diff options
| author | Sanjay Patel <spatel@rotateright.com> | 2018-09-22 14:43:55 +0000 |
|---|---|---|
| committer | Sanjay Patel <spatel@rotateright.com> | 2018-09-22 14:43:55 +0000 |
| commit | 09e02fbf519aa19a85d65d144578cdb91b85865a (patch) | |
| tree | b85909409d03167dab5543a7fc4428800af09b81 /llvm/lib/Transforms/Utils/ImportedFunctionsInliningStatistics.cpp | |
| parent | f3b94728c708df60741bf6c6f054f10c4474fd7f (diff) | |
| download | bcm5719-llvm-09e02fbf519aa19a85d65d144578cdb91b85865a.tar.gz bcm5719-llvm-09e02fbf519aa19a85d65d144578cdb91b85865a.zip | |
[InstCombine][x86] try even harder to convert blendv intrinsic to generic IR (PR38814)
Follow-up to rL342324 (D52059):
Missing optimizations with blendv are shown in:
https://bugs.llvm.org/show_bug.cgi?id=38814
This is an easier and more powerful solution than adding pattern matching for a few
special cases in the backend. The potential danger with this transform in IR is that
the condition value can get separated from the select, and the backend might not be
able to make a blendv out of it again.
llvm-svn: 342806
Diffstat (limited to 'llvm/lib/Transforms/Utils/ImportedFunctionsInliningStatistics.cpp')
0 files changed, 0 insertions, 0 deletions
