ramdisk, cmdline and bootmode + sysfs entries. + "reset" Initiates device reset. + "shutdown" Initiates card OS shutdown. + ========== =================================================== What: /sys/class/mic/mic(x)/shutdown_status Date: October 2013 @@ -69,12 +78,15 @@ Description: OS can shutdown because of various reasons. When read, this entry provides the status on why the card OS was shutdown. Possible values are: - "nop" - shutdown status is not applicable, when the card OS is - "online" - "crashed" - Shutdown because of a HW or SW crash. - "halted" - Shutdown because of a halt command. - "poweroff" - Shutdown because of a poweroff command. - "restart" - Shutdown because of a restart command. + + ========== =================================================== + "nop" shutdown status is not applicable, when the card OS + is "online" + "crashed" Shutdown because of a HW or SW crash. + "halted" Shutdown because of a halt command. + "poweroff" Shutdown because of a poweroff command. + "restart" Shutdown because of a restart command. + ========== =================================================== What: /sys/class/mic/mic(x)/cmdline Date: October 2013 diff --git a/Documentation/ABI/testing/sysfs-class-ocxl b/Documentation/ABI/testing/sysfs-class-ocxl index ae1276efa45a..bf33f4fda58f 100644 --- a/Documentation/ABI/testing/sysfs-class-ocxl +++ b/Documentation/ABI/testing/sysfs-class-ocxl @@ -11,8 +11,11 @@ Contact: linuxppc-dev@lists.ozlabs.org Description: read only Number of contexts for the AFU, in the format / where: + + ==== =============================================== n: number of currently active contexts, for debug max: maximum number of contexts supported by the AFU + ==== =============================================== What: /sys/class/ocxl//pp_mmio_size Date: January 2018 diff --git a/Documentation/ABI/testing/sysfs-class-power b/Documentation/ABI/testing/sysfs-class-power index dbccb2fcd7ce..d4319a04c302 100644 --- a/Documentation/ABI/testing/sysfs-class-power +++ b/Documentation/ABI/testing/sysfs-class-power @@ -1,4 +1,4 @@ -===== General Properties ===== +**General Properties** What: /sys/class/power_supply//manufacturer Date: May 2007 @@ -72,6 +72,7 @@ Description: critically low). Access: Read, Write + Valid values: 0 - 100 (percent) What: /sys/class/power_supply//capacity_error_margin @@ -96,7 +97,9 @@ Description: Coarse representation of battery capacity. Access: Read - Valid values: "Unknown", "Critical", "Low", "Normal", "High", + + Valid values: + "Unknown", "Critical", "Low", "Normal", "High", "Full" What: /sys/class/power_supply//current_avg @@ -139,6 +142,7 @@ Description: throttling for thermal cooling or improving battery health. Access: Read, Write + Valid values: Represented in microamps What: /sys/class/power_supply//charge_control_limit_max @@ -148,6 +152,7 @@ Description: Maximum legal value for the charge_control_limit property. Access: Read + Valid values: Represented in microamps What: /sys/class/power_supply//charge_control_start_threshold @@ -168,6 +173,7 @@ Description: stop. Access: Read, Write + Valid values: 0 - 100 (percent) What: /sys/class/power_supply//charge_type @@ -183,7 +189,9 @@ Description: different algorithm. Access: Read, Write - Valid values: "Unknown", "N/A", "Trickle", "Fast", "Standard", + + Valid values: + "Unknown", "N/A", "Trickle", "Fast", "Standard", "Adaptive", "Custom" What: /sys/class/power_supply//charge_term_current @@ -194,6 +202,7 @@ Description: when the battery is considered full and charging should end. Access: Read + Valid values: Represented in microamps What: /sys/class/power_supply//health @@ -204,7 +213,9 @@ Description: functionality. Access: Read - Valid values: "Unknown", "Good", "Overheat", "Dead", + + Valid values: + "Unknown", "Good", "Overheat", "Dead", "Over voltage", "Unspecified failure", "Cold", "Watchdog timer expire", "Safety timer expire", "Over current", "Calibration required", "Warm", @@ -218,6 +229,7 @@ Description: for a battery charge cycle. Access: Read + Valid values: Represented in microamps What: /sys/class/power_supply//present @@ -227,9 +239,13 @@ Description: Reports whether a battery is present or not in the system. Access: Read + Valid values: + + == ======= 0: Absent 1: Present + == ======= What: /sys/class/power_supply//status Date: May 2007 @@ -240,7 +256,9 @@ Description: used to enable/disable charging to the battery. Access: Read, Write - Valid values: "Unknown", "Charging", "Discharging", + + Valid values: + "Unknown", "Charging", "Discharging", "Not charging", "Full" What: /sys/class/power_supply//technology @@ -250,7 +268,9 @@ Description: Describes the battery technology supported by the supply. Access: Read - Valid values: "Unknown", "NiMH", "Li-ion", "Li-poly", "LiFe", + + Valid values: + "Unknown", "NiMH", "Li-ion", "Li-poly", "LiFe", "NiCd", "LiMn" What: /sys/class/power_supply//temp @@ -260,6 +280,7 @@ Description: Reports the current TBAT battery temperature reading. Access: Read + Valid values: Represented in 1/10 Degrees Celsius What: /sys/class/power_supply//temp_alert_max @@ -274,6 +295,7 @@ Description: critically high, and charging has stopped). Access: Read + Valid values: Represented in 1/10 Degrees Celsius What: /sys/class/power_supply//temp_alert_min @@ -289,6 +311,7 @@ Description: remedy the situation). Access: Read + Valid values: Represented in 1/10 Degrees Celsius What: /sys/class/power_supply//temp_max @@ -299,6 +322,7 @@ Description: charging. Access: Read + Valid values: Represented in 1/10 Degrees Celsius What: /sys/class/power_supply//temp_min @@ -309,6 +333,7 @@ Description: charging. Access: Read + Valid values: Represented in 1/10 Degrees Celsius What: /sys/class/power_supply//voltage_avg, @@ -320,6 +345,7 @@ Description: which they average readings to smooth out the reported value. Access: Read + Valid values: Represented in microvolts What: /sys/class/power_supply//voltage_max, @@ -330,6 +356,7 @@ Description: during charging. Access: Read + Valid values: Represented in microvolts What: /sys/class/power_supply//voltage_min, @@ -340,6 +367,7 @@ Description: during discharging. Access: Read + Valid values: Represented in microvolts What: /sys/class/power_supply//voltage_now, @@ -350,9 +378,10 @@ Description: This value is not averaged/smoothed. Access: Read + Valid values: Represented in microvolts -===== USB Properties ===== +**USB Properties** What: /sys/class/power_supply//current_avg Date: May 2007 @@ -363,6 +392,7 @@ Description: average readings to smooth out the reported value. Access: Read + Valid values: Represented in microamps @@ -373,6 +403,7 @@ Description: Reports the maximum IBUS current the supply can support. Access: Read + Valid values: Represented in microamps What: /sys/class/power_supply//current_now @@ -385,6 +416,7 @@ Description: within the reported min/max range. Access: Read, Write + Valid values: Represented in microamps What: /sys/class/power_supply//input_current_limit @@ -399,6 +431,7 @@ Description: solved using power limit use input_current_limit. Access: Read, Write + Valid values: Represented in microamps What: /sys/class/power_supply//input_voltage_limit @@ -441,10 +474,14 @@ Description: USB supply so voltage and current can be controlled). Access: Read, Write + Valid values: + + == ================================================== 0: Offline 1: Online Fixed - Fixed Voltage Supply 2: Online Programmable - Programmable Voltage Supply + == ================================================== What: /sys/class/power_supply//temp Date: May 2007 @@ -455,6 +492,7 @@ Description: TJUNC temperature of an IC) Access: Read + Valid values: Represented in 1/10 Degrees Celsius What: /sys/class/power_supply//temp_alert_max @@ -470,6 +508,7 @@ Description: remedy the situation). Access: Read + Valid values: Represented in 1/10 Degrees Celsius What: /sys/class/power_supply//temp_alert_min @@ -485,6 +524,7 @@ Description: accordingly to remedy the situation). Access: Read + Valid values: Represented in 1/10 Degrees Celsius What: /sys/class/power_supply//temp_max @@ -494,6 +534,7 @@ Description: Reports the maximum allowed supply temperature for operation. Access: Read + Valid values: Represented in 1/10 Degrees Celsius What: /sys/class/power_supply//temp_min @@ -503,6 +544,7 @@ Description: Reports the mainimum allowed supply temperature for operation. Access: Read + Valid values: Represented in 1/10 Degrees Celsius What: /sys/class/power_supply//usb_type @@ -514,7 +556,9 @@ Description: is attached. Access: Read-Only - Valid values: "Unknown", "SDP", "DCP", "CDP", "ACA", "C", "PD", + + Valid values: + "Unknown", "SDP", "DCP", "CDP", "ACA", "C", "PD", "PD_DRP", "PD_PPS", "BrickID" What: /sys/class/power_supply//voltage_max @@ -524,6 +568,7 @@ Description: Reports the maximum VBUS voltage the supply can support. Access: Read + Valid values: Represented in microvolts What: /sys/class/power_supply//voltage_min @@ -533,6 +578,7 @@ Description: Reports the minimum VBUS voltage the supply can support. Access: Read + Valid values: Represented in microvolts What: /sys/class/power_supply//voltage_now @@ -545,9 +591,10 @@ Description: within the reported min/max range. Access: Read, Write + Valid values: Represented in microvolts -===== Device Specific Properties ===== +**Device Specific Properties** What: /sys/class/power/ds2760-battery.*/charge_now Date: May 2010 @@ -581,6 +628,7 @@ Description: will drop to 0 A) and will trigger interrupt. Valid values: + - 5, 6 or 7 (hours), - 0: disabled. @@ -595,6 +643,7 @@ Description: will drop to 0 A) and will trigger interrupt. Valid values: + - 4 - 16 (hours), step by 2 (rounded down) - 0: disabled. @@ -609,6 +658,7 @@ Description: interrupt and start top-off charging mode. Valid values: + - 100000 - 200000 (microamps), step by 25000 (rounded down) - 200000 - 350000 (microamps), step by 50000 (rounded down) - 0: disabled. @@ -624,6 +674,7 @@ Description: will drop to 0 A) and will trigger interrupt. Valid values: + - 0 - 70 (minutes), step by 10 (rounded down) What: /sys/class/power_supply/bq24257-charger/ovp_voltage @@ -637,6 +688,7 @@ Description: device datasheet for details. Valid values: + - 6000000, 6500000, 7000000, 8000000, 9000000, 9500000, 10000000, 10500000 (all uV) @@ -652,6 +704,7 @@ Description: lower than the set value. See device datasheet for details. Valid values: + - 4200000, 4280000, 4360000, 4440000, 4520000, 4600000, 4680000, 4760000 (all uV) @@ -666,6 +719,7 @@ Description: the charger operates normally. See device datasheet for details. Valid values: + - 1: enabled - 0: disabled @@ -681,6 +735,7 @@ Description: from the system. See device datasheet for details. Valid values: + - 1: enabled - 0: disabled diff --git a/Documentation/ABI/testing/sysfs-class-power-twl4030 b/Documentation/ABI/testing/sysfs-class-power-twl4030 index b4fd32d210c5..7ac36dba87bc 100644 --- a/Documentation/ABI/testing/sysfs-class-power-twl4030 +++ b/Documentation/ABI/testing/sysfs-class-power-twl4030 @@ -4,18 +4,20 @@ Description: Writing to this can disable charging. Possible values are: - "auto" - draw power as appropriate for detected - power source and battery status. - "off" - do not draw any power. - "continuous" - - activate mode described as "linear" in - TWL data sheets. This uses whatever - current is available and doesn't switch off - when voltage drops. - This is useful for unstable power sources - such as bicycle dynamo, but care should - be taken that battery is not over-charged. + ============= =========================================== + "auto" draw power as appropriate for detected + power source and battery status. + "off" do not draw any power. + "continuous" activate mode described as "linear" in + TWL data sheets. This uses whatever + current is available and doesn't switch off + when voltage drops. + + This is useful for unstable power sources + such as bicycle dynamo, but care should + be taken that battery is not over-charged. + ============= =========================================== What: /sys/class/power_supply/twl4030_ac/mode Description: @@ -23,6 +25,9 @@ Description: Writing to this can disable charging. Possible values are: - "auto" - draw power as appropriate for detected - power source and battery status. - "off" - do not draw any power. + + ====== =========================================== + "auto" draw power as appropriate for detected + power source and battery status. + "off" do not draw any power. + ====== =========================================== diff --git a/Documentation/ABI/testing/sysfs-class-rc b/Documentation/ABI/testing/sysfs-class-rc index 6c0d6c8cb911..9c8ff7910858 100644 --- a/Documentation/ABI/testing/sysfs-class-rc +++ b/Documentation/ABI/testing/sysfs-class-rc @@ -21,15 +21,22 @@ KernelVersion: 2.6.36 Contact: Mauro Carvalho Chehab Description: Reading this file returns a list of available protocols, - something like: + something like:: + "rc5 [rc6] nec jvc [sony]" + Enabled protocols are shown in [] brackets. + Writing "+proto" will add a protocol to the list of enabled protocols. + Writing "-proto" will remove a protocol from the list of enabled protocols. + Writing "proto" will enable only "proto". + Writing "none" will disable all protocols. + Write fails with EINVAL if an invalid protocol combination or unknown protocol name is used. @@ -39,11 +46,13 @@ KernelVersion: 3.15 Contact: Mauro Carvalho Chehab Description: Sets the scancode filter expected value. + Use in combination with /sys/class/rc/rcN/filter_mask to set the expected value of the bits set in the filter mask. If the hardware supports it then scancodes which do not match the filter will be ignored. Otherwise the write will fail with an error. + This value may be reset to 0 if the current protocol is altered. What: /sys/class/rc/rcN/filter_mask @@ -56,9 +65,11 @@ Description: of the scancode which should be compared against the expected value. A value of 0 disables the filter to allow all valid scancodes to be processed. + If the hardware supports it then scancodes which do not match the filter will be ignored. Otherwise the write will fail with an error. + This value may be reset to 0 if the current protocol is altered. What: /sys/class/rc/rcN/wakeup_protocols @@ -67,15 +78,22 @@ KernelVersion: 4.11 Contact: Mauro Carvalho Chehab Description: Reading this file returns a list of available protocols to use - for the wakeup filter, something like: + for the wakeup filter, something like:: + "rc-5 nec nec-x rc-6-0 rc-6-6a-24 [rc-6-6a-32] rc-6-mce" + Note that protocol variants are listed, so "nec", "sony", "rc-5", "rc-6" have their different bit length encodings listed if available. + The enabled wakeup protocol is shown in [] brackets. + Only one protocol can be selected at a time. + Writing "proto" will use "proto" for wakeup events. + Writing "none" will disable wakeup. + Write fails with EINVAL if an invalid protocol combination or unknown protocol name is used, or if wakeup is not supported by the hardware. @@ -86,13 +104,17 @@ KernelVersion: 3.15 Contact: Mauro Carvalho Chehab Description: Sets the scancode wakeup filter expected value. + Use in combination with /sys/class/rc/rcN/wakeup_filter_mask to set the expected value of the bits set in the wakeup filter mask to trigger a system wake event. + If the hardware supports it and wakeup_filter_mask is not 0 then scancodes which match the filter will wake the system from e.g. suspend to RAM or power off. + Otherwise the write will fail with an error. + This value may be reset to 0 if the wakeup protocol is altered. What: /sys/class/rc/rcN/wakeup_filter_mask @@ -101,11 +123,15 @@ KernelVersion: 3.15 Contact: Mauro Carvalho Chehab Description: Sets the scancode wakeup filter mask of bits to compare. + Use in combination with /sys/class/rc/rcN/wakeup_filter to set the bits of the scancode which should be compared against the expected value to trigger a system wake event. + If the hardware supports it and wakeup_filter_mask is not 0 then scancodes which match the filter will wake the system from e.g. suspend to RAM or power off. + Otherwise the write will fail with an error. + This value may be reset to 0 if the wakeup protocol is altered. diff --git a/Documentation/ABI/testing/sysfs-class-scsi_host b/Documentation/ABI/testing/sysfs-class-scsi_host index bafc59fd7b69..7c98d8f43c45 100644 --- a/Documentation/ABI/testing/sysfs-class-scsi_host +++ b/Documentation/ABI/testing/sysfs-class-scsi_host @@ -56,8 +56,9 @@ Description: management) on top, which makes it match the Windows IRST (Intel Rapid Storage Technology) driver settings. This setting is also close to min_power, except that: + a) It does not use host-initiated slumber mode, but it does - allow device-initiated slumber + allow device-initiated slumber b) It does not enable low power device sleep mode (DevSlp). What: /sys/class/scsi_host/hostX/em_message @@ -70,8 +71,8 @@ Description: protocol, writes and reads correspond to the LED message format as defined in the AHCI spec. - The user must turn sw_activity (under /sys/block/*/device/) OFF - it they wish to control the activity LED via the em_message + The user must turn sw_activity (under `/sys/block/*/device/`) + OFF it they wish to control the activity LED via the em_message file. em_message_type: (RO) Displays the current enclosure management diff --git a/Documentation/ABI/testing/sysfs-class-typec b/Documentation/ABI/testing/sysfs-class-typec index b834671522d6..b7794e02ad20 100644 --- a/Documentation/ABI/testing/sysfs-class-typec +++ b/Documentation/ABI/testing/sysfs-class-typec @@ -40,10 +40,13 @@ Description: attribute will not return until the operation has finished. Valid values: - - source (The port will behave as source only DFP port) - - sink (The port will behave as sink only UFP port) - - dual (The port will behave as dual-role-data and + + ====== ============================================== + source (The port will behave as source only DFP port) + sink (The port will behave as sink only UFP port) + dual (The port will behave as dual-role-data and dual-role-power port) + ====== ============================================== What: /sys/class/typec//vconn_source Date: April 2017 @@ -59,6 +62,7 @@ Description: generates uevent KOBJ_CHANGE. Valid values: + - "no" when the port is not the VCONN Source - "yes" when the port is the VCONN Source @@ -72,6 +76,7 @@ Description: power operation mode should show "usb_power_delivery". Valid values: + - default - 1.5A - 3.0A @@ -191,6 +196,7 @@ Date: April 2017 Contact: Heikki Krogerus Description: Shows type of the plug on the cable: + - type-a - Standard A - type-b - Standard B - type-c diff --git a/Documentation/ABI/testing/sysfs-devices-platform-ACPI-TAD b/Documentation/ABI/testing/sysfs-devices-platform-ACPI-TAD index 7e43cdce9a52..f7b360a61b21 100644 --- a/Documentation/ABI/testing/sysfs-devices-platform-ACPI-TAD +++ b/Documentation/ABI/testing/sysfs-devices-platform-ACPI-TAD @@ -7,6 +7,7 @@ Description: (RO) Hexadecimal bitmask of the TAD attributes are reported by the platform firmware (see ACPI 6.2, section 9.18.2): + ======= ====================================================== BIT(0): AC wakeup implemented if set BIT(1): DC wakeup implemented if set BIT(2): Get/set real time features implemented if set @@ -16,6 +17,7 @@ Description: BIT(6): The AC timer wakes up from S5 if set BIT(7): The DC timer wakes up from S4 if set BIT(8): The DC timer wakes up from S5 if set + ======= ====================================================== The other bits are reserved. @@ -62,9 +64,11 @@ Description: timer status with the following meaning of bits (see ACPI 6.2, Section 9.18.5): + ======= ====================================================== Bit(0): The timer has expired if set. Bit(1): The timer has woken up the system from a sleep state (S3 or S4/S5 if supported) if set. + ======= ====================================================== The other bits are reserved. diff --git a/Documentation/ABI/testing/sysfs-devices-platform-docg3 b/Documentation/ABI/testing/sysfs-devices-platform-docg3 index 8aa36716882f..378c42694bfb 100644 --- a/Documentation/ABI/testing/sysfs-devices-platform-docg3 +++ b/Documentation/ABI/testing/sysfs-devices-platform-docg3 @@ -9,8 +9,10 @@ Description: The protection has information embedded whether it blocks reads, writes or both. The result is: - 0 -> the DPS is not keylocked - 1 -> the DPS is keylocked + + - 0 -> the DPS is not keylocked + - 1 -> the DPS is keylocked + Users: None identified so far. What: /sys/devices/platform/docg3/f[0-3]_dps[01]_protection_key @@ -27,8 +29,12 @@ Description: Entering the correct value toggle the lock, and can be observed through f[0-3]_dps[01]_is_keylocked. Possible values are: + - 8 bytes + Typical values are: + - "00000000" - "12345678" + Users: None identified so far. diff --git a/Documentation/ABI/testing/sysfs-devices-platform-sh_mobile_lcdc_fb b/Documentation/ABI/testing/sysfs-devices-platform-sh_mobile_lcdc_fb index 2107082426da..e45ac2e865d5 100644 --- a/Documentation/ABI/testing/sysfs-devices-platform-sh_mobile_lcdc_fb +++ b/Documentation/ABI/testing/sysfs-devices-platform-sh_mobile_lcdc_fb @@ -17,10 +17,10 @@ Description: to overlay planes. Selects the composition mode for the overlay. Possible values - are + are: - 0 - Alpha Blending - 1 - ROP3 + - 0 - Alpha Blending + - 1 - ROP3 What: /sys/devices/platform/sh_mobile_lcdc_fb.[0-3]/graphics/fb[0-9]/ovl_position Date: May 2012 @@ -30,7 +30,7 @@ Description: to overlay planes. Stores the x,y overlay position on the display in pixels. The - position format is `[0-9]+,[0-9]+'. + position format is `[0-9]+,[0-9]+`. What: /sys/devices/platform/sh_mobile_lcdc_fb.[0-3]/graphics/fb[0-9]/ovl_rop3 Date: May 2012 diff --git a/Documentation/ABI/testing/sysfs-devices-system-cpu b/Documentation/ABI/testing/sysfs-devices-system-cpu index b555df825447..274c337ec6a9 100644 --- a/Documentation/ABI/testing/sysfs-devices-system-cpu +++ b/Documentation/ABI/testing/sysfs-devices-system-cpu @@ -151,23 +151,28 @@ Description: The processor idle states which are available for use have the following attributes: - name: (RO) Name of the idle state (string). + ======== ==== ================================================= + name: (RO) Name of the idle state (string). latency: (RO) The latency to exit out of this idle state (in - microseconds). + microseconds). - power: (RO) The power consumed while in this idle state (in - milliwatts). + power: (RO) The power consumed while in this idle state (in + milliwatts). - time: (RO) The total time spent in this idle state (in microseconds). + time: (RO) The total time spent in this idle state + (in microseconds). - usage: (RO) Number of times this state was entered (a count). + usage: (RO) Number of times this state was entered (a count). - above: (RO) Number of times this state was entered, but the - observed CPU idle duration was too short for it (a count). + above: (RO) Number of times this state was entered, but the + observed CPU idle duration was too short for it + (a count). - below: (RO) Number of times this state was entered, but the - observed CPU idle duration was too long for it (a count). + below: (RO) Number of times this state was entered, but the + observed CPU idle duration was too long for it + (a count). + ======== ==== ================================================= What: /sys/devices/system/cpu/cpuX/cpuidle/stateN/desc Date: February 2008 @@ -290,6 +295,7 @@ Description: Processor frequency boosting control This switch controls the boost setting for the whole system. Boosting allows the CPU and the firmware to run at a frequency beyound it's nominal limit. + More details can be found in Documentation/admin-guide/pm/cpufreq.rst @@ -337,43 +343,57 @@ Contact: Sudeep Holla Description: Parameters for the CPU cache attributes allocation_policy: - - WriteAllocate: allocate a memory location to a cache line - on a cache miss because of a write - - ReadAllocate: allocate a memory location to a cache line + - WriteAllocate: + allocate a memory location to a cache line + on a cache miss because of a write + - ReadAllocate: + allocate a memory location to a cache line on a cache miss because of a read - - ReadWriteAllocate: both writeallocate and readallocate + - ReadWriteAllocate: + both writeallocate and readallocate - attributes: LEGACY used only on IA64 and is same as write_policy + attributes: + LEGACY used only on IA64 and is same as write_policy - coherency_line_size: the minimum amount of data in bytes that gets + coherency_line_size: + the minimum amount of data in bytes that gets transferred from memory to cache - level: the cache hierarchy in the multi-level cache configuration + level: + the cache hierarchy in the multi-level cache configuration - number_of_sets: total number of sets in the cache, a set is a + number_of_sets: + total number of sets in the cache, a set is a collection of cache lines with the same cache index - physical_line_partition: number of physical cache line per cache tag + physical_line_partition: + number of physical cache line per cache tag - shared_cpu_list: the list of logical cpus sharing the cache + shared_cpu_list: + the list of logical cpus sharing the cache - shared_cpu_map: logical cpu mask containing the list of cpus sharing + shared_cpu_map: + logical cpu mask containing the list of cpus sharing the cache - size: the total cache size in kB + size: + the total cache size in kB type: - Instruction: cache that only holds instructions - Data: cache that only caches data - Unified: cache that holds both data and instructions - ways_of_associativity: degree of freedom in placing a particular block - of memory in the cache + ways_of_associativity: + degree of freedom in placing a particular block + of memory in the cache write_policy: - - WriteThrough: data is written to both the cache line + - WriteThrough: + data is written to both the cache line and to the block in the lower-level memory - - WriteBack: data is written only to the cache line and + - WriteBack: + data is written only to the cache line and the modified cache line is written to main memory only when it is replaced @@ -414,30 +434,30 @@ Description: POWERNV CPUFreq driver's frequency throttle stats directory and throttle attributes exported in the 'throttle_stats' directory: - turbo_stat : This file gives the total number of times the max - frequency is throttled to lower frequency in turbo (at and above - nominal frequency) range of frequencies. + frequency is throttled to lower frequency in turbo (at and above + nominal frequency) range of frequencies. - sub_turbo_stat : This file gives the total number of times the - max frequency is throttled to lower frequency in sub-turbo(below - nominal frequency) range of frequencies. + max frequency is throttled to lower frequency in sub-turbo(below + nominal frequency) range of frequencies. - unthrottle : This file gives the total number of times the max - frequency is unthrottled after being throttled. + frequency is unthrottled after being throttled. - powercap : This file gives the total number of times the max - frequency is throttled due to 'Power Capping'. + frequency is throttled due to 'Power Capping'. - overtemp : This file gives the total number of times the max - frequency is throttled due to 'CPU Over Temperature'. + frequency is throttled due to 'CPU Over Temperature'. - supply_fault : This file gives the total number of times the - max frequency is throttled due to 'Power Supply Failure'. + max frequency is throttled due to 'Power Supply Failure'. - overcurrent : This file gives the total number of times the - max frequency is throttled due to 'Overcurrent'. + max frequency is throttled due to 'Overcurrent'. - occ_reset : This file gives the total number of times the max - frequency is throttled due to 'OCC Reset'. + frequency is throttled due to 'OCC Reset'. The sysfs attributes representing different throttle reasons like powercap, overtemp, supply_fault, overcurrent and occ_reset map to @@ -469,8 +489,9 @@ What: /sys/devices/system/cpu/cpuX/regs/ Date: June 2016 Contact: Linux ARM Kernel Mailing list Description: AArch64 CPU registers + 'identification' directory exposes the CPU ID registers for - identifying model and revision of the CPU. + identifying model and revision of the CPU. What: /sys/devices/system/cpu/cpu#/cpu_capacity Date: December 2016 @@ -497,9 +518,11 @@ Description: Information about CPU vulnerabilities vulnerabilities. The output of those files reflects the state of the CPUs in the system. Possible output values: + ================ ============================================== "Not affected" CPU is not affected by the vulnerability "Vulnerable" CPU is affected and no mitigation in effect "Mitigation: $M" CPU is affected and mitigation $M is in effect + ================ ============================================== See also: Documentation/admin-guide/hw-vuln/index.rst @@ -515,12 +538,14 @@ Description: Control Symetric Multi Threading (SMT) control: Read/write interface to control SMT. Possible values: + ================ ========================================= "on" SMT is enabled "off" SMT is disabled "forceoff" SMT is force disabled. Cannot be changed. "notsupported" SMT is not supported by the CPU "notimplemented" SMT runtime toggling is not implemented for the architecture + ================ ========================================= If control status is "forceoff" or "notsupported" writes are rejected. diff --git a/Documentation/ABI/testing/sysfs-devices-system-ibm-rtl b/Documentation/ABI/testing/sysfs-devices-system-ibm-rtl index 470def06ab0a..1a8ee26e92ae 100644 --- a/Documentation/ABI/testing/sysfs-devices-system-ibm-rtl +++ b/Documentation/ABI/testing/sysfs-devices-system-ibm-rtl @@ -5,8 +5,10 @@ Contact: Vernon Mauery Description: The state file allows a means by which to change in and out of Premium Real-Time Mode (PRTM), as well as the ability to query the current state. - 0 => PRTM off - 1 => PRTM enabled + + - 0 => PRTM off + - 1 => PRTM enabled + Users: The ibm-prtm userspace daemon uses this interface. diff --git a/Documentation/ABI/testing/sysfs-driver-bd9571mwv-regulator b/Documentation/ABI/testing/sysfs-driver-bd9571mwv-regulator index 4d63a7904b94..42214b4ff14a 100644 --- a/Documentation/ABI/testing/sysfs-driver-bd9571mwv-regulator +++ b/Documentation/ABI/testing/sysfs-driver-bd9571mwv-regulator @@ -6,11 +6,13 @@ Description: Read/write the current state of DDR Backup Mode, which controls if DDR power rails will be kept powered during system suspend. ("on"/"1" = enabled, "off"/"0" = disabled). Two types of power switches (or control signals) can be used: + A. With a momentary power switch (or pulse signal), DDR Backup Mode is enabled by default when available, as the PMIC will be configured only during system suspend. B. With a toggle power switch (or level signal), the following steps must be followed exactly: + 1. Configure PMIC for backup mode, to change the role of the accessory power switch from a power switch to a wake-up switch, @@ -20,8 +22,10 @@ Description: Read/write the current state of DDR Backup Mode, which controls 3. Suspend system, 4. Switch accessory power switch on, to resume the system. + DDR Backup Mode must be explicitly enabled by the user, to invoke step 1. + See also Documentation/devicetree/bindings/mfd/bd9571mwv.txt. Users: User space applications for embedded boards equipped with a BD9571MWV PMIC. diff --git a/Documentation/ABI/testing/sysfs-driver-genwqe b/Documentation/ABI/testing/sysfs-driver-genwqe index 64ac6d567c4b..69d855dc4c47 100644 --- a/Documentation/ABI/testing/sysfs-driver-genwqe +++ b/Documentation/ABI/testing/sysfs-driver-genwqe @@ -29,8 +29,12 @@ What: /sys/class/genwqe/genwqe_card/reload_bitstream Date: May 2014 Contact: klebers@linux.vnet.ibm.com Description: Interface to trigger a PCIe card reset to reload the bitstream. + + :: + sudo sh -c 'echo 1 > \ /sys/class/genwqe/genwqe0_card/reload_bitstream' + If successfully, the card will come back with the bitstream set on 'next_bitstream'. @@ -64,8 +68,11 @@ Description: Base clock frequency of the card. What: /sys/class/genwqe/genwqe_card/device/sriov_numvfs Date: Oct 2013 Contact: haver@linux.vnet.ibm.com -Description: Enable VFs (1..15): +Description: Enable VFs (1..15):: + sudo sh -c 'echo 15 > \ /sys/bus/pci/devices/0000\:1b\:00.0/sriov_numvfs' - Disable VFs: + + Disable VFs:: + Write a 0 into the same sysfs entry. diff --git a/Documentation/ABI/testing/sysfs-driver-hid-logitech-lg4ff b/Documentation/ABI/testing/sysfs-driver-hid-logitech-lg4ff index 305dffd229a8..de07be314efc 100644 --- a/Documentation/ABI/testing/sysfs-driver-hid-logitech-lg4ff +++ b/Documentation/ABI/testing/sysfs-driver-hid-logitech-lg4ff @@ -12,7 +12,9 @@ KernelVersion: 4.1 Contact: Michal Malý Description: Displays a set of alternate modes supported by a wheel. Each mode is listed as follows: + Tag: Mode Name + Currently active mode is marked with an asterisk. List also contains an abstract item "native" which always denotes the native mode of the wheel. Echoing the mode tag switches the @@ -24,24 +26,30 @@ Description: Displays a set of alternate modes supported by a wheel. Each This entry is not created for devices that have only one mode. Currently supported mode switches: - Driving Force Pro: + + Driving Force Pro:: + DF-EX --> DFP - G25: + G25:: + DF-EX --> DFP --> G25 - G27: + G27:: + DF-EX <*> DFP <-> G25 <-> G27 DF-EX <*--------> G25 <-> G27 DF-EX <*----------------> G27 - G29: + G29:: + DF-EX <*> DFP <-> G25 <-> G27 <-> G29 DF-EX <*--------> G25 <-> G27 <-> G29 DF-EX <*----------------> G27 <-> G29 DF-EX <*------------------------> G29 - DFGT: + DFGT:: + DF-EX <*> DFP <-> DFGT DF-EX <*--------> DFGT diff --git a/Documentation/ABI/testing/sysfs-driver-hid-wiimote b/Documentation/ABI/testing/sysfs-driver-hid-wiimote index 39dfa5cb1cc5..cd7b82a5c27d 100644 --- a/Documentation/ABI/testing/sysfs-driver-hid-wiimote +++ b/Documentation/ABI/testing/sysfs-driver-hid-wiimote @@ -39,9 +39,13 @@ Description: While a device is initialized by the wiimote driver, we perform Other strings for each device-type are available and may be added if new device-specific detections are added. Currently supported are: - gen10: First Wii Remote generation - gen20: Second Wii Remote Plus generation (builtin MP) + + ============= ======================================= + gen10: First Wii Remote generation + gen20: Second Wii Remote Plus generation + (builtin MP) balanceboard: Wii Balance Board + ============= ======================================= What: /sys/bus/hid/drivers/wiimote//bboard_calib Date: May 2013 @@ -54,6 +58,7 @@ Description: This attribute is only provided if the device was detected as a First, 0kg values for all 4 sensors are written, followed by the 17kg values for all 4 sensors and last the 34kg values for all 4 sensors. + Calibration data is already applied by the kernel to all input values but may be used by user-space to perform other transformations. @@ -68,9 +73,11 @@ Description: This attribute is only provided if the device was detected as a is prefixed with a +/-. Each value is a signed 16bit number. Data is encoded as decimal numbers and specifies the offsets of the analog sticks of the pro-controller. + Calibration data is already applied by the kernel to all input values but may be used by user-space to perform other transformations. + Calibration data is detected by the kernel during device setup. You can write "scan\n" into this file to re-trigger calibration. You can also write data directly in the form "x1:y1 x2:y2" to diff --git a/Documentation/ABI/testing/sysfs-driver-samsung-laptop b/Documentation/ABI/testing/sysfs-driver-samsung-laptop index 34d3a3359cf4..28c9c040de5d 100644 --- a/Documentation/ABI/testing/sysfs-driver-samsung-laptop +++ b/Documentation/ABI/testing/sysfs-driver-samsung-laptop @@ -9,10 +9,12 @@ Description: Some Samsung laptops have different "performance levels" their fans quiet at all costs. Reading from this file will show the current performance level. Writing to the file can change this value. + Valid options: - "silent" - "normal" - "overclock" + - "silent" + - "normal" + - "overclock" + Note that not all laptops support all of these options. Specifically, not all support the "overclock" option, and it's still unknown if this value even changes @@ -25,8 +27,9 @@ Contact: Corentin Chary Description: Max battery charge level can be modified, battery cycle life can be extended by reducing the max battery charge level. - 0 means normal battery mode (100% charge) - 1 means battery life extender mode (80% charge) + + - 0 means normal battery mode (100% charge) + - 1 means battery life extender mode (80% charge) What: /sys/devices/platform/samsung/usb_charge Date: December 1, 2011 diff --git a/Documentation/ABI/testing/sysfs-driver-toshiba_acpi b/Documentation/ABI/testing/sysfs-driver-toshiba_acpi index f34221b52b14..e5a438d84e1f 100644 --- a/Documentation/ABI/testing/sysfs-driver-toshiba_acpi +++ b/Documentation/ABI/testing/sysfs-driver-toshiba_acpi @@ -4,10 +4,12 @@ KernelVersion: 3.15 Contact: Azael Avalos Description: This file controls the keyboard backlight operation mode, valid values are: + * 0x1 -> FN-Z * 0x2 -> AUTO (also called TIMER) * 0x8 -> ON * 0x10 -> OFF + Note that from kernel 3.16 onwards this file accepts all listed parameters, kernel 3.15 only accepts the first two (FN-Z and AUTO). @@ -41,8 +43,10 @@ KernelVersion: 3.15 Contact: Azael Avalos Description: This files controls the status of the touchpad and pointing stick (if available), valid values are: + * 0 -> OFF * 1 -> ON + Users: KToshiba What: /sys/devices/LNXSYSTM:00/LNXSYBUS:00/TOS{1900,620{0,7,8}}:00/available_kbd_modes @@ -51,10 +55,12 @@ KernelVersion: 3.16 Contact: Azael Avalos Description: This file shows the supported keyboard backlight modes the system supports, which can be: + * 0x1 -> FN-Z * 0x2 -> AUTO (also called TIMER) * 0x8 -> ON * 0x10 -> OFF + Note that not all keyboard types support the listed modes. See the entry named "available_kbd_modes" Users: KToshiba @@ -65,6 +71,7 @@ KernelVersion: 3.16 Contact: Azael Avalos Description: This file shows the current keyboard backlight type, which can be: + * 1 -> Type 1, supporting modes FN-Z and AUTO * 2 -> Type 2, supporting modes TIMER, ON and OFF Users: KToshiba @@ -75,10 +82,12 @@ KernelVersion: 4.0 Contact: Azael Avalos Description: This file controls the USB Sleep & Charge charging mode, which can be: + * 0 -> Disabled (0x00) * 1 -> Alternate (0x09) * 2 -> Auto (0x21) * 3 -> Typical (0x11) + Note that from kernel 4.1 onwards this file accepts all listed values, kernel 4.0 only supports the first three. Note that this feature only works when connected to power, if @@ -93,8 +102,10 @@ Contact: Azael Avalos Description: This file controls the USB Sleep Functions under battery, and set the level at which point they will be disabled, accepted values can be: + * 0 -> Disabled * 1-100 -> Battery level to disable sleep functions + Currently it prints two values, the first one indicates if the feature is enabled or disabled, while the second one shows the current battery level set. @@ -107,8 +118,10 @@ Date: January 23, 2015 KernelVersion: 4.0 Contact: Azael Avalos Description: This file controls the USB Rapid Charge state, which can be: + * 0 -> Disabled * 1 -> Enabled + Note that toggling this value requires a reboot for changes to take effect. Users: KToshiba @@ -118,8 +131,10 @@ Date: January 23, 2015 KernelVersion: 4.0 Contact: Azael Avalos Description: This file controls the Sleep & Music state, which values can be: + * 0 -> Disabled * 1 -> Enabled + Note that this feature only works when connected to power, if you want to use it under battery, see the entry named "sleep_functions_on_battery" @@ -138,6 +153,7 @@ KernelVersion: 4.0 Contact: Azael Avalos Description: This file controls the state of the internal fan, valid values are: + * 0 -> OFF * 1 -> ON @@ -147,8 +163,10 @@ KernelVersion: 4.0 Contact: Azael Avalos Description: This file controls the Special Functions (hotkeys) operation mode, valid values are: + * 0 -> Normal Operation * 1 -> Special Functions + In the "Normal Operation" mode, the F{1-12} keys are as usual and the hotkeys are accessed via FN-F{1-12}. In the "Special Functions" mode, the F{1-12} keys trigger the @@ -163,8 +181,10 @@ KernelVersion: 4.0 Contact: Azael Avalos Description: This file controls whether the laptop should turn ON whenever the LID is opened, valid values are: + * 0 -> Disabled * 1 -> Enabled + Note that toggling this value requires a reboot for changes to take effect. Users: KToshiba @@ -174,8 +194,10 @@ Date: February 12, 2015 KernelVersion: 4.0 Contact: Azael Avalos Description: This file controls the USB 3 functionality, valid values are: + * 0 -> Disabled (Acts as a regular USB 2) * 1 -> Enabled (Full USB 3 functionality) + Note that toggling this value requires a reboot for changes to take effect. Users: KToshiba @@ -188,10 +210,14 @@ Description: This file controls the Cooling Method feature. Reading this file prints two values, the first is the actual cooling method and the second is the maximum cooling method supported. When the maximum cooling method is ONE, valid values are: + * 0 -> Maximum Performance * 1 -> Battery Optimized + When the maximum cooling method is TWO, valid values are: + * 0 -> Maximum Performance * 1 -> Performance * 2 -> Battery Optimized + Users: KToshiba diff --git a/Documentation/ABI/testing/sysfs-driver-toshiba_haps b/Documentation/ABI/testing/sysfs-driver-toshiba_haps index a662370b4dbf..c938690ce10d 100644 --- a/Documentation/ABI/testing/sysfs-driver-toshiba_haps +++ b/Documentation/ABI/testing/sysfs-driver-toshiba_haps @@ -4,10 +4,12 @@ KernelVersion: 3.17 Contact: Azael Avalos Description: This file controls the built-in accelerometer protection level, valid values are: + * 0 -> Disabled * 1 -> Low * 2 -> Medium * 3 -> High + The default potection value is set to 2 (Medium). Users: KToshiba diff --git a/Documentation/ABI/testing/sysfs-driver-wacom b/Documentation/ABI/testing/sysfs-driver-wacom index afc48fc163b5..16acaa5712ec 100644 --- a/Documentation/ABI/testing/sysfs-driver-wacom +++ b/Documentation/ABI/testing/sysfs-driver-wacom @@ -79,7 +79,9 @@ Description: When the Wacom Intuos 4 is connected over Bluetooth, the image has to contain 256 bytes (64x32 px 1 bit colour). The format is also scrambled, like in the USB mode, and it can - be summarized by converting 76543210 into GECA6420. + be summarized by converting:: + + 76543210 into GECA6420. HGFEDCBA HFDB7531 What: /sys/bus/hid/devices/::./wacom_remote/unpair_remote diff --git a/Documentation/ABI/testing/sysfs-firmware-acpi b/Documentation/ABI/testing/sysfs-firmware-acpi index 613f42a9d5cd..e4afc2538210 100644 --- a/Documentation/ABI/testing/sysfs-firmware-acpi +++ b/Documentation/ABI/testing/sysfs-firmware-acpi @@ -12,11 +12,14 @@ Description: image: The image bitmap. Currently a 32-bit BMP. status: 1 if the image is valid, 0 if firmware invalidated it. type: 0 indicates image is in BMP format. + + ======== =================================================== version: The version of the BGRT. Currently 1. xoffset: The number of pixels between the left of the screen and the left edge of the image. yoffset: The number of pixels between the top of the screen and the top edge of the image. + ======== =================================================== What: /sys/firmware/acpi/hotplug/ Date: February 2013 @@ -33,12 +36,14 @@ Description: The following setting is available to user space for each hotplug profile: + ======== ======================================================= enabled: If set, the ACPI core will handle notifications of - hotplug events associated with the given class of - devices and will allow those devices to be ejected with - the help of the _EJ0 control method. Unsetting it - effectively disables hotplug for the correspoinding - class of devices. + hotplug events associated with the given class of + devices and will allow those devices to be ejected with + the help of the _EJ0 control method. Unsetting it + effectively disables hotplug for the correspoinding + class of devices. + ======== ======================================================= The value of the above attribute is an integer number: 1 (set) or 0 (unset). Attempts to write any other values to it will @@ -71,86 +76,90 @@ Description: To figure out where all the SCI's are coming from, /sys/firmware/acpi/interrupts contains a file listing every possible source, and the count of how many - times it has triggered. - - $ cd /sys/firmware/acpi/interrupts - $ grep . * - error: 0 - ff_gbl_lock: 0 enable - ff_pmtimer: 0 invalid - ff_pwr_btn: 0 enable - ff_rt_clk: 2 disable - ff_slp_btn: 0 invalid - gpe00: 0 invalid - gpe01: 0 enable - gpe02: 108 enable - gpe03: 0 invalid - gpe04: 0 invalid - gpe05: 0 invalid - gpe06: 0 enable - gpe07: 0 enable - gpe08: 0 invalid - gpe09: 0 invalid - gpe0A: 0 invalid - gpe0B: 0 invalid - gpe0C: 0 invalid - gpe0D: 0 invalid - gpe0E: 0 invalid - gpe0F: 0 invalid - gpe10: 0 invalid - gpe11: 0 invalid - gpe12: 0 invalid - gpe13: 0 invalid - gpe14: 0 invalid - gpe15: 0 invalid - gpe16: 0 invalid - gpe17: 1084 enable - gpe18: 0 enable - gpe19: 0 invalid - gpe1A: 0 invalid - gpe1B: 0 invalid - gpe1C: 0 invalid - gpe1D: 0 invalid - gpe1E: 0 invalid - gpe1F: 0 invalid - gpe_all: 1192 - sci: 1194 - sci_not: 0 - - sci - The number of times the ACPI SCI - has been called and claimed an interrupt. - - sci_not - The number of times the ACPI SCI - has been called and NOT claimed an interrupt. - - gpe_all - count of SCI caused by GPEs. - - gpeXX - count for individual GPE source - - ff_gbl_lock - Global Lock - - ff_pmtimer - PM Timer - - ff_pwr_btn - Power Button - - ff_rt_clk - Real Time Clock - - ff_slp_btn - Sleep Button - - error - an interrupt that can't be accounted for above. - - invalid: it's either a GPE or a Fixed Event that - doesn't have an event handler. - - disable: the GPE/Fixed Event is valid but disabled. - - enable: the GPE/Fixed Event is valid and enabled. - - Root has permission to clear any of these counters. Eg. - # echo 0 > gpe11 - - All counters can be cleared by clearing the total "sci": - # echo 0 > sci + times it has triggered:: + + $ cd /sys/firmware/acpi/interrupts + $ grep . * + error: 0 + ff_gbl_lock: 0 enable + ff_pmtimer: 0 invalid + ff_pwr_btn: 0 enable + ff_rt_clk: 2 disable + ff_slp_btn: 0 invalid + gpe00: 0 invalid + gpe01: 0 enable + gpe02: 108 enable + gpe03: 0 invalid + gpe04: 0 invalid + gpe05: 0 invalid + gpe06: 0 enable + gpe07: 0 enable + gpe08: 0 invalid + gpe09: 0 invalid + gpe0A: 0 invalid + gpe0B: 0 invalid + gpe0C: 0 invalid + gpe0D: 0 invalid + gpe0E: 0 invalid + gpe0F: 0 invalid + gpe10: 0 invalid + gpe11: 0 invalid + gpe12: 0 invalid + gpe13: 0 invalid + gpe14: 0 invalid + gpe15: 0 invalid + gpe16: 0 invalid + gpe17: 1084 enable + gpe18: 0 enable + gpe19: 0 invalid + gpe1A: 0 invalid + gpe1B: 0 invalid + gpe1C: 0 invalid + gpe1D: 0 invalid + gpe1E: 0 invalid + gpe1F: 0 invalid + gpe_all: 1192 + sci: 1194 + sci_not: 0 + + =========== ================================================== + sci The number of times the ACPI SCI + has been called and claimed an interrupt. + + sci_not The number of times the ACPI SCI + has been called and NOT claimed an interrupt. + + gpe_all count of SCI caused by GPEs. + + gpeXX count for individual GPE source + + ff_gbl_lock Global Lock + + ff_pmtimer PM Timer + + ff_pwr_btn Power Button + + ff_rt_clk Real Time Clock + + ff_slp_btn Sleep Button + + error an interrupt that can't be accounted for above. + + invalid it's either a GPE or a Fixed Event that + doesn't have an event handler. + + disable the GPE/Fixed Event is valid but disabled. + + enable the GPE/Fixed Event is valid and enabled. + =========== ================================================== + + Root has permission to clear any of these counters. Eg.:: + + # echo 0 > gpe11 + + All counters can be cleared by clearing the total "sci":: + + # echo 0 > sci None of these counters has an effect on the function of the system, they are simply statistics. @@ -165,32 +174,34 @@ Description: Let's take power button fixed event for example, please kill acpid and other user space applications so that the machine won't shutdown - when pressing the power button. - # cat ff_pwr_btn - 0 enabled - # press the power button for 3 times; - # cat ff_pwr_btn - 3 enabled - # echo disable > ff_pwr_btn - # cat ff_pwr_btn - 3 disabled - # press the power button for 3 times; - # cat ff_pwr_btn - 3 disabled - # echo enable > ff_pwr_btn - # cat ff_pwr_btn - 4 enabled - /* - * this is because the status bit is set even if the enable bit is cleared, - * and it triggers an ACPI fixed event when the enable bit is set again - */ - # press the power button for 3 times; - # cat ff_pwr_btn - 7 enabled - # echo disable > ff_pwr_btn - # press the power button for 3 times; - # echo clear > ff_pwr_btn /* clear the status bit */ - # echo disable > ff_pwr_btn - # cat ff_pwr_btn - 7 enabled + when pressing the power button:: + + # cat ff_pwr_btn + 0 enabled + # press the power button for 3 times; + # cat ff_pwr_btn + 3 enabled + # echo disable > ff_pwr_btn + # cat ff_pwr_btn + 3 disabled + # press the power button for 3 times; + # cat ff_pwr_btn + 3 disabled + # echo enable > ff_pwr_btn + # cat ff_pwr_btn + 4 enabled + /* + * this is because the status bit is set even if the enable + * bit is cleared, and it triggers an ACPI fixed event when + * the enable bit is set again + */ + # press the power button for 3 times; + # cat ff_pwr_btn + 7 enabled + # echo disable > ff_pwr_btn + # press the power button for 3 times; + # echo clear > ff_pwr_btn /* clear the status bit */ + # echo disable > ff_pwr_btn + # cat ff_pwr_btn + 7 enabled diff --git a/Documentation/ABI/testing/sysfs-firmware-dmi-entries b/Documentation/ABI/testing/sysfs-firmware-dmi-entries index 210ad44b95a5..fe0289c87768 100644 --- a/Documentation/ABI/testing/sysfs-firmware-dmi-entries +++ b/Documentation/ABI/testing/sysfs-firmware-dmi-entries @@ -33,7 +33,7 @@ Description: doesn't matter), they will be represented in sysfs as entries "T-0" through "T-(N-1)": - Example entry directories: + Example entry directories:: /sys/firmware/dmi/entries/17-0 /sys/firmware/dmi/entries/17-1 @@ -50,61 +50,65 @@ Description: Each DMI entry in sysfs has the common header values exported as attributes: - handle : The 16bit 'handle' that is assigned to this + ======== ================================================= + handle The 16bit 'handle' that is assigned to this entry by the firmware. This handle may be referred to by other entries. - length : The length of the entry, as presented in the + length The length of the entry, as presented in the entry itself. Note that this is _not the total count of bytes associated with the - entry_. This value represents the length of + entry. This value represents the length of the "formatted" portion of the entry. This "formatted" region is sometimes followed by the "unformatted" region composed of nul terminated strings, with termination signalled by a two nul characters in series. - raw : The raw bytes of the entry. This includes the + raw The raw bytes of the entry. This includes the "formatted" portion of the entry, the "unformatted" strings portion of the entry, and the two terminating nul characters. - type : The type of the entry. This value is the same + type The type of the entry. This value is the same as found in the directory name. It indicates how the rest of the entry should be interpreted. - instance: The instance ordinal of the entry for the + instance The instance ordinal of the entry for the given type. This value is the same as found in the parent directory name. - position: The ordinal position (zero-based) of the entry + position The ordinal position (zero-based) of the entry within the entirety of the DMI entry table. + ======== ================================================= - === Entry Specialization === + **Entry Specialization** Some entry types may have other information available in sysfs. Not all types are specialized. - --- Type 15 - System Event Log --- + **Type 15 - System Event Log** This entry allows the firmware to export a log of events the system has taken. This information is typically backed by nvram, but the implementation details are abstracted by this table. This entry's data - is exported in the directory: + is exported in the directory:: - /sys/firmware/dmi/entries/15-0/system_event_log + /sys/firmware/dmi/entries/15-0/system_event_log and has the following attributes (documented in the SMBIOS / DMI specification under "System Event Log (Type 15)": - area_length - header_start_offset - data_start_offset - access_method - status - change_token - access_method_address - header_format - per_log_type_descriptor_length - type_descriptors_supported_count + - area_length + - header_start_offset + - data_start_offset + - access_method + - status + - change_token + - access_method_address + - header_format + - per_log_type_descriptor_length + - type_descriptors_supported_count As well, the kernel exports the binary attribute: - raw_event_log : The raw binary bits of the event log + ============= ==================================== + raw_event_log The raw binary bits of the event log as described by the DMI entry. + ============= ==================================== diff --git a/Documentation/ABI/testing/sysfs-firmware-gsmi b/Documentation/ABI/testing/sysfs-firmware-gsmi index 0faa0aaf4b6a..7a558354c1ee 100644 --- a/Documentation/ABI/testing/sysfs-firmware-gsmi +++ b/Documentation/ABI/testing/sysfs-firmware-gsmi @@ -20,7 +20,7 @@ Description: This directory has the same layout (and underlying implementation as /sys/firmware/efi/vars. - See Documentation/ABI/*/sysfs-firmware-efi-vars + See `Documentation/ABI/*/sysfs-firmware-efi-vars` for more information on how to interact with this structure. diff --git a/Documentation/ABI/testing/sysfs-firmware-memmap b/Documentation/ABI/testing/sysfs-firmware-memmap index eca0d65087dc..1f6f4d3a32c0 100644 --- a/Documentation/ABI/testing/sysfs-firmware-memmap +++ b/Documentation/ABI/testing/sysfs-firmware-memmap @@ -20,7 +20,7 @@ Description: the raw memory map to userspace. The structure is as follows: Under /sys/firmware/memmap there - are subdirectories with the number of the entry as their name: + are subdirectories with the number of the entry as their name:: /sys/firmware/memmap/0 /sys/firmware/memmap/1 @@ -34,14 +34,16 @@ Description: Each directory contains three files: - start : The start address (as hexadecimal number with the + ======== ===================================================== + start The start address (as hexadecimal number with the '0x' prefix). - end : The end address, inclusive (regardless whether the + end The end address, inclusive (regardless whether the firmware provides inclusive or exclusive ranges). - type : Type of the entry as string. See below for a list of + type Type of the entry as string. See below for a list of valid types. + ======== ===================================================== - So, for example: + So, for example:: /sys/firmware/memmap/0/start /sys/firmware/memmap/0/end @@ -57,9 +59,8 @@ Description: - reserved Following shell snippet can be used to display that memory - map in a human-readable format: + map in a human-readable format:: - -------------------- 8< ---------------------------------------- #!