[PATCH] D148036: [SLP]Improve reduction cost model for scalars.

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon added inline comments.


================
Comment at: llvm/test/Transforms/SLPVectorizer/X86/horizontal-minmax.ll:1126
+; DEFAULT-NEXT:    [[M:%.*]] = tail call i64 @llvm.umax.i64(i64 [[M32]], i64 [[M10]])
+; DEFAULT-NEXT:    ret i64 [[M]]
 ;
----------------
Makes sense: https://gcc.godbolt.org/z/fKbGnzEr8


================
Comment at: llvm/test/Transforms/SLPVectorizer/X86/horizontal-smax.ll:37
+; AVX-NEXT:    [[TMP3:%.*]] = call i32 @llvm.vector.reduce.smax.v4i32(<4 x i32> [[TMP2]])
+; AVX-NEXT:    ret i32 [[TMP3]]
 ;
----------------
This looks to be about right: https://gcc.godbolt.org/z/sq99696Y7

You can add additional SSE test levels if you want to be certain?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D148036/new/

https://reviews.llvm.org/D148036