diff --git a/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp b/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
--- a/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
+++ b/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
@@ -4592,7 +4592,19 @@
   };
 
   InstructionsState S = getSameOpcode(VL);
-  if (Depth == RecursionMaxDepth) {
+
+  // Gather if we hit the RecursionMaxDepth, unless this is a load (or z/sext of
+  // a load), in which case peek through to include it in the tree, without
+  // ballooning over-budget.
+  if (Depth >= RecursionMaxDepth &&
+      !(S.MainOp && isa<Instruction>(S.MainOp) && S.MainOp == S.AltOp &&
+        VL.size() >= 4 &&
+        (match(S.MainOp, m_Load(m_Value())) || all_of(VL, [&S](const Value *I) {
+           return match(I,
+                        m_OneUse(m_ZExtOrSExt(m_OneUse(m_Load(m_Value()))))) &&
+                  cast<Instruction>(I)->getOpcode() ==
+                      cast<Instruction>(S.MainOp)->getOpcode();
+         })))) {
     LLVM_DEBUG(dbgs() << "SLP: Gathering due to max recursion depth.\n");
     if (TryToFindDuplicates(S))
       newTreeEntry(VL, None /*not vectorized*/, S, UserTreeIdx,
diff --git a/llvm/test/Transforms/SLPVectorizer/AArch64/loadorder.ll b/llvm/test/Transforms/SLPVectorizer/AArch64/loadorder.ll
--- a/llvm/test/Transforms/SLPVectorizer/AArch64/loadorder.ll
+++ b/llvm/test/Transforms/SLPVectorizer/AArch64/loadorder.ll
@@ -1239,103 +1239,102 @@
 ; CHECK-NEXT:    [[IDX_EXT63:%.*]] = sext i32 [[ST2:%.*]] to i64
 ; CHECK-NEXT:    [[ARRAYIDX3:%.*]] = getelementptr inbounds i8, i8* [[P1:%.*]], i64 4
 ; CHECK-NEXT:    [[ARRAYIDX5:%.*]] = getelementptr inbounds i8, i8* [[P2:%.*]], i64 4
-; CHECK-NEXT:    [[TMP0:%.*]] = bitcast i8* [[P1]] to <4 x i8>*
-; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x i8>, <4 x i8>* [[TMP0]], align 1
-; CHECK-NEXT:    [[TMP2:%.*]] = bitcast i8* [[P2]] to <4 x i8>*
-; CHECK-NEXT:    [[TMP3:%.*]] = load <4 x i8>, <4 x i8>* [[TMP2]], align 1
 ; CHECK-NEXT:    [[ADD_PTR:%.*]] = getelementptr inbounds i8, i8* [[P1]], i64 [[IDX_EXT]]
 ; CHECK-NEXT:    [[ADD_PTR64:%.*]] = getelementptr inbounds i8, i8* [[P2]], i64 [[IDX_EXT63]]
 ; CHECK-NEXT:    [[ARRAYIDX3_1:%.*]] = getelementptr inbounds i8, i8* [[ADD_PTR]], i64 4
 ; CHECK-NEXT:    [[ARRAYIDX5_1:%.*]] = getelementptr inbounds i8, i8* [[ADD_PTR64]], i64 4
-; CHECK-NEXT:    [[TMP4:%.*]] = bitcast i8* [[ADD_PTR]] to <4 x i8>*
-; CHECK-NEXT:    [[TMP5:%.*]] = load <4 x i8>, <4 x i8>* [[TMP4]], align 1
-; CHECK-NEXT:    [[TMP6:%.*]] = bitcast i8* [[ADD_PTR64]] to <4 x i8>*
-; CHECK-NEXT:    [[TMP7:%.*]] = load <4 x i8>, <4 x i8>* [[TMP6]], align 1
 ; CHECK-NEXT:    [[ADD_PTR_1:%.*]] = getelementptr inbounds i8, i8* [[ADD_PTR]], i64 [[IDX_EXT]]
 ; CHECK-NEXT:    [[ADD_PTR64_1:%.*]] = getelementptr inbounds i8, i8* [[ADD_PTR64]], i64 [[IDX_EXT63]]
