kernel/linux.git/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c, branch v6.1.168

drm/amdgpu: Fix fence put before wait in amdgpu_amdkfd_submit_ib

2026-04-11T12:16:12+00:00

[ Upstream commit 7150850146ebfa4ca998f653f264b8df6f7f85be ] amdgpu_amdkfd_submit_ib() submits a GPU job and gets a fence from amdgpu_ib_schedule(). This fence is used to wait for job completion. Currently, the code drops the fence reference using dma_fence_put() before calling dma_fence_wait(). If dma_fence_put() releases the last reference, the fence may be freed before dma_fence_wait() is called. This can lead to a use-after-free. Fix this by waiting on the fence first and releasing the reference only after dma_fence_wait() completes. Fixes the below: drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c:697 amdgpu_amdkfd_submit_ib() warn: passing freed memory 'f' (line 696) Fixes: 9ae55f030dc5 ("drm/amdgpu: Follow up change to previous drm scheduler change.") Cc: Felix Kuehling Cc: Dan Carpenter Cc: Christian König Cc: Alex Deucher Signed-off-by: Srinivasan Shanmugam Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit 8b9e5259adc385b61a6590a13b82ae0ac2bd3482) Signed-off-by: Sasha Levin

drm/amdgpu: cleanup scheduler job initialization v2

2026-01-11T14:19:22+00:00

[ Upstream commit f7d66fb2ea43a3016e78a700a2ca6c77a74579f9 ] Init the DRM scheduler base class while allocating the job. This makes the whole handling much more cleaner. v2: fix coding style Signed-off-by: Christian König Reviewed-by: Luben Tuikov Link: https://patchwork.freedesktop.org/patch/msgid/20221014084641.128280-7-christian.koenig@amd.com Stable-dep-of: 4fa944255be5 ("drm/amdgpu: add missing lock to amdgpu_ttm_access_memory_sdma") Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

drm/amdkfd: amdkfd_free_gtt_mem clear the correct pointer

2024-11-17T14:07:22+00:00

commit c86ad39140bbcb9dc75a10046c2221f657e8083b upstream. Pass pointer reference to amdgpu_bo_unref to clear the correct pointer, otherwise amdgpu_bo_unref clear the local variable, the original pointer not set to NULL, this could cause use-after-free bug. Signed-off-by: Philip Yang Reviewed-by: Felix Kuehling Acked-by: Christian König Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin Signed-off-by: Vamsi Krishna Brahmajosyula Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: disable GFXOFF during compute for GFX11

2022-11-02T21:16:11+00:00

Temporary workaround to fix issues observed in some compute applications when GFXOFF is enabled on GFX11. Signed-off-by: Graham Sider Acked-by: Alex Deucher Reviewed-by: Harish Kasiviswanathan Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org # 6.0.x

Revert "drm/amdgpu: let mode2 reset fallback to default when failure"

2022-10-19T02:08:33+00:00

This reverts commit dac6b80818ac2353631c5a33d140d8d5508e2957. This commit reverted the AMDGPU_SKIP_MODE2_RESET as it conflicts with the original design of reset handler. Will redesign it. Fixes: dac6b80818ac23 ("drm/amdgpu: let mode2 reset fallback to default when failure") Signed-off-by: Victor Zhao Reviewed-by: Lijo Lazar Signed-off-by: Alex Deucher

drm/amdgpu: Correct amdgpu_amdkfd_total_mem_size calculation

2022-10-06T16:08:18+00:00

amdkfd_total_mem_size is the size of total GPUs vram plus system memory to estimate page tables memory usage and leave enough VRAM room for page tables allocation. Calculate amdkfd_total_mem_size in amdgpu_amdkfd_device_probe is incorrect because adev->gmc.real_vram_size is still 0 called from amdgpu_device_ip_early_init. Move the calculation to amdgpu_amdkfd_device_init to get the correct VRAM size. Do reverse calculation in amdgpu_amdkfd_device_fini_sw to support hot-unplugging GPUs. Signed-off-by: Philip Yang Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher

drm/amdgpu: add page retirement handling for CPU RAS

2022-09-29T13:42:36+00:00

Do RAS page retirement in poison consumption handler unconditionally. Signed-off-by: Tao Zhou Reviewed-by: Hawking Zhang Signed-off-by: Alex Deucher

drm/amdgpu: add gang submit frontend v6

2022-09-20T16:40:46+00:00

Allows submitting jobs as gang which needs to run on multiple engines at the same time. All members of the gang get the same implicit, explicit and VM dependencies. So no gang member will start running until everything else is ready. The last job is considered the gang leader (usually a submission to the GFX ring) and used for signaling output dependencies. Each job is remembered individually as user of a buffer object, so there is no joining of work at the end. v2: rebase and fix review comments from Andrey and Yogesh v3: use READ instead of BOOKKEEP for now because of VM unmaps, set gang leader only when necessary v4: fix order of pushing jobs and adding fences found by Trigger. v5: fix job index calculation and adding IBs to jobs v6: fix typo found by Alex Signed-off-by: Christian König Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher

drm/amdgpu: cleanup coding style in amdgpu_amdkfd.c

2022-09-13T18:32:58+00:00

Fix everything checkpatch.pl complained about in amdgpu_amdkfd.c Signed-off-by: Jingyu Wang Signed-off-by: Alex Deucher

drm/amdgpu: let mode2 reset fallback to default when failure

2022-08-16T22:14:31+00:00

- introduce AMDGPU_SKIP_MODE2_RESET flag - let mode2 reset fallback to default reset method if failed v2: move this part out from the asic specific part Signed-off-by: Victor Zhao Acked-by: Andrey Grodzovsky Signed-off-by: Alex Deucher