[llvm] 0c00dbb - [LoopVectorize] Regenerate test checks (NFC)
Nikita Popov via llvm-commits
llvm-commits at lists.llvm.org
Fri May 13 07:41:56 PDT 2022
Author: Nikita Popov
Date: 2022-05-13T16:41:48+02:00
New Revision: 0c00dbb97557e6ff177beeab3b2ea583b3eac482
URL: https://github.com/llvm/llvm-project/commit/0c00dbb97557e6ff177beeab3b2ea583b3eac482
DIFF: https://github.com/llvm/llvm-project/commit/0c00dbb97557e6ff177beeab3b2ea583b3eac482.diff
LOG: [LoopVectorize] Regenerate test checks (NFC)
Added:
Modified:
llvm/test/Transforms/LoopVectorize/induction.ll
Removed:
################################################################################
diff --git a/llvm/test/Transforms/LoopVectorize/induction.ll b/llvm/test/Transforms/LoopVectorize/induction.ll
index 4bfb94fc87fb..049cc2e38ea4 100644
--- a/llvm/test/Transforms/LoopVectorize/induction.ll
+++ b/llvm/test/Transforms/LoopVectorize/induction.ll
@@ -20,8 +20,8 @@ define void @multi_int_induction(i32* %A, i32 %N) {
; CHECK: vector.ph:
; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[TMP2]], 2
; CHECK-NEXT: [[N_VEC:%.*]] = sub i64 [[TMP2]], [[N_MOD_VF]]
-; CHECK-NEXT: [[CAST_CRD:%.*]] = trunc i64 [[N_VEC]] to i32
-; CHECK-NEXT: [[IND_END:%.*]] = add i32 190, [[CAST_CRD]]
+; CHECK-NEXT: [[CAST_VTC:%.*]] = trunc i64 [[N_VEC]] to i32
+; CHECK-NEXT: [[IND_END:%.*]] = add i32 190, [[CAST_VTC]]
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
@@ -64,8 +64,8 @@ define void @multi_int_induction(i32* %A, i32 %N) {
; IND-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; IND: vector.ph:
; IND-NEXT: [[N_VEC:%.*]] = and i64 [[TMP2]], 8589934590
-; IND-NEXT: [[CAST_CRD:%.*]] = trunc i64 [[N_VEC]] to i32
-; IND-NEXT: [[IND_END:%.*]] = add i32 [[CAST_CRD]], 190
+; IND-NEXT: [[CAST_VTC:%.*]] = trunc i64 [[N_VEC]] to i32
+; IND-NEXT: [[IND_END:%.*]] = add i32 [[CAST_VTC]], 190
; IND-NEXT: br label [[VECTOR_BODY:%.*]]
; IND: vector.body:
; IND-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
@@ -106,8 +106,8 @@ define void @multi_int_induction(i32* %A, i32 %N) {
; UNROLL-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; UNROLL: vector.ph:
; UNROLL-NEXT: [[N_VEC:%.*]] = and i64 [[TMP2]], 8589934588
-; UNROLL-NEXT: [[CAST_CRD:%.*]] = trunc i64 [[N_VEC]] to i32
-; UNROLL-NEXT: [[IND_END:%.*]] = add i32 [[CAST_CRD]], 190
+; UNROLL-NEXT: [[CAST_VTC:%.*]] = trunc i64 [[N_VEC]] to i32
+; UNROLL-NEXT: [[IND_END:%.*]] = add i32 [[CAST_VTC]], 190
; UNROLL-NEXT: br label [[VECTOR_BODY:%.*]]
; UNROLL: vector.body:
; UNROLL-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
@@ -153,8 +153,8 @@ define void @multi_int_induction(i32* %A, i32 %N) {
; UNROLL-NO-IC: vector.ph:
; UNROLL-NO-IC-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[TMP2]], 4
; UNROLL-NO-IC-NEXT: [[N_VEC:%.*]] = sub i64 [[TMP2]], [[N_MOD_VF]]
-; UNROLL-NO-IC-NEXT: [[CAST_CRD:%.*]] = trunc i64 [[N_VEC]] to i32
-; UNROLL-NO-IC-NEXT: [[IND_END:%.*]] = add i32 190, [[CAST_CRD]]
+; UNROLL-NO-IC-NEXT: [[CAST_VTC:%.*]] = trunc i64 [[N_VEC]] to i32
+; UNROLL-NO-IC-NEXT: [[IND_END:%.*]] = add i32 190, [[CAST_VTC]]
; UNROLL-NO-IC-NEXT: br label [[VECTOR_BODY:%.*]]
; UNROLL-NO-IC: vector.body:
; UNROLL-NO-IC-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
@@ -203,8 +203,8 @@ define void @multi_int_induction(i32* %A, i32 %N) {
; INTERLEAVE-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; INTERLEAVE: vector.ph:
; INTERLEAVE-NEXT: [[N_VEC:%.*]] = and i64 [[TMP2]], 8589934584
-; INTERLEAVE-NEXT: [[CAST_CRD:%.*]] = trunc i64 [[N_VEC]] to i32
-; INTERLEAVE-NEXT: [[IND_END:%.*]] = add i32 [[CAST_CRD]], 190
+; INTERLEAVE-NEXT: [[CAST_VTC:%.*]] = trunc i64 [[N_VEC]] to i32
+; INTERLEAVE-NEXT: [[IND_END:%.*]] = add i32 [[CAST_VTC]], 190
; INTERLEAVE-NEXT: br label [[VECTOR_BODY:%.*]]
; INTERLEAVE: vector.body:
; INTERLEAVE-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
@@ -2166,12 +2166,12 @@ define i32 @scalarize_induction_variable_05(i32* %a, i32 %x, i1 %c, i32 %n) {
; UNROLL: vector.ph:
; UNROLL-NEXT: [[N_VEC:%.*]] = and i32 [[SMAX]], 2147483644
; UNROLL-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i1> poison, i1 [[C:%.*]], i64 0
-; UNROLL-NEXT: [[BROADCAST_SPLATINSERT6:%.*]] = insertelement <2 x i1> poison, i1 [[C]], i64 0
+; UNROLL-NEXT: [[BROADCAST_SPLATINSERT5:%.*]] = insertelement <2 x i1> poison, i1 [[C]], i64 0
; UNROLL-NEXT: br label [[VECTOR_BODY:%.*]]
; UNROLL: vector.body:
-; UNROLL-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_UDIV_CONTINUE11:%.*]] ]
-; UNROLL-NEXT: [[VEC_PHI:%.*]] = phi <2 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP28:%.*]], [[PRED_UDIV_CONTINUE11]] ]
-; UNROLL-NEXT: [[VEC_PHI2:%.*]] = phi <2 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP29:%.*]], [[PRED_UDIV_CONTINUE11]] ]
+; UNROLL-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_UDIV_CONTINUE10:%.*]] ]
+; UNROLL-NEXT: [[VEC_PHI:%.*]] = phi <2 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP28:%.*]], [[PRED_UDIV_CONTINUE10]] ]
+; UNROLL-NEXT: [[VEC_PHI1:%.*]] = phi <2 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP29:%.*]], [[PRED_UDIV_CONTINUE10]] ]
; UNROLL-NEXT: [[TMP0:%.*]] = or i32 [[INDEX]], 2
; UNROLL-NEXT: [[TMP1:%.*]] = sext i32 [[INDEX]] to i64
; UNROLL-NEXT: [[TMP2:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 [[TMP1]]
@@ -2179,7 +2179,7 @@ define i32 @scalarize_induction_variable_05(i32* %a, i32 %x, i1 %c, i32 %n) {
; UNROLL-NEXT: [[WIDE_LOAD:%.*]] = load <2 x i32>, <2 x i32>* [[TMP3]], align 4
; UNROLL-NEXT: [[TMP4:%.*]] = getelementptr inbounds i32, i32* [[TMP2]], i64 2
; UNROLL-NEXT: [[TMP5:%.*]] = bitcast i32* [[TMP4]] to <2 x i32>*
-; UNROLL-NEXT: [[WIDE_LOAD3:%.*]] = load <2 x i32>, <2 x i32>* [[TMP5]], align 4
+; UNROLL-NEXT: [[WIDE_LOAD2:%.*]] = load <2 x i32>, <2 x i32>* [[TMP5]], align 4
; UNROLL-NEXT: br i1 [[C]], label [[PRED_UDIV_IF:%.*]], label [[PRED_UDIV_CONTINUE:%.*]]
; UNROLL: pred.udiv.if:
; UNROLL-NEXT: [[TMP6:%.*]] = extractelement <2 x i32> [[WIDE_LOAD]], i64 0
@@ -2188,40 +2188,40 @@ define i32 @scalarize_induction_variable_05(i32* %a, i32 %x, i1 %c, i32 %n) {
; UNROLL-NEXT: br label [[PRED_UDIV_CONTINUE]]
; UNROLL: pred.udiv.continue:
; UNROLL-NEXT: [[TMP9:%.*]] = phi <2 x i32> [ poison, [[VECTOR_BODY]] ], [ [[TMP8]], [[PRED_UDIV_IF]] ]
-; UNROLL-NEXT: br i1 [[C]], label [[PRED_UDIV_IF4:%.*]], label [[PRED_UDIV_CONTINUE5:%.*]]
+; UNROLL-NEXT: br i1 [[C]], label [[PRED_UDIV_IF3:%.*]], label [[PRED_UDIV_CONTINUE4:%.*]]
; UNROLL: pred.udiv.if3:
; UNROLL-NEXT: [[TMP10:%.*]] = or i32 [[INDEX]], 1
; UNROLL-NEXT: [[TMP11:%.*]] = extractelement <2 x i32> [[WIDE_LOAD]], i64 1
; UNROLL-NEXT: [[TMP12:%.*]] = udiv i32 [[TMP11]], [[TMP10]]
; UNROLL-NEXT: [[TMP13:%.*]] = insertelement <2 x i32> [[TMP9]], i32 [[TMP12]], i64 1
-; UNROLL-NEXT: br label [[PRED_UDIV_CONTINUE5]]
+; UNROLL-NEXT: br label [[PRED_UDIV_CONTINUE4]]
; UNROLL: pred.udiv.continue4:
-; UNROLL-NEXT: [[TMP14:%.*]] = phi <2 x i32> [ [[TMP9]], [[PRED_UDIV_CONTINUE]] ], [ [[TMP13]], [[PRED_UDIV_IF4]] ]
-; UNROLL-NEXT: br i1 [[C]], label [[PRED_UDIV_IF8:%.*]], label [[PRED_UDIV_CONTINUE9:%.*]]
+; UNROLL-NEXT: [[TMP14:%.*]] = phi <2 x i32> [ [[TMP9]], [[PRED_UDIV_CONTINUE]] ], [ [[TMP13]], [[PRED_UDIV_IF3]] ]
+; UNROLL-NEXT: br i1 [[C]], label [[PRED_UDIV_IF7:%.*]], label [[PRED_UDIV_CONTINUE8:%.*]]
; UNROLL: pred.udiv.if7:
-; UNROLL-NEXT: [[TMP15:%.*]] = extractelement <2 x i32> [[WIDE_LOAD3]], i64 0
+; UNROLL-NEXT: [[TMP15:%.*]] = extractelement <2 x i32> [[WIDE_LOAD2]], i64 0
; UNROLL-NEXT: [[TMP16:%.*]] = udiv i32 [[TMP15]], [[TMP0]]
; UNROLL-NEXT: [[TMP17:%.*]] = insertelement <2 x i32> poison, i32 [[TMP16]], i64 0
-; UNROLL-NEXT: br label [[PRED_UDIV_CONTINUE9]]
+; UNROLL-NEXT: br label [[PRED_UDIV_CONTINUE8]]
; UNROLL: pred.udiv.continue8:
-; UNROLL-NEXT: [[TMP18:%.*]] = phi <2 x i32> [ poison, [[PRED_UDIV_CONTINUE5]] ], [ [[TMP17]], [[PRED_UDIV_IF8]] ]
-; UNROLL-NEXT: br i1 [[C]], label [[PRED_UDIV_IF10:%.*]], label [[PRED_UDIV_CONTINUE11]]
+; UNROLL-NEXT: [[TMP18:%.*]] = phi <2 x i32> [ poison, [[PRED_UDIV_CONTINUE4]] ], [ [[TMP17]], [[PRED_UDIV_IF7]] ]