 ; CHECK-NEXT:    [[ARRAYIDX3_2:%.*]] = getelementptr inbounds i8, i8* [[ADD_PTR_1]], i64 4
 ; CHECK-NEXT:    [[ARRAYIDX5_2:%.*]] = getelementptr inbounds i8, i8* [[ADD_PTR64_1]], i64 4
-; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i8* [[ADD_PTR_1]] to <4 x i8>*
-; CHECK-NEXT:    [[TMP9:%.*]] = load <4 x i8>, <4 x i8>* [[TMP8]], align 1
-; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i8* [[ADD_PTR64_1]] to <4 x i8>*
-; CHECK-NEXT:    [[TMP11:%.*]] = load <4 x i8>, <4 x i8>* [[TMP10]], align 1
 ; CHECK-NEXT:    [[ADD_PTR_2:%.*]] = getelementptr inbounds i8, i8* [[ADD_PTR_1]], i64 [[IDX_EXT]]
 ; CHECK-NEXT:    [[ADD_PTR64_2:%.*]] = getelementptr inbounds i8, i8* [[ADD_PTR64_1]], i64 [[IDX_EXT63]]
 ; CHECK-NEXT:    [[ARRAYIDX3_3:%.*]] = getelementptr inbounds i8, i8* [[ADD_PTR_2]], i64 4
 ; CHECK-NEXT:    [[ARRAYIDX5_3:%.*]] = getelementptr inbounds i8, i8* [[ADD_PTR64_2]], i64 4
-; CHECK-NEXT:    [[TMP12:%.*]] = bitcast i8* [[ADD_PTR_2]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast i8* [[P1]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x i8>, <4 x i8>* [[TMP0]], align 1
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast i8* [[P2]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP3:%.*]] = load <4 x i8>, <4 x i8>* [[TMP2]], align 1
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast i8* [[ARRAYIDX3]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP5:%.*]] = load <4 x i8>, <4 x i8>* [[TMP4]], align 1
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast i8* [[ARRAYIDX5]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP7:%.*]] = load <4 x i8>, <4 x i8>* [[TMP6]], align 1
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i8* [[ADD_PTR]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP9:%.*]] = load <4 x i8>, <4 x i8>* [[TMP8]], align 1
+; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i8* [[ADD_PTR64]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP11:%.*]] = load <4 x i8>, <4 x i8>* [[TMP10]], align 1
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast i8* [[ARRAYIDX3_1]] to <4 x i8>*
 ; CHECK-NEXT:    [[TMP13:%.*]] = load <4 x i8>, <4 x i8>* [[TMP12]], align 1
-; CHECK-NEXT:    [[TMP14:%.*]] = bitcast i8* [[ADD_PTR64_2]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast i8* [[ARRAYIDX5_1]] to <4 x i8>*
 ; CHECK-NEXT:    [[TMP15:%.*]] = load <4 x i8>, <4 x i8>* [[TMP14]], align 1
-; CHECK-NEXT:    [[TMP16:%.*]] = bitcast i8* [[ARRAYIDX3]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast i8* [[ADD_PTR_1]] to <4 x i8>*
 ; CHECK-NEXT:    [[TMP17:%.*]] = load <4 x i8>, <4 x i8>* [[TMP16]], align 1
-; CHECK-NEXT:    [[TMP18:%.*]] = bitcast i8* [[ARRAYIDX3_1]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP18:%.*]] = bitcast i8* [[ADD_PTR64_1]] to <4 x i8>*
 ; CHECK-NEXT:    [[TMP19:%.*]] = load <4 x i8>, <4 x i8>* [[TMP18]], align 1
 ; CHECK-NEXT:    [[TMP20:%.*]] = bitcast i8* [[ARRAYIDX3_2]] to <4 x i8>*
