[PATCH] D48477: [SLPVectorizer] Relax alternate opcodes to accept any BinaryOperator pair

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon created this revision.
RKSimon added reviewers: ABataev, dtemirbulatov, spatel, fhahn, javed.absar.

SLP currently only accepts (F)Add/(F)Sub alternate counterpart ops to be merged into an alternate shuffle.

This patch proposes to relax this to accept any pair of BinaryOperator opcodes instead, assuming the target's cost model accepts the vectorization+shuffle.


Repository:
  rL LLVM

https://reviews.llvm.org/D48477

Files:
  lib/Transforms/Vectorize/SLPVectorizer.cpp
  test/Transforms/SLPVectorizer/AArch64/transpose.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D48477.152453.patch
Type: text/x-patch
Size: 5588 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180622/f5cff5fb/attachment.bin>