[llvm] [LV] Vectorize FMax w/o fast-math flags. (PR #146711)

Florian Hahn via llvm-commits llvm-commits at lists.llvm.org
Mon Jul 7 02:39:02 PDT 2025


================
@@ -8909,6 +8914,9 @@ VPlanPtr LoopVectorizationPlanner::tryToBuildVPlanWithVPRecipes(
 
   // Adjust the recipes for any inloop reductions.
   adjustRecipesForReductions(Plan, RecipeBuilder, Range.Start);
+  if (!VPlanTransforms::runPass(
+          VPlanTransforms::handleFMaxReductionsWithoutFastMath, *Plan))
+    return nullptr;
----------------
fhahn wrote:

See response here https://github.com/llvm/llvm-project/pull/146711/files/598c59f3c7d565ff7994c816f52c0aaeed138402#r2189438081

https://github.com/llvm/llvm-project/pull/146711


More information about the llvm-commits mailing list