+; UNROLL-NEXT: br i1 [[C]], label [[PRED_UDIV_IF9:%.*]], label [[PRED_UDIV_CONTINUE10]]
; UNROLL: pred.udiv.if9:
; UNROLL-NEXT: [[TMP19:%.*]] = or i32 [[INDEX]], 3
-; UNROLL-NEXT: [[TMP20:%.*]] = extractelement <2 x i32> [[WIDE_LOAD3]], i64 1
+; UNROLL-NEXT: [[TMP20:%.*]] = extractelement <2 x i32> [[WIDE_LOAD2]], i64 1
; UNROLL-NEXT: [[TMP21:%.*]] = udiv i32 [[TMP20]], [[TMP19]]
; UNROLL-NEXT: [[TMP22:%.*]] = insertelement <2 x i32> [[TMP18]], i32 [[TMP21]], i64 1
-; UNROLL-NEXT: br label [[PRED_UDIV_CONTINUE11]]
+; UNROLL-NEXT: br label [[PRED_UDIV_CONTINUE10]]
; UNROLL: pred.udiv.continue10:
-; UNROLL-NEXT: [[TMP23:%.*]] = phi <2 x i32> [ [[TMP18]], [[PRED_UDIV_CONTINUE9]] ], [ [[TMP22]], [[PRED_UDIV_IF10]] ]
+; UNROLL-NEXT: [[TMP23:%.*]] = phi <2 x i32> [ [[TMP18]], [[PRED_UDIV_CONTINUE8]] ], [ [[TMP22]], [[PRED_UDIV_IF9]] ]
; UNROLL-NEXT: [[TMP24:%.*]] = xor <2 x i1> [[BROADCAST_SPLATINSERT]], <i1 true, i1 poison>
; UNROLL-NEXT: [[TMP25:%.*]] = shufflevector <2 x i1> [[TMP24]], <2 x i1> poison, <2 x i32> zeroinitializer
-; UNROLL-NEXT: [[TMP26:%.*]] = xor <2 x i1> [[BROADCAST_SPLATINSERT6]], <i1 true, i1 poison>
+; UNROLL-NEXT: [[TMP26:%.*]] = xor <2 x i1> [[BROADCAST_SPLATINSERT5]], <i1 true, i1 poison>
; UNROLL-NEXT: [[TMP27:%.*]] = shufflevector <2 x i1> [[TMP26]], <2 x i1> poison, <2 x i32> zeroinitializer
; UNROLL-NEXT: [[PREDPHI:%.*]] = select <2 x i1> [[TMP25]], <2 x i32> [[WIDE_LOAD]], <2 x i32> [[TMP14]]
-; UNROLL-NEXT: [[PREDPHI12:%.*]] = select <2 x i1> [[TMP27]], <2 x i32> [[WIDE_LOAD3]], <2 x i32> [[TMP23]]
+; UNROLL-NEXT: [[PREDPHI11:%.*]] = select <2 x i1> [[TMP27]], <2 x i32> [[WIDE_LOAD2]], <2 x i32> [[TMP23]]
; UNROLL-NEXT: [[TMP28]] = add <2 x i32> [[PREDPHI]], [[VEC_PHI]]
-; UNROLL-NEXT: [[TMP29]] = add <2 x i32> [[PREDPHI12]], [[VEC_PHI2]]
+; UNROLL-NEXT: [[TMP29]] = add <2 x i32> [[PREDPHI11]], [[VEC_PHI1]]
; UNROLL-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
; UNROLL-NEXT: [[TMP30:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; UNROLL-NEXT: br i1 [[TMP30]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP24:![0-9]+]]
@@ -2264,13 +2264,13 @@ define i32 @scalarize_induction_variable_05(i32* %a, i32 %x, i1 %c, i32 %n) {
; UNROLL-NO-IC-NEXT: [[N_VEC:%.*]] = sub i32 [[SMAX]], [[N_MOD_VF]]
; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i1> poison, i1 [[C:%.*]], i32 0
; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x i1> [[BROADCAST_SPLATINSERT]], <2 x i1> poison, <2 x i32> zeroinitializer
-; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLATINSERT6:%.*]] = insertelement <2 x i1> poison, i1 [[C]], i32 0
-; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLAT7:%.*]] = shufflevector <2 x i1> [[BROADCAST_SPLATINSERT6]], <2 x i1> poison, <2 x i32> zeroinitializer
+; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLATINSERT5:%.*]] = insertelement <2 x i1> poison, i1 [[C]], i32 0
+; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLAT6:%.*]] = shufflevector <2 x i1> [[BROADCAST_SPLATINSERT5]], <2 x i1> poison, <2 x i32> zeroinitializer
; UNROLL-NO-IC-NEXT: br label [[VECTOR_BODY:%.*]]
; UNROLL-NO-IC: vector.body:
-; UNROLL-NO-IC-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_UDIV_CONTINUE11:%.*]] ]
-; UNROLL-NO-IC-NEXT: [[VEC_PHI:%.*]] = phi <2 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP32:%.*]], [[PRED_UDIV_CONTINUE11]] ]
-; UNROLL-NO-IC-NEXT: [[VEC_PHI2:%.*]] = phi <2 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP33:%.*]], [[PRED_UDIV_CONTINUE11]] ]
+; UNROLL-NO-IC-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_UDIV_CONTINUE10:%.*]] ]
+; UNROLL-NO-IC-NEXT: [[VEC_PHI:%.*]] = phi <2 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP32:%.*]], [[PRED_UDIV_CONTINUE10]] ]
+; UNROLL-NO-IC-NEXT: [[VEC_PHI1:%.*]] = phi <2 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP33:%.*]], [[PRED_UDIV_CONTINUE10]] ]
; UNROLL-NO-IC-NEXT: [[TMP0:%.*]] = add i32 [[INDEX]], 0
; UNROLL-NO-IC-NEXT: [[TMP1:%.*]] = add i32 [[INDEX]], 2
; UNROLL-NO-IC-NEXT: [[TMP2:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[TMP0]]
@@ -2280,7 +2280,7 @@ define i32 @scalarize_induction_variable_05(i32* %a, i32 %x, i1 %c, i32 %n) {
; UNROLL-NO-IC-NEXT: [[WIDE_LOAD:%.*]] = load <2 x i32>, <2 x i32>* [[TMP5]], align 4
; UNROLL-NO-IC-NEXT: [[TMP6:%.*]] = getelementptr inbounds i32, i32* [[TMP2]], i32 2
; UNROLL-NO-IC-NEXT: [[TMP7:%.*]] = bitcast i32* [[TMP6]] to <2 x i32>*
-; UNROLL-NO-IC-NEXT: [[WIDE_LOAD3:%.*]] = load <2 x i32>, <2 x i32>* [[TMP7]], align 4
+; UNROLL-NO-IC-NEXT: [[WIDE_LOAD2:%.*]] = load <2 x i32>, <2 x i32>* [[TMP7]], align 4
; UNROLL-NO-IC-NEXT: [[TMP8:%.*]] = extractelement <2 x i1> [[BROADCAST_SPLAT]], i32 0
; UNROLL-NO-IC-NEXT: br i1 [[TMP8]], label [[PRED_UDIV_IF:%.*]], label [[PRED_UDIV_CONTINUE:%.*]]
; UNROLL-NO-IC: pred.udiv.if:
@@ -2291,40 +2291,40 @@ define i32 @scalarize_induction_variable_05(i32* %a, i32 %x, i1 %c, i32 %n) {
; UNROLL-NO-IC: pred.udiv.continue:
; UNROLL-NO-IC-NEXT: [[TMP12:%.*]] = phi <2 x i32> [ poison, [[VECTOR_BODY]] ], [ [[TMP11]], [[PRED_UDIV_IF]] ]
; UNROLL-NO-IC-NEXT: [[TMP13:%.*]] = extractelement <2 x i1> [[BROADCAST_SPLAT]], i32 1
-; UNROLL-NO-IC-NEXT: br i1 [[TMP13]], label [[PRED_UDIV_IF4:%.*]], label [[PRED_UDIV_CONTINUE5:%.*]]
+; UNROLL-NO-IC-NEXT: br i1 [[TMP13]], label [[PRED_UDIV_IF3:%.*]], label [[PRED_UDIV_CONTINUE4:%.*]]
; UNROLL-NO-IC: pred.udiv.if3:
; UNROLL-NO-IC-NEXT: [[TMP14:%.*]] = add i32 [[INDEX]], 1
; UNROLL-NO-IC-NEXT: [[TMP15:%.*]] = extractelement <2 x i32> [[WIDE_LOAD]], i32 1
; UNROLL-NO-IC-NEXT: [[TMP16:%.*]] = udiv i32 [[TMP15]], [[TMP14]]
; UNROLL-NO-IC-NEXT: [[TMP17:%.*]] = insertelement <2 x i32> [[TMP12]], i32 [[TMP16]], i32 1
-; UNROLL-NO-IC-NEXT: br label [[PRED_UDIV_CONTINUE5]]
+; UNROLL-NO-IC-NEXT: br label [[PRED_UDIV_CONTINUE4]]
; UNROLL-NO-IC: pred.udiv.continue4:
-; UNROLL-NO-IC-NEXT: [[TMP18:%.*]] = phi <2 x i32> [ [[TMP12]], [[PRED_UDIV_CONTINUE]] ], [ [[TMP17]], [[PRED_UDIV_IF4]] ]
-; UNROLL-NO-IC-NEXT: [[TMP19:%.*]] = extractelement <2 x i1> [[BROADCAST_SPLAT7]], i32 0
-; UNROLL-NO-IC-NEXT: br i1 [[TMP19]], label [[PRED_UDIV_IF8:%.*]], label [[PRED_UDIV_CONTINUE9:%.*]]
+; UNROLL-NO-IC-NEXT: [[TMP18:%.*]] = phi <2 x i32> [ [[TMP12]], [[PRED_UDIV_CONTINUE]] ], [ [[TMP17]], [[PRED_UDIV_IF3]] ]
+; UNROLL-NO-IC-NEXT: [[TMP19:%.*]] = extractelement <2 x i1> [[BROADCAST_SPLAT6]], i32 0
+; UNROLL-NO-IC-NEXT: br i1 [[TMP19]], label [[PRED_UDIV_IF7:%.*]], label [[PRED_UDIV_CONTINUE8:%.*]]
; UNROLL-NO-IC: pred.udiv.if7:
-; UNROLL-NO-IC-NEXT: [[TMP20:%.*]] = extractelement <2 x i32> [[WIDE_LOAD3]], i32 0
+; UNROLL-NO-IC-NEXT: [[TMP20:%.*]] = extractelement <2 x i32> [[WIDE_LOAD2]], i32 0
; UNROLL-NO-IC-NEXT: [[TMP21:%.*]] = udiv i32 [[TMP20]], [[TMP1]]
; UNROLL-NO-IC-NEXT: [[TMP22:%.*]] = insertelement <2 x i32> poison, i32 [[TMP21]], i32 0
-; UNROLL-NO-IC-NEXT: br label [[PRED_UDIV_CONTINUE9]]
+; UNROLL-NO-IC-NEXT: br label [[PRED_UDIV_CONTINUE8]]
; UNROLL-NO-IC: pred.udiv.continue8:
-; UNROLL-NO-IC-NEXT: [[TMP23:%.*]] = phi <2 x i32> [ poison, [[PRED_UDIV_CONTINUE5]] ], [ [[TMP22]], [[PRED_UDIV_IF8]] ]
-; UNROLL-NO-IC-NEXT: [[TMP24:%.*]] = extractelement <2 x i1> [[BROADCAST_SPLAT7]], i32 1
-; UNROLL-NO-IC-NEXT: br i1 [[TMP24]], label [[PRED_UDIV_IF10:%.*]], label [[PRED_UDIV_CONTINUE11]]
+; UNROLL-NO-IC-NEXT: [[TMP23:%.*]] = phi <2 x i32> [ poison, [[PRED_UDIV_CONTINUE4]] ], [ [[TMP22]], [[PRED_UDIV_IF7]] ]
+; UNROLL-NO-IC-NEXT: [[TMP24:%.*]] = extractelement <2 x i1> [[BROADCAST_SPLAT6]], i32 1
+; UNROLL-NO-IC-NEXT: br i1 [[TMP24]], label [[PRED_UDIV_IF9:%.*]], label [[PRED_UDIV_CONTINUE10]]
; UNROLL-NO-IC: pred.udiv.if9:
; UNROLL-NO-IC-NEXT: [[TMP25:%.*]] = add i32 [[INDEX]], 3
-; UNROLL-NO-IC-NEXT: [[TMP26:%.*]] = extractelement <2 x i32> [[WIDE_LOAD3]], i32 1
+; UNROLL-NO-IC-NEXT: [[TMP26:%.*]] = extractelement <2 x i32> [[WIDE_LOAD2]], i32 1
; UNROLL-NO-IC-NEXT: [[TMP27:%.*]] = udiv i32 [[TMP26]], [[TMP25]]
; UNROLL-NO-IC-NEXT: [[TMP28:%.*]] = insertelement <2 x i32> [[TMP23]], i32 [[TMP27]], i32 1
-; UNROLL-NO-IC-NEXT: br label [[PRED_UDIV_CONTINUE11]]
+; UNROLL-NO-IC-NEXT: br label [[PRED_UDIV_CONTINUE10]]
