[llvm] 883eb88 - [LV] Add extra uniformity tests with LSHR and AND.

[Florian Hahn via llvm-commits]

Author: Florian Hahn
Date: 2023-04-26T19:51:35+01:00
New Revision: 883eb88caed04b269da7ba69265fd7c4dc815231

URL: https://github.com/llvm/llvm-project/commit/883eb88caed04b269da7ba69265fd7c4dc815231
DIFF: https://github.com/llvm/llvm-project/commit/883eb88caed04b269da7ba69265fd7c4dc815231.diff

LOG: [LV] Add extra uniformity tests with LSHR and AND.

Extra tests for D148841 based on the tests added in
95539186c82604f783.

Added: 
    llvm/test/Transforms/LoopVectorize/uniform_across_vf_induction1_and.ll
    llvm/test/Transforms/LoopVectorize/uniform_across_vf_induction1_lshr.ll

Modified: 
    

Removed: 
    


################################################################################
diff  --git a/llvm/test/Transforms/LoopVectorize/uniform_across_vf_induction1_and.ll b/llvm/test/Transforms/LoopVectorize/uniform_across_vf_induction1_and.ll
new file mode 100644
index 000000000000..be9a87be9461
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/uniform_across_vf_induction1_and.ll
@@ -0,0 +1,756 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 2
+; RUN: opt -passes=loop-vectorize -force-vector-interleave=1 -force-vector-width=2 %s -S | FileCheck %s
+
+; Tests for checking uniformity within a VF.
+
+; for (iv = 0 ; ; iv += 1) B[iv] = A[i&-1] + 42;
+define void @ld_and_neg1_step1_start0_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_and_neg1_step1_start0_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = and i64 [[TMP0]], -1
+; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i64, ptr [[TMP2]], i32 0
+; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <2 x i64>, ptr [[TMP3]], align 8
+; CHECK-NEXT:    [[TMP4:%.*]] = add nsw <2 x i64> [[WIDE_LOAD]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[TMP5]], i32 0
+; CHECK-NEXT:    store <2 x i64> [[TMP4]], ptr [[TMP6]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1000, 1000
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1000, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[AND:%.*]] = and i64 [[IV]], -1
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[AND]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv_next, %loop ]
+  %and = and i64 %iv, -1
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %and
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 1
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 0 ; ; iv += 1) B[iv] = A[iv&-2] + 42;
+define void @ld_and_neg2_step1_start0_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_and_neg2_step1_start0_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 0, i64 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = and <2 x i64> [[VEC_IND]], <i64 -2, i64 -2>
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <2 x i64> [[TMP1]], i32 0
+; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <2 x i64> [[TMP1]], i32 1
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP4]]
+; CHECK-NEXT:    [[TMP6:%.*]] = load i64, ptr [[TMP3]], align 8
+; CHECK-NEXT:    [[TMP7:%.*]] = load i64, ptr [[TMP5]], align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <2 x i64> poison, i64 [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <2 x i64> [[TMP8]], i64 [[TMP7]], i32 1
+; CHECK-NEXT:    [[TMP10:%.*]] = add nsw <2 x i64> [[TMP9]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP11:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[TMP11]], i32 0
+; CHECK-NEXT:    store <2 x i64> [[TMP10]], ptr [[TMP12]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 2, i64 2>
+; CHECK-NEXT:    [[TMP13:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[TMP13]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1000, 1000
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1000, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[AND:%.*]] = and i64 [[IV]], -2
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[AND]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv_next, %loop ]
+  %and = and i64 %iv, -2
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %and
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 1
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 0 ; ; iv += 1) B[iv] = A[iv&-3] + 42;
+define void @ld_and_neg3_step1_start0_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_and_neg3_step1_start0_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 0, i64 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = and <2 x i64> [[VEC_IND]], <i64 -3, i64 -3>
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <2 x i64> [[TMP1]], i32 0
+; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <2 x i64> [[TMP1]], i32 1
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP4]]
+; CHECK-NEXT:    [[TMP6:%.*]] = load i64, ptr [[TMP3]], align 8
+; CHECK-NEXT:    [[TMP7:%.*]] = load i64, ptr [[TMP5]], align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <2 x i64> poison, i64 [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <2 x i64> [[TMP8]], i64 [[TMP7]], i32 1
+; CHECK-NEXT:    [[TMP10:%.*]] = add nsw <2 x i64> [[TMP9]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP11:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[TMP11]], i32 0
+; CHECK-NEXT:    store <2 x i64> [[TMP10]], ptr [[TMP12]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 2, i64 2>
+; CHECK-NEXT:    [[TMP13:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[TMP13]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1000, 1000
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1000, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[AND:%.*]] = and i64 [[IV]], -3
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[AND]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP7:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv_next, %loop ]
+  %and = and i64 %iv, -3
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %and
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 1
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 0 ; ; iv += 2) B[iv] = A[iv&-1] + 42;
