diff options
| author | Maarten Lankhorst <dev@lankhorst.se> | 2026-01-08 11:10:20 +0100 |
|---|---|---|
| committer | Maarten Lankhorst <dev@lankhorst.se> | 2026-01-12 16:28:47 +0100 |
| commit | 9086170bfb925c90e298134791a9851ed626b873 (patch) | |
| tree | ad0600ba947042511f72b97c234c02b968f0adb3 /drivers/gpu | |
| parent | a7ae0836917ca617434c416f9c61dc6024d04949 (diff) | |
drm/xe: Do not dereference ggtt_node in xe_bo.c
A careful inspection of __xe_ggtt_insert_bo_at() shows that
the ggtt_node can always be seen as inserted from xe_bo.c
due to the way error handling is performed.
The checks are also a little bit too paranoid, since we
never create a bo with ggtt_node[id] initialised but not
inserted into the GGTT, which can be seen by looking at
__xe_ggtt_insert_bo_at()
Additionally, the size of the GGTT is never bigger than 4 GB,
so adding a check at that level is incorrect.
Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patch.msgid.link/20260108101014.579906-14-dev@lankhorst.se
Diffstat (limited to 'drivers/gpu')
| -rw-r--r-- | drivers/gpu/drm/xe/xe_bo.c | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c index 408c74216fdf..917e50c276ac 100644 --- a/drivers/gpu/drm/xe/xe_bo.c +++ b/drivers/gpu/drm/xe/xe_bo.c @@ -1715,7 +1715,7 @@ static void xe_ttm_bo_destroy(struct ttm_buffer_object *ttm_bo) xe_assert(xe, list_empty(&ttm_bo->base.gpuva.list)); for_each_tile(tile, xe, id) - if (bo->ggtt_node[id] && bo->ggtt_node[id]->base.size) + if (bo->ggtt_node[id]) xe_ggtt_remove_bo(tile->mem.ggtt, bo); #ifdef CONFIG_PROC_FS @@ -3609,8 +3609,8 @@ void xe_bo_put(struct xe_bo *bo) might_lock(&bo->client->bos_lock); #endif for_each_tile(tile, xe_bo_device(bo), id) - if (bo->ggtt_node[id] && bo->ggtt_node[id]->ggtt) - xe_ggtt_might_lock(bo->ggtt_node[id]->ggtt); + if (bo->ggtt_node[id]) + xe_ggtt_might_lock(tile->mem.ggtt); drm_gem_object_put(&bo->ttm.base); } } |