; UNROLL-NO-IC: pred.udiv.continue10:
-; UNROLL-NO-IC-NEXT: [[TMP29:%.*]] = phi <2 x i32> [ [[TMP23]], [[PRED_UDIV_CONTINUE9]] ], [ [[TMP28]], [[PRED_UDIV_IF10]] ]
+; UNROLL-NO-IC-NEXT: [[TMP29:%.*]] = phi <2 x i32> [ [[TMP23]], [[PRED_UDIV_CONTINUE8]] ], [ [[TMP28]], [[PRED_UDIV_IF9]] ]
; UNROLL-NO-IC-NEXT: [[TMP30:%.*]] = xor <2 x i1> [[BROADCAST_SPLAT]], <i1 true, i1 true>
-; UNROLL-NO-IC-NEXT: [[TMP31:%.*]] = xor <2 x i1> [[BROADCAST_SPLAT7]], <i1 true, i1 true>
+; UNROLL-NO-IC-NEXT: [[TMP31:%.*]] = xor <2 x i1> [[BROADCAST_SPLAT6]], <i1 true, i1 true>
; UNROLL-NO-IC-NEXT: [[PREDPHI:%.*]] = select <2 x i1> [[TMP30]], <2 x i32> [[WIDE_LOAD]], <2 x i32> [[TMP18]]
-; UNROLL-NO-IC-NEXT: [[PREDPHI12:%.*]] = select <2 x i1> [[TMP31]], <2 x i32> [[WIDE_LOAD3]], <2 x i32> [[TMP29]]
+; UNROLL-NO-IC-NEXT: [[PREDPHI11:%.*]] = select <2 x i1> [[TMP31]], <2 x i32> [[WIDE_LOAD2]], <2 x i32> [[TMP29]]
; UNROLL-NO-IC-NEXT: [[TMP32]] = add <2 x i32> [[PREDPHI]], [[VEC_PHI]]
-; UNROLL-NO-IC-NEXT: [[TMP33]] = add <2 x i32> [[PREDPHI12]], [[VEC_PHI2]]
+; UNROLL-NO-IC-NEXT: [[TMP33]] = add <2 x i32> [[PREDPHI11]], [[VEC_PHI1]]
; UNROLL-NO-IC-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
; UNROLL-NO-IC-NEXT: [[TMP34:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; UNROLL-NO-IC-NEXT: br i1 [[TMP34]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP24:![0-9]+]]
@@ -2364,12 +2364,12 @@ define i32 @scalarize_induction_variable_05(i32* %a, i32 %x, i1 %c, i32 %n) {
; INTERLEAVE: vector.ph:
; INTERLEAVE-NEXT: [[N_VEC:%.*]] = and i32 [[SMAX]], 2147483640
; INTERLEAVE-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i1> poison, i1 [[C:%.*]], i64 0
-; INTERLEAVE-NEXT: [[BROADCAST_SPLATINSERT10:%.*]] = insertelement <4 x i1> poison, i1 [[C]], i64 0
+; INTERLEAVE-NEXT: [[BROADCAST_SPLATINSERT9:%.*]] = insertelement <4 x i1> poison, i1 [[C]], i64 0
; INTERLEAVE-NEXT: br label [[VECTOR_BODY:%.*]]
; INTERLEAVE: vector.body:
-; INTERLEAVE-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_UDIV_CONTINUE19:%.*]] ]
-; INTERLEAVE-NEXT: [[VEC_PHI:%.*]] = phi <4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP48:%.*]], [[PRED_UDIV_CONTINUE19]] ]
-; INTERLEAVE-NEXT: [[VEC_PHI2:%.*]] = phi <4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP49:%.*]], [[PRED_UDIV_CONTINUE19]] ]
+; INTERLEAVE-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_UDIV_CONTINUE18:%.*]] ]
+; INTERLEAVE-NEXT: [[VEC_PHI:%.*]] = phi <4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP48:%.*]], [[PRED_UDIV_CONTINUE18]] ]
+; INTERLEAVE-NEXT: [[VEC_PHI1:%.*]] = phi <4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP49:%.*]], [[PRED_UDIV_CONTINUE18]] ]
; INTERLEAVE-NEXT: [[TMP0:%.*]] = or i32 [[INDEX]], 4
; INTERLEAVE-NEXT: [[TMP1:%.*]] = sext i32 [[INDEX]] to i64
; INTERLEAVE-NEXT: [[TMP2:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 [[TMP1]]
@@ -2377,7 +2377,7 @@ define i32 @scalarize_induction_variable_05(i32* %a, i32 %x, i1 %c, i32 %n) {
; INTERLEAVE-NEXT: [[WIDE_LOAD:%.*]] = load <4 x i32>, <4 x i32>* [[TMP3]], align 4
; INTERLEAVE-NEXT: [[TMP4:%.*]] = getelementptr inbounds i32, i32* [[TMP2]], i64 4
; INTERLEAVE-NEXT: [[TMP5:%.*]] = bitcast i32* [[TMP4]] to <4 x i32>*
-; INTERLEAVE-NEXT: [[WIDE_LOAD3:%.*]] = load <4 x i32>, <4 x i32>* [[TMP5]], align 4
+; INTERLEAVE-NEXT: [[WIDE_LOAD2:%.*]] = load <4 x i32>, <4 x i32>* [[TMP5]], align 4
; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF:%.*]], label [[PRED_UDIV_CONTINUE:%.*]]
; INTERLEAVE: pred.udiv.if:
; INTERLEAVE-NEXT: [[TMP6:%.*]] = extractelement <4 x i32> [[WIDE_LOAD]], i64 0
@@ -2386,76 +2386,76 @@ define i32 @scalarize_induction_variable_05(i32* %a, i32 %x, i1 %c, i32 %n) {
; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE]]
; INTERLEAVE: pred.udiv.continue:
; INTERLEAVE-NEXT: [[TMP9:%.*]] = phi <4 x i32> [ poison, [[VECTOR_BODY]] ], [ [[TMP8]], [[PRED_UDIV_IF]] ]
-; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF4:%.*]], label [[PRED_UDIV_CONTINUE5:%.*]]
+; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF3:%.*]], label [[PRED_UDIV_CONTINUE4:%.*]]
; INTERLEAVE: pred.udiv.if3:
; INTERLEAVE-NEXT: [[TMP10:%.*]] = or i32 [[INDEX]], 1
; INTERLEAVE-NEXT: [[TMP11:%.*]] = extractelement <4 x i32> [[WIDE_LOAD]], i64 1
; INTERLEAVE-NEXT: [[TMP12:%.*]] = udiv i32 [[TMP11]], [[TMP10]]
; INTERLEAVE-NEXT: [[TMP13:%.*]] = insertelement <4 x i32> [[TMP9]], i32 [[TMP12]], i64 1
-; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE5]]
+; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE4]]
; INTERLEAVE: pred.udiv.continue4:
-; INTERLEAVE-NEXT: [[TMP14:%.*]] = phi <4 x i32> [ [[TMP9]], [[PRED_UDIV_CONTINUE]] ], [ [[TMP13]], [[PRED_UDIV_IF4]] ]
-; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF6:%.*]], label [[PRED_UDIV_CONTINUE7:%.*]]
+; INTERLEAVE-NEXT: [[TMP14:%.*]] = phi <4 x i32> [ [[TMP9]], [[PRED_UDIV_CONTINUE]] ], [ [[TMP13]], [[PRED_UDIV_IF3]] ]
+; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF5:%.*]], label [[PRED_UDIV_CONTINUE6:%.*]]
; INTERLEAVE: pred.udiv.if5:
; INTERLEAVE-NEXT: [[TMP15:%.*]] = or i32 [[INDEX]], 2
; INTERLEAVE-NEXT: [[TMP16:%.*]] = extractelement <4 x i32> [[WIDE_LOAD]], i64 2
; INTERLEAVE-NEXT: [[TMP17:%.*]] = udiv i32 [[TMP16]], [[TMP15]]
; INTERLEAVE-NEXT: [[TMP18:%.*]] = insertelement <4 x i32> [[TMP14]], i32 [[TMP17]], i64 2
-; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE7]]
+; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE6]]
; INTERLEAVE: pred.udiv.continue6:
-; INTERLEAVE-NEXT: [[TMP19:%.*]] = phi <4 x i32> [ [[TMP14]], [[PRED_UDIV_CONTINUE5]] ], [ [[TMP18]], [[PRED_UDIV_IF6]] ]
-; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF8:%.*]], label [[PRED_UDIV_CONTINUE9:%.*]]
+; INTERLEAVE-NEXT: [[TMP19:%.*]] = phi <4 x i32> [ [[TMP14]], [[PRED_UDIV_CONTINUE4]] ], [ [[TMP18]], [[PRED_UDIV_IF5]] ]
+; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF7:%.*]], label [[PRED_UDIV_CONTINUE8:%.*]]
; INTERLEAVE: pred.udiv.if7:
; INTERLEAVE-NEXT: [[TMP20:%.*]] = or i32 [[INDEX]], 3
; INTERLEAVE-NEXT: [[TMP21:%.*]] = extractelement <4 x i32> [[WIDE_LOAD]], i64 3
; INTERLEAVE-NEXT: [[TMP22:%.*]] = udiv i32 [[TMP21]], [[TMP20]]
; INTERLEAVE-NEXT: [[TMP23:%.*]] = insertelement <4 x i32> [[TMP19]], i32 [[TMP22]], i64 3
-; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE9]]
+; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE8]]
; INTERLEAVE: pred.udiv.continue8:
-; INTERLEAVE-NEXT: [[TMP24:%.*]] = phi <4 x i32> [ [[TMP19]], [[PRED_UDIV_CONTINUE7]] ], [ [[TMP23]], [[PRED_UDIV_IF8]] ]
-; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF12:%.*]], label [[PRED_UDIV_CONTINUE13:%.*]]
+; INTERLEAVE-NEXT: [[TMP24:%.*]] = phi <4 x i32> [ [[TMP19]], [[PRED_UDIV_CONTINUE6]] ], [ [[TMP23]], [[PRED_UDIV_IF7]] ]
+; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF11:%.*]], label [[PRED_UDIV_CONTINUE12:%.*]]
; INTERLEAVE: pred.udiv.if11:
-; INTERLEAVE-NEXT: [[TMP25:%.*]] = extractelement <4 x i32> [[WIDE_LOAD3]], i64 0
+; INTERLEAVE-NEXT: [[TMP25:%.*]] = extractelement <4 x i32> [[WIDE_LOAD2]], i64 0
; INTERLEAVE-NEXT: [[TMP26:%.*]] = udiv i32 [[TMP25]], [[TMP0]]
; INTERLEAVE-NEXT: [[TMP27:%.*]] = insertelement <4 x i32> poison, i32 [[TMP26]], i64 0
-; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE13]]
+; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE12]]
; INTERLEAVE: pred.udiv.continue12:
-; INTERLEAVE-NEXT: [[TMP28:%.*]] = phi <4 x i32> [ poison, [[PRED_UDIV_CONTINUE9]] ], [ [[TMP27]], [[PRED_UDIV_IF12]] ]
-; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF14:%.*]], label [[PRED_UDIV_CONTINUE15:%.*]]
+; INTERLEAVE-NEXT: [[TMP28:%.*]] = phi <4 x i32> [ poison, [[PRED_UDIV_CONTINUE8]] ], [ [[TMP27]], [[PRED_UDIV_IF11]] ]
+; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF13:%.*]], label [[PRED_UDIV_CONTINUE14:%.*]]
; INTERLEAVE: pred.udiv.if13:
; INTERLEAVE-NEXT: [[TMP29:%.*]] = or i32 [[INDEX]], 5
-; INTERLEAVE-NEXT: [[TMP30:%.*]] = extractelement <4 x i32> [[WIDE_LOAD3]], i64 1
+; INTERLEAVE-NEXT: [[TMP30:%.*]] = extractelement <4 x i32> [[WIDE_LOAD2]], i64 1
; INTERLEAVE-NEXT: [[TMP31:%.*]] = udiv i32 [[TMP30]], [[TMP29]]
