summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
AgeCommit message (Collapse)AuthorFilesLines
2024-05-18drm/amdkfd: Ensure gpu_id is uniqueHarish Kasiviswanathan1-6/+34
gpu_id needs to be unique for user space to identify GPUs via KFD interface. In the current implementation there is a very small probability of having non unique gpu_ids. v2: Add check to confirm if gpu_id is unique. If not unique, find one Changed commit header to reflect the above v3: Use crc16 as suggested-by: Lijo Lazar <lijo.lazar@amd.com> Ensure that gpu_id != 0 Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-13drm/amdkfd: Use dev_error intead of pr_errorHarish Kasiviswanathan1-11/+8
No functional change. This will help in moving gpu_id creation to next step while still being able to identify the correct GPU Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-09drm/amdkfd: Reconcile the definition and use of oem_id in struct ↵Michael Chen1-2/+1
kfd_topology_device Currently oem_id is defined as uint8_t[6] and casted to uint64_t* in some use case. This would lead code scanner to complain about access beyond. Re-define it in union to enforce 8-byte size and alignment to avoid potential issue. Signed-off-by: Michael Chen <michael.chen@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-08Revert "drm/amdkfd: Add partition id field to location_id"Lijo Lazar1-3/+2
This reverts commit c37ce764cd492f044dcdbb39616298f02b0dbc7f. RCCL library is currently not treating spatial partitions differently, hence this change is causing issues. Revert temporarily till RCCL implementation is ready for spatial partitions. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Jonathan Kim <jonathan.kim@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-02drm/amdkfd: enable single alu ops for gfx12Jonathan Kim1-0/+4
GFX12 debugging requires setting up precise ALU operation for catching ALU exceptions. Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Tested-by: Lancelot Six <lancelot.six@amd.com> Reviewed-by: Eric Huang <jinhuieric.huang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-05-02drm/amdgpu: Add gfx v9_4_4 ip blockHawking Zhang1-2/+4
Add gfx v9_4_4 ip block support Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-02-26drm/amdkfd: Add partition id field to location_idLijo Lazar1-2/+3
On devices which have multi-partition nodes, keep partition id in location_id[31:28]. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Jonathan Kim <jonathan.kim@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-02-13drm/amdkfd: Fix L2 cache size reporting in GFX9.4.3Kent Russell1-6/+4
Its currently incorrectly multiplied by number of XCCs in the partition Fixes: be457b2252b6 ("drm/amdkfd: Update cache info for GFX 9.4.3") Signed-off-by: Kent Russell <kent.russell@amd.com> Reviewed-by: Mukul Joshi <mukul.joshi@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-02-07drm/amdkfd: Add cache line sizes to KFD topologyJoseph Greathouse1-0/+3
The KFD topology includes cache line size, but we have not been filling that information out unless we are parsing a CRAT table. Fill in this information for the devices where we have cache information structs, and pipe this information to the topology sysfs files. v2: squash in fix from Joe (Alex) Signed-off-by: Joseph Greathouse <Joseph.Greathouse@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-01-06drm/amdkfd: Fix iterator used outside loop in 'kfd_add_peer_prop()'Srinivasan Shanmugam1-11/+13
Fix the following about iterator use: drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_topology.c:1456 kfd_add_peer_prop() warn: iterator used outside loop: 'iolink3' Cc: Felix Kuehling <Felix.Kuehling@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-01-06drm/amdkfd: Confirm list is non-empty before utilizing list_first_entry in ↵Srinivasan Shanmugam1-9/+12