/bin/bash cd /sys/firmware/memmap for dir in * ; do @@ -68,4 +69,3 @@ Description: type=$(cat $dir/type) printf "%016x-%016x (%s)\n" $start $[ $end +1] "$type" done - -------------------- >8 ---------------------------------------- diff --git a/Documentation/ABI/testing/sysfs-fs-ext4 b/Documentation/ABI/testing/sysfs-fs-ext4 index 78604db56279..99e3d92f8299 100644 --- a/Documentation/ABI/testing/sysfs-fs-ext4 +++ b/Documentation/ABI/testing/sysfs-fs-ext4 @@ -45,8 +45,8 @@ Description: parameter will have their blocks allocated out of a block group specific preallocation pool, so that small files are packed closely together. Each large file - will have its blocks allocated out of its own unique - preallocation pool. + will have its blocks allocated out of its own unique + preallocation pool. What: /sys/fs/ext4//inode_readahead_blks Date: March 2008 diff --git a/Documentation/ABI/testing/sysfs-hypervisor-xen b/Documentation/ABI/testing/sysfs-hypervisor-xen index 53b7b2ea7515..4dbe0c49b393 100644 --- a/Documentation/ABI/testing/sysfs-hypervisor-xen +++ b/Documentation/ABI/testing/sysfs-hypervisor-xen @@ -15,14 +15,17 @@ KernelVersion: 4.3 Contact: Boris Ostrovsky Description: If running under Xen: Describes mode that Xen's performance-monitoring unit (PMU) - uses. Accepted values are - "off" -- PMU is disabled - "self" -- The guest can profile itself - "hv" -- The guest can profile itself and, if it is + uses. Accepted values are: + + ====== ============================================ + "off" PMU is disabled + "self" The guest can profile itself + "hv" The guest can profile itself and, if it is privileged (e.g. dom0), the hypervisor - "all" -- The guest can profile itself, the hypervisor + "all" The guest can profile itself, the hypervisor and all other guests. Only available to privileged guests. + ====== ============================================ What: /sys/hypervisor/pmu/pmu_features Date: August 2015 diff --git a/Documentation/ABI/testing/sysfs-kernel-boot_params b/Documentation/ABI/testing/sysfs-kernel-boot_params index eca38ce2852d..7f9bda453c4d 100644 --- a/Documentation/ABI/testing/sysfs-kernel-boot_params +++ b/Documentation/ABI/testing/sysfs-kernel-boot_params @@ -23,16 +23,17 @@ Description: The /sys/kernel/boot_params directory contains two representation of setup_data type. "data" file is the binary representation of setup_data payload. - The whole boot_params directory structure is like below: - /sys/kernel/boot_params - |__ data - |__ setup_data - | |__ 0 - | | |__ data - | | |__ type - | |__ 1 - | |__ data - | |__ type - |__ version + The whole boot_params directory structure is like below:: + + /sys/kernel/boot_params + |__ data + |__ setup_data + | |__ 0 + | | |__ data + | | |__ type + | |__ 1 + | |__ data + | |__ type + |__ version Users: Kexec diff --git a/Documentation/ABI/testing/sysfs-kernel-mm-hugepages b/Documentation/ABI/testing/sysfs-kernel-mm-hugepages index fdaa2162fae1..294387e2c7fb 100644 --- a/Documentation/ABI/testing/sysfs-kernel-mm-hugepages +++ b/Documentation/ABI/testing/sysfs-kernel-mm-hugepages @@ -7,9 +7,11 @@ Description: of the hugepages supported by the kernel/CPU combination. Under these directories are a number of files: - nr_hugepages - nr_overcommit_hugepages - free_hugepages - surplus_hugepages - resv_hugepages + + - nr_hugepages + - nr_overcommit_hugepages + - free_hugepages + - surplus_hugepages + - resv_hugepages + See Documentation/admin-guide/mm/hugetlbpage.rst for details. diff --git a/Documentation/ABI/testing/sysfs-platform-asus-laptop b/Documentation/ABI/testing/sysfs-platform-asus-laptop index 8b0e8205a6a2..c78d358dbdbe 100644 --- a/Documentation/ABI/testing/sysfs-platform-asus-laptop +++ b/Documentation/ABI/testing/sysfs-platform-asus-laptop @@ -4,13 +4,16 @@ KernelVersion: 2.6.20 Contact: "Corentin Chary" Description: This file allows display switching. The value - is composed by 4 bits and defined as follow: - 4321 - |||`- LCD - ||`-- CRT - |`--- TV - `---- DVI - Ex: - 0 (0000b) means no display + is composed by 4 bits and defined as follow:: + + 4321 + |||`- LCD + ||`-- CRT + |`--- TV + `---- DVI + + Ex: + - 0 (0000b) means no display - 3 (0011b) CRT+LCD. What: /sys/devices/platform/asus_laptop/gps @@ -28,8 +31,10 @@ Contact: "Corentin Chary" Description: Some models like the W1N have a LED display that can be used to display several items of information. - To control the LED display, use the following : + To control the LED display, use the following:: + echo 0x0T000DDD > /sys/devices/platform/asus_laptop/ + where T control the 3 letters display, and DDD the 3 digits display. The DDD table can be found in Documentation/admin-guide/laptops/asus-laptop.rst diff --git a/Documentation/ABI/testing/sysfs-platform-asus-wmi b/Documentation/ABI/testing/sysfs-platform-asus-wmi index 1efac0ddb417..04885738cf15 100644 --- a/Documentation/ABI/testing/sysfs-platform-asus-wmi +++ b/Documentation/ABI/testing/sysfs-platform-asus-wmi @@ -5,6 +5,7 @@ Contact: "Corentin Chary" Description: Change CPU clock configuration (write-only). There are three available clock configuration: + * 0 -> Super Performance Mode * 1 -> High Performance Mode * 2 -> Power Saving Mode diff --git a/Documentation/ABI/testing/sysfs-platform-at91 b/Documentation/ABI/testing/sysfs-platform-at91 index 4cc6a865ae66..b146be74b8e0 100644 --- a/Documentation/ABI/testing/sysfs-platform-at91 +++ b/Documentation/ABI/testing/sysfs-platform-at91 @@ -18,8 +18,10 @@ Description: In order to use an extended can_id add the CAN_EFF_FLAG (0x80000000U) to the can_id. Example: - - standard id 0x7ff: - echo 0x7ff > /sys/class/net/can0/mb0_id + - standard id 0x7ff:: - - extended id 0x1fffffff: - echo 0x9fffffff > /sys/class/net/can0/mb0_id + echo 0x7ff > /sys/class/net/can0/mb0_id + + - extended id 0x1fffffff:: + + echo 0x9fffffff > /sys/class/net/can0/mb0_id diff --git a/Documentation/ABI/testing/sysfs-platform-eeepc-laptop b/Documentation/ABI/testing/sysfs-platform-eeepc-laptop index 5b026c69587a..70dbe0733cf6 100644 --- a/Documentation/ABI/testing/sysfs-platform-eeepc-laptop +++ b/Documentation/ABI/testing/sysfs-platform-eeepc-laptop @@ -4,9 +4,11 @@ KernelVersion: 2.6.26 Contact: "Corentin Chary" Description: This file allows display switching. + - 1 = LCD - 2 = CRT - 3 = LCD+CRT + If you run X11, you should use xrandr instead. What: /sys/devices/platform/eeepc/camera @@ -30,16 +32,20 @@ Contact: "Corentin Chary" Description: Change CPU clock configuration. On the Eee PC 1000H there are three available clock configuration: + * 0 -> Super Performance Mode * 1 -> High Performance Mode * 2 -> Power Saving Mode + On Eee PC 701 there is only 2 available clock configurations. Available configuration are listed in available_cpufv file. Reading this file will show the raw hexadecimal value which - is defined as follow: - | 8 bit | 8 bit | - | `---- Current mode - `------------ Availables modes + is defined as follow:: + + | 8 bit | 8 bit | + | `---- Current mode + `------------ Availables modes + For example, 0x301 means: mode 1 selected, 3 available modes. What: /sys/devices/platform/eeepc/available_cpufv diff --git a/Documentation/ABI/testing/sysfs-platform-ideapad-laptop b/Documentation/ABI/testing/sysfs-platform-ideapad-laptop index 1b31be3f996a..fd2ac02bc5bd 100644 --- a/Documentation/ABI/testing/sysfs-platform-ideapad-laptop +++ b/Documentation/ABI/testing/sysfs-platform-ideapad-laptop @@ -12,6 +12,7 @@ Contact: "Maxim Mikityanskiy " Description: Change fan mode There are four available modes: + * 0 -> Super Silent Mode * 1 -> Standard Mode * 2 -> Dust Cleaning @@ -32,9 +33,11 @@ KernelVersion: 4.18 Contact: "Oleg Keri " Description: Control fn-lock mode. + * 1 -> Switched On * 0 -> Switched Off - For example: - # echo "0" > \ - /sys/bus/pci/devices/0000:00:1f.0/PNP0C09:00/VPC2004:00/fn_lock + For example:: + + # echo "0" > \ + /sys/bus/pci/devices/0000:00:1f.0/PNP0C09:00/VPC2004:00/fn_lock diff --git a/Documentation/ABI/testing/sysfs-platform-intel-wmi-thunderbolt b/Documentation/ABI/testing/sysfs-platform-intel-wmi-thunderbolt index 8af65059d519..e19144fd5d86 100644 --- a/Documentation/ABI/testing/sysfs-platform-intel-wmi-thunderbolt +++ b/Documentation/ABI/testing/sysfs-platform-intel-wmi-thunderbolt @@ -7,5 +7,6 @@ Description: Thunderbolt controllers to turn on or off when no devices are connected (write-only) There are two available states: + * 0 -> Force power disabled * 1 -> Force power enabled diff --git a/Documentation/ABI/testing/sysfs-platform-sst-atom b/Documentation/ABI/testing/sysfs-platform-sst-atom index 0d07c0395660..d5f6e21f0e42 100644 --- a/Documentation/ABI/testing/sysfs-platform-sst-atom +++ b/Documentation/ABI/testing/sysfs-platform-sst-atom @@ -5,13 +5,22 @@ Contact: "Sebastien Guiriec" Description: LPE Firmware version for SST driver on all atom plaforms (BYT/CHT/Merrifield/BSW). - If the FW has never been loaded it will display: + If the FW has never been loaded it will display:: + "FW not yet loaded" - If FW has been loaded it will display: + + If FW has been loaded it will display:: + "v01.aa.bb.cc" + aa: Major version is reflecting SoC version: + + === ============= 0d: BYT FW 0b: BSW FW 07: Merrifield FW + === ============= + bb: Minor version + cc: Build version diff --git a/Documentation/ABI/testing/sysfs-platform-usbip-vudc b/Documentation/ABI/testing/sysfs-platform-usbip-vudc index 81fcfb454913..53622d3ba27c 100644 --- a/Documentation/ABI/testing/sysfs-platform-usbip-vudc +++ b/Documentation/ABI/testing/sysfs-platform-usbip-vudc @@ -16,10 +16,13 @@ Contact: Krzysztof Opasiak Description: Current status of the device. Allowed values: - 1 - Device is available and can be exported - 2 - Device is currently exported - 3 - Fatal error occurred during communication - with peer + + == ========================================== + 1 Device is available and can be exported + 2 Device is currently exported + 3 Fatal error occurred during communication + with peer + == ========================================== What: /sys/devices/platform/usbip-vudc.%d/usbip_sockfd Date: April 2016 diff --git a/Documentation/ABI/testing/sysfs-ptp b/Documentation/ABI/testing/sysfs-ptp index a17f817a9309..2363ad810ddb 100644 --- a/Documentation/ABI/testing/sysfs-ptp +++ b/Documentation/ABI/testing/sysfs-ptp @@ -69,7 +69,7 @@ Description: pin offered by the PTP hardware clock. The file name is the hardware dependent pin name. Reading from this file produces two numbers, the assigned function (see - the PTP_PF_ enumeration values in linux/ptp_clock.h) + the `PTP_PF_` enumeration values in linux/ptp_clock.h) and the channel number. The function and channel assignment may be changed by two writing numbers into the file. -- cgit v1.2.3 From 906e4af9359dff0e24aeaf3b92226847cb5c8185 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:40 +0100 Subject: docs: ABI: sysfs-uevent: make it compatible with ReST output - Replace " by ``, in order to use monospaced fonts; - mark literal blocks as such. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/63904cc9b6a8581c5fc2ea1dca5d925874c67372.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/testing/sysfs-uevent | 28 ++++++++++++++++------------ 1 file changed, 16 insertions(+), 12 deletions(-) diff --git a/Documentation/ABI/testing/sysfs-uevent b/Documentation/ABI/testing/sysfs-uevent index aa39f8d7bcdf..0b6227706b35 100644 --- a/Documentation/ABI/testing/sysfs-uevent +++ b/Documentation/ABI/testing/sysfs-uevent @@ -6,42 +6,46 @@ Description: Enable passing additional variables for synthetic uevents that are generated by writing /sys/.../uevent file. - Recognized extended format is ACTION [UUID [KEY=VALUE ...]. + Recognized extended format is:: - The ACTION is compulsory - it is the name of the uevent action - ("add", "change", "remove"). There is no change compared to - previous functionality here. The rest of the extended format - is optional. + ACTION [UUID [KEY=VALUE ...] + + The ACTION is compulsory - it is the name of the uevent + action (``add``, ``change``, ``remove``). There is no change + compared to previous functionality here. The rest of the + extended format is optional. You need to pass UUID first before any KEY=VALUE pairs. - The UUID must be in "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx" + The UUID must be in ``xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx`` format where 'x' is a hex digit. The UUID is considered to be a transaction identifier so it's possible to use the same UUID value for one or more synthetic uevents in which case we logically group these uevents together for any userspace listeners. The UUID value appears in uevent as - "SYNTH_UUID=xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx" environment + ``SYNTH_UUID=xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx`` environment variable. If UUID is not passed in, the generated synthetic uevent gains - "SYNTH_UUID=0" environment variable automatically. + ``SYNTH_UUID=0`` environment variable automatically. The KEY=VALUE pairs can contain alphanumeric characters only. + It's possible to define zero or more pairs - each pair is then delimited by a space character ' '. Each pair appears in - synthetic uevent as "SYNTH_ARG_KEY=VALUE". That means the KEY - name gains "SYNTH_ARG_" prefix to avoid possible collisions + synthetic uevent as ``SYNTH_ARG_KEY=VALUE``. That means the KEY + name gains ``SYNTH_ARG_`` prefix to avoid possible collisions with existing variables. - Example of valid sequence written to the uevent file: + Example of valid sequence written to the uevent file:: add fe4d7c9d-b8c6-4a70-9ef1-3d8a58d18eed A=1 B=abc - This generates synthetic uevent including these variables: + This generates synthetic uevent including these variables:: ACTION=add SYNTH_ARG_A=1 SYNTH_ARG_B=abc SYNTH_UUID=fe4d7c9d-b8c6-4a70-9ef1-3d8a58d18eed + Users: udev, userspace tools generating synthetic uevents -- cgit v1.2.3 From 642514df1bc1c6a9318b537670fdb55a6ee38e45 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:41 +0100 Subject: docs: ABI: make it parse ABI/stable as ReST-compatible files Now that the stable ABI files are compatible with ReST, parse them without converting complex descriptions as literal blocks nor escaping special characters. Please notice that escaping special characters will probably be needed at descriptions, at least for the asterisk character. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/59ccbaa75ff05f23e701dd9a0bbe118e9343a553.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/admin-guide/abi-stable.rst | 1 + Documentation/sphinx/kernel_abi.py | 8 ++++++-- 2 files changed, 7 insertions(+), 2 deletions(-) diff --git a/Documentation/admin-guide/abi-stable.rst b/Documentation/admin-guide/abi-stable.rst index 7495d7a35048..70490736e0d3 100644 --- a/Documentation/admin-guide/abi-stable.rst +++ b/Documentation/admin-guide/abi-stable.rst @@ -11,3 +11,4 @@ Most interfaces (like syscalls) are expected to never change and always be available. .. kernel-abi:: $srctree/Documentation/ABI/stable + :rst: diff --git a/Documentation/sphinx/kernel_abi.py b/Documentation/sphinx/kernel_abi.py index ce5f3b0ae811..f3da859c9878 100644 --- a/Documentation/sphinx/kernel_abi.py +++ b/Documentation/sphinx/kernel_abi.py @@ -73,12 +73,13 @@ class KernelCmd(Directive): u"""KernelABI (``kernel-abi``) directive""" required_arguments = 1 - optional_arguments = 0 + optional_arguments = 2 has_content = False final_argument_whitespace = True option_spec = { - "debug" : directives.flag + "debug" : directives.flag, + "rst" : directives.unchanged } def run(self): @@ -92,6 +93,9 @@ class KernelCmd(Directive): cmd = "get_abi.pl rest --enable-lineno --dir " cmd += self.arguments[0] + if 'rst' in self.options: + cmd += " --rst-source" + srctree = path.abspath(os.environ["srctree"]) fname = cmd -- cgit v1.2.3 From c12759ff2dd029ce74b026e6b6bc62060383a5a7 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:42 +0100 Subject: docs: ABI: create a 2-depth index for ABI That helps to identify what ABI files are adding titles. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/e267b36ae7f32bab2a86f1da6b40bb3e62c877d4.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/admin-guide/abi.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/admin-guide/abi.rst b/Documentation/admin-guide/abi.rst index 3b9645c77469..bcab3ef2597c 100644 --- a/Documentation/admin-guide/abi.rst +++ b/Documentation/admin-guide/abi.rst @@ -3,7 +3,7 @@ Linux ABI description ===================== .. toctree:: - :maxdepth: 1 + :maxdepth: 2 abi-stable abi-testing -- cgit v1.2.3 From 7832dea738dc3ea4a40860dc8b9f83ca1e092ab9 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:43 +0100 Subject: docs: ABI: don't escape ReST-incompatible chars from obsolete and removed With just a single fix, the contents there can be parsed properly without the need to escape any ReST incompatible stuff. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/472f4574b6aa2ff4de5a819db1a4a5c9a34f5168.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/obsolete/sysfs-gpio | 2 ++ Documentation/admin-guide/abi-obsolete.rst | 1 + Documentation/admin-guide/abi-removed.rst | 1 + 3 files changed, 4 insertions(+) diff --git a/Documentation/ABI/obsolete/sysfs-gpio b/Documentation/ABI/obsolete/sysfs-gpio index e0d4e5e2dd90..b8b0fd341c17 100644 --- a/Documentation/ABI/obsolete/sysfs-gpio +++ b/Documentation/ABI/obsolete/sysfs-gpio @@ -13,6 +13,8 @@ Description: GPIOs are identified as they are inside the kernel, using integers in the range 0..INT_MAX. See Documentation/admin-guide/gpio for more information. + :: + /sys/class/gpio /export ... asks the kernel to export a GPIO to userspace /unexport ... to return a GPIO to the kernel diff --git a/Documentation/admin-guide/abi-obsolete.rst b/Documentation/admin-guide/abi-obsolete.rst index cda9168445a5..d095867899c5 100644 --- a/Documentation/admin-guide/abi-obsolete.rst +++ b/Documentation/admin-guide/abi-obsolete.rst @@ -8,3 +8,4 @@ The description of the interface will document the reason why it is obsolete and when it can be expected to be removed. .. kernel-abi:: $srctree/Documentation/ABI/obsolete + :rst: diff --git a/Documentation/admin-guide/abi-removed.rst b/Documentation/admin-guide/abi-removed.rst index 497978fc9632..f7e9e43023c1 100644 --- a/Documentation/admin-guide/abi-removed.rst +++ b/Documentation/admin-guide/abi-removed.rst @@ -2,3 +2,4 @@ ABI removed symbols =================== .. kernel-abi:: $srctree/Documentation/ABI/removed + :rst: -- cgit v1.2.3 From 50865d041eead18b110e589674ff91ffcd243e7e Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:44 +0100 Subject: docs: abi-testing.rst: enable --rst-sources when building docs Now that ABI/testing documents were fixed, add --rst-sources to the ABI/testing too. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/9242473fd3df785565bb6084b1b814cc15074fb2.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/admin-guide/abi-testing.rst | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/admin-guide/abi-testing.rst b/Documentation/admin-guide/abi-testing.rst index 5c886fc50b9e..b205b16a72d0 100644 --- a/Documentation/admin-guide/abi-testing.rst +++ b/Documentation/admin-guide/abi-testing.rst @@ -17,3 +17,4 @@ name to the description of these interfaces, so that the kernel developers can easily notify them if any changes occur. .. kernel-abi:: $srctree/Documentation/ABI/testing + :rst: -- cgit v1.2.3 From 75442fb0ccaacddf1654a5304401a9f556c61004 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:45 +0100 Subject: docs: Kconfig/Makefile: add a check for broken ABI files The files under Documentation/ABI should follow the syntax as defined at Documentation/ABI/README. Allow checking if they're following the syntax by running the ABI parser script on COMPILE_TEST. With that, when there's a problem with a file under Documentation/ABI, it would produce a warning like: Warning: file ./Documentation/ABI/testing/sysfs-bus-pci-devices-aer_stats#14: What '/sys/bus/pci/devices//aer_stats/aer_rootport_total_err_cor' doesn't have a description Warning: file ./Documentation/ABI/testing/sysfs-bus-pci-devices-aer_stats#21: What '/sys/bus/pci/devices//aer_stats/aer_rootport_total_err_fatal' doesn't have a description Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/57a38de85cb4b548857207cf1fc1bf1ee08613c9.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/Kconfig | 10 ++++++++++ Documentation/Makefile | 5 +++++ lib/Kconfig.debug | 2 ++ scripts/get_abi.pl | 14 +++++++++++--- 4 files changed, 28 insertions(+), 3 deletions(-) diff --git a/Documentation/Kconfig b/Documentation/Kconfig index 66046fa1c341..e549a61f4d96 100644 --- a/Documentation/Kconfig +++ b/Documentation/Kconfig @@ -10,4 +10,14 @@ config WARN_MISSING_DOCUMENTS If unsure, select 'N'. +config WARN_ABI_ERRORS + bool "Warn if there are errors at ABI files" + depends on COMPILE_TEST + help + The files under Documentation/ABI should follow what's + described at Documentation/ABI/README. Yet, as they're manually + written, it would be possible that some of those files would + have errors that would break them for being parsed by + scripts/get_abi.pl. Add a check to verify them. + If unsure, select 'N'. diff --git a/Documentation/Makefile b/Documentation/Makefile index 6b12dd82f712..6a59a13d3c53 100644 --- a/Documentation/Makefile +++ b/Documentation/Makefile @@ -10,6 +10,11 @@ ifeq ($(CONFIG_WARN_MISSING_DOCUMENTS),y) $(shell $(srctree)/scripts/documentation-file-ref-check --warn) endif +# Check for broken ABI files +ifeq ($(CONFIG_WARN_ABI_ERRORS),y) +$(shell $(srctree)/scripts/get_abi.pl validate --dir $(srctree)/Documentation/ABI) +endif + # You can set these variables from the command line. SPHINXBUILD = sphinx-build SPHINXOPTS = diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index d7a7bc3b6098..c789b39ed527 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -2446,4 +2446,6 @@ config HYPERV_TESTING endmenu # "Kernel Testing and Coverage" +source "Documentation/Kconfig" + endmenu # Kernel hacking diff --git a/scripts/get_abi.pl b/scripts/get_abi.pl index ff4f9f82ecad..f6adf4b38a2b 100755 --- a/scripts/get_abi.pl +++ b/scripts/get_abi.pl @@ -50,7 +50,15 @@ my %symbols; sub parse_error($$$$) { my ($file, $ln, $msg, $data) = @_; - print STDERR "file $file#$ln: $msg at\n\t$data"; + $data =~ s/\s+$/\n/; + + print STDERR "Warning: file $file#$ln:\n\t$msg"; + + if ($data ne "") { + print STDERR ". Line\n\t\t$data"; + } else { + print STDERR "\n"; + } } # @@ -110,7 +118,7 @@ sub parse_abi { # Invalid, but it is a common mistake if ($new_tag eq "where") { - parse_error($file, $ln, "tag 'Where' is invalid. Should be 'What:' instead", $_); + parse_error($file, $ln, "tag 'Where' is invalid. Should be 'What:' instead", ""); $new_tag = "what"; } @@ -225,7 +233,7 @@ sub parse_abi { } # Everything else is error - parse_error($file, $ln, "Unexpected line:", $_); + parse_error($file, $ln, "Unexpected content", $_); } $data{$nametag}->{description} =~ s/^\n+// if ($data{$nametag}->{description}); if ($what) { -- cgit v1.2.3 From b1faa368bb202649259fd8afed78bcc31e58e293 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:46 +0100 Subject: docs: ABI: convert testing/configfs-acpi to ReST There are some problems with this file when a ReST content is produced. Fix it. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/f56daf94b80f1051438e8c787ba04552adb66e67.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/testing/configfs-acpi | 34 ++++++++++++++++++++++++--------- 1 file changed, 25 insertions(+), 9 deletions(-) diff --git a/Documentation/ABI/testing/configfs-acpi b/Documentation/ABI/testing/configfs-acpi index 4ab4e99aa863..c09b640c3cb1 100644 --- a/Documentation/ABI/testing/configfs-acpi +++ b/Documentation/ABI/testing/configfs-acpi @@ -14,7 +14,8 @@ Description: This group contains the configuration for user defined ACPI tables. The attributes of a user define table are: - aml - a binary attribute that the user can use to + aml + - a binary attribute that the user can use to fill in the ACPI aml definitions. Once the aml data is written to this file and the file is closed the table will be loaded and ACPI devices @@ -26,11 +27,26 @@ Description: The rest of the attributes are read-only and are valid only after the table has been loaded by filling the aml entry: - signature - ASCII table signature - length - length of table in bytes, including the header - revision - ACPI Specification minor version number - oem_id - ASCII OEM identification - oem_table_id - ASCII OEM table identification - oem_revision - OEM revision number - asl_compiler_id - ASCII ASL compiler vendor ID - asl_compiler_revision - ASL compiler version + signature + - ASCII table signature + + length + - length of table in bytes, including the header + + revision + - ACPI Specification minor version number + + oem_id + - ASCII OEM identification + + oem_table_id + - ASCII OEM table identification + + oem_revision + - OEM revision number + + asl_compiler_id + - ASCII ASL compiler vendor ID + + asl_compiler_revision + - ASL compiler version -- cgit v1.2.3 From 002a9c2f99c77677635d2ab2e906fc6ba3f7ab67 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:47 +0100 Subject: docs: ABI: fix syntax to be parsed using ReST notation There are a number of new changes at the ABI files that cause them to produce warnings when generating ABI output. Fix them. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/55a89f423cf122982c462d257722e44d6ece4b36.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- .../ABI/stable/sysfs-driver-firmware-zynqmp | 62 +++++++++------ Documentation/ABI/testing/debugfs-moxtet | 28 +++++-- Documentation/ABI/testing/debugfs-turris-mox-rwtm | 6 +- .../testing/sysfs-bus-event_source-devices-dfl_fme | 14 ++-- Documentation/ABI/testing/sysfs-bus-most | 6 +- Documentation/ABI/testing/sysfs-class-devfreq | 6 +- Documentation/ABI/testing/sysfs-class-ocxl | 14 ++-- Documentation/ABI/testing/sysfs-class-power-wilco | 18 +++-- Documentation/ABI/testing/sysfs-class-rnbd-client | 93 ++++++++++++---------- Documentation/ABI/testing/sysfs-class-rtrs-client | 23 +++--- .../testing/sysfs-devices-platform-stratix10-rsu | 10 +++ Documentation/ABI/testing/sysfs-driver-w1_therm | 75 +++++++++++------ Documentation/ABI/testing/sysfs-platform-dfl-fme | 14 +++- Documentation/ABI/testing/sysfs-platform-dptf | 11 ++- .../testing/sysfs-platform-intel-wmi-sbl-fw-update | 1 + .../ABI/testing/sysfs-platform-mellanox-bootctl | 50 +++++++----- Documentation/ABI/testing/sysfs-platform-wilco-ec | 1 + 17 files changed, 271 insertions(+), 161 deletions(-) diff --git a/Documentation/ABI/stable/sysfs-driver-firmware-zynqmp b/Documentation/ABI/stable/sysfs-driver-firmware-zynqmp index 00fa04c76ff3..f5724bb5b462 100644 --- a/Documentation/ABI/stable/sysfs-driver-firmware-zynqmp +++ b/Documentation/ABI/stable/sysfs-driver-firmware-zynqmp @@ -12,13 +12,15 @@ Description: resets. Three registers are used by the FSBL and other Xilinx software products: GLOBAL_GEN_STORAGE{4:6}. - Usage: - # cat /sys/devices/platform/firmware\:zynqmp-firmware/ggs0 - # echo > /sys/devices/platform/firmware\:zynqmp-firmware/ggs0 + Usage:: + + # cat /sys/devices/platform/firmware\:zynqmp-firmware/ggs0 + # echo > /sys/devices/platform/firmware\:zynqmp-firmware/ggs0 + + Example:: - Example: - # cat /sys/devices/platform/firmware\:zynqmp-firmware/ggs0 - # echo 0x1234ABCD > /sys/devices/platform/firmware\:zynqmp-firmware/ggs0 + # cat /sys/devices/platform/firmware\:zynqmp-firmware/ggs0 + # echo 0x1234ABCD > /sys/devices/platform/firmware\:zynqmp-firmware/ggs0 Users: Xilinx @@ -39,13 +41,15 @@ Description: software products: PERS_GLOB_GEN_STORAGE{4:7}. Register is reset only by a POR reset. - Usage: - # cat /sys/devices/platform/firmware\:zynqmp-firmware/pggs0 - # echo > /sys/devices/platform/firmware\:zynqmp-firmware/pggs0 + Usage:: + + # cat /sys/devices/platform/firmware\:zynqmp-firmware/pggs0 + # echo > /sys/devices/platform/firmware\:zynqmp-firmware/pggs0 + + Example:: - Example: - # cat /sys/devices/platform/firmware\:zynqmp-firmware/pggs0 - # echo 0x1234ABCD > /sys/devices/platform/firmware\:zynqmp-firmware/pggs0 + # cat /sys/devices/platform/firmware\:zynqmp-firmware/pggs0 + # echo 0x1234ABCD > /sys/devices/platform/firmware\:zynqmp-firmware/pggs0 Users: Xilinx @@ -61,23 +65,28 @@ Description: Following are available shutdown scopes(subtypes): - subsystem: Only the APU along with all of its peripherals + subsystem: + Only the APU along with all of its peripherals not used by other processing units will be shut down. This may result in the FPD power domain being shut down provided that no other processing unit uses FPD peripherals or DRAM. - ps_only: The complete PS will be shut down, including the + ps_only: + The complete PS will be shut down, including the RPU, PMU, etc. Only the PL domain (FPGA) remains untouched. - system: The complete system/device is shut down. + system: + The complete system/device is shut down. - Usage: - # cat /sys/devices/platform/firmware\:zynqmp-firmware/shutdown_scope - # echo > /sys/devices/platform/firmware\:zynqmp-firmware/shutdown_scope + Usage:: + + # cat /sys/devices/platform/firmware\:zynqmp-firmware/shutdown_scope + # echo > /sys/devices/platform/firmware\:zynqmp-firmware/shutdown_scope + + Example:: - Example: - # cat /sys/devices/platform/firmware\:zynqmp-firmware/shutdown_scope - # echo "subsystem" > /sys/devices/platform/firmware\:zynqmp-firmware/shutdown_scope + # cat /sys/devices/platform/firmware\:zynqmp-firmware/shutdown_scope + # echo "subsystem" > /sys/devices/platform/firmware\:zynqmp-firmware/shutdown_scope Users: Xilinx @@ -94,10 +103,13 @@ Description: system restart. Usage: - Set healthy bit - # echo 1 > /sys/devices/platform/firmware\:zynqmp-firmware/health_status - Unset healthy bit - # echo 0 > /sys/devices/platform/firmware\:zynqmp-firmware/health_status + Set healthy bit:: + + # echo 1 > /sys/devices/platform/firmware\:zynqmp-firmware/health_status + + Unset healthy bit:: + + # echo 0 > /sys/devices/platform/firmware\:zynqmp-firmware/health_status Users: Xilinx diff --git a/Documentation/ABI/testing/debugfs-moxtet b/Documentation/ABI/testing/debugfs-moxtet index 67b1717794d8..15dfaecb2892 100644 --- a/Documentation/ABI/testing/debugfs-moxtet +++ b/Documentation/ABI/testing/debugfs-moxtet @@ -5,10 +5,16 @@ Contact: Marek Behún Description: (R) Read input from the shift registers, in hexadecimal. Returns N+1 bytes, where N is the number of Moxtet connected modules. The first byte is from the CPU board itself. - Example: 101214 - 10: CPU board with SD card - 12: 2 = PCIe module, 1 = IRQ not active - 14: 4 = Peridot module, 1 = IRQ not active + + Example:: + + 101214 + + == ======================================= + 10 CPU board with SD card + 12 2 = PCIe module, 1 = IRQ not active + 14 4 = Peridot module, 1 = IRQ not active + == ======================================= What: /sys/kernel/debug/moxtet/output Date: March 2019 @@ -17,7 +23,13 @@ Contact: Marek Behún Description: (RW) Read last written value to the shift registers, in hexadecimal, or write values to the shift registers, also in hexadecimal. - Example: 0102 - 01: 01 was last written, or is to be written, to the - first module's shift register - 02: the same for second module + + Example:: + + 0102 + + == ================================================ + 01 01 was last written, or is to be written, to the + first module's shift register + 02 the same for second module + == ================================================ diff --git a/Documentation/ABI/testing/debugfs-turris-mox-rwtm b/Documentation/ABI/testing/debugfs-turris-mox-rwtm index 2b3255ee68fd..c8f7dadd591c 100644 --- a/Documentation/ABI/testing/debugfs-turris-mox-rwtm +++ b/Documentation/ABI/testing/debugfs-turris-mox-rwtm @@ -2,8 +2,10 @@ What: /sys/kernel/debug/turris-mox-rwtm/do_sign Date: Jun 2020 KernelVersion: 5.8 Contact: Marek Behún -Description: (W) Message to sign with the ECDSA private key stored in +Description: (W) + Message to sign with the ECDSA private key stored in device's OTP. The message must be exactly 64 bytes (since this is intended for SHA-512 hashes). - (R) The resulting signature, 136 bytes. This contains the R and + (R) + The resulting signature, 136 bytes. This contains the R and S values of the ECDSA signature, both in big-endian format. diff --git a/Documentation/ABI/testing/sysfs-bus-event_source-devices-dfl_fme b/Documentation/ABI/testing/sysfs-bus-event_source-devices-dfl_fme index c9278a3b3df1..63a32ddcb95e 100644 --- a/Documentation/ABI/testing/sysfs-bus-event_source-devices-dfl_fme +++ b/Documentation/ABI/testing/sysfs-bus-event_source-devices-dfl_fme @@ -8,13 +8,13 @@ Description: Read-only. Attribute group to describe the magic bits Each attribute under this group defines a bit range of the perf_event_attr.config. All supported attributes are listed - below. + below:: event = "config:0-11" - event ID evtype = "config:12-15" - event type portid = "config:16-23" - event source - For example, + For example:: fab_mmio_read = "event=0x06,evtype=0x02,portid=0xff" @@ -40,11 +40,11 @@ Description: Read-only. Attribute group to describe performance monitoring All supported performance monitoring events are listed below. - Basic events (evtype=0x00) + Basic events (evtype=0x00):: clock = "event=0x00,evtype=0x00,portid=0xff" - Cache events (evtype=0x01) + Cache events (evtype=0x01):: cache_read_hit = "event=0x00,evtype=0x01,portid=0xff" cache_read_miss = "event=0x01,evtype=0x01,portid=0xff" @@ -59,7 +59,7 @@ Description: Read-only. Attribute group to describe performance monitoring cache_rx_req_stall = "event=0x09,evtype=0x01,portid=0xff" cache_eviction = "event=0x0a,evtype=0x01,portid=0xff" - Fabric events (evtype=0x02) + Fabric events (evtype=0x02):: fab_pcie0_read = "event=0x00,evtype=0x02,portid=0xff" fab_pcie0_write = "event=0x01,evtype=0x02,portid=0xff" @@ -78,7 +78,7 @@ Description: Read-only. Attribute group to describe performance monitoring fab_port_mmio_read = "event=0x06,evtype=0x02,portid=?" fab_port_mmio_write = "event=0x07,evtype=0x02,portid=?" - VTD events (evtype=0x03) + VTD events (evtype=0x03):: vtd_port_read_transaction = "event=0x00,evtype=0x03,portid=?" vtd_port_write_transaction = "event=0x01,evtype=0x03,portid=?" @@ -88,7 +88,7 @@ Description: Read-only. Attribute group to describe performance monitoring vtd_port_devtlb_2m_fill = "event=0x05,evtype=0x03,portid=?" vtd_port_devtlb_1g_fill = "event=0x06,evtype=0x03,portid=?" - VTD SIP events (evtype=0x04) + VTD SIP events (evtype=0x04):: vtd_sip_iotlb_4k_hit = "event=0x00,evtype=0x04,portid=0xff" vtd_sip_iotlb_2m_hit = "event=0x01,evtype=0x04,portid=0xff" diff --git a/Documentation/ABI/testing/sysfs-bus-most b/Documentation/ABI/testing/sysfs-bus-most index ec0a603d804b..38cc03e408e7 100644 --- a/Documentation/ABI/testing/sysfs-bus-most +++ b/Documentation/ABI/testing/sysfs-bus-most @@ -235,7 +235,8 @@ KernelVersion: 4.15 Contact: Christian Gromm Description: This is to read back the configured direction of the channel. - The following strings will be accepted: + The following strings will be accepted:: + 'tx', 'rx' Users: @@ -246,7 +247,8 @@ KernelVersion: 4.15 Contact: Christian Gromm Description: This is to read back the configured data type of the channel. - The following strings will be accepted: + The following strings will be accepted:: + 'control', 'async', 'sync', diff --git a/Documentation/ABI/testing/sysfs-class-devfreq b/Documentation/ABI/testing/sysfs-class-devfreq index deefffb3bbe4..b8ebff4b1c4c 100644 --- a/Documentation/ABI/testing/sysfs-class-devfreq +++ b/Documentation/ABI/testing/sysfs-class-devfreq @@ -62,7 +62,8 @@ Description: driver should provide the list of available frequencies with its profile. If need to reset the statistics of devfreq behavior on a specific device, enter 0(zero) to 'trans_stat' - as following: + as following:: + echo 0 > /sys/class/devfreq/.../trans_stat What: /sys/class/devfreq/.../userspace/set_freq @@ -117,6 +118,7 @@ Description: This work timer is used by devfreq workqueue in order to monitor the device status such as utilization. The user can change the work timer on runtime according to their demand - as following: + as following:: + echo deferrable > /sys/class/devfreq/.../timer echo delayed > /sys/class/devfreq/.../timer diff --git a/Documentation/ABI/testing/sysfs-class-ocxl b/Documentation/ABI/testing/sysfs-class-ocxl index bf33f4fda58f..847a7edc3113 100644 --- a/Documentation/ABI/testing/sysfs-class-ocxl +++ b/Documentation/ABI/testing/sysfs-class-ocxl @@ -13,8 +13,8 @@ Description: read only where: ==== =============================================== - n: number of currently active contexts, for debug - max: maximum number of contexts supported by the AFU + n number of currently active contexts, for debug + max maximum number of contexts supported by the AFU ==== =============================================== What: /sys/class/ocxl//pp_mmio_size @@ -43,7 +43,9 @@ Contact: linuxppc-dev@lists.ozlabs.org Description: read/write Control whether the FPGA is reloaded on a link reset. Enabled through a vendor-specific logic block on the FPGA. - 0 Do not reload FPGA image from flash - 1 Reload FPGA image from flash - unavailable - The device does not support this capability + + =========== =========================================== + 0 Do not reload FPGA image from flash + 1 Reload FPGA image from flash + unavailable The device does not support this capability + =========== =========================================== diff --git a/Documentation/ABI/testing/sysfs-class-power-wilco b/Documentation/ABI/testing/sysfs-class-power-wilco index 84fde1d0ada0..82af180fcaab 100644 --- a/Documentation/ABI/testing/sysfs-class-power-wilco +++ b/Documentation/ABI/testing/sysfs-class-power-wilco @@ -4,17 +4,23 @@ KernelVersion: 5.2 Description: What charging algorithm to use: - Standard: Fully charges battery at a standard rate. - Adaptive: Battery settings adaptively optimized based on + Standard: + Fully charges battery at a standard rate. + Adaptive: + Battery settings adaptively optimized based on typical battery usage pattern. - Fast: Battery charges over a shorter period. - Trickle: Extends battery lifespan, intended for users who + Fast: + Battery charges over a shorter period. + Trickle: + Extends battery lifespan, intended for users who primarily use their Chromebook while connected to AC. - Custom: A low and high threshold percentage is specified. + Custom: + A low and high threshold percentage is specified. Charging begins when level drops below charge_control_start_threshold, and ceases when level is above charge_control_end_threshold. - Long Life: Customized charge rate for last longer battery life. + Long Life: + Customized charge rate for last longer battery life. On Wilco device this mode is pre-configured in the factory through EC's private PID. Swiching to a different mode will be denied by Wilco EC when Long Life mode is enabled. diff --git a/Documentation/ABI/testing/sysfs-class-rnbd-client b/Documentation/ABI/testing/sysfs-class-rnbd-client index c084f203b41e..00c0286733d4 100644 --- a/Documentation/ABI/testing/sysfs-class-rnbd-client +++ b/Documentation/ABI/testing/sysfs-class-rnbd-client @@ -5,62 +5,70 @@ Contact: Jack Wang Danil Kipnis Usage: echo "sessname= path=<[srcaddr,]dstaddr> - > [path=<[srcaddr,]dstaddr>] device_path= - > [access_mode=] > map_device - > - > addr ::= [ ip: | ip: | gid: ] + # cat /sys/class/rnbd-client/ctl/map_device + + > Usage: echo "sessname= path=<[srcaddr,]dstaddr> + > [path=<[srcaddr,]dstaddr>] device_path= + > [access_mode=] > map_device + > + > addr ::= [ ip: | ip: | gid: ] What: /sys/class/rnbd-client/ctl/map_device Date: Feb 2020 KernelVersion: 5.7 Contact: Jack Wang Danil Kipnis -Description: Expected format is the following: +Description: Expected format is the following:: - sessname= - path=<[srcaddr,]dstaddr> [path=<[srcaddr,]dstaddr> ...] - device_path= - [access_mode=] + sessname= + path=<[srcaddr,]dstaddr> [path=<[srcaddr,]dstaddr> ...] + device_path= + [access_mode=] Where: - sessname: accepts a string not bigger than 256 chars, which identifies - a given session on the client and on the server. - I.e. "clt_hostname-srv_hostname" could be a natural choice. + sessname: + accepts a string not bigger than 256 chars, which identifies + a given session on the client and on the server. + I.e. "clt_hostname-srv_hostname" could be a natural choice. + + path: + describes a connection between the client and the server by + specifying destination and, when required, the source address. + The addresses are to be provided in the following format:: - path: describes a connection between the client and the server by - specifying destination and, when required, the source address. - The addresses are to be provided in the following format: + ip: + ip: + gid: - ip: - ip: - gid: + for example:: - for example: + path=ip:10.0.0.66 - path=ip:10.0.0.66 The single addr is treated as the destination. The connection will be established to this server from any client IP address. - path=ip:10.0.0.66,ip:10.0.1.66 + :: + + path=ip:10.0.0.66,ip:10.0.1.66 + First addr is the source address and the second is the destination. If multiple "path=" options are specified multiple connection will be established and data will be sent according to the selected multipath policy (see RTRS mp_policy sysfs entry description). - device_path: Path to the block device on the server side. Path is specified - relative to the directory on server side configured in the - 'dev_search_path' module parameter of the rnbd_server. - The rnbd_server prepends the received from client - with and tries to open the - / block device. On success, - a /dev/rnbd device file, a /sys/block/rnbd_client/rnbd/ - directory and an entry in /sys/class/rnbd-client/ctl/devices - will be created. + device_path: + Path to the block device on the server side. Path is specified + relative to the directory on server side configured in the + 'dev_search_path' module parameter of the rnbd_server. + The rnbd_server prepends the received from client + with and tries to open the + / block device. On success, + a /dev/rnbd device file, a /sys/block/rnbd_client/rnbd/ + directory and an entry in /sys/class/rnbd-client/ctl/devices + will be created. If 'dev_search_path' contains '%SESSNAME%', then each session can have different devices namespace, e.g. server was configured with @@ -68,11 +76,12 @@ Description: Expected format is the following: client has this string "sessname=blya device_path=sda", then server will try to open: /run/rnbd-devs/blya/sda. - access_mode: the access_mode parameter specifies if the device is to be - mapped as "ro" read-only or "rw" read-write. The server allows - a device to be exported in rw mode only once. The "migration" - access mode has to be specified if a second mapping in read-write - mode is desired. + access_mode: + the access_mode parameter specifies if the device is to be + mapped as "ro" read-only or "rw" read-write. The server allows + a device to be exported in rw mode only once. The "migration" + access mode has to be specified if a second mapping in read-write + mode is desired. By default "rw" is used. @@ -91,7 +100,7 @@ Description: Expected format is the following: is the same as the device name. By extracting the last part of the path the path to the device /dev/ can be build. - o /dev/block/$(cat /sys/class/rnbd-client/ctl/devices//dev) + * /dev/block/$(cat /sys/class/rnbd-client/ctl/devices//dev) How to find the of the device is described on the next section. @@ -106,6 +115,6 @@ Description: For each device mapped on the client a new symbolic link is created The of each device is created as follows: - If the 'device_path' provided during mapping contains slashes ("/"), - they are replaced by exclamation mark ("!") and used as as the - . Otherwise, the will be the same as the - "device_path" provided. + they are replaced by exclamation mark ("!") and used as as the + . Otherwise, the will be the same as the + "device_path" provided. diff --git a/Documentation/ABI/testing/sysfs-class-rtrs-client b/Documentation/ABI/testing/sysfs-class-rtrs-client index e7e718db8941..0f7165aab251 100644 --- a/Documentation/ABI/testing/sysfs-class-rtrs-client +++ b/Documentation/ABI/testing/sysfs-class-rtrs-client @@ -10,10 +10,10 @@ Date: Feb 2020 KernelVersion: 5.7 Contact: Jack Wang Danil Kipnis Description: RW, adds a new path (connection) to an existing session. Expected format is the - following: + following:: - <[source addr,]destination addr> - *addr ::= [ ip: | gid: ] + <[source addr,]destination addr> + *addr ::= [ ip: | gid: ] What: /sys/class/rtrs-client//max_reconnect_attempts Date: Feb 2020 @@ -29,10 +29,10 @@ Contact: Jack Wang Danil Kipnis /paths/ Date: Feb 2020 @@ -109,8 +109,11 @@ Description: RTRS expects that each HCA IRQ is pinned to a separate CPU. If it's not the case, the processing of an I/O response could be processed on a different CPU than where it was originally submitted. This file shows how many interrupts where generated on a non expected CPU. - "from:" is the CPU on which the IRQ was expected, but not generated. - "to:" is the CPU on which the IRQ was generated, but not expected. + + "from:" + is the CPU on which the IRQ was expected, but not generated. + "to:" + is the CPU on which the IRQ was generated, but not expected. What: /sys/class/rtrs-client//paths//stats/reconnects Date: Feb 2020 @@ -125,7 +128,7 @@ Date: Feb 2020 KernelVersion: 5.7 Contact: Jack Wang Danil Kipnis Description: Contains statistics regarding rdma operations and inflight operations. - The output consists of 6 values: + The output consists of 6 values:: - \ - + \ + diff --git a/Documentation/ABI/testing/sysfs-devices-platform-stratix10-rsu b/Documentation/ABI/testing/sysfs-devices-platform-stratix10-rsu index a8daceb4a956..ee253b033280 100644 --- a/Documentation/ABI/testing/sysfs-devices-platform-stratix10-rsu +++ b/Documentation/ABI/testing/sysfs-devices-platform-stratix10-rsu @@ -102,6 +102,8 @@ Description: b[15:0] inform firmware the current software execution stage. + + == =========================================== 0 the first stage bootloader didn't run or didn't reach the point of launching second stage bootloader. @@ -111,21 +113,29 @@ Description: 2 both first and second stage bootloader ran and the operating system launch was attempted. + == =========================================== b[16] + == =========================================== 1 firmware to reset current image retry counter. 