+define void @ld_and_neg1_step2_start0_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_and_neg1_step2_start0_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 0, i64 2>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = mul i64 [[INDEX]], 2
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[OFFSET_IDX]], 2
+; CHECK-NEXT:    [[TMP2:%.*]] = and <2 x i64> [[VEC_IND]], <i64 -1, i64 -1>
+; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <2 x i64> [[TMP2]], i32 0
+; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP3]]
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <2 x i64> [[TMP2]], i32 1
+; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP5]]
+; CHECK-NEXT:    [[TMP7:%.*]] = load i64, ptr [[TMP4]], align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = load i64, ptr [[TMP6]], align 8
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <2 x i64> poison, i64 [[TMP7]], i32 0
+; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <2 x i64> [[TMP9]], i64 [[TMP8]], i32 1
+; CHECK-NEXT:    [[TMP11:%.*]] = add nsw <2 x i64> [[TMP10]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <2 x i64> [[TMP11]], i32 0
+; CHECK-NEXT:    store i64 [[TMP14]], ptr [[TMP12]], align 8
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <2 x i64> [[TMP11]], i32 1
+; CHECK-NEXT:    store i64 [[TMP15]], ptr [[TMP13]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 4, i64 4>
+; CHECK-NEXT:    [[TMP16:%.*]] = icmp eq i64 [[INDEX_NEXT]], 500
+; CHECK-NEXT:    br i1 [[TMP16]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 500, 500
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1000, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[AND:%.*]] = and i64 [[IV]], -1
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[AND]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 2
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP9:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv_next, %loop ]
+  %and = and i64 %iv, -1
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %and
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 2
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 0 ; ; iv += 2) B[iv] = A[iv&-2] + 42;
+define void @ld_and_neg2_step2_start0_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_and_neg2_step2_start0_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = mul i64 [[INDEX]], 2
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[OFFSET_IDX]], 2
+; CHECK-NEXT:    [[TMP2:%.*]] = and i64 [[TMP0]], 1
+; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = load i64, ptr [[TMP3]], align 8
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i64> poison, i64 [[TMP4]], i64 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x i64> [[BROADCAST_SPLATINSERT]], <2 x i64> poison, <2 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP5:%.*]] = add nsw <2 x i64> [[BROADCAST_SPLAT]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP7:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <2 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    store i64 [[TMP8]], ptr [[TMP6]], align 8
+; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <2 x i64> [[TMP5]], i32 1
+; CHECK-NEXT:    store i64 [[TMP9]], ptr [[TMP7]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT]], 500
+; CHECK-NEXT:    br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 500, 500
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1000, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[AND:%.*]] = and i64 [[IV]], 1
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[AND]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 2
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP11:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv_next, %loop ]
+  %and = and i64 %iv, 1
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %and
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 2
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 0 ; ; iv += 3) B[iv] = A[iv&-1] + 42;
+define void @ld_and_neg1_step3_start0_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_and_neg1_step3_start0_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 0, i64 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = mul i64 [[INDEX]], 3
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[OFFSET_IDX]], 3
+; CHECK-NEXT:    [[TMP2:%.*]] = and <2 x i64> [[VEC_IND]], <i64 -1, i64 -1>
+; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <2 x i64> [[TMP2]], i32 0
+; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP3]]
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <2 x i64> [[TMP2]], i32 1
+; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP5]]
+; CHECK-NEXT:    [[TMP7:%.*]] = load i64, ptr [[TMP4]], align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = load i64, ptr [[TMP6]], align 8
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <2 x i64> poison, i64 [[TMP7]], i32 0
+; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <2 x i64> [[TMP9]], i64 [[TMP8]], i32 1
+; CHECK-NEXT:    [[TMP11:%.*]] = add nsw <2 x i64> [[TMP10]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <2 x i64> [[TMP11]], i32 0
+; CHECK-NEXT:    store i64 [[TMP14]], ptr [[TMP12]], align 8
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <2 x i64> [[TMP11]], i32 1
+; CHECK-NEXT:    store i64 [[TMP15]], ptr [[TMP13]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 6, i64 6>
+; CHECK-NEXT:    [[TMP16:%.*]] = icmp eq i64 [[INDEX_NEXT]], 332
+; CHECK-NEXT:    br i1 [[TMP16]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 333, 332
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 996, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[AND:%.*]] = and i64 [[IV]], -1
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[AND]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 3
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP13:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv_next, %loop ]
+  %and = and i64 %iv, -1
