linux/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c, branch v6.12

drm/amdgpu: nuke the VM PD/PT shadow handling

2024-09-18T20:15:06Z

This was only used as workaround for recovering the page tables after VRAM was lost and is no longer necessary after the function amdgpu_vm_bo_reset_state_machine() started to do the same. Compute never used shadows either, so the only proplematic case left is SVM and that is most likely not recoverable in any way when VRAM is lost. Signed-off-by: Christian König Acked-by: Lijo Lazar Signed-off-by: Alex Deucher

drm/amdgpu: Fix missing check pcie_p2p module param

2024-09-17T14:04:57Z

The module param pcie_p2p should be checked for kfd p2p feature, so add it. Fixes: 75f0efbc4b3b ("drm/amdgpu: Take IOMMU remapping into account for p2p checks") Signed-off-by: Bob Zhou Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher

drm/amd/amdgpu: move drain_workqueue before shutdown is set

2024-08-29T17:39:07Z

[background] when unloading amdgpu driver right after running a workload, drain_workqueue is causing "Fence fallback timer expired on ring sdma0.0". Under sriov, this issue will cause sriov full access timeout and a reset happening. move drain_workqueue before shutdown is set to allow ih process and before enter full access under sriov to avoid full access time cost. Signed-off-by: Victor Zhao Reviewed-by: Feifei Xu Signed-off-by: Alex Deucher

drm/amdgpu: skip printing vram_lost if needed

2024-08-29T17:38:53Z

The vm lost status can only be obtained after a GPU reset occurs, but sometimes a dev core dump can be happened before GPU reset. So a new argument is added to tell the dev core dump implementation whether to skip printing the vram_lost status in the dump. And this patch is also trying to decouple the core dump function from the GPU reset function, by replacing the argument amdgpu_reset_context with amdgpu_job to specify the context for core dump. V2: Inform user if VRAM lost check is skipped so users don't assume VRAM wasn't lost (Alex) Signed-off-by: Trigger Huang Suggested-by: Alex Deucher Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher

Merge tag 'amd-drm-next-6.12-2024-08-26' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

2024-08-27T12:33:12Z

amd-drm-next-6.12-2024-08-26: amdgpu: - SDMA devcoredump support - DCN 4.0.1 updates - DC SUBVP fixes - Refactor OPP in DC - Refactor MMHUBBUB in DC - DC DML 2.1 updates - DC FAMS2 updates - RAS updates - GFX12 updates - VCN 4.0.3 updates - JPEG 4.0.3 updates - Enable wave kill (soft recovery) for compute queues - Clean up CP error interrupt handling - Enable CP bad opcode interrupts - VCN 4.x fixes - VCN 5.x fixes - GPU reset fixes - Fix vbios embedded EDID size handling - SMU 14.x updates - Misc code cleanups and spelling fixes - VCN devcoredump support - ISP MFD i2c support - DC vblank fixes - GFX 12 fixes - PSR fixes - Convert vbios embedded EDID to drm_edid - DCN 3.5 updates - DMCUB updates - Cursor fixes - Overdrive support for SMU 14.x - GFX CP padding optimizations - DCC fixes - DSC fixes - Preliminary per queue reset infrastructure - Initial per queue reset support for GFX 9 - Initial per queue reset support for GFX 7, 8 - DCN 3.2 fixes - DP MST fixes - SR-IOV fixes - GFX 9.4.3/4 devcoredump support - Add process isolation framework - Enable process isolation support for GFX 9.4.3/4 - Take IOMMU remapping into account for P2P DMA checks amdkfd: - CRIU fixes - Improved input validation for user queues - HMM fix - Enable process isolation support for GFX 9.4.3/4 - Initial per queue reset support for GFX 9 - Allow users to target recommended SDMA engines radeon: - remove .load and drm_dev_alloc - Fix vbios embedded EDID size handling - Convert vbios embedded EDID to drm_edid - Use GEM references instead of TTM - r100 cp init cleanup - Fix potential overflows in evergreen CS offset tracking UAPI: - KFD support for targetting queues on recommended SDMA engines Proposed userspace: https://github.com/ROCm/ROCR-Runtime/commit/2f588a24065f41c208c3701945e20be746d8faf7 https://github.com/ROCm/ROCR-Runtime/commit/eb30a5bbc7719c6ffcf2d2dd2878bc53a47b3f30 drm/buddy: - Add start address support for trim function From: Alex Deucher Link: https://patchwork.freedesktop.org/patch/msgid/20240826201528.55307-1-alexander.deucher@amd.com

