summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorfarah kassabri <fkassabri@habana.ai>2020-10-14 15:17:36 +0300
committerOded Gabbay <ogabbay@kernel.org>2020-11-30 11:47:32 +0300
commiteb10b897e45968740fde22626843177b7b2f2191 (patch)
treec7c22d418055dfbdeaedb554d22852158a2d0031
parentba7e389c30c6bfb78857ba7a6f8d2cd4bbf5bde7 (diff)
downloadlinux-eb10b897e45968740fde22626843177b7b2f2191.tar.xz
habanalabs: reset device upon fw read failure
failure in reading pre-boot verion is not handled correctly, upon failure we need to reset the device in order to be able to reinstall the driver. Signed-off-by: farah kassabri <fkassabri@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
-rw-r--r--drivers/misc/habanalabs/common/firmware_if.c4
-rw-r--r--drivers/misc/habanalabs/common/habanalabs.h2
-rw-r--r--drivers/misc/habanalabs/common/pci.c5
-rw-r--r--drivers/misc/habanalabs/gaudi/gaudi.c27
-rw-r--r--drivers/misc/habanalabs/goya/goya.c22
5 files changed, 38 insertions, 22 deletions
diff --git a/drivers/misc/habanalabs/common/firmware_if.c b/drivers/misc/habanalabs/common/firmware_if.c
index fb9d901d5059..2fc12e529241 100644
--- a/drivers/misc/habanalabs/common/firmware_if.c
+++ b/drivers/misc/habanalabs/common/firmware_if.c
@@ -607,7 +607,9 @@ int hl_fw_read_preboot_status(struct hl_device *hdev, u32 cpu_boot_status_reg,
return -EIO;
}
- hdev->asic_funcs->read_device_fw_version(hdev, FW_COMP_PREBOOT);
+ rc = hdev->asic_funcs->read_device_fw_version(hdev, FW_COMP_PREBOOT);
+ if (rc)
+ return rc;
security_status = RREG32(cpu_security_boot_status_reg);
diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index f00891ddc3ad..0f4f8ef2cfce 100644
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -927,7 +927,7 @@ struct hl_asic_funcs {
void (*ctx_fini)(struct hl_ctx *ctx);
int (*get_clk_rate)(struct hl_device *hdev, u32 *cur_clk, u32 *max_clk);
u32 (*get_queue_id_for_cq)(struct hl_device *hdev, u32 cq_idx);
- void (*read_device_fw_version)(struct hl_device *hdev,
+ int (*read_device_fw_version)(struct hl_device *hdev,
enum hl_fw_component fwc);
int (*load_firmware_to_device)(struct hl_device *hdev);
int (*load_boot_fit_to_device)(struct hl_device *hdev);
diff --git a/drivers/misc/habanalabs/common/pci.c b/drivers/misc/habanalabs/common/pci.c
index 211f3190f8d7..02152d85cf19 100644
--- a/drivers/misc/habanalabs/common/pci.c
+++ b/drivers/misc/habanalabs/common/pci.c
@@ -390,8 +390,11 @@ int hl_pci_init(struct hl_device *hdev, u32 cpu_boot_status_reg,
rc = hl_fw_read_preboot_status(hdev, cpu_boot_status_reg,
cpu_security_boot_status_reg, boot_err0_reg,
preboot_ver_timeout);
- if (rc)
+ if (rc) {
+ dev_err(hdev->dev, "Failed to read preboot version\n");
+ hdev->asic_funcs->hw_fini(hdev, true);
goto unmap_pci_bars;
+ }
return 0;
diff --git a/drivers/misc/habanalabs/gaudi/gaudi.c b/drivers/misc/habanalabs/gaudi/gaudi.c
index ecfcfdfdd6b0..6aa3e38e3598 100644
--- a/drivers/misc/habanalabs/gaudi/gaudi.c
+++ b/drivers/misc/habanalabs/gaudi/gaudi.c
@@ -3603,7 +3603,7 @@ static int gaudi_load_boot_fit_to_device(struct hl_device *hdev)
return hl_fw_load_fw_to_device(hdev, GAUDI_BOOT_FIT_FILE, dst, 0, 0);
}
-static void gaudi_read_device_fw_version(struct hl_device *hdev,
+static int gaudi_read_device_fw_version(struct hl_device *hdev,
