[PATCH 2/2] drm/ttm: fix ttm_bo_cleanup_refs_or_queue once more
zhoucm1
david1.zhou at amd.com
Thu Sep 7 07:33:25 UTC 2017
Acked-by: Chunming Zhou <david1.zhou at amd.com>
On 2017年09月05日 03:02, Christian König wrote:
> From: Christian König <christian.koenig at amd.com>
>
> With shared reservation objects __ttm_bo_reserve() can easily fail even on
> destroyed BOs. This prevents correct handling when we need to individualize
> the reservation object.
>
> Fix this by individualizing the object before even trying to reserve it.
>
> Signed-off-by: Christian König <christian.koenig at amd.com>
> ---
> drivers/gpu/drm/ttm/ttm_bo.c | 32 +++++++++++++++++---------------
> 1 file changed, 17 insertions(+), 15 deletions(-)
>
> diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
> index 180ce62..bee77d3 100644
> --- a/drivers/gpu/drm/ttm/ttm_bo.c
> +++ b/drivers/gpu/drm/ttm/ttm_bo.c
> @@ -440,28 +440,29 @@ static void ttm_bo_cleanup_refs_or_queue(struct ttm_buffer_object *bo)
> struct ttm_bo_global *glob = bo->glob;
> int ret;
>
> + ret = ttm_bo_individualize_resv(bo);
> + if (ret) {
> + /* Last resort, if we fail to allocate memory for the
> + * fences block for the BO to become idle
> + */
> + reservation_object_wait_timeout_rcu(bo->resv, true, false,
> + 30 * HZ);
> + spin_lock(&glob->lru_lock);
> + goto error;
> + }
> +
> spin_lock(&glob->lru_lock);
> ret = __ttm_bo_reserve(bo, false, true, NULL);
> -
> if (!ret) {
> - if (!ttm_bo_wait(bo, false, true)) {
> + if (reservation_object_test_signaled_rcu(&bo->ttm_resv, true)) {
> ttm_bo_del_from_lru(bo);
> spin_unlock(&glob->lru_lock);
> + if (bo->resv != &bo->ttm_resv)
> + reservation_object_unlock(&bo->ttm_resv);
> ttm_bo_cleanup_memtype_use(bo);
> -
> return;
> }
>
> - ret = ttm_bo_individualize_resv(bo);
> - if (ret) {
> - /* Last resort, if we fail to allocate memory for the
> - * fences block for the BO to become idle and free it.
> - */
> - spin_unlock(&glob->lru_lock);
> - ttm_bo_wait(bo, true, true);
> - ttm_bo_cleanup_memtype_use(bo);
> - return;
> - }
> ttm_bo_flush_all_fences(bo);
>
> /*
> @@ -474,11 +475,12 @@ static void ttm_bo_cleanup_refs_or_queue(struct ttm_buffer_object *bo)
> ttm_bo_add_to_lru(bo);
> }
>
> - if (bo->resv != &bo->ttm_resv)
> - reservation_object_unlock(&bo->ttm_resv);
> __ttm_bo_unreserve(bo);
> }
> + if (bo->resv != &bo->ttm_resv)
> + reservation_object_unlock(&bo->ttm_resv);
>
> +error:
> kref_get(&bo->list_kref);
> list_add_tail(&bo->ddestroy, &bdev->ddestroy);
> spin_unlock(&glob->lru_lock);
More information about the amd-gfx
mailing list