; INTERLEAVE-NEXT: [[TMP32:%.*]] = insertelement <4 x i32> [[TMP28]], i32 [[TMP31]], i64 1
-; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE15]]
+; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE14]]
; INTERLEAVE: pred.udiv.continue14:
-; INTERLEAVE-NEXT: [[TMP33:%.*]] = phi <4 x i32> [ [[TMP28]], [[PRED_UDIV_CONTINUE13]] ], [ [[TMP32]], [[PRED_UDIV_IF14]] ]
-; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF16:%.*]], label [[PRED_UDIV_CONTINUE17:%.*]]
+; INTERLEAVE-NEXT: [[TMP33:%.*]] = phi <4 x i32> [ [[TMP28]], [[PRED_UDIV_CONTINUE12]] ], [ [[TMP32]], [[PRED_UDIV_IF13]] ]
+; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF15:%.*]], label [[PRED_UDIV_CONTINUE16:%.*]]
; INTERLEAVE: pred.udiv.if15:
; INTERLEAVE-NEXT: [[TMP34:%.*]] = or i32 [[INDEX]], 6
-; INTERLEAVE-NEXT: [[TMP35:%.*]] = extractelement <4 x i32> [[WIDE_LOAD3]], i64 2
+; INTERLEAVE-NEXT: [[TMP35:%.*]] = extractelement <4 x i32> [[WIDE_LOAD2]], i64 2
; INTERLEAVE-NEXT: [[TMP36:%.*]] = udiv i32 [[TMP35]], [[TMP34]]
; INTERLEAVE-NEXT: [[TMP37:%.*]] = insertelement <4 x i32> [[TMP33]], i32 [[TMP36]], i64 2
-; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE17]]
+; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE16]]
; INTERLEAVE: pred.udiv.continue16:
-; INTERLEAVE-NEXT: [[TMP38:%.*]] = phi <4 x i32> [ [[TMP33]], [[PRED_UDIV_CONTINUE15]] ], [ [[TMP37]], [[PRED_UDIV_IF16]] ]
-; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF18:%.*]], label [[PRED_UDIV_CONTINUE19]]
+; INTERLEAVE-NEXT: [[TMP38:%.*]] = phi <4 x i32> [ [[TMP33]], [[PRED_UDIV_CONTINUE14]] ], [ [[TMP37]], [[PRED_UDIV_IF15]] ]
+; INTERLEAVE-NEXT: br i1 [[C]], label [[PRED_UDIV_IF17:%.*]], label [[PRED_UDIV_CONTINUE18]]
; INTERLEAVE: pred.udiv.if17:
; INTERLEAVE-NEXT: [[TMP39:%.*]] = or i32 [[INDEX]], 7
-; INTERLEAVE-NEXT: [[TMP40:%.*]] = extractelement <4 x i32> [[WIDE_LOAD3]], i64 3
+; INTERLEAVE-NEXT: [[TMP40:%.*]] = extractelement <4 x i32> [[WIDE_LOAD2]], i64 3
; INTERLEAVE-NEXT: [[TMP41:%.*]] = udiv i32 [[TMP40]], [[TMP39]]
; INTERLEAVE-NEXT: [[TMP42:%.*]] = insertelement <4 x i32> [[TMP38]], i32 [[TMP41]], i64 3
-; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE19]]
+; INTERLEAVE-NEXT: br label [[PRED_UDIV_CONTINUE18]]
; INTERLEAVE: pred.udiv.continue18:
-; INTERLEAVE-NEXT: [[TMP43:%.*]] = phi <4 x i32> [ [[TMP38]], [[PRED_UDIV_CONTINUE17]] ], [ [[TMP42]], [[PRED_UDIV_IF18]] ]
+; INTERLEAVE-NEXT: [[TMP43:%.*]] = phi <4 x i32> [ [[TMP38]], [[PRED_UDIV_CONTINUE16]] ], [ [[TMP42]], [[PRED_UDIV_IF17]] ]
; INTERLEAVE-NEXT: [[TMP44:%.*]] = xor <4 x i1> [[BROADCAST_SPLATINSERT]], <i1 true, i1 poison, i1 poison, i1 poison>
; INTERLEAVE-NEXT: [[TMP45:%.*]] = shufflevector <4 x i1> [[TMP44]], <4 x i1> poison, <4 x i32> zeroinitializer
-; INTERLEAVE-NEXT: [[TMP46:%.*]] = xor <4 x i1> [[BROADCAST_SPLATINSERT10]], <i1 true, i1 poison, i1 poison, i1 poison>
+; INTERLEAVE-NEXT: [[TMP46:%.*]] = xor <4 x i1> [[BROADCAST_SPLATINSERT9]], <i1 true, i1 poison, i1 poison, i1 poison>
; INTERLEAVE-NEXT: [[TMP47:%.*]] = shufflevector <4 x i1> [[TMP46]], <4 x i1> poison, <4 x i32> zeroinitializer
; INTERLEAVE-NEXT: [[PREDPHI:%.*]] = select <4 x i1> [[TMP45]], <4 x i32> [[WIDE_LOAD]], <4 x i32> [[TMP24]]
-; INTERLEAVE-NEXT: [[PREDPHI20:%.*]] = select <4 x i1> [[TMP47]], <4 x i32> [[WIDE_LOAD3]], <4 x i32> [[TMP43]]
+; INTERLEAVE-NEXT: [[PREDPHI19:%.*]] = select <4 x i1> [[TMP47]], <4 x i32> [[WIDE_LOAD2]], <4 x i32> [[TMP43]]
; INTERLEAVE-NEXT: [[TMP48]] = add <4 x i32> [[PREDPHI]], [[VEC_PHI]]
-; INTERLEAVE-NEXT: [[TMP49]] = add <4 x i32> [[PREDPHI20]], [[VEC_PHI2]]
+; INTERLEAVE-NEXT: [[TMP49]] = add <4 x i32> [[PREDPHI19]], [[VEC_PHI1]]
; INTERLEAVE-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 8
; INTERLEAVE-NEXT: [[TMP50:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; INTERLEAVE-NEXT: br i1 [[TMP50]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP24:![0-9]+]]
@@ -2539,10 +2539,10 @@ define void @iv_vector_and_scalar_users(%pair.i16* %p, i32 %a, i32 %n) {
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT: [[VEC_IND1:%.*]] = phi <2 x i32> [ <i32 0, i32 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT2:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ <i32 0, i32 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP3:%.*]] = add i64 [[INDEX]], 0
; CHECK-NEXT: [[TMP4:%.*]] = add i64 [[INDEX]], 1
-; CHECK-NEXT: [[TMP5:%.*]] = add <2 x i32> [[BROADCAST_SPLAT]], [[VEC_IND1]]
+; CHECK-NEXT: [[TMP5:%.*]] = add <2 x i32> [[BROADCAST_SPLAT]], [[VEC_IND]]
; CHECK-NEXT: [[TMP6:%.*]] = trunc <2 x i32> [[TMP5]] to <2 x i16>
; CHECK-NEXT: [[TMP7:%.*]] = getelementptr inbounds [[PAIR_I16:%.*]], %pair.i16* [[P:%.*]], i64 [[TMP3]], i32 1
; CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds [[PAIR_I16]], %pair.i16* [[P]], i64 [[TMP4]], i32 1
@@ -2551,7 +2551,7 @@ define void @iv_vector_and_scalar_users(%pair.i16* %p, i32 %a, i32 %n) {
; CHECK-NEXT: [[TMP10:%.*]] = extractelement <2 x i16> [[TMP6]], i32 1
; CHECK-NEXT: store i16 [[TMP10]], i16* [[TMP8]], align 2
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
-; CHECK-NEXT: [[VEC_IND_NEXT2]] = add <2 x i32> [[VEC_IND1]], <i32 2, i32 2>
+; CHECK-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
; CHECK-NEXT: [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT: br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP26:![0-9]+]]
; CHECK: middle.block:
@@ -2588,9 +2588,9 @@ define void @iv_vector_and_scalar_users(%pair.i16* %p, i32 %a, i32 %n) {
; IND-NEXT: br label [[VECTOR_BODY:%.*]]
; IND: vector.body:
; IND-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; IND-NEXT: [[VEC_IND1:%.*]] = phi <2 x i32> [ <i32 0, i32 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT2:%.*]], [[VECTOR_BODY]] ]
+; IND-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ <i32 0, i32 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; IND-NEXT: [[TMP3:%.*]] = or i64 [[INDEX]], 1
-; IND-NEXT: [[TMP4:%.*]] = add <2 x i32> [[BROADCAST_SPLAT]], [[VEC_IND1]]
+; IND-NEXT: [[TMP4:%.*]] = add <2 x i32> [[BROADCAST_SPLAT]], [[VEC_IND]]
; IND-NEXT: [[TMP5:%.*]] = trunc <2 x i32> [[TMP4]] to <2 x i16>
; IND-NEXT: [[TMP6:%.*]] = getelementptr inbounds [[PAIR_I16:%.*]], %pair.i16* [[P:%.*]], i64 [[INDEX]], i32 1
; IND-NEXT: [[TMP7:%.*]] = getelementptr inbounds [[PAIR_I16]], %pair.i16* [[P]], i64 [[TMP3]], i32 1
@@ -2599,7 +2599,7 @@ define void @iv_vector_and_scalar_users(%pair.i16* %p, i32 %a, i32 %n) {
; IND-NEXT: [[TMP9:%.*]] = extractelement <2 x i16> [[TMP5]], i64 1
; IND-NEXT: store i16 [[TMP9]], i16* [[TMP7]], align 2
; IND-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
-; IND-NEXT: [[VEC_IND_NEXT2]] = add <2 x i32> [[VEC_IND1]], <i32 2, i32 2>
+; IND-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
; IND-NEXT: [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; IND-NEXT: br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP26:![0-9]+]]
; IND: middle.block:
@@ -2633,18 +2633,18 @@ define void @iv_vector_and_scalar_users(%pair.i16* %p, i32 %a, i32 %n) {
; UNROLL-NEXT: [[N_VEC:%.*]] = and i64 [[TMP2]], 8589934588
; UNROLL-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[A:%.*]], i64 0
; UNROLL-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x i32> [[BROADCAST_SPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
-; UNROLL-NEXT: [[BROADCAST_SPLATINSERT6:%.*]] = insertelement <2 x i32> poison, i32 [[A]], i64 0
+; UNROLL-NEXT: [[BROADCAST_SPLATINSERT2:%.*]] = insertelement <2 x i32> poison, i32 [[A]], i64 0
; UNROLL-NEXT: br label [[VECTOR_BODY:%.*]]
; UNROLL: vector.body:
; UNROLL-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; UNROLL-NEXT: [[VEC_IND2:%.*]] = phi <2 x i32> [ <i32 0, i32 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT5:%.*]], [[VECTOR_BODY]] ]
+; UNROLL-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ <i32 0, i32 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; UNROLL-NEXT: [[TMP3:%.*]] = or i64 [[INDEX]], 1
; UNROLL-NEXT: [[TMP4:%.*]] = or i64 [[INDEX]], 2
; UNROLL-NEXT: [[TMP5:%.*]] = or i64 [[INDEX]], 3
-; UNROLL-NEXT: [[TMP6:%.*]] = add <2 x i32> [[BROADCAST_SPLAT]], [[VEC_IND2]]
