[PATCH] D76844: [InstCombine] try to reduce shuffle with bitcasted operand
Sanjay Patel via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Thu Mar 26 13:02:56 PDT 2020
spatel marked an inline comment as done.
spatel added inline comments.
================
Comment at: llvm/lib/Transforms/InstCombine/InstCombineVectorOps.cpp:1937-1939
+ SmallVector<Constant *, 16> ScaledMaskC;
+ for (unsigned i = 0; i != VWidth; ++i)
+ ScaledMaskC.push_back(Builder.getInt32(ScaledMask[i]));
----------------
lebedev.ri wrote:
> Same comment about reserve+range-based-for, although i hope that will get obsoleted by D72467?
Yes, if we convert the mask from a constant operand to a special constant, then this and other code gets simplified away.
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D76844/new/
https://reviews.llvm.org/D76844
More information about the llvm-commits
mailing list