diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2019-07-02 19:15:45 +0000 |
|---|---|---|
| committer | Matt Arsenault <Matthew.Arsenault@amd.com> | 2019-07-02 19:15:45 +0000 |
| commit | 5fe851b6cd90ceaa4cf468d9403b3920e1d0ae15 (patch) | |
| tree | d5be01bc00f8101862c7d350c6e7c22028de38b9 /llvm/lib/AsmParser/LLLexer.cpp | |
| parent | e6768d613adb3313722d02aa5740f33ab11b60fe (diff) | |
| download | bcm5719-llvm-5fe851b6cd90ceaa4cf468d9403b3920e1d0ae15.tar.gz bcm5719-llvm-5fe851b6cd90ceaa4cf468d9403b3920e1d0ae15.zip | |
AMDGPU: Custom lower vector_shuffle for v4i16/v4f16
Ordinarily it is lowered as a build_vector of each extract_vector_elt,
which in turn get lowered to bitcasts and bit shifts. Very little
understand the lowered extract pattern, resulting in much worse
code. We treat concat_vectors of v2i16 as legal, so prefer that.
llvm-svn: 364959
Diffstat (limited to 'llvm/lib/AsmParser/LLLexer.cpp')
0 files changed, 0 insertions, 0 deletions
