drm/i915: bind objects to the global gtt only when needed

And track the existence of such a binding similar to the aliasing
ppgtt case. Speeds up binding/unbinding in the common case where we
only need a ppgtt binding (which is accessed in a cpu coherent fashion
by the gpu) and no gloabl gtt binding (which needs uc writes for the
ptes).

This patch just puts the required tracking in place.

v2: Check that global gtt mappings exist in the error_state capture
code (with Chris Wilson's llc reloc patches batchbuffers are no longer
relocated as mappable in all situations, so this matters). Suggested
by Chris Wilson.

v3: Adapted to Chris' latest llc-reloc patches.

v4: Fix a bug in the i915 error state capture code noticed by Chris
Wilson.

Reviewed-and-tested-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

4 files changed, 17 insertions, 3 deletions

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 3619f76..b6098b0 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -882,6 +882,7 @@ struct drm_i915_gem_object {
 	unsigned int cache_level:2;
 
 	unsigned int has_aliasing_ppgtt_mapping:1;
+	unsigned int has_global_gtt_mapping:1;
 
 	struct page **pages;
 
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 031ca5b..69009d1 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1153,6 +1153,9 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 			goto unlock;
 	}
 
+	if (!obj->has_global_gtt_mapping)
+		i915_gem_gtt_bind_object(obj, obj->cache_level);
+
 	if (obj->tiling_mode == I915_TILING_NONE)
 		ret = i915_gem_object_put_fence(obj);
 	else
@@ -2097,7 +2100,8 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj)
 
 	trace_i915_gem_object_unbind(obj);
 
-	i915_gem_gtt_unbind_object(obj);
+	if (obj->has_global_gtt_mapping)
+		i915_gem_gtt_unbind_object(obj);
 	if (obj->has_aliasing_ppgtt_mapping) {
 		i915_ppgtt_unbind_object(dev_priv->mm.aliasing_ppgtt, obj);
 		obj->has_aliasing_ppgtt_mapping = 0;
@@ -2952,7 +2956,8 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 				return ret;
 		}
 
-		i915_gem_gtt_bind_object(obj, cache_level);
+		if (obj->has_global_gtt_mapping)
+			i915_gem_gtt_bind_object(obj, cache_level);
 		if (obj->has_aliasing_ppgtt_mapping)
 			i915_ppgtt_bind_object(dev_priv->mm.aliasing_ppgtt,
 					       obj, cache_level);
@@ -3342,6 +3347,9 @@ i915_gem_object_pin(struct drm_i915_gem_object *obj,
 			return ret;
 	}
 
+	if (!obj->has_global_gtt_mapping && map_and_fenceable)
+		i915_gem_gtt_bind_object(obj, obj->cache_level);
+
 	if (obj->pin_count++ == 0) {
 		if (!obj->active)
 			list_move_tail(&obj->mm_list,
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index bf33eaf..72be806 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -394,12 +394,16 @@ void i915_gem_gtt_bind_object(struct drm_i915_gem_object *obj,
 				       obj->base.size >> PAGE_SHIFT,
 				       obj->pages,
 				       agp_type);
+
+	obj->has_global_gtt_mapping = 1;
 }
 
 void i915_gem_gtt_unbind_object(struct drm_i915_gem_object *obj)
 {
 	intel_gtt_clear_range(obj->gtt_space->start >> PAGE_SHIFT,
 			      obj->base.size >> PAGE_SHIFT);
+
+	obj->has_global_gtt_mapping = 0;
 }
 
 void i915_gem_gtt_finish_object(struct drm_i915_gem_object *obj)
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 4c89149..998116e 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -729,7 +729,8 @@ i915_error_object_create(struct drm_i915_private *dev_priv,
 			goto unwind;
 
 		local_irq_save(flags);
-		if (reloc_offset < dev_priv->mm.gtt_mappable_end) {
+		if (reloc_offset < dev_priv->mm.gtt_mappable_end &&
+		    src->has_global_gtt_mapping) {
 			void __iomem *s;
 
 			/* Simply ignore tiling or any overlapping fence.