From: Eric Anholt <eric@anholt.net>
Date: Tue, 14 Jun 2011 23:43:09 +0000 (-0700)
Subject: Revert "drm/i915: Kill GTT mappings when moving from GTT domain"
X-Git-Tag: MMI-PSA29.97-13-9~19068^2~5
X-Git-Url: https://git.stricted.de/?a=commitdiff_plain;h=e92d03bff9a0d0bcbb812c9b1290ca96c9338d45;p=GitHub%2FMotorolaMobilityLLC%2Fkernel-slsi.git

Revert "drm/i915: Kill GTT mappings when moving from GTT domain"

This reverts commit 4a684a4117abd756291969336af454e8a958802f.
Userland has always been required to set the object's domain to GTT
before using it through a GTT mapping, it's not something that the
kernel is supposed to enforce.  (The pagefault support is so that we
can handle multiple mappings without userland having to pin across
them, not so that userland can use GTT after GPU domains without
telling the kernel).

Fixes 19.2% +/- 0.8% (n=6) performance regression in cairo-gl
firefox-talos-gfx on my T420 latop.

Signed-off-by: Keith Packard <keithp@keithp.com>
---

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 94c84d744100..c6389de53161 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1219,11 +1219,11 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 		ret = i915_gem_object_bind_to_gtt(obj, 0, true);
 		if (ret)
 			goto unlock;
-	}
 
-	ret = i915_gem_object_set_to_gtt_domain(obj, write);
-	if (ret)
-		goto unlock;
+		ret = i915_gem_object_set_to_gtt_domain(obj, write);
+		if (ret)
+			goto unlock;
+	}
 
 	if (obj->tiling_mode == I915_TILING_NONE)
 		ret = i915_gem_object_put_fence(obj);
@@ -2926,8 +2926,6 @@ i915_gem_object_flush_gtt_write_domain(struct drm_i915_gem_object *obj)
 	 */
 	wmb();
 
-	i915_gem_release_mmap(obj);
-
 	old_write_domain = obj->base.write_domain;
 	obj->base.write_domain = 0;
 
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 20a4cc5b818f..4934cf84c320 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -187,10 +187,6 @@ i915_gem_object_set_to_gpu_domain(struct drm_i915_gem_object *obj,
 	if ((flush_domains | invalidate_domains) & I915_GEM_DOMAIN_CPU)
 		i915_gem_clflush_object(obj);
 
-	/* blow away mappings if mapped through GTT */
-	if ((flush_domains | invalidate_domains) & I915_GEM_DOMAIN_GTT)
-		i915_gem_release_mmap(obj);
-
 	if (obj->base.pending_write_domain)
 		cd->flips |= atomic_read(&obj->pending_flip);