diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2019-06-13 10:30:12 +0000 |
|---|---|---|
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2019-06-13 10:30:12 +0000 |
| commit | e1aea858968364a3685d31042a46b470ec33458b (patch) | |
| tree | 4a7ee406a4eba9fca5bd33bb772a86f2d69f1c48 /llvm/test/CodeGen/NVPTX/loop-vectorize.ll | |
| parent | 076ae0d2e2f6cee455f49a131b32d14d457f603c (diff) | |
| download | bcm5719-llvm-e1aea858968364a3685d31042a46b470ec33458b.tar.gz bcm5719-llvm-e1aea858968364a3685d31042a46b470ec33458b.zip | |
[X86][SSE] Add SSE4A nt store tests on X86 as well as X64
We should be able to use MOVNTSD (f64) instead of MOVNTI (i32) to reduce the number of ops 32-bit targets
Pulled out of D63246
llvm-svn: 363247
Diffstat (limited to 'llvm/test/CodeGen/NVPTX/loop-vectorize.ll')
0 files changed, 0 insertions, 0 deletions