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %and
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 3
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 0 ; ; iv += 3) B[iv] = A[i&-2] + 42;
+define void @ld_and_neg2_step3_start0_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_and_neg2_step3_start0_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 0, i64 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = mul i64 [[INDEX]], 3
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[OFFSET_IDX]], 3
+; CHECK-NEXT:    [[TMP2:%.*]] = and <2 x i64> [[VEC_IND]], <i64 -2, i64 -2>
+; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <2 x i64> [[TMP2]], i32 0
+; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP3]]
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <2 x i64> [[TMP2]], i32 1
+; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP5]]
+; CHECK-NEXT:    [[TMP7:%.*]] = load i64, ptr [[TMP4]], align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = load i64, ptr [[TMP6]], align 8
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <2 x i64> poison, i64 [[TMP7]], i32 0
+; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <2 x i64> [[TMP9]], i64 [[TMP8]], i32 1
+; CHECK-NEXT:    [[TMP11:%.*]] = add nsw <2 x i64> [[TMP10]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <2 x i64> [[TMP11]], i32 0
+; CHECK-NEXT:    store i64 [[TMP14]], ptr [[TMP12]], align 8
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <2 x i64> [[TMP11]], i32 1
+; CHECK-NEXT:    store i64 [[TMP15]], ptr [[TMP13]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 6, i64 6>
+; CHECK-NEXT:    [[TMP16:%.*]] = icmp eq i64 [[INDEX_NEXT]], 332
+; CHECK-NEXT:    br i1 [[TMP16]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP14:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 333, 332
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 996, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[AND:%.*]] = and i64 [[IV]], -2
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[AND]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 3
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP15:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv_next, %loop ]
+  %and = and i64 %iv, -2
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %and
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 3
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 1 ; ; iv += 1) B[iv] = A[iv&-2] + 42;
+define void @ld_and_neg2_step1_start1_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_and_neg2_step1_start1_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 1, i64 2>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i64 1, [[INDEX]]
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = and <2 x i64> [[VEC_IND]], <i64 -2, i64 -2>
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <2 x i64> [[TMP1]], i32 0
+; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <2 x i64> [[TMP1]], i32 1
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP4]]
+; CHECK-NEXT:    [[TMP6:%.*]] = load i64, ptr [[TMP3]], align 8
+; CHECK-NEXT:    [[TMP7:%.*]] = load i64, ptr [[TMP5]], align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <2 x i64> poison, i64 [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <2 x i64> [[TMP8]], i64 [[TMP7]], i32 1
+; CHECK-NEXT:    [[TMP10:%.*]] = add nsw <2 x i64> [[TMP9]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP11:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[TMP11]], i32 0
+; CHECK-NEXT:    store <2 x i64> [[TMP10]], ptr [[TMP12]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 2, i64 2>
+; CHECK-NEXT:    [[TMP13:%.*]] = icmp eq i64 [[INDEX_NEXT]], 998
+; CHECK-NEXT:    br i1 [[TMP13]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP16:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 999, 998
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 999, [[MIDDLE_BLOCK]] ], [ 1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[AND:%.*]] = and i64 [[IV]], -2
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[AND]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP17:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 1, %entry ], [ %iv_next, %loop ]
+  %and = and i64 %iv, -2
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %and
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 1
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 1 ; ; iv += 2) B[iv] = A[iv&-2] + 42;
+define void @ld_and_neg2_step2_start1_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_and_neg2_step2_start1_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 1, i64 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = mul i64 [[INDEX]], 2
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i64 1, [[TMP0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP2:%.*]] = add i64 [[OFFSET_IDX]], 2
+; CHECK-NEXT:    [[TMP3:%.*]] = and <2 x i64> [[VEC_IND]], <i64 -2, i64 -2>
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i32 0
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP4]]
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <2 x i64> [[TMP3]], i32 1
+; CHECK-NEXT:    [[TMP7:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP6]]
+; CHECK-NEXT:    [[TMP8:%.*]] = load i64, ptr [[TMP5]], align 8
+; CHECK-NEXT:    [[TMP9:%.*]] = load i64, ptr [[TMP7]], align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <2 x i64> poison, i64 [[TMP8]], i32 0
+; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <2 x i64> [[TMP10]], i64 [[TMP9]], i32 1
+; CHECK-NEXT:    [[TMP12:%.*]] = add nsw <2 x i64> [[TMP11]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <2 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP15]], ptr [[TMP13]], align 8
+; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <2 x i64> [[TMP12]], i32 1
+; CHECK-NEXT:    store i64 [[TMP16]], ptr [[TMP14]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 4, i64 4>
+; CHECK-NEXT:    [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], 498
