From b8e6d3f68c3bd1ac54492e210ece87475e7f862b Mon Sep 17 00:00:00 2001 From: Shashank Sharma Date: Wed, 20 Nov 2024 18:04:33 +0100 Subject: [PATCH 3091/5403] drm/amdgpu: handle eviction fence race MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The eviction process can get into a race condition between the eviction fence suspend work (which replaces the old fence with new) and kms_close (which destroys the fence and doesn't expect a new one). This patch: - adds a flag to indicate that fd is closing, so fence replacement is not required (evf_mgr->fd_closing) - adds a flush_work() during the ev_fence_destroy routine V2: Addressed review comments from Christian: - Do not use mutex to sync - Use flush_work and wait for suspend_work to be done V3: Fixed state machine for queue->active, which adds into race between suspend/resume and queue ops Cc: Alex Deucher Cc: Christian König Reviewed-by: Christian König Signed-off-by: Shashank Sharma Signed-off-by: Arvind Yadav Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_eviction_fence.c | 7 +++++++ drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 4 +++- drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c | 3 ++- drivers/gpu/drm/amd/amdgpu/mes_v11_0_userqueue.c | 9 +++++---- 4 files changed, 17 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_eviction_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_eviction_fence.c index 189afb872775..c22767a75348 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_eviction_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_eviction_fence.c @@ -117,6 +117,10 @@ amdgpu_eviction_fence_suspend_worker(struct work_struct *work) /* Signal old eviction fence */ amdgpu_eviction_fence_signal(evf_mgr); + /* Do not replace eviction fence is fd is getting closed */ + if (evf_mgr->fd_closing) + return; + /* Prepare the objects to replace eviction fence */ drm_exec_init(&exec, DRM_EXEC_IGNORE_DUPLICATES, 0); drm_exec_until_all_locked(&exec) { @@ -199,6 +203,9 @@ void amdgpu_eviction_fence_destroy(struct amdgpu_eviction_fence_mgr *evf_mgr) { struct amdgpu_eviction_fence *ev_fence; + /* Wait for any pending work to execute */ + flush_delayed_work(&evf_mgr->suspend_work); + spin_lock(&evf_mgr->ev_fence_lock); ev_fence = evf_mgr->ev_fence; spin_unlock(&evf_mgr->ev_fence_lock); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 700442584f97..f8370da08038 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -1490,10 +1490,12 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev, amdgpu_bo_unreserve(pd); } + fpriv->evf_mgr.fd_closing = true; + amdgpu_userq_mgr_fini(&fpriv->userq_mgr); amdgpu_eviction_fence_destroy(&fpriv->evf_mgr); + amdgpu_ctx_mgr_fini(&fpriv->ctx_mgr); amdgpu_vm_fini(adev, &fpriv->vm); - amdgpu_userq_mgr_fini(&fpriv->userq_mgr); if (pasid) amdgpu_pasid_free_delayed(pd->tbo.base.resv, pasid); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c index 57e502d014a1..cba51bdf2e2c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userqueue.c @@ -614,9 +614,10 @@ void amdgpu_userq_mgr_fini(struct amdgpu_userq_mgr *userq_mgr) cancel_delayed_work(&userq_mgr->resume_work); + mutex_lock(&userq_mgr->userq_mutex); idr_for_each_entry(&userq_mgr->userq_idr, queue, queue_id) amdgpu_userqueue_cleanup(userq_mgr, queue, queue_id); - idr_destroy(&userq_mgr->userq_idr); + mutex_unlock(&userq_mgr->userq_mutex); mutex_destroy(&userq_mgr->userq_mutex); } diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0_userqueue.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0_userqueue.c index ee0a757fcfa3..b1b7bc47d39f 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0_userqueue.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0_userqueue.c @@ -139,6 +139,7 @@ static int mes_v11_0_userq_map(struct amdgpu_userq_mgr *uq_mgr, return r; } + queue->queue_active = true; DRM_DEBUG_DRIVER("Queue (doorbell:%d) mapped successfully\n", userq_props->doorbell_index); return 0; } @@ -160,6 +161,7 @@ static void mes_v11_0_userq_unmap(struct amdgpu_userq_mgr *uq_mgr, amdgpu_mes_unlock(&adev->mes); if (r) DRM_ERROR("Failed to unmap queue in HW, err (%d)\n", r); + queue->queue_active = false; } static int mes_v11_0_userq_create_ctx_space(struct amdgpu_userq_mgr *uq_mgr, @@ -331,7 +333,6 @@ static int mes_v11_0_userq_mqd_create(struct amdgpu_userq_mgr *uq_mgr, goto free_ctx; } - queue->queue_active = true; return 0; free_ctx: @@ -350,12 +351,12 @@ static void mes_v11_0_userq_mqd_destroy(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_usermode_queue *queue) { - mes_v11_0_userq_unmap(uq_mgr, queue); - amdgpu_bo_unref(&queue->wptr_obj.obj); + if (queue->queue_active) + mes_v11_0_userq_unmap(uq_mgr, queue); + amdgpu_userqueue_destroy_object(uq_mgr, &queue->fw_obj); kfree(queue->userq_prop); amdgpu_userqueue_destroy_object(uq_mgr, &queue->mqd); - queue->queue_active = false; } static int mes_v11_0_userq_suspend(struct amdgpu_userq_mgr *uq_mgr, -- 2.54.0