The SIMD16 replicated FB write message only works if we don't need the
color calculator to mask our framebuffer writes. Previously, we bailed
on it if color_mask wasn't <true, true, true, true>. However, this was
needlessly strict for formats with fewer than four components - only the
components that actually exist matter.
WebGL Aquarium attempts to clear a BGRX texture with the ColorMask set
to <true, true, true, false>. This will work perfectly fine with the
replicated data message; we just bailed unnecessarily.
Improves performance of WebGL Aquarium on Iris Pro (at 1920x1080) by
abound 50%, and Bay Trail (at 1366x768) by over 70% (using Chrome 24).
v2: Use _mesa_format_has_color_component() to properly handle ALPHA
formats (and generally be less fragile).
Signed-off-by: Kenneth Graunke <kenn...@whitecape.org>
---
src/mesa/drivers/dri/i965/brw_blorp_clear.cpp | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
We do support rendering to alpha formats, so the old patch was broken.
This one ought to work, and still gives the same performance improvement.
diff --git a/src/mesa/drivers/dri/i965/brw_blorp_clear.cpp
b/src/mesa/drivers/dri/i965/brw_blorp_clear.cpp
index 3c6c57b..fd18b45 100644
--- a/src/mesa/drivers/dri/i965/brw_blorp_clear.cpp
+++ b/src/mesa/drivers/dri/i965/brw_blorp_clear.cpp
@@ -231,7 +231,8 @@ brw_blorp_clear_params::brw_blorp_clear_params(struct
brw_context *brw,
* state. This is not documented.
*/
for (int i = 0; i < 4; i++) {
- if (!color_mask[i]) {
+ if (_mesa_format_has_color_component(irb->mt->format, i) &&
+ !color_mask[i]) {
color_write_disable[i] = true;
wm_prog_key.use_simd16_replicated_data = false;
}
--
1.9.0
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
