diff options
author | Ian Romanick <[email protected]> | 2018-09-12 17:16:50 -0700 |
---|---|---|
committer | Ian Romanick <[email protected]> | 2018-09-17 00:38:22 -0700 |
commit | df9dbc03d3df37c10f2d7b73124f8b3fd7a2518f (patch) | |
tree | 6e64b111bc87f2e6e8e6e485aa6b5be55ad51e77 /src | |
parent | 0dd8189f159258ad8684036b0d696dbb58a268ca (diff) |
i965/fs: Don't propagate conditional modifiers from integer compares to adds
No shader-db changes on any Intel platform... which probably explains
why no bugs have been bisected to this problem since it landed in Mesa
18.1. :( The commit mentioned below is in 18.2, so 18.1 would need a
slightly different fix (due to code refactoring).
Signed-off-by: Ian Romanick <[email protected]>
Fixes: 77f269bb560 "i965/fs: Refactor propagation of conditional modifiers from compares to adds"
Reviewed-by: Alejandro PiƱeiro <[email protected]> (reviewed the original patch)
Cc: Matt Turner <[email protected]> (reviewed the original patch)
Diffstat (limited to 'src')
-rw-r--r-- | src/intel/compiler/brw_fs_cmod_propagation.cpp | 10 |
1 files changed, 9 insertions, 1 deletions
diff --git a/src/intel/compiler/brw_fs_cmod_propagation.cpp b/src/intel/compiler/brw_fs_cmod_propagation.cpp index 5b74f267359..5fb522f810f 100644 --- a/src/intel/compiler/brw_fs_cmod_propagation.cpp +++ b/src/intel/compiler/brw_fs_cmod_propagation.cpp @@ -211,9 +211,17 @@ opt_cmod_propagation_local(const gen_device_info *devinfo, bblock_t *block) /* A CMP with a second source of zero can match with anything. A CMP * with a second source that is not zero can only match with an ADD * instruction. + * + * Only apply this optimization to float-point sources. It can fail for + * integers. For inputs a = 0x80000000, b = 4, int(0x80000000) < 4, but + * int(0x80000000) - 4 overflows and results in 0x7ffffffc. that's not + * less than zero, so the flags get set differently than for (a < b). */ if (inst->opcode == BRW_OPCODE_CMP && !inst->src[1].is_zero()) { - progress = cmod_propagate_cmp_to_add(devinfo, block, inst) || progress; + if (brw_reg_type_is_floating_point(inst->src[0].type) && + cmod_propagate_cmp_to_add(devinfo, block, inst)) + progress = true; + continue; } |