0 no action. + == =========================================== b[17] + == =========================================== 1 firmware to clear RSU log 0 no action. + == =========================================== b[18] this is negative logic + + == =========================================== 1 no action 0 firmware record the notify code defined in b[15:0]. + == =========================================== What: /sys/devices/platform/stratix10-rsu.0/dcmf0 Date: June 2020 diff --git a/Documentation/ABI/testing/sysfs-driver-w1_therm b/Documentation/ABI/testing/sysfs-driver-w1_therm index 8873bbb075cb..6a37dc33ffdb 100644 --- a/Documentation/ABI/testing/sysfs-driver-w1_therm +++ b/Documentation/ABI/testing/sysfs-driver-w1_therm @@ -22,8 +22,10 @@ Description: device data to its embedded EEPROM, either restore data embedded in device EEPROM. Be aware that devices support limited EEPROM writing cycles (typical 50k) + * 'save': save device RAM to EEPROM * 'restore': restore EEPROM data in device RAM + Users: any user space application which wants to communicate with w1_term device @@ -33,9 +35,11 @@ Date: May 2020 Contact: Akira Shimahara Description: (RO) return the power status by asking the device + * '0': device parasite powered * '1': device externally powered * '-xx': xx is kernel error when reading power status + Users: any user space application which wants to communicate with w1_term device @@ -49,10 +53,12 @@ Description: will be changed only in device RAM, so it will be cleared when power is lost. Trigger a 'save' to EEPROM command to keep values after power-on. Read or write are : + * '9..14': device resolution in bit - or resolution to set in bit + or resolution to set in bit * '-xx': xx is kernel error when reading the resolution * Anything else: do nothing + Some DS18B20 clones are fixed in 12-bit resolution, so the actual resolution is read back from the chip and verified. Error is reported if the results differ. @@ -65,16 +71,18 @@ Date: May 2020 Contact: Akira Shimahara Description: (RO) return the temperature in 1/1000 degC. + * If a bulk read has been triggered, it will directly - return the temperature computed when the bulk read - occurred, if available. If not yet available, nothing - is returned (a debug kernel message is sent), you - should retry later on. + return the temperature computed when the bulk read + occurred, if available. If not yet available, nothing + is returned (a debug kernel message is sent), you + should retry later on. * If no bulk read has been triggered, it will trigger - a conversion and send the result. Note that the - conversion duration depend on the resolution (if - device support this feature). It takes 94ms in 9bits - resolution, 750ms for 12bits. + a conversion and send the result. Note that the + conversion duration depend on the resolution (if + device support this feature). It takes 94ms in 9bits + resolution, 750ms for 12bits. + Users: any user space application which wants to communicate with w1_term device @@ -86,12 +94,14 @@ Description: (RW) return the temperature in 1/1000 degC. *read*: return 2 lines with the hexa output data sent on the bus, return the CRC check and temperature in 1/1000 degC - *write* : + *write*: + * '0' : save the 2 or 3 bytes to the device EEPROM - (i.e. TH, TL and config register) + (i.e. TH, TL and config register) * '9..14' : set the device resolution in RAM - (if supported) + (if supported) * Anything else: do nothing + refer to Documentation/w1/slaves/w1_therm.rst for detailed information. Users: any user space application which wants to communicate with @@ -103,14 +113,21 @@ Date: May 2020 Contact: Akira Shimahara Description: (RW) trigger a bulk read conversion. read the status + *read*: - * '-1': conversion in progress on at least 1 sensor - * '1' : conversion complete but at least one sensor + * '-1': + conversion in progress on at least 1 sensor + * '1' : + conversion complete but at least one sensor value has not been read yet - * '0' : no bulk operation. Reading temperature will + * '0' : + no bulk operation. Reading temperature will trigger a conversion on each device - *write*: 'trigger': trigger a bulk read on all supporting + + *write*: + 'trigger': trigger a bulk read on all supporting devices on the bus + Note that if a bulk read is sent but one sensor is not read immediately, the next access to temperature on this device will return the temperature measured at the time of issue @@ -128,14 +145,19 @@ Description: reset to default (datasheet) conversion time for a new resolution. - *read*: Actual conversion time in milliseconds. *write*: - '0': Set the default conversion time from the datasheet. - '1': Measure and set the conversion time. Make a single + *read*: + Actual conversion time in milliseconds. + + *write*: + * '0': + Set the default conversion time from the datasheet. + * '1': + Measure and set the conversion time. Make a single temperature conversion, measure an actual value. Increase it by 20% for temperature range. A new conversion time can be obtained by reading this same attribute. - other positive value: + * other positive value: Set the conversion time in milliseconds. Users: An application using the w1_term device @@ -148,16 +170,21 @@ Description: (RW) Control optional driver settings. Bit masks to read/write (bitwise OR): - 1: Enable check for conversion success. If byte 6 of + == ============================================================ + 1 Enable check for conversion success. If byte 6 of scratchpad memory is 0xC after conversion, and temperature reads 85.00 (powerup value) or 127.94 (insufficient power) - return a conversion error. - 2: Enable poll for conversion completion. Generate read cycles + 2 Enable poll for conversion completion. Generate read cycles after the conversion start and wait for 1's. In parasite power mode this feature is not available. + == ============================================================ + + *read*: + Currently selected features. - *read*: Currently selected features. - *write*: Select features. + *write*: + Select features. Users: An application using the w1_term device diff --git a/Documentation/ABI/testing/sysfs-platform-dfl-fme b/Documentation/ABI/testing/sysfs-platform-dfl-fme index 3683cb1cdc3d..d6ab34e81b9b 100644 --- a/Documentation/ABI/testing/sysfs-platform-dfl-fme +++ b/Documentation/ABI/testing/sysfs-platform-dfl-fme @@ -113,8 +113,11 @@ KernelVersion: 5.5 Contact: Wu Hao Description: Read-Only. Read this file to get the name of hwmon device, it supports values: - 'dfl_fme_thermal' - thermal hwmon device name - 'dfl_fme_power' - power hwmon device name + + ================= ========================= + 'dfl_fme_thermal' thermal hwmon device name + 'dfl_fme_power' power hwmon device name + ================= ========================= What: /sys/bus/platform/devices/dfl-fme.0/hwmon/hwmonX/temp1_input Date: October 2019 @@ -169,8 +172,11 @@ KernelVersion: 5.5 Contact: Wu Hao Description: Read-Only. Read this file to get the policy of hardware threshold1 (see 'temp1_max'). It only supports two values (policies): - 0 - AP2 state (90% throttling) - 1 - AP1 state (50% throttling) + + == ========================== + 0 AP2 state (90% throttling) + 1 AP1 state (50% throttling) + == ========================== What: /sys/bus/platform/devices/dfl-fme.0/hwmon/hwmonX/power1_input Date: October 2019 diff --git a/Documentation/ABI/testing/sysfs-platform-dptf b/Documentation/ABI/testing/sysfs-platform-dptf index 2cbc660d163b..141834342a4d 100644 --- a/Documentation/ABI/testing/sysfs-platform-dptf +++ b/Documentation/ABI/testing/sysfs-platform-dptf @@ -27,12 +27,15 @@ KernelVersion: v4.10 Contact: linux-acpi@vger.kernel.org Description: (RO) Display the platform power source + + ========= ============================ bits[3:0] Current power source - 0x00 = DC - 0x01 = AC - 0x02 = USB - 0x03 = Wireless Charger + - 0x00 = DC + - 0x01 = AC + - 0x02 = USB + - 0x03 = Wireless Charger bits[7:4] Power source sequence number + ========= ============================ What: /sys/bus/platform/devices/INT3407:00/dptf_power/battery_steady_power Date: Jul, 2016 diff --git a/Documentation/ABI/testing/sysfs-platform-intel-wmi-sbl-fw-update b/Documentation/ABI/testing/sysfs-platform-intel-wmi-sbl-fw-update index 5aa618987cad..02ae1e9bbfc8 100644 --- a/Documentation/ABI/testing/sysfs-platform-intel-wmi-sbl-fw-update +++ b/Documentation/ABI/testing/sysfs-platform-intel-wmi-sbl-fw-update @@ -8,5 +8,6 @@ Description: of 0 and userspace can signal SBL to update firmware, on next reboot, by writing a value of 1. There are two available states: + * 0 -> Skip firmware update while rebooting * 1 -> Attempt firmware update on next reboot diff --git a/Documentation/ABI/testing/sysfs-platform-mellanox-bootctl b/Documentation/ABI/testing/sysfs-platform-mellanox-bootctl index 401d202f478b..e79ca22e2f45 100644 --- a/Documentation/ABI/testing/sysfs-platform-mellanox-bootctl +++ b/Documentation/ABI/testing/sysfs-platform-mellanox-bootctl @@ -5,10 +5,13 @@ Contact: "Liming Sun " Description: The Life-cycle state of the SoC, which could be one of the following values. - Production - Production state and can be updated to secure - GA Secured - Secure chip and not able to change state - GA Non-Secured - Non-Secure chip and not able to change state - RMA - Return Merchandise Authorization + + ============== ============================================= + Production Production state and can be updated to secure + GA Secured Secure chip and not able to change state + GA Non-Secured Non-Secure chip and not able to change state + RMA Return Merchandise Authorization + ============== ============================================= What: /sys/bus/platform/devices/MLNXBF04:00/post_reset_wdog Date: Oct 2019 @@ -25,10 +28,13 @@ KernelVersion: 5.5 Contact: "Liming Sun " Description: The source of the boot stream for the next reset. It could be - one of the following values. - external - boot from external source (USB or PCIe) - emmc - boot from the onchip eMMC - emmc_legacy - boot from the onchip eMMC in legacy (slow) mode + one of the following values: + + =========== =============================================== + external boot from external source (USB or PCIe) + emmc boot from the onchip eMMC + emmc_legacy boot from the onchip eMMC in legacy (slow) mode + =========== =============================================== What: /sys/bus/platform/devices/MLNXBF04:00/second_reset_action Date: Oct 2019 @@ -38,11 +44,14 @@ Description: Update the source of the boot stream after next reset. It could be one of the following values and will be applied after next reset. - external - boot from external source (USB or PCIe) - emmc - boot from the onchip eMMC - emmc_legacy - boot from the onchip eMMC in legacy (slow) mode - swap_emmc - swap the primary / secondary boot partition - none - cancel the action + + =========== =============================================== + external boot from external source (USB or PCIe) + emmc boot from the onchip eMMC + emmc_legacy boot from the onchip eMMC in legacy (slow) mode + swap_emmc swap the primary / secondary boot partition + none cancel the action + =========== =============================================== What: /sys/bus/platform/devices/MLNXBF04:00/secure_boot_fuse_state Date: Oct 2019 @@ -50,9 +59,12 @@ KernelVersion: 5.5 Contact: "Liming Sun " Description: The state of eFuse versions with the following values. - InUse - burnt, valid and currently in use - Used - burnt and valid - Free - not burnt and free to use - Skipped - not burnt but not free (skipped) - Wasted - burnt and invalid - Invalid - not burnt but marked as valid (error state). + + ======= =============================================== + InUse burnt, valid and currently in use + Used burnt and valid + Free not burnt and free to use + Skipped not burnt but not free (skipped) + Wasted burnt and invalid + Invalid not burnt but marked as valid (error state). + ======= =============================================== diff --git a/Documentation/ABI/testing/sysfs-platform-wilco-ec b/Documentation/ABI/testing/sysfs-platform-wilco-ec index 5f60b184a5a5..4439d0644091 100644 --- a/Documentation/ABI/testing/sysfs-platform-wilco-ec +++ b/Documentation/ABI/testing/sysfs-platform-wilco-ec @@ -39,6 +39,7 @@ Description: which affects charging via the special USB PowerShare port (marked with a small lightning bolt or battery icon) when in low power states: + - In S0, the port will always provide power. - In S0ix, if usb_charge is enabled, then power will be supplied to the port when on AC or if battery is > 50%. -- cgit v1.2.3 From 97f26cab455e1868edb0b74ce714dcb52444b63b Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:48 +0100 Subject: docs: ABI: vdso: use the right format for ABI This ABI is not following the format described at ABI/README. Use it, filling in the blanks with the git log that added it, and using the current e-mail from Andy. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/28c7cf3a71e15fb7499b70ec8f38c2efaaf4add2.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/stable/vdso | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/Documentation/ABI/stable/vdso b/Documentation/ABI/stable/vdso index 73ed1240a5c0..951838d42781 100644 --- a/Documentation/ABI/stable/vdso +++ b/Documentation/ABI/stable/vdso @@ -1,3 +1,9 @@ +What: vDSO +Date: July 2011 +KernelVersion: 3.0 +Contact: Andy Lutomirski +Description: + On some architectures, when the kernel loads any userspace program it maps an ELF DSO into that program's address space. This DSO is called the vDSO and it often contains useful and highly-optimized alternatives -- cgit v1.2.3 From ed8c39d43983d19f181ff47af0374c4e252f84bd Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:49 +0100 Subject: docs: ABI: sysfs-bus-nvdimm: use the right format for ABI This ABI is not following the format described at ABI/README. Use it, filling in the blanks with the git log that added it, and using the current e-mail from Dan. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/9ec379cbf6dcf65ce3039c3671baf7bcaea532f4.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/testing/sysfs-bus-nvdimm | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/Documentation/ABI/testing/sysfs-bus-nvdimm b/Documentation/ABI/testing/sysfs-bus-nvdimm index d64380262be8..bff84a16812a 100644 --- a/Documentation/ABI/testing/sysfs-bus-nvdimm +++ b/Documentation/ABI/testing/sysfs-bus-nvdimm @@ -1,2 +1,8 @@ +What: nvdimm +Date: July 2020 +KernelVersion: 5.8 +Contact: Dan Williams +Description: + The libnvdimm sub-system implements a common sysfs interface for platform nvdimm resources. See Documentation/driver-api/nvdimm/. -- cgit v1.2.3 From 54a19b4d3fe0fa0a31b46cd60951e8177cac25fa Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:50 +0100 Subject: docs: ABI: cleanup several ABI documents There are some ABI documents that, while they don't generate any warnings, they have issues when parsed by get_abi.pl script on its output result. Address them, in order to provide a clean output. Reviewed-by: Tom Rix # for fpga-manager Reviewed-By: Kajol Jain # for sysfs-bus-event_source-devices-hv_gpci and sysfs-bus-event_source-devices-hv_24x7 Acked-by: Jonathan Cameron #for IIO Acked-by: Oded Gabbay # for Habanalabs Acked-by: Vaibhav Jain # for sysfs-bus-papr-pmem Acked-by: Cezary Rojewski # for catpt Acked-by: Suzuki K Poulose Acked-by: Ilya Dryomov # for rbd Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/5bc78e5b68ed1e9e39135173857cb2e753be868f.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/obsolete/sysfs-class-dax | 8 +- .../ABI/obsolete/sysfs-driver-hid-roccat-pyra | 3 + Documentation/ABI/removed/devfs | 1 + Documentation/ABI/removed/raw1394 | 1 + Documentation/ABI/removed/sysfs-class-rfkill | 2 +- Documentation/ABI/removed/video1394 | 1 + Documentation/ABI/stable/firewire-cdev | 65 +++--- Documentation/ABI/stable/sysfs-acpi-pmprofile | 4 +- Documentation/ABI/stable/sysfs-bus-w1 | 1 + Documentation/ABI/stable/sysfs-class-tpm | 4 +- Documentation/ABI/stable/sysfs-driver-speakup | 4 + Documentation/ABI/testing/configfs-most | 135 ++++++++---- Documentation/ABI/testing/configfs-usb-gadget-ecm | 12 +- Documentation/ABI/testing/configfs-usb-gadget-eem | 10 +- .../ABI/testing/configfs-usb-gadget-loopback | 6 +- .../ABI/testing/configfs-usb-gadget-mass-storage | 18 +- Documentation/ABI/testing/configfs-usb-gadget-midi | 14 +- .../ABI/testing/configfs-usb-gadget-printer | 6 +- .../ABI/testing/configfs-usb-gadget-sourcesink | 18 +- .../ABI/testing/configfs-usb-gadget-subset | 10 +- Documentation/ABI/testing/configfs-usb-gadget-uac2 | 14 +- Documentation/ABI/testing/configfs-usb-gadget-uvc | 2 +- Documentation/ABI/testing/debugfs-cec-error-inj | 2 +- .../ABI/testing/debugfs-driver-habanalabs | 12 +- Documentation/ABI/testing/debugfs-pfo-nx-crypto | 28 +-- Documentation/ABI/testing/debugfs-pktcdvd | 2 +- Documentation/ABI/testing/debugfs-turris-mox-rwtm | 10 +- Documentation/ABI/testing/debugfs-wilco-ec | 21 +- Documentation/ABI/testing/dell-smbios-wmi | 32 +-- Documentation/ABI/testing/gpio-cdev | 13 +- Documentation/ABI/testing/procfs-diskstats | 6 +- Documentation/ABI/testing/procfs-smaps_rollup | 48 ++--- Documentation/ABI/testing/pstore | 19 +- Documentation/ABI/testing/sysfs-block-rnbd | 4 +- Documentation/ABI/testing/sysfs-bus-acpi | 1 + .../ABI/testing/sysfs-bus-coresight-devices-etb10 | 5 +- Documentation/ABI/testing/sysfs-bus-css | 3 + Documentation/ABI/testing/sysfs-bus-dfl | 2 + .../testing/sysfs-bus-event_source-devices-hv_24x7 | 6 +- .../testing/sysfs-bus-event_source-devices-hv_gpci | 7 +- Documentation/ABI/testing/sysfs-bus-fcoe | 68 +++--- Documentation/ABI/testing/sysfs-bus-fsl-mc | 12 +- .../ABI/testing/sysfs-bus-i2c-devices-fsa9480 | 26 ++- Documentation/ABI/testing/sysfs-bus-i3c | 2 + Documentation/ABI/testing/sysfs-bus-iio | 19 +- Documentation/ABI/testing/sysfs-bus-iio-adc-hi8435 | 5 + Documentation/ABI/testing/sysfs-bus-iio-adc-stm32 | 3 + .../ABI/testing/sysfs-bus-iio-distance-srf08 | 7 +- .../ABI/testing/sysfs-bus-iio-frequency-ad9523 | 2 + .../ABI/testing/sysfs-bus-iio-frequency-adf4371 | 10 +- .../ABI/testing/sysfs-bus-iio-health-afe440x | 12 +- .../ABI/testing/sysfs-bus-iio-light-isl29018 | 6 +- .../ABI/testing/sysfs-bus-intel_th-devices-gth | 11 +- Documentation/ABI/testing/sysfs-bus-papr-pmem | 23 ++- Documentation/ABI/testing/sysfs-bus-pci | 22 +- .../ABI/testing/sysfs-bus-pci-devices-catpt | 1 + .../ABI/testing/sysfs-bus-pci-drivers-ehci_hcd | 4 +- Documentation/ABI/testing/sysfs-bus-rbd | 37 ++-- Documentation/ABI/testing/sysfs-bus-siox | 3 + Documentation/ABI/testing/sysfs-bus-thunderbolt | 18 +- Documentation/ABI/testing/sysfs-bus-usb | 2 + .../testing/sysfs-class-backlight-driver-lm3533 | 26 ++- Documentation/ABI/testing/sysfs-class-bdi | 1 - Documentation/ABI/testing/sysfs-class-chromeos | 15 +- Documentation/ABI/testing/sysfs-class-cxl | 8 +- Documentation/ABI/testing/sysfs-class-devlink | 30 +-- Documentation/ABI/testing/sysfs-class-extcon | 34 +-- Documentation/ABI/testing/sysfs-class-fpga-manager | 5 +- Documentation/ABI/testing/sysfs-class-gnss | 2 + Documentation/ABI/testing/sysfs-class-led | 1 + .../ABI/testing/sysfs-class-led-driver-el15203000 | 30 +-- .../ABI/testing/sysfs-class-led-driver-lm3533 | 44 ++-- Documentation/ABI/testing/sysfs-class-led-flash | 27 ++- .../ABI/testing/sysfs-class-led-trigger-netdev | 7 + .../ABI/testing/sysfs-class-led-trigger-usbport | 1 + Documentation/ABI/testing/sysfs-class-leds-gt683r | 8 +- Documentation/ABI/testing/sysfs-class-net | 61 ++++-- Documentation/ABI/testing/sysfs-class-net-cdc_ncm | 6 +- Documentation/ABI/testing/sysfs-class-net-phydev | 2 + Documentation/ABI/testing/sysfs-class-pktcdvd | 38 ++-- Documentation/ABI/testing/sysfs-class-power | 12 +- Documentation/ABI/testing/sysfs-class-power-mp2629 | 1 + .../ABI/testing/sysfs-class-power-twl4030 | 4 +- Documentation/ABI/testing/sysfs-class-rapidio | 48 +++-- Documentation/ABI/testing/sysfs-class-regulator | 36 ++-- Documentation/ABI/testing/sysfs-class-remoteproc | 14 +- .../sysfs-class-rtc-rtc0-device-rtc_calibration | 1 + Documentation/ABI/testing/sysfs-class-uwb_rc | 13 +- Documentation/ABI/testing/sysfs-class-watchdog | 7 +- Documentation/ABI/testing/sysfs-dev | 7 +- Documentation/ABI/testing/sysfs-devices-mapping | 41 ++-- Documentation/ABI/testing/sysfs-devices-memory | 15 +- .../testing/sysfs-devices-platform-_UDC_-gadget | 10 +- .../ABI/testing/sysfs-devices-platform-ipmi | 52 ++--- Documentation/ABI/testing/sysfs-devices-system-cpu | 4 +- Documentation/ABI/testing/sysfs-driver-hid-lenovo | 10 + Documentation/ABI/testing/sysfs-driver-hid-ntrig | 13 +- .../ABI/testing/sysfs-driver-hid-roccat-kone | 19 ++ Documentation/ABI/testing/sysfs-driver-hid-wiimote | 1 + .../ABI/testing/sysfs-driver-input-exc3000 | 2 + .../ABI/testing/sysfs-driver-jz4780-efuse | 6 +- Documentation/ABI/testing/sysfs-driver-pciback | 6 +- Documentation/ABI/testing/sysfs-driver-ufs | 228 ++++++++++++++++----- Documentation/ABI/testing/sysfs-driver-w1_ds28e17 | 3 + Documentation/ABI/testing/sysfs-firmware-acpi | 16 +- Documentation/ABI/testing/sysfs-firmware-efi-esrt | 28 ++- .../ABI/testing/sysfs-firmware-efi-runtime-map | 14 +- .../ABI/testing/sysfs-firmware-qemu_fw_cfg | 20 +- Documentation/ABI/testing/sysfs-firmware-sfi | 6 +- Documentation/ABI/testing/sysfs-firmware-sgi_uv | 6 +- Documentation/ABI/testing/sysfs-fs-f2fs | 48 +++-- Documentation/ABI/testing/sysfs-kernel-mm-ksm | 5 +- Documentation/ABI/testing/sysfs-kernel-slab | 3 + Documentation/ABI/testing/sysfs-module | 17 +- .../ABI/testing/sysfs-platform-dell-laptop | 10 +- .../ABI/testing/sysfs-platform-dell-smbios | 4 +- .../ABI/testing/sysfs-platform-i2c-demux-pinctrl | 4 +- Documentation/ABI/testing/sysfs-platform-kim | 1 + .../ABI/testing/sysfs-platform-phy-rcar-gen3-usb2 | 10 +- .../ABI/testing/sysfs-platform-renesas_usb3 | 10 +- Documentation/ABI/testing/sysfs-power | 21 +- Documentation/ABI/testing/sysfs-profiling | 2 +- Documentation/ABI/testing/sysfs-wusb_cbaf | 3 +- Documentation/ABI/testing/usb-charger-uevent | 84 ++++---- Documentation/ABI/testing/usb-uevent | 32 +-- scripts/get_abi.pl | 2 - 126 files changed, 1327 insertions(+), 771 deletions(-) diff --git a/Documentation/ABI/obsolete/sysfs-class-dax b/Documentation/ABI/obsolete/sysfs-class-dax index 2cb9fc5e8bd1..0faf1354cd05 100644 --- a/Documentation/ABI/obsolete/sysfs-class-dax +++ b/Documentation/ABI/obsolete/sysfs-class-dax @@ -8,11 +8,11 @@ Description: Device DAX is the device-centric analogue of Filesystem system. Device DAX is strict, precise and predictable. Specifically this interface: - 1/ Guarantees fault granularity with respect to a given - page size (pte, pmd, or pud) set at configuration time. + 1. Guarantees fault granularity with respect to a given + page size (pte, pmd, or pud) set at configuration time. - 2/ Enforces deterministic behavior by being strict about - what fault scenarios are supported. + 2. Enforces deterministic behavior by being strict about + what fault scenarios are supported. The /sys/class/dax/ interface enumerates all the device-dax instances in the system. The ABI is diff --git a/Documentation/ABI/obsolete/sysfs-driver-hid-roccat-pyra b/Documentation/ABI/obsolete/sysfs-driver-hid-roccat-pyra index 5d41ebadf15e..66545c587a64 100644 --- a/Documentation/ABI/obsolete/sysfs-driver-hid-roccat-pyra +++ b/Documentation/ABI/obsolete/sysfs-driver-hid-roccat-pyra @@ -7,10 +7,13 @@ Description: It is possible to switch the cpi setting of the mouse with the setting reported by the mouse. This number has to be further processed to receive the real dpi value: + ===== ==== VALUE DPI + ===== ==== 1 400 2 800 4 1600 + ===== ==== This file is readonly. Has never been used. If bookkeeping is done, it's done in userland tools. diff --git a/Documentation/ABI/removed/devfs b/Documentation/ABI/removed/devfs index 0020c49933c4..24fb35adf277 100644 --- a/Documentation/ABI/removed/devfs +++ b/Documentation/ABI/removed/devfs @@ -5,6 +5,7 @@ Description: devfs has been unmaintained for a number of years, has unfixable races, contains a naming policy within the kernel that is against the LSB, and can be replaced by using udev. + The files fs/devfs/*, include/linux/devfs_fs*.h were removed, along with the assorted devfs function calls throughout the kernel tree. diff --git a/Documentation/ABI/removed/raw1394 b/Documentation/ABI/removed/raw1394 index ec333e676322..9ec7ec493920 100644 --- a/Documentation/ABI/removed/raw1394 +++ b/Documentation/ABI/removed/raw1394 @@ -7,6 +7,7 @@ Description: to implement sensible device security policies, and its low level of abstraction that required userspace clients to duplicate much of the kernel's ieee1394 core functionality. + Replaced by /dev/fw*, i.e. the ABI of firewire-core. diff --git a/Documentation/ABI/removed/sysfs-class-rfkill b/Documentation/ABI/removed/sysfs-class-rfkill index 9c08c7f98ffb..f25174eafd55 100644 --- a/Documentation/ABI/removed/sysfs-class-rfkill +++ b/Documentation/ABI/removed/sysfs-class-rfkill @@ -10,4 +10,4 @@ Description: This file was deprecated because there no longer was a way to claim just control over a single rfkill instance. This file was scheduled to be removed in 2012, and was removed in 2016. -Values: 0: Kernel handles events +Values: 0: Kernel handles events diff --git a/Documentation/ABI/removed/video1394 b/Documentation/ABI/removed/video1394 index c39c25aee77b..1905d35a6619 100644 --- a/Documentation/ABI/removed/video1394 +++ b/Documentation/ABI/removed/video1394 @@ -8,6 +8,7 @@ Description: performance issues in its first generation. Any video1394 user had to use raw1394 + libraw1394 too because video1394 did not provide asynchronous I/O for device discovery and configuration. + Replaced by /dev/fw*, i.e. the ABI of firewire-core. diff --git a/Documentation/ABI/stable/firewire-cdev b/Documentation/ABI/stable/firewire-cdev index c9e8ff026154..261f85b13154 100644 --- a/Documentation/ABI/stable/firewire-cdev +++ b/Documentation/ABI/stable/firewire-cdev @@ -16,6 +16,7 @@ Description: different scope: - The 1394 node which is associated with the file: + - Asynchronous request transmission - Get the Configuration ROM - Query node ID @@ -23,6 +24,7 @@ Description: and local node - The 1394 bus (i.e. "card") to which the node is attached to: + - Isochronous stream transmission and reception - Asynchronous stream transmission and reception - Asynchronous broadcast request transmission @@ -35,6 +37,7 @@ Description: - Bus reset initiation, bus reset event reception - All 1394 buses: + - Allocation of IEEE 1212 address ranges on the local link layers, reception of inbound requests to such an address range, asynchronous response transmission @@ -59,50 +62,50 @@ Description: The following file operations are supported: open(2) - Currently the only useful flags are O_RDWR. + Currently the only useful flags are O_RDWR. ioctl(2) - Initiate various actions. Some take immediate effect, others - are performed asynchronously while or after the ioctl returns. - See the inline documentation in for - descriptions of all ioctls. + Initiate various actions. Some take immediate effect, others + are performed asynchronously while or after the ioctl returns. + See the inline documentation in for + descriptions of all ioctls. poll(2), select(2), epoll_wait(2) etc. - Watch for events to become available to be read. + Watch for events to become available to be read. read(2) - Receive various events. There are solicited events like - outbound asynchronous transaction completion or isochronous - buffer completion, and unsolicited events such as bus resets, - request reception, or PHY packet reception. Always use a read - buffer which is large enough to receive the largest event that - could ever arrive. See for descriptions - of all event types and for which ioctls affect reception of - events. + Receive various events. There are solicited events like + outbound asynchronous transaction completion or isochronous + buffer completion, and unsolicited events such as bus resets, + request reception, or PHY packet reception. Always use a read + buffer which is large enough to receive the largest event that + could ever arrive. See for descriptions + of all event types and for which ioctls affect reception of + events. mmap(2) - Allocate a DMA buffer for isochronous reception or transmission - and map it into the process address space. The arguments should - be used as follows: addr = NULL, length = the desired buffer - size, i.e. number of packets times size of largest packet, - prot = at least PROT_READ for reception and at least PROT_WRITE - for transmission, flags = MAP_SHARED, fd = the handle to the - /dev/fw*, offset = 0. + Allocate a DMA buffer for isochronous reception or transmission + and map it into the process address space. The arguments should + be used as follows: addr = NULL, length = the desired buffer + size, i.e. number of packets times size of largest packet, + prot = at least PROT_READ for reception and at least PROT_WRITE + for transmission, flags = MAP_SHARED, fd = the handle to the + /dev/fw*, offset = 0. Isochronous reception works in packet-per-buffer fashion except for multichannel reception which works in buffer-fill mode. munmap(2) - Unmap the isochronous I/O buffer from the process address space. + Unmap the isochronous I/O buffer from the process address space. close(2) - Besides stopping and freeing I/O contexts that were associated - with the file descriptor, back out any changes to the local - nodes' Configuration ROM. Deallocate isochronous channels and - bandwidth at the IRM that were marked for kernel-assisted - re- and deallocation. - -Users: libraw1394 - libdc1394 - libhinawa + Besides stopping and freeing I/O contexts that were associated + with the file descriptor, back out any changes to the local + nodes' Configuration ROM. Deallocate isochronous channels and + bandwidth at the IRM that were marked for kernel-assisted + re- and deallocation. + +Users: libraw1394; + libdc1394; + libhinawa; tools like linux-firewire-utils, fwhack, ... diff --git a/Documentation/ABI/stable/sysfs-acpi-pmprofile b/Documentation/ABI/stable/sysfs-acpi-pmprofile index fd97d22b677f..2d6314f0e4e4 100644 --- a/Documentation/ABI/stable/sysfs-acpi-pmprofile +++ b/Documentation/ABI/stable/sysfs-acpi-pmprofile @@ -1,8 +1,8 @@ -What: /sys/firmware/acpi/pm_profile +What: /sys/firmware/acpi/pm_profile Date: 03-Nov-2011 KernelVersion: v3.2 Contact: linux-acpi@vger.kernel.org -Description: The ACPI pm_profile sysfs interface exports the platform +Description: The ACPI pm_profile sysfs interface exports the platform power management (and performance) requirement expectations as provided by BIOS. The integer value is directly passed as retrieved from the FADT ACPI table. diff --git a/Documentation/ABI/stable/sysfs-bus-w1 b/Documentation/ABI/stable/sysfs-bus-w1 index 992dfb183ed0..5cd5e872bcae 100644 --- a/Documentation/ABI/stable/sysfs-bus-w1 +++ b/Documentation/ABI/stable/sysfs-bus-w1 @@ -6,6 +6,7 @@ Description: Bus scanning interval, microseconds component. control systems are attached/generate presence for as short as 100 ms - hence the tens-to-hundreds milliseconds scan intervals are required. + see Documentation/w1/w1-generic.rst for detailed information. Users: any user space application which wants to know bus scanning interval diff --git a/Documentation/ABI/stable/sysfs-class-tpm b/Documentation/ABI/stable/sysfs-class-tpm index ec464cf7861a..91ca63ec7581 100644 --- a/Documentation/ABI/stable/sysfs-class-tpm +++ b/Documentation/ABI/stable/sysfs-class-tpm @@ -191,6 +191,6 @@ Contact: linux-integrity@vger.kernel.org Description: The "tpm_version_major" property shows the TCG spec major version implemented by the TPM device. - Example output: + Example output:: - 2 + 2 diff --git a/Documentation/ABI/stable/sysfs-driver-speakup b/Documentation/ABI/stable/sysfs-driver-speakup index c6a32c434ce9..792f58ba327d 100644 --- a/Documentation/ABI/stable/sysfs-driver-speakup +++ b/Documentation/ABI/stable/sysfs-driver-speakup @@ -69,6 +69,7 @@ Description: Controls if typing interrupts output from speakup. With speakup if for example the say screen command is used before the entire screen is read. + With no_interrupt set to one, if the say screen command is used, and one then types on the keyboard, speakup will continue to say the whole screen regardless until @@ -215,8 +216,10 @@ Description: This file contains names for key states. Again, these are part of the help system. For instance, if you had pressed speakup + keypad 3, you would hear: "speakup keypad 3 is go to bottom edge." + The speakup key is depressed, so the name of the key state is speakup. + This part of the message comes from the states collection. What: /sys/accessibility/speakup/i18n/characters @@ -297,6 +300,7 @@ KernelVersion: 2.6 Contact: speakup@linux-speakup.org Description: Controls if punctuation is spoken by speakup, or by the synthesizer. + For example, speakup speaks ">" as "greater", while the espeak synthesizer used by the soft driver speaks "greater than". Zero lets speakup speak the punctuation. One lets the diff --git a/Documentation/ABI/testing/configfs-most b/Documentation/ABI/testing/configfs-most index ed67a4d9f6d6..bc6b8bd18da4 100644 --- a/Documentation/ABI/testing/configfs-most +++ b/Documentation/ABI/testing/configfs-most @@ -15,22 +15,28 @@ KernelVersion: 5.2 Description: The attributes: - buffer_size configure the buffer size for this channel + buffer_size + configure the buffer size for this channel - subbuffer_size configure the sub-buffer size for this channel + subbuffer_size + configure the sub-buffer size for this channel (needed for synchronous and isochrnous data) - num_buffers configure number of buffers used for this + num_buffers + configure number of buffers used for this channel - datatype configure type of data that will travel over + datatype + configure type of data that will travel over this channel - direction configure whether this link will be an input + direction + configure whether this link will be an input or output - dbr_size configure DBR data buffer size (this is used + dbr_size + configure DBR data buffer size (this is used for MediaLB communication only) packets_per_xact @@ -39,18 +45,23 @@ Description: transmitted via USB (this is used for USB communication only) - device name of the device the link is to be attached to + device + name of the device the link is to be attached to - channel name of the channel the link is to be attached to + channel + name of the channel the link is to be attached to - comp_params pass parameters needed by some components + comp_params + pass parameters needed by some components - create_link write '1' to this attribute to trigger the + create_link + write '1' to this attribute to trigger the creation of the link. In case of speculative configuration, the creation is post-poned until a physical device is being attached to the bus. - destroy_link write '1' to this attribute to destroy an + destroy_link + write '1' to this attribute to destroy an active link What: /sys/kernel/config/most_video/ @@ -59,22 +70,28 @@ KernelVersion: 5.2 Description: The attributes: - buffer_size configure the buffer size for this channel + buffer_size + configure the buffer size for this channel - subbuffer_size configure the sub-buffer size for this channel + subbuffer_size + configure the sub-buffer size for this channel (needed for synchronous and isochrnous data) - num_buffers configure number of buffers used for this + num_buffers + configure number of buffers used for this channel - datatype configure type of data that will travel over + datatype + configure type of data that will travel over this channel - direction configure whether this link will be an input + direction + configure whether this link will be an input or output - dbr_size configure DBR data buffer size (this is used + dbr_size + configure DBR data buffer size (this is used for MediaLB communication only) packets_per_xact @@ -83,18 +100,23 @@ Description: transmitted via USB (this is used for USB communication only) - device name of the device the link is to be attached to + device + name of the device the link is to be attached to - channel name of the channel the link is to be attached to + channel + name of the channel the link is to be attached to - comp_params pass parameters needed by some components + comp_params + pass parameters needed by some components - create_link write '1' to this attribute to trigger the + create_link + write '1' to this attribute to trigger the creation of the link. In case of speculative configuration, the creation is post-poned until a physical device is being attached to the bus. - destroy_link write '1' to this attribute to destroy an + destroy_link + write '1' to this attribute to destroy an active link What: /sys/kernel/config/most_net/ @@ -103,22 +125,28 @@ KernelVersion: 5.2 Description: The attributes: - buffer_size configure the buffer size for this channel + buffer_size + configure the buffer size for this channel - subbuffer_size configure the sub-buffer size for this channel + subbuffer_size + configure the sub-buffer size for this channel (needed for synchronous and isochrnous data) - num_buffers configure number of buffers used for this + num_buffers + configure number of buffers used for this channel - datatype configure type of data that will travel over + datatype + configure type of data that will travel over this channel - direction configure whether this link will be an input + direction + configure whether this link will be an input or output - dbr_size configure DBR data buffer size (this is used + dbr_size + configure DBR data buffer size (this is used for MediaLB communication only) packets_per_xact @@ -127,18 +155,23 @@ Description: transmitted via USB (this is used for USB communication only) - device name of the device the link is to be attached to + device + name of the device the link is to be attached to - channel name of the channel the link is to be attached to + channel + name of the channel the link is to be attached to - comp_params pass parameters needed by some components + comp_params + pass parameters needed by some components - create_link write '1' to this attribute to trigger the + create_link + write '1' to this attribute to trigger the creation of the link. In case of speculative configuration, the creation is post-poned until a physical device is being attached to the bus. - destroy_link write '1' to this attribute to destroy an + destroy_link + write '1' to this attribute to destroy an active link What: /sys/kernel/config/most_sound/ @@ -147,7 +180,8 @@ KernelVersion: 5.2 Description: The attributes: - create_card write '1' to this attribute to trigger the + create_card + write '1' to this attribute to trigger the registration of the sound card with the ALSA subsystem. @@ -157,22 +191,28 @@ KernelVersion: 5.2 Description: The attributes: - buffer_size configure the buffer size for this channel + buffer_size + configure the buffer size for this channel - subbuffer_size configure the sub-buffer size for this channel + subbuffer_size + configure the sub-buffer size for this channel (needed for synchronous and isochrnous data) - num_buffers configure number of buffers used for this + num_buffers + configure number of buffers used for this channel - datatype configure type of data that will travel over + datatype + configure type of data that will travel over this channel - direction configure whether this link will be an input + direction + configure whether this link will be an input or output - dbr_size configure DBR data buffer size (this is used + dbr_size + configure DBR data buffer size (this is used for MediaLB communication only) packets_per_xact @@ -181,16 +221,21 @@ Description: transmitted via USB (this is used for USB communication only) - device name of the device the link is to be attached to + device + name of the device the link is to be attached to - channel name of the channel the link is to be attached to + channel + name of the channel the link is to be attached to - comp_params pass parameters needed by some components + comp_params + pass parameters needed by some components - create_link write '1' to this attribute to trigger the + create_link + write '1' to this attribute to trigger the creation of the link. In case of speculative configuration, the creation is post-poned until a physical device is being attached to the bus. - destroy_link write '1' to this attribute to destroy an + destroy_link + write '1' to this attribute to destroy an active link diff --git a/Documentation/ABI/testing/configfs-usb-gadget-ecm b/Documentation/ABI/testing/configfs-usb-gadget-ecm index 0addf7704b4c..272bc1e4ce2e 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget-ecm +++ b/Documentation/ABI/testing/configfs-usb-gadget-ecm @@ -4,13 +4,17 @@ KernelVersion: 3.11 Description: The attributes: - ifname - network device interface name associated with + ifname + - network device interface name associated with this function instance - qmult - queue length multiplier for high and + qmult + - queue length multiplier for high and super speed - host_addr - MAC address of host's end of this + host_addr + - MAC address of host's end of this Ethernet over USB link - dev_addr - MAC address of device's end of this + dev_addr + - MAC address of device's end of this Ethernet over USB link diff --git a/Documentation/ABI/testing/configfs-usb-gadget-eem b/Documentation/ABI/testing/configfs-usb-gadget-eem index a4c57158fcde..178c3d5fb647 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget-eem +++ b/Documentation/ABI/testing/configfs-usb-gadget-eem @@ -4,11 +4,13 @@ KernelVersion: 3.11 Description: The attributes: - ifname - network device interface name associated with + ========== ============================================= + ifname network device interface name associated with this function instance - qmult - queue length multiplier for high and + qmult queue length multiplier for high and super speed - host_addr - MAC address of host's end of this + host_addr MAC address of host's end of this Ethernet over USB link - dev_addr - MAC address of device's end of this + dev_addr MAC address of device's end of this Ethernet over USB link + ========== ============================================= diff --git a/Documentation/ABI/testing/configfs-usb-gadget-loopback b/Documentation/ABI/testing/configfs-usb-gadget-loopback index 06beefbcf061..e6c6ba5ac7ff 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget-loopback +++ b/Documentation/ABI/testing/configfs-usb-gadget-loopback @@ -4,5 +4,7 @@ KernelVersion: 3.13 Description: The attributes: - qlen - depth of loopback queue - buflen - buffer length + ======= ======================= + qlen depth of loopback queue + buflen buffer length + ======= ======================= diff --git a/Documentation/ABI/testing/configfs-usb-gadget-mass-storage b/Documentation/ABI/testing/configfs-usb-gadget-mass-storage index 9931fb0d63ba..c86b63a7bb43 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget-mass-storage +++ b/Documentation/ABI/testing/configfs-usb-gadget-mass-storage @@ -4,12 +4,14 @@ KernelVersion: 3.13 Description: The attributes: - stall - Set to permit function to halt bulk endpoints. + =========== ============================================== + stall Set to permit function to halt bulk endpoints. Disabled on some USB devices known not to work correctly. You should set it to true. - num_buffers - Number of pipeline buffers. Valid numbers + num_buffers Number of pipeline buffers. Valid numbers are 2..4. Available only if CONFIG_USB_GADGET_DEBUG_FILES is set. + =========== ============================================== What: /config/usb-gadget/gadget/functions/mass_storage.name/lun.name Date: Oct 2013 @@ -17,15 +19,17 @@ KernelVersion: 3.13 Description: The attributes: - file - The path to the backing file for the LUN. + =========== ============================================== + file The path to the backing file for the LUN. Required if LUN is not marked as removable. - ro - Flag specifying access to the LUN shall be + ro Flag specifying access to the LUN shall be read-only. This is implied if CD-ROM emulation is enabled as well as when it was impossible to open "filename" in R/W mode. - removable - Flag specifying that LUN shall be indicated as + removable Flag specifying that LUN shall be indicated as being removable. - cdrom - Flag specifying that LUN shall be reported as + cdrom Flag specifying that LUN shall be reported as being a CD-ROM. - nofua - Flag specifying that FUA flag + nofua Flag specifying that FUA flag in SCSI WRITE(10,12) + =========== ============================================== diff --git a/Documentation/ABI/testing/configfs-usb-gadget-midi b/Documentation/ABI/testing/configfs-usb-gadget-midi index 6b341df7249c..07389cddd51a 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget-midi +++ b/Documentation/ABI/testing/configfs-usb-gadget-midi @@ -4,9 +4,11 @@ KernelVersion: 3.19 Description: The attributes: - index - index value for the USB MIDI adapter - id - ID string for the USB MIDI adapter - buflen - MIDI buffer length - qlen - USB read request queue length - in_ports - number of MIDI input ports - out_ports - number of MIDI output ports + ========== ==================================== + index index value for the USB MIDI adapter + id ID string for the USB MIDI adapter + buflen MIDI buffer length + qlen USB read request queue length + in_ports number of MIDI input ports + out_ports number of MIDI output ports + ========== ==================================== diff --git a/Documentation/ABI/testing/configfs-usb-gadget-printer b/Documentation/ABI/testing/configfs-usb-gadget-printer index 6b0714e3c605..7aa731bac2da 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget-printer +++ b/Documentation/ABI/testing/configfs-usb-gadget-printer @@ -4,6 +4,8 @@ KernelVersion: 4.1 Description: The attributes: - pnp_string - Data to be passed to the host in pnp string - q_len - Number of requests per endpoint + ========== =========================================== + pnp_string Data to be passed to the host in pnp string + q_len Number of requests per endpoint + ========== =========================================== diff --git a/Documentation/ABI/testing/configfs-usb-gadget-sourcesink b/Documentation/ABI/testing/configfs-usb-gadget-sourcesink index f56335af2d88..1f3d31b607b7 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget-sourcesink +++ b/Documentation/ABI/testing/configfs-usb-gadget-sourcesink @@ -4,11 +4,13 @@ KernelVersion: 3.13 Description: The attributes: - pattern - 0 (all zeros), 1 (mod63), 2 (none) - isoc_interval - 1..16 - isoc_maxpacket - 0 - 1023 (fs), 0 - 1024 (hs/ss) - isoc_mult - 0..2 (hs/ss only) - isoc_maxburst - 0..15 (ss only) - buflen - buffer length - bulk_qlen - depth of queue for bulk - iso_qlen - depth of queue for iso + ============== ================================== + pattern 0 (all zeros), 1 (mod63), 2 (none) + isoc_interval 1..16 + isoc_maxpacket 0 - 1023 (fs), 0 - 1024 (hs/ss) + isoc_mult 0..2 (hs/ss only) + isoc_maxburst 0..15 (ss only) + buflen buffer length + bulk_qlen depth of queue for bulk + iso_qlen depth of queue for iso + ============== ================================== diff --git a/Documentation/ABI/testing/configfs-usb-gadget-subset b/Documentation/ABI/testing/configfs-usb-gadget-subset index 9373e2c51ea4..0061b864351f 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget-subset +++ b/Documentation/ABI/testing/configfs-usb-gadget-subset @@ -4,11 +4,13 @@ KernelVersion: 3.11 Description: The attributes: - ifname - network device interface name associated with + ========== ============================================= + ifname network device interface name associated with this function instance - qmult - queue length multiplier for high and + qmult queue length multiplier for high and super speed - host_addr - MAC address of host's end of this + host_addr MAC address of host's end of this Ethernet over USB link - dev_addr - MAC address of device's end of this + dev_addr MAC address of device's end of this Ethernet over USB link + ========== ============================================= diff --git a/Documentation/ABI/testing/configfs-usb-gadget-uac2 b/Documentation/ABI/testing/configfs-usb-gadget-uac2 index 2bfdd4efa9bd..d4356c8b8cd6 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget-uac2 +++ b/Documentation/ABI/testing/configfs-usb-gadget-uac2 @@ -4,9 +4,11 @@ KernelVersion: 3.18 Description: The attributes: - c_chmask - capture channel mask - c_srate - capture sampling rate - c_ssize - capture sample size (bytes) - p_chmask - playback channel mask - p_srate - playback sampling rate - p_ssize - playback sample size (bytes) + ========= ============================ + c_chmask capture channel mask + c_srate capture sampling rate + c_ssize capture sample size (bytes) + p_chmask playback channel mask + p_srate playback sampling rate + p_ssize playback sample size (bytes) + ========= ============================ diff --git a/Documentation/ABI/testing/configfs-usb-gadget-uvc b/Documentation/ABI/testing/configfs-usb-gadget-uvc index cee81b0347bb..ac5e11af79a8 100644 --- a/Documentation/ABI/testing/configfs-usb-gadget-uvc +++ b/Documentation/ABI/testing/configfs-usb-gadget-uvc @@ -55,7 +55,7 @@ Description: Default output terminal descriptors ============== ============================================= iTerminal index of string descriptor - bSourceID id of the terminal to which this terminal + bSourceID id of the terminal to which this terminal is connected bAssocTerminal id of the input terminal to which this output terminal is associated diff --git a/Documentation/ABI/testing/debugfs-cec-error-inj b/Documentation/ABI/testing/debugfs-cec-error-inj index 5afcd78fbdb7..8debcb08a3b5 100644 --- a/Documentation/ABI/testing/debugfs-cec-error-inj +++ b/Documentation/ABI/testing/debugfs-cec-error-inj @@ -23,7 +23,7 @@ error injections without having to know the details of the driver-specific commands. Note that the output of 'error-inj' shall be valid as input to 'error-inj'. -So this must work: +So this must work:: $ cat error-inj >einj.txt $ cat einj.txt >error-inj diff --git a/Documentation/ABI/testing/debugfs-driver-habanalabs b/Documentation/ABI/testing/debugfs-driver-habanalabs index 2e9ae311e02d..c5d678d39144 100644 --- a/Documentation/ABI/testing/debugfs-driver-habanalabs +++ b/Documentation/ABI/testing/debugfs-driver-habanalabs @@ -20,9 +20,13 @@ Description: Allow the root user to disable/enable in runtime the clock The user can supply a bitmask value, each bit represents a different engine to disable/enable its clock gating feature. The bitmask is composed of 20 bits: - 0 - 7 : DMA channels - 8 - 11 : MME engines - 12 - 19 : TPC engines + + ======= ============ + 0 - 7 DMA channels + 8 - 11 MME engines + 12 - 19 TPC engines + ======= ============ + The bit's location of a specific engine can be determined using (1 << GAUDI_ENGINE_ID_*). GAUDI_ENGINE_ID_* values are defined in uapi habanalabs.h file in enum gaudi_engine_id @@ -59,6 +63,7 @@ Description: Allows the root user to read or write directly through the the generic Linux user-space PCI mapping) because the DDR bar is very small compared to the DDR memory and only the driver can move the bar before and after the transaction. + If the IOMMU is disabled, it also allows the root user to read or write from the host a device VA of a host mapped memory @@ -73,6 +78,7 @@ Description: Allows the root user to read or write 64 bit data directly the generic Linux user-space PCI mapping) because the DDR bar is very small compared to the DDR memory and only the driver can move the bar before and after the transaction. + If the IOMMU is disabled, it also allows the root user to read or write from the host a device VA of a host mapped memory diff --git a/Documentation/ABI/testing/debugfs-pfo-nx-crypto b/Documentation/ABI/testing/debugfs-pfo-nx-crypto index 685d5a448423..f75a655c1531 100644 --- a/Documentation/ABI/testing/debugfs-pfo-nx-crypto +++ b/Documentation/ABI/testing/debugfs-pfo-nx-crypto @@ -4,42 +4,42 @@ KernelVersion: 3.4 Contact: Kent Yoder Description: - These debugfs interfaces are built by the nx-crypto driver, built in +These debugfs interfaces are built by the nx-crypto driver, built in arch/powerpc/crypto/nx. Error Detection =============== errors: -- A u32 providing a total count of errors since the driver was loaded. The -only errors counted here are those returned from the hcall, H_COP_OP. + A u32 providing a total count of errors since the driver was loaded. The + only errors counted here are those returned from the hcall, H_COP_OP. last_error: -- The most recent non-zero return code from the H_COP_OP hcall. -EBUSY is not -recorded here (the hcall will retry until -EBUSY goes away). + The most recent non-zero return code from the H_COP_OP hcall. -EBUSY is not + recorded here (the hcall will retry until -EBUSY goes away). last_error_pid: -- The process ID of the process who received the most recent error from the -hcall. + The process ID of the process who received the most recent error from the + hcall. Device Use ========== aes_bytes: -- The total number of bytes encrypted using AES in any of the driver's -supported modes. + The total number of bytes encrypted using AES in any of the driver's + supported modes. aes_ops: -- The total number of AES operations submitted to the hardware. + The total number of AES operations submitted to the hardware. sha256_bytes: -- The total number of bytes hashed by the hardware using SHA-256. + The total number of bytes hashed by the hardware using SHA-256. sha256_ops: -- The total number of SHA-256 operations submitted to the hardware. + The total number of SHA-256 operations submitted to the hardware. sha512_bytes: -- The total number of bytes hashed by the hardware using SHA-512. + The total number of bytes hashed by the hardware using SHA-512. sha512_ops: -- The total number of SHA-512 operations submitted to the hardware. + The total number of SHA-512 operations submitted to the hardware. diff --git a/Documentation/ABI/testing/debugfs-pktcdvd b/Documentation/ABI/testing/debugfs-pktcdvd index 787907d70462..f6f65a4faea0 100644 --- a/Documentation/ABI/testing/debugfs-pktcdvd +++ b/Documentation/ABI/testing/debugfs-pktcdvd @@ -10,7 +10,7 @@ these files in debugfs: /sys/kernel/debug/pktcdvd/pktcdvd[0-7]/ ==== ====== ==================================== - info (0444) Lots of driver statistics and infos. + info 0444 Lots of driver statistics and infos. ==== ====== ==================================== Example:: diff --git a/Documentation/ABI/testing/debugfs-turris-mox-rwtm b/Documentation/ABI/testing/debugfs-turris-mox-rwtm index c8f7dadd591c..ad08f535af3b 100644 --- a/Documentation/ABI/testing/debugfs-turris-mox-rwtm +++ b/Documentation/ABI/testing/debugfs-turris-mox-rwtm @@ -2,10 +2,12 @@ What: /sys/kernel/debug/turris-mox-rwtm/do_sign Date: Jun 2020 KernelVersion: 5.8 Contact: Marek Behún -Description: (W) - Message to sign with the ECDSA private key stored in +Description: + + === =========================================================== + (W) Message to sign with the ECDSA private key stored in device's OTP. The message must be exactly 64 bytes (since this is intended for SHA-512 hashes). - (R) - The resulting signature, 136 bytes. This contains the R and + (R) The resulting signature, 136 bytes. This contains the R and S values of the ECDSA signature, both in big-endian format. + === =========================================================== diff --git a/Documentation/ABI/testing/debugfs-wilco-ec b/Documentation/ABI/testing/debugfs-wilco-ec index 9d8d9d2def5b..682e3c09ef4d 100644 --- a/Documentation/ABI/testing/debugfs-wilco-ec +++ b/Documentation/ABI/testing/debugfs-wilco-ec @@ -27,16 +27,17 @@ Description: for writing, two for the type and at least a single byte of data. - Example: - // Request EC info type 3 (EC firmware build date) - // Corresponds with sending type 0x00f0 with - // MBOX = [38, 00, 03, 00] - $ echo 00 f0 38 00 03 00 > /sys/kernel/debug/wilco_ec/raw - // View the result. The decoded ASCII result "12/21/18" is - // included after the raw hex. - // Corresponds with MBOX = [00, 00, 31, 32, 2f, 32, 31, 38, ...] - $ cat /sys/kernel/debug/wilco_ec/raw - 00 00 31 32 2f 32 31 2f 31 38 00 38 00 01 00 2f 00 ..12/21/18.8... + Example:: + + // Request EC info type 3 (EC firmware build date) + // Corresponds with sending type 0x00f0 with + // MBOX = [38, 00, 03, 00] + $ echo 00 f0 38 00 03 00 > /sys/kernel/debug/wilco_ec/raw + // View the result. The decoded ASCII result "12/21/18" is + // included after the raw hex. + // Corresponds with MBOX = [00, 00, 31, 32, 2f, 32, 31, 38, ...] + $ cat /sys/kernel/debug/wilco_ec/raw + 00 00 31 32 2f 32 31 2f 31 38 00 38 00 01 00 2f 00 ..12/21/18.8... Note that the first 16 bytes of the received MBOX[] will be printed, even if some of the data is junk, and skipping bytes diff --git a/Documentation/ABI/testing/dell-smbios-wmi b/Documentation/ABI/testing/dell-smbios-wmi index fc919ce16008..5f3a0dc67050 100644 --- a/Documentation/ABI/testing/dell-smbios-wmi +++ b/Documentation/ABI/testing/dell-smbios-wmi @@ -10,29 +10,29 @@ Description: 1) To perform an SMBIOS call from userspace, you'll need to - first determine the minimum size of the calling interface - buffer for your machine. - Platforms that contain larger buffers can return larger - objects from the system firmware. - Commonly this size is either 4k or 32k. + first determine the minimum size of the calling interface + buffer for your machine. + Platforms that contain larger buffers can return larger + objects from the system firmware. + Commonly this size is either 4k or 32k. - To determine the size of the buffer read() a u64 dword from - the WMI character device /dev/wmi/dell-smbios. + To determine the size of the buffer read() a u64 dword from + the WMI character device /dev/wmi/dell-smbios. 