[CI] drm/i915: Fix assert in i915_ggtt_pin

* [CI] drm/i915: Fix assert in i915_ggtt_pin
@ 2022-04-29 14:07 ` Tvrtko Ursulin
  0 siblings, 0 replies; 6+ messages in thread
From: Tvrtko Ursulin @ 2022-04-29 14:07 UTC (permalink / raw)
  To: Intel-gfx; +Cc: dri-devel

From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

Use lockdep_assert_not_held to simplify and correct the code. Otherwise
false positive are hit if lock state is uknown like after a previous
taint.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reported-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
---
It's not pretty but it fired again and it's distracting so it will have to
do for now.
---
 drivers/gpu/drm/i915/i915_vma.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
index 162e8d83691b..e782ebfcc0ca 100644
--- a/drivers/gpu/drm/i915/i915_vma.c
+++ b/drivers/gpu/drm/i915/i915_vma.c
@@ -1565,9 +1565,7 @@ int i915_ggtt_pin(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
 	if (ww)
 		return __i915_ggtt_pin(vma, ww, align, flags);
 
-#ifdef CONFIG_LOCKDEP
-	WARN_ON(dma_resv_held(vma->obj->base.resv));
-#endif
+	lockdep_assert_not_held(&vma->obj->base.resv->lock.base);
 
 	for_i915_gem_ww(&_ww, err, true) {
 		err = i915_gem_object_lock(vma->obj, &_ww);
-- 
2.32.0


^ permalink raw reply related	[flat|nested] 6+ messages in thread