 ; CHECK-NEXT:    [[TMP21:%.*]] = load <4 x i8>, <4 x i8>* [[TMP20]], align 1
-; CHECK-NEXT:    [[TMP22:%.*]] = bitcast i8* [[ARRAYIDX3_3]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP22:%.*]] = bitcast i8* [[ARRAYIDX5_2]] to <4 x i8>*
 ; CHECK-NEXT:    [[TMP23:%.*]] = load <4 x i8>, <4 x i8>* [[TMP22]], align 1
-; CHECK-NEXT:    [[TMP24:%.*]] = shufflevector <4 x i8> [[TMP23]], <4 x i8> [[TMP21]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP25:%.*]] = shufflevector <4 x i8> [[TMP19]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP26:%.*]] = shufflevector <16 x i8> [[TMP24]], <16 x i8> [[TMP25]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 16, i32 17, i32 18, i32 19, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP27:%.*]] = shufflevector <4 x i8> [[TMP17]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP28:%.*]] = shufflevector <16 x i8> [[TMP26]], <16 x i8> [[TMP27]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 16, i32 17, i32 18, i32 19>
-; CHECK-NEXT:    [[TMP29:%.*]] = zext <16 x i8> [[TMP28]] to <16 x i32>
-; CHECK-NEXT:    [[TMP30:%.*]] = bitcast i8* [[ARRAYIDX5]] to <4 x i8>*
-; CHECK-NEXT:    [[TMP31:%.*]] = load <4 x i8>, <4 x i8>* [[TMP30]], align 1
-; CHECK-NEXT:    [[TMP32:%.*]] = bitcast i8* [[ARRAYIDX5_1]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP24:%.*]] = bitcast i8* [[ADD_PTR_2]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP25:%.*]] = load <4 x i8>, <4 x i8>* [[TMP24]], align 1
+; CHECK-NEXT:    [[TMP26:%.*]] = shufflevector <4 x i8> [[TMP25]], <4 x i8> [[TMP17]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP27:%.*]] = shufflevector <4 x i8> [[TMP9]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP28:%.*]] = shufflevector <16 x i8> [[TMP26]], <16 x i8> [[TMP27]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 16, i32 17, i32 18, i32 19, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP29:%.*]] = shufflevector <4 x i8> [[TMP1]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP30:%.*]] = shufflevector <16 x i8> [[TMP28]], <16 x i8> [[TMP29]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 16, i32 17, i32 18, i32 19>
+; CHECK-NEXT:    [[TMP31:%.*]] = zext <16 x i8> [[TMP30]] to <16 x i32>
+; CHECK-NEXT:    [[TMP32:%.*]] = bitcast i8* [[ADD_PTR64_2]] to <4 x i8>*
 ; CHECK-NEXT:    [[TMP33:%.*]] = load <4 x i8>, <4 x i8>* [[TMP32]], align 1
-; CHECK-NEXT:    [[TMP34:%.*]] = bitcast i8* [[ARRAYIDX5_2]] to <4 x i8>*
-; CHECK-NEXT:    [[TMP35:%.*]] = load <4 x i8>, <4 x i8>* [[TMP34]], align 1
-; CHECK-NEXT:    [[TMP36:%.*]] = bitcast i8* [[ARRAYIDX5_3]] to <4 x i8>*
-; CHECK-NEXT:    [[TMP37:%.*]] = load <4 x i8>, <4 x i8>* [[TMP36]], align 1
