[PATCH] D140514: [VPlan] Use VPDominatorTree in VPlanVerifier .

Florian Hahn via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Wed Jan 25 04:17:30 PST 2023


fhahn marked 6 inline comments as done.
fhahn added inline comments.


================
Comment at: llvm/lib/Transforms/Vectorize/LoopVectorize.cpp:9165
 
+#ifndef NDEBUG
   assert(VPlanVerifier::verifyPlanIsValid(*Plan) && "VPlan is invalid");
----------------
Ayal wrote:
> nit: can be done independent of this patch.
It's not needed any longer now that the VPDominatorTree is computed inside the call. I'll remove it.


================
Comment at: llvm/lib/Transforms/Vectorize/VPlanVerifier.cpp:212
           continue;
 
         // If the user is in the same block, check it comes after R in the
----------------
Ayal wrote:
> nit: can introduce a method that takes two recipes (along with their numberings) and says if one dominates the other, whether they belong to the same VPBB's or not. Analogous to DominatorTree interfaces.
That would indeed be a useful utility. I'll prepare that separately.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D140514/new/

https://reviews.llvm.org/D140514



More information about the llvm-commits mailing list