kernel/linux.git/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c, branch v6.12.107

drm/amdgpu: move debug_vm handling to amdgpu_cs_parser_fini

2026-08-23T12:22:48+00:00

[ Upstream commit d8726ef11512754a68c0ab53c57634a569b8feff ] The commit referenced below restarts the CS if the validation is still in progress. When debug_vm is enabled, all BOs from the CS are invalidated so we will hit an infinite loop. To avoid that, defer BO invalidation to amdgpu_cs_parser_fini. Fixes: 59720bfd8c6d ("drm/amdgpu: restart the CS if some parts of the VM are still invalidated") Signed-off-by: Pierre-Eric Pelloux-Prayer Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit 8c990ee9daa295462df24982ce6878db997a380a) Cc: stable@vger.kernel.org Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: remove unused function parameter

2026-08-23T12:22:48+00:00

[ Upstream commit a541a6e865ecd8dfd8df6eeb134cc20e7139d329 ] amdgpu_vm_bo_invalidate doesn't use the adev parameter and not all callers have a reference to adev handy, so remove it for cleanliness. Signed-off-by: Yunxiang Li Reviewed-by: Christian König Link: https://patchwork.freedesktop.org/patch/msgid/20241219151411.1150-5-Yunxiang.Li@amd.com Signed-off-by: Christian König Stable-dep-of: d8726ef11512 ("drm/amdgpu: move debug_vm handling to amdgpu_cs_parser_fini") Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: disallow multiple FENCE chunks in one submit

2026-08-23T12:22:41+00:00

commit 931cd1d1baeae68e8eb2c23bc1f3d8934dca6241 upstream. amdgpu_cs_pass1() dispatches on chunk_id once per chunk without rejecting repeated ids. p->uf_bo is a single-slot field, so a submission carrying two AMDGPU_CHUNK_ID_FENCE chunks runs amdgpu_cs_p1_user_fence() twice, and the second run overwrites p->uf_bo with a freshly referenced BO without dropping the reference taken by the first. amdgpu_cs_parser_fini() only unrefs the final p->uf_bo, so every FENCE chunk but the last leaks a BO reference. The leaked BO outlives handle close and process exit. Reject duplicate FENCE chunks the same way commit fec5f8e8c6bc ("drm/amdgpu: disallow multiple BO_HANDLES chunks in one submit") did for p->bo_list. Fixes: d38ceaf99ed0 ("drm/amdgpu: add core driver (v4)") Reported-by: Yuhao Jiang Assisted-by: Claude:claude-opus-5 Cc: stable@vger.kernel.org Reviewed-by: Christian König Signed-off-by: Junrui Luo Signed-off-by: Alex Deucher (cherry picked from commit 665b1fc2a1845206408f9a2c6da67101789edb82) Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: reject oversized IBs with per-ring packet limits

2026-08-23T12:22:37+00:00

commit fd37f9dd5b5ab70a46fa7bc76623c0528d602b27 upstream. On GFX rings, amdgpu_cs_p2_ib() passed user-supplied ib_bytes through to ib->length_dw without a limit, while ring_emit_ib() encodes length into packet fields. Oversized values can corrupt adjacent control bits and destabilize command submission. Add a per-ring IB packet size limit helper and reject command submissions exceeding the corresponding dword limit before IB allocation. Use the documented 20-bit limit for GFX/compute/SDMA/VPE, and apply the MM fallback limit for other ring types. Signed-off-by: Candice Li Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit 7f48fa2cf62e3fa6c9c3870aa74988f773247e52) Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman

drm/exec: Remove the index parameter from drm_exec_for_each_locked_obj[_reverse]

2026-08-09T18:23:26+00:00

[ Upstream commit ce44b78512e9102aea54ff6b6e521d6c8de9f31c ] Nobody makes any use of it. Possible internal future users can instead use the _index variable. External users shouldn't use it since the array it's pointing into is internal drm_exec state. v2: - Use a unique id for the loop variable (Christian) Assisted-by: GitHub Copilot:claude-sonnet-4.6 Signed-off-by: Thomas Hellström Reviewed-by: Christian König Link: https://patch.msgid.link/20260520101616.41284-2-thomas.hellstrom@linux.intel.com Stable-dep-of: af80e2bfde93 ("drm/xe: Wait on external BO kernel fences in exec IOCTL") Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

drm/sched: Store the drm client_id in drm_sched_fence

2026-08-09T18:23:25+00:00

[ Upstream commit 2956554823cedb390b7ec4534afa898176317638 ] This will be used in a later commit to trace the drm client_id in some of the gpu_scheduler trace events. This requires changing all the users of drm_sched_job_init to add an extra parameter. The newly added drm_client_id field in the drm_sched_fence is a bit of a duplicate of the owner one. One suggestion I received was to merge those 2 fields - this can't be done right now as amdgpu uses some special values (AMDGPU_FENCE_OWNER_*) that can't really be translated into a client id. Christian is working on getting rid of those; when it's done we should be able to squash owner/drm_client_id together. Reviewed-by: Christian König Signed-off-by: Pierre-Eric Pelloux-Prayer Signed-off-by: Philipp Stanner Link: https://lore.kernel.org/r/20250526125505.2360-3-pierre-eric.pelloux-prayer@amd.com Stable-dep-of: 3e864bf2a32a ("drm/amdgpu/gfx: fix cleaner shader IB buffer overflow") Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: validate CP_GFX_SHADOW chunk size in CS pass1