kfd_topology.c Before using list_first_entry, make sure to check that list is not empty, if list is empty return -ENODATA. Fixes the below: drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_topology.c:1347 kfd_create_indirect_link_prop() warn: can 'gpu_link' even be NULL? drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_topology.c:1428 kfd_add_peer_prop() warn: can 'iolink1' even be NULL? drivers/gpu/drm/amd/amdgpu/../amdkfd/kfd_topology.c:1433 kfd_add_peer_prop() warn: can 'iolink2' even be NULL? Fixes: 0f28cca87e9a ("drm/amdkfd: Extend KFD device topology to surface peer-to-peer links") Cc: Felix Kuehling <Felix.Kuehling@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Suggested-by: Felix Kuehling <Felix.Kuehling@amd.com> Suggested-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-11-08Merge tag 'drm-next-2023-11-07' of git://anongit.freedesktop.org/drm/drmLinus Torvalds1-2/+16
Pull more drm updates from Dave Airlie: "Geert pointed out I missed the renesas reworks in my main pull, so this pull contains the renesas next work for atomic conversion and DT support. It also contains a bunch of amdgpu and some small ssd13xx fixes. renesas: - atomic conversion - DT support ssd13xx: - dt binding fix for ssd132x - Initialize ssd130x crtc_state to NULL. amdgpu: - Fix RAS support check - RAS fixes - MES fixes - SMU13 fixes - Contiguous memory allocation fix - BACO fixes - GPU reset fixes - Min power limit fixes - GFX11 fixes - USB4/TB hotplug fixes - ARM regression fix - GFX9.4.3 fixes - KASAN/KCSAN stack size check fixes - SR-IOV fixes - SMU14 fixes - PSP13 fixes - Display blend fixes - Flexible array size fixes amdkfd: - GPUVM fix radeon: - Flexible array size fixes" * tag 'drm-next-2023-11-07' of git://anongit.freedesktop.org/drm/drm: (83 commits) drm/amd/display: Enable fast update on blendTF change drm/amd/display: Fix blend LUT programming drm/amd/display: Program plane color setting correctly drm/amdgpu: Query and report boot status drm/amdgpu: Add psp v13 function to query boot status drm/amd/swsmu: remove fw version check in sw_init. drm/amd/swsmu: update smu v14_0_0 driver if and metrics table drm/amdgpu: Add C2PMSG_109/126 reg field shift/masks drm/amdgpu: Optimize the asic type fix code drm/amdgpu: fix GRBM read timeout when do mes_self_test drm/amdgpu: check recovery status of xgmi hive in ras_reset_error_count drm/amd/pm: only check sriov vf flag once when creating hwmon sysfs drm/amdgpu: Attach eviction fence on alloc drm/amdkfd: Improve amdgpu_vm_handle_moved drm/amd/display: Increase frame warning limit with KASAN or KCSAN in dml2 drm/amd/display: Avoid NULL dereference of timing generator drm/amdkfd: Update cache info for GFX 9.4.3 drm/amdkfd: Populate cache info for GFX 9.4.3 drm/amdgpu: don't put MQDs in VRAM on ARM | ARM64 drm/amdgpu/smu13: drop compute workload workaround ...
2023-11-03drm/amdkfd: Update cache info for GFX 9.4.3Mukul Joshi1-2/+16
Update cache info reporting based on compute and memory partitioning modes. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-11-01Merge tag 'drm-next-2023-10-31-1' of git://anongit.freedesktop.org/drm/drmLinus Torvalds1-24/+20
Pull drm updates from Dave Airlie: "Highlights: - AMD adds some more upcoming HW platforms - Intel made Meteorlake stable and started adding Lunarlake - nouveau has a bunch of display rework in prepartion for the NVIDIA GSP firmware support - msm adds a7xx support - habanalabs has finished migration to accel subsystem Detail summary: kernel: - add initial vmemdup-user-array core: - fix platform remove() to return void - drm_file owner updated to reflect owner - move size calcs to drm buddy allocator - let GPUVM build as a module - allow variable number of run-queues in scheduler edid: - handle bad h/v sync_end in