enum hl_fw_component fwc)
{
const char *name;
@@ -3623,7 +3623,7 @@ static void gaudi_read_device_fw_version(struct hl_device *hdev,
break;
default:
dev_warn(hdev->dev, "Undefined FW component: %d\n", fwc);
- return;
+ return -EIO;
}
ver_off &= ~((u32)SRAM_BASE_ADDR);
@@ -3635,7 +3635,10 @@ static void gaudi_read_device_fw_version(struct hl_device *hdev,
dev_err(hdev->dev, "%s version offset (0x%x) is above SRAM\n",
name, ver_off);
strcpy(dest, "unavailable");
+ return -EIO;
}
+
+ return 0;
}
static int gaudi_init_cpu(struct hl_device *hdev)
@@ -3925,16 +3928,18 @@ static void gaudi_hw_fini(struct hl_device *hdev, bool hard_reset)
WREG32(mmPSOC_GLOBAL_CONF_BOOT_STRAP_PINS, boot_strap);
- gaudi->hw_cap_initialized &= ~(HW_CAP_CPU | HW_CAP_CPU_Q |
- HW_CAP_HBM | HW_CAP_PCI_DMA |
- HW_CAP_MME | HW_CAP_TPC_MASK |
- HW_CAP_HBM_DMA | HW_CAP_PLL |
- HW_CAP_NIC_MASK | HW_CAP_MMU |
- HW_CAP_SRAM_SCRAMBLER |
- HW_CAP_HBM_SCRAMBLER |
- HW_CAP_CLK_GATE);
+ if (gaudi) {
+ gaudi->hw_cap_initialized &= ~(HW_CAP_CPU | HW_CAP_CPU_Q |
+ HW_CAP_HBM | HW_CAP_PCI_DMA |
+ HW_CAP_MME | HW_CAP_TPC_MASK |
+ HW_CAP_HBM_DMA | HW_CAP_PLL |
+ HW_CAP_NIC_MASK | HW_CAP_MMU |
+ HW_CAP_SRAM_SCRAMBLER |
+ HW_CAP_HBM_SCRAMBLER |
+ HW_CAP_CLK_GATE);
- memset(gaudi->events_stat, 0, sizeof(gaudi->events_stat));
+ memset(gaudi->events_stat, 0, sizeof(gaudi->events_stat));
+ }
}
static int gaudi_suspend(struct hl_device *hdev)
diff --git a/drivers/misc/habanalabs/goya/goya.c b/drivers/misc/habanalabs/goya/goya.c
index ab467943cef3..a0580b8e1f05 100644
--- a/drivers/misc/habanalabs/goya/goya.c
+++ b/drivers/misc/habanalabs/goya/goya.c
@@ -2341,7 +2341,7 @@ static int goya_load_boot_fit_to_device(struct hl_device *hdev)
* FW component passes an offset from SRAM_BASE_ADDR in SCRATCHPAD_xx.
* The version string should be located by that offset.
*/
-static void goya_read_device_fw_version(struct hl_device *hdev,
+static int goya_read_device_fw_version(struct hl_device *hdev,
enum hl_fw_component fwc)
{
const char *name;
@@ -2361,7 +2361,7 @@ static void goya_read_device_fw_version(struct hl_device *hdev,
break;
default:
dev_warn(hdev->dev, "Undefined FW component: %d\n", fwc);
- return;
+ return -EIO;
}
ver_off &= ~((u32)SRAM_BASE_ADDR);
@@ -2373,7 +2373,11 @@ static void goya_read_device_fw_version(struct hl_device *hdev,
dev_err(hdev->dev, "%s version offset (0x%x) is above SRAM\n",
name, ver_off);
strcpy(dest, "unavailable");
+
+ return -EIO;
}
+
+ return 0;
}
static int goya_init_cpu(struct hl_device *hdev)
@@ -2644,12 +2648,14 @@ static void goya_hw_fini(struct hl_device *hdev, bool hard_reset)
WREG32(mmPSOC_GLOBAL_CONF_SW_BTM_FSM,
0xA << PSOC_GLOBAL_CONF_SW_BTM_FSM_CTRL_SHIFT);
- goya->hw_cap_initialized &= ~(HW_CAP_CPU | HW_CAP_CPU_Q |
- HW_CAP_DDR_0 | HW_CAP_DDR_1 |
- HW_CAP_DMA | HW_CAP_MME |
- HW_CAP_MMU | HW_CAP_TPC_MBIST |
- HW_CAP_GOLDEN | HW_CAP_TPC);
- memset(goya->events_stat, 0, sizeof(goya->events_stat));
+ if (goya) {
+ goya->hw_cap_initialized &= ~(HW_CAP_CPU | HW_CAP_CPU_Q |
+ HW_CAP_DDR_0 | HW_CAP_DDR_1 |
+ HW_CAP_DMA | HW_CAP_MME |
+ HW_CAP_MMU | HW_CAP_TPC_MBIST |
+ HW_CAP_GOLDEN | HW_CAP_TPC);
+ memset(goya->events_stat, 0, sizeof(goya->events_stat));
+ }
}
int goya_suspend(struct hl_device *hdev)