We can avoid an early clflush when pwriting if we use the current CPU
write domain rather than moving the object to the GTT domain for the
purposes of the pwrite. This has the advantage of not flushing the
presumably hot data that we want to upload into the bo, and of ascribing
the clflush to the execution when profiling.
Signed-off-by: Chris Wilson <ch...@chris-wilson.co.uk>
---
drivers/gpu/drm/i915/i915_gem.c | 3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 528ff7d..c10bf39 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -941,7 +941,8 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data,
if (obj_priv->phys_obj)
ret = i915_gem_phys_pwrite(dev, obj, args, file_priv);
else if (obj_priv->tiling_mode == I915_TILING_NONE &&
- dev->gtt_total != 0) {
+ dev->gtt_total != 0 &&
+ obj->write_domain != I915_GEM_DOMAIN_CPU) {
ret = i915_gem_gtt_pwrite_fast(dev, obj, args, file_priv);
if (ret == -EFAULT) {
ret = i915_gem_gtt_pwrite_slow(dev, obj, args,
--
1.7.1
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