2) After you've determined the minimum size of the calling - interface buffer, you can allocate a structure that represents - the structure documented above. + interface buffer, you can allocate a structure that represents + the structure documented above. 3) In the 'length' object store the size of the buffer you - determined above and allocated. + determined above and allocated. 4) In this buffer object, prepare as necessary for the SMBIOS - call you're interested in. Typically SMBIOS buffers have - "class", "select", and "input" defined to values that coincide - with the data you are interested in. - Documenting class/select/input values is outside of the scope - of this documentation. Check with the libsmbios project for - further documentation on these values. + call you're interested in. Typically SMBIOS buffers have + "class", "select", and "input" defined to values that coincide + with the data you are interested in. + Documenting class/select/input values is outside of the scope + of this documentation. Check with the libsmbios project for + further documentation on these values. 6) Run the call by using ioctl() as described in the header. diff --git a/Documentation/ABI/testing/gpio-cdev b/Documentation/ABI/testing/gpio-cdev index 7b265fbb47e3..66bdcd188b6c 100644 --- a/Documentation/ABI/testing/gpio-cdev +++ b/Documentation/ABI/testing/gpio-cdev @@ -12,15 +12,16 @@ Description: The following file operations are supported: open(2) - Currently the only useful flags are O_RDWR. + Currently the only useful flags are O_RDWR. ioctl(2) - Initiate various actions. - See the inline documentation in [include/uapi] - for descriptions of all ioctls. + Initiate various actions. + + See the inline documentation in [include/uapi] + for descriptions of all ioctls. close(2) - Stops and free up the I/O contexts that was associated - with the file descriptor. + Stops and free up the I/O contexts that was associated + with the file descriptor. Users: TBD diff --git a/Documentation/ABI/testing/procfs-diskstats b/Documentation/ABI/testing/procfs-diskstats index df5a3a8c1edf..e58d641443d3 100644 --- a/Documentation/ABI/testing/procfs-diskstats +++ b/Documentation/ABI/testing/procfs-diskstats @@ -35,7 +35,9 @@ Description: Kernel 5.5+ appends two more fields for flush requests: - 19 - flush requests completed successfully - 20 - time spent flushing + == ===================================== + 19 flush requests completed successfully + 20 time spent flushing + == ===================================== For more details refer to Documentation/admin-guide/iostats.rst diff --git a/Documentation/ABI/testing/procfs-smaps_rollup b/Documentation/ABI/testing/procfs-smaps_rollup index 046978193368..a4e31c465194 100644 --- a/Documentation/ABI/testing/procfs-smaps_rollup +++ b/Documentation/ABI/testing/procfs-smaps_rollup @@ -14,28 +14,28 @@ Description: For more details, see Documentation/filesystems/proc.rst and the procfs man page. - Typical output looks like this: + Typical output looks like this:: - 00100000-ff709000 ---p 00000000 00:00 0 [rollup] - Size: 1192 kB - KernelPageSize: 4 kB - MMUPageSize: 4 kB - Rss: 884 kB - Pss: 385 kB - Pss_Anon: 301 kB - Pss_File: 80 kB - Pss_Shmem: 4 kB - Shared_Clean: 696 kB - Shared_Dirty: 0 kB - Private_Clean: 120 kB - Private_Dirty: 68 kB - Referenced: 884 kB - Anonymous: 68 kB - LazyFree: 0 kB - AnonHugePages: 0 kB - ShmemPmdMapped: 0 kB - Shared_Hugetlb: 0 kB - Private_Hugetlb: 0 kB - Swap: 0 kB - SwapPss: 0 kB - Locked: 385 kB + 00100000-ff709000 ---p 00000000 00:00 0 [rollup] + Size: 1192 kB + KernelPageSize: 4 kB + MMUPageSize: 4 kB + Rss: 884 kB + Pss: 385 kB + Pss_Anon: 301 kB + Pss_File: 80 kB + Pss_Shmem: 4 kB + Shared_Clean: 696 kB + Shared_Dirty: 0 kB + Private_Clean: 120 kB + Private_Dirty: 68 kB + Referenced: 884 kB + Anonymous: 68 kB + LazyFree: 0 kB + AnonHugePages: 0 kB + ShmemPmdMapped: 0 kB + Shared_Hugetlb: 0 kB + Private_Hugetlb: 0 kB + Swap: 0 kB + SwapPss: 0 kB + Locked: 385 kB diff --git a/Documentation/ABI/testing/pstore b/Documentation/ABI/testing/pstore index d45209abdb1b..5b02540781a2 100644 --- a/Documentation/ABI/testing/pstore +++ b/Documentation/ABI/testing/pstore @@ -9,25 +9,25 @@ Description: Generic interface to platform dependent persistent storage. provide a generic interface to show records captured in the dying moments. In the case of a panic the last part of the console log is captured, but other interesting - data can also be saved. + data can also be saved:: - # mount -t pstore -o kmsg_bytes=8000 - /sys/fs/pstore + # mount -t pstore -o kmsg_bytes=8000 - /sys/fs/pstore - $ ls -l /sys/fs/pstore/ - total 0 - -r--r--r-- 1 root root 7896 Nov 30 15:38 dmesg-erst-1 + $ ls -l /sys/fs/pstore/ + total 0 + -r--r--r-- 1 root root 7896 Nov 30 15:38 dmesg-erst-1 Different users of this interface will result in different filename prefixes. Currently two are defined: - "dmesg" - saved console log - "mce" - architecture dependent data from fatal h/w error + - "dmesg" - saved console log + - "mce" - architecture dependent data from fatal h/w error Once the information in a file has been read, removing the file will signal to the underlying persistent storage - device that it can reclaim the space for later re-use. + device that it can reclaim the space for later re-use:: - $ rm /sys/fs/pstore/dmesg-erst-1 + $ rm /sys/fs/pstore/dmesg-erst-1 The expectation is that all files in /sys/fs/pstore/ will be saved elsewhere and erased from persistent store @@ -44,4 +44,3 @@ Description: Generic interface to platform dependent persistent storage. backends are available, the preferred backend may be set by passing the pstore.backend= argument to the kernel at boot time. - diff --git a/Documentation/ABI/testing/sysfs-block-rnbd b/Documentation/ABI/testing/sysfs-block-rnbd index 8f070b47f361..14a6fe9422b3 100644 --- a/Documentation/ABI/testing/sysfs-block-rnbd +++ b/Documentation/ABI/testing/sysfs-block-rnbd @@ -9,9 +9,9 @@ Description: To unmap a volume, "normal" or "force" has to be written to: is using the device. When "force" is used, the device is also unmapped when device is in use. All I/Os that are in progress will fail. - Example: + Example:: - # echo "normal" > /sys/block/rnbd0/rnbd/unmap_device + # echo "normal" > /sys/block/rnbd0/rnbd/unmap_device What: /sys/block/rnbd/rnbd/state Date: Feb 2020 diff --git a/Documentation/ABI/testing/sysfs-bus-acpi b/Documentation/ABI/testing/sysfs-bus-acpi index c78603497b97..58abacf59b2a 100644 --- a/Documentation/ABI/testing/sysfs-bus-acpi +++ b/Documentation/ABI/testing/sysfs-bus-acpi @@ -5,6 +5,7 @@ Description: This attribute indicates the full path of ACPI namespace object associated with the device object. For example, \_SB_.PCI0. + This file is not present for device objects representing fixed ACPI hardware features (like power and sleep buttons). diff --git a/Documentation/ABI/testing/sysfs-bus-coresight-devices-etb10 b/Documentation/ABI/testing/sysfs-bus-coresight-devices-etb10 index b5f526081711..3e92cbd3fd83 100644 --- a/Documentation/ABI/testing/sysfs-bus-coresight-devices-etb10 +++ b/Documentation/ABI/testing/sysfs-bus-coresight-devices-etb10 @@ -4,7 +4,10 @@ KernelVersion: 3.19 Contact: Mathieu Poirier Description: (RW) Add/remove a sink from a trace path. There can be multiple source for a single sink. - ex: echo 1 > /sys/bus/coresight/devices/20010000.etb/enable_sink + + ex:: + + echo 1 > /sys/bus/coresight/devices/20010000.etb/enable_sink What: /sys/bus/coresight/devices/.etb/trigger_cntr Date: November 2014 diff --git a/Documentation/ABI/testing/sysfs-bus-css b/Documentation/ABI/testing/sysfs-bus-css index 966f8504bd7b..12a733fe357f 100644 --- a/Documentation/ABI/testing/sysfs-bus-css +++ b/Documentation/ABI/testing/sysfs-bus-css @@ -20,6 +20,7 @@ Contact: Cornelia Huck Description: Contains the ids of the channel paths used by this subchannel, as reported by the channel subsystem during subchannel recognition. + Note: This is an I/O-subchannel specific attribute. Users: s390-tools, HAL @@ -31,6 +32,7 @@ Description: Contains the PIM/PAM/POM values, as reported by the channel subsystem when last queried by the common I/O layer (this implies that this attribute is not necessarily in sync with the values current in the channel subsystem). + Note: This is an I/O-subchannel specific attribute. Users: s390-tools, HAL @@ -53,6 +55,7 @@ Description: This file allows the driver for a device to be specified. When opt-out of driver binding using a driver_override name such as "none". Only a single driver may be specified in the override, there is no support for parsing delimiters. + Note that unlike the mechanism of the same name for pci, this file does not allow to override basic matching rules. I.e., the driver must still match the subchannel type of the device. diff --git a/Documentation/ABI/testing/sysfs-bus-dfl b/Documentation/ABI/testing/sysfs-bus-dfl index 23543be904f2..b0265ab17200 100644 --- a/Documentation/ABI/testing/sysfs-bus-dfl +++ b/Documentation/ABI/testing/sysfs-bus-dfl @@ -4,6 +4,7 @@ KernelVersion: 5.10 Contact: Xu Yilun Description: Read-only. It returns type of DFL FIU of the device. Now DFL supports 2 FIU types, 0 for FME, 1 for PORT. + Format: 0x%x What: /sys/bus/dfl/devices/dfl_dev.X/feature_id @@ -12,4 +13,5 @@ KernelVersion: 5.10 Contact: Xu Yilun Description: Read-only. It returns feature identifier local to its DFL FIU type. + Format: 0x%x diff --git a/Documentation/ABI/testing/sysfs-bus-event_source-devices-hv_24x7 b/Documentation/ABI/testing/sysfs-bus-event_source-devices-hv_24x7 index 2273627df190..de390a010af8 100644 --- a/Documentation/ABI/testing/sysfs-bus-event_source-devices-hv_24x7 +++ b/Documentation/ABI/testing/sysfs-bus-event_source-devices-hv_24x7 @@ -7,7 +7,7 @@ Description: Read-only. Attribute group to describe the magic bits Each attribute under this group defines a bit range of the perf_event_attr.config. All supported attributes are listed - below. + below:: chip = "config:16-31" core = "config:16-31" @@ -16,9 +16,9 @@ Description: Read-only. Attribute group to describe the magic bits offset = "config:32-63" vcpu = "config:16-31" - For example, + For example:: - PM_PB_CYC = "domain=1,offset=0x80,chip=?,lpar=0x0" + PM_PB_CYC = "domain=1,offset=0x80,chip=?,lpar=0x0" In this event, '?' after chip specifies that this value will be provided by user while running this event. diff --git a/Documentation/ABI/testing/sysfs-bus-event_source-devices-hv_gpci b/Documentation/ABI/testing/sysfs-bus-event_source-devices-hv_gpci index 6a023b42486c..12e2bf92783f 100644 --- a/Documentation/ABI/testing/sysfs-bus-event_source-devices-hv_gpci +++ b/Documentation/ABI/testing/sysfs-bus-event_source-devices-hv_gpci @@ -7,7 +7,7 @@ Description: Read-only. Attribute group to describe the magic bits Each attribute under this group defines a bit range of the perf_event_attr.config. All supported attributes are listed - below. + below:: counter_info_version = "config:16-23" length = "config:24-31" @@ -20,9 +20,9 @@ Description: Read-only. Attribute group to describe the magic bits secondary_index = "config:0-15" starting_index = "config:32-63" - For example, + For example:: - processor_core_utilization_instructions_completed = "request=0x94, + processor_core_utilization_instructions_completed = "request=0x94, phys_processor_idx=?,counter_info_version=0x8, length=8,offset=0x18" @@ -36,6 +36,7 @@ Description: '0' if the hypervisor is configured to forbid access to event counters being accumulated by other guests and to physical domain event counters. + '1' if that access is allowed. What: /sys/bus/event_source/devices/hv_gpci/interface/ga diff --git a/Documentation/ABI/testing/sysfs-bus-fcoe b/Documentation/ABI/testing/sysfs-bus-fcoe index 657df13b100d..8fe787cc4ab7 100644 --- a/Documentation/ABI/testing/sysfs-bus-fcoe +++ b/Documentation/ABI/testing/sysfs-bus-fcoe @@ -3,16 +3,19 @@ Date: August 2012 KernelVersion: TBD Contact: Robert Love , devel@open-fcoe.org Description: The FCoE bus. Attributes in this directory are control interfaces. + Attributes: - ctlr_create: 'FCoE Controller' instance creation interface. Writing an + ctlr_create: + 'FCoE Controller' instance creation interface. Writing an to this file will allocate and populate sysfs with a fcoe_ctlr_device (ctlr_X). The user can then configure any per-port settings and finally write to the fcoe_ctlr_device's 'start' attribute to begin the kernel's discovery and login process. - ctlr_destroy: 'FCoE Controller' instance removal interface. Writing a + ctlr_destroy: + 'FCoE Controller' instance removal interface. Writing a fcoe_ctlr_device's sysfs name to this file will log the fcoe_ctlr_device out of the fabric or otherwise connected FCoE devices. It will also free all kernel memory allocated @@ -32,11 +35,13 @@ Description: 'FCoE Controller' instances on the fcoe bus. Attributes: - fcf_dev_loss_tmo: Device loss timeout period (see below). Changing + fcf_dev_loss_tmo: + Device loss timeout period (see below). Changing this value will change the dev_loss_tmo for all FCFs discovered by this controller. - mode: Display or change the FCoE Controller's mode. Possible + mode: + Display or change the FCoE Controller's mode. Possible modes are 'Fabric' and 'VN2VN'. If a FCoE Controller is started in 'Fabric' mode then FIP FCF discovery is initiated and ultimately a fabric login is attempted. @@ -44,23 +49,30 @@ Attributes: FIP VN2VN discovery and login is performed. A FCoE Controller only supports one mode at a time. - enabled: Whether an FCoE controller is enabled or disabled. + enabled: + Whether an FCoE controller is enabled or disabled. 0 if disabled, 1 if enabled. Writing either 0 or 1 to this file will enable or disable the FCoE controller. - lesb/link_fail: Link Error Status Block (LESB) link failure count. + lesb/link_fail: + Link Error Status Block (LESB) link failure count. - lesb/vlink_fail: Link Error Status Block (LESB) virtual link + lesb/vlink_fail: + Link Error Status Block (LESB) virtual link failure count. - lesb/miss_fka: Link Error Status Block (LESB) missed FCoE + lesb/miss_fka: + Link Error Status Block (LESB) missed FCoE Initialization Protocol (FIP) Keep-Alives (FKA). - lesb/symb_err: Link Error Status Block (LESB) symbolic error count. + lesb/symb_err: + Link Error Status Block (LESB) symbolic error count. - lesb/err_block: Link Error Status Block (LESB) block error count. + lesb/err_block: + Link Error Status Block (LESB) block error count. - lesb/fcs_error: Link Error Status Block (LESB) Fibre Channel + lesb/fcs_error: + Link Error Status Block (LESB) Fibre Channel Services error count. Notes: ctlr_X (global increment starting at 0) @@ -75,31 +87,41 @@ Description: 'FCoE FCF' instances on the fcoe bus. A FCF is a Fibre Channel Fibre Channel frames into a FC fabric. It can also take outbound FC frames and pack them in Ethernet packets to be sent to their destination on the Ethernet segment. + Attributes: - fabric_name: Identifies the fabric that the FCF services. + fabric_name: + Identifies the fabric that the FCF services. - switch_name: Identifies the FCF. + switch_name: + Identifies the FCF. - priority: The switch's priority amongst other FCFs on the same + priority: + The switch's priority amongst other FCFs on the same fabric. - selected: 1 indicates that the switch has been selected for use; + selected: + 1 indicates that the switch has been selected for use; 0 indicates that the switch will not be used. - fc_map: The Fibre Channel MAP + fc_map: + The Fibre Channel MAP - vfid: The Virtual Fabric ID + vfid: + The Virtual Fabric ID - mac: The FCF's MAC address + mac: + The FCF's MAC address - fka_period: The FIP Keep-Alive period + fka_period: + The FIP Keep-Alive period fabric_state: The internal kernel state - "Unknown" - Initialization value - "Disconnected" - No link to the FCF/fabric - "Connected" - Host is connected to the FCF - "Deleted" - FCF is being removed from the system + + - "Unknown" - Initialization value + - "Disconnected" - No link to the FCF/fabric + - "Connected" - Host is connected to the FCF + - "Deleted" - FCF is being removed from the system dev_loss_tmo: The device loss timeout period for this FCF. diff --git a/Documentation/ABI/testing/sysfs-bus-fsl-mc b/Documentation/ABI/testing/sysfs-bus-fsl-mc index 80256b8b4f26..bf3c6af6ad89 100644 --- a/Documentation/ABI/testing/sysfs-bus-fsl-mc +++ b/Documentation/ABI/testing/sysfs-bus-fsl-mc @@ -6,8 +6,10 @@ Description: the driver to attempt to bind to the device found at this location. The format for the location is Object.Id and is the same as found in /sys/bus/fsl-mc/devices/. - For example: - # echo dpni.2 > /sys/bus/fsl-mc/drivers/fsl_dpaa2_eth/bind + + For example:: + + # echo dpni.2 > /sys/bus/fsl-mc/drivers/fsl_dpaa2_eth/bind What: /sys/bus/fsl-mc/drivers/.../unbind Date: December 2016 @@ -17,5 +19,7 @@ Description: driver to attempt to unbind from the device found at this location. The format for the location is Object.Id and is the same as found in /sys/bus/fsl-mc/devices/. - For example: - # echo dpni.2 > /sys/bus/fsl-mc/drivers/fsl_dpaa2_eth/unbind + + For example:: + + # echo dpni.2 > /sys/bus/fsl-mc/drivers/fsl_dpaa2_eth/unbind diff --git a/Documentation/ABI/testing/sysfs-bus-i2c-devices-fsa9480 b/Documentation/ABI/testing/sysfs-bus-i2c-devices-fsa9480 index 9de269bb0ae5..42dfc9399d2d 100644 --- a/Documentation/ABI/testing/sysfs-bus-i2c-devices-fsa9480 +++ b/Documentation/ABI/testing/sysfs-bus-i2c-devices-fsa9480 @@ -3,19 +3,25 @@ Date: February 2011 Contact: Minkyu Kang Description: show what device is attached - NONE - no device - USB - USB device is attached - UART - UART is attached - CHARGER - Charger is attaced - JIG - JIG is attached + + ======= ====================== + NONE no device + USB USB device is attached + UART UART is attached + CHARGER Charger is attaced + JIG JIG is attached + ======= ====================== What: /sys/bus/i2c/devices/.../switch Date: February 2011 Contact: Minkyu Kang Description: show or set the state of manual switch - VAUDIO - switch to VAUDIO path - UART - switch to UART path - AUDIO - switch to AUDIO path - DHOST - switch to DHOST path - AUTO - switch automatically by device + + ======= ============================== + VAUDIO switch to VAUDIO path + UART switch to UART path + AUDIO switch to AUDIO path + DHOST switch to DHOST path + AUTO switch automatically by device + ======= ============================== diff --git a/Documentation/ABI/testing/sysfs-bus-i3c b/Documentation/ABI/testing/sysfs-bus-i3c index 2f332ec36f82..1f4a2662335b 100644 --- a/Documentation/ABI/testing/sysfs-bus-i3c +++ b/Documentation/ABI/testing/sysfs-bus-i3c @@ -84,6 +84,7 @@ Description: by space. Modes can be "hdr-ddr", "hdr-tsp" and "hdr-tsl". See the I3C specification for more details about these HDR modes. + This entry describes the HDRCAP of the master controller driving the bus. @@ -135,6 +136,7 @@ Description: Expose the HDR (High Data Rate) capabilities of a device. Returns a list of supported HDR mode, each element is separated by space. Modes can be "hdr-ddr", "hdr-tsp" and "hdr-tsl". + See the I3C specification for more details about these HDR modes. diff --git a/Documentation/ABI/testing/sysfs-bus-iio b/Documentation/ABI/testing/sysfs-bus-iio index e3df71987eff..df42bed09f25 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio +++ b/Documentation/ABI/testing/sysfs-bus-iio @@ -15,6 +15,7 @@ Description: based on hardware generated events (e.g. data ready) or provided by a separate driver for other hardware (e.g. periodic timer, GPIO or high resolution timer). + Contains trigger type specific elements. These do not generalize well and hence are not documented in this file. X is the IIO index of the trigger. @@ -666,6 +667,7 @@ Description: [Y][_name]__thresh_falling_value may take different values, but the device can only enable both thresholds or neither. + Note the driver will assume the last p events requested are to be enabled where p is how many it supports (which may vary depending on the exact set requested. So if you want to be @@ -720,6 +722,7 @@ Description: [Y][_name]__roc_falling_value may take different values, but the device can only enable both rate of change thresholds or neither. + Note the driver will assume the last p events requested are to be enabled where p is however many it supports (which may vary depending on the exact set requested. So if you want to be @@ -775,9 +778,11 @@ Description: Specifies the value of threshold that the device is comparing against for the events enabled by Y[_name]_thresh[_rising|falling]_en. + If separate attributes exist for the two directions, but direction is not specified for this attribute, then a single threshold value applies to both directions. + The raw or input element of the name indicates whether the value is in raw device units or in processed units (as _raw and _input do on sysfs direct channel read attributes). @@ -860,6 +865,7 @@ Description: If separate attributes exist for the two directions, but direction is not specified for this attribute, then a single hysteresis value applies to both directions. + For falling events the hysteresis is added to the _value attribute for this event to get the upper threshold for when the event goes back to normal, for rising events the hysteresis is subtracted from the _value @@ -906,6 +912,7 @@ Description: Specifies the value of rate of change threshold that the device is comparing against for the events enabled by [Y][_name]_roc[_rising|falling]_en. + If separate attributes exist for the two directions, but direction is not specified for this attribute, then a single threshold value applies to both directions. @@ -1305,6 +1312,7 @@ Description: Proximity measurement indicating that some object is near the sensor, usually by observing reflectivity of infrared or ultrasound emitted. + Often these sensors are unit less and as such conversion to SI units is not possible. Higher proximity measurements indicate closer objects, and vice versa. Units after @@ -1450,9 +1458,12 @@ Contact: linux-iio@vger.kernel.org Description: A single positive integer specifying the maximum number of scan elements to wait for. + Poll will block until the watermark is reached. + Blocking read will wait until the minimum between the requested read amount or the low water mark is available. + Non-blocking read will retrieve the available samples from the buffer even if there are less samples then watermark level. This allows the application to block on poll with a timeout and read @@ -1481,11 +1492,13 @@ Description: device settings allows it (e.g. if a trigger is set that samples data differently that the hardware fifo does then hardware fifo will not enabled). + If the hardware fifo is enabled and the level of the hardware fifo reaches the hardware fifo watermark level the device will flush its hardware fifo to the device buffer. Doing a non blocking read on the device when no samples are present in the device buffer will also force a flush. + When the hardware fifo is enabled there is no need to use a trigger to use buffer mode since the watermark settings guarantees that the hardware fifo is flushed to the device @@ -1523,6 +1536,7 @@ Description: A single positive integer specifying the minimum watermark level for the hardware fifo of this device. If the device does not have a hardware fifo this entry is not present. + If the user sets buffer/watermark to a value less than this one, then the hardware watermark will remain unset. @@ -1533,6 +1547,7 @@ Description: A single positive integer specifying the maximum watermark level for the hardware fifo of this device. If the device does not have a hardware fifo this entry is not present. + If the user sets buffer/watermark to a value greater than this one, then the hardware watermark will be capped at this value. @@ -1544,6 +1559,7 @@ Description: levels for the hardware fifo. This entry is optional and if it is not present it means that all the values between hwfifo_watermark_min and hwfifo_watermark_max are supported. + If the user sets buffer/watermark to a value greater than hwfifo_watermak_min but not equal to any of the values in this list, the driver will chose an appropriate value for the @@ -1605,7 +1621,8 @@ KernelVersion: 4.1.0 Contact: linux-iio@vger.kernel.org Description: '1' (enable) or '0' (disable) specifying the enable - of heater function. Same reading values apply + of heater function. Same reading values apply. + This ABI is especially applicable for humidity sensors to heatup the device and get rid of any condensation in some humidity environment diff --git a/Documentation/ABI/testing/sysfs-bus-iio-adc-hi8435 b/Documentation/ABI/testing/sysfs-bus-iio-adc-hi8435 index f30b4c424fb6..4b01150af397 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio-adc-hi8435 +++ b/Documentation/ABI/testing/sysfs-bus-iio-adc-hi8435 @@ -19,9 +19,11 @@ Description: is separately set for "GND-Open" and "Supply-Open" modes. Channels 0..31 have common low threshold values, but could have different sensing_modes. + The low voltage threshold range is between 2..21V. Hysteresis between low and high thresholds can not be lower then 2 and can not be odd. + If falling threshold results hysteresis to odd value then rising threshold is automatically subtracted by one. @@ -34,10 +36,13 @@ Description: this value then the threshold rising event is pushed. Depending on in_voltageY_sensing_mode the high voltage threshold is separately set for "GND-Open" and "Supply-Open" modes. + Channels 0..31 have common high threshold values, but could have different sensing_modes. + The high voltage threshold range is between 3..22V. Hysteresis between low and high thresholds can not be lower then 2 and can not be odd. + If rising threshold results hysteresis to odd value then falling threshold is automatically appended by one. diff --git a/Documentation/ABI/testing/sysfs-bus-iio-adc-stm32 b/Documentation/ABI/testing/sysfs-bus-iio-adc-stm32 index efe4c85e3c8b..1975c7a1af34 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio-adc-stm32 +++ b/Documentation/ABI/testing/sysfs-bus-iio-adc-stm32 @@ -5,10 +5,13 @@ Description: The STM32 ADC can be configured to use external trigger sources (e.g. timers, pwm or exti gpio). Then, it can be tuned to start conversions on external trigger by either: + - "rising-edge" - "falling-edge" - "both-edges". + Reading returns current trigger polarity. + Writing value before enabling conversions sets trigger polarity. What: /sys/bus/iio/devices/triggerX/trigger_polarity_available diff --git a/Documentation/ABI/testing/sysfs-bus-iio-distance-srf08 b/Documentation/ABI/testing/sysfs-bus-iio-distance-srf08 index a133fd8d081a..40df5c9fef99 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio-distance-srf08 +++ b/Documentation/ABI/testing/sysfs-bus-iio-distance-srf08 @@ -15,8 +15,11 @@ Description: first object echoed in meters. Default value is 6.020. This setting limits the time the driver is waiting for a echo. + Showing the range of available values is represented as the minimum value, the step and the maximum value, all enclosed in square brackets. - Example: - [0.043 0.043 11.008] + + Example:: + + [0.043 0.043 11.008] diff --git a/Documentation/ABI/testing/sysfs-bus-iio-frequency-ad9523 b/Documentation/ABI/testing/sysfs-bus-iio-frequency-ad9523 index a91aeabe7b24..d065cda7dd96 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio-frequency-ad9523 +++ b/Documentation/ABI/testing/sysfs-bus-iio-frequency-ad9523 @@ -8,7 +8,9 @@ KernelVersion: 3.4.0 Contact: linux-iio@vger.kernel.org Description: Reading returns either '1' or '0'. + '1' means that the clock in question is present. + '0' means that the clock is missing. What: /sys/bus/iio/devices/iio:deviceX/pllY_locked diff --git a/Documentation/ABI/testing/sysfs-bus-iio-frequency-adf4371 b/Documentation/ABI/testing/sysfs-bus-iio-frequency-adf4371 index 302de64cb424..544548ee794c 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio-frequency-adf4371 +++ b/Documentation/ABI/testing/sysfs-bus-iio-frequency-adf4371 @@ -27,12 +27,12 @@ What: /sys/bus/iio/devices/iio:deviceX/out_altvoltageY_name KernelVersion: Contact: linux-iio@vger.kernel.org Description: - Reading returns the datasheet name for channel Y: + Reading returns the datasheet name for channel Y:: - out_altvoltage0_name: RF8x - out_altvoltage1_name: RFAUX8x - out_altvoltage2_name: RF16x - out_altvoltage3_name: RF32x + out_altvoltage0_name: RF8x + out_altvoltage1_name: RFAUX8x + out_altvoltage2_name: RF16x + out_altvoltage3_name: RF32x What: /sys/bus/iio/devices/iio:deviceX/out_altvoltageY_powerdown KernelVersion: diff --git a/Documentation/ABI/testing/sysfs-bus-iio-health-afe440x b/Documentation/ABI/testing/sysfs-bus-iio-health-afe440x index 6adba9058b22..66b621f10223 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio-health-afe440x +++ b/Documentation/ABI/testing/sysfs-bus-iio-health-afe440x @@ -6,10 +6,14 @@ Description: Get measured values from the ADC for these stages. Y is the specific stage number corresponding to datasheet stage names as follows: - 1 -> LED2 - 2 -> ALED2/LED3 - 3 -> LED1 - 4 -> ALED1/LED4 + + == ========== + 1 LED2 + 2 ALED2/LED3 + 3 LED1 + 4 ALED1/LED4 + == ========== + Note that channels 5 and 6 represent LED2-ALED2 and LED1-ALED1 respectively which simply helper channels containing the calculated difference in the value of stage 1 - 2 and 3 - 4. diff --git a/Documentation/ABI/testing/sysfs-bus-iio-light-isl29018 b/Documentation/ABI/testing/sysfs-bus-iio-light-isl29018 index f0ce0a0476ea..220206a20d98 100644 --- a/Documentation/ABI/testing/sysfs-bus-iio-light-isl29018 +++ b/Documentation/ABI/testing/sysfs-bus-iio-light-isl29018 @@ -15,5 +15,7 @@ Description: Scheme 0 has wider dynamic range, Scheme 1 proximity detection is less affected by the ambient IR noise variation. - 0 Sensing IR from LED and ambient - 1 Sensing IR from LED with ambient IR rejection + == ============================================= + 0 Sensing IR from LED and ambient + 1 Sensing IR from LED with ambient IR rejection + == ============================================= diff --git a/Documentation/ABI/testing/sysfs-bus-intel_th-devices-gth b/Documentation/ABI/testing/sysfs-bus-intel_th-devices-gth index 22d0843849a8..b7b2278fe042 100644 --- a/Documentation/ABI/testing/sysfs-bus-intel_th-devices-gth +++ b/Documentation/ABI/testing/sysfs-bus-intel_th-devices-gth @@ -10,10 +10,13 @@ Date: June 2015 KernelVersion: 4.3 Contact: Alexander Shishkin Description: (RO) Output port type: - 0: not present, - 1: MSU (Memory Storage Unit) - 2: CTP (Common Trace Port) - 4: PTI (MIPI PTI). + + == ========================= + 0 not present, + 1 MSU (Memory Storage Unit) + 2 CTP (Common Trace Port) + 4 PTI (MIPI PTI). + == ========================= What: /sys/bus/intel_th/devices/-gth/outputs/[0-7]_drop Date: June 2015 diff --git a/Documentation/ABI/testing/sysfs-bus-papr-pmem b/Documentation/ABI/testing/sysfs-bus-papr-pmem index c1a67275c43f..8316c33862a0 100644 --- a/Documentation/ABI/testing/sysfs-bus-papr-pmem +++ b/Documentation/ABI/testing/sysfs-bus-papr-pmem @@ -11,19 +11,26 @@ Description: at 'Documentation/powerpc/papr_hcalls.rst' . Below are the flags reported in this sysfs file: - * "not_armed" : Indicates that NVDIMM contents will not + * "not_armed" + Indicates that NVDIMM contents will not survive a power cycle. - * "flush_fail" : Indicates that NVDIMM contents + * "flush_fail" + Indicates that NVDIMM contents couldn't be flushed during last shut-down event. - * "restore_fail": Indicates that NVDIMM contents + * "restore_fail" + Indicates that NVDIMM contents couldn't be restored during NVDIMM initialization. - * "encrypted" : NVDIMM contents are encrypted. - * "smart_notify": There is health event for the NVDIMM. - * "scrubbed" : Indicating that contents of the + * "encrypted" + NVDIMM contents are encrypted. + * "smart_notify" + There is health event for the NVDIMM. + * "scrubbed" + Indicating that contents of the NVDIMM have been scrubbed. - * "locked" : Indicating that NVDIMM contents cant + * "locked" + Indicating that NVDIMM contents cant be modified until next power cycle. What: /sys/bus/nd/devices/nmemX/papr/perf_stats @@ -51,4 +58,4 @@ Description: * "MedWDur " : Media Write Duration * "CchRHCnt" : Cache Read Hit Count * "CchWHCnt" : Cache Write Hit Count - * "FastWCnt" : Fast Write Count \ No newline at end of file + * "FastWCnt" : Fast Write Count diff --git a/Documentation/ABI/testing/sysfs-bus-pci b/Documentation/ABI/testing/sysfs-bus-pci index 450296cc7948..77ad9ec3c801 100644 --- a/Documentation/ABI/testing/sysfs-bus-pci +++ b/Documentation/ABI/testing/sysfs-bus-pci @@ -7,8 +7,10 @@ Description: this location. This is useful for overriding default bindings. The format for the location is: DDDD:BB:DD.F. That is Domain:Bus:Device.Function and is the same as - found in /sys/bus/pci/devices/. For example: - # echo 0000:00:19.0 > /sys/bus/pci/drivers/foo/bind + found in /sys/bus/pci/devices/. For example:: + + # echo 0000:00:19.0 > /sys/bus/pci/drivers/foo/bind + (Note: kernels before 2.6.28 may require echo -n). What: /sys/bus/pci/drivers/.../unbind @@ -20,8 +22,10 @@ Description: this location. This may be useful when overriding default bindings. The format for the location is: DDDD:BB:DD.F. That is Domain:Bus:Device.Function and is the same as - found in /sys/bus/pci/devices/. For example: - # echo 0000:00:19.0 > /sys/bus/pci/drivers/foo/unbind + found in /sys/bus/pci/devices/. For example:: + + # echo 0000:00:19.0 > /sys/bus/pci/drivers/foo/unbind + (Note: kernels before 2.6.28 may require echo -n). What: /sys/bus/pci/drivers/.../new_id @@ -38,8 +42,9 @@ Description: Class, Class Mask, and Private Driver Data. The Vendor ID and Device ID fields are required, the rest are optional. Upon successfully adding an ID, the driver will probe - for the device and attempt to bind to it. For example: - # echo "8086 10f5" > /sys/bus/pci/drivers/foo/new_id + for the device and attempt to bind to it. For example:: + + # echo "8086 10f5" > /sys/bus/pci/drivers/foo/new_id What: /sys/bus/pci/drivers/.../remove_id Date: February 2009 @@ -54,8 +59,9 @@ Description: required, the rest are optional. After successfully removing an ID, the driver will no longer support the device. This is useful to ensure auto probing won't - match the driver to the device. For example: - # echo "8086 10f5" > /sys/bus/pci/drivers/foo/remove_id + match the driver to the device. For example:: + + # echo "8086 10f5" > /sys/bus/pci/drivers/foo/remove_id What: /sys/bus/pci/rescan Date: January 2009 diff --git a/Documentation/ABI/testing/sysfs-bus-pci-devices-catpt b/Documentation/ABI/testing/sysfs-bus-pci-devices-catpt index 8a200f4eefbd..f85db86d63e8 100644 --- a/Documentation/ABI/testing/sysfs-bus-pci-devices-catpt +++ b/Documentation/ABI/testing/sysfs-bus-pci-devices-catpt @@ -4,6 +4,7 @@ Contact: Cezary Rojewski Description: Version of AudioDSP firmware ASoC catpt driver is communicating with. + Format: %d.%d.%d.%d, type:major:minor:build. What: /sys/devices/pci0000:00//fw_info diff --git a/Documentation/ABI/testing/sysfs-bus-pci-drivers-ehci_hcd b/Documentation/ABI/testing/sysfs-bus-pci-drivers-ehci_hcd index 60c60fa624b2..c90d97a80855 100644 --- a/Documentation/ABI/testing/sysfs-bus-pci-drivers-ehci_hcd +++ b/Documentation/ABI/testing/sysfs-bus-pci-drivers-ehci_hcd @@ -21,11 +21,11 @@ Description: number returns the port to normal operation. For example: To force the high-speed device attached to - port 4 on bus 2 to run at full speed: + port 4 on bus 2 to run at full speed:: echo 4 >/sys/bus/usb/devices/usb2/../companion - To return the port to high-speed operation: + To return the port to high-speed operation:: echo -4 >/sys/bus/usb/devices/usb2/../companion diff --git a/Documentation/ABI/testing/sysfs-bus-rbd b/Documentation/ABI/testing/sysfs-bus-rbd index cc30bee8b5f4..417a2fe21be1 100644 --- a/Documentation/ABI/testing/sysfs-bus-rbd +++ b/Documentation/ABI/testing/sysfs-bus-rbd @@ -7,6 +7,8 @@ Description: Usage: [] + Example:: + $ echo "192.168.0.1 name=admin rbd foo" > /sys/bus/rbd/add The snapshot name can be "-" or omitted to map the image @@ -23,6 +25,8 @@ Description: Usage: [force] + Example:: + $ echo 2 > /sys/bus/rbd/remove Optional "force" argument which when passed will wait for @@ -80,26 +84,29 @@ Date: Oct, 2010 KernelVersion: v2.6.37 Contact: Sage Weil Description: - size: (RO) The size (in bytes) of the mapped block + + ============== ================================================ + size (RO) The size (in bytes) of the mapped block device. - major: (RO) The block device major number. + major (RO) The block device major number. - client_id: (RO) The ceph unique client id that was assigned + client_id (RO) The ceph unique client id that was assigned for this specific session. - pool: (RO) The name of the storage pool where this rbd + pool (RO) The name of the storage pool where this rbd image resides. An rbd image name is unique within its pool. - name: (RO) The name of the rbd image. + name (RO) The name of the rbd image. - refresh: (WO) Writing to this file will reread the image + refresh (WO) Writing to this file will reread the image header data and set all relevant data structures accordingly. - current_snap: (RO) The current snapshot for which the device + current_snap (RO) The current snapshot for which the device is mapped. + ============== ================================================ What: /sys/bus/rbd/devices//pool_id @@ -117,11 +124,13 @@ Date: Oct, 2012 KernelVersion: v3.7 Contact: Sage Weil Description: - image_id: (RO) The unique id for the rbd image. (For rbd + ========= =============================================== + image_id (RO) The unique id for the rbd image. (For rbd image format 1 this is empty.) - features: (RO) A hexadecimal encoding of the feature bits + features (RO) A hexadecimal encoding of the feature bits for this image. + ========= =============================================== What: /sys/bus/rbd/devices//parent @@ -149,14 +158,16 @@ Date: Aug, 2016 KernelVersion: v4.9 Contact: Sage Weil Description: - snap_id: (RO) The current snapshot's id. + ============ ================================================ + snap_id (RO) The current snapshot's id. - config_info: (RO) The string written into + config_info (RO) The string written into /sys/bus/rbd/add{,_single_major}. - cluster_fsid: (RO) The ceph cluster UUID. + cluster_fsid (RO) The ceph cluster UUID. - client_addr: (RO) The ceph unique client + client_addr (RO) The ceph unique client entity_addr_t (address + nonce). The format is
:/: '1.2.3.4:1234/5678' or '[1:2:3:4:5:6:7:8]:1234/5678'. + ============ ================================================ diff --git a/Documentation/ABI/testing/sysfs-bus-siox b/Documentation/ABI/testing/sysfs-bus-siox index c2a403f20b90..50e80238f30d 100644 --- a/Documentation/ABI/testing/sysfs-bus-siox +++ b/Documentation/ABI/testing/sysfs-bus-siox @@ -8,6 +8,7 @@ Description: When the file contains a "1" the bus is operated and periodically does a push-pull cycle to write and read data from the connected devices. + When writing a "0" or "1" the bus moves to the described state. What: /sys/bus/siox/devices/siox-X/device_add @@ -21,8 +22,10 @@ Description: to add a new device dynamically. is the name that is used to match to a driver (similar to the platform bus). and define the length of the input and output shift register in bytes respectively. + defines the 4 bit device type that is check to identify connection problems. + The new device is added to the end of the existing chain. What: /sys/bus/siox/devices/siox-X/device_remove diff --git a/Documentation/ABI/testing/sysfs-bus-thunderbolt b/Documentation/ABI/testing/sysfs-bus-thunderbolt index 171127294674..0b4ab9e4b8f4 100644 --- a/Documentation/ABI/testing/sysfs-bus-thunderbolt +++ b/Documentation/ABI/testing/sysfs-bus-thunderbolt @@ -193,10 +193,11 @@ Description: When new NVM image is written to the non-active NVM verification fails an error code is returned instead. This file will accept writing values "1" or "2" + - Writing "1" will flush the image to the storage - area and authenticate the image in one action. + area and authenticate the image in one action. - Writing "2" will run some basic validation on the image - and flush it to the storage area. + and flush it to the storage area. When read holds status of the last authentication operation if an error occurred during the process. This @@ -213,9 +214,11 @@ Description: This contains name of the property directory the XDomain question. Following directories are already reserved by the Apple XDomain specification: - network: IP/ethernet over Thunderbolt - targetdm: Target disk mode protocol over Thunderbolt - extdisp: External display mode protocol over Thunderbolt + ======== =============================================== + network IP/ethernet over Thunderbolt + targetdm Target disk mode protocol over Thunderbolt + extdisp External display mode protocol over Thunderbolt + ======== =============================================== What: /sys/bus/thunderbolt/devices/./modalias Date: Jan 2018 @@ -293,7 +296,8 @@ Description: For supported devices, automatically authenticate the new Thunderbo image when the device is disconnected from the host system. This file will accept writing values "1" or "2" + - Writing "1" will flush the image to the storage - area and prepare the device for authentication on disconnect. + area and prepare the device for authentication on disconnect. - Writing "2" will run some basic validation on the image - and flush it to the storage area. + and flush it to the storage area. diff --git a/Documentation/ABI/testing/sysfs-bus-usb b/Documentation/ABI/testing/sysfs-bus-usb index e449b8374f6a..bf2c1968525f 100644 --- a/Documentation/ABI/testing/sysfs-bus-usb +++ b/Documentation/ABI/testing/sysfs-bus-usb @@ -9,6 +9,7 @@ Description: by writing INTERFACE to /sys/bus/usb/drivers_probe This allows to avoid side-effects with drivers that need multiple interfaces. + A deauthorized interface cannot be probed or claimed. What: /sys/bus/usb/devices/usbX/interface_authorized_default @@ -216,6 +217,7 @@ Description: - Bit 0 of this field selects the "old" enumeration scheme, as it is considerably faster (it only causes one USB reset instead of 2). + The old enumeration scheme can also be selected globally using /sys/module/usbcore/parameters/old_scheme_first, but it is often not desirable as the new scheme was introduced to diff --git a/Documentation/ABI/testing/sysfs-class-backlight-driver-lm3533 b/Documentation/ABI/testing/sysfs-class-backlight-driver-lm3533 index c0e0a9ae7b3d..8251e78abc49 100644 --- a/Documentation/ABI/testing/sysfs-class-backlight-driver-lm3533 +++ b/Documentation/ABI/testing/sysfs-class-backlight-driver-lm3533 @@ -6,8 +6,10 @@ Description: Get the ALS output channel used as input in ALS-current-control mode (0, 1), where: - 0 - out_current0 (backlight 0) - 1 - out_current1 (backlight 1) + == ========================== + 0 out_current0 (backlight 0) + 1 out_current1 (backlight 1) + == ========================== What: /sys/class/backlight//als_en Date: May 2012 @@ -30,8 +32,10 @@ Contact: Johan Hovold Description: Set the brightness-mapping mode (0, 1), where: - 0 - exponential mode - 1 - linear mode + == ================ + 0 exponential mode + 1 linear mode + == ================ What: /sys/class/backlight//pwm Date: April 2012 @@ -40,9 +44,11 @@ Contact: Johan Hovold Description: Set the PWM-input control mask (5 bits), where: - bit 5 - PWM-input enabled in Zone 4 - bit 4 - PWM-input enabled in Zone 3 - bit 3 - PWM-input enabled in Zone 2 - bit 2 - PWM-input enabled in Zone 1 - bit 1 - PWM-input enabled in Zone 0 - bit 0 - PWM-input enabled + ===== =========================== + bit 5 PWM-input enabled in Zone 4 + bit 4 PWM-input enabled in Zone 3 + bit 3 PWM-input enabled in Zone 2 + bit 2 PWM-input enabled in Zone 1 + bit 1 PWM-input enabled in Zone 0 + bit 0 PWM-input enabled + ===== =========================== diff --git a/Documentation/ABI/testing/sysfs-class-bdi b/Documentation/ABI/testing/sysfs-class-bdi index d773d5697cf5..5402bd74ba43 100644 --- a/Documentation/ABI/testing/sysfs-class-bdi +++ b/Documentation/ABI/testing/sysfs-class-bdi @@ -24,7 +24,6 @@ default filesystems which do not provide their own BDI. Files under /sys/class/bdi// ---------------------------------- read_ahead_kb (read-write) diff --git a/Documentation/ABI/testing/sysfs-class-chromeos b/Documentation/ABI/testing/sysfs-class-chromeos index 5819699d66ec..74ece942722e 100644 --- a/Documentation/ABI/testing/sysfs-class-chromeos +++ b/Documentation/ABI/testing/sysfs-class-chromeos @@ -17,13 +17,14 @@ Date: August 2015 KernelVersion: 4.2 Description: Tell the EC to reboot in various ways. Options are: - "cancel": Cancel a pending reboot. - "ro": Jump to RO without rebooting. - "rw": Jump to RW without rebooting. - "cold": Cold reboot. - "disable-jump": Disable jump until next reboot. - "hibernate": Hibernate the EC. - "at-shutdown": Reboot after an AP shutdown. + + - "cancel": Cancel a pending reboot. + - "ro": Jump to RO without rebooting. + - "rw": Jump to RW without rebooting. + - "cold": Cold reboot. + - "disable-jump": Disable jump until next reboot. + - "hibernate": Hibernate the EC. + - "at-shutdown": Reboot after an AP shutdown. What: /sys/class/chromeos//version Date: August 2015 diff --git a/Documentation/ABI/testing/sysfs-class-cxl b/Documentation/ABI/testing/sysfs-class-cxl index a6f51a104c44..818f55970efb 100644 --- a/Documentation/ABI/testing/sysfs-class-cxl +++ b/Documentation/ABI/testing/sysfs-class-cxl @@ -217,6 +217,7 @@ Description: read/write card. A power cycle is required to load the image. "none" could be useful for debugging because the trace arrays are preserved. + "user" and "factory" means PERST will cause either the user or user or factory image to be loaded. Default is to reload on PERST whichever image the card has @@ -240,8 +241,11 @@ Contact: linuxppc-dev@lists.ozlabs.org Description: read/write Trust that when an image is reloaded via PERST, it will not have changed. - 0 = don't trust, the image may be different (default) - 1 = trust that the image will not change. + + == ================================================= + 0 don't trust, the image may be different (default) + 1 trust that the image will not change. + == ================================================= Users: https://github.com/ibm-capi/libcxl What: /sys/class/cxl//psl_timebase_synced diff --git a/Documentation/ABI/testing/sysfs-class-devlink b/Documentation/ABI/testing/sysfs-class-devlink index 64791b65c9a3..b662f747c83e 100644 --- a/Documentation/ABI/testing/sysfs-class-devlink +++ b/Documentation/ABI/testing/sysfs-class-devlink @@ -18,9 +18,9 @@ Description: This will be one of the following strings: - 'consumer unbind' - 'supplier unbind' - 'never' + - 'consumer unbind' + - 'supplier unbind' + - 'never' 'consumer unbind' means the device link will be removed when the consumer's driver is unbound from the consumer device. @@ -49,8 +49,10 @@ Description: This will be one of the following strings: - '0' - Does not affect runtime power management - '1' - Affects runtime power management + === ======================================== + '0' Does not affect runtime power management + '1' Affects runtime power management + === ======================================== What: /sys/class/devlink/.../status Date: May 2020 @@ -68,13 +70,13 @@ Description: This will be one of the following strings: - 'not tracked' - 'dormant' - 'available' - 'consumer probing' - 'active' - 'supplier unbinding' - 'unknown' + - 'not tracked' + - 'dormant' + - 'available' + - 'consumer probing' + - 'active' + - 'supplier unbinding' + - 'unknown' 'not tracked' means this device link does not track the status and has no impact on the binding, unbinding and syncing the @@ -114,8 +116,10 @@ Description: This will be one of the following strings: + === ================================ '0' - '1' - Affects runtime power management + '1' Affects runtime power management + === ================================ '0' means the device link can affect other device behaviors like binding/unbinding, suspend/resume, runtime power diff --git a/Documentation/ABI/testing/sysfs-class-extcon b/Documentation/ABI/testing/sysfs-class-extcon index 57a726232912..fde0fecd5de9 100644 --- a/Documentation/ABI/testing/sysfs-class-extcon +++ b/Documentation/ABI/testing/sysfs-class-extcon @@ -39,19 +39,22 @@ Description: callback. If the default callback for showing function is used, the - format is like this: - # cat state - USB_OTG=1 - HDMI=0 - TA=1 - EAR_JACK=0 - # + format is like this:: + + # cat state + USB_OTG=1 + HDMI=0 + TA=1 + EAR_JACK=0 + # + In this example, the extcon device has USB_OTG and TA cables attached and HDMI and EAR_JACK cables detached. In order to update the state of an extcon device, enter a hex - state number starting with 0x: - # echo 0xHEX > state + state number starting with 0x:: + + # echo 0xHEX > state This updates the whole state of the extcon device. Inputs of all the methods are required to meet the @@ -84,12 +87,13 @@ Contact: MyungJoo Ham Description: Shows the relations of mutually exclusiveness. For example, if the mutually_exclusive array of extcon device is - {0x3, 0x5, 0xC, 0x0}, then the output is: - # ls mutually_exclusive/ - 0x3 - 0x5 - 0xc - # + {0x3, 0x5, 0xC, 0x0}, then the output is:: + + # ls mutually_exclusive/ + 0x3 + 0x5 + 0xc + # Note that mutually_exclusive is a sub-directory of the extcon device and the file names under the mutually_exclusive diff --git a/Documentation/ABI/testing/sysfs-class-fpga-manager b/Documentation/ABI/testing/sysfs-class-fpga-manager index 5284fa33d4c5..d78689c357a5 100644 --- a/Documentation/ABI/testing/sysfs-class-fpga-manager +++ b/Documentation/ABI/testing/sysfs-class-fpga-manager @@ -28,8 +28,7 @@ Description: Read fpga manager state as a string. * firmware request = firmware class request in progress * firmware request error = firmware request failed * write init = preparing FPGA for programming - * write init error = Error while preparing FPGA for - programming + * write init error = Error while preparing FPGA for programming * write = FPGA ready to receive image data * write error = Error while programming * write complete = Doing post programming steps @@ -47,7 +46,7 @@ Description: Read fpga manager status as a string. programming errors to userspace. This is a list of strings for the supported status. - * reconfig operation error - invalid operations detected by + * reconfig operation error - invalid operations detected by reconfiguration hardware. e.g. start reconfiguration with errors not cleared diff --git a/Documentation/ABI/testing/sysfs-class-gnss b/Documentation/ABI/testing/sysfs-class-gnss index 2467b6900eae..c8553d972edd 100644 --- a/Documentation/ABI/testing/sysfs-class-gnss +++ b/Documentation/ABI/testing/sysfs-class-gnss @@ -6,9 +6,11 @@ Description: The GNSS receiver type. The currently identified types reflect the protocol(s) supported by the receiver: + ====== =========== "NMEA" NMEA 0183 "SiRF" SiRF Binary "UBX" UBX + ====== =========== Note that also non-"NMEA" type receivers typically support a subset of NMEA 0183 with vendor extensions (e.g. to allow diff --git a/Documentation/ABI/testing/sysfs-class-led b/Documentation/ABI/testing/sysfs-class-led