+; CHECK-NEXT:    br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP18:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 499, 498
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 997, [[MIDDLE_BLOCK]] ], [ 1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[AND:%.*]] = and i64 [[IV]], -2
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[AND]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 2
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP19:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 1, %entry ], [ %iv_next, %loop ]
+  %and = and i64 %iv, -2
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %and
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 2
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 1 ; ; iv += 3) B[iv] = A[iv&-2] + 42;
+define void @ld_and_neg2_step3_start1_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_and_neg2_step3_start1_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 1, i64 4>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = mul i64 [[INDEX]], 3
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i64 1, [[TMP0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP2:%.*]] = add i64 [[OFFSET_IDX]], 3
+; CHECK-NEXT:    [[TMP3:%.*]] = and <2 x i64> [[VEC_IND]], <i64 -2, i64 -2>
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i32 0
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP4]]
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <2 x i64> [[TMP3]], i32 1
+; CHECK-NEXT:    [[TMP7:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP6]]
+; CHECK-NEXT:    [[TMP8:%.*]] = load i64, ptr [[TMP5]], align 8
+; CHECK-NEXT:    [[TMP9:%.*]] = load i64, ptr [[TMP7]], align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <2 x i64> poison, i64 [[TMP8]], i32 0
+; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <2 x i64> [[TMP10]], i64 [[TMP9]], i32 1
+; CHECK-NEXT:    [[TMP12:%.*]] = add nsw <2 x i64> [[TMP11]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <2 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP15]], ptr [[TMP13]], align 8
+; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <2 x i64> [[TMP12]], i32 1
+; CHECK-NEXT:    store i64 [[TMP16]], ptr [[TMP14]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 6, i64 6>
+; CHECK-NEXT:    [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], 332
+; CHECK-NEXT:    br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP20:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 333, 332
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 997, [[MIDDLE_BLOCK]] ], [ 1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[AND:%.*]] = and i64 [[IV]], -2
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[AND]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 3
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP21:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 1, %entry ], [ %iv_next, %loop ]
+  %and = and i64 %iv, -2
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %and
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 3
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 1 ; ; iv += 3) B[iv] = A[iv&-3] + 42;
+define void @ld_and_neg3_step3_start1_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_and_neg3_step3_start1_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 1, i64 4>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = mul i64 [[INDEX]], 3
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i64 1, [[TMP0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP2:%.*]] = add i64 [[OFFSET_IDX]], 3
+; CHECK-NEXT:    [[TMP3:%.*]] = and <2 x i64> [[VEC_IND]], <i64 -3, i64 -3>
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i32 0
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP4]]
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <2 x i64> [[TMP3]], i32 1
+; CHECK-NEXT:    [[TMP7:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP6]]
+; CHECK-NEXT:    [[TMP8:%.*]] = load i64, ptr [[TMP5]], align 8
+; CHECK-NEXT:    [[TMP9:%.*]] = load i64, ptr [[TMP7]], align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <2 x i64> poison, i64 [[TMP8]], i32 0
+; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <2 x i64> [[TMP10]], i64 [[TMP9]], i32 1
+; CHECK-NEXT:    [[TMP12:%.*]] = add nsw <2 x i64> [[TMP11]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <2 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP15]], ptr [[TMP13]], align 8
+; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <2 x i64> [[TMP12]], i32 1
+; CHECK-NEXT:    store i64 [[TMP16]], ptr [[TMP14]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 6, i64 6>
+; CHECK-NEXT:    [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], 332
+; CHECK-NEXT:    br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP22:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 333, 332
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 997, [[MIDDLE_BLOCK]] ], [ 1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[AND:%.*]] = and i64 [[IV]], -3
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[AND]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 3
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP23:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 1, %entry ], [ %iv_next, %loop ]
+  %and = and i64 %iv, -3
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %and
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 3
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}

diff  --git a/llvm/test/Transforms/LoopVectorize/uniform_across_vf_induction1_lshr.ll b/llvm/test/Transforms/LoopVectorize/uniform_across_vf_induction1_lshr.ll
new file mode 100644
index 000000000000..737336377adb
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/uniform_across_vf_induction1_lshr.ll
@@ -0,0 +1,748 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 2
+; RUN: opt -passes=loop-vectorize -force-vector-interleave=1 -force-vector-width=2 %s -S | FileCheck %s
+
+; Tests for checking uniformity within a VF.
