[PATCH] D57779: [SLP] Add support for throttling.

Mon Apr 26 08:14:45 PDT 2021

ABataev added inline comments.

================
Comment at: llvm/test/Transforms/SLPVectorizer/X86/arith-fix.ll:357-361
+; AVX1-NEXT:    [[TMP53:%.*]] = insertelement <4 x i32> poison, i32 [[R12]], i32 0
+; AVX1-NEXT:    [[TMP54:%.*]] = insertelement <4 x i32> [[TMP53]], i32 [[R13]], i32 1
+; AVX1-NEXT:    [[TMP55:%.*]] = insertelement <4 x i32> [[TMP54]], i32 [[R14]], i32 2
+; AVX1-NEXT:    [[TMP56:%.*]] = insertelement <4 x i32> [[TMP55]], i32 [[R15]], i32 3
+; AVX1-NEXT:    store <4 x i32> [[TMP56]], <4 x i32>* bitcast (i32* getelementptr inbounds ([16 x i32], [16 x i32]* @c32, i32 0, i64 12) to <4 x i32>*), align 4
----------------
Looks like it does not respect `MinTreeSize` option anymore. And it is strange that such code sequence gets profitable for vectorization (scalar cost is 8, vector cost is 9)

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D57779/new/

https://reviews.llvm.org/D57779