index 65e040978f73..0ed5c2629c6f 100644 --- a/Documentation/ABI/testing/sysfs-class-led +++ b/Documentation/ABI/testing/sysfs-class-led @@ -47,6 +47,7 @@ Contact: Richard Purdie Description: Set the trigger for this LED. A trigger is a kernel based source of LED events. + You can change triggers in a similar manner to the way an IO scheduler is chosen. Trigger specific parameters can appear in /sys/class/leds/ once a given trigger is selected. For diff --git a/Documentation/ABI/testing/sysfs-class-led-driver-el15203000 b/Documentation/ABI/testing/sysfs-class-led-driver-el15203000 index 69befe947d7e..da546e86deb5 100644 --- a/Documentation/ABI/testing/sysfs-class-led-driver-el15203000 +++ b/Documentation/ABI/testing/sysfs-class-led-driver-el15203000 @@ -27,23 +27,23 @@ Description: ^ | - 0 On -|----+ +----+ +--- + 0 On -|----+ +----+ +--- | | | | | Off-| +-------------------+ +-------------------+ | - 1 On -| +----+ +----+ + 1 On -| +----+ +----+ | | | | | Off |----+ +-------------------+ +------------------ | - 2 On -| +----+ +----+ + 2 On -| +----+ +----+ | | | | | Off-|---------+ +-------------------+ +------------- | - 3 On -| +----+ +----+ + 3 On -| +----+ +----+ | | | | | Off-|--------------+ +-------------------+ +-------- | - 4 On -| +----+ +----+ + 4 On -| +----+ +----+ | | | | | Off-|-------------------+ +-------------------+ +--- | @@ -55,23 +55,23 @@ Description: ^ | - 0 On -| +-------------------+ +-------------------+ + 0 On -| +-------------------+ +-------------------+ | | | | | Off-|----+ +----+ +--- | - 1 On -|----+ +-------------------+ +------------------ + 1 On -|----+ +-------------------+ +------------------ | | | | | Off | +----+ +----+ | - 2 On -|---------+ +-------------------+ +------------- + 2 On -|---------+ +-------------------+ +------------- | | | | | Off-| +----+ +----+ | - 3 On -|--------------+ +-------------------+ +-------- + 3 On -|--------------+ +-------------------+ +-------- | | | | | Off-| +----+ +----+ | - 4 On -|-------------------+ +-------------------+ +--- + 4 On -|-------------------+ +-------------------+ +--- | | | | | Off-| +----+ +----+ | @@ -83,23 +83,23 @@ Description: ^ | - 0 On -|----+ +-------- + 0 On -|----+ +-------- | | | Off-| +---------------------------------------+ | - 1 On -| +----+ +----+ + 1 On -| +----+ +----+ | | | | | Off |----+ +-----------------------------+ +-------- | - 2 On -| +----+ +----+ + 2 On -| +----+ +----+ | | | | | Off-|---------+ +-------------------+ +------------- | - 3 On -| +----+ +----+ + 3 On -| +----+ +----+ | | | | | Off-|--------------+ +---------+ +------------------ | - 4 On -| +---------+ + 4 On -| +---------+ | | | Off-|-------------------+ +----------------------- | diff --git a/Documentation/ABI/testing/sysfs-class-led-driver-lm3533 b/Documentation/ABI/testing/sysfs-class-led-driver-lm3533 index e4c89b261546..e38a835d0a85 100644 --- a/Documentation/ABI/testing/sysfs-class-led-driver-lm3533 +++ b/Documentation/ABI/testing/sysfs-class-led-driver-lm3533 @@ -6,8 +6,10 @@ Description: Set the ALS output channel to use as input in ALS-current-control mode (1, 2), where: - 1 - out_current1 - 2 - out_current2 + == ============ + 1 out_current1 + 2 out_current2 + == ============ What: /sys/class/leds//als_en Date: May 2012 @@ -24,14 +26,16 @@ Contact: Johan Hovold Description: Set the pattern generator fall and rise times (0..7), where: - 0 - 2048 us - 1 - 262 ms - 2 - 524 ms - 3 - 1.049 s - 4 - 2.097 s - 5 - 4.194 s - 6 - 8.389 s - 7 - 16.78 s + == ======= + 0 2048 us + 1 262 ms + 2 524 ms + 3 1.049 s + 4 2.097 s + 5 4.194 s + 6 8.389 s + 7 16.78 s + == ======= What: /sys/class/leds//id Date: April 2012 @@ -47,8 +51,10 @@ Contact: Johan Hovold Description: Set the brightness-mapping mode (0, 1), where: - 0 - exponential mode - 1 - linear mode + == ================ + 0 exponential mode + 1 linear mode + == ================ What: /sys/class/leds//pwm Date: April 2012 @@ -57,9 +63,11 @@ Contact: Johan Hovold Description: Set the PWM-input control mask (5 bits), where: - bit 5 - PWM-input enabled in Zone 4 - bit 4 - PWM-input enabled in Zone 3 - bit 3 - PWM-input enabled in Zone 2 - bit 2 - PWM-input enabled in Zone 1 - bit 1 - PWM-input enabled in Zone 0 - bit 0 - PWM-input enabled + ===== =========================== + bit 5 PWM-input enabled in Zone 4 + bit 4 PWM-input enabled in Zone 3 + bit 3 PWM-input enabled in Zone 2 + bit 2 PWM-input enabled in Zone 1 + bit 1 PWM-input enabled in Zone 0 + bit 0 PWM-input enabled + ===== =========================== diff --git a/Documentation/ABI/testing/sysfs-class-led-flash b/Documentation/ABI/testing/sysfs-class-led-flash index 220a0270b47b..11e5677c3672 100644 --- a/Documentation/ABI/testing/sysfs-class-led-flash +++ b/Documentation/ABI/testing/sysfs-class-led-flash @@ -55,26 +55,35 @@ Description: read only Flash faults are re-read after strobing the flash. Possible flash faults: - * led-over-voltage - flash controller voltage to the flash LED + * led-over-voltage + flash controller voltage to the flash LED has exceeded the limit specific to the flash controller - * flash-timeout-exceeded - the flash strobe was still on when + * flash-timeout-exceeded + the flash strobe was still on when the timeout set by the user has expired; not all flash controllers may set this in all such conditions - * controller-over-temperature - the flash controller has + * controller-over-temperature + the flash controller has overheated - * controller-short-circuit - the short circuit protection + * controller-short-circuit + the short circuit protection of the flash controller has been triggered - * led-power-supply-over-current - current in the LED power + * led-power-supply-over-current + current in the LED power supply has exceeded the limit specific to the flash controller - * indicator-led-fault - the flash controller has detected + * indicator-led-fault + the flash controller has detected a short or open circuit condition on the indicator LED - * led-under-voltage - flash controller voltage to the flash + * led-under-voltage + flash controller voltage to the flash LED has been below the minimum limit specific to the flash - * controller-under-voltage - the input voltage of the flash + * controller-under-voltage + the input voltage of the flash controller is below the limit under which strobing the flash at full current will not be possible; the condition persists until this flag is no longer set - * led-over-temperature - the temperature of the LED has exceeded + * led-over-temperature + the temperature of the LED has exceeded its allowed upper limit diff --git a/Documentation/ABI/testing/sysfs-class-led-trigger-netdev b/Documentation/ABI/testing/sysfs-class-led-trigger-netdev index 451af6d6768c..646540950e38 100644 --- a/Documentation/ABI/testing/sysfs-class-led-trigger-netdev +++ b/Documentation/ABI/testing/sysfs-class-led-trigger-netdev @@ -19,18 +19,23 @@ KernelVersion: 4.16 Contact: linux-leds@vger.kernel.org Description: Signal the link state of the named network device. + If set to 0 (default), the LED's normal state is off. + If set to 1, the LED's normal state reflects the link state of the named network device. Setting this value also immediately changes the LED state. + What: /sys/class/leds//tx Date: Dec 2017 KernelVersion: 4.16 Contact: linux-leds@vger.kernel.org Description: Signal transmission of data on the named network device. + If set to 0 (default), the LED will not blink on transmission. + If set to 1, the LED will blink for the milliseconds specified in interval to signal transmission. @@ -40,6 +45,8 @@ KernelVersion: 4.16 Contact: linux-leds@vger.kernel.org Description: Signal reception of data on the named network device. + If set to 0 (default), the LED will not blink on reception. + If set to 1, the LED will blink for the milliseconds specified in interval to signal reception. diff --git a/Documentation/ABI/testing/sysfs-class-led-trigger-usbport b/Documentation/ABI/testing/sysfs-class-led-trigger-usbport index f440e690daef..eb81152b8348 100644 --- a/Documentation/ABI/testing/sysfs-class-led-trigger-usbport +++ b/Documentation/ABI/testing/sysfs-class-led-trigger-usbport @@ -8,5 +8,6 @@ Description: selected for the USB port trigger. Selecting ports makes trigger observing them for any connected devices and lighting on LED if there are any. + Echoing "1" value selects USB port. Echoing "0" unselects it. Current state can be also read. diff --git a/Documentation/ABI/testing/sysfs-class-leds-gt683r b/Documentation/ABI/testing/sysfs-class-leds-gt683r index 6adab27f646e..b57ffb26e722 100644 --- a/Documentation/ABI/testing/sysfs-class-leds-gt683r +++ b/Documentation/ABI/testing/sysfs-class-leds-gt683r @@ -7,9 +7,11 @@ Description: of one LED will update the mode of its two sibling devices as well. Possible values are: - 0 - normal - 1 - audio - 2 - breathing + == ========= + 0 normal + 1 audio + 2 breathing + == ========= Normal: LEDs are fully on when enabled Audio: LEDs brightness depends on sound level diff --git a/Documentation/ABI/testing/sysfs-class-net b/Documentation/ABI/testing/sysfs-class-net index 3b404577f380..7670012ae9b6 100644 --- a/Documentation/ABI/testing/sysfs-class-net +++ b/Documentation/ABI/testing/sysfs-class-net @@ -4,10 +4,13 @@ KernelVersion: 3.17 Contact: netdev@vger.kernel.org Description: Indicates the name assignment type. Possible values are: - 1: enumerated by the kernel, possibly in an unpredictable way - 2: predictably named by the kernel - 3: named by userspace - 4: renamed + + == ========================================================== + 1 enumerated by the kernel, possibly in an unpredictable way + 2 predictably named by the kernel + 3 named by userspace + 4 renamed + == ========================================================== What: /sys/class/net//addr_assign_type Date: July 2010 @@ -15,10 +18,13 @@ KernelVersion: 3.2 Contact: netdev@vger.kernel.org Description: Indicates the address assignment type. Possible values are: - 0: permanent address - 1: randomly generated - 2: stolen from another device - 3: set using dev_set_mac_address + + == ============================= + 0 permanent address + 1 randomly generated + 2 stolen from another device + 3 set using dev_set_mac_address + == ============================= What: /sys/class/net//addr_len Date: April 2005 @@ -51,9 +57,12 @@ Description: Default value 0 does not forward any link local frames. Restricted bits: - 0: 01-80-C2-00-00-00 Bridge Group Address used for STP - 1: 01-80-C2-00-00-01 (MAC Control) 802.3 used for MAC PAUSE - 2: 01-80-C2-00-00-02 (Link Aggregation) 802.3ad + + == ======================================================== + 0 01-80-C2-00-00-00 Bridge Group Address used for STP + 1 01-80-C2-00-00-01 (MAC Control) 802.3 used for MAC PAUSE + 2 01-80-C2-00-00-02 (Link Aggregation) 802.3ad + == ======================================================== Any values not setting these bits can be used. Take special care when forwarding control frames e.g. 802.1X-PAE or LLDP. @@ -74,8 +83,11 @@ Contact: netdev@vger.kernel.org Description: Indicates the current physical link state of the interface. Posssible values are: - 0: physical link is down - 1: physical link is up + + == ===================== + 0 physical link is down + 1 physical link is up + == ===================== Note: some special devices, e.g: bonding and team drivers will allow this attribute to be written to force a link state for @@ -131,8 +143,11 @@ Contact: netdev@vger.kernel.org Description: Indicates whether the interface is under test. Possible values are: - 0: interface is not being tested - 1: interface is being tested + + == ============================= + 0 interface is not being tested + 1 interface is being tested + == ============================= When an interface is under test, it cannot be expected to pass packets as normal. @@ -144,8 +159,11 @@ Contact: netdev@vger.kernel.org Description: Indicates the interface latest or current duplex value. Possible values are: - half: half duplex - full: full duplex + + ==== =========== + half half duplex + full full duplex + ==== =========== Note: This attribute is only valid for interfaces that implement the ethtool get_link_ksettings method (mostly Ethernet). @@ -196,8 +214,11 @@ Description: Indicates the interface link mode, as a decimal number. This attribute should be used in conjunction with 'dormant' attribute to determine the interface usability. Possible values: - 0: default link mode - 1: dormant link mode + + == ================= + 0 default link mode + 1 dormant link mode + == ================= What: /sys/class/net//mtu Date: April 2005 @@ -226,7 +247,9 @@ KernelVersion: 2.6.17 Contact: netdev@vger.kernel.org Description: Indicates the interface RFC2863 operational state as a string. + Possible values are: + "unknown", "notpresent", "down", "lowerlayerdown", "testing", "dormant", "up". diff --git a/Documentation/ABI/testing/sysfs-class-net-cdc_ncm b/Documentation/ABI/testing/sysfs-class-net-cdc_ncm index f7be0e88b139..06416d0e163d 100644 --- a/Documentation/ABI/testing/sysfs-class-net-cdc_ncm +++ b/Documentation/ABI/testing/sysfs-class-net-cdc_ncm @@ -91,9 +91,9 @@ Date: May 2014 KernelVersion: 3.16 Contact: Bjørn Mork Description: - Bit 0: 16-bit NTB supported (set to 1) - Bit 1: 32-bit NTB supported - Bits 2 – 15: reserved (reset to zero; must be ignored by host) + - Bit 0: 16-bit NTB supported (set to 1) + - Bit 1: 32-bit NTB supported + - Bits 2 – 15: reserved (reset to zero; must be ignored by host) What: /sys/class/net//cdc_ncm/dwNtbInMaxSize Date: May 2014 diff --git a/Documentation/ABI/testing/sysfs-class-net-phydev b/Documentation/ABI/testing/sysfs-class-net-phydev index 206cbf538b59..40ced0ea4316 100644 --- a/Documentation/ABI/testing/sysfs-class-net-phydev +++ b/Documentation/ABI/testing/sysfs-class-net-phydev @@ -35,7 +35,9 @@ Description: Ethernet driver during bus enumeration, encoded in string. This interface mode is used to configure the Ethernet MAC with the appropriate mode for its data lines to the PHY hardware. + Possible values are: + (not available), mii, gmii, sgmii, tbi, rev-mii, rmii, rgmii, rgmii-id, rgmii-rxid, rgmii-txid, rtbi, smii xgmii, moca, qsgmii, trgmii, 1000base-x, 2500base-x, rxaui, diff --git a/Documentation/ABI/testing/sysfs-class-pktcdvd b/Documentation/ABI/testing/sysfs-class-pktcdvd index dde4f26d0780..ba1ce626591d 100644 --- a/Documentation/ABI/testing/sysfs-class-pktcdvd +++ b/Documentation/ABI/testing/sysfs-class-pktcdvd @@ -11,15 +11,17 @@ KernelVersion: 2.6.20 Contact: Thomas Maier Description: - add: (WO) Write a block device id (major:minor) to + ========== ============================================== + add (WO) Write a block device id (major:minor) to create a new pktcdvd device and map it to the block device. - remove: (WO) Write the pktcdvd device id (major:minor) + remove (WO) Write the pktcdvd device id (major:minor) to remove the pktcdvd device. - device_map: (RO) Shows the device mapping in format: + device_map (RO) Shows the device mapping in format: pktcdvd[0-7] + ========== ============================================== What: /sys/class/pktcdvd/pktcdvd[0-7]/dev @@ -65,29 +67,31 @@ Date: Oct. 2006 KernelVersion: 2.6.20 Contact: Thomas Maier Description: - size: (RO) Contains the size of the bio write queue. + ============== ================================================ + size (RO) Contains the size of the bio write queue. - congestion_off: (RW) If bio write queue size is below this mark, + congestion_off (RW) If bio write queue size is below this mark, accept new bio requests from the block layer. - congestion_on: (RW) If bio write queue size is higher as this + congestion_on (RW) If bio write queue size is higher as this mark, do no longer accept bio write requests from the block layer and wait till the pktcdvd device has processed enough bio's so that bio write queue size is below congestion off mark. A value of <= 0 disables congestion control. + ============== ================================================ Example: -------- -To use the pktcdvd sysfs interface directly, you can do: - -# create a new pktcdvd device mapped to /dev/hdc -echo "22:0" >/sys/class/pktcdvd/add -cat /sys/class/pktcdvd/device_map -# assuming device pktcdvd0 was created, look at stat's -cat /sys/class/pktcdvd/pktcdvd0/stat/kb_written -# print the device id of the mapped block device -fgrep pktcdvd0 /sys/class/pktcdvd/device_map -# remove device, using pktcdvd0 device id 253:0 -echo "253:0" >/sys/class/pktcdvd/remove +To use the pktcdvd sysfs interface directly, you can do:: + + # create a new pktcdvd device mapped to /dev/hdc + echo "22:0" >/sys/class/pktcdvd/add + cat /sys/class/pktcdvd/device_map + # assuming device pktcdvd0 was created, look at stat's + cat /sys/class/pktcdvd/pktcdvd0/stat/kb_written + # print the device id of the mapped block device + fgrep pktcdvd0 /sys/class/pktcdvd/device_map + # remove device, using pktcdvd0 device id 253:0 + echo "253:0" >/sys/class/pktcdvd/remove diff --git a/Documentation/ABI/testing/sysfs-class-power b/Documentation/ABI/testing/sysfs-class-power index d4319a04c302..d68ad528a8e5 100644 --- a/Documentation/ABI/testing/sysfs-class-power +++ b/Documentation/ABI/testing/sysfs-class-power @@ -43,7 +43,9 @@ Date: May 2007 Contact: linux-pm@vger.kernel.org Description: Fine grain representation of battery capacity. + Access: Read + Valid values: 0 - 100 (percent) What: /sys/class/power_supply//capacity_alert_max @@ -58,6 +60,7 @@ Description: low). Access: Read, Write + Valid values: 0 - 100 (percent) What: /sys/class/power_supply//capacity_alert_min @@ -88,6 +91,7 @@ Description: completely useless. Access: Read + Valid values: 0 - 100 (percent) What: /sys/class/power_supply//capacity_level @@ -111,6 +115,7 @@ Description: which they average readings to smooth out the reported value. Access: Read + Valid values: Represented in microamps. Negative values are used for discharging batteries, positive values for charging batteries. @@ -131,6 +136,7 @@ Description: This value is not averaged/smoothed. Access: Read + Valid values: Represented in microamps. Negative values are used for discharging batteries, positive values for charging batteries. @@ -383,7 +389,7 @@ Description: **USB Properties** -What: /sys/class/power_supply//current_avg +What: /sys/class/power_supply//current_avg Date: May 2007 Contact: linux-pm@vger.kernel.org Description: @@ -449,6 +455,7 @@ Description: solved using power limit use input_voltage_limit. Access: Read, Write + Valid values: Represented in microvolts What: /sys/class/power_supply//input_power_limit @@ -462,6 +469,7 @@ Description: limit only for problems that can be solved using power limit. Access: Read, Write + Valid values: Represented in microwatts What: /sys/class/power_supply//online, @@ -747,6 +755,7 @@ Description: manufactured. Access: Read + Valid values: Reported as integer What: /sys/class/power_supply//manufacture_month @@ -756,6 +765,7 @@ Description: Reports the month when the device has been manufactured. Access: Read + Valid values: 1-12 What: /sys/class/power_supply//manufacture_day diff --git a/Documentation/ABI/testing/sysfs-class-power-mp2629 b/Documentation/ABI/testing/sysfs-class-power-mp2629 index 327a07e22805..914d67caac0d 100644 --- a/Documentation/ABI/testing/sysfs-class-power-mp2629 +++ b/Documentation/ABI/testing/sysfs-class-power-mp2629 @@ -5,4 +5,5 @@ Description: Represents a battery impedance compensation to accelerate charging. Access: Read, Write + Valid values: Represented in milli-ohms. Valid range is [0, 140]. diff --git a/Documentation/ABI/testing/sysfs-class-power-twl4030 b/Documentation/ABI/testing/sysfs-class-power-twl4030 index 7ac36dba87bc..b52f7023f8ba 100644 --- a/Documentation/ABI/testing/sysfs-class-power-twl4030 +++ b/Documentation/ABI/testing/sysfs-class-power-twl4030 @@ -6,9 +6,9 @@ Description: Possible values are: ============= =========================================== - "auto" draw power as appropriate for detected + "auto" draw power as appropriate for detected power source and battery status. - "off" do not draw any power. + "off" do not draw any power. "continuous" activate mode described as "linear" in TWL data sheets. This uses whatever current is available and doesn't switch off diff --git a/Documentation/ABI/testing/sysfs-class-rapidio b/Documentation/ABI/testing/sysfs-class-rapidio index 8716beeb16c1..19aefb21b639 100644 --- a/Documentation/ABI/testing/sysfs-class-rapidio +++ b/Documentation/ABI/testing/sysfs-class-rapidio @@ -6,6 +6,7 @@ Description: The /sys/class/rapidio_port subdirectory contains individual subdirectories named as "rapidioN" where N = mport ID registered with RapidIO subsystem. + NOTE: An mport ID is not a RapidIO destination ID assigned to a given local mport device. @@ -16,7 +17,9 @@ Contact: Matt Porter , Alexandre Bounine Description: (RO) reports RapidIO common transport system size: + 0 = small (8-bit destination ID, max. 256 devices), + 1 = large (16-bit destination ID, max. 65536 devices). What: /sys/class/rapidio_port/rapidioN/port_destid @@ -25,31 +28,32 @@ KernelVersion: v3.15 Contact: Matt Porter , Alexandre Bounine Description: - (RO) reports RapidIO destination ID assigned to the given - RapidIO mport device. If value 0xFFFFFFFF is returned this means - that no valid destination ID have been assigned to the mport - (yet). Normally, before enumeration/discovery have been executed - only fabric enumerating mports have a valid destination ID - assigned to them using "hdid=..." rapidio module parameter. + +(RO) reports RapidIO destination ID assigned to the given +RapidIO mport device. If value 0xFFFFFFFF is returned this means +that no valid destination ID have been assigned to the mport +(yet). Normally, before enumeration/discovery have been executed +only fabric enumerating mports have a valid destination ID +assigned to them using "hdid=..." rapidio module parameter. After enumeration or discovery was performed for a given mport device, the corresponding subdirectory will also contain subdirectories for each child RapidIO device connected to the mport. The example below shows mport device subdirectory with several child RapidIO -devices attached to it. - -[rio@rapidio ~]$ ls /sys/class/rapidio_port/rapidio0/ -l -total 0 -drwxr-xr-x 3 root root 0 Feb 11 15:10 00:e:0001 -drwxr-xr-x 3 root root 0 Feb 11 15:10 00:e:0004 -drwxr-xr-x 3 root root 0 Feb 11 15:10 00:e:0007 -drwxr-xr-x 3 root root 0 Feb 11 15:10 00:s:0002 -drwxr-xr-x 3 root root 0 Feb 11 15:10 00:s:0003 -drwxr-xr-x 3 root root 0 Feb 11 15:10 00:s:0005 -lrwxrwxrwx 1 root root 0 Feb 11 15:11 device -> ../../../0000:01:00.0 --r--r--r-- 1 root root 4096 Feb 11 15:11 port_destid -drwxr-xr-x 2 root root 0 Feb 11 15:11 power -lrwxrwxrwx 1 root root 0 Feb 11 15:04 subsystem -> ../../../../../../class/rapidio_port --r--r--r-- 1 root root 4096 Feb 11 15:11 sys_size --rw-r--r-- 1 root root 4096 Feb 11 15:04 uevent +devices attached to it:: + + [rio@rapidio ~]$ ls /sys/class/rapidio_port/rapidio0/ -l + total 0 + drwxr-xr-x 3 root root 0 Feb 11 15:10 00:e:0001 + drwxr-xr-x 3 root root 0 Feb 11 15:10 00:e:0004 + drwxr-xr-x 3 root root 0 Feb 11 15:10 00:e:0007 + drwxr-xr-x 3 root root 0 Feb 11 15:10 00:s:0002 + drwxr-xr-x 3 root root 0 Feb 11 15:10 00:s:0003 + drwxr-xr-x 3 root root 0 Feb 11 15:10 00:s:0005 + lrwxrwxrwx 1 root root 0 Feb 11 15:11 device -> ../../../0000:01:00.0 + -r--r--r-- 1 root root 4096 Feb 11 15:11 port_destid + drwxr-xr-x 2 root root 0 Feb 11 15:11 power + lrwxrwxrwx 1 root root 0 Feb 11 15:04 subsystem -> ../../../../../../class/rapidio_port + -r--r--r-- 1 root root 4096 Feb 11 15:11 sys_size + -rw-r--r-- 1 root root 4096 Feb 11 15:04 uevent diff --git a/Documentation/ABI/testing/sysfs-class-regulator b/Documentation/ABI/testing/sysfs-class-regulator index bc578bc60628..8516f08806dd 100644 --- a/Documentation/ABI/testing/sysfs-class-regulator +++ b/Documentation/ABI/testing/sysfs-class-regulator @@ -35,13 +35,13 @@ Description: This will be one of the following strings: - off - on - error - fast - normal - idle - standby + - off + - on + - error + - fast + - normal + - idle + - standby "off" means the regulator is not supplying power to the system. @@ -74,9 +74,9 @@ Description: This will be one of the following strings: - 'voltage' - 'current' - 'unknown' + - 'voltage' + - 'current' + - 'unknown' 'voltage' means the regulator output voltage can be controlled by software. @@ -129,11 +129,11 @@ Description: The opmode value can be one of the following strings: - 'fast' - 'normal' - 'idle' - 'standby' - 'unknown' + - 'fast' + - 'normal' + - 'idle' + - 'standby' + - 'unknown' The modes are described in include/linux/regulator/consumer.h @@ -360,9 +360,9 @@ Description: This will be one of the following strings: - 'enabled' - 'disabled' - 'unknown' + - 'enabled' + - 'disabled' + - 'unknown' 'enabled' means the regulator is in bypass mode. diff --git a/Documentation/ABI/testing/sysfs-class-remoteproc b/Documentation/ABI/testing/sysfs-class-remoteproc index 066b9b6f4924..0c9ee55098b8 100644 --- a/Documentation/ABI/testing/sysfs-class-remoteproc +++ b/Documentation/ABI/testing/sysfs-class-remoteproc @@ -16,11 +16,11 @@ Description: Remote processor state Reports the state of the remote processor, which will be one of: - "offline" - "suspended" - "running" - "crashed" - "invalid" + - "offline" + - "suspended" + - "running" + - "crashed" + - "invalid" "offline" means the remote processor is powered off. @@ -38,8 +38,8 @@ Description: Remote processor state Writing this file controls the state of the remote processor. The following states can be written: - "start" - "stop" + - "start" + - "stop" Writing "start" will attempt to start the processor running the firmware indicated by, or written to, diff --git a/Documentation/ABI/testing/sysfs-class-rtc-rtc0-device-rtc_calibration b/Documentation/ABI/testing/sysfs-class-rtc-rtc0-device-rtc_calibration index ec950c93e5c6..ee8ed6494a01 100644 --- a/Documentation/ABI/testing/sysfs-class-rtc-rtc0-device-rtc_calibration +++ b/Documentation/ABI/testing/sysfs-class-rtc-rtc0-device-rtc_calibration @@ -7,6 +7,7 @@ Description: Attribute for calibrating ST-Ericsson AB8500 Real Time Clock calibrate the AB8500.s 32KHz Real Time Clock. Every 60 seconds the AB8500 will correct the RTC's value by adding to it the value of this attribute. + The range of the attribute is -127 to +127 in units of 30.5 micro-seconds (half-parts-per-million of the 32KHz clock) Users: The /vendor/st-ericsson/base_utilities/core/rtc_calibration diff --git a/Documentation/ABI/testing/sysfs-class-uwb_rc b/Documentation/ABI/testing/sysfs-class-uwb_rc index a0578751c1e3..6c5dcad21e19 100644 --- a/Documentation/ABI/testing/sysfs-class-uwb_rc +++ b/Documentation/ABI/testing/sysfs-class-uwb_rc @@ -66,11 +66,14 @@ Description: [] to start (or stop) scanning on a channel. is one of: - 0 - scan - 1 - scan outside BP - 2 - scan while inactive - 3 - scanning disabled - 4 - scan (with start time of ) + + == ======================================= + 0 scan + 1 scan outside BP + 2 scan while inactive + 3 scanning disabled + 4 scan (with start time of ) + == ======================================= What: /sys/class/uwb_rc/uwbN/mac_address Date: July 2008 diff --git a/Documentation/ABI/testing/sysfs-class-watchdog b/Documentation/ABI/testing/sysfs-class-watchdog index 9860a8b2ba75..585caecda3a5 100644 --- a/Documentation/ABI/testing/sysfs-class-watchdog +++ b/Documentation/ABI/testing/sysfs-class-watchdog @@ -91,10 +91,13 @@ Description: h/w strapping (for WDT2 only). At alternate flash the 'access_cs0' sysfs node provides: - ast2400: a way to get access to the primary SPI flash + + ast2400: + a way to get access to the primary SPI flash chip at CS0 after booting from the alternate chip at CS1. - ast2500: a way to restore the normal address mapping + ast2500: + a way to restore the normal address mapping from (CS0->CS1, CS1->CS0) to (CS0->CS0, CS1->CS1). diff --git a/Documentation/ABI/testing/sysfs-dev b/Documentation/ABI/testing/sysfs-dev index a9f2b8b0530f..d1739063e762 100644 --- a/Documentation/ABI/testing/sysfs-dev +++ b/Documentation/ABI/testing/sysfs-dev @@ -9,9 +9,10 @@ Description: The /sys/dev tree provides a method to look up the sysfs the form ":". These links point to the corresponding sysfs path for the given device. - Example: - $ readlink /sys/dev/block/8:32 - ../../block/sdc + Example:: + + $ readlink /sys/dev/block/8:32 + ../../block/sdc Entries in /sys/dev/char and /sys/dev/block will be dynamically created and destroyed as devices enter and diff --git a/Documentation/ABI/testing/sysfs-devices-mapping b/Documentation/ABI/testing/sysfs-devices-mapping index 490ccfd67f12..8d202bac9394 100644 --- a/Documentation/ABI/testing/sysfs-devices-mapping +++ b/Documentation/ABI/testing/sysfs-devices-mapping @@ -8,26 +8,27 @@ Description: block. For example, on 4-die Xeon platform with up to 6 IIO stacks per die and, therefore, 6 IIO PMON blocks per die, the mapping of - IIO PMON block 0 exposes as the following: + IIO PMON block 0 exposes as the following:: - $ ls /sys/devices/uncore_iio_0/die* - -r--r--r-- /sys/devices/uncore_iio_0/die0 - -r--r--r-- /sys/devices/uncore_iio_0/die1 - -r--r--r-- /sys/devices/uncore_iio_0/die2 - -r--r--r-- /sys/devices/uncore_iio_0/die3 + $ ls /sys/devices/uncore_iio_0/die* + -r--r--r-- /sys/devices/uncore_iio_0/die0 + -r--r--r-- /sys/devices/uncore_iio_0/die1 + -r--r--r-- /sys/devices/uncore_iio_0/die2 + -r--r--r-- /sys/devices/uncore_iio_0/die3 - $ tail /sys/devices/uncore_iio_0/die* - ==> /sys/devices/uncore_iio_0/die0 <== - 0000:00 - ==> /sys/devices/uncore_iio_0/die1 <== - 0000:40 - ==> /sys/devices/uncore_iio_0/die2 <== - 0000:80 - ==> /sys/devices/uncore_iio_0/die3 <== - 0000:c0 + $ tail /sys/devices/uncore_iio_0/die* + ==> /sys/devices/uncore_iio_0/die0 <== + 0000:00 + ==> /sys/devices/uncore_iio_0/die1 <== + 0000:40 + ==> /sys/devices/uncore_iio_0/die2 <== + 0000:80 + ==> /sys/devices/uncore_iio_0/die3 <== + 0000:c0 - Which means: - IIO PMU 0 on die 0 belongs to PCI RP on bus 0x00, domain 0x0000 - IIO PMU 0 on die 1 belongs to PCI RP on bus 0x40, domain 0x0000 - IIO PMU 0 on die 2 belongs to PCI RP on bus 0x80, domain 0x0000 - IIO PMU 0 on die 3 belongs to PCI RP on bus 0xc0, domain 0x0000 + Which means:: + + IIO PMU 0 on die 0 belongs to PCI RP on bus 0x00, domain 0x0000 + IIO PMU 0 on die 1 belongs to PCI RP on bus 0x40, domain 0x0000 + IIO PMU 0 on die 2 belongs to PCI RP on bus 0x80, domain 0x0000 + IIO PMU 0 on die 3 belongs to PCI RP on bus 0xc0, domain 0x0000 diff --git a/Documentation/ABI/testing/sysfs-devices-memory b/Documentation/ABI/testing/sysfs-devices-memory index deef3b5723cf..2da2b1fba2c1 100644 --- a/Documentation/ABI/testing/sysfs-devices-memory +++ b/Documentation/ABI/testing/sysfs-devices-memory @@ -47,16 +47,19 @@ Description: online/offline state of the memory section. When written, root can toggle the the online/offline state of a removable memory section (see removable file description above) - using the following commands. - # echo online > /sys/devices/system/memory/memoryX/state - # echo offline > /sys/devices/system/memory/memoryX/state + using the following commands:: + + # echo online > /sys/devices/system/memory/memoryX/state + # echo offline > /sys/devices/system/memory/memoryX/state For example, if /sys/devices/system/memory/memory22/removable contains a value of 1 and /sys/devices/system/memory/memory22/state contains the string "online" the following command can be executed by - by root to offline that section. - # echo offline > /sys/devices/system/memory/memory22/state + by root to offline that section:: + + # echo offline > /sys/devices/system/memory/memory22/state + Users: hotplug memory remove tools http://www.ibm.com/developerworks/wikis/display/LinuxP/powerpc-utils @@ -78,6 +81,7 @@ Description: For example, the following symbolic link is created for memory section 9 on node0: + /sys/devices/system/memory/memory9/node0 -> ../../node/node0 @@ -90,4 +94,5 @@ Description: points to the corresponding /sys/devices/system/memory/memoryY memory section directory. For example, the following symbolic link is created for memory section 9 on node0. + /sys/devices/system/node/node0/memory9 -> ../../memory/memory9 diff --git a/Documentation/ABI/testing/sysfs-devices-platform-_UDC_-gadget b/Documentation/ABI/testing/sysfs-devices-platform-_UDC_-gadget index d548eaac230a..40f29a01fd14 100644 --- a/Documentation/ABI/testing/sysfs-devices-platform-_UDC_-gadget +++ b/Documentation/ABI/testing/sysfs-devices-platform-_UDC_-gadget @@ -3,8 +3,9 @@ Date: April 2010 Contact: Fabien Chouteau Description: Show the suspend state of an USB composite gadget. - 1 -> suspended - 0 -> resumed + + - 1 -> suspended + - 0 -> resumed (_UDC_ is the name of the USB Device Controller driver) @@ -17,5 +18,6 @@ Description: Storage mode. Possible values are: - 1 -> ignore the FUA flag - 0 -> obey the FUA flag + + - 1 -> ignore the FUA flag + - 0 -> obey the FUA flag diff --git a/Documentation/ABI/testing/sysfs-devices-platform-ipmi b/Documentation/ABI/testing/sysfs-devices-platform-ipmi index afb5db856e1c..07df0ddc0b69 100644 --- a/Documentation/ABI/testing/sysfs-devices-platform-ipmi +++ b/Documentation/ABI/testing/sysfs-devices-platform-ipmi @@ -123,38 +123,40 @@ KernelVersion: v4.15 Contact: openipmi-developer@lists.sourceforge.net Description: - idles: (RO) Number of times the interface was + ====================== ======================================== + idles (RO) Number of times the interface was idle while being polled. - watchdog_pretimeouts: (RO) Number of watchdog pretimeouts. + watchdog_pretimeouts (RO) Number of watchdog pretimeouts. - complete_transactions: (RO) Number of completed messages. + complete_transactions (RO) Number of completed messages. - events: (RO) Number of IPMI events received from + events (RO) Number of IPMI events received from the hardware. - interrupts: (RO) Number of interrupts the driver + interrupts (RO) Number of interrupts the driver handled. - hosed_count: (RO) Number of times the hardware didn't + hosed_count (RO) Number of times the hardware didn't follow the state machine. - long_timeouts: (RO) Number of times the driver + long_timeouts (RO) Number of times the driver requested a timer while nothing was in progress. - flag_fetches: (RO) Number of times the driver + flag_fetches (RO) Number of times the driver requested flags from the hardware. - attentions: (RO) Number of time the driver got an + attentions (RO) Number of time the driver got an ATTN from the hardware. - incoming_messages: (RO) Number of asynchronous messages + incoming_messages (RO) Number of asynchronous messages received. - short_timeouts: (RO) Number of times the driver + short_timeouts (RO) Number of times the driver requested a timer while an operation was in progress. + ====================== ======================================== What: /sys/devices/platform/ipmi_si.*/interrupts_enabled @@ -201,38 +203,40 @@ Date: Sep, 2017 KernelVersion: v4.15 Contact: openipmi-developer@lists.sourceforge.net Description: - hosed: (RO) Number of times the hardware didn't + ====================== ======================================== + hosed (RO) Number of times the hardware didn't follow the state machine. - alerts: (RO) Number of alerts received. + alerts (RO) Number of alerts received. - sent_messages: (RO) Number of total messages sent. + sent_messages (RO) Number of total messages sent. - sent_message_parts: (RO) Number of message parts sent. + sent_message_parts (RO) Number of message parts sent. Messages may be broken into parts if they are long. - received_messages: (RO) Number of message responses + received_messages (RO) Number of message responses received. - received_message_parts: (RO) Number of message fragments + received_message_parts (RO) Number of message fragments received. - events: (RO) Number of received events. + events (RO) Number of received events. - watchdog_pretimeouts: (RO) Number of watchdog pretimeouts. + watchdog_pretimeouts (RO) Number of watchdog pretimeouts. - flag_fetches: (RO) Number of times a flag fetch was + flag_fetches (RO) Number of times a flag fetch was requested. - send_retries: (RO) Number of time a message was + send_retries (RO) Number of time a message was retried. - receive_retries: (RO) Number of times the receive of a + receive_retries (RO) Number of times the receive of a message was retried. - send_errors: (RO) Number of times the send of a + send_errors (RO) Number of times the send of a message failed. - receive_errors: (RO) Number of errors in receiving + receive_errors (RO) Number of errors in receiving messages. + ====================== ======================================== diff --git a/Documentation/ABI/testing/sysfs-devices-system-cpu b/Documentation/ABI/testing/sysfs-devices-system-cpu index 274c337ec6a9..1a04ca8162ad 100644 --- a/Documentation/ABI/testing/sysfs-devices-system-cpu +++ b/Documentation/ABI/testing/sysfs-devices-system-cpu @@ -169,7 +169,7 @@ Description: observed CPU idle duration was too short for it (a count). - below: (RO) Number of times this state was entered, but the + below: (RO) Number of times this state was entered, but the observed CPU idle duration was too long for it (a count). ======== ==== ================================================= @@ -601,7 +601,7 @@ Description: Secure Virtual Machine Facility in POWER9 and newer processors. i.e., it is a Secure Virtual Machine. -What: /sys/devices/system/cpu/cpuX/purr +What: /sys/devices/system/cpu/cpuX/purr Date: Apr 2005 Contact: Linux for PowerPC mailing list Description: PURR ticks for this CPU since the system boot. diff --git a/Documentation/ABI/testing/sysfs-driver-hid-lenovo b/Documentation/ABI/testing/sysfs-driver-hid-lenovo index 53a0725962e1..aee85ca1f6be 100644 --- a/Documentation/ABI/testing/sysfs-driver-hid-lenovo +++ b/Documentation/ABI/testing/sysfs-driver-hid-lenovo @@ -3,14 +3,18 @@ Date: July 2011 Contact: linux-input@vger.kernel.org Description: This controls if mouse clicks should be generated if the trackpoint is quickly pressed. How fast this press has to be is being controlled by press_speed. + Values are 0 or 1. + Applies to Thinkpad USB Keyboard with TrackPoint. What: /sys/bus/usb/devices/-:./::./dragging Date: July 2011 Contact: linux-input@vger.kernel.org Description: If this setting is enabled, it is possible to do dragging by pressing the trackpoint. This requires press_to_select to be enabled. + Values are 0 or 1. + Applies to Thinkpad USB Keyboard with TrackPoint. What: /sys/bus/usb/devices/-:./::./release_to_select @@ -25,7 +29,9 @@ Date: July 2011 Contact: linux-input@vger.kernel.org Description: This setting controls if the mouse click events generated by pressing the trackpoint (if press_to_select is enabled) generate a left or right mouse button click. + Values are 0 or 1. + Applies to Thinkpad USB Keyboard with TrackPoint. What: /sys/bus/usb/devices/-:./::./sensitivity @@ -39,12 +45,16 @@ What: /sys/bus/usb/devices/-:./-:./::./fn_lock Date: July 2014 Contact: linux-input@vger.kernel.org Description: This setting controls whether Fn Lock is enabled on the keyboard (i.e. if F1 is Mute or F1) + Values are 0 or 1 + Applies to ThinkPad Compact (USB|Bluetooth) Keyboard with TrackPoint. diff --git a/Documentation/ABI/testing/sysfs-driver-hid-ntrig b/Documentation/ABI/testing/sysfs-driver-hid-ntrig index e574a5625efe..0e323a5cec6c 100644 --- a/Documentation/ABI/testing/sysfs-driver-hid-ntrig +++ b/Documentation/ABI/testing/sysfs-driver-hid-ntrig @@ -29,12 +29,13 @@ Contact: linux-input@vger.kernel.org Description: Threholds to override activation slack. - activation_width: (RW) Width threshold to immediately + ================= ===================================== + activation_width (RW) Width threshold to immediately start processing touch events. - activation_height: (RW) Height threshold to immediately + activation_height (RW) Height threshold to immediately start processing touch events. - + ================= ===================================== What: /sys/bus/hid/drivers/ntrig//min_width What: /sys/bus/hid/drivers/ntrig//min_height @@ -44,11 +45,13 @@ Contact: linux-input@vger.kernel.org Description: Minimum size contact accepted. - min_width: (RW) Minimum touch contact width to decide + ========== =========================================== + min_width (RW) Minimum touch contact width to decide activation and activity. - min_height: (RW) Minimum touch contact height to decide + min_height (RW) Minimum touch contact height to decide activation and activity. + ========== =========================================== What: /sys/bus/hid/drivers/ntrig//sensor_physical_width diff --git a/Documentation/ABI/testing/sysfs-driver-hid-roccat-kone b/Documentation/ABI/testing/sysfs-driver-hid-roccat-kone index 8f7982c70d72..11cd9bf0ad18 100644 --- a/Documentation/ABI/testing/sysfs-driver-hid-roccat-kone +++ b/Documentation/ABI/testing/sysfs-driver-hid-roccat-kone @@ -3,17 +3,21 @@ Date: March 2010 Contact: Stefan Achatz Description: It is possible to switch the dpi setting of the mouse with the press of a button. + When read, this file returns the raw number of the actual dpi setting reported by the mouse. This number has to be further processed to receive the real dpi value: + ===== ===== VALUE DPI + ===== ===== 1 800 2 1200 3 1600 4 2000 5 2400 6 3200 + ===== ===== This file is readonly. Users: http://roccat.sourceforge.net @@ -22,6 +26,7 @@ What: /sys/bus/usb/devices/-:./ Description: When read, this file returns the number of the actual profile. + This file is readonly. Users: http://roccat.sourceforge.net @@ -33,6 +38,7 @@ Description: When read, this file returns the raw integer version number of the further usage in other programs. To receive the real version number the decimal point has to be shifted 2 positions to the left. E.g. a returned value of 138 means 1.38 + This file is readonly. Users: http://roccat.sourceforge.net @@ -43,10 +49,13 @@ Description: The mouse can store 5 profiles which can be switched by the press of a button. A profile holds information like button mappings, sensitivity, the colors of the 5 leds and light effects. + When read, these files return the respective profile. The returned data is 975 bytes in size. + When written, this file lets one write the respective profile data back to the mouse. The data has to be 975 bytes long. + The mouse will reject invalid data, whereas the profile number stored in the profile doesn't need to fit the number of the store. @@ -58,6 +67,7 @@ Contact: Stefan Achatz Description: When read, this file returns the settings stored in the mouse. The size of the data is 36 bytes and holds information like the startup_profile, tcu state and calibration_data. + When written, this file lets write settings back to the mouse. The data has to be 36 bytes long. The mouse will reject invalid data. @@ -67,8 +77,10 @@ What: /sys/bus/usb/devices/-:./ Description: The integer value of this attribute ranges from 1 to 5. + When read, this attribute returns the number of the profile that's active when the mouse is powered on. + When written, this file sets the number of the startup profile and the mouse activates this profile immediately. Users: http://roccat.sourceforge.net @@ -80,9 +92,12 @@ Description: The mouse has a "Tracking Control Unit" which lets the user calibrate the laser power to fit the mousepad surface. When read, this file returns the current state of the TCU, where 0 means off and 1 means on. + Writing 0 in this file will switch the TCU off. + Writing 1 in this file will start the calibration which takes around 6 seconds to complete and activates the TCU. + Users: http://roccat.sourceforge.net What: /sys/bus/usb/devices/-:./::./kone/roccatkone/weight @@ -93,14 +108,18 @@ Description: The mouse can be equipped with one of four supplied weights and its value can be read out. When read, this file returns the raw value returned by the mouse which eases further processing in other software. + The values map to the weights as follows: + ===== ====== VALUE WEIGHT + ===== ====== 0 none 1 5g 2 10g 3 15g 4 20g + ===== ====== This file is readonly. Users: http://roccat.sourceforge.net diff --git a/Documentation/ABI/testing/sysfs-driver-hid-wiimote b/Documentation/ABI/testing/sysfs-driver-hid-wiimote index cd7b82a5c27d..3bf43d9dcdfe 100644 --- a/Documentation/ABI/testing/sysfs-driver-hid-wiimote +++ b/Documentation/ABI/testing/sysfs-driver-hid-wiimote @@ -20,6 +20,7 @@ Description: This file contains the currently connected and initialized the official Nintendo Nunchuck extension and classic is the Nintendo Classic Controller extension. The motionp extension can be combined with the other two. + Starting with kernel-version 3.11 Motion Plus hotplugging is supported and if detected, it's no longer reported as static extension. You will get uevent notifications for the motion-plus diff --git a/Documentation/ABI/testing/sysfs-driver-input-exc3000 b/Documentation/ABI/testing/sysfs-driver-input-exc3000 index 3d316d54f81c..cd7c578aef2c 100644 --- a/Documentation/ABI/testing/sysfs-driver-input-exc3000 +++ b/Documentation/ABI/testing/sysfs-driver-input-exc3000 @@ -4,6 +4,7 @@ Contact: linux-input@vger.kernel.org Description: Reports the firmware version provided by the touchscreen, for example "00_T6" on a EXC80H60 Access: Read + Valid values: Represented as string What: /sys/bus/i2c/devices/xxx/model @@ -12,4 +13,5 @@ Contact: linux-input@vger.kernel.org Description: Reports the model identification provided by the touchscreen, for example "Orion_1320" on a EXC80H60 Access: Read + Valid values: Represented as string diff --git a/Documentation/ABI/testing/sysfs-driver-jz4780-efuse b/Documentation/ABI/testing/sysfs-driver-jz4780-efuse index bb6f5d6ceea0..4cf595d681e6 100644 --- a/Documentation/ABI/testing/sysfs-driver-jz4780-efuse +++ b/Documentation/ABI/testing/sysfs-driver-jz4780-efuse @@ -4,7 +4,9 @@ Contact: PrasannaKumar Muralidharan Description: read-only access to the efuse on the Ingenic JZ4780 SoC The SoC has a one time programmable 8K efuse that is split into segments. The driver supports read only. - The segments are + The segments are: + + ===== ======== ================= 0x000 64 bit Random Number 0x008 128 bit Ingenic Chip ID 0x018 128 bit Customer ID @@ -12,5 +14,7 @@ Description: read-only access to the efuse on the Ingenic JZ4780 SoC 0x1E0 8 bit Protect Segment 0x1E1 2296 bit HDMI Key 0x300 2048 bit Security boot key + ===== ======== ================= + Users: any user space application which wants to read the Chip and Customer ID diff --git a/Documentation/ABI/testing/sysfs-driver-pciback b/Documentation/ABI/testing/sysfs-driver-pciback index 73308c2b81b0..49f5fd0c8bbd 100644 --- a/Documentation/ABI/testing/sysfs-driver-pciback +++ b/Documentation/ABI/testing/sysfs-driver-pciback @@ -7,8 +7,10 @@ Description: the format of DDDD:BB:DD.F-REG:SIZE:MASK will allow the guest to write and read from the PCI device. That is Domain:Bus: Device.Function-Register:Size:Mask (Domain is optional). - For example: - #echo 00:19.0-E0:2:FF > /sys/bus/pci/drivers/pciback/quirks + For example:: + + #echo 00:19.0-E0:2:FF > /sys/bus/pci/drivers/pciback/quirks + will allow the guest to read and write to the configuration register 0x0E. diff --git a/Documentation/ABI/testing/sysfs-driver-ufs b/Documentation/ABI/testing/sysfs-driver-ufs index d1a352194d2e..adc0d0e91607 100644 --- a/Documentation/ABI/testing/sysfs-driver-ufs +++ b/Documentation/ABI/testing/sysfs-driver-ufs @@ -18,6 +18,7 @@ Contact: Stanislav Nijnikov Description: This file shows the device type. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/device_class @@ -26,6 +27,7 @@ Contact: Stanislav Nijnikov Description: This file shows the device class. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/device_sub_class @@ -34,6 +36,7 @@ Contact: Stanislav Nijnikov Description: This file shows the UFS storage subclass. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/protocol @@ -43,6 +46,7 @@ Description: This file shows the protocol supported by an UFS device. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/number_of_luns @@ -51,6 +55,7 @@ Contact: Stanislav Nijnikov Description: This file shows number of logical units. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/number_of_wluns @@ -60,6 +65,7 @@ Description: This file shows number of well known logical units. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/boot_enable @@ -69,6 +75,7 @@ Description: This file shows value that indicates whether the device is enabled for boot. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/descriptor_access_enable @@ -79,6 +86,7 @@ Description: This file shows value that indicates whether the device of the boot sequence. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/initial_power_mode @@ -88,6 +96,7 @@ Description: This file shows value that defines the power mode after device initialization or hardware reset. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/high_priority_lun @@ -96,6 +105,7 @@ Contact: Stanislav Nijnikov Description: This file shows the high priority lun. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/secure_removal_type @@ -104,6 +114,7 @@ Contact: Stanislav Nijnikov Description: This file shows the secure removal type. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/support_security_lun @@ -113,6 +124,7 @@ Description: This file shows whether the security lun is supported. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/bkops_termination_latency @@ -122,6 +134,7 @@ Description: This file shows the background operations termination latency. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/initial_active_icc_level @@ -130,6 +143,7 @@ Contact: Stanislav Nijnikov Description: This file shows the initial active ICC level. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/specification_version @@ -138,6 +152,7 @@ Contact: Stanislav Nijnikov Description: This file shows the specification version. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/manufacturing_date @@ -147,6 +162,7 @@ Description: This file shows the manufacturing date in BCD format. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/manufacturer_id @@ -155,6 +171,7 @@ Contact: Stanislav Nijnikov Description: This file shows the manufacturee ID. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/rtt_capability @@ -164,6 +181,7 @@ Description: This file shows the maximum number of outstanding RTTs supported by the device. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/rtc_update @@ -173,6 +191,7 @@ Description: This file shows the frequency and method of the realtime clock update. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/ufs_features @@ -182,6 +201,7 @@ Description: This file shows which features are supported by the device. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/ffu_timeout @@ -190,6 +210,7 @@ Contact: Stanislav Nijnikov Description: This file shows the FFU timeout. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/queue_depth @@ -198,6 +219,7 @@ Contact: Stanislav Nijnikov Description: This file shows the device queue depth. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/device_version @@ -206,6 +228,7 @@ Contact: Stanislav Nijnikov Description: This file shows the device version. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/number_of_secure_wpa @@ -215,6 +238,7 @@ Description: This file shows number of secure write protect areas supported by the device. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/psa_max_data_size @@ -225,6 +249,7 @@ Description: This file shows the maximum amount of data that may be This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/psa_state_timeout @@ -234,6 +259,7 @@ Description: This file shows the command maximum timeout for a change in PSA state. This is one of the UFS device descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. @@ -244,6 +270,7 @@ Description: This file shows the MIPI UniPro version number in BCD format. This is one of the UFS interconnect descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/interconnect_descriptor/mphy_version @@ -253,6 +280,7 @@ Description: This file shows the MIPI M-PHY version number in BCD format. This is one of the UFS interconnect descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. @@ -264,6 +292,7 @@ Description: This file shows the total memory quantity available to of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/max_number_of_luns @@ -273,6 +302,7 @@ Description: This file shows the maximum number of logical units supported by the UFS device. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/segment_size @@ -281,6 +311,7 @@ Contact: Stanislav Nijnikov Description: This file shows the segment size. