From: Karolina Stolarek karolina.stolarek@intel.com
[ Upstream commit 3b401e30c249849d803de6c332dad2a595a58658 ]
With the current cleanup flow, we could trigger a NULL pointer dereference if there is a delayed destruction of a BO with a system resource that gets executed on drain_workqueue() call, as we attempt to free a resource using an already released resource manager.
Remove the device from the device list and drain its workqueue before releasing the system domain manager in ttm_device_fini().
Signed-off-by: Karolina Stolarek karolina.stolarek@intel.com Reviewed-by: Christian König christian.koenig@amd.com Link: https://patchwork.freedesktop.org/patch/msgid/20231016121525.2237838-1-karol... Signed-off-by: Christian König christian.koenig@amd.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/gpu/drm/ttm/ttm_device.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/ttm/ttm_device.c b/drivers/gpu/drm/ttm/ttm_device.c index 2df59b3c2ea16..291cdd893fc50 100644 --- a/drivers/gpu/drm/ttm/ttm_device.c +++ b/drivers/gpu/drm/ttm/ttm_device.c @@ -234,10 +234,6 @@ void ttm_device_fini(struct ttm_device *bdev) struct ttm_resource_manager *man; unsigned i;
- man = ttm_manager_type(bdev, TTM_PL_SYSTEM); - ttm_resource_manager_set_used(man, false); - ttm_set_driver_manager(bdev, TTM_PL_SYSTEM, NULL); - mutex_lock(&ttm_global_mutex); list_del(&bdev->device_list); mutex_unlock(&ttm_global_mutex); @@ -247,6 +243,10 @@ void ttm_device_fini(struct ttm_device *bdev) if (ttm_bo_delayed_delete(bdev, true)) pr_debug("Delayed destroy list was clean\n");
+ man = ttm_manager_type(bdev, TTM_PL_SYSTEM); + ttm_resource_manager_set_used(man, false); + ttm_set_driver_manager(bdev, TTM_PL_SYSTEM, NULL); + spin_lock(&bdev->lru_lock); for (i = 0; i < TTM_MAX_BO_PRIORITY; ++i) if (list_empty(&man->lru[0]))