2026-08-03T09:17:16+00:00

commit 84c4c36acd5c4b2558b5069f869a165b2c655c84 upstream. Add a minimum-length check for the AMDGPU_CHUNK_ID_CP_GFX_SHADOW chunk in amdgpu_cs_pass1(), matching the gate already present for the IB, FENCE and BO_HANDLES chunk types. The CP_GFX_SHADOW case previously shared a bare break with the dependency and syncobj chunk types, which do not dereference a fixed-size struct. When userspace submits this chunk with length_dw == 0, vmemdup_array_user() is called with size 0 and returns ZERO_SIZE_PTR, which passes the IS_ERR() check. amdgpu_cs_p2_shadow() then dereferences chunk->kdata as a struct drm_amdgpu_cs_chunk_cp_gfx_shadow (reading shadow->flags), faulting on the ZERO_SIZE_PTR and causing a NULL-pointer dereference. This is reachable by an unprivileged process in the render group. Reject undersized chunks with -EINVAL during pass1 so the bad submission is rejected before pass2 ever dereferences the data. Fixes: ac9287055ff1 ("drm/amdgpu: add gfx shadow CS IOCTL support") Reviewed-by: Alex Deucher Signed-off-by: Mario Limonciello Signed-off-by: Alex Deucher (cherry picked from commit 7f61b2eef7415eccdb40850aca0de94211948657) Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: restart the CS if some parts of the VM are still invalidated

2026-06-19T11:42:35+00:00

commit 40396ffdf6120e2380706c59e1a84d7e765a37b6 upstream. Make sure that we only submit work with full up to date VM page tables. Backport to 7.1 and older. Signed-off-by: Christian König Reviewed-by: Vitaly Prosyak Tested-by: Vitaly Prosyak Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit 59720bfd8c6dbebeb8d5a7ab64241b007efd9213) Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: Fix NULL pointer dereference in VRAM logic for APU devices

2025-11-24T09:35:47+00:00

[ Upstream commit 883f309add55060233bf11c1ea6947140372920f ] Previously, APU platforms (and other scenarios with uninitialized VRAM managers) triggered a NULL pointer dereference in `ttm_resource_manager_usage()`. The root cause is not that the `struct ttm_resource_manager *man` pointer itself is NULL, but that `man->bdev` (the backing device pointer within the manager) remains uninitialized (NULL) on APUs—since APUs lack dedicated VRAM and do not fully set up VRAM manager structures. When `ttm_resource_manager_usage()` attempts to acquire `man->bdev->lru_lock`, it dereferences the NULL `man->bdev`, leading to a kernel OOPS. 1. **amdgpu_cs.c**: Extend the existing bandwidth control check in `amdgpu_cs_get_threshold_for_moves()` to include a check for `ttm_resource_manager_used()`. If the manager is not used (uninitialized `bdev`), return 0 for migration thresholds immediately—skipping VRAM-specific logic that would trigger the NULL dereference. 2. **amdgpu_kms.c**: Update the `AMDGPU_INFO_VRAM_USAGE` ioctl and memory info reporting to use a conditional: if the manager is used, return the real VRAM usage; otherwise, return 0. This avoids accessing `man->bdev` when it is NULL. 3. **amdgpu_virt.c**: Modify the vf2pf (virtual function to physical function) data write path. Use `ttm_resource_manager_used()` to check validity: if the manager is usable, calculate `fb_usage` from VRAM usage; otherwise, set `fb_usage` to 0 (APUs have no discrete framebuffer to report). This approach is more robust than APU-specific checks because it: - Works for all scenarios where the VRAM manager is uninitialized (not just APUs), - Aligns with TTM's design by using its native helper function, - Preserves correct behavior for discrete GPUs (which have fully initialized `man->bdev` and pass the `ttm_resource_manager_used()` check). v4: use ttm_resource_manager_used(&adev->mman.vram_mgr.manager) instead of checking the adev->gmc.is_app_apu flag (Christian) Reviewed-by: Christian König Suggested-by: Lijo Lazar Signed-off-by: Jesse Zhang Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdgpu: reject gang submissions under SRIOV

2025-11-13T20:34:25+00:00

[ Upstream commit d7ddcf921e7d0d8ebe82e89635bc9dc26ba9540d ] Gang submission means that the kernel driver guarantees that multiple submissions are executed on the HW at the same time on different engines. Background is that those submissions then depend on each other and each can't finish stand alone. SRIOV now uses world switch to preempt submissions on the engines to allow sharing the HW resources between multiple VFs. The problem is now that the SRIOV world switch can't know about such inter dependencies and will cause a timeout if it waits for a partially running gang submission. To conclude SRIOV and gang submissions are fundamentally incompatible at the moment. For now just disable them. Signed-off-by: Christian König Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin