On 25/03/2021 15:27, Christian König wrote:
The BO might be NULL in this function, use the bdev directly.
Signed-off-by: Christian König christian.koenig@amd.com Reported-by: Colin Ian King colin.king@canonical.com Fixes: a1f091f8ef2b ("drm/ttm: switch to per device LRU lock")
drivers/gpu/drm/ttm/ttm_bo.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index 2d2ac532987e..6ab7b66ce36d 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -625,7 +625,7 @@ int ttm_mem_evict_first(struct ttm_device *bdev, unsigned i; int ret;
- spin_lock(&bo->bdev->lru_lock);
- spin_lock(&bdev->lru_lock); for (i = 0; i < TTM_MAX_BO_PRIORITY; ++i) { list_for_each_entry(bo, &man->lru[i], lru) { bool busy;
@@ -662,7 +662,7 @@ int ttm_mem_evict_first(struct ttm_device *bdev, if (!bo) { if (busy_bo && !ttm_bo_get_unless_zero(busy_bo)) busy_bo = NULL;
spin_unlock(&bo->bdev->lru_lock);
spin_unlock(&bdev->lru_lock);@@ -676,7 +676,7 @@ int ttm_mem_evict_first(struct ttm_device *bdev, return ret; }
- spin_unlock(&bo->bdev->lru_lock);
spin_unlock(&bdev->lru_lock);
ret = ttm_bo_evict(bo, ctx); if (locked)
Looks good to me. Thanks.
Reviewed-by: Colin Ian King colin.king@canonical.com