EDIDs panfrost: - add Boris as maintainer fbdev: - use fb_ops helpers more - only allow logo use from fbcon - rename fb_pgproto to pgprot_framebuffer - add HPD state to drm_connector_oob_hotplug_event - convert to fbdev i/o mem helpers i915: - Enable meteorlake by default - Early Xe2 LPD/Lunarlake display enablement - Rework subplatforms into IP version checks - GuC based TLB invalidation for Meteorlake - Display rework for future Xe driver integration - LNL FBC features - LNL display feature capability reads - update recommended fw versions for DG2+ - drop fastboot module parameter - added deviceid for Arrowlake-S - drop preproduction workarounds - don't disable preemption for resets - cleanup inlines in headers - PXP firmware loading fix - Fix sg list lengths - DSC PPS state readout/verification - Add more RPL P/U PCI IDs - Add new DG2-G12 stepping - DP enhanced framing support to state checker - Improve shared link bandwidth management - stop using GEM macros in display code - refactor related code into display code - locally enable W=1 warnings - remove PSR watchdog timers on LNL amdgpu: - RAS/FRU EEPROM updatse - IP discovery updatses - GC 11.5 support - DCN 3.5 support - VPE 6.1 support - NBIO 7.11 support - DML2 support - lots of IP updates - use flexible arrays for bo list handling - W=1 fixes - Enable seamless boot in more cases - Enable context type property for HDMI - Rework GPUVM TLB flushing - VCN IB start/size alignment fixes amdkfd: - GC 10/11 fixes - GC 11.5 support - use partial migration in GPU faults radeon: - W=1 Fixes - fix some possible buffer overflow/NULL derefs nouveau: - update uapi for NO_PREFETCH - scheduler/fence fixes - rework suspend/resume for GSP-RM - rework display in preparation for GSP-RM habanalabs: - uapi: expose tsc clock - uapi: block access to eventfd through control device - uapi: force dma-buf export to PAGE_SIZE alignments - complete move to accel subsystem - move firmware interface include files - perform hard reset on PCIe AXI drain event - optimise user interrupt handling msm: - DP: use existing helpers for DPCD - DPU: interrupts reworked - gpu: a7xx (a730/a740) support - decouple msm_drv from kms for headless devices mediatek: - MT8188 dsi/dp/edp support - DDP GAMMA - 12 bit LUT support - connector dynamic selection capability rockchip: - rv1126 mipi-dsi/vop support - add planar formats ast: - rename constants panels: - Mitsubishi AA084XE01 - JDI LPM102A188A - LTK050H3148W-CTA6 ivpu: - power management fixes qaic: - add detach slice bo api komeda: - add NV12 writeback tegra: - support NVSYNC/NHSYNC - host1x suspend fixes ili9882t: - separate into own driver" * tag 'drm-next-2023-10-31-1' of git://anongit.freedesktop.org/drm/drm: (1803 commits) drm/amdgpu: Remove unused variables from amdgpu_show_fdinfo drm/amdgpu: Remove duplicate fdinfo fields drm/amd/amdgpu: avoid to disable gfxhub interrupt when driver is unloaded drm/amdgpu: Add EXT_COHERENT support for APU and NUMA systems drm/amdgpu: Retrieve CE count from ce_count_lo_chip in EccInfo table drm/amdgpu: Identify data parity error corrected in replay mode drm/amdgpu: Fix typo in IP discovery parsing drm/amd/display: fix S/G display enablement drm/amdxcp: fix amdxcp unloads incompletely drm/amd/amdgpu: fix the GPU power print error in pm info drm/amdgpu: Use pcie domain of xcc acpi objects drm/amd: check num of link levels when update pcie param drm/amdgpu: Add a read to GFX v9.4.3 ring test drm/amd/pm: call smu_cmn_get_smc_version in is_mode1_reset_supported. drm/amdgpu: get RAS poison status from DF v4_6_2 drm/amdgpu: Use discovery table's subrevision drm/amd/display: 3.2.256 drm/amd/display: add interface to query SubVP status drm/amd/display: Read before writing Backlight Mode Set Register drm/amd/display: Disable SYMCLK32_SE RCO on DCN314 ...
