diff options
| author | Craig Topper <craig.topper@intel.com> | 2019-08-07 21:16:10 +0000 |
|---|---|---|
| committer | Craig Topper <craig.topper@intel.com> | 2019-08-07 21:16:10 +0000 |
| commit | 7f7ef0208b57a4d253cd8b07053460f40ad7cbc8 (patch) | |
| tree | b05475a04ead756451dbe459b9b9d518663260b1 /llvm/docs/AMDGPUModifierSyntax.rst | |
| parent | d47be4da5ab2dd68e89701b2f3c9b355ab95ef36 (diff) | |
| download | bcm5719-llvm-7f7ef0208b57a4d253cd8b07053460f40ad7cbc8.tar.gz bcm5719-llvm-7f7ef0208b57a4d253cd8b07053460f40ad7cbc8.zip | |
[X86] Allow pack instructions to be used for 512->256 truncates when -mprefer-vector-width=256 is causing 512-bit vectors to be split
If we're splitting the 512-bit vector anyway and we have zero/sign bits, then we might as well use pack instructions to concat and truncate at once.
Differential Revision: https://reviews.llvm.org/D65904
llvm-svn: 368210
Diffstat (limited to 'llvm/docs/AMDGPUModifierSyntax.rst')
0 files changed, 0 insertions, 0 deletions