-; UNROLL-NEXT: [[TMP7:%.*]] = add <2 x i32> [[BROADCAST_SPLATINSERT6]], <i32 2, i32 poison>
+; UNROLL-NEXT: [[TMP6:%.*]] = add <2 x i32> [[BROADCAST_SPLAT]], [[VEC_IND]]
+; UNROLL-NEXT: [[TMP7:%.*]] = add <2 x i32> [[BROADCAST_SPLATINSERT2]], <i32 2, i32 poison>
; UNROLL-NEXT: [[TMP8:%.*]] = shufflevector <2 x i32> [[TMP7]], <2 x i32> poison, <2 x i32> zeroinitializer
-; UNROLL-NEXT: [[TMP9:%.*]] = add <2 x i32> [[TMP8]], [[VEC_IND2]]
+; UNROLL-NEXT: [[TMP9:%.*]] = add <2 x i32> [[TMP8]], [[VEC_IND]]
; UNROLL-NEXT: [[TMP10:%.*]] = trunc <2 x i32> [[TMP6]] to <2 x i16>
; UNROLL-NEXT: [[TMP11:%.*]] = trunc <2 x i32> [[TMP9]] to <2 x i16>
; UNROLL-NEXT: [[TMP12:%.*]] = getelementptr inbounds [[PAIR_I16:%.*]], %pair.i16* [[P:%.*]], i64 [[INDEX]], i32 1
@@ -2660,7 +2660,7 @@ define void @iv_vector_and_scalar_users(%pair.i16* %p, i32 %a, i32 %n) {
; UNROLL-NEXT: [[TMP19:%.*]] = extractelement <2 x i16> [[TMP11]], i64 1
; UNROLL-NEXT: store i16 [[TMP19]], i16* [[TMP15]], align 2
; UNROLL-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
-; UNROLL-NEXT: [[VEC_IND_NEXT5]] = add <2 x i32> [[VEC_IND2]], <i32 4, i32 4>
+; UNROLL-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[VEC_IND]], <i32 4, i32 4>
; UNROLL-NEXT: [[TMP20:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; UNROLL-NEXT: br i1 [[TMP20]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP26:![0-9]+]]
; UNROLL: middle.block:
@@ -2695,19 +2695,19 @@ define void @iv_vector_and_scalar_users(%pair.i16* %p, i32 %a, i32 %n) {
; UNROLL-NO-IC-NEXT: [[N_VEC:%.*]] = sub i64 [[TMP2]], [[N_MOD_VF]]
; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[A:%.*]], i32 0
; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x i32> [[BROADCAST_SPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
-; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLATINSERT6:%.*]] = insertelement <2 x i32> poison, i32 [[A]], i32 0
-; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLAT7:%.*]] = shufflevector <2 x i32> [[BROADCAST_SPLATINSERT6]], <2 x i32> poison, <2 x i32> zeroinitializer
+; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLATINSERT2:%.*]] = insertelement <2 x i32> poison, i32 [[A]], i32 0
+; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLAT3:%.*]] = shufflevector <2 x i32> [[BROADCAST_SPLATINSERT2]], <2 x i32> poison, <2 x i32> zeroinitializer
; UNROLL-NO-IC-NEXT: br label [[VECTOR_BODY:%.*]]
; UNROLL-NO-IC: vector.body:
; UNROLL-NO-IC-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; UNROLL-NO-IC-NEXT: [[VEC_IND2:%.*]] = phi <2 x i32> [ <i32 0, i32 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT5:%.*]], [[VECTOR_BODY]] ]
-; UNROLL-NO-IC-NEXT: [[STEP_ADD3:%.*]] = add <2 x i32> [[VEC_IND2]], <i32 2, i32 2>
+; UNROLL-NO-IC-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ <i32 0, i32 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; UNROLL-NO-IC-NEXT: [[STEP_ADD:%.*]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
; UNROLL-NO-IC-NEXT: [[TMP3:%.*]] = add i64 [[INDEX]], 0
; UNROLL-NO-IC-NEXT: [[TMP4:%.*]] = add i64 [[INDEX]], 1
; UNROLL-NO-IC-NEXT: [[TMP5:%.*]] = add i64 [[INDEX]], 2
; UNROLL-NO-IC-NEXT: [[TMP6:%.*]] = add i64 [[INDEX]], 3
-; UNROLL-NO-IC-NEXT: [[TMP7:%.*]] = add <2 x i32> [[BROADCAST_SPLAT]], [[VEC_IND2]]
-; UNROLL-NO-IC-NEXT: [[TMP8:%.*]] = add <2 x i32> [[BROADCAST_SPLAT7]], [[STEP_ADD3]]
+; UNROLL-NO-IC-NEXT: [[TMP7:%.*]] = add <2 x i32> [[BROADCAST_SPLAT]], [[VEC_IND]]
+; UNROLL-NO-IC-NEXT: [[TMP8:%.*]] = add <2 x i32> [[BROADCAST_SPLAT3]], [[STEP_ADD]]
; UNROLL-NO-IC-NEXT: [[TMP9:%.*]] = trunc <2 x i32> [[TMP7]] to <2 x i16>
; UNROLL-NO-IC-NEXT: [[TMP10:%.*]] = trunc <2 x i32> [[TMP8]] to <2 x i16>
; UNROLL-NO-IC-NEXT: [[TMP11:%.*]] = getelementptr inbounds [[PAIR_I16:%.*]], %pair.i16* [[P:%.*]], i64 [[TMP3]], i32 1
@@ -2723,7 +2723,7 @@ define void @iv_vector_and_scalar_users(%pair.i16* %p, i32 %a, i32 %n) {
; UNROLL-NO-IC-NEXT: [[TMP18:%.*]] = extractelement <2 x i16> [[TMP10]], i32 1
; UNROLL-NO-IC-NEXT: store i16 [[TMP18]], i16* [[TMP14]], align 2
; UNROLL-NO-IC-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
-; UNROLL-NO-IC-NEXT: [[VEC_IND_NEXT5]] = add <2 x i32> [[STEP_ADD3]], <i32 2, i32 2>
+; UNROLL-NO-IC-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[STEP_ADD]], <i32 2, i32 2>
; UNROLL-NO-IC-NEXT: [[TMP19:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; UNROLL-NO-IC-NEXT: br i1 [[TMP19]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP26:![0-9]+]]
; UNROLL-NO-IC: middle.block:
@@ -2757,11 +2757,11 @@ define void @iv_vector_and_scalar_users(%pair.i16* %p, i32 %a, i32 %n) {
; INTERLEAVE-NEXT: [[N_VEC:%.*]] = and i64 [[TMP2]], 8589934584
; INTERLEAVE-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[A:%.*]], i64 0
; INTERLEAVE-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
-; INTERLEAVE-NEXT: [[BROADCAST_SPLATINSERT6:%.*]] = insertelement <4 x i32> poison, i32 [[A]], i64 0
+; INTERLEAVE-NEXT: [[BROADCAST_SPLATINSERT2:%.*]] = insertelement <4 x i32> poison, i32 [[A]], i64 0
; INTERLEAVE-NEXT: br label [[VECTOR_BODY:%.*]]
; INTERLEAVE: vector.body:
; INTERLEAVE-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; INTERLEAVE-NEXT: [[VEC_IND2:%.*]] = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT5:%.*]], [[VECTOR_BODY]] ]
+; INTERLEAVE-NEXT: [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; INTERLEAVE-NEXT: [[TMP3:%.*]] = or i64 [[INDEX]], 1
; INTERLEAVE-NEXT: [[TMP4:%.*]] = or i64 [[INDEX]], 2
; INTERLEAVE-NEXT: [[TMP5:%.*]] = or i64 [[INDEX]], 3
@@ -2769,10 +2769,10 @@ define void @iv_vector_and_scalar_users(%pair.i16* %p, i32 %a, i32 %n) {
; INTERLEAVE-NEXT: [[TMP7:%.*]] = or i64 [[INDEX]], 5
; INTERLEAVE-NEXT: [[TMP8:%.*]] = or i64 [[INDEX]], 6
; INTERLEAVE-NEXT: [[TMP9:%.*]] = or i64 [[INDEX]], 7
-; INTERLEAVE-NEXT: [[TMP10:%.*]] = add <4 x i32> [[BROADCAST_SPLAT]], [[VEC_IND2]]
-; INTERLEAVE-NEXT: [[TMP11:%.*]] = add <4 x i32> [[BROADCAST_SPLATINSERT6]], <i32 4, i32 poison, i32 poison, i32 poison>
+; INTERLEAVE-NEXT: [[TMP10:%.*]] = add <4 x i32> [[BROADCAST_SPLAT]], [[VEC_IND]]
+; INTERLEAVE-NEXT: [[TMP11:%.*]] = add <4 x i32> [[BROADCAST_SPLATINSERT2]], <i32 4, i32 poison, i32 poison, i32 poison>
; INTERLEAVE-NEXT: [[TMP12:%.*]] = shufflevector <4 x i32> [[TMP11]], <4 x i32> poison, <4 x i32> zeroinitializer
-; INTERLEAVE-NEXT: [[TMP13:%.*]] = add <4 x i32> [[TMP12]], [[VEC_IND2]]
+; INTERLEAVE-NEXT: [[TMP13:%.*]] = add <4 x i32> [[TMP12]], [[VEC_IND]]
; INTERLEAVE-NEXT: [[TMP14:%.*]] = trunc <4 x i32> [[TMP10]] to <4 x i16>
; INTERLEAVE-NEXT: [[TMP15:%.*]] = trunc <4 x i32> [[TMP13]] to <4 x i16>
; INTERLEAVE-NEXT: [[TMP16:%.*]] = getelementptr inbounds [[PAIR_I16:%.*]], %pair.i16* [[P:%.*]], i64 [[INDEX]], i32 1
@@ -2800,7 +2800,7 @@ define void @iv_vector_and_scalar_users(%pair.i16* %p, i32 %a, i32 %n) {
; INTERLEAVE-NEXT: [[TMP31:%.*]] = extractelement <4 x i16> [[TMP15]], i64 3
; INTERLEAVE-NEXT: store i16 [[TMP31]], i16* [[TMP23]], align 2
; INTERLEAVE-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
-; INTERLEAVE-NEXT: [[VEC_IND_NEXT5]] = add <4 x i32> [[VEC_IND2]], <i32 8, i32 8, i32 8, i32 8>
+; INTERLEAVE-NEXT: [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], <i32 8, i32 8, i32 8, i32 8>
; INTERLEAVE-NEXT: [[TMP32:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; INTERLEAVE-NEXT: br i1 [[TMP32]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP26:![0-9]+]]
; INTERLEAVE: middle.block:
@@ -3290,8 +3290,8 @@ define i32 @testoverflowcheck() {
; CHECK: vector.ph:
; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i32 [[TMP3]], 2
; CHECK-NEXT: [[N_VEC:%.*]] = sub i32 [[TMP3]], [[N_MOD_VF]]
-; CHECK-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; CHECK-NEXT: [[IND_END:%.*]] = add i8 [[DOTPR_I]], [[CAST_CRD]]
+; CHECK-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; CHECK-NEXT: [[IND_END:%.*]] = add i8 [[DOTPR_I]], [[CAST_VTC]]
; CHECK-NEXT: [[TMP4:%.*]] = insertelement <2 x i32> <i32 -1, i32 -1>, i32 [[C_PROMOTED_I]], i32 0
; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[TMP0]], i32 0