2023-10-31Merge tag 'x86-core-2023-10-29-v2' of ↵Linus Torvalds1-1/+1
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip Pull x86 core updates from Thomas Gleixner: - Limit the hardcoded topology quirk for Hygon CPUs to those which have a model ID less than 4. The newer models have the topology CPUID leaf 0xB correctly implemented and are not affected. - Make SMT control more robust against enumeration failures SMT control was added to allow controlling SMT at boottime or runtime. The primary purpose was to provide a simple mechanism to disable SMT in the light of speculation attack vectors. It turned out that the code is sensible to enumeration failures and worked only by chance for XEN/PV. XEN/PV has no real APIC enumeration which means the primary thread mask is not set up correctly. By chance a XEN/PV boot ends up with smp_num_siblings == 2, which makes the hotplug control stay at its default value "enabled". So the mask is never evaluated. The ongoing rework of the topology evaluation caused XEN/PV to end up with smp_num_siblings == 1, which sets the SMT control to "not supported" and the empty primary thread mask causes the hotplug core to deny the bringup of the APS. Make the decision logic more robust and take 'not supported' and 'not implemented' into account for the decision whether a CPU should be booted or not. - Fake primary thread mask for XEN/PV Pretend that all XEN/PV vCPUs are primary threads, which makes the usage of the primary thread mask valid on XEN/PV. That is consistent with because all of the topology information on XEN/PV is fake or even non-existent. - Encapsulate topology information in cpuinfo_x86 Move the randomly scattered topology data into a separate data structure for readability and as a preparatory step for the topology evaluation overhaul. - Consolidate APIC ID data type to u32 It's fixed width hardware data and not randomly u16, int, unsigned long or whatever developers decided to use. - Cure the abuse of cpuinfo for persisting logical IDs. Per CPU cpuinfo is used to persist the logical package and die IDs. That's really not the right place simply because cpuinfo is subject to be reinitialized when a CPU goes through an offline/online cycle. Use separate per CPU data for the persisting to enable the further topology management rework. It will be removed once the new topology management is in place. - Provide a debug interface for inspecting topology information Useful in general and extremly helpful for validating the topology management rework in terms of correctness or "bug" compatibility. * tag 'x86-core-2023-10-29-v2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (23 commits) x86/apic, x86/hyperv: Use u32 in hv_snp_boot_ap() too x86/cpu: Provide debug interface x86/cpu/topology: Cure the abuse of cpuinfo for persisting logical ids x86/apic: Use u32 for wakeup_secondary_cpu[_64]() x86/apic: Use u32 for [gs]et_apic_id() x86/apic: Use u32 for phys_pkg_id() x86/apic: Use u32 for cpu_present_to_apicid() x86/apic: Use u32 for check_apicid_used() x86/apic: Use u32 for APIC IDs in global data x86/apic: Use BAD_APICID consistently x86/cpu: Move cpu_l[l2]c_id into topology info x86/cpu: Move logical package and die IDs into topology info x86/cpu: Remove pointless evaluation of x86_coreid_bits x86/cpu: Move cu_id into topology info x86/cpu: Move cpu_core_id into topology info hwmon: (fam15h_power) Use topology_core_id() scsi: lpfc: Use topology_core_id() x86/cpu: Move cpu_die_id into topology info x86/cpu: Move phys_proc_id into topology info x86/cpu: Encapsulate topology information in cpuinfo_x86 ...