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/allocation_unit_size @@ -289,6 +320,7 @@ Contact: Stanislav Nijnikov Description: This file shows the allocation unit size. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/min_addressable_block_size @@ -298,6 +330,7 @@ Description: This file shows the minimum addressable block size. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/optimal_read_block_size @@ -307,6 +340,7 @@ Description: This file shows the optimal read block size. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/optimal_write_block_size @@ -316,6 +350,7 @@ Description: This file shows the optimal write block size. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/max_in_buffer_size @@ -325,6 +360,7 @@ Description: This file shows the maximum data-in buffer size. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/max_out_buffer_size @@ -334,6 +370,7 @@ Description: This file shows the maximum data-out buffer size. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/rpmb_rw_size @@ -343,6 +380,7 @@ Description: This file shows the maximum number of RPMB frames allowed in Security Protocol In/Out. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/dyn_capacity_resource_policy @@ -352,6 +390,7 @@ Description: This file shows the dynamic capacity resource policy. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/data_ordering @@ -361,6 +400,7 @@ Description: This file shows support for out-of-order data transfer. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/max_number_of_contexts @@ -370,6 +410,7 @@ Description: This file shows maximum available number of contexts which are supported by the device. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/sys_data_tag_unit_size @@ -378,6 +419,7 @@ Contact: Stanislav Nijnikov Description: This file shows system data tag unit size. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/sys_data_tag_resource_size @@ -388,6 +430,7 @@ Description: This file shows maximum storage area size allocated by This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/secure_removal_types @@ -397,6 +440,7 @@ Description: This file shows supported secure removal types. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/memory_types @@ -406,6 +450,7 @@ Description: This file shows supported memory types. This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/*_memory_max_alloc_units @@ -416,6 +461,7 @@ Description: This file shows the maximum number of allocation units for enhanced type 1-4). This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/*_memory_capacity_adjustment_factor @@ -426,6 +472,7 @@ Description: This file shows the memory capacity adjustment factor for enhanced type 1-4). This is one of the UFS geometry descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. @@ -436,6 +483,7 @@ Description: This file shows preend of life information. This is one of the UFS health descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/health_descriptor/life_time_estimation_a @@ -445,6 +493,7 @@ Description: This file shows indication of the device life time (method a). This is one of the UFS health descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/health_descriptor/life_time_estimation_b @@ -454,6 +503,7 @@ Description: This file shows indication of the device life time (method b). This is one of the UFS health descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. @@ -464,6 +514,7 @@ Description: This file shows maximum VCC, VCCQ and VCCQ2 value for active ICC levels from 0 to 15. This is one of the UFS power descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. @@ -473,6 +524,7 @@ Contact: Stanislav Nijnikov Description: This file contains a device manufactureer name string. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/string_descriptors/product_name @@ -480,6 +532,7 @@ Date: February 2018 Contact: Stanislav Nijnikov Description: This file contains a product name string. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/string_descriptors/oem_id @@ -487,6 +540,7 @@ Date: February 2018 Contact: Stanislav Nijnikov Description: This file contains a OEM ID string. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/string_descriptors/serial_number @@ -495,6 +549,7 @@ Contact: Stanislav Nijnikov Description: This file contains a device serial number string. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/string_descriptors/product_revision @@ -503,6 +558,7 @@ Contact: Stanislav Nijnikov Description: This file contains a product revision string. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. @@ -512,6 +568,7 @@ Contact: Stanislav Nijnikov Description: This file shows boot LUN information. This is one of the UFS unit descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/class/scsi_device/*/device/unit_descriptor/lun_write_protect @@ -520,6 +577,7 @@ Contact: Stanislav Nijnikov Description: This file shows LUN write protection status. This is one of the UFS unit descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/class/scsi_device/*/device/unit_descriptor/lun_queue_depth @@ -528,6 +586,7 @@ Contact: Stanislav Nijnikov Description: This file shows LUN queue depth. This is one of the UFS unit descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/class/scsi_device/*/device/unit_descriptor/psa_sensitive @@ -536,6 +595,7 @@ Contact: Stanislav Nijnikov Description: This file shows PSA sensitivity. This is one of the UFS unit descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/class/scsi_device/*/device/unit_descriptor/lun_memory_type @@ -544,6 +604,7 @@ Contact: Stanislav Nijnikov Description: This file shows LUN memory type. This is one of the UFS unit descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/class/scsi_device/*/device/unit_descriptor/data_reliability @@ -553,6 +614,7 @@ Description: This file defines the device behavior when a power failure occurs during a write operation. This is one of the UFS unit descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/class/scsi_device/*/device/unit_descriptor/logical_block_size @@ -562,6 +624,7 @@ Description: This file shows the size of addressable logical blocks (calculated as an exponent with base 2). This is one of the UFS unit descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/class/scsi_device/*/device/unit_descriptor/logical_block_count @@ -571,6 +634,7 @@ Description: This file shows total number of addressable logical blocks. This is one of the UFS unit descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/class/scsi_device/*/device/unit_descriptor/erase_block_size @@ -579,6 +643,7 @@ Contact: Stanislav Nijnikov Description: This file shows the erase block size. This is one of the UFS unit descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/class/scsi_device/*/device/unit_descriptor/provisioning_type @@ -587,6 +652,7 @@ Contact: Stanislav Nijnikov Description: This file shows the thin provisioning type. This is one of the UFS unit descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/class/scsi_device/*/device/unit_descriptor/physical_memory_resourse_count @@ -595,6 +661,7 @@ Contact: Stanislav Nijnikov Description: This file shows the total physical memory resources. This is one of the UFS unit descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/class/scsi_device/*/device/unit_descriptor/context_capabilities @@ -603,6 +670,7 @@ Contact: Stanislav Nijnikov Description: This file shows the context capabilities. This is one of the UFS unit descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. What: /sys/class/scsi_device/*/device/unit_descriptor/large_unit_granularity @@ -611,6 +679,7 @@ Contact: Stanislav Nijnikov Description: This file shows the granularity of the LUN. This is one of the UFS unit descriptor parameters. The full information about the descriptor could be found at UFS specifications 2.1. + The file is read only. @@ -619,6 +688,7 @@ Date: February 2018 Contact: Stanislav Nijnikov Description: This file shows the device init status. The full information about the flag could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/flags/permanent_wpe @@ -627,6 +697,7 @@ Contact: Stanislav Nijnikov Description: This file shows whether permanent write protection is enabled. The full information about the flag could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/flags/power_on_wpe @@ -636,6 +707,7 @@ Description: This file shows whether write protection is enabled on all logical units configured as power on write protected. The full information about the flag could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/flags/bkops_enable @@ -644,6 +716,7 @@ Contact: Stanislav Nijnikov Description: This file shows whether the device background operations are enabled. The full information about the flag could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/flags/life_span_mode_enable @@ -652,6 +725,7 @@ Contact: Stanislav Nijnikov Description: This file shows whether the device life span mode is enabled. The full information about the flag could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/flags/phy_resource_removal @@ -660,6 +734,7 @@ Contact: Stanislav Nijnikov Description: This file shows whether physical resource removal is enable. The full information about the flag could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/flags/busy_rtc @@ -668,6 +743,7 @@ Contact: Stanislav Nijnikov Description: This file shows whether the device is executing internal operation related to real time clock. The full information about the flag could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/flags/disable_fw_update @@ -676,6 +752,7 @@ Contact: Stanislav Nijnikov Description: This file shows whether the device FW update is permanently disabled. The full information about the flag could be found at UFS specifications 2.1. + The file is read only. @@ -685,6 +762,7 @@ Contact: Stanislav Nijnikov Description: This file provides the boot lun enabled UFS device attribute. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/current_power_mode @@ -693,6 +771,7 @@ Contact: Stanislav Nijnikov Description: This file provides the current power mode UFS device attribute. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/active_icc_level @@ -701,6 +780,7 @@ Contact: Stanislav Nijnikov Description: This file provides the active icc level UFS device attribute. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/ooo_data_enabled @@ -709,6 +789,7 @@ Contact: Stanislav Nijnikov Description: This file provides the out of order data transfer enabled UFS device attribute. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/bkops_status @@ -717,6 +798,7 @@ Contact: Stanislav Nijnikov Description: This file provides the background operations status UFS device attribute. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/purge_status @@ -725,6 +807,7 @@ Contact: Stanislav Nijnikov Description: This file provides the purge operation status UFS device attribute. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/max_data_in_size @@ -733,6 +816,7 @@ Contact: Stanislav Nijnikov Description: This file shows the maximum data size in a DATA IN UPIU. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/max_data_out_size @@ -741,6 +825,7 @@ Contact: Stanislav Nijnikov Description: This file shows the maximum number of bytes that can be requested with a READY TO TRANSFER UPIU. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/reference_clock_frequency @@ -749,6 +834,7 @@ Contact: Stanislav Nijnikov Description: This file provides the reference clock frequency UFS device attribute. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/configuration_descriptor_lock @@ -765,6 +851,7 @@ Description: This file provides the maximum current number of outstanding RTTs in device that is allowed. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/exception_event_control @@ -773,6 +860,7 @@ Contact: Stanislav Nijnikov Description: This file provides the exception event control UFS device attribute. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/exception_event_status @@ -781,6 +869,7 @@ Contact: Stanislav Nijnikov Description: This file provides the exception event status UFS device attribute. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/ffu_status @@ -789,6 +878,7 @@ Contact: Stanislav Nijnikov Description: This file provides the ffu status UFS device attribute. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/psa_state @@ -796,6 +886,7 @@ Date: February 2018 Contact: Stanislav Nijnikov Description: This file show the PSA feature status. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/psa_data_size @@ -805,6 +896,7 @@ Description: This file shows the amount of data that the host plans to load to all logical units in pre-soldering state. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. @@ -815,6 +907,7 @@ Description: This file shows the The amount of physical memory needed to be removed from the physical memory resources pool of the particular logical unit. The full information about the attribute could be found at UFS specifications 2.1. + The file is read only. @@ -824,24 +917,28 @@ Contact: Subhash Jadavani Description: This entry could be used to set or show the UFS device runtime power management level. The current driver implementation supports 6 levels with next target states: - 0 - an UFS device will stay active, an UIC link will - stay active - 1 - an UFS device will stay active, an UIC link will - hibernate - 2 - an UFS device will moved to sleep, an UIC link will - stay active - 3 - an UFS device will moved to sleep, an UIC link will - hibernate - 4 - an UFS device will be powered off, an UIC link will - hibernate - 5 - an UFS device will be powered off, an UIC link will - be powered off + + == ==================================================== + 0 an UFS device will stay active, an UIC link will + stay active + 1 an UFS device will stay active, an UIC link will + hibernate + 2 an UFS device will moved to sleep, an UIC link will + stay active + 3 an UFS device will moved to sleep, an UIC link will + hibernate + 4 an UFS device will be powered off, an UIC link will + hibernate + 5 an UFS device will be powered off, an UIC link will + be powered off + == ==================================================== What: /sys/bus/platform/drivers/ufshcd/*/rpm_target_dev_state Date: February 2018 Contact: Subhash Jadavani Description: This entry shows the target power mode of an UFS device for the chosen runtime power management level. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/rpm_target_link_state @@ -849,6 +946,7 @@ Date: February 2018 Contact: Subhash Jadavani Description: This entry shows the target state of an UFS UIC link for the chosen runtime power management level. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/spm_lvl @@ -857,24 +955,28 @@ Contact: Subhash Jadavani Description: This entry could be used to set or show the UFS device system power management level. The current driver implementation supports 6 levels with next target states: - 0 - an UFS device will stay active, an UIC link will - stay active - 1 - an UFS device will stay active, an UIC link will - hibernate - 2 - an UFS device will moved to sleep, an UIC link will - stay active - 3 - an UFS device will moved to sleep, an UIC link will - hibernate - 4 - an UFS device will be powered off, an UIC link will - hibernate - 5 - an UFS device will be powered off, an UIC link will - be powered off + + == ==================================================== + 0 an UFS device will stay active, an UIC link will + stay active + 1 an UFS device will stay active, an UIC link will + hibernate + 2 an UFS device will moved to sleep, an UIC link will + stay active + 3 an UFS device will moved to sleep, an UIC link will + hibernate + 4 an UFS device will be powered off, an UIC link will + hibernate + 5 an UFS device will be powered off, an UIC link will + be powered off + == ==================================================== What: /sys/bus/platform/drivers/ufshcd/*/spm_target_dev_state Date: February 2018 Contact: Subhash Jadavani Description: This entry shows the target power mode of an UFS device for the chosen system power management level. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/spm_target_link_state @@ -882,18 +984,21 @@ Date: February 2018 Contact: Subhash Jadavani Description: This entry shows the target state of an UFS UIC link for the chosen system power management level. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/wb_presv_us_en Date: June 2020 Contact: Asutosh Das Description: This entry shows if preserve user-space was configured + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/wb_shared_alloc_units Date: June 2020 Contact: Asutosh Das Description: This entry shows the shared allocated units of WB buffer + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/device_descriptor/wb_type @@ -901,6 +1006,7 @@ Date: June 2020 Contact: Asutosh Das Description: This entry shows the configured WB type. 0x1 for shared buffer mode. 0x0 for dedicated buffer mode. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/wb_buff_cap_adj @@ -910,6 +1016,7 @@ Description: This entry shows the total user-space decrease in shared buffer mode. The value of this parameter is 3 for TLC NAND when SLC mode is used as WriteBooster Buffer. 2 for MLC NAND. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/wb_max_alloc_units @@ -917,6 +1024,7 @@ Date: June 2020 Contact: Asutosh Das Description: This entry shows the Maximum total WriteBooster Buffer size which is supported by the entire device. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/wb_max_wb_luns @@ -924,6 +1032,7 @@ Date: June 2020 Contact: Asutosh Das Description: This entry shows the maximum number of luns that can support WriteBooster. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/wb_sup_red_type @@ -937,46 +1046,59 @@ Description: The supportability of user space reduction mode preserve user space type. 02h: Device can be configured in either user space reduction type or preserve user space type. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/geometry_descriptor/wb_sup_wb_type Date: June 2020 Contact: Asutosh Das Description: The supportability of WriteBooster Buffer type. - 00h: LU based WriteBooster Buffer configuration - 01h: Single shared WriteBooster Buffer - configuration - 02h: Supporting both LU based WriteBooster - Buffer and Single shared WriteBooster Buffer - configuration + + === ========================================================== + 00h LU based WriteBooster Buffer configuration + 01h Single shared WriteBooster Buffer configuration + 02h Supporting both LU based WriteBooster. + Buffer and Single shared WriteBooster Buffer configuration + === ========================================================== + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/flags/wb_enable Date: June 2020 Contact: Asutosh Das Description: This entry shows the status of WriteBooster. - 0: WriteBooster is not enabled. - 1: WriteBooster is enabled + + == ============================ + 0 WriteBooster is not enabled. + 1 WriteBooster is enabled + == ============================ + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/flags/wb_flush_en Date: June 2020 Contact: Asutosh Das Description: This entry shows if flush is enabled. - 0: Flush operation is not performed. - 1: Flush operation is performed. + + == ================================= + 0 Flush operation is not performed. + 1 Flush operation is performed. + == ================================= + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/flags/wb_flush_during_h8 Date: June 2020 Contact: Asutosh Das Description: Flush WriteBooster Buffer during hibernate state. - 0: Device is not allowed to flush the - WriteBooster Buffer during link hibernate - state. - 1: Device is allowed to flush the - WriteBooster Buffer during link hibernate - state + + == ================================================= + 0 Device is not allowed to flush the + WriteBooster Buffer during link hibernate state. + 1 Device is allowed to flush the + WriteBooster Buffer during link hibernate state. + == ================================================= + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/wb_avail_buf @@ -984,23 +1106,30 @@ Date: June 2020 Contact: Asutosh Das Description: This entry shows the amount of unused WriteBooster buffer available. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/wb_cur_buf Date: June 2020 Contact: Asutosh Das Description: This entry shows the amount of unused current buffer. + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/wb_flush_status Date: June 2020 Contact: Asutosh Das Description: This entry shows the flush operation status. - 00h: idle - 01h: Flush operation in progress - 02h: Flush operation stopped prematurely. - 03h: Flush operation completed successfully - 04h: Flush operation general failure + + + === ====================================== + 00h idle + 01h Flush operation in progress + 02h Flush operation stopped prematurely. + 03h Flush operation completed successfully + 04h Flush operation general failure + === ====================================== + The file is read only. What: /sys/bus/platform/drivers/ufshcd/*/attributes/wb_life_time_est @@ -1008,9 +1137,13 @@ Date: June 2020 Contact: Asutosh Das Description: This entry shows an indication of the WriteBooster Buffer lifetime based on the amount of performed program/erase cycles - 01h: 0% - 10% WriteBooster Buffer life time used + + === ============================================= + 01h 0% - 10% WriteBooster Buffer life time used ... - 0Ah: 90% - 100% WriteBooster Buffer life time used + 0Ah 90% - 100% WriteBooster Buffer life time used + === ============================================= + The file is read only. What: /sys/class/scsi_device/*/device/unit_descriptor/wb_buf_alloc_units @@ -1018,4 +1151,5 @@ Date: June 2020 Contact: Asutosh Das Description: This entry shows the configured size of WriteBooster buffer. 0400h corresponds to 4GB. + The file is read only. diff --git a/Documentation/ABI/testing/sysfs-driver-w1_ds28e17 b/Documentation/ABI/testing/sysfs-driver-w1_ds28e17 index d301e7017afe..e92aba4eb594 100644 --- a/Documentation/ABI/testing/sysfs-driver-w1_ds28e17 +++ b/Documentation/ABI/testing/sysfs-driver-w1_ds28e17 @@ -5,7 +5,9 @@ Contact: Jan Kandziora Description: When written, this file sets the I2C speed on the connected DS28E17 chip. When read, it reads the current setting from the DS28E17 chip. + Valid values: 100, 400, 900 [kBaud]. + Default 100, can be set by w1_ds28e17.speed= module parameter. Users: w1_ds28e17 driver @@ -17,5 +19,6 @@ Description: When written, this file sets the multiplier used to calculate the busy timeout for I2C operations on the connected DS28E17 chip. When read, returns the current setting. Valid values: 1 to 9. + Default 1, can be set by w1_ds28e17.stretch= module parameter. Users: w1_ds28e17 driver diff --git a/Documentation/ABI/testing/sysfs-firmware-acpi b/Documentation/ABI/testing/sysfs-firmware-acpi index e4afc2538210..b16d30a71709 100644 --- a/Documentation/ABI/testing/sysfs-firmware-acpi +++ b/Documentation/ABI/testing/sysfs-firmware-acpi @@ -81,11 +81,11 @@ Description: $ cd /sys/firmware/acpi/interrupts $ grep . * error: 0 - ff_gbl_lock: 0 enable - ff_pmtimer: 0 invalid - ff_pwr_btn: 0 enable - ff_rt_clk: 2 disable - ff_slp_btn: 0 invalid + ff_gbl_lock: 0 enable + ff_pmtimer: 0 invalid + ff_pwr_btn: 0 enable + ff_rt_clk: 2 disable + ff_slp_btn: 0 invalid gpe00: 0 invalid gpe01: 0 enable gpe02: 108 enable @@ -118,9 +118,9 @@ Description: gpe1D: 0 invalid gpe1E: 0 invalid gpe1F: 0 invalid - gpe_all: 1192 - sci: 1194 - sci_not: 0 + gpe_all: 1192 + sci: 1194 + sci_not: 0 =========== ================================================== sci The number of times the ACPI SCI diff --git a/Documentation/ABI/testing/sysfs-firmware-efi-esrt b/Documentation/ABI/testing/sysfs-firmware-efi-esrt index 6e431d1a4e79..31b57676d4ad 100644 --- a/Documentation/ABI/testing/sysfs-firmware-efi-esrt +++ b/Documentation/ABI/testing/sysfs-firmware-efi-esrt @@ -35,10 +35,13 @@ What: /sys/firmware/efi/esrt/entries/entry$N/fw_type Date: February 2015 Contact: Peter Jones Description: What kind of firmware entry this is: - 0 - Unknown - 1 - System Firmware - 2 - Device Firmware - 3 - UEFI Driver + + == =============== + 0 Unknown + 1 System Firmware + 2 Device Firmware + 3 UEFI Driver + == =============== What: /sys/firmware/efi/esrt/entries/entry$N/fw_class Date: February 2015 @@ -71,11 +74,14 @@ Date: February 2015 Contact: Peter Jones Description: The result of the last firmware update attempt for the firmware resource entry. - 0 - Success - 1 - Insufficient resources - 2 - Incorrect version - 3 - Invalid format - 4 - Authentication error - 5 - AC power event - 6 - Battery power event + + == ====================== + 0 Success + 1 Insufficient resources + 2 Incorrect version + 3 Invalid format + 4 Authentication error + 5 AC power event + 6 Battery power event + == ====================== diff --git a/Documentation/ABI/testing/sysfs-firmware-efi-runtime-map b/Documentation/ABI/testing/sysfs-firmware-efi-runtime-map index c61b9b348e99..9c4d581be396 100644 --- a/Documentation/ABI/testing/sysfs-firmware-efi-runtime-map +++ b/Documentation/ABI/testing/sysfs-firmware-efi-runtime-map @@ -14,7 +14,7 @@ Description: Switching efi runtime services to virtual mode requires /sys/firmware/efi/runtime-map/ is the directory the kernel exports that information in. - subdirectories are named with the number of the memory range: + subdirectories are named with the number of the memory range:: /sys/firmware/efi/runtime-map/0 /sys/firmware/efi/runtime-map/1 @@ -24,11 +24,13 @@ Description: Switching efi runtime services to virtual mode requires Each subdirectory contains five files: - attribute : The attributes of the memory range. - num_pages : The size of the memory range in pages. - phys_addr : The physical address of the memory range. - type : The type of the memory range. - virt_addr : The virtual address of the memory range. + ========= ========================================= + attribute The attributes of the memory range. + num_pages The size of the memory range in pages. + phys_addr The physical address of the memory range. + type The type of the memory range. + virt_addr The virtual address of the memory range. + ========= ========================================= Above values are all hexadecimal numbers with the '0x' prefix. Users: Kexec diff --git a/Documentation/ABI/testing/sysfs-firmware-qemu_fw_cfg b/Documentation/ABI/testing/sysfs-firmware-qemu_fw_cfg index 011dda4f8e8a..ee0d6dbc810e 100644 --- a/Documentation/ABI/testing/sysfs-firmware-qemu_fw_cfg +++ b/Documentation/ABI/testing/sysfs-firmware-qemu_fw_cfg @@ -15,7 +15,7 @@ Description: to the fw_cfg device can be found in "docs/specs/fw_cfg.txt" in the QEMU source tree. - === SysFS fw_cfg Interface === + **SysFS fw_cfg Interface** The fw_cfg sysfs interface described in this document is only intended to display discoverable blobs (i.e., those registered @@ -31,7 +31,7 @@ Description: /sys/firmware/qemu_fw_cfg/rev - --- Discoverable fw_cfg blobs by selector key --- + **Discoverable fw_cfg blobs by selector key** All discoverable blobs listed in the fw_cfg file directory are displayed as entries named after their unique selector key @@ -45,24 +45,26 @@ Description: Each such fw_cfg sysfs entry has the following values exported as attributes: - name : The 56-byte nul-terminated ASCII string used as the + ==== ==================================================== + name The 56-byte nul-terminated ASCII string used as the blob's 'file name' in the fw_cfg directory. - size : The length of the blob, as given in the fw_cfg + size The length of the blob, as given in the fw_cfg directory. - key : The value of the blob's selector key as given in the + key The value of the blob's selector key as given in the fw_cfg directory. This value is the same as used in the parent directory name. - raw : The raw bytes of the blob, obtained by selecting the + raw The raw bytes of the blob, obtained by selecting the entry via the control register, and reading a number of bytes equal to the blob size from the data register. + ==== ==================================================== - --- Listing fw_cfg blobs by file name --- + **Listing fw_cfg blobs by file name** While the fw_cfg device does not impose any specific naming convention on the blobs registered in the file directory, QEMU developers have traditionally used path name semantics - to give each blob a descriptive name. For example: + to give each blob a descriptive name. For example:: "bootorder" "genroms/kvmvapic.bin" @@ -81,7 +83,7 @@ Description: of directories matching the path name components of fw_cfg blob names, ending in symlinks to the by_key entry for each "basename", as illustrated below (assume current directory is - /sys/firmware): + /sys/firmware):: qemu_fw_cfg/by_name/bootorder -> ../by_key/38 qemu_fw_cfg/by_name/etc/e820 -> ../../by_key/35 diff --git a/Documentation/ABI/testing/sysfs-firmware-sfi b/Documentation/ABI/testing/sysfs-firmware-sfi index 4be7d44aeacf..5210e0f06ddb 100644 --- a/Documentation/ABI/testing/sysfs-firmware-sfi +++ b/Documentation/ABI/testing/sysfs-firmware-sfi @@ -9,7 +9,7 @@ Description: http://simplefirmware.org/documentation While the tables are used by the kernel, user-space - can observe them this way: + can observe them this way:: - # cd /sys/firmware/sfi/tables - # cat $TABLENAME > $TABLENAME.bin + # cd /sys/firmware/sfi/tables + # cat $TABLENAME > $TABLENAME.bin diff --git a/Documentation/ABI/testing/sysfs-firmware-sgi_uv b/Documentation/ABI/testing/sysfs-firmware-sgi_uv index 4573fd4b7876..66800baab096 100644 --- a/Documentation/ABI/testing/sysfs-firmware-sgi_uv +++ b/Documentation/ABI/testing/sysfs-firmware-sgi_uv @@ -5,7 +5,7 @@ Description: The /sys/firmware/sgi_uv directory contains information about the SGI UV platform. - Under that directory are a number of files: + Under that directory are a number of files:: partition_id coherence_id @@ -14,7 +14,7 @@ Description: SGI UV systems can be partitioned into multiple physical machines, which each partition running a unique copy of the operating system. Each partition will have a unique - partition id. To display the partition id, use the command: + partition id. To display the partition id, use the command:: cat /sys/firmware/sgi_uv/partition_id @@ -22,6 +22,6 @@ Description: A partitioned SGI UV system can have one or more coherence domain. The coherence id indicates which coherence domain this partition is in. To display the coherence id, use the - command: + command:: cat /sys/firmware/sgi_uv/coherence_id diff --git a/Documentation/ABI/testing/sysfs-fs-f2fs b/Documentation/ABI/testing/sysfs-fs-f2fs index 834d0becae6d..67b3ed8e8c2f 100644 --- a/Documentation/ABI/testing/sysfs-fs-f2fs +++ b/Documentation/ABI/testing/sysfs-fs-f2fs @@ -20,10 +20,13 @@ What: /sys/fs/f2fs//gc_idle Date: July 2013 Contact: "Namjae Jeon" Description: Controls the victim selection policy for garbage collection. - Setting gc_idle = 0(default) will disable this option. Setting - gc_idle = 1 will select the Cost Benefit approach & setting - gc_idle = 2 will select the greedy approach & setting - gc_idle = 3 will select the age-threshold based approach. + Setting gc_idle = 0(default) will disable this option. Setting: + + =========== =============================================== + gc_idle = 1 will select the Cost Benefit approach & setting + gc_idle = 2 will select the greedy approach & setting + gc_idle = 3 will select the age-threshold based approach. + =========== =============================================== What: /sys/fs/f2fs//reclaim_segments Date: October 2013 @@ -46,10 +49,17 @@ Date: November 2013 Contact: "Jaegeuk Kim" Description: Controls the in-place-update policy. updates in f2fs. User can set: - 0x01: F2FS_IPU_FORCE, 0x02: F2FS_IPU_SSR, - 0x04: F2FS_IPU_UTIL, 0x08: F2FS_IPU_SSR_UTIL, - 0x10: F2FS_IPU_FSYNC, 0x20: F2FS_IPU_ASYNC, - 0x40: F2FS_IPU_NOCACHE. + + ==== ================= + 0x01 F2FS_IPU_FORCE + 0x02 F2FS_IPU_SSR + 0x04 F2FS_IPU_UTIL + 0x08 F2FS_IPU_SSR_UTIL + 0x10 F2FS_IPU_FSYNC + 0x20 F2FS_IPU_ASYNC, + 0x40 F2FS_IPU_NOCACHE + ==== ================= + Refer segment.h for details. What: /sys/fs/f2fs//min_ipu_util @@ -332,18 +342,28 @@ Date: April 2020 Contact: "Jaegeuk Kim" Description: Give a way to attach REQ_META|FUA to data writes given temperature-based bits. Now the bits indicate: - * REQ_META | REQ_FUA | - * 5 | 4 | 3 | 2 | 1 | 0 | - * Cold | Warm | Hot | Cold | Warm | Hot | + + +-------------------+-------------------+ + | REQ_META | REQ_FUA | + +------+------+-----+------+------+-----+ + | 5 | 4 | 3 | 2 | 1 | 0 | + +------+------+-----+------+------+-----+ + | Cold | Warm | Hot | Cold | Warm | Hot | + +------+------+-----+------+------+-----+ What: /sys/fs/f2fs//node_io_flag Date: June 2020 Contact: "Jaegeuk Kim" Description: Give a way to attach REQ_META|FUA to node writes given temperature-based bits. Now the bits indicate: - * REQ_META | REQ_FUA | - * 5 | 4 | 3 | 2 | 1 | 0 | - * Cold | Warm | Hot | Cold | Warm | Hot | + + +-------------------+-------------------+ + | REQ_META | REQ_FUA | + +------+------+-----+------+------+-----+ + | 5 | 4 | 3 | 2 | 1 | 0 | + +------+------+-----+------+------+-----+ + | Cold | Warm | Hot | Cold | Warm | Hot | + +------+------+-----+------+------+-----+ What: /sys/fs/f2fs//iostat_period_ms Date: April 2020 diff --git a/Documentation/ABI/testing/sysfs-kernel-mm-ksm b/Documentation/ABI/testing/sysfs-kernel-mm-ksm index dfc13244cda3..1c9bed5595f5 100644 --- a/Documentation/ABI/testing/sysfs-kernel-mm-ksm +++ b/Documentation/ABI/testing/sysfs-kernel-mm-ksm @@ -34,8 +34,9 @@ Description: Kernel Samepage Merging daemon sysfs interface in a tree. run: write 0 to disable ksm, read 0 while ksm is disabled. - write 1 to run ksm, read 1 while ksm is running. - write 2 to disable ksm and unmerge all its pages. + + - write 1 to run ksm, read 1 while ksm is running. + - write 2 to disable ksm and unmerge all its pages. sleep_millisecs: how many milliseconds ksm should sleep between scans. diff --git a/Documentation/ABI/testing/sysfs-kernel-slab b/Documentation/ABI/testing/sysfs-kernel-slab index ed35833ad7f0..c9f12baf8baa 100644 --- a/Documentation/ABI/testing/sysfs-kernel-slab +++ b/Documentation/ABI/testing/sysfs-kernel-slab @@ -346,6 +346,7 @@ Description: number of objects per slab. If a slab cannot be allocated because of fragmentation, SLUB will retry with the minimum order possible depending on its characteristics. + When debug_guardpage_minorder=N (N > 0) parameter is specified (see Documentation/admin-guide/kernel-parameters.rst), the minimum possible order is used and this sysfs entry can not be used to change @@ -361,6 +362,7 @@ Description: new slab has not been possible at the cache's order and instead fallen back to its minimum possible order. It can be written to clear the current count. + Available when CONFIG_SLUB_STATS is enabled. What: /sys/kernel/slab/cache/partial @@ -410,6 +412,7 @@ Description: slab from a remote node as opposed to allocating a new slab on the local node. This reduces the amount of wasted memory over the entire system but can be expensive. + Available when CONFIG_NUMA is enabled. What: /sys/kernel/slab/cache/sanity_checks diff --git a/Documentation/ABI/testing/sysfs-module b/Documentation/ABI/testing/sysfs-module index 0aac02e7fb0e..353c0db5bc1f 100644 --- a/Documentation/ABI/testing/sysfs-module +++ b/Documentation/ABI/testing/sysfs-module @@ -17,14 +17,15 @@ KernelVersion: 3.1 Contact: Kirill Smelkov Description: Maximum time allowed for periodic transfers per microframe (μs) - [ USB 2.0 sets maximum allowed time for periodic transfers per + Note: + USB 2.0 sets maximum allowed time for periodic transfers per microframe to be 80%, that is 100 microseconds out of 125 microseconds (full microframe). However there are cases, when 80% max isochronous bandwidth is too limiting. For example two video streams could require 110 microseconds of isochronous bandwidth per microframe to work - together. ] + together. Through this setting it is possible to raise the limit so that the host controller would allow allocating more than 100 @@ -45,8 +46,10 @@ Date: Jan 2012 KernelVersion:»·3.3 Contact: Kay Sievers Description: Module taint flags: - P - proprietary module - O - out-of-tree module - F - force-loaded module - C - staging driver module - E - unsigned module + == ===================== + P proprietary module + O out-of-tree module + F force-loaded module + C staging driver module + E unsigned module + == ===================== diff --git a/Documentation/ABI/testing/sysfs-platform-dell-laptop b/Documentation/ABI/testing/sysfs-platform-dell-laptop index 9b917c7453de..82bcfe9df66e 100644 --- a/Documentation/ABI/testing/sysfs-platform-dell-laptop +++ b/Documentation/ABI/testing/sysfs-platform-dell-laptop @@ -34,9 +34,12 @@ Description: this file. To disable a trigger, write its name preceded by '-' instead. - For example, to enable the keyboard as trigger run: + For example, to enable the keyboard as trigger run:: + echo +keyboard > /sys/class/leds/dell::kbd_backlight/start_triggers - To disable it: + + To disable it:: + echo -keyboard > /sys/class/leds/dell::kbd_backlight/start_triggers Note that not all the available triggers can be configured. @@ -57,7 +60,8 @@ Description: with any the above units. If no unit is specified, the value is assumed to be expressed in seconds. - For example, to set the timeout to 10 minutes run: + For example, to set the timeout to 10 minutes run:: + echo 10m > /sys/class/leds/dell::kbd_backlight/stop_timeout Note that when this file is read, the returned value might be diff --git a/Documentation/ABI/testing/sysfs-platform-dell-smbios b/Documentation/ABI/testing/sysfs-platform-dell-smbios index 205d3b6361e0..e6e0f7f834a7 100644 --- a/Documentation/ABI/testing/sysfs-platform-dell-smbios +++ b/Documentation/ABI/testing/sysfs-platform-dell-smbios @@ -13,8 +13,8 @@ Description: For example the token ID "5" would be available as the following attributes: - 0005_location - 0005_value + - 0005_location + - 0005_value Tokens will vary from machine to machine, and only tokens available on that machine will be diff --git a/Documentation/ABI/testing/sysfs-platform-i2c-demux-pinctrl b/Documentation/ABI/testing/sysfs-platform-i2c-demux-pinctrl index c394b808be19..b6a138b50d99 100644 --- a/Documentation/ABI/testing/sysfs-platform-i2c-demux-pinctrl +++ b/Documentation/ABI/testing/sysfs-platform-i2c-demux-pinctrl @@ -5,9 +5,9 @@ Contact: Wolfram Sang Description: Reading the file will give you a list of masters which can be selected for a demultiplexed bus. The format is - ":". Example from a Renesas Lager board: + ":". Example from a Renesas Lager board:: - 0:/i2c@e6500000 1:/i2c@e6508000 + 0:/i2c@e6500000 1:/i2c@e6508000 What: /sys/devices/platform//current_master Date: January 2016 diff --git a/Documentation/ABI/testing/sysfs-platform-kim b/Documentation/ABI/testing/sysfs-platform-kim index c1653271872a..a7f81de68046 100644 --- a/Documentation/ABI/testing/sysfs-platform-kim +++ b/Documentation/ABI/testing/sysfs-platform-kim @@ -5,6 +5,7 @@ Contact: "Pavan Savoy" Description: Name of the UART device at which the WL128x chip is connected. example: "/dev/ttyS0". + The device name flows down to architecture specific board initialization file from the SFI/ATAGS bootloader firmware. The name exposed is read from the user-space diff --git a/Documentation/ABI/testing/sysfs-platform-phy-rcar-gen3-usb2 b/Documentation/ABI/testing/sysfs-platform-phy-rcar-gen3-usb2 index 6212697bbf6f..bc510ccc37a7 100644 --- a/Documentation/ABI/testing/sysfs-platform-phy-rcar-gen3-usb2 +++ b/Documentation/ABI/testing/sysfs-platform-phy-rcar-gen3-usb2 @@ -7,9 +7,11 @@ Description: The file can show/change the phy mode for role swap of usb. Write the following strings to change the mode: - "host" - switching mode from peripheral to host. - "peripheral" - switching mode from host to peripheral. + + - "host" - switching mode from peripheral to host. + - "peripheral" - switching mode from host to peripheral. Read the file, then it shows the following strings: - "host" - The mode is host now. - "peripheral" - The mode is peripheral now. + + - "host" - The mode is host now. + - "peripheral" - The mode is peripheral now. diff --git a/Documentation/ABI/testing/sysfs-platform-renesas_usb3 b/Documentation/ABI/testing/sysfs-platform-renesas_usb3 index 5621c15d5dc0..8af5b9c3fabb 100644 --- a/Documentation/ABI/testing/sysfs-platform-renesas_usb3 +++ b/Documentation/ABI/testing/sysfs-platform-renesas_usb3 @@ -7,9 +7,11 @@ Description: The file can show/change the drd mode of usb. Write the following string to change the mode: - "host" - switching mode from peripheral to host. - "peripheral" - switching mode from host to peripheral. + + - "host" - switching mode from peripheral to host. + - "peripheral" - switching mode from host to peripheral. Read the file, then it shows the following strings: - "host" - The mode is host now. - "peripheral" - The mode is peripheral now. + + - "host" - The mode is host now. + - "peripheral" - The mode is peripheral now. diff --git a/Documentation/ABI/testing/sysfs-power b/Documentation/ABI/testing/sysfs-power index 5e6ead29124c..51c0f578bfce 100644 --- a/Documentation/ABI/testing/sysfs-power +++ b/Documentation/ABI/testing/sysfs-power @@ -47,14 +47,18 @@ Description: suspend-to-disk mechanism. Reading from this file returns the name of the method by which the system will be put to sleep on the next suspend. There are four methods supported: + 'firmware' - means that the memory image will be saved to disk by some firmware, in which case we also assume that the firmware will handle the system suspend. + 'platform' - the memory image will be saved by the kernel and the system will be put to sleep by the platform driver (e.g. ACPI or other PM registers). + 'shutdown' - the memory image will be saved by the kernel and the system will be powered off. + 'reboot' - the memory image will be saved by the kernel and the system will be rebooted. @@ -74,12 +78,12 @@ Description: The suspend-to-disk method may be chosen by writing to this file one of the accepted strings: - 'firmware' - 'platform' - 'shutdown' - 'reboot' - 'testproc' - 'test' + - 'firmware' + - 'platform' + - 'shutdown' + - 'reboot' + - 'testproc' + - 'test' It will only change to 'firmware' or 'platform' if the system supports that. @@ -114,9 +118,9 @@ Description: string representing a nonzero integer into it. To use this debugging feature you should attempt to suspend - the machine, then reboot it and run + the machine, then reboot it and run:: - dmesg -s 1000000 | grep 'hash matches' + dmesg -s 1000000 | grep 'hash matches' If you do not get any matches (or they appear to be false positives), it is possible that the last PM event point @@ -244,6 +248,7 @@ Description: wakeup sources created with the help of /sys/power/wake_lock. When a string is written to /sys/power/wake_unlock, it will be assumed to represent the name of a wakeup source to deactivate. + If a wakeup source object of that name exists and is active at the moment, it will be deactivated. diff --git a/Documentation/ABI/testing/sysfs-profiling b/Documentation/ABI/testing/sysfs-profiling index 8a8e466eb2c0..e39dd3a0ceef 100644 --- a/Documentation/ABI/testing/sysfs-profiling +++ b/Documentation/ABI/testing/sysfs-profiling @@ -5,7 +5,7 @@ Description: /sys/kernel/profiling is the runtime equivalent of the boot-time profile= option. - You can get the same effect running: + You can get the same effect running:: echo 2 > /sys/kernel/profiling diff --git a/Documentation/ABI/testing/sysfs-wusb_cbaf b/Documentation/ABI/testing/sysfs-wusb_cbaf index a99c5f86a37a..2969d3694ec0 100644 --- a/Documentation/ABI/testing/sysfs-wusb_cbaf +++ b/Documentation/ABI/testing/sysfs-wusb_cbaf @@ -45,7 +45,8 @@ Description: 7. Device is unplugged. References: - [WUSB-AM] Association Models Supplement to the + [WUSB-AM] + Association Models Supplement to the Certified Wireless Universal Serial Bus Specification, version 1.0. diff --git a/Documentation/ABI/testing/usb-charger-uevent b/Documentation/ABI/testing/usb-charger-uevent index 419a92dd0d86..1db89b0cf80f 100644 --- a/Documentation/ABI/testing/usb-charger-uevent +++ b/Documentation/ABI/testing/usb-charger-uevent @@ -3,44 +3,52 @@ Date: 2020-01-14 KernelVersion: 5.6 Contact: linux-usb@vger.kernel.org Description: There are two USB charger states: - USB_CHARGER_ABSENT - USB_CHARGER_PRESENT + + - USB_CHARGER_ABSENT + - USB_CHARGER_PRESENT + There are five USB charger types: - USB_CHARGER_UNKNOWN_TYPE: Charger type is unknown - USB_CHARGER_SDP_TYPE: Standard Downstream Port - USB_CHARGER_CDP_TYPE: Charging Downstream Port - USB_CHARGER_DCP_TYPE: Dedicated Charging Port - USB_CHARGER_ACA_TYPE: Accessory Charging Adapter + + ======================== ========================== + USB_CHARGER_UNKNOWN_TYPE Charger type is unknown + USB_CHARGER_SDP_TYPE Standard Downstream Port + USB_CHARGER_CDP_TYPE Charging Downstream Port + USB_CHARGER_DCP_TYPE Dedicated Charging Port + USB_CHARGER_ACA_TYPE Accessory Charging Adapter + ======================== ========================== + https://www.usb.org/document-library/battery-charging-v12-spec-and-adopters-agreement - Here are two examples taken using udevadm monitor -p when - USB charger is online: - UDEV change /devices/soc0/usbphynop1 (platform) - ACTION=change - DEVPATH=/devices/soc0/usbphynop1 - DRIVER=usb_phy_generic - MODALIAS=of:Nusbphynop1T(null)Cusb-nop-xceiv - OF_COMPATIBLE_0=usb-nop-xceiv - OF_COMPATIBLE_N=1 - OF_FULLNAME=/usbphynop1 - OF_NAME=usbphynop1 - SEQNUM=2493 - SUBSYSTEM=platform - USB_CHARGER_STATE=USB_CHARGER_PRESENT - USB_CHARGER_TYPE=USB_CHARGER_SDP_TYPE - USEC_INITIALIZED=227422826 - - USB charger is offline: - KERNEL change /devices/soc0/usbphynop1 (platform) - ACTION=change - DEVPATH=/devices/soc0/usbphynop1 - DRIVER=usb_phy_generic - MODALIAS=of:Nusbphynop1T(null)Cusb-nop-xceiv - OF_COMPATIBLE_0=usb-nop-xceiv - OF_COMPATIBLE_N=1 - OF_FULLNAME=/usbphynop1 - OF_NAME=usbphynop1 - SEQNUM=2494 - SUBSYSTEM=platform - USB_CHARGER_STATE=USB_CHARGER_ABSENT - USB_CHARGER_TYPE=USB_CHARGER_UNKNOWN_TYPE + Here are two examples taken using ``udevadm monitor -p`` when + USB charger is online:: + + UDEV change /devices/soc0/usbphynop1 (platform) + ACTION=change + DEVPATH=/devices/soc0/usbphynop1 + DRIVER=usb_phy_generic + MODALIAS=of:Nusbphynop1T(null)Cusb-nop-xceiv + OF_COMPATIBLE_0=usb-nop-xceiv + OF_COMPATIBLE_N=1 + OF_FULLNAME=/usbphynop1 + OF_NAME=usbphynop1 + SEQNUM=2493 + SUBSYSTEM=platform + USB_CHARGER_STATE=USB_CHARGER_PRESENT + USB_CHARGER_TYPE=USB_CHARGER_SDP_TYPE + USEC_INITIALIZED=227422826 + + USB charger is offline:: + + KERNEL change /devices/soc0/usbphynop1 (platform) + ACTION=change + DEVPATH=/devices/soc0/usbphynop1 + DRIVER=usb_phy_generic + MODALIAS=of:Nusbphynop1T(null)Cusb-nop-xceiv + OF_COMPATIBLE_0=usb-nop-xceiv + OF_COMPATIBLE_N=1 + OF_FULLNAME=/usbphynop1 + OF_NAME=usbphynop1 + SEQNUM=2494 + SUBSYSTEM=platform + USB_CHARGER_STATE=USB_CHARGER_ABSENT + USB_CHARGER_TYPE=USB_CHARGER_UNKNOWN_TYPE diff --git a/Documentation/ABI/testing/usb-uevent b/Documentation/ABI/testing/usb-uevent index d35c3cad892c..2b8eca4bf2b1 100644 --- a/Documentation/ABI/testing/usb-uevent +++ b/Documentation/ABI/testing/usb-uevent @@ -6,22 +6,22 @@ Description: When the USB Host Controller has entered a state where it is no longer functional a uevent will be raised. The uevent will contain ACTION=offline and ERROR=DEAD. - Here is an example taken using udevadm monitor -p: + Here is an example taken using udevadm monitor -p:: - KERNEL[130.428945] offline /devices/pci0000:00/0000:00:10.0/usb2 (usb) - ACTION=offline - BUSNUM=002 - DEVNAME=/dev/bus/usb/002/001 - DEVNUM=001 - DEVPATH=/devices/pci0000:00/0000:00:10.0/usb2 - DEVTYPE=usb_device - DRIVER=usb - ERROR=DEAD - MAJOR=189 - MINOR=128 - PRODUCT=1d6b/2/414 - SEQNUM=2168 - SUBSYSTEM=usb - TYPE=9/0/1 + KERNEL[130.428945] offline /devices/pci0000:00/0000:00:10.0/usb2 (usb) + ACTION=offline + BUSNUM=002 + DEVNAME=/dev/bus/usb/002/001 + DEVNUM=001 + DEVPATH=/devices/pci0000:00/0000:00:10.0/usb2 + DEVTYPE=usb_device + DRIVER=usb + ERROR=DEAD + MAJOR=189 + MINOR=128 + PRODUCT=1d6b/2/414 + SEQNUM=2168 + SUBSYSTEM=usb + TYPE=9/0/1 Users: chromium-os-dev@chromium.org diff --git a/scripts/get_abi.pl b/scripts/get_abi.pl index f6adf4b38a2b..d09bb37dd236 100755 --- a/scripts/get_abi.pl +++ b/scripts/get_abi.pl @@ -316,8 +316,6 @@ sub output_rest { $len = length($name) if (length($name) > $len); } - print "What:\n\n"; - print "+-" . "-" x $len . "-+\n"; foreach my $name (@names) { printf "| %s", $name . " " x ($len - length($name)) . " |\n"; -- cgit v1.2.3 From 4119f0dfbd27722216352a7349930a4b54032fe5 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:51 +0100 Subject: docs: ABI: change read/write attributes Unfortunately, (R) and (W) are valid markups for enumerated lists, as described at: https://docutils.sourceforge.io/docs/ref/rst/restructuredtext.html#enumerated-lists So, we ned to replace them by: (R) -> (Read) (W) -> (Write) As otherwise, (R) will be displayed as R., with is not what it is desired. There's no need to touch (RO) and (RW). Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/6e81ad8064f3ed4f8dc265086fdf1c618043f935.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/testing/debugfs-moxtet | 2 +- Documentation/ABI/testing/debugfs-turris-mox-rwtm | 15 +++-- .../ABI/testing/sysfs-bus-coresight-devices-cti | 60 ++++++++--------- .../ABI/testing/sysfs-bus-coresight-devices-etb10 | 16 ++--- .../ABI/testing/sysfs-bus-coresight-devices-etm3x | 10 +-- .../ABI/testing/sysfs-bus-coresight-devices-etm4x | 76 +++++++++++----------- .../ABI/testing/sysfs-bus-coresight-devices-stm | 2 +- .../ABI/testing/sysfs-bus-coresight-devices-tmc | 20 +++--- Documentation/ABI/testing/sysfs-bus-moxtet-devices | 6 +- .../ABI/testing/sysfs-firmware-turris-mox-rwtm | 10 +-- 10 files changed, 109 insertions(+), 108 deletions(-) diff --git a/Documentation/ABI/testing/debugfs-moxtet b/Documentation/ABI/testing/debugfs-moxtet index 15dfaecb2892..6eee10c3d5a1 100644 --- a/Documentation/ABI/testing/debugfs-moxtet +++ b/Documentation/ABI/testing/debugfs-moxtet @@ -2,7 +2,7 @@ What: /sys/kernel/debug/moxtet/input Date: March 2019 KernelVersion: 5.3 Contact: Marek Behún -Description: (R) Read input from the shift registers, in hexadecimal. +Description: (Read) Read input from the shift registers, in hexadecimal. Returns N+1 bytes, where N is the number of Moxtet connected modules. The first byte is from the CPU board itself. diff --git a/Documentation/ABI/testing/debugfs-turris-mox-rwtm b/Documentation/ABI/testing/debugfs-turris-mox-rwtm index ad08f535af3b..326df1b74707 100644 --- a/Documentation/ABI/testing/debugfs-turris-mox-rwtm +++ b/Documentation/ABI/testing/debugfs-turris-mox-rwtm @@ -4,10 +4,11 @@ KernelVersion: 5.8 Contact: Marek Behún Description: - === =========================================================== - (W) Message to sign with the ECDSA private key stored in - device's OTP. The message must be exactly 64 bytes (since - this is intended for SHA-512 hashes). - (R) The resulting signature, 136 bytes. This contains the R and - S values of the ECDSA signature, both in big-endian format. - === =========================================================== + ======= =========================================================== + (Write) Message to sign with the ECDSA private key stored in + device's OTP. The message must be exactly 64 bytes + (since this is intended for SHA-512 hashes). + (Read) The resulting signature, 136 bytes. This contains the + R and S values of the ECDSA signature, both in + big-endian format. + ======= =========================================================== diff --git a/Documentation/ABI/testing/sysfs-bus-coresight-devices-cti b/Documentation/ABI/testing/sysfs-bus-coresight-devices-cti index 9d11502b4390..bf2869c413e7 100644 --- a/Documentation/ABI/testing/sysfs-bus-coresight-devices-cti +++ b/Documentation/ABI/testing/sysfs-bus-coresight-devices-cti @@ -8,50 +8,50 @@ What: /sys/bus/coresight/devices//powered Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) Indicate if the CTI hardware is powered. +Description: (Read) Indicate if the CTI hardware is powered. What: /sys/bus/coresight/devices//ctmid Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) Display the associated CTM ID +Description: (Read) Display the associated CTM ID What: /sys/bus/coresight/devices//nr_trigger_cons Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) Number of devices connected to triggers on this CTI +Description: (Read) Number of devices connected to triggers on this CTI What: /sys/bus/coresight/devices//triggers/name Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) Name of connected device +Description: (Read) Name of connected device What: /sys/bus/coresight/devices//triggers/in_signals Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) Input trigger signals from connected device +Description: (Read) Input trigger signals from connected device What: /sys/bus/coresight/devices//triggers/in_types Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) Functional types for the input trigger signals +Description: (Read) Functional types for the input trigger signals from connected device What: /sys/bus/coresight/devices//triggers/out_signals Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) Output trigger signals to connected device +Description: (Read) Output trigger signals to connected device What: /sys/bus/coresight/devices//triggers/out_types Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) Functional types for the output trigger signals +Description: (Read) Functional types for the output trigger signals to connected device What: /sys/bus/coresight/devices//regs/inout_sel @@ -88,7 +88,7 @@ What: /sys/bus/coresight/devices//regs/intack Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (W) Write the INTACK register. +Description: (Write) Write the INTACK register. What: /sys/bus/coresight/devices//regs/appset Date: March 2020 @@ -101,99 +101,99 @@ What: /sys/bus/coresight/devices//regs/appclear Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (W) Write APPCLEAR register to deactivate channel. +Description: (Write) Write APPCLEAR register to deactivate channel. What: /sys/bus/coresight/devices//regs/apppulse Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (W) Write APPPULSE to pulse a channel active for one clock +Description: (Write) Write APPPULSE to pulse a channel active for one clock cycle. What: /sys/bus/coresight/devices//regs/chinstatus Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) Read current status of channel inputs. +Description: (Read) Read current status of channel inputs. What: /sys/bus/coresight/devices//regs/choutstatus Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) read current status of channel outputs. +Description: (Read) read current status of channel outputs. What: /sys/bus/coresight/devices//regs/triginstatus Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) read current status of input trigger signals +Description: (Read) read current status of input trigger signals What: /sys/bus/coresight/devices//regs/trigoutstatus Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) read current status of output trigger signals. +Description: (Read) read current status of output trigger signals. What: /sys/bus/coresight/devices//channels/trigin_attach Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (W) Attach a CTI input trigger to a CTM channel. +Description: (Write) Attach a CTI input trigger to a CTM channel. What: /sys/bus/coresight/devices//channels/trigin_detach Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (W) Detach a CTI input trigger from a CTM channel. +Description: (Write) Detach a CTI input trigger from a CTM channel. What: /sys/bus/coresight/devices//channels/trigout_attach Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (W) Attach a CTI output trigger to a CTM channel. +Description: (Write) Attach a CTI output trigger to a CTM channel. What: /sys/bus/coresight/devices//channels/trigout_detach Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (W) Detach a CTI output trigger from a CTM channel. +Description: (Write) Detach a CTI output trigger from a CTM channel. What: /sys/bus/coresight/devices//channels/chan_gate_enable Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (RW) Enable CTIGATE for single channel (W) or list enabled +Description: (RW) Enable CTIGATE for single channel (Write) or list enabled channels through the gate (R). What: /sys/bus/coresight/devices//channels/chan_gate_disable Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (W) Disable CTIGATE for single channel. +Description: (Write) Disable CTIGATE for single channel. What: /sys/bus/coresight/devices//channels/chan_set Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (W) Activate a single channel. +Description: (Write) Activate a single channel. What: /sys/bus/coresight/devices//channels/chan_clear Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (W) Deactivate a single channel. +Description: (Write) Deactivate a single channel. What: /sys/bus/coresight/devices//channels/chan_pulse Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (W) Pulse a single channel - activate for a single clock cycle. +Description: (Write) Pulse a single channel - activate for a single clock cycle. What: /sys/bus/coresight/devices//channels/trigout_filtered Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) List of output triggers filtered across all connections. +Description: (Read) List of output triggers filtered across all connections. What: /sys/bus/coresight/devices//channels/trig_filter_enable Date: March 2020 @@ -205,13 +205,13 @@ What: /sys/bus/coresight/devices//channels/chan_inuse Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) show channels with at least one attached trigger signal. +Description: (Read) show channels with at least one attached trigger signal. What: /sys/bus/coresight/devices//channels/chan_free Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) show channels with no attached trigger signals. +Description: (Read) show channels with no attached trigger signals. What: /sys/bus/coresight/devices//channels/chan_xtrigs_sel Date: March 2020 @@ -224,18 +224,18 @@ What: /sys/bus/coresight/devices//channels/chan_xtrigs_in Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) Read to see input triggers connected to selected view +Description: (Read) Read to see input triggers connected to selected view channel. What: /sys/bus/coresight/devices//channels/chan_xtrigs_out Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (R) Read to see output triggers connected to selected view +Description: (Read) Read to see output triggers connected to selected view channel. What: /sys/bus/coresight/devices//channels/chan_xtrigs_reset Date: March 2020 KernelVersion 5.7 Contact: Mike Leach or Mathieu Poirier -Description: (W) Clear all channel / trigger programming. +Description: (Write) Clear all channel / trigger programming. diff --git a/Documentation/ABI/testing/sysfs-bus-coresight-devices-etb10 b/Documentation/ABI/testing/sysfs-bus-coresight-devices-etb10 index 3e92cbd3fd83..9a383f6a74eb 100644 --- a/Documentation/ABI/testing/sysfs-bus-coresight-devices-etb10 +++ b/Documentation/ABI/testing/sysfs-bus-coresight-devices-etb10 @@ -23,21 +23,21 @@ What: /sys/bus/coresight/devices/.etb/mgmt/rdp Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Defines the depth, in words, of the trace RAM in powers of +Description: (Read) Defines the depth, in words, of the trace RAM in powers of 2. The value is read directly from HW register RDP, 0x004. What: /sys/bus/coresight/devices/.etb/mgmt/sts Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Shows the value held by the ETB status register. The value +Description: (Read) Shows the value held by the ETB status register. The value is read directly from HW register STS, 0x00C. What: /sys/bus/coresight/devices/.etb/mgmt/rrp Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Shows the value held by the ETB RAM Read Pointer register +Description: (Read) Shows the value held by the ETB RAM Read Pointer register that is used to read entries from the Trace RAM over the APB interface. The value is read directly from HW register RRP, 0x014. @@ -46,7 +46,7 @@ What: /sys/bus/coresight/devices/.etb/mgmt/rwp Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Shows the value held by the ETB RAM Write Pointer register +Description: (Read) Shows the value held by the ETB RAM Write Pointer register that is used to sets the write pointer to write entries from the CoreSight bus into the Trace RAM. The value is read directly from HW register RWP, 0x018. @@ -55,21 +55,21 @@ What: /sys/bus/coresight/devices/.etb/mgmt/trg Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Similar to "trigger_cntr" above except that this value is +Description: (Read) Similar to "trigger_cntr" above except that this value is read directly from HW register TRG, 0x01C. What: /sys/bus/coresight/devices/.etb/mgmt/ctl Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Shows the value held by the ETB Control register. The value +Description: (Read) Shows the value held by the ETB Control register. The value is read directly from HW register CTL, 0x020. What: /sys/bus/coresight/devices/.etb/mgmt/ffsr Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Shows the value held by the ETB Formatter and Flush Status +Description: (Read) Shows the value held by the ETB Formatter and Flush Status register. The value is read directly from HW register FFSR, 0x300. @@ -77,6 +77,6 @@ What: /sys/bus/coresight/devices/.etb/mgmt/ffcr Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Shows the value held by the ETB Formatter and Flush Control +Description: (Read) Shows the value held by the ETB Formatter and Flush Control register. The value is read directly from HW register FFCR, 0x304. diff --git a/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm3x b/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm3x index 924265a1295d..651602a61eac 100644 --- a/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm3x +++ b/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm3x @@ -146,28 +146,28 @@ What: /sys/bus/coresight/devices/.[etm|ptm]/nr_addr_cmp Date: November 2014 KernelVersion: 3.19 Contact: Mathieu Poirier -Description: (R) Provides the number of address comparators pairs accessible +Description: (Read) Provides the number of address comparators pairs accessible on a trace unit, as specified by bit 3:0 of register ETMCCR. What: /sys/bus/coresight/devices/.[etm|ptm]/nr_cntr Date: November 2014 KernelVersion: 3.19 Contact: Mathieu Poirier -Description: (R) Provides the number of counters accessible on a trace unit, +Description: (Read) Provides the number of counters accessible on a trace unit, as specified by bit 15:13 of register ETMCCR. What: /sys/bus/coresight/devices/.[etm|ptm]/nr_ctxid_cmp Date: November 2014 KernelVersion: 3.19 Contact: Mathieu Poirier -Description: (R) Provides the number of context ID comparator available on a +Description: (Read) Provides the number of context ID comparator available on a trace unit, as specified by bit 25:24 of register ETMCCR. What: /sys/bus/coresight/devices/.[etm|ptm]/reset Date: November 2014 KernelVersion: 3.19 Contact: Mathieu Poirier -Description: (W) Cancels all configuration on a trace unit and set it back +Description: (Write) Cancels all configuration on a trace unit and set it back to its boot configuration. What: /sys/bus/coresight/devices/.[etm|ptm]/seq_12_event @@ -216,7 +216,7 @@ What: /sys/bus/coresight/devices/.[etm|ptm]/curr_seq_state Date: November 2014 KernelVersion: 3.19 Contact: Mathieu Poirier -Description: (R) Holds the current state of the sequencer. +Description: (Read) Holds the current state of the sequencer. What: /sys/bus/coresight/devices/.[etm|ptm]/sync_freq Date: November 2014 diff --git a/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm4x b/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm4x index 614874e2cf53..881f0cd99ce4 100644 --- a/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm4x +++ b/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm4x @@ -12,75 +12,75 @@ What: /sys/bus/coresight/devices/etm/cpu Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) The CPU this tracing entity is associated with. +Description: (Read) The CPU this tracing entity is associated with. What: /sys/bus/coresight/devices/etm/nr_pe_cmp Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Indicates the number of PE comparator inputs that are +Description: (Read) Indicates the number of PE comparator inputs that are available for tracing. What: /sys/bus/coresight/devices/etm/nr_addr_cmp Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Indicates the number of address comparator pairs that are +Description: (Read) Indicates the number of address comparator pairs that are available for tracing. What: /sys/bus/coresight/devices/etm/nr_cntr Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Indicates the number of counters that are available for +Description: (Read) Indicates the number of counters that are available for tracing. What: /sys/bus/coresight/devices/etm/nr_ext_inp Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Indicates how many external inputs are implemented. +Description: (Read) Indicates how many external inputs are implemented. What: /sys/bus/coresight/devices/etm/numcidc Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Indicates the number of Context ID comparators that are +Description: (Read) Indicates the number of Context ID comparators that are available for tracing. What: /sys/bus/coresight/devices/etm/numvmidc Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Indicates the number of VMID comparators that are available +Description: (Read) Indicates the number of VMID comparators that are available for tracing. What: /sys/bus/coresight/devices/etm/nrseqstate Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Indicates the number of sequencer states that are +Description: (Read) Indicates the number of sequencer states that are implemented. What: /sys/bus/coresight/devices/etm/nr_resource Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Indicates the number of resource selection pairs that are +Description: (Read) Indicates the number of resource selection pairs that are available for tracing. What: /sys/bus/coresight/devices/etm/nr_ss_cmp Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Indicates the number of single-shot comparator controls that +Description: (Read) Indicates the number of single-shot comparator controls that are available for tracing. What: /sys/bus/coresight/devices/etm/reset Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (W) Cancels all configuration on a trace unit and set it back +Description: (Write) Cancels all configuration on a trace unit and set it back to its boot configuration. What: /sys/bus/coresight/devices/etm/mode @@ -300,7 +300,7 @@ What: /sys/bus/coresight/devices/etm/addr_cmp_view Date: December 2019 KernelVersion: 5.5 Contact: Mathieu Poirier -Description: (R) Print the current settings for the selected address +Description: (Read) Print the current settings for the selected address comparator. What: /sys/bus/coresight/devices/etm/sshot_idx @@ -319,7 +319,7 @@ What: /sys/bus/coresight/devices/etm/sshot_status Date: December 2019 KernelVersion: 5.5 Contact: Mathieu Poirier -Description: (R) Print the current value of the selected single shot +Description: (Read) Print the current value of the selected single shot status register. What: /sys/bus/coresight/devices/etm/sshot_pe_ctrl @@ -333,111 +333,111 @@ What: /sys/bus/coresight/devices/etm/mgmt/trcoslsr Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Print the content of the OS Lock Status Register (0x304). +Description: (Read) Print the content of the OS Lock Status Register (0x304). The value it taken directly from the HW. What: /sys/bus/coresight/devices/etm/mgmt/trcpdcr Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Print the content of the Power Down Control Register +Description: (Read) Print the content of the Power Down Control Register (0x310). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/mgmt/trcpdsr Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Print the content of the Power Down Status Register +Description: (Read) Print the content of the Power Down Status Register (0x314). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/mgmt/trclsr Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Print the content of the SW Lock Status Register +Description: (Read) Print the content of the SW Lock Status Register (0xFB4). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/mgmt/trcauthstatus Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Print the content of the Authentication Status Register +Description: (Read) Print the content of the Authentication Status Register (0xFB8). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/mgmt/trcdevid Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Print the content of the Device ID Register +Description: (Read) Print the content of the Device ID Register (0xFC8). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/mgmt/trcdevtype Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Print the content of the Device Type Register +Description: (Read) Print the content of the Device Type Register (0xFCC). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/mgmt/trcpidr0 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Print the content of the Peripheral ID0 Register +Description: (Read) Print the content of the Peripheral ID0 Register (0xFE0). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/mgmt/trcpidr1 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Print the content of the Peripheral ID1 Register +Description: (Read) Print the content of the Peripheral ID1 Register (0xFE4). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/mgmt/trcpidr2 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Print the content of the Peripheral ID2 Register +Description: (Read) Print the content of the Peripheral ID2 Register (0xFE8). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/mgmt/trcpidr3 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Print the content of the Peripheral ID3 Register +Description: (Read) Print the content of the Peripheral ID3 Register (0xFEC). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/mgmt/trcconfig Date: February 2016 KernelVersion: 4.07 Contact: Mathieu Poirier -Description: (R) Print the content of the trace configuration register +Description: (Read) Print the content of the trace configuration register (0x010) as currently set by SW. What: /sys/bus/coresight/devices/etm/mgmt/trctraceid Date: February 2016 KernelVersion: 4.07 Contact: Mathieu Poirier -Description: (R) Print the content of the trace ID register (0x040). +Description: (Read) Print the content of the trace ID register (0x040). What: /sys/bus/coresight/devices/etm/trcidr/trcidr0 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Returns the tracing capabilities of the trace unit (0x1E0). +Description: (Read) Returns the tracing capabilities of the trace unit (0x1E0). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/trcidr/trcidr1 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Returns the tracing capabilities of the trace unit (0x1E4). +Description: (Read) Returns the tracing capabilities of the trace unit (0x1E4). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/trcidr/trcidr2 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Returns the maximum size of the data value, data address, +Description: (Read) Returns the maximum size of the data value, data address, VMID, context ID and instuction address in the trace unit (0x1E8). The value is taken directly from the HW. @@ -445,7 +445,7 @@ What: /sys/bus/coresight/devices/etm/trcidr/trcidr3 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Returns the value associated with various resources +Description: (Read) Returns the value associated with various resources available to the trace unit. See the Trace Macrocell architecture specification for more details (0x1E8). The value is taken directly from the HW. @@ -454,42 +454,42 @@ What: /sys/bus/coresight/devices/etm/trcidr/trcidr4 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Returns how many resources the trace unit supports (0x1F0). +Description: (Read) Returns how many resources the trace unit supports (0x1F0). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/trcidr/trcidr5 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Returns how many resources the trace unit supports (0x1F4). +Description: (Read) Returns how many resources the trace unit supports (0x1F4). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/trcidr/trcidr8 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Returns the maximum speculation depth of the instruction +Description: (Read) Returns the maximum speculation depth of the instruction trace stream. (0x180). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/trcidr/trcidr9 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Returns the number of P0 right-hand keys that the trace unit +Description: (Read) Returns the number of P0 right-hand keys that the trace unit can use (0x184). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/trcidr/trcidr10 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Returns the number of P1 right-hand keys that the trace unit +Description: (Read) Returns the number of P1 right-hand keys that the trace unit can use (0x188). The value is taken directly from the HW. What: /sys/bus/coresight/devices/etm/trcidr/trcidr11 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Returns the number of special P1 right-hand keys that the +Description: (Read) Returns the number of special P1 right-hand keys that the trace unit can use (0x18C). The value is taken directly from the HW. @@ -497,7 +497,7 @@ What: /sys/bus/coresight/devices/etm/trcidr/trcidr12 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Returns the number of conditional P1 right-hand keys that +Description: (Read) Returns the number of conditional P1 right-hand keys that the trace unit can use (0x190). The value is taken directly from the HW. @@ -505,6 +505,6 @@ What: /sys/bus/coresight/devices/etm/trcidr/trcidr13 Date: April 2015 KernelVersion: 4.01 Contact: Mathieu Poirier -Description: (R) Returns the number of special conditional P1 right-hand keys +Description: (Read) Returns the number of special conditional P1 right-hand keys that the trace unit can use (0x194). The value is taken directly from the HW. diff --git a/Documentation/ABI/testing/sysfs-bus-coresight-devices-stm b/Documentation/ABI/testing/sysfs-bus-coresight-devices-stm index 1dffabe7f48d..53e1f4815d64 100644 --- a/Documentation/ABI/testing/sysfs-bus-coresight-devices-stm +++ b/Documentation/ABI/testing/sysfs-bus-coresight-devices-stm @@ -42,7 +42,7 @@ What: /sys/bus/coresight/devices/.stm/status Date: April 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) List various control and status registers. The specific +Description: (Read) List various control and status registers. The specific layout and content is driver specific. What: /sys/bus/coresight/devices/.stm/traceid diff --git a/Documentation/ABI/testing/sysfs-bus-coresight-devices-tmc b/Documentation/ABI/testing/sysfs-bus-coresight-devices-tmc index ab49b9ac3bcb..6aa527296c71 100644 --- a/Documentation/ABI/testing/sysfs-bus-coresight-devices-tmc +++ b/Documentation/ABI/testing/sysfs-bus-coresight-devices-tmc @@ -11,21 +11,21 @@ What: /sys/bus/coresight/devices/.tmc/mgmt/rsz Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Defines the size, in 32-bit words, of the local RAM buffer. +Description: (Read) Defines the size, in 32-bit words, of the local RAM buffer. The value is read directly from HW register RSZ, 0x004. What: /sys/bus/coresight/devices/.tmc/mgmt/sts Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Shows the value held by the TMC status register. The value +Description: (Read) Shows the value held by the TMC status register. The value is read directly from HW register STS, 0x00C. What: /sys/bus/coresight/devices/.tmc/mgmt/rrp Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Shows the value held by the TMC RAM Read Pointer register +Description: (Read) Shows the value held by the TMC RAM Read Pointer register that is used to read entries from the Trace RAM over the APB interface. The value is read directly from HW register RRP, 0x014. @@ -34,7 +34,7 @@ What: /sys/bus/coresight/devices/.tmc/mgmt/rwp Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Shows the value held by the TMC RAM Write Pointer register +Description: (Read) Shows the value held by the TMC RAM Write Pointer register that is used to sets the write pointer to write entries from the CoreSight bus into the Trace RAM. The value is read directly from HW register RWP, 0x018. @@ -43,21 +43,21 @@ What: /sys/bus/coresight/devices/.tmc/mgmt/trg Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Similar to "trigger_cntr" above except that this value is +Description: (Read) Similar to "trigger_cntr" above except that this value is read directly from HW register TRG, 0x01C. What: /sys/bus/coresight/devices/.tmc/mgmt/ctl Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Shows the value held by the TMC Control register. The value +Description: (Read) Shows the value held by the TMC Control register. The value is read directly from HW register CTL, 0x020. What: /sys/bus/coresight/devices/.tmc/mgmt/ffsr Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Shows the value held by the TMC Formatter and Flush Status +Description: (Read) Shows the value held by the TMC Formatter and Flush Status register. The value is read directly from HW register FFSR, 0x300. @@ -65,7 +65,7 @@ What: /sys/bus/coresight/devices/.tmc/mgmt/ffcr Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Shows the value held by the TMC Formatter and Flush Control +Description: (Read) Shows the value held by the TMC Formatter and Flush Control register. The value is read directly from HW register FFCR, 0x304. @@ -73,7 +73,7 @@ What: /sys/bus/coresight/devices/.tmc/mgmt/mode Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Shows the value held by the TMC Mode register, which +Description: (Read) Shows the value held by the TMC Mode register, which indicate the mode the device has been configured to enact. The The value is read directly from the MODE register, 0x028. @@ -81,7 +81,7 @@ What: /sys/bus/coresight/devices/.tmc/mgmt/devid Date: March 2016 KernelVersion: 4.7 Contact: Mathieu Poirier -Description: (R) Indicates the capabilities of the Coresight TMC. +Description: (Read) Indicates the capabilities of the Coresight TMC. The value is read directly from the DEVID register, 0xFC8, What: /sys/bus/coresight/devices/.tmc/buffer_size diff --git a/Documentation/ABI/testing/sysfs-bus-moxtet-devices b/Documentation/ABI/testing/sysfs-bus-moxtet-devices index 355958527fa3..4a6d61b44f3f 100644 --- a/Documentation/ABI/testing/sysfs-bus-moxtet-devices +++ b/Documentation/ABI/testing/sysfs-bus-moxtet-devices @@ -2,16 +2,16 @@ What: /sys/bus/moxtet/devices/moxtet-./module_description Date: March 2019 KernelVersion: 5.3 Contact: Marek Behún -Description: (R) Moxtet module description. Format: string +Description: (Read) Moxtet module description. Format: string What: /sys/bus/moxtet/devices/moxtet-./module_id Date: March 2019 KernelVersion: 5.3 Contact: Marek Behún -Description: (R) Moxtet module ID. Format: %x +Description: (Read) Moxtet module ID. Format: %x What: /sys/bus/moxtet/devices/moxtet-./module_name Date: March 2019 KernelVersion: 5.3 Contact: Marek Behún -Description: (R) Moxtet module name. Format: string +Description: (Read) Moxtet module name. Format: string diff --git a/Documentation/ABI/testing/sysfs-firmware-turris-mox-rwtm b/Documentation/ABI/testing/sysfs-firmware-turris-mox-rwtm index 15595fab88d1..b8631f5a29c4 100644 --- a/Documentation/ABI/testing/sysfs-firmware-turris-mox-rwtm +++ b/Documentation/ABI/testing/sysfs-firmware-turris-mox-rwtm @@ -2,21 +2,21 @@ What: /sys/firmware/turris-mox-rwtm/board_version Date: August 2019 KernelVersion: 5.4 Contact: Marek Behún -Description: (R) Board version burned into eFuses of this Turris Mox board. +Description: (Read) Board version burned into eFuses of this Turris Mox board. Format: %i What: /sys/firmware/turris-mox-rwtm/mac_address* Date: August 2019 KernelVersion: 5.4 Contact: Marek Behún -Description: (R) MAC addresses burned into eFuses of this Turris Mox board. +Description: (Read) MAC addresses burned into eFuses of this Turris Mox board. Format: %pM What: /sys/firmware/turris-mox-rwtm/pubkey Date: August 2019 KernelVersion: 5.4 Contact: Marek Behún -Description: (R) ECDSA public key (in pubkey hex compressed form) computed +Description: (Read) ECDSA public key (in pubkey hex compressed form) computed as pair to the ECDSA private key burned into eFuses of this Turris Mox Board. Format: string @@ -25,7 +25,7 @@ What: /sys/firmware/turris-mox-rwtm/ram_size Date: August 2019 KernelVersion: 5.4 Contact: Marek Behún -Description: (R) RAM size in MiB of this Turris Mox board as was detected +Description: (Read) RAM size in MiB of this Turris Mox board as was detected during manufacturing and burned into eFuses. Can be 512 or 1024. Format: %i @@ -33,5 +33,5 @@ What: /sys/firmware/turris-mox-rwtm/serial_number Date: August 2019 KernelVersion: 5.4 Contact: Marek Behún -Description: (R) Serial number burned into eFuses of this Turris Mox device. +Description: (Read) Serial number burned into eFuses of this Turris Mox device. Format: %016X -- cgit v1.2.3 From 637ae1fb5dca38649eed31f943114486bacce416 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:52 +0100 Subject: docs: ABI: stable: remove a duplicated documentation Perhaps due to a wrong cut-and-paste, this entry: What: /sys/bus/vmbus/devices//channels//cpu was added twice by the same patch, one following the other. Remove the duplication. Fixes: c2e5df616e1a ("vmbus: add per-channel sysfs info") Acked-by: Wei Liu Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/091e8de5543c280ceb47edcb3ab6d0e9f3fa085b.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/stable/sysfs-bus-vmbus | 7 ------- 1 file changed, 7 deletions(-) diff --git a/Documentation/ABI/stable/sysfs-bus-vmbus b/Documentation/ABI/stable/sysfs-bus-vmbus index 8e8d167eca31..c27b7b89477c 100644 --- a/Documentation/ABI/stable/sysfs-bus-vmbus +++ b/Documentation/ABI/stable/sysfs-bus-vmbus @@ -63,13 +63,6 @@ Contact: Stephen Hemminger Description: VCPU (sub)channel is affinitized to Users: tools/hv/lsvmbus and other debugging tools -What: /sys/bus/vmbus/devices//channels//cpu -Date: September. 2017 -KernelVersion: 4.14 -Contact: Stephen Hemminger -Description: VCPU (sub)channel is affinitized to -Users: tools/hv/lsvmbus and other debugging tools - What: /sys/bus/vmbus/devices//channels//in_mask Date: September. 2017 KernelVersion: 4.14 -- cgit v1.2.3 From edef2fa29739172643dbe9f44fb38d23793816d6 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:53 +0100 Subject: docs: ABI: unify /sys/class/leds//brightness documentation This ABI is defined twice, one for normal leds and another one for multicolor ones. Ensure that just one definition is there at ABI. Acked-by: Pavel Machek Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/1115022e7f172b19ee8610f2ad28cc2f0ca93592.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/testing/sysfs-class-led | 25 +++++++++++++++++++--- .../ABI/testing/sysfs-class-led-multicolor | 23 ++++++-------------- 2 files changed, 28 insertions(+), 20 deletions(-) diff --git a/Documentation/ABI/testing/sysfs-class-led b/Documentation/ABI/testing/sysfs-class-led index 0ed5c2629c6f..2e24ac3bd7ef 100644 --- a/Documentation/ABI/testing/sysfs-class-led +++ b/Documentation/ABI/testing/sysfs-class-led @@ -3,9 +3,26 @@ Date: March 2006 KernelVersion: 2.6.17 Contact: Richard Purdie Description: - Set the brightness of the LED. Most LEDs don't - have hardware brightness support, so will just be turned on for - non-zero brightness settings. The value is between 0 and + Set the brightness of the LED. + + Most LEDs don't have hardware brightness support, so will + just be turned on for non-zero brightness settings. + + .. Note:: + + For multicolor LEDs, writing to this file will update all + LEDs within the group to a calculated percentage of what + each color LED intensity is set to. + + The percentage is calculated for each grouped LED via + the equation below:: + + led_brightness = brightness * multi_intensity/max_brightness + + For additional details please refer to + Documentation/leds/leds-class-multicolor.rst. + + The value is between 0 and /sys/class/leds//max_brightness. Writing 0 to this file clears active trigger. @@ -13,6 +30,8 @@ Description: Writing non-zero to this file while trigger is active changes the top brightness trigger is going to use. + + What: /sys/class/leds//max_brightness Date: March 2006 KernelVersion: 2.6.17 diff --git a/Documentation/ABI/testing/sysfs-class-led-multicolor b/Documentation/ABI/testing/sysfs-class-led-multicolor index eeeddcbdbbe3..16fc827b10cb 100644 --- a/Documentation/ABI/testing/sysfs-class-led-multicolor +++ b/Documentation/ABI/testing/sysfs-class-led-multicolor @@ -1,20 +1,3 @@ -What: /sys/class/leds//brightness -Date: March 2020 -KernelVersion: 5.9 -Contact: Dan Murphy -Description: read/write - Writing to this file will update all LEDs within the group to a - calculated percentage of what each color LED intensity is set - to. The percentage is calculated for each grouped LED via the - equation below: - - led_brightness = brightness * multi_intensity/max_brightness - - For additional details please refer to - Documentation/leds/leds-class-multicolor.rst. - - The value of the LED is from 0 to - /sys/class/leds//max_brightness. What: /sys/class/leds//multi_index Date: March 2020 @@ -25,6 +8,9 @@ Description: read as an array of strings as they are indexed in the multi_intensity file. + For additional details please refer to + Documentation/leds/leds-class-multicolor.rst. + What: /sys/class/leds//multi_intensity Date: March 2020 KernelVersion: 5.9 @@ -33,3 +19,6 @@ Description: read/write This file contains array of integers. Order of components is described by the multi_index array. The maximum intensity should not exceed /sys/class/leds//max_brightness. + + For additional details please refer to + Documentation/leds/leds-class-multicolor.rst. -- cgit v1.2.3 From 9e0b29adfb3323ffd250b74373c68209e291b8e5 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:54 +0100 Subject: docs: ABI: sysfs-class-power: unify duplicated properties The ABI is not supposed to have duplicated entries, as warned by get_abi.pl: $ ./scripts/get_abi.pl validate 2>&1|grep sysfs-class-power Warning: /sys/class/power_supply//current_avg is defined 2 times: Documentation/ABI/testing/sysfs-class-power:108 Documentation/ABI/testing/sysfs-class-power:391 Warning: /sys/class/power_supply//current_max is defined 2 times: Documentation/ABI/testing/sysfs-class-power:121 Documentation/ABI/testing/sysfs-class-power:404 Warning: /sys/class/power_supply//current_now is defined 2 times: Documentation/ABI/testing/sysfs-class-power:130 Documentation/ABI/testing/sysfs-class-power:414 Warning: /sys/class/power_supply//temp is defined 2 times: Documentation/ABI/testing/sysfs-class-power:281 Documentation/ABI/testing/sysfs-class-power:493 Warning: /sys/class/power_supply//temp_alert_max is defined 2 times: Documentation/ABI/testing/sysfs-class-power:291 Documentation/ABI/testing/sysfs-class-power:505 Warning: /sys/class/power_supply//temp_alert_min is defined 2 times: Documentation/ABI/testing/sysfs-class-power:306 Documentation/ABI/testing/sysfs-class-power:521 Warning: /sys/class/power_supply//temp_max is defined 2 times: Documentation/ABI/testing/sysfs-class-power:322 Documentation/ABI/testing/sysfs-class-power:537 Warning: /sys/class/power_supply//temp_min is defined 2 times: Documentation/ABI/testing/sysfs-class-power:333 Documentation/ABI/testing/sysfs-class-power:547 Warning: /sys/class/power_supply//voltage_max is defined 2 times: Documentation/ABI/testing/sysfs-class-power:356 Documentation/ABI/testing/sysfs-class-power:571 Warning: /sys/class/power_supply//voltage_min is defined 2 times: Documentation/ABI/testing/sysfs-class-power:367 Documentation/ABI/testing/sysfs-class-power:581 Warning: /sys/class/power_supply//voltage_now is defined 2 times: Documentation/ABI/testing/sysfs-class-power:378 Documentation/ABI/testing/sysfs-class-power:591 Yet, both USB and Battery share a common set of charging-related properties. Unify the entries for such properties in order to avoid duplication, while preserving the battery and USB-specific data properly documented. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/bcdf5f76326ea48a990a7cac612af216c387537d.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/testing/sysfs-class-power | 486 +++++++++++++--------------- 1 file changed, 223 insertions(+), 263 deletions(-) diff --git a/Documentation/ABI/testing/sysfs-class-power b/Documentation/ABI/testing/sysfs-class-power index d68ad528a8e5..ca830c6cd809 100644 --- a/Documentation/ABI/testing/sysfs-class-power +++ b/Documentation/ABI/testing/sysfs-class-power @@ -36,7 +36,229 @@ Description: Access: Read Valid values: "Battery", "UPS", "Mains", "USB", "Wireless" -===== Battery Properties ===== +**Battery and USB properties** + +What: /sys/class/power_supply//current_avg +Date: May 2007 +Contact: linux-pm@vger.kernel.org +Description: + Battery: + + Reports an average IBAT current reading for the battery, over + a fixed period. Normally devices will provide a fixed interval + in which they average readings to smooth out the reported + value. + + USB: + + Reports an average IBUS current reading over a fixed period. + Normally devices will provide a fixed interval in which they + average readings to smooth out the reported value. + + Access: Read + + Valid values: Represented in microamps. Negative values are + used for discharging batteries, positive values for charging + batteries and for USB IBUS current. + +What: /sys/class/power_supply//current_max +Date: October 2010 +Contact: linux-pm@vger.kernel.org +Description: + Battery: + + Reports the maximum IBAT current allowed into the battery. + + USB: + + Reports the maximum IBUS current the supply can support. + + Access: Read + Valid values: Represented in microamps + +What: /sys/class/power_supply//current_now +Date: May 2007 +Contact: linux-pm@vger.kernel.org +Description: + + Battery: + + Reports an instant, single IBAT current reading for the + battery. This value is not averaged/smoothed. + + Access: Read + + USB: + + Reports the IBUS current supplied now. This value is generally + read-only reporting, unless the 'online' state of the supply + is set to be programmable, in which case this value can be set + within the reported min/max range. + + Access: Read, Write + + Valid values: Represented in microamps. Negative values are + used for discharging batteries, positive values for charging + batteries and for USB IBUS current. + +What: /sys/class/power_supply//temp +Date: May 2007 +Contact: linux-pm@vger.kernel.org +Description: + Battery: + + Reports the current TBAT battery temperature reading. + + USB: + + Reports the current supply temperature reading. This would + normally be the internal temperature of the device itself + (e.g TJUNC temperature of an IC) + + Access: Read + + Valid values: Represented in 1/10 Degrees Celsius + +What: /sys/class/power_supply//temp_alert_max +Date: July 2012 +Contact: linux-pm@vger.kernel.org +Description: + Battery: + + Maximum TBAT temperature trip-wire value where the supply will + notify user-space of the event. + + USB: + + Maximum supply temperature trip-wire value where the supply + will notify user-space of the event. + + This is normally used for the charging scenario where + user-space needs to know if the temperature has crossed an + upper threshold so it can take appropriate action (e.g. warning + user that the temperature is critically high, and charging has + stopped). + + Access: Read + + Valid values: Represented in 1/10 Degrees Celsius + +What: /sys/class/power_supply//temp_alert_min +Date: July 2012 +Contact: linux-pm@vger.kernel.org +Description: + + Battery: + + Minimum TBAT temperature trip-wire value where the supply will + notify user-space of the event. + + USB: + + Minimum supply temperature trip-wire value where the supply + will notify user-space of the event. + + This is normally used for the charging scenario where user-space + needs to know if the temperature has crossed a lower threshold + so it can take appropriate action (e.g. warning user that + temperature level is high, and charging current has been + reduced accordingly to remedy the situation). + + Access: Read + + Valid values: Represented in 1/10 Degrees Celsius + +What: /sys/class/power_supply//temp_max +Date: July 2014 +Contact: linux-pm@vger.kernel.org +Description: + Battery: + + Reports the maximum allowed TBAT battery temperature for + charging. + + USB: + + Reports the maximum allowed supply temperature for operation. + + Access: Read + + Valid values: Represented in 1/10 Degrees Celsius + +What: /sys/class/power_supply//temp_min +Date: July 2014 +Contact: linux-pm@vger.kernel.org +Description: + Battery: + + Reports the minimum allowed TBAT battery temperature for + charging. + + USB: + + Reports the minimum allowed supply temperature for operation. + + Access: Read + + Valid values: Represented in 1/10 Degrees Celsius + +What: /sys/class/power_supply//voltage_max, +Date: January 2008 +Contact: linux-pm@vger.kernel.org +Description: + Battery: + + Reports the maximum safe VBAT voltage permitted for the + battery, during charging. + + USB: + + Reports the maximum VBUS voltage the supply can support. + + Access: Read + + Valid values: Represented in microvolts + +What: /sys/class/power_supply//voltage_min, +Date: January 2008 +Contact: linux-pm@vger.kernel.org +Description: + Battery: + + Reports the minimum safe VBAT voltage permitted for the + battery, during discharging. + + USB: + + Reports the minimum VBUS voltage the supply can support. + + Access: Read + + Valid values: Represented in microvolts + +What: /sys/class/power_supply//voltage_now, +Date: May 2007 +Contact: linux-pm@vger.kernel.org +Description: + Battery: + + Reports an instant, single VBAT voltage reading for the + battery. This value is not averaged/smoothed. + + Access: Read + + USB: + + Reports the VBUS voltage supplied now. This value is generally + read-only reporting, unless the 'online' state of the supply + is set to be programmable, in which case this value can be set + within the reported min/max range. + + Access: Read, Write + + Valid values: Represented in microvolts + +**Battery Properties** What: /sys/class/power_supply//capacity Date: May 2007 @@ -106,40 +328,6 @@ Description: "Unknown", "Critical", "Low", "Normal", "High", "Full" -What: /sys/class/power_supply//current_avg -Date: May 2007 -Contact: linux-pm@vger.kernel.org -Description: - Reports an average IBAT current reading for the battery, over a - fixed period. Normally devices will provide a fixed interval in - which they average readings to smooth out the reported value. - - Access: Read - - Valid values: Represented in microamps. Negative values are used - for discharging batteries, positive values for charging batteries. - -What: /sys/class/power_supply//current_max -Date: October 2010 -Contact: linux-pm@vger.kernel.org -Description: - Reports the maximum IBAT current allowed into the battery. - - Access: Read - Valid values: Represented in microamps - -What: /sys/class/power_supply//current_now -Date: May 2007 -Contact: linux-pm@vger.kernel.org -Description: - Reports an instant, single IBAT current reading for the battery. - This value is not averaged/smoothed. - - Access: Read - - Valid values: Represented in microamps. Negative values are used - for discharging batteries, positive values for charging batteries. - What: /sys/class/power_supply//charge_control_limit Date: Oct 2012 Contact: linux-pm@vger.kernel.org @@ -279,68 +467,6 @@ Description: "Unknown", "NiMH", "Li-ion", "Li-poly", "LiFe", "NiCd", "LiMn" -What: /sys/class/power_supply//temp -Date: May 2007 -Contact: linux-pm@vger.kernel.org -Description: - Reports the current TBAT battery temperature reading. - - Access: Read - - Valid values: Represented in 1/10 Degrees Celsius - -What: /sys/class/power_supply//temp_alert_max -Date: July 2012 -Contact: linux-pm@vger.kernel.org -Description: - Maximum TBAT temperature trip-wire value where the supply will - notify user-space of the event. This is normally used for the - battery charging scenario where user-space needs to know the - battery temperature has crossed an upper threshold so it can - take appropriate action (e.g. warning user that battery level is - critically high, and charging has stopped). - - Access: Read - - Valid values: Represented in 1/10 Degrees Celsius - -What: /sys/class/power_supply//temp_alert_min -Date: July 2012 -Contact: linux-pm@vger.kernel.org -Description: - Minimum TBAT temperature trip-wire value where the supply will - notify user-space of the event. This is normally used for the - battery charging scenario where user-space needs to know the - battery temperature has crossed a lower threshold so it can take - appropriate action (e.g. warning user that battery level is - high, and charging current has been reduced accordingly to - remedy the situation). - - Access: Read - - Valid values: Represented in 1/10 Degrees Celsius - -What: /sys/class/power_supply//temp_max -Date: July 2014 -Contact: linux-pm@vger.kernel.org -Description: - Reports the maximum allowed TBAT battery temperature for - charging. - - Access: Read - - Valid values: Represented in 1/10 Degrees Celsius - -What: /sys/class/power_supply//temp_min -Date: July 2014 -Contact: linux-pm@vger.kernel.org -Description: - Reports the minimum allowed TBAT battery temperature for - charging. - - Access: Read - - Valid values: Represented in 1/10 Degrees Celsius What: /sys/class/power_supply//voltage_avg, Date: May 2007 @@ -354,77 +480,8 @@ Description: Valid values: Represented in microvolts -What: /sys/class/power_supply//voltage_max, -Date: January 2008 -Contact: linux-pm@vger.kernel.org -Description: - Reports the maximum safe VBAT voltage permitted for the battery, - during charging. - - Access: Read - - Valid values: Represented in microvolts - -What: /sys/class/power_supply//voltage_min, -Date: January 2008 -Contact: linux-pm@vger.kernel.org -Description: - Reports the minimum safe VBAT voltage permitted for the battery, - during discharging. - - Access: Read - - Valid values: Represented in microvolts - -What: /sys/class/power_supply//voltage_now, -Date: May 2007 -Contact: linux-pm@vger.kernel.org -Description: - Reports an instant, single VBAT voltage reading for the battery. - This value is not averaged/smoothed. - - Access: Read - - Valid values: Represented in microvolts - **USB Properties** -What: /sys/class/power_supply//current_avg -Date: May 2007 -Contact: linux-pm@vger.kernel.org -Description: - Reports an average IBUS current reading over a fixed period. - Normally devices will provide a fixed interval in which they - average readings to smooth out the reported value. - - Access: Read - - Valid values: Represented in microamps - - -What: /sys/class/power_supply//current_max -Date: October 2010 -Contact: linux-pm@vger.kernel.org -Description: - Reports the maximum IBUS current the supply can support. - - Access: Read - - Valid values: Represented in microamps - -What: /sys/class/power_supply//current_now -Date: May 2007 -Contact: linux-pm@vger.kernel.org -Description: - Reports the IBUS current supplied now. This value is generally - read-only reporting, unless the 'online' state of the supply - is set to be programmable, in which case this value can be set - within the reported min/max range. - - Access: Read, Write - - Valid values: Represented in microamps - What: /sys/class/power_supply//input_current_limit Date: July 2014 Contact: linux-pm@vger.kernel.org @@ -491,70 +548,6 @@ Description: 2: Online Programmable - Programmable Voltage Supply == ================================================== -What: /sys/class/power_supply//temp -Date: May 2007 -Contact: linux-pm@vger.kernel.org -Description: - Reports the current supply temperature reading. This would - normally be the internal temperature of the device itself (e.g - TJUNC temperature of an IC) - - Access: Read - - Valid values: Represented in 1/10 Degrees Celsius - -What: /sys/class/power_supply//temp_alert_max -Date: July 2012 -Contact: linux-pm@vger.kernel.org -Description: - Maximum supply temperature trip-wire value where the supply will - notify user-space of the event. This is normally used for the - charging scenario where user-space needs to know the supply - temperature has crossed an upper threshold so it can take - appropriate action (e.g. warning user that the supply - temperature is critically high, and charging has stopped to - remedy the situation). - - Access: Read - - Valid values: Represented in 1/10 Degrees Celsius - -What: /sys/class/power_supply//temp_alert_min -Date: July 2012 -Contact: linux-pm@vger.kernel.org -Description: - Minimum supply temperature trip-wire value where the supply will - notify user-space of the event. This is normally used for the - charging scenario where user-space needs to know the supply - temperature has crossed a lower threshold so it can take - appropriate action (e.g. warning user that the supply - temperature is high, and charging current has been reduced - accordingly to remedy the situation). - - Access: Read - - Valid values: Represented in 1/10 Degrees Celsius - -What: /sys/class/power_supply//temp_max -Date: July 2014 -Contact: linux-pm@vger.kernel.org -Description: - Reports the maximum allowed supply temperature for operation. - - Access: Read - - Valid values: Represented in 1/10 Degrees Celsius - -What: /sys/class/power_supply//temp_min -Date: July 2014 -Contact: linux-pm@vger.kernel.org -Description: - Reports the mainimum allowed supply temperature for operation. - - Access: Read - - Valid values: Represented in 1/10 Degrees Celsius - What: /sys/class/power_supply//usb_type Date: March 2018 Contact: linux-pm@vger.kernel.org @@ -569,39 +562,6 @@ Description: "Unknown", "SDP", "DCP", "CDP", "ACA", "C", "PD", "PD_DRP", "PD_PPS", "BrickID" -What: /sys/class/power_supply//voltage_max -Date: January 2008 -Contact: linux-pm@vger.kernel.org -Description: - Reports the maximum VBUS voltage the supply can support. - - Access: Read - - Valid values: Represented in microvolts - -What: /sys/class/power_supply//voltage_min -Date: January 2008 -Contact: linux-pm@vger.kernel.org -Description: - Reports the minimum VBUS voltage the supply can support. - - Access: Read - - Valid values: Represented in microvolts - -What: /sys/class/power_supply//voltage_now -Date: May 2007 -Contact: linux-pm@vger.kernel.org -Description: - Reports the VBUS voltage supplied now. This value is generally - read-only reporting, unless the 'online' state of the supply - is set to be programmable, in which case this value can be set - within the reported min/max range. - - Access: Read, Write - - Valid values: Represented in microvolts - **Device Specific Properties** What: /sys/class/power/ds2760-battery.*/charge_now -- cgit v1.2.3 From 1b3023ef54d4fc8e8a034d2178ef80bddd7ba206 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:55 +0100 Subject: docs: ABI: sysfs-c2port: remove a duplicated entry As warned by scripts/get_abi.pl: Warning: /sys/class/c2port/c2portX/flash_erase is defined 2 times: Documentation/ABI/testing/sysfs-c2port:60 Documentation/ABI/testing/sysfs-c2port:68 This entry was added twice at the same patch. Probalby a cut-and paste issue. Fixes: 4e17e1db9647 ("Add c2 port support") Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/4ea2b31ddc76161fbae004f2f2ed91dfb757703f.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/testing/sysfs-c2port | 7 ------- 1 file changed, 7 deletions(-) diff --git a/Documentation/ABI/testing/sysfs-c2port b/Documentation/ABI/testing/sysfs-c2port index 716cffc457e9..f7b8cf6e4398 100644 --- a/Documentation/ABI/testing/sysfs-c2port +++ b/Documentation/ABI/testing/sysfs-c2port @@ -66,13 +66,6 @@ Description: the "erase" command on the on-board flash of the connected micro. -What: /sys/class/c2port/c2portX/flash_erase -Date: October 2008 -Contact: Rodolfo Giometti -Description: - The /sys/class/c2port/c2portX/flash_erase file show the - on-board flash size of the connected micro. - What: /sys/class/c2port/c2portX/reset Date: October 2008 Contact: Rodolfo Giometti -- cgit v1.2.3 From bd6e8441c2fc4d653d115516f55b1700284bc42d Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:56 +0100 Subject: docs: ABI: sysfs-class-backlight: unify ABI documentation Both adp8860 and adp8870 define some extensions to the backlight class. This causes warnings: Warning: /sys/class/backlight//ambient_light_level is defined 2 times: /sys/class/backlight//ambient_light_level:8 /sys/class/backlight//ambient_light_level:30 Warning: /sys/class/backlight//ambient_light_zone is defined 2 times: /sys/class/backlight//ambient_light_zone:18 /sys/class/backlight//ambient_light_zone:40 As ABI definitions shouldn't be duplicated. Unfortunately, the ABI is dependent on the specific device features. As such, ambient_light_level range is somewhat different among the supported devices. The ambient_light_zone is even worse: the meanings of each preset are different, and there's no ABI to retrieve the supported types nor their meanins. Unfortunately, it is too late to fix it without causing regressions, as this has been used since Kernel v2.6.35. Rewrite those ABI documentation using the current documentation as a reference, and double-checking at the datasheets: https://www.analog.com/media/en/technical-documentation/data-sheets/ADP8870.pdf https://www.analog.com/media/en/technical-documentation/data-sheets/ADP8860.pdf in order to properly document the differences between those two drivers. