[Intel-gfx] [PATCH] i915: Set object to gtt domain when faulting it back in

[Kristian Høgsberg]

From: Kristian Høgsberg <krh at redhat.com>

When a GEM object is evicted from the GTT we set it to the CPU domain,
as it might get swapped in and out or ever mmapped regularly.  If the
object is mmapped through the GTT it can still get evicted in this way
by other objects requiring GTT space.  When the GTT mapping is touched
again we fault it back into the GTT, but fail to set it back to the
GTT domain.  This means we fail to flush any cached CPU writes to the
pages backing the object which will then happen "eventually", typically
after we write to the page through the uncached GTT mapping.

Signed-off-by: Kristian Høgsberg <krh at redhat.com>
---
 drivers/gpu/drm/i915/i915_gem.c |    7 +++++++
 1 files changed, 7 insertions(+), 0 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 4a24c90..2500510 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1145,6 +1145,13 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 			mutex_unlock(&dev->struct_mutex);
 			return VM_FAULT_SIGBUS;
 		}
+
+		ret = i915_gem_object_set_to_gtt_domain(obj, write);
+		if (ret) {
+			mutex_unlock(&dev->struct_mutex);
+			return VM_FAULT_SIGBUS;
+		}
+
 		list_add(&obj_priv->list, &dev_priv->mm.inactive_list);
 	}
 
-- 
1.6.2.2