-; CHECK-NEXT:    [[TMP38:%.*]] = shufflevector <4 x i8> [[TMP37]], <4 x i8> [[TMP35]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP39:%.*]] = shufflevector <4 x i8> [[TMP33]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP40:%.*]] = shufflevector <16 x i8> [[TMP38]], <16 x i8> [[TMP39]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 16, i32 17, i32 18, i32 19, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP41:%.*]] = shufflevector <4 x i8> [[TMP31]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP42:%.*]] = shufflevector <16 x i8> [[TMP40]], <16 x i8> [[TMP41]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 16, i32 17, i32 18, i32 19>
-; CHECK-NEXT:    [[TMP43:%.*]] = zext <16 x i8> [[TMP42]] to <16 x i32>
-; CHECK-NEXT:    [[TMP44:%.*]] = shufflevector <4 x i8> [[TMP13]], <4 x i8> [[TMP9]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP45:%.*]] = shufflevector <4 x i8> [[TMP5]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP46:%.*]] = shufflevector <16 x i8> [[TMP44]], <16 x i8> [[TMP45]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 16, i32 17, i32 18, i32 19, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP47:%.*]] = shufflevector <4 x i8> [[TMP1]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP48:%.*]] = shufflevector <16 x i8> [[TMP46]], <16 x i8> [[TMP47]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 16, i32 17, i32 18, i32 19>
-; CHECK-NEXT:    [[TMP49:%.*]] = zext <16 x i8> [[TMP48]] to <16 x i32>
-; CHECK-NEXT:    [[TMP50:%.*]] = shufflevector <4 x i8> [[TMP15]], <4 x i8> [[TMP11]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP51:%.*]] = shufflevector <4 x i8> [[TMP7]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP52:%.*]] = shufflevector <16 x i8> [[TMP50]], <16 x i8> [[TMP51]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 16, i32 17, i32 18, i32 19, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP53:%.*]] = shufflevector <4 x i8> [[TMP3]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
-; CHECK-NEXT:    [[TMP54:%.*]] = shufflevector <16 x i8> [[TMP52]], <16 x i8> [[TMP53]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 16, i32 17, i32 18, i32 19>
-; CHECK-NEXT:    [[TMP55:%.*]] = zext <16 x i8> [[TMP54]] to <16 x i32>
-; CHECK-NEXT:    [[TMP56:%.*]] = sub nsw <16 x i32> [[TMP49]], [[TMP55]]
-; CHECK-NEXT:    [[TMP57:%.*]] = sub nsw <16 x i32> [[TMP29]], [[TMP43]]
+; CHECK-NEXT:    [[TMP34:%.*]] = shufflevector <4 x i8> [[TMP33]], <4 x i8> [[TMP19]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP35:%.*]] = shufflevector <4 x i8> [[TMP11]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP36:%.*]] = shufflevector <16 x i8> [[TMP34]], <16 x i8> [[TMP35]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 16, i32 17, i32 18, i32 19, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP37:%.*]] = shufflevector <4 x i8> [[TMP3]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP38:%.*]] = shufflevector <16 x i8> [[TMP36]], <16 x i8> [[TMP37]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 16, i32 17, i32 18, i32 19>
+; CHECK-NEXT:    [[TMP39:%.*]] = zext <16 x i8> [[TMP38]] to <16 x i32>
