diff options
| author | Sunil Khatri <sunil.khatri@amd.com> | 2026-05-18 19:58:08 +0530 |
|---|---|---|
| committer | Alex Deucher <alexander.deucher@amd.com> | 2026-05-27 11:58:31 -0400 |
| commit | cf4aafdccefccc7f8236fed028d06725246e289e (patch) | |
| tree | 8022e148e5b6dad4fd6d9dcc3435a23e1c092952 | |
| parent | a00caed2302c604c19a5cab781e34d7ba4fa7558 (diff) | |
drm/amdgpu/userq: make sure queue is valid in the hang_detect_work
Thread 1: Running amdgpu_userq_destroy which eventually remove
the queue from door bell and set userq_mgr = NULL.
Thread2: An interrupt might have scheduled the hang_detect_work
which still need userq_mgr to be valid but could get an NULL
ptrs.
To fix that make sure we cancel the hang_detect_work again before
setting userq_mgr to NULL.
Along with that we also need all the queue va to remain valid till
we could be running anything on the queue and hence moving the
userq_va post hang_detect handler is cancelled.
Signed-off-by: Sunil Khatri <sunil.khatri@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 1a66ceb98b137d18d303b9889f0e7d8c4db73943)
| -rw-r--r-- | drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 14 |
1 files changed, 7 insertions, 7 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index 5da107dffcce..8d7dad1d30eb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -427,8 +427,6 @@ static void amdgpu_userq_cleanup(struct amdgpu_usermode_queue *queue) xa_erase_irq(&adev->userq_doorbell_xa, queue->doorbell_index); amdgpu_userq_fence_driver_free(queue); queue->fence_drv = NULL; - queue->userq_mgr = NULL; - list_del(&queue->userq_va_list); up_read(&adev->reset_domain->sem); } @@ -619,11 +617,6 @@ amdgpu_userq_destroy(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_usermode_que /* Cancel any pending hang detection work and cleanup */ cancel_delayed_work_sync(&queue->hang_detect_work); - - amdgpu_bo_reserve(vm->root.bo, true); - amdgpu_userq_buffer_vas_list_cleanup(adev, queue); - amdgpu_bo_unreserve(vm->root.bo); - mutex_lock(&uq_mgr->userq_mutex); amdgpu_userq_wait_for_last_fence(queue); @@ -635,6 +628,13 @@ amdgpu_userq_destroy(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_usermode_que amdgpu_userq_cleanup(queue); mutex_unlock(&uq_mgr->userq_mutex); + cancel_delayed_work_sync(&queue->hang_detect_work); + amdgpu_bo_reserve(vm->root.bo, true); + amdgpu_userq_buffer_vas_list_cleanup(adev, queue); + amdgpu_bo_unreserve(vm->root.bo); + list_del(&queue->userq_va_list); + queue->userq_mgr = NULL; + amdgpu_bo_reserve(queue->db_obj.obj, true); amdgpu_bo_unpin(queue->db_obj.obj); amdgpu_bo_unreserve(queue->db_obj.obj); |