[PATCH] D121618: [LV] Move code to place pointer induction increment to VPlan post-processing.

[Ayal Zaks via Phabricator via llvm-commits]

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

Looks fine to me, with a couple of nits.



================
Comment at: llvm/lib/Transforms/Vectorize/LoopVectorize.cpp:9602
   // A pointer induction, performed by using a gep
-  BasicBlock *LoopLatch =
-      State.LI->getLoopFor(State.CFG.PrevBB)->getLoopLatch();
+  const DataLayout &DL = CanonicalIV->getModule()->getDataLayout();
+  Instruction *InductionLoc = &*State.Builder.GetInsertPoint();
----------------
nit: can NewPointerPhi  be used instead of CanonicalIV, which seems less relevant?


================
Comment at: llvm/lib/Transforms/Vectorize/VPlan.cpp:990
+      PHINode *Phi = nullptr;
+      if (auto *WidenPhi = dyn_cast<VPWidenPointerInductionRecipe>(&R)) {
+        // TODO: Split off the case that all users of a pointer phi are scalar
----------------
nit: reverse the condition to handle the simpler int or fp case first.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D121618/new/

https://reviews.llvm.org/D121618