Commit 74097f9f authored by Philip Yang's avatar Philip Yang Committed by Alex Deucher

drm/amdkfd: Process notifier release callback don't take mutex

Move process queues cleanup to deferred work kfd_process_wq_release, to
avoid potential deadlock circular locking warning:

 WARNING: possible circular locking dependency detected
               the existing dependency chain (in reverse order) is:
      -> #2
        ((work_completion)(&svms->deferred_list_work)){+.+.}-{0:0}:
        __flush_work+0x343/0x4a0
        svm_range_list_lock_and_flush_work+0x39/0xc0
        svm_range_set_attr+0xe8/0x1080 [amdgpu]
        kfd_ioctl+0x19b/0x600 [amdgpu]
        __x64_sys_ioctl+0x81/0xb0
        do_syscall_64+0x34/0x80
        entry_SYSCALL_64_after_hwframe+0x44/0xae

      -> #1 (&info->lock#2){+.+.}-{3:3}:
        __mutex_lock+0xa4/0x940
        amdgpu_amdkfd_gpuvm_acquire_process_vm+0x2e3/0x590
        kfd_process_device_init_vm+0x61/0x200 [amdgpu]
        kfd_ioctl_acquire_vm+0x83/0xb0 [amdgpu]
        kfd_ioctl+0x19b/0x600 [amdgpu]
        __x64_sys_ioctl+0x81/0xb0
        do_syscall_64+0x34/0x80
       entry_SYSCALL_64_after_hwframe+0x44/0xae

      -> #0 (&process->mutex){+.+.}-{3:3}:
        __lock_acquire+0x1365/0x23d0
        lock_acquire+0xc9/0x2e0
        __mutex_lock+0xa4/0x940
        kfd_process_notifier_release+0x96/0xe0 [amdgpu]
        __mmu_notifier_release+0x94/0x210
        exit_mmap+0x35/0x1f0
        mmput+0x63/0x120
        svm_range_deferred_list_work+0x177/0x2c0 [amdgpu]
        process_one_work+0x2a4/0x600
        worker_thread+0x39/0x3e0
        kthread+0x16d/0x1a0

  Possible unsafe locking scenario:

      CPU0                    CPU1
        ----                    ----
   lock((work_completion)(&svms->deferred_list_work));
                                lock(&info->lock#2);
             lock((work_completion)(&svms->deferred_list_work));
   lock(&process->mutex);
Signed-off-by: default avatarPhilip Yang <Philip.Yang@amd.com>
Reviewed-by: default avatarFelix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: default avatarAlex Deucher <alexander.deucher@amd.com>
parent 0593ad21
...@@ -1115,6 +1115,15 @@ static void kfd_process_wq_release(struct work_struct *work) ...@@ -1115,6 +1115,15 @@ static void kfd_process_wq_release(struct work_struct *work)
struct kfd_process *p = container_of(work, struct kfd_process, struct kfd_process *p = container_of(work, struct kfd_process,
release_work); release_work);
kfd_process_dequeue_from_all_devices(p);
pqm_uninit(&p->pqm);
/* Signal the eviction fence after user mode queues are
* destroyed. This allows any BOs to be freed without
* triggering pointless evictions or waiting for fences.
*/
dma_fence_signal(p->ef);
kfd_process_remove_sysfs(p); kfd_process_remove_sysfs(p);
kfd_iommu_unbind_process(p); kfd_iommu_unbind_process(p);
...@@ -1179,20 +1188,8 @@ static void kfd_process_notifier_release(struct mmu_notifier *mn, ...@@ -1179,20 +1188,8 @@ static void kfd_process_notifier_release(struct mmu_notifier *mn,
cancel_delayed_work_sync(&p->eviction_work); cancel_delayed_work_sync(&p->eviction_work);
cancel_delayed_work_sync(&p->restore_work); cancel_delayed_work_sync(&p->restore_work);
mutex_lock(&p->mutex);
kfd_process_dequeue_from_all_devices(p);
pqm_uninit(&p->pqm);
/* Indicate to other users that MM is no longer valid */ /* Indicate to other users that MM is no longer valid */
p->mm = NULL; p->mm = NULL;
/* Signal the eviction fence after user mode queues are
* destroyed. This allows any BOs to be freed without
* triggering pointless evictions or waiting for fences.
*/
dma_fence_signal(p->ef);
mutex_unlock(&p->mutex);
mmu_notifier_put(&p->mmu_notifier); mmu_notifier_put(&p->mmu_notifier);
} }
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment