[PATCH] D101109: [SLP]Improve multinode analysis.

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon added a comment.

Have you been able to investigate any of these instruction increase regressions?



================
Comment at: llvm/test/Transforms/SLPVectorizer/AArch64/transpose-inseltpoison.ll:87
 ; CHECK-NEXT:    [[TMP8:%.*]] = shufflevector <4 x i32> [[TMP6]], <4 x i32> [[TMP7]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
-; CHECK-NEXT:    [[TMP9:%.*]] = add <4 x i32> [[TMP8]], [[TMP5]]
+; CHECK-NEXT:    [[TMP9:%.*]] = add <4 x i32> [[TMP3]], [[TMP8]]
 ; CHECK-NEXT:    ret <4 x i32> [[TMP9]]
----------------
Regression?


================
Comment at: llvm/test/Transforms/SLPVectorizer/AArch64/transpose-inseltpoison.ll:231
+; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <4 x i32> [[V0]], <4 x i32> undef, <4 x i32> <i32 1, i32 0, i32 2, i32 3>
+; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <4 x i32> [[V1]], <4 x i32> undef, <4 x i32> <i32 1, i32 0, i32 2, i32 3>
 ; CHECK-NEXT:    [[TMP3:%.*]] = sub <4 x i32> [[TMP1]], [[TMP2]]
----------------
Regression?


================
Comment at: llvm/test/Transforms/SLPVectorizer/AArch64/transpose.ll:83
+; CHECK-NEXT:    [[TMP4:%.*]] = shufflevector <4 x i32> [[V0]], <4 x i32> undef, <4 x i32> <i32 1, i32 0, i32 2, i32 3>
+; CHECK-NEXT:    [[TMP5:%.*]] = shufflevector <4 x i32> [[V1]], <4 x i32> undef, <4 x i32> <i32 1, i32 0, i32 2, i32 3>
+; CHECK-NEXT:    [[TMP6:%.*]] = add <4 x i32> [[TMP4]], [[TMP5]]
----------------
Regression?


================
Comment at: llvm/test/Transforms/SLPVectorizer/AArch64/transpose.ll:231
+; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <4 x i32> [[V0]], <4 x i32> undef, <4 x i32> <i32 1, i32 0, i32 2, i32 3>
+; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <4 x i32> [[V1]], <4 x i32> undef, <4 x i32> <i32 1, i32 0, i32 2, i32 3>
 ; CHECK-NEXT:    [[TMP3:%.*]] = sub <4 x i32> [[TMP1]], [[TMP2]]
----------------
Regression?


================
Comment at: llvm/test/Transforms/SLPVectorizer/X86/lookahead.ll:592
 ; CHECK-NEXT:    [[TMP2:%.*]] = load <2 x double>, <2 x double>* [[TMP1]], align 4
+; CHECK-NEXT:    [[SHUFFLE:%.*]] = shufflevector <2 x double> [[TMP2]], <2 x double> poison, <2 x i32> <i32 1, i32 0>
 ; CHECK-NEXT:    [[LOADVEC:%.*]] = load <2 x double>, <2 x double>* [[VECPTR1:%.*]], align 4
----------------
Regression?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D101109/new/

https://reviews.llvm.org/D101109