2023-10-10x86/cpu: Encapsulate topology information in cpuinfo_x86Thomas Gleixner1-1/+1
The topology related information is randomly scattered across cpuinfo_x86. Create a new structure cpuinfo_topo and move in a first step initial_apicid and apicid into it. Aside of being better readable this is in preparation for replacing the horribly fragile CPU topology evaluation code further down the road. Consolidate APIC ID fields to u32 as that represents the hardware type. No functional change. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Tested-by: Juergen Gross <jgross@suse.com> Tested-by: Sohil Mehta <sohil.mehta@intel.com> Tested-by: Michael Kelley <mikelley@microsoft.com> Tested-by: Peter Zijlstra (Intel) <peterz@infradead.org> Tested-by: Zhang Rui <rui.zhang@intel.com> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lore.kernel.org/r/20230814085112.269787744@linutronix.de
2023-10-03drm/amdkfd: drop struct kfd_cu_infoAlex Deucher1-29/+20
I think this was an abstraction back from when kfd supported both radeon and amdgpu. Since we just support amdgpu now, there is no more need for this and we can use the amdgpu structures directly. This also avoids having the kfd_cu_info structures on the stack when inlining which can blow up the stack. Cc: Arnd Bergmann <arnd@kernel.org> Acked-by: Arnd Bergmann <arnd@arndb.de> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-10-03drm/amdkfd: reduce stack size in kfd_topology_add_device()Alex Deucher1-4/+9
kfd_topology.c:2082:1: warning: the frame size of 1440 bytes is larger than 1024 bytes Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2866 Cc: Arnd Bergmann <arnd@kernel.org> Acked-by: Arnd Bergmann <arnd@arndb.de> Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-09-12drm/amdkfd: Update cache info reporting for GFX v9.4.3Mukul Joshi1-36/+46
Update cache info reporting in sysfs to report the correct number of CUs and associated cache information based on different spatial partitioning modes. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-09-12drm/amdgpu: Store CU info from all XCCs for GFX v9.4.3Mukul Joshi1-6/+5
Currently, we store CU info only for a single XCC assuming that it is the same for all XCCs. However, that may not be true. As a result, store CU info for all XCCs. This info is later used for CU masking. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-09-12drm/amdkfd: Update cache info reporting for GFX v9.4.3Mukul Joshi1-36/+46
Update cache info reporting in sysfs to report the correct number of CUs and associated cache information based on different spatial partitioning modes. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-09-12drm/amdgpu: Store CU info from all XCCs for GFX v9.4.3Mukul Joshi1-6/+5
Currently, we store CU info only for a single XCC assuming that it is the same for all XCCs. However, that may not be true. As a result, store CU info for all XCCs. This info is later used for CU masking. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-08-16drm/amdgpu: skip xcp drm device allocation when out of drm resourceJames Zhu1-1/+8
Return 0 when drm device alloc failed with -ENOSPC in order to allow amdgpu drive loading. But the xcp without drm device node assigned won't be visiable in user space. This helps amdgpu driver loading on system which has more than 64 nodes, the current limitation. The proposal to add more drm nodes is discussed in public, which will support up to 2^20 nodes totally. kernel drm: https://lore.kernel.org/lkml/20230724211428.3831636-1-michal.winiarski@intel.com/T/ libdrm: https://gitlab.freedesktop.org/mesa/drm/-/merge_requests/305 Signed-off-by: James Zhu <James.Zhu@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-08-11drm/amdkfd: drop IOMMUv2 supportAlex Deucher1-96/+15
Now that we use the dGPU path for all APUs, drop the IOMMUv2 support. v2: drop the now unused queue manager functions for gfx7/8 APUs Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Tested-by: Mike Lothian <mike@fireburn.co.uk> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-07-27drm/amdkfd: fix and enable ttmp setup for gfx11Jonathan Kim1-7/+5
The MES cached process context must be cleared on adding any queue for the first time. For proper debug support, the MES will clear it's cached process context on the first call to SET_SHADER_DEBUGGER. This allows TTMPs to be pesistently enabled in a safe manner. Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Eric Huang <jinhuieric@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-07-12drm/amdkfd: report dispatch id always saved in ttmps after gc9.4.2Jonathan Kim1-2/+3
