[PATCH] D29826: [SLP] General improvements of SLP vectorization process.

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon added inline comments.


================
Comment at: lib/Transforms/Vectorize/SLPVectorizer.cpp:4152
 
-bool SLPVectorizerPass::tryToVectorize(BinaryOperator *V, BoUpSLP &R) {
-  if (!V)
+bool SLPVectorizerPass::tryToVectorize(Instruction *I, BoUpSLP &R) {
+  if (!I)
----------------
This BinaryOperator->Instruction change (along with associated changes in tryToVectorizeHorReductionOrInstOperands and vectorizeRootInstruction) look like they can pulled out and committed without any problem.


================
Comment at: test/Transforms/SLPVectorizer/AArch64/gather-root.ll:1
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -slp-vectorizer -S | FileCheck %s --check-prefix=DEFAULT
----------------
Regenerate the current trunk codegen with the script (it looks like you've edited below), commit that and rebase to show the diffs in this patch.


https://reviews.llvm.org/D29826