+; CHECK-NEXT:    [[TMP40:%.*]] = sub nsw <16 x i32> [[TMP31]], [[TMP39]]
+; CHECK-NEXT:    [[TMP41:%.*]] = bitcast i8* [[ARRAYIDX3_3]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP42:%.*]] = load <4 x i8>, <4 x i8>* [[TMP41]], align 1
+; CHECK-NEXT:    [[TMP43:%.*]] = shufflevector <4 x i8> [[TMP42]], <4 x i8> [[TMP21]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP44:%.*]] = shufflevector <4 x i8> [[TMP13]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP45:%.*]] = shufflevector <16 x i8> [[TMP43]], <16 x i8> [[TMP44]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 16, i32 17, i32 18, i32 19, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP46:%.*]] = shufflevector <4 x i8> [[TMP5]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP47:%.*]] = shufflevector <16 x i8> [[TMP45]], <16 x i8> [[TMP46]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 16, i32 17, i32 18, i32 19>
+; CHECK-NEXT:    [[TMP48:%.*]] = zext <16 x i8> [[TMP47]] to <16 x i32>
+; CHECK-NEXT:    [[TMP49:%.*]] = bitcast i8* [[ARRAYIDX5_3]] to <4 x i8>*
+; CHECK-NEXT:    [[TMP50:%.*]] = load <4 x i8>, <4 x i8>* [[TMP49]], align 1
+; CHECK-NEXT:    [[TMP51:%.*]] = shufflevector <4 x i8> [[TMP50]], <4 x i8> [[TMP23]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP52:%.*]] = shufflevector <4 x i8> [[TMP15]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP53:%.*]] = shufflevector <16 x i8> [[TMP51]], <16 x i8> [[TMP52]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 16, i32 17, i32 18, i32 19, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP54:%.*]] = shufflevector <4 x i8> [[TMP7]], <4 x i8> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT:    [[TMP55:%.*]] = shufflevector <16 x i8> [[TMP53]], <16 x i8> [[TMP54]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 16, i32 17, i32 18, i32 19>
+; CHECK-NEXT:    [[TMP56:%.*]] = zext <16 x i8> [[TMP55]] to <16 x i32>
+; CHECK-NEXT:    [[TMP57:%.*]] = sub nsw <16 x i32> [[TMP48]], [[TMP56]]
 ; CHECK-NEXT:    [[TMP58:%.*]] = shl nsw <16 x i32> [[TMP57]], <i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16>
-; CHECK-NEXT:    [[TMP59:%.*]] = add nsw <16 x i32> [[TMP58]], [[TMP56]]
-; CHECK-NEXT:    [[TMP60:%.*]] = shufflevector <16 x i32> [[TMP59]], <16 x i32> poison, <16 x i32> <i32 3, i32 7, i32 11, i32 15, i32 6, i32 2, i32 10, i32 14, i32 5, i32 1, i32 9, i32 13, i32 4, i32 0, i32 8, i32 12>
-; CHECK-NEXT:    [[TMP61:%.*]] = shufflevector <16 x i32> [[TMP60]], <16 x i32> poison, <16 x i32> <i32 5, i32 4, i32 6, i32 7, i32 1, i32 0, i32 2, i32 3, i32 12, i32 13, i32 14, i32 15, i32 8, i32 9, i32 10, i32 11>
-; CHECK-NEXT:    [[TMP62:%.*]] = add nsw <16 x i32> [[TMP60]], [[TMP61]]
-; CHECK-NEXT:    [[TMP63:%.*]] = sub nsw <16 x i32> [[TMP60]], [[TMP61]]
-; CHECK-NEXT:    [[TMP64:%.*]] = shufflevector <16 x i32> [[TMP62]], <16 x i32> [[TMP63]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 20, i32 21, i32 22, i32 23, i32 8, i32 9, i32 10, i32 11, i32 28, i32 29, i32 30, i32 31>