drm/amdgpu: Take IOMMU remapping into account for p2p checks

2024-08-23T14:53:05Z

when trying to enable p2p the amdgpu_device_is_peer_accessible() checks the condition where address_mask overlaps the aper_base and hence returns 0, due to which the p2p disables for this platform IOMMU should remap the BAR addresses so the device can access them. Hence check if peer_adev is remapping DMA v5: (Felix, Alex) - fixing comment as per Alex feedback - refactor code as per Felix v4: (Alex) - fix the comment and description v3: - remove iommu_remap variable v2: (Alex) - Fix as per review comments - add new function amdgpu_device_check_iommu_remap to check if iommu remap Signed-off-by: Rahul Jain Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdgpu: Implement Enforce Isolation Handler for KGD/KFD serialization

2024-08-21T02:07:35Z

This commit introduces the Enforce Isolation Handler designed to enforce shader isolation on AMD GPUs, which helps to prevent data leakage between different processes. The handler counts the number of emitted fences for each GFX and compute ring. If there are any fences, it schedules the `enforce_isolation_work` to be run after a delay of `GFX_SLICE_PERIOD`. If there are no fences, it signals the Kernel Fusion Driver (KFD) to resume the runqueue. The function is synchronized using the `enforce_isolation_mutex`. This commit also introduces a reference count mechanism (kfd_sch_req_count) to keep track of the number of requests to enable the KFD scheduler. When a request to enable the KFD scheduler is made, the reference count is decremented. When the reference count reaches zero, a delayed work is scheduled to enforce isolation after a delay of GFX_SLICE_PERIOD. When a request to disable the KFD scheduler is made, the function first checks if the reference count is zero. If it is, it cancels the delayed work for enforcing isolation and checks if the KFD scheduler is active. If the KFD scheduler is active, it sends a request to stop the KFD scheduler and sets the KFD scheduler state to inactive. Then, it increments the reference count. The function is synchronized using the kfd_sch_mutex to ensure that the KFD scheduler state and reference count are updated atomically. Cc: Christian König Cc: Alex Deucher Signed-off-by: Alex Deucher Signed-off-by: Srinivasan Shanmugam Suggested-by: Christian König Suggested-by: Alex Deucher

drm/amdgpu: Add enforce_isolation sysfs attribute

2024-08-21T02:06:52Z

This commit adds a new sysfs attribute 'enforce_isolation' to control the 'enforce_isolation' setting per GPU. The attribute can be read and written, and accepts values 0 (disabled) and 1 (enabled). When 'enforce_isolation' is enabled, reserved VMIDs are allocated for each ring. When it's disabled, the reserved VMIDs are freed. The set function locks a mutex before changing the 'enforce_isolation' flag and the VMIDs, and unlocks it afterwards. This ensures that these operations are atomic and prevents race conditions and other concurrency issues. Cc: Christian König Cc: Alex Deucher Signed-off-by: Srinivasan Shanmugam Suggested-by: Alex Deucher Signed-off-by: Alex Deucher

drm/amdgpu: Make enforce_isolation setting per GPU

2024-08-16T18:27:45Z

This commit makes enforce_isolation setting to be per GPU and per partition by adding the enforce_isolation array to the adev structure. The adev variable is set based on the global enforce_isolation module parameter during device initialization. In amdgpu_ids.c, the adev->enforce_isolation value for the current GPU is used to determine whether to enforce isolation between graphics and compute processes on that GPU. In amdgpu_ids.c, the adev->enforce_isolation value for the current GPU and partition is used to determine whether to enforce isolation between graphics and compute processes on that GPU and partition. This allows the enforce_isolation setting to be controlled individually for each GPU and each partition, which is useful in a system with multiple GPUs and partitions where different isolation settings might be desired for different GPUs and partitions. v2: fix loop in amdgpu_vmid_mgr_init() (Alex) Cc: Christian König Cc: Alex Deucher Signed-off-by: Srinivasan Shanmugam Signed-off-by: Alex Deucher Suggested-by: Christian König

drm/amdgpu/gfx11: add a mutex for the gfx semaphore

2024-08-16T18:24:33Z

This will be used in more places in the future so add a mutex. Acked-by: Vitaly Prosyak Signed-off-by: Alex Deucher