; CHECK-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x i32> [[BROADCAST_SPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
@@ -3334,8 +3334,8 @@ define i32 @testoverflowcheck() {
; IND-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; IND: vector.ph:
; IND-NEXT: [[N_VEC:%.*]] = and i32 [[TMP3]], 510
-; IND-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; IND-NEXT: [[IND_END:%.*]] = add i8 [[DOTPR_I]], [[CAST_CRD]]
+; IND-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; IND-NEXT: [[IND_END:%.*]] = add i8 [[DOTPR_I]], [[CAST_VTC]]
; IND-NEXT: [[TMP4:%.*]] = insertelement <2 x i32> <i32 poison, i32 -1>, i32 [[C_PROMOTED_I]], i64 0
; IND-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[TMP0]], i64 0
; IND-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x i32> [[BROADCAST_SPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
@@ -3376,8 +3376,8 @@ define i32 @testoverflowcheck() {
; UNROLL-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; UNROLL: vector.ph:
; UNROLL-NEXT: [[N_VEC:%.*]] = and i32 [[TMP3]], 508
-; UNROLL-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; UNROLL-NEXT: [[IND_END:%.*]] = add i8 [[DOTPR_I]], [[CAST_CRD]]
+; UNROLL-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; UNROLL-NEXT: [[IND_END:%.*]] = add i8 [[DOTPR_I]], [[CAST_VTC]]
; UNROLL-NEXT: [[TMP4:%.*]] = insertelement <2 x i32> <i32 poison, i32 -1>, i32 [[C_PROMOTED_I]], i64 0
; UNROLL-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[TMP0]], i64 0
; UNROLL-NEXT: [[BROADCAST_SPLATINSERT2:%.*]] = insertelement <2 x i32> poison, i32 [[TMP0]], i64 0
@@ -3421,8 +3421,8 @@ define i32 @testoverflowcheck() {
; UNROLL-NO-IC: vector.ph:
; UNROLL-NO-IC-NEXT: [[N_MOD_VF:%.*]] = urem i32 [[TMP3]], 4
; UNROLL-NO-IC-NEXT: [[N_VEC:%.*]] = sub i32 [[TMP3]], [[N_MOD_VF]]
-; UNROLL-NO-IC-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; UNROLL-NO-IC-NEXT: [[IND_END:%.*]] = add i8 [[DOTPR_I]], [[CAST_CRD]]
+; UNROLL-NO-IC-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; UNROLL-NO-IC-NEXT: [[IND_END:%.*]] = add i8 [[DOTPR_I]], [[CAST_VTC]]
; UNROLL-NO-IC-NEXT: [[TMP4:%.*]] = insertelement <2 x i32> <i32 -1, i32 -1>, i32 [[C_PROMOTED_I]], i32 0
; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[TMP0]], i32 0
; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x i32> [[BROADCAST_SPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
@@ -3470,8 +3470,8 @@ define i32 @testoverflowcheck() {
; INTERLEAVE-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; INTERLEAVE: vector.ph:
; INTERLEAVE-NEXT: [[N_VEC:%.*]] = and i32 [[TMP3]], 504
-; INTERLEAVE-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; INTERLEAVE-NEXT: [[IND_END:%.*]] = add i8 [[DOTPR_I]], [[CAST_CRD]]
+; INTERLEAVE-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; INTERLEAVE-NEXT: [[IND_END:%.*]] = add i8 [[DOTPR_I]], [[CAST_VTC]]
; INTERLEAVE-NEXT: [[TMP4:%.*]] = insertelement <4 x i32> <i32 poison, i32 -1, i32 -1, i32 -1>, i32 [[C_PROMOTED_I]], i64 0
; INTERLEAVE-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[TMP0]], i64 0
; INTERLEAVE-NEXT: [[BROADCAST_SPLATINSERT2:%.*]] = insertelement <4 x i32> poison, i32 [[TMP0]], i64 0
@@ -3552,25 +3552,25 @@ define void @wrappingindvars1(i8 %t, i32 %len, i32 *%A) {
; CHECK: vector.ph:
; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i32 [[TMP0]], 2
; CHECK-NEXT: [[N_VEC:%.*]] = sub i32 [[TMP0]], [[N_MOD_VF]]
-; CHECK-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; CHECK-NEXT: [[IND_END:%.*]] = add i8 [[T]], [[CAST_CRD]]
+; CHECK-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; CHECK-NEXT: [[IND_END:%.*]] = add i8 [[T]], [[CAST_VTC]]
; CHECK-NEXT: [[IND_END3:%.*]] = add i32 [[EXT]], [[N_VEC]]
-; CHECK-NEXT: [[DOTSPLATINSERT4:%.*]] = insertelement <2 x i32> poison, i32 [[EXT]], i32 0
-; CHECK-NEXT: [[DOTSPLAT5:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT4]], <2 x i32> poison, <2 x i32> zeroinitializer
-; CHECK-NEXT: [[INDUCTION6:%.*]] = add <2 x i32> [[DOTSPLAT5]], <i32 0, i32 1>
+; CHECK-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[EXT]], i32 0
+; CHECK-NEXT: [[DOTSPLAT:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
+; CHECK-NEXT: [[INDUCTION:%.*]] = add <2 x i32> [[DOTSPLAT]], <i32 0, i32 1>
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT: [[VEC_IND7:%.*]] = phi <2 x i32> [ [[INDUCTION6]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT8:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP12:%.*]] = trunc i32 [[INDEX]] to i8
; CHECK-NEXT: [[OFFSET_IDX:%.*]] = add i8 [[T]], [[TMP12]]
; CHECK-NEXT: [[TMP13:%.*]] = add i8 [[OFFSET_IDX]], 0
; CHECK-NEXT: [[TMP14:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i8 [[TMP13]]
; CHECK-NEXT: [[TMP15:%.*]] = getelementptr inbounds i32, i32* [[TMP14]], i32 0
; CHECK-NEXT: [[TMP16:%.*]] = bitcast i32* [[TMP15]] to <2 x i32>*
-; CHECK-NEXT: store <2 x i32> [[VEC_IND7]], <2 x i32>* [[TMP16]], align 4
+; CHECK-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP16]], align 4
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 2
-; CHECK-NEXT: [[VEC_IND_NEXT8]] = add <2 x i32> [[VEC_IND7]], <i32 2, i32 2>
+; CHECK-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
; CHECK-NEXT: [[TMP17:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT: br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP36:![0-9]+]]
; CHECK: middle.block:
@@ -3621,24 +3621,24 @@ define void @wrappingindvars1(i8 %t, i32 %len, i32 *%A) {
; IND-NEXT: br i1 [[TMP11]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
; IND: vector.ph:
; IND-NEXT: [[N_VEC:%.*]] = and i32 [[TMP0]], -2
-; IND-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; IND-NEXT: [[IND_END:%.*]] = add i8 [[CAST_CRD]], [[T]]
+; IND-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; IND-NEXT: [[IND_END:%.*]] = add i8 [[CAST_VTC]], [[T]]
; IND-NEXT: [[IND_END3:%.*]] = add i32 [[N_VEC]], [[EXT]]
-; IND-NEXT: [[DOTSPLATINSERT4:%.*]] = insertelement <2 x i32> poison, i32 [[EXT]], i64 0
-; IND-NEXT: [[DOTSPLAT5:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT4]], <2 x i32> poison, <2 x i32> zeroinitializer
-; IND-NEXT: [[INDUCTION6:%.*]] = add nuw nsw <2 x i32> [[DOTSPLAT5]], <i32 0, i32 1>
+; IND-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[EXT]], i64 0
+; IND-NEXT: [[DOTSPLAT:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
+; IND-NEXT: [[INDUCTION:%.*]] = add nuw nsw <2 x i32> [[DOTSPLAT]], <i32 0, i32 1>
; IND-NEXT: br label [[VECTOR_BODY:%.*]]
; IND: vector.body:
; IND-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; IND-NEXT: [[VEC_IND7:%.*]] = phi <2 x i32> [ [[INDUCTION6]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT8:%.*]], [[VECTOR_BODY]] ]
+; IND-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; IND-NEXT: [[TMP12:%.*]] = trunc i32 [[INDEX]] to i8
; IND-NEXT: [[OFFSET_IDX:%.*]] = add i8 [[TMP12]], [[T]]
; IND-NEXT: [[TMP13:%.*]] = sext i8 [[OFFSET_IDX]] to i64
; IND-NEXT: [[TMP14:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 [[TMP13]]
; IND-NEXT: [[TMP15:%.*]] = bitcast i32* [[TMP14]] to <2 x i32>*
-; IND-NEXT: store <2 x i32> [[VEC_IND7]], <2 x i32>* [[TMP15]], align 4
+; IND-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP15]], align 4
; IND-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 2
-; IND-NEXT: [[VEC_IND_NEXT8]] = add <2 x i32> [[VEC_IND7]], <i32 2, i32 2>
+; IND-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
; IND-NEXT: [[TMP16:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; IND-NEXT: br i1 [[TMP16]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP36:![0-9]+]]
; IND: middle.block:
@@ -3690,28 +3690,28 @@ define void @wrappingindvars1(i8 %t, i32 %len, i32 *%A) {
; UNROLL-NEXT: br i1 [[TMP11]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
; UNROLL: vector.ph:
; UNROLL-NEXT: [[N_VEC:%.*]] = and i32 [[TMP0]], -4
-; UNROLL-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; UNROLL-NEXT: [[IND_END:%.*]] = add i8 [[CAST_CRD]], [[T]]
+; UNROLL-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; UNROLL-NEXT: [[IND_END:%.*]] = add i8 [[CAST_VTC]], [[T]]