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/342195ad5a819d9bcfcebc133c77ab69b4211672.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/testing/sysfs-class-backlight | 60 ++++++++++++++++++++++ .../ABI/testing/sysfs-class-backlight-adp8860 | 21 +------- .../testing/sysfs-class-backlight-driver-adp8870 | 30 ++--------- 3 files changed, 65 insertions(+), 46 deletions(-) diff --git a/Documentation/ABI/testing/sysfs-class-backlight b/Documentation/ABI/testing/sysfs-class-backlight index 3ab175a3f5cb..1fc86401bf95 100644 --- a/Documentation/ABI/testing/sysfs-class-backlight +++ b/Documentation/ABI/testing/sysfs-class-backlight @@ -24,3 +24,63 @@ Description: non-linear The brightness changes non-linearly with each step. Brightness controls should use a linear mapping for a linear perception. + +What: /sys/class/backlight//ambient_light_level +Date: Apr, 2010 +KernelVersion: v2.6.35 +Contact: Michael Hennerich +Description: + (RO) Get conversion value of the light sensor. + + The value is automatically updated every 80 ms when the + light sensor is enabled. + + The value range is device-driver specific: + + For ADP8870: + + It returns integer between 0 (dark) and 8000 (max ambient + brightness). + + For ADP8860: + + It returns a 13-bits integer. + +What: /sys/class/backlight//ambient_light_zone +Date: Apr, 2010 +KernelVersion: v2.6.35 +Contact: Michael Hennerich , + device-drivers-devel@blackfin.uclinux.org + +Description: + (RW) Read or write the specific brightness level at which the + backlight operates. + + The value meaning is device-driver specific: + + For ADP8860: + + == ========================== + 0 Off: Backlight set to 0 mA + 1 Level 1: daylight + 2 Level 2: bright + 3 Level 3: dark + == ========================== + + For ADP8870: + + == ========================== + 0 Off: Backlight set to 0 mA + 1 Level 1: daylight + 2 Level 2: bright + 3 Level 3: office + 4 Level 4: indoor + 5 Level 5: dark + == ========================== + + Writing 0 returns to normal/automatic ambient light level + operation. + + It can be enabled by writing the value stored in + /sys/class/backlight//max_brightness to + /sys/class/backlight//brightness. diff --git a/Documentation/ABI/testing/sysfs-class-backlight-adp8860 b/Documentation/ABI/testing/sysfs-class-backlight-adp8860 index 54d61c788b1b..6610ac73f9ba 100644 --- a/Documentation/ABI/testing/sysfs-class-backlight-adp8860 +++ b/Documentation/ABI/testing/sysfs-class-backlight-adp8860 @@ -6,25 +6,8 @@ adp8860, adp8861 and adp8863 devices: daylight (level 1), office (level 2) and dark (level 3). By default the brightness operates at the daylight brightness level. -What: /sys/class/backlight//ambient_light_level -Date: Apr, 2010 -KernelVersion: v2.6.35 -Contact: Michael Hennerich -Description: - (RO) 13-bit conversion value for the first light sensor—high - byte (Bit 12 to Bit 8). The value is updated every 80 ms (when - the light sensor is enabled). - - -What: /sys/class/backlight//ambient_light_zone -Date: Apr, 2010 -KernelVersion: v2.6.35 -Contact: Michael Hennerich -Description: - (RW) Read or write the specific level at which the backlight - operates. Value "0" enables automatic ambient light sensing, and - values "1", "2" or "3" set the control to daylight, office or - dark respectively. +See also /sys/class/backlight//ambient_light_level and +/sys/class/backlight//ambient_light_zone. What: /sys/class/backlight//l1_daylight_max diff --git a/Documentation/ABI/testing/sysfs-class-backlight-driver-adp8870 b/Documentation/ABI/testing/sysfs-class-backlight-driver-adp8870 index 33e648808117..b08ca912cad4 100644 --- a/Documentation/ABI/testing/sysfs-class-backlight-driver-adp8870 +++ b/Documentation/ABI/testing/sysfs-class-backlight-driver-adp8870 @@ -1,3 +1,6 @@ +See also /sys/class/backlight//ambient_light_level and +/sys/class/backlight//ambient_light_zone. + What: /sys/class/backlight//_max What: /sys/class/backlight//l1_daylight_max What: /sys/class/backlight//l2_bright_max @@ -27,30 +30,3 @@ Description: set to 0. Full off when the backlight is disabled. This file will also show the dim brightness level stored for this . - -What: /sys/class/backlight//ambient_light_level -Date: May 2011 -KernelVersion: 3.0 -Contact: device-drivers-devel@blackfin.uclinux.org -Description: - Get conversion value of the light sensor. - This value is updated every 80 ms (when the light sensor - is enabled). Returns integer between 0 (dark) and - 8000 (max ambient brightness) - -What: /sys/class/backlight//ambient_light_zone -Date: May 2011 -KernelVersion: 3.0 -Contact: device-drivers-devel@blackfin.uclinux.org -Description: - Get/Set current ambient light zone. Reading returns - integer between 1..5 (1 = daylight, 2 = bright, ..., 5 = dark). - Writing a value between 1..5 forces the backlight controller - to enter the corresponding ambient light zone. - Writing 0 returns to normal/automatic ambient light level - operation. The ambient light sensing feature on these devices - is an extension to the API documented in - Documentation/ABI/stable/sysfs-class-backlight. - It can be enabled by writing the value stored in - /sys/class/backlight//max_brightness to - /sys/class/backlight//brightness. -- cgit v1.2.3 From 26a07553041eb728678348a2bb0cd703ccd11083 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:57 +0100 Subject: docs: ABI: sysfs-class-led-trigger-pattern: remove hw_pattern duplication The ABI files are supposed to be unique. Yet, in the specific case of hw_pattern, there are some duplicated entries as warned by scripts/get_abi.pl: Warning: /sys/class/leds//hw_pattern is defined 3 times: Documentation/ABI/testing/sysfs-class-led-trigger-pattern:14 Documentation/ABI/testing/sysfs-class-led-driver-sc27xx:0 Documentation/ABI/testing/sysfs-class-led-driver-el15203000:0 Drop the duplication from the ABI files, moving the specific definitions to files inside Documentation/leds. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/038e57881550550b298e598f8f9b7f20515cbe15.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- .../ABI/testing/sysfs-class-led-driver-el15203000 | 135 -------------------- .../ABI/testing/sysfs-class-led-driver-sc27xx | 22 ---- .../ABI/testing/sysfs-class-led-trigger-pattern | 4 +- Documentation/leds/index.rst | 1 + Documentation/leds/leds-el15203000.rst | 140 +++++++++++++++++++++ Documentation/leds/leds-sc27xx.rst | 27 ++++ 6 files changed, 170 insertions(+), 159 deletions(-) delete mode 100644 Documentation/ABI/testing/sysfs-class-led-driver-sc27xx create mode 100644 Documentation/leds/leds-el15203000.rst create mode 100644 Documentation/leds/leds-sc27xx.rst diff --git a/Documentation/ABI/testing/sysfs-class-led-driver-el15203000 b/Documentation/ABI/testing/sysfs-class-led-driver-el15203000 index da546e86deb5..04f3ffdc5936 100644 --- a/Documentation/ABI/testing/sysfs-class-led-driver-el15203000 +++ b/Documentation/ABI/testing/sysfs-class-led-driver-el15203000 @@ -1,138 +1,3 @@ -What: /sys/class/leds//hw_pattern -Date: September 2019 -KernelVersion: 5.5 -Description: - Specify a hardware pattern for the EL15203000 LED. - The LEDs board supports only predefined patterns by firmware - for specific LEDs. - - Breathing mode for Screen frame light tube:: - - "0 4000 1 4000" - - ^ - | - Max-| --- - | / \ - | / \ - | / \ / - | / \ / - Min-|- --- - | - 0------4------8--> time (sec) - - Cascade mode for Pipe LED:: - - "1 800 2 800 4 800 8 800 16 800" - - ^ - | - 0 On -|----+ +----+ +--- - | | | | | - Off-| +-------------------+ +-------------------+ - | - 1 On -| +----+ +----+ - | | | | | - Off |----+ +-------------------+ +------------------ - | - 2 On -| +----+ +----+ - | | | | | - Off-|---------+ +-------------------+ +------------- - | - 3 On -| +----+ +----+ - | | | | | - Off-|--------------+ +-------------------+ +-------- - | - 4 On -| +----+ +----+ - | | | | | - Off-|-------------------+ +-------------------+ +--- - | - 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) - - Inverted cascade mode for Pipe LED:: - - "30 800 29 800 27 800 23 800 15 800" - - ^ - | - 0 On -| +-------------------+ +-------------------+ - | | | | | - Off-|----+ +----+ +--- - | - 1 On -|----+ +-------------------+ +------------------ - | | | | | - Off | +----+ +----+ - | - 2 On -|---------+ +-------------------+ +------------- - | | | | | - Off-| +----+ +----+ - | - 3 On -|--------------+ +-------------------+ +-------- - | | | | | - Off-| +----+ +----+ - | - 4 On -|-------------------+ +-------------------+ +--- - | | | | | - Off-| +----+ +----+ - | - 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) - - Bounce mode for Pipe LED:: - - "1 800 2 800 4 800 8 800 16 800 16 800 8 800 4 800 2 800 1 800" - - ^ - | - 0 On -|----+ +-------- - | | | - Off-| +---------------------------------------+ - | - 1 On -| +----+ +----+ - | | | | | - Off |----+ +-----------------------------+ +-------- - | - 2 On -| +----+ +----+ - | | | | | - Off-|---------+ +-------------------+ +------------- - | - 3 On -| +----+ +----+ - | | | | | - Off-|--------------+ +---------+ +------------------ - | - 4 On -| +---------+ - | | | - Off-|-------------------+ +----------------------- - | - 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) - - Inverted bounce mode for Pipe LED:: - - "30 800 29 800 27 800 23 800 15 800 15 800 23 800 27 800 29 800 30 800" - - ^ - | - 0 On -| +---------------------------------------+ - | | | - Off-|----+ +-------- - | - 1 On -|----+ +-----------------------------+ +-------- - | | | | | - Off | +----+ +----+ - | - 2 On -|---------+ +-------------------+ +------------- - | | | | | - Off-| +----+ +----+ - | - 3 On -|--------------+ +---------+ +------------------ - | | | | | - Off-| +----+ +----+ - | - 4 On -|-------------------+ +----------------------- - | | | - Off-| +---------+ - | - 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) - What: /sys/class/leds//repeat Date: September 2019 KernelVersion: 5.5 diff --git a/Documentation/ABI/testing/sysfs-class-led-driver-sc27xx b/Documentation/ABI/testing/sysfs-class-led-driver-sc27xx deleted file mode 100644 index 215482379580..000000000000 --- a/Documentation/ABI/testing/sysfs-class-led-driver-sc27xx +++ /dev/null @@ -1,22 +0,0 @@ -What: /sys/class/leds//hw_pattern -Date: September 2018 -KernelVersion: 4.20 -Description: - Specify a hardware pattern for the SC27XX LED. For the SC27XX - LED controller, it only supports 4 stages to make a single - hardware pattern, which is used to configure the rise time, - high time, fall time and low time for the breathing mode. - - For the breathing mode, the SC27XX LED only expects one brightness - for the high stage. To be compatible with the hardware pattern - format, we should set brightness as 0 for rise stage, fall - stage and low stage. - - - Min stage duration: 125 ms - - Max stage duration: 31875 ms - - Since the stage duration step is 125 ms, the duration should be - a multiplier of 125, like 125ms, 250ms, 375ms, 500ms ... 31875ms. - - Thus the format of the hardware pattern values should be: - "0 rise_duration brightness high_duration 0 fall_duration 0 low_duration". diff --git a/Documentation/ABI/testing/sysfs-class-led-trigger-pattern b/Documentation/ABI/testing/sysfs-class-led-trigger-pattern index bd92ef9d6faa..d91a07767adf 100644 --- a/Documentation/ABI/testing/sysfs-class-led-trigger-pattern +++ b/Documentation/ABI/testing/sysfs-class-led-trigger-pattern @@ -23,8 +23,8 @@ Description: Since different LED hardware can have different semantics of hardware patterns, each driver is expected to provide its own - description for the hardware patterns in their ABI documentation - file. + description for the hardware patterns in their documentation + file at Documentation/leds/. What: /sys/class/leds//repeat Date: September 2018 diff --git a/Documentation/leds/index.rst b/Documentation/leds/index.rst index bc70c6aa7138..53e6090454af 100644 --- a/Documentation/leds/index.rst +++ b/Documentation/leds/index.rst @@ -17,6 +17,7 @@ LEDs uleds leds-blinkm + leds-el15203000 leds-lm3556 leds-lp3944 leds-lp5521 diff --git a/Documentation/leds/leds-el15203000.rst b/Documentation/leds/leds-el15203000.rst new file mode 100644 index 000000000000..12c23d79724d --- /dev/null +++ b/Documentation/leds/leds-el15203000.rst @@ -0,0 +1,140 @@ +.. SPDX-License-Identifier: GPL-2.0 + +================================== +Kernel driver for Crane EL15203000 +================================== + +/sys/class/leds//hw_pattern +-------------------------------- + +Specify a hardware pattern for the EL15203000 LED. + +The LEDs board supports only predefined patterns by firmware +for specific LEDs. + +Breathing mode for Screen frame light tube:: + + "0 4000 1 4000" + + ^ + | + Max-| --- + | / \ + | / \ + | / \ / + | / \ / + Min-|- --- + | + 0------4------8--> time (sec) + +Cascade mode for Pipe LED:: + + "1 800 2 800 4 800 8 800 16 800" + + ^ + | + 0 On -|----+ +----+ +--- + | | | | | + Off-| +-------------------+ +-------------------+ + | + 1 On -| +----+ +----+ + | | | | | + Off |----+ +-------------------+ +------------------ + | + 2 On -| +----+ +----+ + | | | | | + Off-|---------+ +-------------------+ +------------- + | + 3 On -| +----+ +----+ + | | | | | + Off-|--------------+ +-------------------+ +-------- + | + 4 On -| +----+ +----+ + | | | | | + Off-|-------------------+ +-------------------+ +--- + | + 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) + +Inverted cascade mode for Pipe LED:: + + "30 800 29 800 27 800 23 800 15 800" + + ^ + | + 0 On -| +-------------------+ +-------------------+ + | | | | | + Off-|----+ +----+ +--- + | + 1 On -|----+ +-------------------+ +------------------ + | | | | | + Off | +----+ +----+ + | + 2 On -|---------+ +-------------------+ +------------- + | | | | | + Off-| +----+ +----+ + | + 3 On -|--------------+ +-------------------+ +-------- + | | | | | + Off-| +----+ +----+ + | + 4 On -|-------------------+ +-------------------+ +--- + | | | | | + Off-| +----+ +----+ + | + 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) + +Bounce mode for Pipe LED:: + + "1 800 2 800 4 800 8 800 16 800 16 800 8 800 4 800 2 800 1 800" + + ^ + | + 0 On -|----+ +-------- + | | | + Off-| +---------------------------------------+ + | + 1 On -| +----+ +----+ + | | | | | + Off |----+ +-----------------------------+ +-------- + | + 2 On -| +----+ +----+ + | | | | | + Off-|---------+ +-------------------+ +------------- + | + 3 On -| +----+ +----+ + | | | | | + Off-|--------------+ +---------+ +------------------ + | + 4 On -| +---------+ + | | | + Off-|-------------------+ +----------------------- + | + 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) + +Inverted bounce mode for Pipe LED:: + + "30 800 29 800 27 800 23 800 15 800 15 800 23 800 27 800 29 800 30 800" + + ^ + | + 0 On -| +---------------------------------------+ + | | | + Off-|----+ +-------- + | + 1 On -|----+ +-----------------------------+ +-------- + | | | | | + Off | +----+ +----+ + | + 2 On -|---------+ +-------------------+ +------------- + | | | | | + Off-| +----+ +----+ + | + 3 On -|--------------+ +---------+ +------------------ + | | | | | + Off-| +----+ +----+ + | + 4 On -|-------------------+ +----------------------- + | | | + Off-| +---------+ + | + 0---0.8--1.6--2.4--3.2---4---4.8--5.6--6.4--7.2---8--> time (sec) diff --git a/Documentation/leds/leds-sc27xx.rst b/Documentation/leds/leds-sc27xx.rst new file mode 100644 index 000000000000..6bdf6ba3c9fd --- /dev/null +++ b/Documentation/leds/leds-sc27xx.rst @@ -0,0 +1,27 @@ +.. SPDX-License-Identifier: GPL-2.0 + +=================================== +Kernel driver for Spreadtrum SC27XX +=================================== + +/sys/class/leds//hw_pattern +-------------------------------- + +Specify a hardware pattern for the SC27XX LED. For the SC27XX +LED controller, it only supports 4 stages to make a single +hardware pattern, which is used to configure the rise time, +high time, fall time and low time for the breathing mode. + +For the breathing mode, the SC27XX LED only expects one brightness +for the high stage. To be compatible with the hardware pattern +format, we should set brightness as 0 for rise stage, fall +stage and low stage. + +- Min stage duration: 125 ms +- Max stage duration: 31875 ms + +Since the stage duration step is 125 ms, the duration should be +a multiplier of 125, like 125ms, 250ms, 375ms, 500ms ... 31875ms. + +Thus the format of the hardware pattern values should be: +"0 rise_duration brightness high_duration 0 fall_duration 0 low_duration". -- cgit v1.2.3 From 2fcce37a3d2f2f3d7fa36aad2b71cbaca135ffd4 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Fri, 30 Oct 2020 08:40:58 +0100 Subject: scripts: get_abi.pl: assume ReST format by default All ABI files are now compatible with ReST format. So, change the script default to assume that the ABI source files are compatible with ReST. Acked-by: Jonathan Corbet Signed-off-by: Mauro Carvalho Chehab Link: https://lore.kernel.org/r/4b0af257010570e0e917d82498e42992bec5e173.1604042072.git.mchehab+huawei@kernel.org Signed-off-by: Greg Kroah-Hartman --- scripts/get_abi.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/get_abi.pl b/scripts/get_abi.pl index d09bb37dd236..2cb592f8eba4 100755 --- a/scripts/get_abi.pl +++ b/scripts/get_abi.pl @@ -18,7 +18,7 @@ my $prefix="Documentation/ABI"; # # If true, assumes that the description is formatted with ReST # -my $description_is_rst = 0; +my $description_is_rst = 1; GetOptions( "debug|d+" => \$debug, -- cgit v1.2.3 From 65ff5cd04551daf2c11c7928e48fc3483391c900 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Fri, 30 Oct 2020 10:42:53 +0800 Subject: blk-mq: mark flush request as IDLE in flush_end_io() Mark flush request as IDLE in its .end_io(), aligning it with how normal requests behave. The flush request stays in in-flight tags if we're not using an IO scheduler, so we need to change its state into IDLE. Otherwise, we will hang in blk_mq_tagset_wait_completed_request() during error recovery because flush the request state is kept as COMPLETED. Reported-by: Yi Zhang Signed-off-by: Ming Lei Tested-by: Yi Zhang Cc: Chao Leng Cc: Sagi Grimberg Signed-off-by: Jens Axboe --- block/blk-flush.c | 1 + 1 file changed, 1 insertion(+) diff --git a/block/blk-flush.c b/block/blk-flush.c index 53abb5c73d99..e32958f0b687 100644 --- a/block/blk-flush.c +++ b/block/blk-flush.c @@ -225,6 +225,7 @@ static void flush_end_io(struct request *flush_rq, blk_status_t error) /* release the tag's ownership to the req cloned from */ spin_lock_irqsave(&fq->mq_flush_lock, flags); + WRITE_ONCE(flush_rq->state, MQ_RQ_IDLE); if (!refcount_dec_and_test(&flush_rq->ref)) { fq->rq_status = error; spin_unlock_irqrestore(&fq->mq_flush_lock, flags); -- cgit v1.2.3 From 22f553842b14a1289c088a79a67fb479d3fa2a4e Mon Sep 17 00:00:00 2001 From: Qais Yousef Date: Tue, 27 Oct 2020 21:51:13 +0000 Subject: KVM: arm64: Handle Asymmetric AArch32 systems On a system without uniform support for AArch32 at EL0, it is possible for the guest to force run AArch32 at EL0 and potentially cause an illegal exception if running on a core without AArch32. Add an extra check so that if we catch the guest doing that, then we prevent it from running again by resetting vcpu->arch.target and return ARM_EXCEPTION_IL. We try to catch this misbehaviour as early as possible and not rely on an illegal exception occuring to signal the problem. Attempting to run a 32bit app in the guest will produce an error from QEMU if the guest exits while running in AArch32 EL0. Tested on Juno by instrumenting the host to fake asym aarch32 and instrumenting KVM to make the asymmetry visible to the guest. [will: Incorporated feedback from Marc] Signed-off-by: Qais Yousef Signed-off-by: Will Deacon Signed-off-by: Marc Zyngier Cc: James Morse Cc: Marc Zyngier Link: https://lore.kernel.org/r/20201021104611.2744565-2-qais.yousef@arm.com Link: https://lore.kernel.org/r/20201027215118.27003-2-will@kernel.org --- arch/arm64/kvm/arm.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c index f56122eedffc..a3b32df1afb0 100644 --- a/arch/arm64/kvm/arm.c +++ b/arch/arm64/kvm/arm.c @@ -808,6 +808,25 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) preempt_enable(); + /* + * The ARMv8 architecture doesn't give the hypervisor + * a mechanism to prevent a guest from dropping to AArch32 EL0 + * if implemented by the CPU. If we spot the guest in such + * state and that we decided it wasn't supposed to do so (like + * with the asymmetric AArch32 case), return to userspace with + * a fatal error. + */ + if (!system_supports_32bit_el0() && vcpu_mode_is_32bit(vcpu)) { + /* + * As we have caught the guest red-handed, decide that + * it isn't fit for purpose anymore by making the vcpu + * invalid. The VMM can try and fix it by issuing a + * KVM_ARM_VCPU_INIT if it really wants to. + */ + vcpu->arch.target = -1; + ret = ARM_EXCEPTION_IL; + } + ret = handle_exit(vcpu, ret); } -- cgit v1.2.3 From d48e3850030623e1c20785bceaaf78f916d0b1a3 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 26 Oct 2020 16:22:56 +0100 Subject: locking/lockdep: Remove more raw_cpu_read() usage I initially thought raw_cpu_read() was OK, since if it is !0 we have IRQs disabled and can't get migrated, so if we get migrated both CPUs must have 0 and it doesn't matter which 0 we read. And while that is true; it isn't the whole store, on pretty much all architectures (except x86) this can result in computing the address for one CPU, getting migrated, the old CPU continuing execution with another task (possibly setting recursion) and then the new CPU reading the value of the old CPU, which is no longer 0. Similer to: baffd723e44d ("lockdep: Revert "lockdep: Use raw_cpu_*() for per-cpu variables"") Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20201026152256.GB2651@hirez.programming.kicks-ass.net --- kernel/locking/lockdep.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index fc206aefa970..11028497d4df 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -84,7 +84,7 @@ static inline bool lockdep_enabled(void) if (!debug_locks) return false; - if (raw_cpu_read(lockdep_recursion)) + if (this_cpu_read(lockdep_recursion)) return false; if (current->lockdep_recursion) -- cgit v1.2.3 From 1a39340865ce505a029b37aeb47a3e4c8db5f6c6 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 27 Oct 2020 13:48:34 +0100 Subject: lockdep: Fix nr_unused_locks accounting Chris reported that commit 24d5a3bffef1 ("lockdep: Fix usage_traceoverflow") breaks the nr_unused_locks validation code triggered by /proc/lockdep_stats. By fully splitting LOCK_USED and LOCK_USED_READ it becomes a bad indicator for accounting nr_unused_locks; simplyfy by using any first bit. Fixes: 24d5a3bffef1 ("lockdep: Fix usage_traceoverflow") Reported-by: Chris Wilson Signed-off-by: Peter Zijlstra (Intel) Tested-by: Chris Wilson Link: https://lkml.kernel.org/r/20201027124834.GL2628@hirez.programming.kicks-ass.net --- kernel/locking/lockdep.c | 14 ++++---------- 1 file changed, 4 insertions(+), 10 deletions(-) diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index 11028497d4df..b71ad8d9f1c9 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -4396,6 +4396,9 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this, if (unlikely(hlock_class(this)->usage_mask & new_mask)) goto unlock; + if (!hlock_class(this)->usage_mask) + debug_atomic_dec(nr_unused_locks); + hlock_class(this)->usage_mask |= new_mask; if (new_bit < LOCK_TRACE_STATES) { @@ -4403,19 +4406,10 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this, return 0; } - switch (new_bit) { - case 0 ... LOCK_USED-1: + if (new_bit < LOCK_USED) { ret = mark_lock_irq(curr, this, new_bit); if (!ret) return 0; - break; - - case LOCK_USED: - debug_atomic_dec(nr_unused_locks); - break; - - default: - break; } unlock: -- cgit v1.2.3 From 8a967d655e406c8a63744a60b221071fad9a736b Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Fri, 30 Oct 2020 13:39:55 -0400 Subject: KVM: x86: replace static const variables with macros Even though the compiler is able to replace static const variables with their value, it will warn about them being unused when Linux is built with W=1. Use good old macros instead, this is not C++. Signed-off-by: Paolo Bonzini --- arch/x86/kvm/mmu/mmu.c | 10 +++++----- arch/x86/kvm/mmu/spte.c | 16 ++++++++-------- arch/x86/kvm/mmu/spte.h | 16 ++++++++-------- 3 files changed, 21 insertions(+), 21 deletions(-) diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index 17587f496ec7..1f96adff8dc4 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -225,7 +225,7 @@ static gfn_t get_mmio_spte_gfn(u64 spte) { u64 gpa = spte & shadow_nonpresent_or_rsvd_lower_gfn_mask; - gpa |= (spte >> shadow_nonpresent_or_rsvd_mask_len) + gpa |= (spte >> SHADOW_NONPRESENT_OR_RSVD_MASK_LEN) & shadow_nonpresent_or_rsvd_mask; return gpa >> PAGE_SHIFT; @@ -591,15 +591,15 @@ static u64 mmu_spte_get_lockless(u64 *sptep) static u64 restore_acc_track_spte(u64 spte) { u64 new_spte = spte; - u64 saved_bits = (spte >> shadow_acc_track_saved_bits_shift) - & shadow_acc_track_saved_bits_mask; + u64 saved_bits = (spte >> SHADOW_ACC_TRACK_SAVED_BITS_SHIFT) + & SHADOW_ACC_TRACK_SAVED_BITS_MASK; WARN_ON_ONCE(spte_ad_enabled(spte)); WARN_ON_ONCE(!is_access_track_spte(spte)); new_spte &= ~shadow_acc_track_mask; - new_spte &= ~(shadow_acc_track_saved_bits_mask << - shadow_acc_track_saved_bits_shift); + new_spte &= ~(SHADOW_ACC_TRACK_SAVED_BITS_MASK << + SHADOW_ACC_TRACK_SAVED_BITS_SHIFT); new_spte |= saved_bits; return new_spte; diff --git a/arch/x86/kvm/mmu/spte.c b/arch/x86/kvm/mmu/spte.c index d9c5665a55e9..fcac2cac78fe 100644 --- a/arch/x86/kvm/mmu/spte.c +++ b/arch/x86/kvm/mmu/spte.c @@ -55,7 +55,7 @@ u64 make_mmio_spte(struct kvm_vcpu *vcpu, u64 gfn, unsigned int access) mask |= shadow_mmio_value | access; mask |= gpa | shadow_nonpresent_or_rsvd_mask; mask |= (gpa & shadow_nonpresent_or_rsvd_mask) - << shadow_nonpresent_or_rsvd_mask_len; + << SHADOW_NONPRESENT_OR_RSVD_MASK_LEN; return mask; } @@ -231,12 +231,12 @@ u64 mark_spte_for_access_track(u64 spte) !spte_can_locklessly_be_made_writable(spte), "kvm: Writable SPTE is not locklessly dirty-trackable\n"); - WARN_ONCE(spte & (shadow_acc_track_saved_bits_mask << - shadow_acc_track_saved_bits_shift), + WARN_ONCE(spte & (SHADOW_ACC_TRACK_SAVED_BITS_MASK << + SHADOW_ACC_TRACK_SAVED_BITS_SHIFT), "kvm: Access Tracking saved bit locations are not zero\n"); - spte |= (spte & shadow_acc_track_saved_bits_mask) << - shadow_acc_track_saved_bits_shift; + spte |= (spte & SHADOW_ACC_TRACK_SAVED_BITS_MASK) << + SHADOW_ACC_TRACK_SAVED_BITS_SHIFT; spte &= ~shadow_acc_track_mask; return spte; @@ -245,7 +245,7 @@ u64 mark_spte_for_access_track(u64 spte) void kvm_mmu_set_mmio_spte_mask(u64 mmio_value, u64 access_mask) { BUG_ON((u64)(unsigned)access_mask != access_mask); - WARN_ON(mmio_value & (shadow_nonpresent_or_rsvd_mask << shadow_nonpresent_or_rsvd_mask_len)); + WARN_ON(mmio_value & (shadow_nonpresent_or_rsvd_mask << SHADOW_NONPRESENT_OR_RSVD_MASK_LEN)); WARN_ON(mmio_value & shadow_nonpresent_or_rsvd_lower_gfn_mask); shadow_mmio_value = mmio_value | SPTE_MMIO_MASK; shadow_mmio_access_mask = access_mask; @@ -306,9 +306,9 @@ void kvm_mmu_reset_all_pte_masks(void) low_phys_bits = boot_cpu_data.x86_phys_bits; if (boot_cpu_has_bug(X86_BUG_L1TF) && !WARN_ON_ONCE(boot_cpu_data.x86_cache_bits >= - 52 - shadow_nonpresent_or_rsvd_mask_len)) { + 52 - SHADOW_NONPRESENT_OR_RSVD_MASK_LEN)) { low_phys_bits = boot_cpu_data.x86_cache_bits - - shadow_nonpresent_or_rsvd_mask_len; + - SHADOW_NONPRESENT_OR_RSVD_MASK_LEN; shadow_nonpresent_or_rsvd_mask = rsvd_bits(low_phys_bits, boot_cpu_data.x86_cache_bits - 1); } diff --git a/arch/x86/kvm/mmu/spte.h b/arch/x86/kvm/mmu/spte.h index 4ecf40e0b8fe..5c75a451c000 100644 --- a/arch/x86/kvm/mmu/spte.h +++ b/arch/x86/kvm/mmu/spte.h @@ -104,20 +104,20 @@ extern u64 __read_mostly shadow_acc_track_mask; */ extern u64 __read_mostly shadow_nonpresent_or_rsvd_mask; +/* + * The number of high-order 1 bits to use in the mask above. + */ +#define SHADOW_NONPRESENT_OR_RSVD_MASK_LEN 5 + /* * The mask/shift to use for saving the original R/X bits when marking the PTE * as not-present for access tracking purposes. We do not save the W bit as the * PTEs being access tracked also need to be dirty tracked, so the W bit will be * restored only when a write is attempted to the page. */ -static const u64 shadow_acc_track_saved_bits_mask = PT64_EPT_READABLE_MASK | - PT64_EPT_EXECUTABLE_MASK; -static const u64 shadow_acc_track_saved_bits_shift = PT64_SECOND_AVAIL_BITS_SHIFT; - -/* - * The number of high-order 1 bits to use in the mask above. - */ -static const u64 shadow_nonpresent_or_rsvd_mask_len = 5; +#define SHADOW_ACC_TRACK_SAVED_BITS_MASK (PT64_EPT_READABLE_MASK | \ + PT64_EPT_EXECUTABLE_MASK) +#define SHADOW_ACC_TRACK_SAVED_BITS_SHIFT PT64_SECOND_AVAIL_BITS_SHIFT /* * In some cases, we need to preserve the GFN of a non-present or reserved -- cgit v1.2.3 From d383b3146d805a743658225c8973f5d38c6fedf4 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 30 Oct 2020 16:14:14 +0100 Subject: KVM: x86: Fix NULL dereference at kvm_msr_ignored_check() The newly introduced kvm_msr_ignored_check() tries to print error or debug messages via vcpu_*() macros, but those may cause Oops when NULL vcpu is passed for KVM_GET_MSRS ioctl. Fix it by replacing the print calls with kvm_*() macros. (Note that this will leave vcpu argument completely unused in the function, but I didn't touch it to make the fix as small as possible. A clean up may be applied later.) Fixes: 12bc2132b15e ("KVM: X86: Do the same ignore_msrs check for feature msrs") BugLink: https://bugzilla.suse.com/show_bug.cgi?id=1178280 Cc: Signed-off-by: Takashi Iwai Message-Id: <20201030151414.20165-1-tiwai@suse.de> Signed-off-by: Paolo Bonzini --- arch/x86/kvm/x86.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 397f599b20e5..f5ede41bf9e6 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -265,13 +265,13 @@ static int kvm_msr_ignored_check(struct kvm_vcpu *vcpu, u32 msr, if (ignore_msrs) { if (report_ignored_msrs) - vcpu_unimpl(vcpu, "ignored %s: 0x%x data 0x%llx\n", - op, msr, data); + kvm_pr_unimpl("ignored %s: 0x%x data 0x%llx\n", + op, msr, data); /* Mask the error */ return 0; } else { - vcpu_debug_ratelimited(vcpu, "unhandled %s: 0x%x data 0x%llx\n", - op, msr, data); + kvm_debug_ratelimited("unhandled %s: 0x%x data 0x%llx\n", + op, msr, data); return -ENOENT; } } -- cgit v1.2.3 From 4acbf5545d5acfeeac6d84e31cb2203ba19223ef Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Tue, 27 Oct 2020 00:54:08 -0500 Subject: Bluetooth: btintel: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- drivers/bluetooth/btintel.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/bluetooth/btintel.h b/drivers/bluetooth/btintel.h index 09346ae308eb..78cc64b42b30 100644 --- a/drivers/bluetooth/btintel.h +++ b/drivers/bluetooth/btintel.h @@ -47,7 +47,7 @@ enum { struct intel_tlv { u8 type; u8 len; - u8 val[0]; + u8 val[]; } __packed; struct intel_version_tlv { -- cgit v1.2.3 From 691f4077d5604af4067ebbea4a27f05e760557b7 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Tue, 27 Oct 2020 16:30:45 -0500 Subject: gve: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. Refactor the code according to the use of a flexible-array member in struct gve_stats_report, instead of a zero-length array, and use the struct_size() helper to calculate the size for the resource allocation. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- drivers/net/ethernet/google/gve/gve_adminq.h | 2 +- drivers/net/ethernet/google/gve/gve_main.c | 5 ++--- 2 files changed, 3 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/google/gve/gve_adminq.h b/drivers/net/ethernet/google/gve/gve_adminq.h index 281de8326bc5..015796a20118 100644 --- a/drivers/net/ethernet/google/gve/gve_adminq.h +++ b/drivers/net/ethernet/google/gve/gve_adminq.h @@ -198,7 +198,7 @@ static_assert(sizeof(struct stats) == 16); struct gve_stats_report { __be64 written_count; - struct stats stats[0]; + struct stats stats[]; }; static_assert(sizeof(struct gve_stats_report) == 8); diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c index 48a433154ce0..02e7d74779f4 100644 --- a/drivers/net/ethernet/google/gve/gve_main.c +++ b/drivers/net/ethernet/google/gve/gve_main.c @@ -116,9 +116,8 @@ static int gve_alloc_stats_report(struct gve_priv *priv) priv->tx_cfg.num_queues; rx_stats_num = (GVE_RX_STATS_REPORT_NUM + NIC_RX_STATS_REPORT_NUM) * priv->rx_cfg.num_queues; - priv->stats_report_len = sizeof(struct gve_stats_report) + - (tx_stats_num + rx_stats_num) * - sizeof(struct stats); + priv->stats_report_len = struct_size(priv->stats_report, stats, + tx_stats_num + rx_stats_num); priv->stats_report = dma_alloc_coherent(&priv->pdev->dev, priv->stats_report_len, &priv->stats_report_bus, GFP_KERNEL); -- cgit v1.2.3 From b21de80ac156044c731889a686135e6d0b7c2bcc Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Tue, 27 Oct 2020 01:15:49 -0500 Subject: mei: hw: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- drivers/misc/mei/hw.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/misc/mei/hw.h b/drivers/misc/mei/hw.h index 8bac86c4d86b..df2fb9520dd8 100644 --- a/drivers/misc/mei/hw.h +++ b/drivers/misc/mei/hw.h @@ -224,7 +224,7 @@ struct mei_ext_hdr { u8 type; u8 length; u8 ext_payload[2]; - u8 hdr[0]; + u8 hdr[]; }; /** @@ -238,7 +238,7 @@ struct mei_ext_meta_hdr { u8 count; u8 size; u8 reserved[2]; - struct mei_ext_hdr hdrs[0]; + struct mei_ext_hdr hdrs[]; }; /* @@ -308,7 +308,7 @@ struct mei_msg_hdr { u32 dma_ring:1; u32 internal:1; u32 msg_complete:1; - u32 extension[0]; + u32 extension[]; } __packed; /* The length is up to 9 bits */ -- cgit v1.2.3 From 290562075d4d9e85b7ff4104f9a634ffc3cccb69 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Tue, 27 Oct 2020 15:28:40 -0500 Subject: net/mlx5: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- include/linux/mlx5/mlx5_ifc.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 651591a2965d..a092346c7b2d 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -5823,7 +5823,7 @@ struct mlx5_ifc_alloc_modify_header_context_in_bits { u8 reserved_at_68[0x10]; u8 num_of_actions[0x8]; - union mlx5_ifc_set_add_copy_action_in_auto_bits actions[0]; + union mlx5_ifc_set_add_copy_action_in_auto_bits actions[]; }; struct mlx5_ifc_dealloc_modify_header_context_out_bits { @@ -9761,7 +9761,7 @@ struct mlx5_ifc_mcda_reg_bits { u8 reserved_at_60[0x20]; - u8 data[0][0x20]; + u8 data[][0x20]; }; enum { -- cgit v1.2.3 From 7206d58a3a538c80b36305d1904de313cf47ef4c Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Tue, 27 Oct 2020 01:33:56 -0500 Subject: net/smc: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- net/smc/smc_clc.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/smc/smc_clc.h b/net/smc/smc_clc.h index b3f46ab79e47..c579d1d5995a 100644 --- a/net/smc/smc_clc.h +++ b/net/smc/smc_clc.h @@ -124,7 +124,7 @@ struct smc_clc_v2_extension { struct smc_clnt_opts_area_hdr hdr; u8 roce[16]; /* RoCEv2 GID */ u8 reserved[16]; - u8 user_eids[0][SMC_MAX_EID_LEN]; + u8 user_eids[][SMC_MAX_EID_LEN]; }; struct smc_clc_msg_proposal_prefix { /* prefix part of clc proposal message*/ @@ -143,7 +143,7 @@ struct smc_clc_msg_smcd { /* SMC-D GID information */ struct smc_clc_smcd_v2_extension { u8 system_eid[SMC_MAX_EID_LEN]; u8 reserved[16]; - struct smc_clc_smcd_gid_chid gidchid[0]; + struct smc_clc_smcd_gid_chid gidchid[]; }; struct smc_clc_msg_proposal { /* clc proposal message sent by Linux */ -- cgit v1.2.3 From a38283da05d321fa1fce38ea3cf41c9f1dbd1f21 Mon Sep 17 00:00:00 2001 From: "Gustavo A. R. Silva" Date: Tue, 27 Oct 2020 13:50:19 -0500 Subject: printk: ringbuffer: Replace zero-length array with flexible-array member MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a regular need in the kernel to provide a way to declare having a dynamically sized set of trailing elements in a structure. Kernel code should always use “flexible array members”[1] for these cases. The older style of one-element or zero-length arrays should no longer be used[2]. [1] https://en.wikipedia.org/wiki/Flexible_array_member [2] https://www.kernel.org/doc/html/v5.9/process/deprecated.html#zero-length-and-one-element-arrays Signed-off-by: Gustavo A. R. Silva --- kernel/printk/printk_ringbuffer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/printk/printk_ringbuffer.c b/kernel/printk/printk_ringbuffer.c index 24a960a89aa8..6b1525685277 100644 --- a/kernel/printk/printk_ringbuffer.c +++ b/kernel/printk/printk_ringbuffer.c @@ -345,7 +345,7 @@ DESC_ID((id) - DESCS_COUNT(desc_ring)) */ struct prb_data_block { unsigned long id; - char data[0]; + char data[]; }; /* -- cgit v1.2.3 From 97a71a5455997d72adfd74e9d575d782d7c2ab9a Mon Sep 17 00:00:00 2001 From: Jim Mattson Date: Mon, 26 Oct 2020 11:09:22 -0700 Subject: KVM: selftests: test behavior of unmapped L2 APIC-access address Add a regression test for commit 671ddc700fd0 ("KVM: nVMX: Don't leak L1 MMIO regions to L2"). First, check to see that an L2 guest can be launched with a valid APIC-access address that is backed by a page of L1 physical memory. Next, set the APIC-access address to a (valid) L1 physical address that is not backed by memory. KVM can't handle this situation, so resuming L2 should result in a KVM exit for internal error (emulation). Signed-off-by: Jim Mattson Reviewed-by: Ricardo Koller Reviewed-by: Peter Shier Message-Id: <20201026180922.3120555-1-jmattson@google.com> Signed-off-by: Paolo Bonzini --- tools/testing/selftests/kvm/.gitignore | 1 + tools/testing/selftests/kvm/Makefile | 1 + tools/testing/selftests/kvm/include/x86_64/vmx.h | 6 + tools/testing/selftests/kvm/lib/x86_64/vmx.c | 9 ++ .../selftests/kvm/x86_64/vmx_apic_access_test.c | 142 +++++++++++++++++++++ 5 files changed, 159 insertions(+) create mode 100644 tools/testing/selftests/kvm/x86_64/vmx_apic_access_test.c diff --git a/tools/testing/selftests/kvm/.gitignore b/tools/testing/selftests/kvm/.gitignore index 307ceaadbbb9..d2c2d6205008 100644 --- a/tools/testing/selftests/kvm/.gitignore +++ b/tools/testing/selftests/kvm/.gitignore @@ -15,6 +15,7 @@ /x86_64/vmx_preemption_timer_test /x86_64/svm_vmcall_test /x86_64/sync_regs_test +/x86_64/vmx_apic_access_test /x86_64/vmx_close_while_nested_test /x86_64/vmx_dirty_log_test /x86_64/vmx_set_nested_state_test diff --git a/tools/testing/selftests/kvm/Makefile b/tools/testing/selftests/kvm/Makefile index 7ebe71fbca53..30afbad36cd5 100644 --- a/tools/testing/selftests/kvm/Makefile +++ b/tools/testing/selftests/kvm/Makefile @@ -49,6 +49,7 @@ TEST_GEN_PROGS_x86_64 += x86_64/state_test TEST_GEN_PROGS_x86_64 += x86_64/vmx_preemption_timer_test TEST_GEN_PROGS_x86_64 += x86_64/svm_vmcall_test TEST_GEN_PROGS_x86_64 += x86_64/sync_regs_test +TEST_GEN_PROGS_x86_64 += x86_64/vmx_apic_access_test TEST_GEN_PROGS_x86_64 += x86_64/vmx_close_while_nested_test TEST_GEN_PROGS_x86_64 += x86_64/vmx_dirty_log_test TEST_GEN_PROGS_x86_64 += x86_64/vmx_set_nested_state_test diff --git a/tools/testing/selftests/kvm/include/x86_64/vmx.h b/tools/testing/selftests/kvm/include/x86_64/vmx.h index 54d624dd6c10..e78d7e26ba61 100644 --- a/tools/testing/selftests/kvm/include/x86_64/vmx.h +++ b/tools/testing/selftests/kvm/include/x86_64/vmx.h @@ -573,6 +573,10 @@ struct vmx_pages { void *eptp_hva; uint64_t eptp_gpa; void *eptp; + + void *apic_access_hva; + uint64_t apic_access_gpa; + void *apic_access; }; union vmx_basic { @@ -615,5 +619,7 @@ void nested_map_memslot(struct vmx_pages *vmx, struct kvm_vm *vm, uint32_t memslot, uint32_t eptp_memslot); void prepare_eptp(struct vmx_pages *vmx, struct kvm_vm *vm, uint32_t eptp_memslot); +void prepare_virtualize_apic_accesses(struct vmx_pages *vmx, struct kvm_vm *vm, + uint32_t eptp_memslot); #endif /* SELFTEST_KVM_VMX_H */ diff --git a/tools/testing/selftests/kvm/lib/x86_64/vmx.c b/tools/testing/selftests/kvm/lib/x86_64/vmx.c index f1e00d43eea2..2448b30e8efa 100644 --- a/tools/testing/selftests/kvm/lib/x86_64/vmx.c +++ b/tools/testing/selftests/kvm/lib/x86_64/vmx.c @@ -542,3 +542,12 @@ void prepare_eptp(struct vmx_pages *vmx, struct kvm_vm *vm, vmx->eptp_hva = addr_gva2hva(vm, (uintptr_t)vmx->eptp); vmx->eptp_gpa = addr_gva2gpa(vm, (uintptr_t)vmx->eptp); } + +void prepare_virtualize_apic_accesses(struct vmx_pages *vmx, struct kvm_vm *vm, + uint32_t eptp_memslot) +{ + vmx->apic_access = (void *)vm_vaddr_alloc(vm, getpagesize(), + 0x10000, 0, 0); + vmx->apic_access_hva = addr_gva2hva(vm, (uintptr_t)vmx->apic_access); + vmx->apic_access_gpa = addr_gva2gpa(vm, (uintptr_t)vmx->apic_access); +} diff --git a/tools/testing/selftests/kvm/x86_64/vmx_apic_access_test.c b/tools/testing/selftests/kvm/x86_64/vmx_apic_access_test.c new file mode 100644 index 000000000000..1f65342d6cb7 --- /dev/null +++ b/tools/testing/selftests/kvm/x86_64/vmx_apic_access_test.c @@ -0,0 +1,142 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * vmx_apic_access_test + * + * Copyright (C) 2020, Google LLC. + * + * This work is licensed under the terms of the GNU GPL, version 2. + * + * The first subtest simply checks to see that an L2 guest can be + * launched with a valid APIC-access address that is backed by a + * page of L1 physical memory. + * + * The second subtest sets the APIC-access address to a (valid) L1 + * physical address that is not backed by memory. KVM can't handle + * this situation, so resuming L2 should result in a KVM exit for + * internal error (emulation). This is not an architectural + * requirement. It is just a shortcoming of KVM. The internal error + * is unfortunate, but it's better than what used to happen! + */ + +#include "test_util.h" +#include "kvm_util.h" +#include "processor.h" +#include "vmx.h" + +#include +#include + +#include "kselftest.h" + +#define VCPU_ID 0 + +/* The virtual machine object. */ +static struct kvm_vm *vm; + +static void l2_guest_code(void) +{ + /* Exit to L1 */ + __asm__ __volatile__("vmcall"); +} + +static void l1_guest_code(struct vmx_pages *vmx_pages, unsigned long high_gpa) +{ +#define L2_GUEST_STACK_SIZE 64 + unsigned long l2_guest_stack[L2_GUEST_STACK_SIZE]; + uint32_t control; + + GUEST_ASSERT(prepare_for_vmx_operation(vmx_pages)); + GUEST_ASSERT(load_vmcs(vmx_pages)); + + /* Prepare the VMCS for L2 execution. */ + prepare_vmcs(vmx_pages, l2_guest_code, + &l2_guest_stack[L2_GUEST_STACK_SIZE]); + control = vmreadz(CPU_BASED_VM_EXEC_CONTROL); + control |= CPU_BASED_ACTIVATE_SECONDARY_CONTROLS; + vmwrite(CPU_BASED_VM_EXEC_CONTROL, control); + control = vmreadz(SECONDARY_VM_EXEC_CONTROL); + control |= SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES; + vmwrite(SECONDARY_VM_EXEC_CONTROL, control); + vmwrite(APIC_ACCESS_ADDR, vmx_pages->apic_access_gpa); + + /* Try to launch L2 with the memory-backed APIC-access address. */ + GUEST_SYNC(vmreadz(APIC_ACCESS_ADDR)); + GUEST_ASSERT(!vmlaunch()); + GUEST_ASSERT(vmreadz(VM_EXIT_REASON) == EXIT_REASON_VMCALL); + + vmwrite(APIC_ACCESS_ADDR, high_gpa); + + /* Try to resume L2 with the unbacked APIC-access address. */ + GUEST_SYNC(vmreadz(APIC_ACCESS_ADDR)); + GUEST_ASSERT(!vmresume()); + GUEST_ASSERT(vmreadz(VM_EXIT_REASON) == EXIT_REASON_VMCALL); + + GUEST_DONE(); +} + +int main(int argc, char *argv[]) +{ + unsigned long apic_access_addr = ~0ul; + unsigned int paddr_width; + unsigned int vaddr_width; + vm_vaddr_t vmx_pages_gva; + unsigned long high_gpa; + struct vmx_pages *vmx; + bool done = false; + + nested_vmx_check_supported(); + + vm = vm_create_default(VCPU_ID, 0, (void *) l1_guest_code); + vcpu_set_cpuid(vm, VCPU_ID, kvm_get_supported_cpuid()); + + kvm_get_cpu_address_width(&paddr_width, &vaddr_width); + high_gpa = (1ul << paddr_width) - getpagesize(); + if ((unsigned long)DEFAULT_GUEST_PHY_PAGES * getpagesize() > high_gpa) { + print_skip("No unbacked physical page available"); + exit(KSFT_SKIP); + } + + vmx = vcpu_alloc_vmx(vm, &vmx_pages_gva); + prepare_virtualize_apic_accesses(vmx, vm, 0); + vcpu_args_set(vm, VCPU_ID, 2, vmx_pages_gva, high_gpa); + + while (!done) { + volatile struct kvm_run *run = vcpu_state(vm, VCPU_ID); + struct ucall uc; + + vcpu_run(vm, VCPU_ID); + if (apic_access_addr == high_gpa) { + TEST_ASSERT(run->exit_reason == + KVM_EXIT_INTERNAL_ERROR, + "Got exit reason other than KVM_EXIT_INTERNAL_ERROR: %u (%s)\n", + run->exit_reason, + exit_reason_str(run->exit_reason)); + TEST_ASSERT(run->internal.suberror == + KVM_INTERNAL_ERROR_EMULATION, + "Got internal suberror other than KVM_INTERNAL_ERROR_EMULATION: %u\n", + run->internal.suberror); + break; + } + TEST_ASSERT(run->exit_reason == KVM_EXIT_IO, + "Got exit_reason other than KVM_EXIT_IO: %u (%s)\n", + run->exit_reason, + exit_reason_str(run->exit_reason)); + + switch (get_ucall(vm, VCPU_ID, &uc)) { + case UCALL_ABORT: + TEST_FAIL("%s at %s:%ld", (const char *)uc.args[0], + __FILE__, uc.args[1]); + /* NOT REACHED */ + case UCALL_SYNC: + apic_access_addr = uc.args[1]; + break; + case UCALL_DONE: + done = true; + break; + default: + TEST_ASSERT(false, "Unknown ucall %lu", uc.cmd); + } + } + kvm_vm_free(vm); + return 0; +} -- cgit v1.2.3 From 064eedf2c50f692088e1418c553084bf9c1432f8 Mon Sep 17 00:00:00 2001 From: Vitaly Kuznetsov Date: Wed, 14 Oct 2020 16:33:46 +0200 Subject: KVM: VMX: eVMCS: make evmcs_sanitize_exec_ctrls() work again It was noticed that evmcs_sanitize_exec_ctrls() is not being executed nowadays despite the code checking 'enable_evmcs' static key looking correct. Turns out, static key magic doesn't work in '__init' section (and it is unclear when things changed) but setup_vmcs_config() is called only once per CPU so we don't really need it to. Switch to checking 'enlightened_vmcs' instead, it is supposed to be in sync with 'enable_evmcs'. Opportunistically make evmcs_sanitize_exec_ctrls '__init' and drop unneeded extra newline from it. Reported-by: Yang Weijiang Signed-off-by: Vitaly Kuznetsov Message-Id: <20201014143346.2430936-1-vkuznets@redhat.com> Signed-off-by: Paolo Bonzini --- arch/x86/kvm/vmx/evmcs.c | 3 +-- arch/x86/kvm/vmx/evmcs.h | 3 +-- arch/x86/kvm/vmx/vmx.c | 4 +++- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/x86/kvm/vmx/evmcs.c b/arch/x86/kvm/vmx/evmcs.c index e5325bd0f304..f3199bb02f22 100644 --- a/arch/x86/kvm/vmx/evmcs.c +++ b/arch/x86/kvm/vmx/evmcs.c @@ -297,14 +297,13 @@ const struct evmcs_field vmcs_field_to_evmcs_1[] = { }; const unsigned int nr_evmcs_1_fields = ARRAY_SIZE(vmcs_field_to_evmcs_1); -void evmcs_sanitize_exec_ctrls(struct vmcs_config *vmcs_conf) +__init void evmcs_sanitize_exec_ctrls(struct vmcs_config *vmcs_conf) { vmcs_conf->pin_based_exec_ctrl &= ~EVMCS1_UNSUPPORTED_PINCTRL; vmcs_conf->cpu_based_2nd_exec_ctrl &= ~EVMCS1_UNSUPPORTED_2NDEXEC; vmcs_conf->vmexit_ctrl &= ~EVMCS1_UNSUPPORTED_VMEXIT_CTRL; vmcs_conf->vmentry_ctrl &= ~EVMCS1_UNSUPPORTED_VMENTRY_CTRL; - } #endif diff --git a/arch/x86/kvm/vmx/evmcs.h b/arch/x86/kvm/vmx/evmcs.h index e5f7a7ebf27d..bd41d9462355 100644 --- a/arch/x86/kvm/vmx/evmcs.h +++ b/arch/x86/kvm/vmx/evmcs.h @@ -185,7 +185,7 @@ static inline void evmcs_load(u64 phys_addr) vp_ap->enlighten_vmentry = 1; } -void evmcs_sanitize_exec_ctrls(struct vmcs_config *vmcs_conf); +__init void evmcs_sanitize_exec_ctrls(struct vmcs_config *vmcs_conf); #else /* !IS_ENABLED(CONFIG_HYPERV) */ static inline void evmcs_write64(unsigned long field, u64 value) {} static inline void evmcs_write32(unsigned long field, u32 value) {} @@ -194,7 +194,6 @@ static inline u64 evmcs_read64(unsigned long field) { return 0; } static inline u32 evmcs_read32(unsigned long field) { return 0; } static inline u16 evmcs_read16(unsigned long field) { return 0; } static inline void evmcs_load(u64 phys_addr) {} -static inline void evmcs_sanitize_exec_ctrls(struct vmcs_config *vmcs_conf) {} static inline void evmcs_touch_msr_bitmap(void) {} #endif /* IS_ENABLED(CONFIG_HYPERV) */ diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c index d14c94d0aff1..61d1cecb6f12 100644 --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -2560,8 +2560,10 @@ static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf, vmcs_conf->vmexit_ctrl = _vmexit_control; vmcs_conf->vmentry_ctrl = _vmentry_control; - if (static_branch_unlikely(&enable_evmcs)) +#if IS_ENABLED(CONFIG_HYPERV) + if (enlightened_vmcs) evmcs_sanitize_exec_ctrls(vmcs_conf); +#endif return 0; } -- cgit v1.2.3 From 0f55b67e5ad8dcdfb9ae2aaed75dd9fbb2bd92b2 Mon Sep 17 00:00:00 2001 From: Andrew Jones Date: Thu, 29 Oct 2020 21:17:00 +0100 Subject: KVM: selftests: Don't require THP to run tests Unless we want to test with THP, then we shouldn't require it to be configured by the host kernel. Unfortunately, even advising with MADV_NOHUGEPAGE does require it, so check for THP first in order to avoid madvise failing with EINVAL. Signed-off-by: Andrew Jones Message-Id: <20201029201703.102716-2-drjones@redhat.com> Signed-off-by: Paolo Bonzini --- tools/testing/selftests/kvm/lib/kvm_util.c | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-) diff --git a/tools/testing/selftests/kvm/lib/kvm_util.c b/tools/testing/selftests/kvm/lib/kvm_util.c index 74776ee228f2..3327cebc1095 100644 --- a/tools/testing/selftests/kvm/lib/kvm_util.c +++ b/tools/testing/selftests/kvm/lib/kvm_util.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #define KVM_UTIL_PGS_PER_HUGEPG 512 @@ -664,13 +665,21 @@ void vm_userspace_mem_region_add(struct kvm_vm *vm, /* As needed perform madvise */ if (src_type == VM_MEM_SRC_ANONYMOUS || src_type == VM_MEM_SRC_ANONYMOUS_THP) { - ret = madvise(region->host_mem, npages * vm->page_size, - src_type == VM_MEM_SRC_ANONYMOUS ? MADV_NOHUGEPAGE : MADV_HUGEPAGE); - TEST_ASSERT(ret == 0, "madvise failed,\n" - " addr: %p\n" - " length: 0x%lx\n" - " src_type: %x", - region->host_mem, npages * vm->page_size, src_type); + struct stat statbuf; + + ret = stat("/sys/kernel/mm/transparent_hugepage", &statbuf); + TEST_ASSERT(ret == 0 || (ret == -1 && errno == ENOENT), + "stat /sys/kernel/mm/transparent_hugepage"); + + TEST_ASSERT(ret == 0 || src_type != VM_MEM_SRC_ANONYMOUS_THP, + "VM_MEM_SRC_ANONYMOUS_THP requires THP to be configured in the host kernel"); + + if (ret == 0) { + ret = madvise(region->host_mem, npages * vm->page_size, + src_type == VM_MEM_SRC_ANONYMOUS ? MADV_NOHUGEPAGE : MADV_HUGEPAGE); + TEST_ASSERT(ret == 0, "madvise failed, addr: %p length: 0x%lx src_type: %x", + region->host_mem, npages * vm->page_size, src_type); + } } region->unused_phy_pages = sparsebit_alloc(); -- cgit v1.2.3 From 9478dec3b5e79a1431e2e2b911e32e52a11c6320 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Sat, 31 Oct 2020 11:38:09 -0400 Subject: KVM: vmx: remove unused variable Reported-by: kernel test robot Signed-off-by: Paolo Bonzini --- arch/x86/kvm/vmx/vmx.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c index 61d1cecb6f12..47b8357b9751 100644 --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -6836,7 +6836,6 @@ static void vmx_free_vcpu(struct kvm_vcpu *vcpu) static int vmx_create_vcpu(struct kvm_vcpu *vcpu) { struct vcpu_vmx *vmx; - unsigned long *msr_bitmap; int i, cpu, err; BUILD_BUG_ON(offsetof(struct vcpu_vmx, vcpu) != 0); @@ -6896,7 +6895,6 @@ static int vmx_create_vcpu(struct kvm_vcpu *vcpu) bitmap_fill(vmx->shadow_msr_intercept.read, MAX_POSSIBLE_PASSTHROUGH_MSRS); bitmap_fill(vmx->shadow_msr_intercept.write, MAX_POSSIBLE_PASSTHROUGH_MSRS); - msr_bitmap = vmx->vmcs01.msr_bitmap; vmx_disable_intercept_for_msr(vcpu, MSR_IA32_TSC, MSR_TYPE_R); vmx_disable_intercept_for_msr(vcpu, MSR_FS_BASE, MSR_TYPE_RW); vmx_disable_intercept_for_msr(vcpu, MSR_GS_BASE, MSR_TYPE_RW); -- cgit v1.2.3 From 3cea11cd5e3b00d91caf0b4730194039b45c5891 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 1 Nov 2020 14:43:51 -0800 Subject: Linux 5.10-rc2 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index ee2284a5cad2..9e7fd6a065a7 100644 --- a/Makefile +++ b/Makefile @@ -2,7 +2,7 @@ VERSION = 5 PATCHLEVEL = 10 SUBLEVEL = 0 -EXTRAVERSION = -rc1 +EXTRAVERSION = -rc2 NAME = Kleptomaniac Octopus # *DOCUMENTATION* -- cgit v1.2.3