diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2017-02-05 22:50:29 +0000 |
|---|---|---|
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2017-02-05 22:50:29 +0000 |
| commit | 380ce75687e9f3fc00d52984dd9d64c11a8ad6eb (patch) | |
| tree | 3f2c81e191fb4ab89f6c70fc99890e5720ad0e25 /llvm/test/CodeGen/X86/fast-isel-select-pseudo-cmov.ll | |
| parent | 134ed9986a1baa03d568dd551e20d2134d3dfeae (diff) | |
| download | bcm5719-llvm-380ce75687e9f3fc00d52984dd9d64c11a8ad6eb.tar.gz bcm5719-llvm-380ce75687e9f3fc00d52984dd9d64c11a8ad6eb.zip | |
[X86][SSE] Replace insert_vector_elt(vec, -1, idx) with shuffle
Similar to what we already do for zero elt insertion, we can quickly rematerialize 'allbits' vectors so to avoid a unnecessary gpr value and insertion into a vector
llvm-svn: 294162
Diffstat (limited to 'llvm/test/CodeGen/X86/fast-isel-select-pseudo-cmov.ll')
0 files changed, 0 insertions, 0 deletions