; UNROLL-NEXT: [[IND_END3:%.*]] = add i32 [[N_VEC]], [[EXT]]
-; UNROLL-NEXT: [[DOTSPLATINSERT5:%.*]] = insertelement <2 x i32> poison, i32 [[EXT]], i64 0
-; UNROLL-NEXT: [[DOTSPLAT6:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT5]], <2 x i32> poison, <2 x i32> zeroinitializer
-; UNROLL-NEXT: [[INDUCTION7:%.*]] = add nuw nsw <2 x i32> [[DOTSPLAT6]], <i32 0, i32 1>
+; UNROLL-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[EXT]], i64 0
+; UNROLL-NEXT: [[DOTSPLAT:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
+; UNROLL-NEXT: [[INDUCTION:%.*]] = add nuw nsw <2 x i32> [[DOTSPLAT]], <i32 0, i32 1>
; UNROLL-NEXT: br label [[VECTOR_BODY:%.*]]
; UNROLL: vector.body:
; UNROLL-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; UNROLL-NEXT: [[VEC_IND8:%.*]] = phi <2 x i32> [ [[INDUCTION7]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT11:%.*]], [[VECTOR_BODY]] ]
-; UNROLL-NEXT: [[STEP_ADD9:%.*]] = add <2 x i32> [[VEC_IND8]], <i32 2, i32 2>
+; UNROLL-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; UNROLL-NEXT: [[STEP_ADD:%.*]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
; UNROLL-NEXT: [[TMP12:%.*]] = trunc i32 [[INDEX]] to i8
; UNROLL-NEXT: [[OFFSET_IDX:%.*]] = add i8 [[TMP12]], [[T]]
; UNROLL-NEXT: [[TMP13:%.*]] = sext i8 [[OFFSET_IDX]] to i64
; UNROLL-NEXT: [[TMP14:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 [[TMP13]]
; UNROLL-NEXT: [[TMP15:%.*]] = bitcast i32* [[TMP14]] to <2 x i32>*
-; UNROLL-NEXT: store <2 x i32> [[VEC_IND8]], <2 x i32>* [[TMP15]], align 4
+; UNROLL-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP15]], align 4
; UNROLL-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, i32* [[TMP14]], i64 2
; UNROLL-NEXT: [[TMP17:%.*]] = bitcast i32* [[TMP16]] to <2 x i32>*
-; UNROLL-NEXT: store <2 x i32> [[STEP_ADD9]], <2 x i32>* [[TMP17]], align 4
+; UNROLL-NEXT: store <2 x i32> [[STEP_ADD]], <2 x i32>* [[TMP17]], align 4
; UNROLL-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
-; UNROLL-NEXT: [[VEC_IND_NEXT11]] = add <2 x i32> [[VEC_IND8]], <i32 4, i32 4>
+; UNROLL-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[VEC_IND]], <i32 4, i32 4>
; UNROLL-NEXT: [[TMP18:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; UNROLL-NEXT: br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP36:![0-9]+]]
; UNROLL: middle.block:
@@ -3765,17 +3765,17 @@ define void @wrappingindvars1(i8 %t, i32 %len, i32 *%A) {
; UNROLL-NO-IC: vector.ph:
; UNROLL-NO-IC-NEXT: [[N_MOD_VF:%.*]] = urem i32 [[TMP0]], 4
; UNROLL-NO-IC-NEXT: [[N_VEC:%.*]] = sub i32 [[TMP0]], [[N_MOD_VF]]
-; UNROLL-NO-IC-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; UNROLL-NO-IC-NEXT: [[IND_END:%.*]] = add i8 [[T]], [[CAST_CRD]]
+; UNROLL-NO-IC-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; UNROLL-NO-IC-NEXT: [[IND_END:%.*]] = add i8 [[T]], [[CAST_VTC]]
; UNROLL-NO-IC-NEXT: [[IND_END3:%.*]] = add i32 [[EXT]], [[N_VEC]]
-; UNROLL-NO-IC-NEXT: [[DOTSPLATINSERT5:%.*]] = insertelement <2 x i32> poison, i32 [[EXT]], i32 0
-; UNROLL-NO-IC-NEXT: [[DOTSPLAT6:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT5]], <2 x i32> poison, <2 x i32> zeroinitializer
-; UNROLL-NO-IC-NEXT: [[INDUCTION7:%.*]] = add <2 x i32> [[DOTSPLAT6]], <i32 0, i32 1>
+; UNROLL-NO-IC-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[EXT]], i32 0
+; UNROLL-NO-IC-NEXT: [[DOTSPLAT:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
+; UNROLL-NO-IC-NEXT: [[INDUCTION:%.*]] = add <2 x i32> [[DOTSPLAT]], <i32 0, i32 1>
; UNROLL-NO-IC-NEXT: br label [[VECTOR_BODY:%.*]]
; UNROLL-NO-IC: vector.body:
; UNROLL-NO-IC-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; UNROLL-NO-IC-NEXT: [[VEC_IND8:%.*]] = phi <2 x i32> [ [[INDUCTION7]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT11:%.*]], [[VECTOR_BODY]] ]
-; UNROLL-NO-IC-NEXT: [[STEP_ADD9:%.*]] = add <2 x i32> [[VEC_IND8]], <i32 2, i32 2>
+; UNROLL-NO-IC-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; UNROLL-NO-IC-NEXT: [[STEP_ADD:%.*]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
; UNROLL-NO-IC-NEXT: [[TMP12:%.*]] = trunc i32 [[INDEX]] to i8
; UNROLL-NO-IC-NEXT: [[OFFSET_IDX:%.*]] = add i8 [[T]], [[TMP12]]
; UNROLL-NO-IC-NEXT: [[TMP13:%.*]] = add i8 [[OFFSET_IDX]], 0
@@ -3784,12 +3784,12 @@ define void @wrappingindvars1(i8 %t, i32 %len, i32 *%A) {
; UNROLL-NO-IC-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, i32* [[A]], i8 [[TMP14]]
; UNROLL-NO-IC-NEXT: [[TMP17:%.*]] = getelementptr inbounds i32, i32* [[TMP15]], i32 0
; UNROLL-NO-IC-NEXT: [[TMP18:%.*]] = bitcast i32* [[TMP17]] to <2 x i32>*
-; UNROLL-NO-IC-NEXT: store <2 x i32> [[VEC_IND8]], <2 x i32>* [[TMP18]], align 4
+; UNROLL-NO-IC-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP18]], align 4
; UNROLL-NO-IC-NEXT: [[TMP19:%.*]] = getelementptr inbounds i32, i32* [[TMP15]], i32 2
; UNROLL-NO-IC-NEXT: [[TMP20:%.*]] = bitcast i32* [[TMP19]] to <2 x i32>*
-; UNROLL-NO-IC-NEXT: store <2 x i32> [[STEP_ADD9]], <2 x i32>* [[TMP20]], align 4
+; UNROLL-NO-IC-NEXT: store <2 x i32> [[STEP_ADD]], <2 x i32>* [[TMP20]], align 4
; UNROLL-NO-IC-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
-; UNROLL-NO-IC-NEXT: [[VEC_IND_NEXT11]] = add <2 x i32> [[STEP_ADD9]], <i32 2, i32 2>
+; UNROLL-NO-IC-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[STEP_ADD]], <i32 2, i32 2>
; UNROLL-NO-IC-NEXT: [[TMP21:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; UNROLL-NO-IC-NEXT: br i1 [[TMP21]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP36:![0-9]+]]
; UNROLL-NO-IC: middle.block:
@@ -3840,28 +3840,28 @@ define void @wrappingindvars1(i8 %t, i32 %len, i32 *%A) {
; INTERLEAVE-NEXT: br i1 [[TMP11]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
; INTERLEAVE: vector.ph:
; INTERLEAVE-NEXT: [[N_VEC:%.*]] = and i32 [[TMP0]], -8
-; INTERLEAVE-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; INTERLEAVE-NEXT: [[IND_END:%.*]] = add i8 [[CAST_CRD]], [[T]]
+; INTERLEAVE-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; INTERLEAVE-NEXT: [[IND_END:%.*]] = add i8 [[CAST_VTC]], [[T]]
; INTERLEAVE-NEXT: [[IND_END3:%.*]] = add i32 [[N_VEC]], [[EXT]]
-; INTERLEAVE-NEXT: [[DOTSPLATINSERT5:%.*]] = insertelement <4 x i32> poison, i32 [[EXT]], i64 0
-; INTERLEAVE-NEXT: [[DOTSPLAT6:%.*]] = shufflevector <4 x i32> [[DOTSPLATINSERT5]], <4 x i32> poison, <4 x i32> zeroinitializer
-; INTERLEAVE-NEXT: [[INDUCTION7:%.*]] = add nuw nsw <4 x i32> [[DOTSPLAT6]], <i32 0, i32 1, i32 2, i32 3>
+; INTERLEAVE-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[EXT]], i64 0
+; INTERLEAVE-NEXT: [[DOTSPLAT:%.*]] = shufflevector <4 x i32> [[DOTSPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
+; INTERLEAVE-NEXT: [[INDUCTION:%.*]] = add nuw nsw <4 x i32> [[DOTSPLAT]], <i32 0, i32 1, i32 2, i32 3>
; INTERLEAVE-NEXT: br label [[VECTOR_BODY:%.*]]
; INTERLEAVE: vector.body:
; INTERLEAVE-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; INTERLEAVE-NEXT: [[VEC_IND8:%.*]] = phi <4 x i32> [ [[INDUCTION7]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT11:%.*]], [[VECTOR_BODY]] ]
-; INTERLEAVE-NEXT: [[STEP_ADD9:%.*]] = add <4 x i32> [[VEC_IND8]], <i32 4, i32 4, i32 4, i32 4>
+; INTERLEAVE-NEXT: [[VEC_IND:%.*]] = phi <4 x i32> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; INTERLEAVE-NEXT: [[STEP_ADD:%.*]] = add <4 x i32> [[VEC_IND]], <i32 4, i32 4, i32 4, i32 4>
; INTERLEAVE-NEXT: [[TMP12:%.*]] = trunc i32 [[INDEX]] to i8
; INTERLEAVE-NEXT: [[OFFSET_IDX:%.*]] = add i8 [[TMP12]], [[T]]
; INTERLEAVE-NEXT: [[TMP13:%.*]] = sext i8 [[OFFSET_IDX]] to i64
; INTERLEAVE-NEXT: [[TMP14:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 [[TMP13]]
; INTERLEAVE-NEXT: [[TMP15:%.*]] = bitcast i32* [[TMP14]] to <4 x i32>*
-; INTERLEAVE-NEXT: store <4 x i32> [[VEC_IND8]], <4 x i32>* [[TMP15]], align 4
+; INTERLEAVE-NEXT: store <4 x i32> [[VEC_IND]], <4 x i32>* [[TMP15]], align 4
; INTERLEAVE-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, i32* [[TMP14]], i64 4
; INTERLEAVE-NEXT: [[TMP17:%.*]] = bitcast i32* [[TMP16]] to <4 x i32>*
-; INTERLEAVE-NEXT: store <4 x i32> [[STEP_ADD9]], <4 x i32>* [[TMP17]], align 4
