summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/i915/i915_gem.c
diff options
context:
space:
mode:
authorChris Wilson <chris@chris-wilson.co.uk>2017-06-16 11:54:55 +0100
committerChris Wilson <chris@chris-wilson.co.uk>2017-06-16 14:52:27 +0100
commit7fc92e96c3eed6004ce8dab5e315264bff85db5a (patch)
tree4440f4b244e2699e387f2abdb1e37c295a50f279 /drivers/gpu/drm/i915/i915_gem.c
parente27ab73d17ef90db3e586a02ce2f03eb660451cd (diff)
drm/i915: Store i915_gem_object_is_coherent() as a bit next to cache-dirty
For ease of use (i.e. avoiding a few checks and function calls), store the object's cache coherency next to the cache is dirty bit. Specifically this patch aims to reduce the frequency of no-op calls to i915_gem_object_clflush() to counter-act the increase of such calls for GPU only objects in the previous patch. v2: Replace cache_dirty & ~cache_coherent with cache_dirty && !cache_coherent as gcc generates much better code for the latter (Tvrtko) Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Dongwon Kim <dongwon.kim@intel.com> Cc: Matt Roper <matthew.d.roper@intel.com> Tested-by: Dongwon Kim <dongwon.kim@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20170616105455.16977-1-chris@chris-wilson.co.uk Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Diffstat (limited to 'drivers/gpu/drm/i915/i915_gem.c')
-rw-r--r--drivers/gpu/drm/i915/i915_gem.c14
1 files changed, 7 insertions, 7 deletions
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index b1504a829c6a..4ae30f74c475 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -52,7 +52,7 @@ static bool cpu_write_needs_clflush(struct drm_i915_gem_object *obj)
if (obj->cache_dirty)
return false;
- if (!i915_gem_object_is_coherent(obj))
+ if (!obj->cache_coherent)
return true;
return obj->pin_display;
@@ -253,7 +253,7 @@ __i915_gem_object_release_shmem(struct drm_i915_gem_object *obj,
if (needs_clflush &&
(obj->base.read_domains & I915_GEM_DOMAIN_CPU) == 0 &&
- !i915_gem_object_is_coherent(obj))
+ !obj->cache_coherent)
drm_clflush_sg(pages);
__start_cpu_write(obj);
@@ -856,8 +856,7 @@ int i915_gem_obj_prepare_shmem_read(struct drm_i915_gem_object *obj,
if (ret)
return ret;
- if (i915_gem_object_is_coherent(obj) ||
- !static_cpu_has(X86_FEATURE_CLFLUSH)) {
+ if (obj->cache_coherent || !static_cpu_has(X86_FEATURE_CLFLUSH)) {
ret = i915_gem_object_set_to_cpu_domain(obj, false);
if (ret)
goto err_unpin;
@@ -909,8 +908,7 @@ int i915_gem_obj_prepare_shmem_write(struct drm_i915_gem_object *obj,
if (ret)
return ret;
- if (i915_gem_object_is_coherent(obj) ||
- !static_cpu_has(X86_FEATURE_CLFLUSH)) {
+ if (obj->cache_coherent || !static_cpu_has(X86_FEATURE_CLFLUSH)) {
ret = i915_gem_object_set_to_cpu_domain(obj, true);
if (ret)
goto err_unpin;
@@ -3684,6 +3682,7 @@ restart:
list_for_each_entry(vma, &obj->vma_list, obj_link)
vma->node.color = cache_level;
obj->cache_level = cache_level;
+ obj->cache_coherent = i915_gem_object_is_coherent(obj);
obj->cache_dirty = true; /* Always invalidate stale cachelines */
return 0;
@@ -4344,7 +4343,8 @@ i915_gem_object_create(struct drm_i915_private *dev_priv, u64 size)
} else
obj->cache_level = I915_CACHE_NONE;
- obj->cache_dirty = !i915_gem_object_is_coherent(obj);
+ obj->cache_coherent = i915_gem_object_is_coherent(obj);
+ obj->cache_dirty = !obj->cache_coherent;
trace_i915_gem_object_create(obj);