The feature to save the dispatch ID in trap temporaries 6 & 7 on context save is unconditionally enabled during MQD initialization. Now that TTMPs are always setup regardless of debug mode for GC 9.4.3, we should report that the dispatch ID is always available for debug/trap handling. Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-07-12drm/amdkfd: restore debugger additional info for gfx v9_4_3Jonathan Kim1-2/+8
The additional information that the KFD reports to the debugger was destroyed when the following commit was merged: "drm/amdkfd: convert switches to IP version checking" Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Acked-by: Amber Lin <amber.lin@amd.com> Signed-off-by: Eric Huang <jinhuieric.huang@amd.com> Reviewed-by: Jonathan Kim <jonathan.kim@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23drm/amdkfd: set coherent host access capability flagAlex Sierra1-0/+4
This flag determines whether the host possesses coherent access to the memory of the device. Signed-off-by: Alex Sierra <alex.sierra@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: fix and enable debugging for gfx11Jonathan Kim1-5/+7
There are a couple of fixes required to enable gfx11 debugging. First, ADD_QUEUE.trap_en is an inappropriate place to toggle a per-process register so move it to SET_SHADER_DEBUGGER.trap_en. When ADD_QUEUE.skip_process_ctx_clear is set, MES will prioritize the SET_SHADER_DEBUGGER.trap_en setting. Second, to preserve correct save/restore priviledged wave states in coordination with the trap enablement setting, resume suspended waves early in the disable call. Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: apply trap workaround for gfx11Jonathan Kim1-2/+5
Due to a HW bug, waves in only half the shader arrays can enter trap. When starting a debug session, relocate all waves to the first shader array of each shader engine and mask off the 2nd shader array as unavailable. When ending a debug session, re-enable the 2nd shader array per shader engine. User CU masking per queue cannot be guaranteed to remain functional if requested during debugging (e.g. user cu mask requests only 2nd shader array as an available resource leading to zero HW resources available) nor can runtime be alerted of any of these changes during execution. Make user CU masking and debugging mutual exclusive with respect to availability. If the debugger tries to attach to a process with a user cu masked queue, return the runtime status as enabled but busy. If the debugger tries to attach and fails to reallocate queue waves to the first shader array of each shader engine, return the runtime status as enabled but with an error. In addition, like any other mutli-process debug supported devices, disable trap temporary setup per-process to avoid performance impact from setup overhead. Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: display debug capabilitiesJonathan Kim1-5/+96
Expose debug capabilities in the KFD topology node's HSA capabilities and debug properties flags. Ensure correct capabilities are exposed based on firmware support. Flag definitions can be referenced in uapi/linux/kfd_sysfs.h. Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: Do not access members of xcp w/o check (v2)Hawking Zhang1-1/+1
Not all the asic needs xcp. ensure check xcp availabity before accessing its member. v2: add missing change in kfd_topology.c Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: Move local_mem_info to kfd_nodeMukul Joshi1-3/+4
We need to track memory usage on a per partition basis. To do that, store the local memory information in KFD node instead of kfd device. v2: squash in fix ("amdkfd: Use mem_id to access mem_partition info") Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: KFD graphics interop support compute partitionPhilip Yang1-18/+0
kfd_ioctl_get_dmabuf use the amdgpu bo xcp_id to get the gpu_id of the KFD node from the exported dmabuf_adev, and then create kfd bo on the correct adev and KFD node when importing the amdgpu bo to KFD. Remove function kfd_device_by_adev, it is not needed as it is the same result as dmabuf_adev->kfd.dev->nodes[0]->id. Signed-off-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: Store drm node minor number for kfd nodesPhilip Yang1-2/+6
