linux/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c, branch v6.4

drm/amdgpu: DROP redundant drm_prime_sg_to_dma_addr_array

2023-04-11T22:03:44Z

For DMA-MAP userptr on other GPUs, the dma address array will be populated in amdgpu_ttm_backend_bind. Remove the redundant call from the driver. v2: update the comment Signed-off-by: Shane Xiao Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher

amd/amdgpu: Inherit coherence flags base on original BO flags

2023-04-11T22:03:44Z

For SG BO to DMA-map userptrs on other GPUs, the SG BO need inherit MTYPEs in PTEs from original BO. If we set the flags, the device can be coherent with the CPUs and other GPUs. v2: 1. Drop unnecessary flags check 2. Remove local variable align Signed-off-by: Shane Xiao Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdgpu: Add userptr bo support for mGPUs when iommu is on

2023-04-11T22:03:44Z

For userptr bo with iommu on, multiple GPUs use same system memory dma mapping address when both adev and bo_adev are in identity mode or in the same iommu group. If RAM direct map to one GPU, other GPUs can share the original BO in order to reduce dma address array usage when RAM can direct map to these GPUs. However, we should explicit check whether RAM can direct map to all these GPUs. This patch fixes a potential issue that where RAM is direct mapped on some but not all GPUs. v2: 1. Update comment 2. Add helper function reuse_dmamap Signed-off-by: Shane Xiao Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher

amdkfd: Memory availability can never be negative

2023-03-07T19:21:56Z

Our assumptions about how much KFD memory is currently available for allocation may be violated by various complexities so we define the reported value as advisory, however we should never report negative availability. Signed-off-by: Daniel Phillips Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdkfd: Implement DMA buf fd export from KFD

2023-02-28T19:30:00Z

Exports a DMA buf fd of a given KFD buffer handle. This is intended for being able to import KFD BOs into GEM contexts to leverage the amdgpu_bo_va API for more flexible virtual address mappings. It will also be used for the new upstreamable RDMA solution coming to UCX and RCCL. The corresponding user mode change (Thunk API and kfdtest) is here: https://github.com/fxkamd/ROCT-Thunk-Interface/commits/fxkamd/dmabuf Signed-off-by: Felix Kuehling Acked-by: Christian König Reviewed-by: Xiaogang Chen Signed-off-by: Alex Deucher

drm/amdgpu: Generalize KFD dmabuf import

2023-02-28T19:30:00Z

Use proper amdgpu_gem_prime_import function to handle all kinds of imports. Remember the dmabuf reference to enable proper multi-GPU attachment to multiple VMs without erroneously re-exporting the underlying BO multiple times. Signed-off-by: Felix Kuehling Acked-by: Christian König Signed-off-by: Alex Deucher

Merge tag 'amd-drm-next-6.3-2023-01-13' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

2023-01-16T05:04:13Z

amd-drm-next-6.3-2023-01-13: amdgpu: - Fix possible segfault in failure case - Rework FW requests to happen in early_init for all IPs so that we don't lose the sbios console if FW is missing - PSR fixes - Misc cleanups - Unload fix - SMU13 fixes amdkfd: - Fix for cleared VRAM BOs - Fix cleanup if GPUVM creation fails - Memory accounting fix - Use resource_size rather than open codeing it - GC11 mGPU fix radeon: - Fix memory leak on shutdown Signed-off-by: Dave Airlie From: Alex Deucher Link: https://patchwork.freedesktop.org/patch/msgid/20230113225911.7776-1-alexander.deucher@amd.com

drm/amdkfd: Page aligned memory reserve size

2023-01-11T21:41:03Z

Use page aligned size to reserve memory usage because page aligned TTM BO size is used to unreserve memory usage, otherwise no page aligned size causes memory usage accounting unbalanced. Change vram_used definition type to int64_t to be able to trigger WARN_ONCE(adev && adev->kfd.vram_used < 0, "..."), to help debug the accounting issue with warning and backtrace. Signed-off-by: Philip Yang Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdkfd: Cleanup vm process info if init vm failed

2023-01-11T21:40:45Z

If acquire_vm failed when initializing KFD vm, set vm->process_info to NULL and free process info, otherwise, the future acquire_vm will always fail as vm->process_info is not NULL. Pass avm as parameter to remove the duplicate code. Signed-off-by: Philip Yang Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdkfd: Fix NULL pointer error for GC 11.0.1 on mGPU

2023-01-10T23:05:03Z

The point bo->kfd_bo is NULL for queue's write pointer BO when creating queue on mGPU. To avoid using the pointer fixes the error. Signed-off-by: Eric Huang Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher