diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2016-12-22 21:40:08 +0000 |
|---|---|---|
| committer | Matt Arsenault <Matthew.Arsenault@amd.com> | 2016-12-22 21:40:08 +0000 |
| commit | 0b26e47345e48335c2c7e390fbb3b409b9f0285a (patch) | |
| tree | 22c5fec0d55f3bb670763d1240b2e08436040e45 /llvm/lib/CodeGen/TargetLoweringBase.cpp | |
| parent | 055f506c5465a671f10eec465869105662e6f5c0 (diff) | |
| download | bcm5719-llvm-0b26e47345e48335c2c7e390fbb3b409b9f0285a.tar.gz bcm5719-llvm-0b26e47345e48335c2c7e390fbb3b409b9f0285a.zip | |
AMDGPU: Invert cmp + select with constant
Canonicalize a select with a constant to the false side. This
enables more instruction shrinking opportunities since an
inline immediate can be used for the false side of v_cndmask_b32_e32.
This seems to usually be better but causes some code size regressions
in some tests.
llvm-svn: 290372
Diffstat (limited to 'llvm/lib/CodeGen/TargetLoweringBase.cpp')
0 files changed, 0 insertions, 0 deletions