-; CHECK-NEXT:    [[TMP65:%.*]] = shufflevector <16 x i32> [[TMP64]], <16 x i32> poison, <16 x i32> <i32 9, i32 8, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 1, i32 0, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
-; CHECK-NEXT:    [[TMP66:%.*]] = add nsw <16 x i32> [[TMP64]], [[TMP65]]
-; CHECK-NEXT:    [[TMP67:%.*]] = sub nsw <16 x i32> [[TMP64]], [[TMP65]]
-; CHECK-NEXT:    [[TMP68:%.*]] = shufflevector <16 x i32> [[TMP66]], <16 x i32> [[TMP67]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
-; CHECK-NEXT:    [[TMP69:%.*]] = shufflevector <16 x i32> [[TMP68]], <16 x i32> poison, <16 x i32> <i32 1, i32 0, i32 3, i32 2, i32 5, i32 4, i32 7, i32 6, i32 9, i32 8, i32 11, i32 10, i32 13, i32 12, i32 15, i32 14>
-; CHECK-NEXT:    [[TMP70:%.*]] = add nsw <16 x i32> [[TMP68]], [[TMP69]]
-; CHECK-NEXT:    [[TMP71:%.*]] = sub nsw <16 x i32> [[TMP68]], [[TMP69]]
-; CHECK-NEXT:    [[TMP72:%.*]] = shufflevector <16 x i32> [[TMP70]], <16 x i32> [[TMP71]], <16 x i32> <i32 0, i32 17, i32 2, i32 19, i32 20, i32 5, i32 6, i32 23, i32 24, i32 9, i32 10, i32 27, i32 28, i32 13, i32 14, i32 31>
-; CHECK-NEXT:    [[TMP73:%.*]] = shufflevector <16 x i32> [[TMP72]], <16 x i32> poison, <16 x i32> <i32 2, i32 3, i32 0, i32 1, i32 7, i32 6, i32 5, i32 4, i32 11, i32 10, i32 9, i32 8, i32 15, i32 14, i32 13, i32 12>
-; CHECK-NEXT:    [[TMP74:%.*]] = add nsw <16 x i32> [[TMP72]], [[TMP73]]
-; CHECK-NEXT:    [[TMP75:%.*]] = sub nsw <16 x i32> [[TMP72]], [[TMP73]]
-; CHECK-NEXT:    [[TMP76:%.*]] = shufflevector <16 x i32> [[TMP74]], <16 x i32> [[TMP75]], <16 x i32> <i32 0, i32 1, i32 18, i32 19, i32 4, i32 5, i32 22, i32 23, i32 8, i32 9, i32 26, i32 27, i32 12, i32 13, i32 30, i32 31>
-; CHECK-NEXT:    [[TMP77:%.*]] = lshr <16 x i32> [[TMP76]], <i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15>
-; CHECK-NEXT:    [[TMP78:%.*]] = and <16 x i32> [[TMP77]], <i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537>
-; CHECK-NEXT:    [[TMP79:%.*]] = mul nuw <16 x i32> [[TMP78]], <i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535>
-; CHECK-NEXT:    [[TMP80:%.*]] = add <16 x i32> [[TMP79]], [[TMP76]]
-; CHECK-NEXT:    [[TMP81:%.*]] = xor <16 x i32> [[TMP80]], [[TMP79]]
-; CHECK-NEXT:    [[TMP82:%.*]] = call i32 @llvm.vector.reduce.add.v16i32(<16 x i32> [[TMP81]])
-; CHECK-NEXT:    [[CONV118:%.*]] = and i32 [[TMP82]], 65535
-; CHECK-NEXT:    [[SHR:%.*]] = lshr i32 [[TMP82]], 16
+; CHECK-NEXT:    [[TMP59:%.*]] = add nsw <16 x i32> [[TMP58]], [[TMP40]]
+; CHECK-NEXT:    [[TMP60:%.*]] = shufflevector <16 x i32> [[TMP59]], <16 x i32> poison, <16 x i32> <i32 1, i32 0, i32 3, i32 2, i32 5, i32 4, i32 7, i32 6, i32 9, i32 8, i32 11, i32 10, i32 13, i32 12, i32 15, i32 14>
+; CHECK-NEXT:    [[TMP61:%.*]] = add nsw <16 x i32> [[TMP59]], [[TMP60]]