+
+; for (iv = 0 ; ; iv += 1) B[iv] = A[iv>>0] + 42;
+define void @ld_lshr0_step1_start0_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_lshr0_step1_start0_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = lshr i64 [[TMP0]], 0
+; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i64, ptr [[TMP2]], i32 0
+; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <2 x i64>, ptr [[TMP3]], align 8
+; CHECK-NEXT:    [[TMP4:%.*]] = add nsw <2 x i64> [[WIDE_LOAD]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[TMP5]], i32 0
+; CHECK-NEXT:    store <2 x i64> [[TMP4]], ptr [[TMP6]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1000, 1000
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1000, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[LSHR:%.*]] = lshr i64 [[IV]], 0
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[LSHR]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv_next, %loop ]
+  %lshr = lshr i64 %iv, 0
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %lshr
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 1
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 0 ; ; iv += 1) B[iv] = A[iv>>1] + 42;
+define void @ld_lshr1_step1_start0_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_lshr1_step1_start0_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 0, i64 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = lshr <2 x i64> [[VEC_IND]], <i64 1, i64 1>
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <2 x i64> [[TMP1]], i32 0
+; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <2 x i64> [[TMP1]], i32 1
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP4]]
+; CHECK-NEXT:    [[TMP6:%.*]] = load i64, ptr [[TMP3]], align 8
+; CHECK-NEXT:    [[TMP7:%.*]] = load i64, ptr [[TMP5]], align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <2 x i64> poison, i64 [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <2 x i64> [[TMP8]], i64 [[TMP7]], i32 1
+; CHECK-NEXT:    [[TMP10:%.*]] = add nsw <2 x i64> [[TMP9]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP11:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[TMP11]], i32 0
+; CHECK-NEXT:    store <2 x i64> [[TMP10]], ptr [[TMP12]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 2, i64 2>
+; CHECK-NEXT:    [[TMP13:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[TMP13]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1000, 1000
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1000, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[LSHR:%.*]] = lshr i64 [[IV]], 1
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[LSHR]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv_next, %loop ]
+  %lshr = lshr i64 %iv, 1
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %lshr
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 1
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 0 ; ; iv += 1) B[iv] = A[iv>>2] + 42;
+define void @ld_lshr2_step1_start0_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_lshr2_step1_start0_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 0, i64 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = lshr <2 x i64> [[VEC_IND]], <i64 2, i64 2>
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <2 x i64> [[TMP1]], i32 0
+; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <2 x i64> [[TMP1]], i32 1
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP4]]
+; CHECK-NEXT:    [[TMP6:%.*]] = load i64, ptr [[TMP3]], align 8
+; CHECK-NEXT:    [[TMP7:%.*]] = load i64, ptr [[TMP5]], align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <2 x i64> poison, i64 [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <2 x i64> [[TMP8]], i64 [[TMP7]], i32 1
+; CHECK-NEXT:    [[TMP10:%.*]] = add nsw <2 x i64> [[TMP9]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP11:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[TMP11]], i32 0
+; CHECK-NEXT:    store <2 x i64> [[TMP10]], ptr [[TMP12]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 2, i64 2>
+; CHECK-NEXT:    [[TMP13:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[TMP13]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1000, 1000
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1000, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[LSHR:%.*]] = lshr i64 [[IV]], 2
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[LSHR]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP7:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv_next, %loop ]
+  %lshr = lshr i64 %iv, 2
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %lshr
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 1
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 0 ; ; iv += 2) B[iv] = A[iv>>0] + 42;
+define void @ld_lshr0_step2_start0_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_lshr0_step2_start0_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 0, i64 2>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = mul i64 [[INDEX]], 2
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[OFFSET_IDX]], 2
+; CHECK-NEXT:    [[TMP2:%.*]] = lshr <2 x i64> [[VEC_IND]], zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <2 x i64> [[TMP2]], i32 0
+; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP3]]
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <2 x i64> [[TMP2]], i32 1
+; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP5]]
+; CHECK-NEXT:    [[TMP7:%.*]] = load i64, ptr [[TMP4]], align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = load i64, ptr [[TMP6]], align 8
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <2 x i64> poison, i64 [[TMP7]], i32 0
+; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <2 x i64> [[TMP9]], i64 [[TMP8]], i32 1
+; CHECK-NEXT:    [[TMP11:%.*]] = add nsw <2 x i64> [[TMP10]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <2 x i64> [[TMP11]], i32 0
+; CHECK-NEXT:    store i64 [[TMP14]], ptr [[TMP12]], align 8
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <2 x i64> [[TMP11]], i32 1
+; CHECK-NEXT:    store i64 [[TMP15]], ptr [[TMP13]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 4, i64 4>
+; CHECK-NEXT:    [[TMP16:%.*]] = icmp eq i64 [[INDEX_NEXT]], 500
+; CHECK-NEXT:    br i1 [[TMP16]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 500, 500
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1000, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[LSHR:%.*]] = lshr i64 [[IV]], 0
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[LSHR]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 2
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP9:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv_next, %loop ]
+  %lshr = lshr i64 %iv, 0
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %lshr
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 2
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 0 ; ; iv += 2) B[iv] = A[iv>>1] + 42;
+define void @ld_lshr1_step2_start0_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_lshr1_step2_start0_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = mul i64 [[INDEX]], 2
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[OFFSET_IDX]], 2
+; CHECK-NEXT:    [[TMP2:%.*]] = lshr i64 [[TMP0]], 1
+; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i64, ptr [[TMP3]], i32 0
+; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <2 x i64>, ptr [[TMP4]], align 8
+; CHECK-NEXT:    [[TMP5:%.*]] = add nsw <2 x i64> [[WIDE_LOAD]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP7:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <2 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    store i64 [[TMP8]], ptr [[TMP6]], align 8
+; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <2 x i64> [[TMP5]], i32 1
+; CHECK-NEXT:    store i64 [[TMP9]], ptr [[TMP7]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT]], 500
+; CHECK-NEXT:    br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 500, 500
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1000, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[LSHR:%.*]] = lshr i64 [[IV]], 1
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[LSHR]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 2
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP11:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv_next, %loop ]
+  %lshr = lshr i64 %iv, 1
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %lshr
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 2
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 0 ; ; iv += 3) B[iv] = A[iv>>0] + 42;
+define void @ld_lshr0_step3_start0_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_lshr0_step3_start0_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 0, i64 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = mul i64 [[INDEX]], 3
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[OFFSET_IDX]], 3
+; CHECK-NEXT:    [[TMP2:%.*]] = lshr <2 x i64> [[VEC_IND]], zeroinitializer
+; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <2 x i64> [[TMP2]], i32 0
+; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP3]]
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <2 x i64> [[TMP2]], i32 1
+; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP5]]
+; CHECK-NEXT:    [[TMP7:%.*]] = load i64, ptr [[TMP4]], align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = load i64, ptr [[TMP6]], align 8
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <2 x i64> poison, i64 [[TMP7]], i32 0
+; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <2 x i64> [[TMP9]], i64 [[TMP8]], i32 1
+; CHECK-NEXT:    [[TMP11:%.*]] = add nsw <2 x i64> [[TMP10]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <2 x i64> [[TMP11]], i32 0
+; CHECK-NEXT:    store i64 [[TMP14]], ptr [[TMP12]], align 8
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <2 x i64> [[TMP11]], i32 1
+; CHECK-NEXT:    store i64 [[TMP15]], ptr [[TMP13]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 6, i64 6>
+; CHECK-NEXT:    [[TMP16:%.*]] = icmp eq i64 [[INDEX_NEXT]], 332
+; CHECK-NEXT:    br i1 [[TMP16]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 333, 332
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 996, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[LSHR:%.*]] = lshr i64 [[IV]], 0
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[LSHR]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 3
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP13:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv_next, %loop ]
+  %lshr = lshr i64 %iv, 0
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %lshr
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 3
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 0 ; ; iv += 3) B[iv] = A[iv>>1] + 42;
+define void @ld_lshr1_step3_start0_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_lshr1_step3_start0_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 0, i64 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = mul i64 [[INDEX]], 3
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[OFFSET_IDX]], 3
+; CHECK-NEXT:    [[TMP2:%.*]] = lshr <2 x i64> [[VEC_IND]], <i64 1, i64 1>
+; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <2 x i64> [[TMP2]], i32 0
+; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP3]]
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <2 x i64> [[TMP2]], i32 1
+; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP5]]
+; CHECK-NEXT:    [[TMP7:%.*]] = load i64, ptr [[TMP4]], align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = load i64, ptr [[TMP6]], align 8
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <2 x i64> poison, i64 [[TMP7]], i32 0
+; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <2 x i64> [[TMP9]], i64 [[TMP8]], i32 1
+; CHECK-NEXT:    [[TMP11:%.*]] = add nsw <2 x i64> [[TMP10]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <2 x i64> [[TMP11]], i32 0
+; CHECK-NEXT:    store i64 [[TMP14]], ptr [[TMP12]], align 8
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <2 x i64> [[TMP11]], i32 1
+; CHECK-NEXT:    store i64 [[TMP15]], ptr [[TMP13]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 6, i64 6>
+; CHECK-NEXT:    [[TMP16:%.*]] = icmp eq i64 [[INDEX_NEXT]], 332
+; CHECK-NEXT:    br i1 [[TMP16]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP14:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 333, 332
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 996, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[LSHR:%.*]] = lshr i64 [[IV]], 1
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[LSHR]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 3
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP15:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv_next, %loop ]
+  %lshr = lshr i64 %iv, 1
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %lshr
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 3
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 1 ; ; iv += 1) B[iv] = A[iv>>1] + 42;
+define void @ld_lshr1_step1_start1_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_lshr1_step1_start1_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 1, i64 2>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i64 1, [[INDEX]]
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = lshr <2 x i64> [[VEC_IND]], <i64 1, i64 1>
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <2 x i64> [[TMP1]], i32 0
+; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <2 x i64> [[TMP1]], i32 1
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP4]]
+; CHECK-NEXT:    [[TMP6:%.*]] = load i64, ptr [[TMP3]], align 8
+; CHECK-NEXT:    [[TMP7:%.*]] = load i64, ptr [[TMP5]], align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <2 x i64> poison, i64 [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <2 x i64> [[TMP8]], i64 [[TMP7]], i32 1
+; CHECK-NEXT:    [[TMP10:%.*]] = add nsw <2 x i64> [[TMP9]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP11:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[TMP11]], i32 0
+; CHECK-NEXT:    store <2 x i64> [[TMP10]], ptr [[TMP12]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 2, i64 2>
+; CHECK-NEXT:    [[TMP13:%.*]] = icmp eq i64 [[INDEX_NEXT]], 998
+; CHECK-NEXT:    br i1 [[TMP13]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP16:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 999, 998
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 999, [[MIDDLE_BLOCK]] ], [ 1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[LSHR:%.*]] = lshr i64 [[IV]], 1
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[LSHR]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP17:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 1, %entry ], [ %iv_next, %loop ]
+  %lshr = lshr i64 %iv, 1
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %lshr
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 1
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 1 ; ; iv += 2) B[iv] = A[iv>>1] + 42;
+define void @ld_lshr1_step2_start1_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_lshr1_step2_start1_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = mul i64 [[INDEX]], 2
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i64 1, [[TMP0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP2:%.*]] = add i64 [[OFFSET_IDX]], 2
+; CHECK-NEXT:    [[TMP3:%.*]] = lshr i64 [[TMP1]], 1
+; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP3]]
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[TMP4]], i32 0
+; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <2 x i64>, ptr [[TMP5]], align 8
+; CHECK-NEXT:    [[TMP6:%.*]] = add nsw <2 x i64> [[WIDE_LOAD]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP7:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <2 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    store i64 [[TMP9]], ptr [[TMP7]], align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <2 x i64> [[TMP6]], i32 1
+; CHECK-NEXT:    store i64 [[TMP10]], ptr [[TMP8]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], 498
+; CHECK-NEXT:    br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP18:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 499, 498
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 997, [[MIDDLE_BLOCK]] ], [ 1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[LSHR:%.*]] = lshr i64 [[IV]], 1
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[LSHR]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 2
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP19:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 1, %entry ], [ %iv_next, %loop ]
+  %lshr = lshr i64 %iv, 1
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %lshr
