[PATCH] D39720: [X86][AVX512] lowering kunpack intrinsic - llvm part
Craig Topper via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Tue Nov 7 21:12:03 PST 2017
craig.topper added a comment.
I don't see any tests that produce kunpckbw after this change.
================
Comment at: lib/IR/AutoUpgrade.cpp:1048
+ uint64_t Shift = CI->getType()->getScalarSizeInBits() / 2;
+ uint64_t And = 0xFF;
+ if( Shift == 16) And = 0xFFFF;
----------------
Can this just be
```
And = (1 << Shift) - 1;
```
https://reviews.llvm.org/D39720
More information about the llvm-commits
mailing list