[PATCH] D114155: [LoopAccessAnalysis][SVE] Bail out for scalable vectors

Paul Walker via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Tue Nov 23 09:28:50 PST 2021


paulwalker-arm accepted this revision.
paulwalker-arm added a comment.
This revision is now accepted and ready to land.

A couple of things to consider but otherwise looks good.



================
Comment at: llvm/lib/Analysis/LoopAccessAnalysis.cpp:1046
+  if (isa<ScalableVectorType>(AccessTy)) {
+    LLVM_DEBUG(dbgs() << "LAA: Unsupported stride - scalable object: "
+                      << *AccessTy << "\n");
----------------
Looking at the other messages suggests "LAA: Bad stride - scalable object: " would be more consistent.


================
Comment at: llvm/test/Analysis/LoopAccessAnalysis/scalable-vector-regression-tests.ll:3-5
+; Regression tests verifying "assumption that TypeSize is not scalable" and
+; "Invalid size request on a scalable vector." are not produced by
+; -load-loop-elim.
----------------
To me this suggest this wants to be a `REQUIRES: asserts` test where the debug messages are verified to be produced/not produced as expected.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114155/new/

https://reviews.llvm.org/D114155



More information about the llvm-commits mailing list