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 2
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 1 ; ; iv += 3) B[iv] = A[iv>>1] + 42;
+define void @ld_lshr1_step3_start1_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_lshr1_step3_start1_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 1, i64 4>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = mul i64 [[INDEX]], 3
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i64 1, [[TMP0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP2:%.*]] = add i64 [[OFFSET_IDX]], 3
+; CHECK-NEXT:    [[TMP3:%.*]] = lshr <2 x i64> [[VEC_IND]], <i64 1, i64 1>
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i32 0
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP4]]
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <2 x i64> [[TMP3]], i32 1
+; CHECK-NEXT:    [[TMP7:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP6]]
+; CHECK-NEXT:    [[TMP8:%.*]] = load i64, ptr [[TMP5]], align 8
+; CHECK-NEXT:    [[TMP9:%.*]] = load i64, ptr [[TMP7]], align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <2 x i64> poison, i64 [[TMP8]], i32 0
+; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <2 x i64> [[TMP10]], i64 [[TMP9]], i32 1
+; CHECK-NEXT:    [[TMP12:%.*]] = add nsw <2 x i64> [[TMP11]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <2 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP15]], ptr [[TMP13]], align 8
+; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <2 x i64> [[TMP12]], i32 1
+; CHECK-NEXT:    store i64 [[TMP16]], ptr [[TMP14]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 6, i64 6>
+; CHECK-NEXT:    [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], 332
+; CHECK-NEXT:    br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP20:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 333, 332
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 997, [[MIDDLE_BLOCK]] ], [ 1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[LSHR:%.*]] = lshr i64 [[IV]], 1
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[LSHR]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 3
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP21:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 1, %entry ], [ %iv_next, %loop ]
+  %lshr = lshr i64 %iv, 1
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %lshr
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 3
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}
+
+; for (iv = 1 ; ; iv += 3) B[iv] = A[iv>>2] + 42;
+define void @ld_lshr2_step3_start1_ind1(ptr noalias %A, ptr noalias %B) {
+; CHECK-LABEL: define void @ld_lshr2_step3_start1_ind1
+; CHECK-SAME: (ptr noalias [[A:%.*]], ptr noalias [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 1, i64 4>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = mul i64 [[INDEX]], 3
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i64 1, [[TMP0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP2:%.*]] = add i64 [[OFFSET_IDX]], 3
+; CHECK-NEXT:    [[TMP3:%.*]] = lshr <2 x i64> [[VEC_IND]], <i64 2, i64 2>
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i32 0
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP4]]
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <2 x i64> [[TMP3]], i32 1
+; CHECK-NEXT:    [[TMP7:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[TMP6]]
+; CHECK-NEXT:    [[TMP8:%.*]] = load i64, ptr [[TMP5]], align 8
+; CHECK-NEXT:    [[TMP9:%.*]] = load i64, ptr [[TMP7]], align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = insertelement <2 x i64> poison, i64 [[TMP8]], i32 0
+; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <2 x i64> [[TMP10]], i64 [[TMP9]], i32 1
+; CHECK-NEXT:    [[TMP12:%.*]] = add nsw <2 x i64> [[TMP11]], <i64 42, i64 42>
+; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <2 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP15]], ptr [[TMP13]], align 8
+; CHECK-NEXT:    [[TMP16:%.*]] = extractelement <2 x i64> [[TMP12]], i32 1
+; CHECK-NEXT:    store i64 [[TMP16]], ptr [[TMP14]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 6, i64 6>
+; CHECK-NEXT:    [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], 332
+; CHECK-NEXT:    br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP22:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 333, 332
+; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 997, [[MIDDLE_BLOCK]] ], [ 1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[LSHR:%.*]] = lshr i64 [[IV]], 2
+; CHECK-NEXT:    [[GEP_LD:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[LSHR]]
+; CHECK-NEXT:    [[LD:%.*]] = load i64, ptr [[GEP_LD]], align 8
+; CHECK-NEXT:    [[CALC:%.*]] = add nsw i64 [[LD]], 42
+; CHECK-NEXT:    [[GEP_ST:%.*]] = getelementptr inbounds i64, ptr [[B]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[CALC]], ptr [[GEP_ST]], align 8
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 3
+; CHECK-NEXT:    [[COND:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[COND]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP23:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+loop:
+  %iv = phi i64 [ 1, %entry ], [ %iv_next, %loop ]
+  %lshr = lshr i64 %iv, 2
+  %gep_ld = getelementptr inbounds i64, ptr %A, i64 %lshr
+  %ld = load i64, ptr %gep_ld, align 8
+  %calc = add nsw i64 %ld, 42
+  %gep_st = getelementptr inbounds i64, ptr %B, i64 %iv
+  store i64 %calc, ptr %gep_st, align 8
+  %iv_next = add nsw i64 %iv, 3
+  %cond = icmp eq i64 %iv_next, 1000
+  br i1 %cond, label %exit, label %loop
+exit:
+  ret void
+}