diff options
| author | Justin Holewinski <jholewinski@nvidia.com> | 2015-01-26 19:52:20 +0000 |
|---|---|---|
| committer | Justin Holewinski <jholewinski@nvidia.com> | 2015-01-26 19:52:20 +0000 |
| commit | d4d2e9bd0ec19cd7fc759d3179aa23d870568142 (patch) | |
| tree | 9772ceeec95c5f4c6c5ab215f4bca60a09f2dbd5 /llvm/lib/Transforms | |
| parent | d8cb6b00c535fac4404d77413e3f437a0dc33a9b (diff) | |
| download | bcm5719-llvm-d4d2e9bd0ec19cd7fc759d3179aa23d870568142.tar.gz bcm5719-llvm-d4d2e9bd0ec19cd7fc759d3179aa23d870568142.zip | |
[NVPTX] Generate a more optimal sequence for select of i1
Instead of creating a pattern like "(p && a) || ((!p) && b)",
just expand the i8 operands to i32 and perform the selp on them.
Fixes PR22246
llvm-svn: 227123
Diffstat (limited to 'llvm/lib/Transforms')
0 files changed, 0 insertions, 0 deletions
