fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2026-05-07 04:58:05 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 4

i965/fs: Don't set exec_all on instructions wider than the original in lower_simd_width.

Browse source 
This could have led to somewhat increased bandwidth usage for lowered
texturing instructions on Gen4 (which is the only case in which
lower_width may be greater than inst->exec_size).  After the previous
patches the invariant mentioned in the comment should no longer be
assumed by any of the other optimization and lowering passes, so the
exec_all() call shouldn't be necessary anymore.

Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
This commit is contained in:
Francisco Jerez 2015-07-27 18:42:31 +03:00
parent eaba922582
commit 4529916dfd
1 changed files with 11 additions and 9 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

20
src/mesa/drivers/dri/i965/brw_fs.cpp
View file

@ -4163,10 +4163,15 @@ fs_visitor::lower_simd_width()
const unsigned lower_width = get_lowered_simd_width(devinfo, inst); const unsigned lower_width = get_lowered_simd_width(devinfo, inst);
if (lower_width != inst->exec_size) { if (lower_width != inst->exec_size) {
/* Builder matching the original instruction. */ /* Builder matching the original instruction. We may also need to
* emit an instruction of width larger than the original, set the
* execution size of the builder to the highest of both for now so
* we're sure that both cases can be handled.
*/
const fs_builder ibld = bld.at(block, inst) const fs_builder ibld = bld.at(block, inst)
.exec_all(inst->force_writemask_all) .exec_all(inst->force_writemask_all)
.group(inst->exec_size, inst->force_sechalf); .group(MAX2(inst->exec_size, lower_width),
inst->force_sechalf);
/* Split the copies in chunks of the execution width of either the /* Split the copies in chunks of the execution width of either the
* original or the lowered instruction, whichever is lower. * original or the lowered instruction, whichever is lower.
@ -4189,14 +4194,11 @@ fs_visitor::lower_simd_width()
split_inst.exec_size = lower_width; split_inst.exec_size = lower_width;
split_inst.eot = inst->eot && i == n - 1; split_inst.eot = inst->eot && i == n - 1;
/* Set exec_all if the lowered width is higher than the original /* Select the correct channel enables for the i-th group, then
* to avoid breaking the compiler invariant that no control * transform the sources and destination and emit the lowered
* flow-masked instruction is wider than the shader's * instruction.
* dispatch_width. Then transform the sources and destination and
* emit the lowered instruction.
*/ */
const fs_builder lbld = ibld.exec_all(lower_width > inst->exec_size) const fs_builder lbld = ibld.group(lower_width, i);
.group(lower_width, i);
for (unsigned j = 0; j < inst->sources; j++) { for (unsigned j = 0; j < inst->sources; j++) {
if (inst->src[j].file != BAD_FILE && if (inst->src[j].file != BAD_FILE &&