[PATCH] D56551: [LoopVectorizer] give more advice in remark about failure to vectorize call

[Sanjay Patel via Phabricator via llvm-commits]

spatel created this revision.
spatel added reviewers: hfinkel, Ayal, efriedma.
Herald added a subscriber: mcrosier.

Something like this is requested by:
https://bugs.llvm.org/show_bug.cgi?id=40265
...and it seems like a common enough case that we should acknowledge it. Not sure if this crosses the line for wordiness in an optimization remark though.


https://reviews.llvm.org/D56551

Files:
  lib/Transforms/Vectorize/LoopVectorizationLegality.cpp


Index: lib/Transforms/Vectorize/LoopVectorizationLegality.cpp
===================================================================
--- lib/Transforms/Vectorize/LoopVectorizationLegality.cpp
+++ lib/Transforms/Vectorize/LoopVectorizationLegality.cpp
@@ -715,7 +715,9 @@
           !(CI->getCalledFunction() && TLI &&
             TLI->isFunctionVectorizable(CI->getCalledFunction()->getName()))) {
         ORE->emit(createMissedAnalysis("CantVectorizeCall", CI)
-                  << "call instruction cannot be vectorized");
+                  << "call instruction cannot be vectorized "
+                     "(if this is a math library call, consider relaxing the "
+                     "floating-point model)");
         LLVM_DEBUG(
             dbgs() << "LV: Found a non-intrinsic, non-libfunc callsite.\n");
         return false;


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D56551.181077.patch
Type: text/x-patch
Size: 838 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190110/065a37d7/attachment.bin>