[SLP] Test for PR35047, NFC.

llvm-svn: 317482

diff --git a/llvm/test/Transforms/SLPVectorizer/X86/load-merge.ll b/llvm/test/Transforms/SLPVectorizer/X86/load-merge.ll
new file mode 100644 (file)
index 0000000..57b5db5
--- /dev/null
+++ b/llvm/test/Transforms/SLPVectorizer/X86/load-merge.ll
@@ -0,0 +1,50 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -slp-vectorizer -slp-vectorize-hor -slp-vectorize-hor-store -S < %s -mtriple=x86_64-apple-macosx -mcpu=haswell | FileCheck %s
+
+;unsigned load_le32(unsigned char *data) {
+;    unsigned le32 = (data[0]<<0) | (data[1]<<8) | (data[2]<<16) | (data[3]<<24);
+;    return le32;
+;}
+
+define i32 @_Z9load_le32Ph(i8* nocapture readonly %data) {
+; CHECK-LABEL: @_Z9load_le32Ph(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = load i8, i8* [[DATA:%.*]], align 1
+; CHECK-NEXT:    [[CONV:%.*]] = zext i8 [[TMP0]] to i32
+; CHECK-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, i8* [[DATA]], i64 1
+; CHECK-NEXT:    [[TMP1:%.*]] = load i8, i8* [[ARRAYIDX1]], align 1
+; CHECK-NEXT:    [[CONV2:%.*]] = zext i8 [[TMP1]] to i32
+; CHECK-NEXT:    [[SHL3:%.*]] = shl nuw nsw i32 [[CONV2]], 8
+; CHECK-NEXT:    [[OR:%.*]] = or i32 [[SHL3]], [[CONV]]
+; CHECK-NEXT:    [[ARRAYIDX4:%.*]] = getelementptr inbounds i8, i8* [[DATA]], i64 2
+; CHECK-NEXT:    [[ARRAYIDX8:%.*]] = getelementptr inbounds i8, i8* [[DATA]], i64 3
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast i8* [[ARRAYIDX4]] to <2 x i8>*
+; CHECK-NEXT:    [[TMP3:%.*]] = load <2 x i8>, <2 x i8>* [[TMP2]], align 1
+; CHECK-NEXT:    [[TMP4:%.*]] = zext <2 x i8> [[TMP3]] to <2 x i32>
+; CHECK-NEXT:    [[TMP5:%.*]] = shl nuw <2 x i32> [[TMP4]], <i32 16, i32 24>
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <2 x i32> [[TMP5]], i32 0
+; CHECK-NEXT:    [[OR7:%.*]] = or i32 [[OR]], [[TMP6]]
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <2 x i32> [[TMP5]], i32 1
+; CHECK-NEXT:    [[OR11:%.*]] = or i32 [[OR7]], [[TMP7]]
+; CHECK-NEXT:    ret i32 [[OR11]]
+;
+entry:
+  %0 = load i8, i8* %data, align 1
+  %conv = zext i8 %0 to i32
+  %arrayidx1 = getelementptr inbounds i8, i8* %data, i64 1
+  %1 = load i8, i8* %arrayidx1, align 1
+  %conv2 = zext i8 %1 to i32
+  %shl3 = shl nuw nsw i32 %conv2, 8
+  %or = or i32 %shl3, %conv
+  %arrayidx4 = getelementptr inbounds i8, i8* %data, i64 2
+  %2 = load i8, i8* %arrayidx4, align 1
+  %conv5 = zext i8 %2 to i32
+  %shl6 = shl nuw nsw i32 %conv5, 16
+  %or7 = or i32 %or, %shl6
+  %arrayidx8 = getelementptr inbounds i8, i8* %data, i64 3
+  %3 = load i8, i8* %arrayidx8, align 1
+  %conv9 = zext i8 %3 to i32
+  %shl10 = shl nuw i32 %conv9, 24
+  %or11 = or i32 %or7, %shl10
+  ret i32 %or11
+}