+; CHECK-NEXT:    [[TMP62:%.*]] = sub nsw <16 x i32> [[TMP59]], [[TMP60]]
+; CHECK-NEXT:    [[TMP63:%.*]] = shufflevector <16 x i32> [[TMP61]], <16 x i32> [[TMP62]], <16 x i32> <i32 3, i32 7, i32 11, i32 15, i32 22, i32 18, i32 26, i32 30, i32 5, i32 1, i32 9, i32 13, i32 20, i32 16, i32 24, i32 28>
+; CHECK-NEXT:    [[TMP64:%.*]] = shufflevector <16 x i32> [[TMP63]], <16 x i32> poison, <16 x i32> <i32 9, i32 8, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 1, i32 0, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+; CHECK-NEXT:    [[TMP65:%.*]] = add nsw <16 x i32> [[TMP63]], [[TMP64]]
+; CHECK-NEXT:    [[TMP66:%.*]] = sub nsw <16 x i32> [[TMP63]], [[TMP64]]
+; CHECK-NEXT:    [[TMP67:%.*]] = shufflevector <16 x i32> [[TMP65]], <16 x i32> [[TMP66]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
+; CHECK-NEXT:    [[TMP68:%.*]] = shufflevector <16 x i32> [[TMP67]], <16 x i32> poison, <16 x i32> <i32 1, i32 0, i32 3, i32 2, i32 5, i32 4, i32 7, i32 6, i32 9, i32 8, i32 11, i32 10, i32 13, i32 12, i32 15, i32 14>
+; CHECK-NEXT:    [[TMP69:%.*]] = add nsw <16 x i32> [[TMP67]], [[TMP68]]
+; CHECK-NEXT:    [[TMP70:%.*]] = sub nsw <16 x i32> [[TMP67]], [[TMP68]]
+; CHECK-NEXT:    [[TMP71:%.*]] = shufflevector <16 x i32> [[TMP69]], <16 x i32> [[TMP70]], <16 x i32> <i32 0, i32 17, i32 2, i32 19, i32 20, i32 5, i32 6, i32 23, i32 24, i32 9, i32 10, i32 27, i32 28, i32 13, i32 14, i32 31>
+; CHECK-NEXT:    [[TMP72:%.*]] = shufflevector <16 x i32> [[TMP71]], <16 x i32> poison, <16 x i32> <i32 2, i32 3, i32 0, i32 1, i32 7, i32 6, i32 5, i32 4, i32 11, i32 10, i32 9, i32 8, i32 15, i32 14, i32 13, i32 12>
+; CHECK-NEXT:    [[TMP73:%.*]] = add nsw <16 x i32> [[TMP71]], [[TMP72]]
+; CHECK-NEXT:    [[TMP74:%.*]] = sub nsw <16 x i32> [[TMP71]], [[TMP72]]
+; CHECK-NEXT:    [[TMP75:%.*]] = shufflevector <16 x i32> [[TMP73]], <16 x i32> [[TMP74]], <16 x i32> <i32 0, i32 1, i32 18, i32 19, i32 4, i32 5, i32 22, i32 23, i32 8, i32 9, i32 26, i32 27, i32 12, i32 13, i32 30, i32 31>
+; CHECK-NEXT:    [[TMP76:%.*]] = lshr <16 x i32> [[TMP75]], <i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15, i32 15>
+; CHECK-NEXT:    [[TMP77:%.*]] = and <16 x i32> [[TMP76]], <i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537, i32 65537>
+; CHECK-NEXT:    [[TMP78:%.*]] = mul nuw <16 x i32> [[TMP77]], <i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535, i32 65535>
+; CHECK-NEXT:    [[TMP79:%.*]] = add <16 x i32> [[TMP78]], [[TMP75]]
+; CHECK-NEXT:    [[TMP80:%.*]] = xor <16 x i32> [[TMP79]], [[TMP78]]
+; CHECK-NEXT:    [[TMP81:%.*]] = call i32 @llvm.vector.reduce.add.v16i32(<16 x i32> [[TMP80]])
+; CHECK-NEXT:    [[CONV118:%.*]] = and i32 [[TMP81]], 65535
+; CHECK-NEXT:    [[SHR:%.*]] = lshr i32 [[TMP81]], 16
 ; CHECK-NEXT:    [[ADD119:%.*]] = add nuw nsw i32 [[CONV118]], [[SHR]]
 ; CHECK-NEXT:    [[SHR120:%.*]] = lshr i32 [[ADD119]], 1
 ; CHECK-NEXT:    ret i32 [[SHR120]]