diff options
| author | Craig Topper <craig.topper@intel.com> | 2017-08-17 15:40:25 +0000 |
|---|---|---|
| committer | Craig Topper <craig.topper@intel.com> | 2017-08-17 15:40:25 +0000 |
| commit | 3a622a14f9eb62b9f8aae5f5e937b5a62dab06a4 (patch) | |
| tree | 1bd8b2dd48d50e804ff4a0deefd420cd46fea0ef /llvm/lib/Transforms/Scalar/LoopUnswitch.cpp | |
| parent | 5960848060d1f6882e7b39075b6b034c6c22066c (diff) | |
| download | bcm5719-llvm-3a622a14f9eb62b9f8aae5f5e937b5a62dab06a4.tar.gz bcm5719-llvm-3a622a14f9eb62b9f8aae5f5e937b5a62dab06a4.zip | |
[AVX512] Don't switch unmasked subvector insert/extract instructions when AVX512DQI is enabled.
There's no reason to switch instructions with and without DQI. It just creates extra isel patterns and test divergences.
There is however value in enabling the masked version of the instructions with DQI.
This required introducing some new multiclasses to enabling this splitting.
Differential Revision: https://reviews.llvm.org/D36661
llvm-svn: 311091
Diffstat (limited to 'llvm/lib/Transforms/Scalar/LoopUnswitch.cpp')
0 files changed, 0 insertions, 0 deletions
