[Intel-gfx] [PATCH 07/11] drm/i915: Fix up map and fenceable for VMA
Ben Widawsky
ben at bwidawsk.net
Tue Jul 9 08:08:38 CEST 2013
formerly: "drm/i915: Create VMAs (part 3.5) - map and fenceable
tracking"
The map_and_fenceable tracking is per object. GTT mapping, and fences
only apply to global GTT. As such, object operations which are not
performed on the global GTT should not effect mappable or fenceable
characteristics.
Functionally, this commit could very well be squashed in to the previous
patch which updated object operations to take a VM argument. This
commit is split out because it's a bit tricky (or at least it was for
me).
Signed-off-by: Ben Widawsky <ben at bwidawsk.net>
---
drivers/gpu/drm/i915/i915_gem.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 21015cd..501c590 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -2635,7 +2635,7 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj,
trace_i915_gem_object_unbind(obj, vm);
- if (obj->has_global_gtt_mapping)
+ if (obj->has_global_gtt_mapping && i915_is_ggtt(vm))
i915_gem_gtt_unbind_object(obj);
if (obj->has_aliasing_ppgtt_mapping) {
i915_ppgtt_unbind_object(dev_priv->mm.aliasing_ppgtt, obj);
@@ -2646,7 +2646,8 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj,
list_del(&obj->mm_list);
/* Avoid an unnecessary call to unbind on rebind. */
- obj->map_and_fenceable = true;
+ if (i915_is_ggtt(vm))
+ obj->map_and_fenceable = true;
vma = i915_gem_obj_to_vma(obj, vm);
list_del(&vma->vma_link);
@@ -3213,7 +3214,9 @@ search_free:
i915_is_ggtt(vm) &&
vma->node.start + obj->base.size <= dev_priv->gtt.mappable_end;
- obj->map_and_fenceable = mappable && fenceable;
+ /* Map and fenceable only changes if the VM is the global GGTT */
+ if (i915_is_ggtt(vm))
+ obj->map_and_fenceable = mappable && fenceable;
trace_i915_gem_object_bind(obj, vm, map_and_fenceable);
i915_gem_verify_gtt(dev);
--
1.8.3.2
More information about the Intel-gfx
mailing list