When you've got a simple solid-color shader that doesn't generate
pixel_x/y interpolation, we were deciding that the first vgrf was the same
as both the undefined pixel_x and pixel_y, and extending its live interval
to avoid the stride problem. That tricked optimization (such as
compute_to_mrf) that tries to see if a particular instruction is the last
use of a variable.
---
src/mesa/drivers/dri/i965/brw_fs_live_variables.cpp | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_fs_live_variables.cpp
b/src/mesa/drivers/dri/i965/brw_fs_live_variables.cpp
index 6151729..c7b1f25 100644
--- a/src/mesa/drivers/dri/i965/brw_fs_live_variables.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs_live_variables.cpp
@@ -86,8 +86,10 @@ fs_live_variables::setup_one_read(bblock_t *block, fs_inst
*inst,
*/
int end_ip = ip;
if (v->dispatch_width == 16 && (reg.stride == 0 ||
- (v->pixel_x.reg == reg.reg ||
- v->pixel_y.reg == reg.reg))) {
+ ((v->pixel_x.file == GRF &&
+ v->pixel_x.reg == reg.reg) ||
+ (v->pixel_y.file == GRF &&
+ v->pixel_y.reg == reg.reg)))) {
end_ip++;
}
--
1.9.0
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