From KFD topology, application will find kfd node with the corresponding drm device node minor number, for example if partition drm node starts from /dev/dri/renderD129, then KFD node 0 with store drm node minor number 129. Application will open drm node /dev/dri/renderD129 to create amdgpu vm for kfd node 0 with the correct vm->mem_id to indicate the memory partition. Signed-off-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: Move pgmap to amdgpu_kfd_dev structurePhilip Yang1-1/+1
VRAM pgmap resource is allocated every time when switching compute partitions because kfd_dev is re-initialized by post_partition_switch, As a result, it causes memory region resource leaking and system memory usage accounting unbalanced. pgmap resource should be allocated and registered only once when loading driver and freed when unloading driver, move it from kfd_dev to amdgpu_kfd_dev. Signed-off-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: Use xcc mask for identifying xccLijo Lazar1-4/+4
Instead of start xcc id and number of xcc per node, use the xcc mask which is the mask of logical ids of xccs belonging to a parition. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: bind cpu and hiveless gpu to a hive if xgmi connectedJonathan Kim1-1/+8
If a CPU and GPU are xGMI connected but the GPU is hiveless with respect to other GPUs, create a new CPU-GPU hive using the GPU's PCI device location ID as the new hive ID to maintain fine grain memory access usage. Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: Report XGMI IOLINKs for GFXIP9.4.3Rajneesh Bhardwaj1-3/+2
GFXIP 9.4.3 could be in APU or carveout mode but we cannot use the xgmi.connected_to_cpu flag to identify the iolinks type. Use appropriate APU or Carveout mode based condition to report xgmi connection in kfd topology. Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: add gpu compute cores io links for gfx9.4.3Jonathan Kim1-0/+4
The PSP TA will only provide xGMI topology info for links between GPU sockets so links between partitions from different sockets will be hardcoded as 3 xGMI hops with 1 hops weighted as xGMI and 2 hops weighted with a new intra-socket weight to indicate the longest possible distance. If the link between a partition and the CPU is non-PCIe, then assume the CPU (CCDs) is located within the same socket as the partition and represent the link as an intra-socket weighted single hop XGMI link with memory bandwidth. Links between partitions within a single socket will be abstracted as single hop xGMI links weighted with the new intra-socket weight and will have memory bandwidth. Finally, use the unused function bits in the location ID to represent the coordinates of the compute partition within its socket. A follow on patch will resolve the requirement for GPU socket xGMI link representation sometime later. Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: Update sysfs node properties for multi XCCMukul Joshi1-2/+4
Update simd_count and array_count node properties to report values multiplied by number of XCCs in the partition. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Tested-by: Amber Lin <Amber.Lin@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: Add spatial partitioning support in KFDMukul Joshi1-3/+5
This patch introduces multi-partition support in KFD. This patch includes: - Support for maximum 8 spatial partitions in KFD. - Initialize one HIQ per partition. - Management of VMID range depending on partition mode. - Management of doorbell aperture space between all partitions. - Each partition does its own queue management, interrupt handling, SMI event reporting. - IOMMU, if enabled with multiple partitions, will only work on first partition. - SPM is only supported on the first partition. - Currently, there is no support for resetting individual partitions. All partitions will reset together. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Tested-by: Amber Lin <Amber.Lin@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: Introduce kfd_node struct (v5)Mukul Joshi1-28/+28
Introduce a new structure, kfd_node, which will now represent a compute node. kfd_node is carved out of kfd_dev structure. kfd_dev struct now will become the parent of kfd_node, and will store common resources such as doorbells, GTT sub-alloctor etc. kfd_node struct will store all resources specific to a compute node, such as device queue manager, interrupt handling etc. This is the first step in adding compute partition support in KFD. v2: introduce kfd_node struct to gc v11 (Hawking) v3: make reference to kfd_dev struct through kfd_node (Morris) v4: use kfd_node instead for kfd isr/mqd functions (Morris) v5: rebase (Alex) Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Tested-by: Amber Lin <Amber.Lin@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Morris Zhang <Shiwu.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-02-24drm/amdkfd: Make kobj_type structures constantThomas Weißschuh1-5/+5
