kernel/linux.git/drivers/gpu/drm/amd/amdkfd, branch v5.12.1

drm/amdkfd: dqm fence memory corruption

2021-04-01T01:53:25+00:00

Amdgpu driver uses 4-byte data type as DQM fence memory, and transmits GPU address of fence memory to microcode through query status PM4 message. However, query status PM4 message definition and microcode processing are all processed according to 8 bytes. Fence memory only allocates 4 bytes of memory, but microcode does write 8 bytes of memory, so there is a memory corruption. Changes since v1: * Change dqm->fence_addr as a u64 pointer to fix this issue, also fix up query_status and amdkfd_fence_wait_timeout function uses 64 bit fence value to make them consistent. Signed-off-by: Qu Huang Reviewed-by: Felix Kuehling Signed-off-by: Felix Kuehling Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org

drm/amdkfd: Fix recursive lock warnings

2021-02-18T21:43:09+00:00

memalloc_nofs_save/restore are no longer sufficient to prevent recursive lock warnings when holding locks that can be taken in MMU notifiers. Use memalloc_noreclaim_save/restore instead. Fixes: f920e413ff9c ("mm: track mmu notifiers in fs_reclaim_acquire/release") CC: Daniel Vetter Reviewed-by: Philip Yang Signed-off-by: Felix Kuehling Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org # 5.10.x

drm/amdkfd: Get unique_id dynamically v2

2021-02-09T20:27:28+00:00

Instead of caching the value during amdgpu_device_init, just call the function directly. This avoids issues where the unique_id hasn't been saved by the time that KFD's topology snapshot is done (e.g. Arcturus). KFD's topology information from the amdgpu_device was initially cached at KFD initialization due to amdkfd and amdgpu being separate modules. Now that they are combined together, we can directly call the functions that we need and avoid this unnecessary duplication and complexity. As a side-effect of this change, we also remove unique_id=0 for CPUs, which is obviously not unique. v2: Drop previous patch printing unique_id in hex Signed-off-by: Kent Russell Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdkfd: Enable userptr support when KFD is enabled

2021-01-21T14:55:16+00:00

ROCm user mode depends on userptr support. Without it, KFD is basically useless. Make sure HSA_AMD selects the same options as DRM_AMDGPU_USERPTR to avoid broken configurations where userptr gets enabled but its dependencies are disabled. Signed-off-by: Felix Kuehling Reviewed-by: Philip Yang Signed-off-by: Alex Deucher

drm/amdkfd: Fix out-of-bounds read in kdf_create_vcrat_image_cpu()

2021-01-14T04:46:34+00:00

KASAN reported a slab-out-of-bounds read of size 1 in kdf_create_vcrat_image_cpu(). This occurs when, for example, when on an x86_64 with a single NUMA node because kfd_fill_iolink_info_for_cpu() is a no-op, but afterwards the sub_type_hdr->length, which is out-of-bounds, is read and multiplied by entries. Fortunately, entries is 0 in this case so the overall crat_table->length is still correct. Check if there were any entries before de-referencing sub_type_hdr which may be pointing to out-of-bounds memory. Fixes: b7b6c38529c9 ("drm/amdkfd: Calculate CPU VCRAT size dynamically (v2)") Suggested-by: Felix Kuehling Signed-off-by: Jeremy Cline Reviewed-by: Felix Kuehling Signed-off-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdkfd: check more client ids in interrupt handler

2021-01-08T20:18:25+00:00

Add check for SExSH clients in kfd interrupt handler. Signed-off-by: Tao Zhou Reviewed-by: Alex Deucher Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdkfd: check both client id and src id in interrupt handlers

2020-12-23T20:07:40+00:00

We can have the same src ids for different client ids so make sure to check both the client id and the source id when handling interrupts. Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdkfd: PCIe atomics required for gfx10

2020-12-17T21:43:14+00:00

GFX10 CP firmware expects PCIe atomics support. Don't enumerate GFX10 devices on platforms (PCIe v2) that don't support PCIe atomics. Currently, some of the applications like clinfo soft hangs on platforms without PCIe atomics support. Signed-off-by: Harish Kasiviswanathan Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdgpu: Fix spelling mistake "Heterogenous" -> "Heterogeneous"

2020-12-15T16:35:53+00:00

There is a spelling mistake in a comment in the Kconfig. Fix it. Signed-off-by: Colin Ian King Signed-off-by: Alex Deucher

drm/amdkfd: correct pipe offset calculation

2020-12-10T21:41:49+00:00

Correct pipe offset calculation in is_pipe_enabled function, it should be done in queues. Signed-off-by: Jiansong Chen Reviewed-by: Hawking Zhang Signed-off-by: Alex Deucher