[PATCH] D125016: [LV] Widen freeze instead of scalarizing it

[Florian Hahn via Phabricator via llvm-commits]

fhahn accepted this revision.
fhahn added a comment.
This revision is now accepted and ready to land.

LGTM, thanks!



================
Comment at: llvm/lib/Transforms/Vectorize/LoopVectorize.cpp:9354
+    for (unsigned Part = 0; Part < State.UF; ++Part) {
+
+      Value *Op = State.get(getOperand(0), Part);
----------------
nit: no need for newline here.


================
Comment at: llvm/test/Transforms/LoopVectorize/AArch64/vector-freeze.ll:20
+  %0 = load i8, i8* %arrayidx, align 1
+  br i1 false, label %if.end, label %lor.lhs.false
+
----------------
Might be good to use a real condition here to make the test more robust with respect to future changes.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D125016/new/

https://reviews.llvm.org/D125016