Since commit ee6d3dd4ed48 ("driver core: make kobj_type constant.") the driver core allows the usage of const struct kobj_type. Take advantage of this to constify the structure definitions to prevent modification at runtime. Signed-off-by: Thomas Weißschuh <linux@weissschuh.net> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-01-04drm/amdkfd: Fix kernel warning during topology setupMukul Joshi1-1/+1
This patch fixes the following kernel warning seen during driver load by correctly initializing the p2plink attr before creating the sysfs file: [ +0.002865] ------------[ cut here ]------------ [ +0.002327] kobject: '(null)' (0000000056260cfb): is not initialized, yet kobject_put() is being called. [ +0.004780] WARNING: CPU: 32 PID: 1006 at lib/kobject.c:718 kobject_put+0xaa/0x1c0 [ +0.001361] Call Trace: [ +0.001234] <TASK> [ +0.001067] kfd_remove_sysfs_node_entry+0x24a/0x2d0 [amdgpu] [ +0.003147] kfd_topology_update_sysfs+0x3d/0x750 [amdgpu] [ +0.002890] kfd_topology_add_device+0xbd7/0xc70 [amdgpu] [ +0.002844] ? lock_release+0x13c/0x2e0 [ +0.001936] ? smu_cmn_send_smc_msg_with_param+0x1e8/0x2d0 [amdgpu] [ +0.003313] ? amdgpu_dpm_get_mclk+0x54/0x60 [amdgpu] [ +0.002703] kgd2kfd_device_init.cold+0x39f/0x4ed [amdgpu] [ +0.002930] amdgpu_amdkfd_device_init+0x13d/0x1f0 [amdgpu] [ +0.002944] amdgpu_device_init.cold+0x1464/0x17b4 [amdgpu] [ +0.002970] ? pci_bus_read_config_word+0x43/0x80 [ +0.002380] amdgpu_driver_load_kms+0x15/0x100 [amdgpu] [ +0.002744] amdgpu_pci_probe+0x147/0x370 [amdgpu] [ +0.002522] local_pci_probe+0x40/0x80 [ +0.001896] work_for_cpu_fn+0x10/0x20 [ +0.001892] process_one_work+0x26e/0x5a0 [ +0.002029] worker_thread+0x1fd/0x3e0 [ +0.001890] ? process_one_work+0x5a0/0x5a0 [ +0.002115] kthread+0xea/0x110 [ +0.001618] ? kthread_complete_and_exit+0x20/0x20 [ +0.002422] ret_from_fork+0x1f/0x30 [ +0.001808] </TASK> [ +0.001103] irq event stamp: 59837 [ +0.001718] hardirqs last enabled at (59849): [<ffffffffb30fab12>] __up_console_sem+0x52/0x60 [ +0.004414] hardirqs last disabled at (59860): [<ffffffffb30faaf7>] __up_console_sem+0x37/0x60 [ +0.004414] softirqs last enabled at (59654): [<ffffffffb307d9c7>] irq_exit_rcu+0xd7/0x130 [ +0.004205] softirqs last disabled at (59649): [<ffffffffb307d9c7>] irq_exit_rcu+0xd7/0x130 [ +0.004203] ---[ end trace 0000000000000000 ]--- Fixes: 0f28cca87e9a ("drm/amdkfd: Extend KFD device topology to surface peer-to-peer links") Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-11-29drm/amdkfd: Remove unnecessary condition in kfd_topology_add_device()Dan Carpenter1-3/+2
We re-arranged this code recently so "ret" is always zero at this point. Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-11-23drm/amdkfd: Release the topology_lock in error caseFelix Kuehling1-55/+65
Move the topology-locked part of kfd_topology_add_device into a separate function to simlpify error handling and release the topology lock consistently. Reported-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Felix Kuehling <felix.kuehling@gmail.com> Signed-off-by: Ma Jun <Jun.Ma2@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-11-10drm/amdkfd: Make kfd_fill_cache_non_crat_info() as staticMa Jun1-1/+1
kfd_fill_cache_non_crat_info() is only used in kfd_topology.c, so make it as static. Signed-off-by: Ma Jun <Jun.Ma2@amd.com> Reviewed-by: Guchun Chen <guchun.chen@amd.com> Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-11-04drm/amdkfd: Fix the warning of array-index-out-of-boundsMa Jun1-13/+232
For some GPUs with more CUs, the original sibling_map[32] in struct crat_subtype_cache is not enough to save the cache information when create the VCRAT table, so skip filling the struct crat_subtype_cache info instead fill struct kfd_cache_properties directly to fix this problem. Signed-off-by: Ma Jun <Jun.Ma2@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-10-27drm/amdkfd: Cleanup kfd_dev structMukul Joshi1-13/+12
Cleanup kfd_dev struct by removing ddev and pdev as both drm_device and pci_dev can be fetched from amdgpu_device. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Tested-by: Amber Lin <Amber.Lin@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
generated by cgit v1.2.3 (git 2.39.0) at 2024-08-29 16:27:12 +0300