[PATCH] D69563: [LV] Strip wrap flags from vectorized reductions

[Denis Antrushin via Phabricator via llvm-commits]

dantrushin marked 9 inline comments as done.
dantrushin added inline comments.


================
Comment at: llvm/lib/Transforms/Vectorize/LoopVectorize.cpp:3729
+    for (Instruction *I : ReduList)
+      if (isa<OverflowingBinaryOperator>(I)) {
+        Value *V = getOrCreateVectorValue(I, Part);
----------------
Ayal wrote:
> Better interchange to loop over all Parts inside. 
> Can then potentially do "if (!isa<OverflowingBinaryOperator>(I)) continue;" if preferred.
I had to fuse loops before interchanging.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D69563/new/

https://reviews.llvm.org/D69563