+; INTERLEAVE-NEXT: store <4 x i32> [[STEP_ADD]], <4 x i32>* [[TMP17]], align 4
; INTERLEAVE-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 8
-; INTERLEAVE-NEXT: [[VEC_IND_NEXT11]] = add <4 x i32> [[VEC_IND8]], <i32 8, i32 8, i32 8, i32 8>
+; INTERLEAVE-NEXT: [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], <i32 8, i32 8, i32 8, i32 8>
; INTERLEAVE-NEXT: [[TMP18:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; INTERLEAVE-NEXT: br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP36:![0-9]+]]
; INTERLEAVE: middle.block:
@@ -3947,26 +3947,26 @@ define void @wrappingindvars2(i8 %t, i32 %len, i32 *%A) {
; CHECK: vector.ph:
; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i32 [[TMP0]], 2
; CHECK-NEXT: [[N_VEC:%.*]] = sub i32 [[TMP0]], [[N_MOD_VF]]
-; CHECK-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; CHECK-NEXT: [[IND_END:%.*]] = add i8 [[T]], [[CAST_CRD]]
+; CHECK-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; CHECK-NEXT: [[IND_END:%.*]] = add i8 [[T]], [[CAST_VTC]]
; CHECK-NEXT: [[TMP12:%.*]] = mul i32 [[N_VEC]], 4
; CHECK-NEXT: [[IND_END2:%.*]] = add i32 [[EXT_MUL]], [[TMP12]]
-; CHECK-NEXT: [[DOTSPLATINSERT4:%.*]] = insertelement <2 x i32> poison, i32 [[EXT_MUL]], i32 0
-; CHECK-NEXT: [[DOTSPLAT5:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT4]], <2 x i32> poison, <2 x i32> zeroinitializer
-; CHECK-NEXT: [[INDUCTION6:%.*]] = add <2 x i32> [[DOTSPLAT5]], <i32 0, i32 4>
+; CHECK-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[EXT_MUL]], i32 0
+; CHECK-NEXT: [[DOTSPLAT:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
+; CHECK-NEXT: [[INDUCTION:%.*]] = add <2 x i32> [[DOTSPLAT]], <i32 0, i32 4>
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT: [[VEC_IND7:%.*]] = phi <2 x i32> [ [[INDUCTION6]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT8:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP13:%.*]] = trunc i32 [[INDEX]] to i8
; CHECK-NEXT: [[OFFSET_IDX:%.*]] = add i8 [[T]], [[TMP13]]
; CHECK-NEXT: [[TMP14:%.*]] = add i8 [[OFFSET_IDX]], 0
; CHECK-NEXT: [[TMP15:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i8 [[TMP14]]
; CHECK-NEXT: [[TMP16:%.*]] = getelementptr inbounds i32, i32* [[TMP15]], i32 0
; CHECK-NEXT: [[TMP17:%.*]] = bitcast i32* [[TMP16]] to <2 x i32>*
-; CHECK-NEXT: store <2 x i32> [[VEC_IND7]], <2 x i32>* [[TMP17]], align 4
+; CHECK-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP17]], align 4
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 2
-; CHECK-NEXT: [[VEC_IND_NEXT8]] = add <2 x i32> [[VEC_IND7]], <i32 8, i32 8>
+; CHECK-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[VEC_IND]], <i32 8, i32 8>
; CHECK-NEXT: [[TMP18:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT: br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP38:![0-9]+]]
; CHECK: middle.block:
@@ -4019,25 +4019,25 @@ define void @wrappingindvars2(i8 %t, i32 %len, i32 *%A) {
; IND-NEXT: br i1 [[TMP11]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
; IND: vector.ph:
; IND-NEXT: [[N_VEC:%.*]] = and i32 [[TMP0]], -2
-; IND-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; IND-NEXT: [[IND_END:%.*]] = add i8 [[CAST_CRD]], [[T]]
+; IND-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; IND-NEXT: [[IND_END:%.*]] = add i8 [[CAST_VTC]], [[T]]
; IND-NEXT: [[TMP12:%.*]] = add i32 [[N_VEC]], [[EXT]]
; IND-NEXT: [[IND_END2:%.*]] = shl i32 [[TMP12]], 2
-; IND-NEXT: [[DOTSPLATINSERT4:%.*]] = insertelement <2 x i32> poison, i32 [[EXT_MUL]], i64 0
-; IND-NEXT: [[DOTSPLAT5:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT4]], <2 x i32> poison, <2 x i32> zeroinitializer
-; IND-NEXT: [[INDUCTION6:%.*]] = add nuw nsw <2 x i32> [[DOTSPLAT5]], <i32 0, i32 4>
+; IND-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[EXT_MUL]], i64 0
+; IND-NEXT: [[DOTSPLAT:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
+; IND-NEXT: [[INDUCTION:%.*]] = add nuw nsw <2 x i32> [[DOTSPLAT]], <i32 0, i32 4>
; IND-NEXT: br label [[VECTOR_BODY:%.*]]
; IND: vector.body:
; IND-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; IND-NEXT: [[VEC_IND7:%.*]] = phi <2 x i32> [ [[INDUCTION6]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT8:%.*]], [[VECTOR_BODY]] ]
+; IND-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; IND-NEXT: [[TMP13:%.*]] = trunc i32 [[INDEX]] to i8
; IND-NEXT: [[OFFSET_IDX:%.*]] = add i8 [[TMP13]], [[T]]
; IND-NEXT: [[TMP14:%.*]] = sext i8 [[OFFSET_IDX]] to i64
; IND-NEXT: [[TMP15:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 [[TMP14]]
; IND-NEXT: [[TMP16:%.*]] = bitcast i32* [[TMP15]] to <2 x i32>*
-; IND-NEXT: store <2 x i32> [[VEC_IND7]], <2 x i32>* [[TMP16]], align 4
+; IND-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP16]], align 4
; IND-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 2
-; IND-NEXT: [[VEC_IND_NEXT8]] = add <2 x i32> [[VEC_IND7]], <i32 8, i32 8>
+; IND-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[VEC_IND]], <i32 8, i32 8>
; IND-NEXT: [[TMP17:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; IND-NEXT: br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP38:![0-9]+]]
; IND: middle.block:
@@ -4091,29 +4091,29 @@ define void @wrappingindvars2(i8 %t, i32 %len, i32 *%A) {
; UNROLL-NEXT: br i1 [[TMP11]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
; UNROLL: vector.ph:
; UNROLL-NEXT: [[N_VEC:%.*]] = and i32 [[TMP0]], -4
-; UNROLL-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; UNROLL-NEXT: [[IND_END:%.*]] = add i8 [[CAST_CRD]], [[T]]
+; UNROLL-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; UNROLL-NEXT: [[IND_END:%.*]] = add i8 [[CAST_VTC]], [[T]]
; UNROLL-NEXT: [[TMP12:%.*]] = add i32 [[N_VEC]], [[EXT]]
; UNROLL-NEXT: [[IND_END2:%.*]] = shl i32 [[TMP12]], 2
-; UNROLL-NEXT: [[DOTSPLATINSERT5:%.*]] = insertelement <2 x i32> poison, i32 [[EXT_MUL]], i64 0
-; UNROLL-NEXT: [[DOTSPLAT6:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT5]], <2 x i32> poison, <2 x i32> zeroinitializer
-; UNROLL-NEXT: [[INDUCTION7:%.*]] = add nuw nsw <2 x i32> [[DOTSPLAT6]], <i32 0, i32 4>
+; UNROLL-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[EXT_MUL]], i64 0
+; UNROLL-NEXT: [[DOTSPLAT:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
+; UNROLL-NEXT: [[INDUCTION:%.*]] = add nuw nsw <2 x i32> [[DOTSPLAT]], <i32 0, i32 4>
; UNROLL-NEXT: br label [[VECTOR_BODY:%.*]]
; UNROLL: vector.body:
; UNROLL-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; UNROLL-NEXT: [[VEC_IND8:%.*]] = phi <2 x i32> [ [[INDUCTION7]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT11:%.*]], [[VECTOR_BODY]] ]
-; UNROLL-NEXT: [[STEP_ADD9:%.*]] = add <2 x i32> [[VEC_IND8]], <i32 8, i32 8>
+; UNROLL-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; UNROLL-NEXT: [[STEP_ADD:%.*]] = add <2 x i32> [[VEC_IND]], <i32 8, i32 8>
; UNROLL-NEXT: [[TMP13:%.*]] = trunc i32 [[INDEX]] to i8
; UNROLL-NEXT: [[OFFSET_IDX:%.*]] = add i8 [[TMP13]], [[T]]
; UNROLL-NEXT: [[TMP14:%.*]] = sext i8 [[OFFSET_IDX]] to i64
; UNROLL-NEXT: [[TMP15:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 [[TMP14]]
; UNROLL-NEXT: [[TMP16:%.*]] = bitcast i32* [[TMP15]] to <2 x i32>*
-; UNROLL-NEXT: store <2 x i32> [[VEC_IND8]], <2 x i32>* [[TMP16]], align 4
+; UNROLL-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP16]], align 4
; UNROLL-NEXT: [[TMP17:%.*]] = getelementptr inbounds i32, i32* [[TMP15]], i64 2
; UNROLL-NEXT: [[TMP18:%.*]] = bitcast i32* [[TMP17]] to <2 x i32>*
-; UNROLL-NEXT: store <2 x i32> [[STEP_ADD9]], <2 x i32>* [[TMP18]], align 4
+; UNROLL-NEXT: store <2 x i32> [[STEP_ADD]], <2 x i32>* [[TMP18]], align 4
; UNROLL-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
-; UNROLL-NEXT: [[VEC_IND_NEXT11]] = add <2 x i32> [[VEC_IND8]], <i32 16, i32 16>
+; UNROLL-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[VEC_IND]], <i32 16, i32 16>
; UNROLL-NEXT: [[TMP19:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; UNROLL-NEXT: br i1 [[TMP19]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP38:![0-9]+]]
; UNROLL: middle.block:
@@ -4169,18 +4169,18 @@ define void @wrappingindvars2(i8 %t, i32 %len, i32 *%A) {
; UNROLL-NO-IC: vector.ph:
; UNROLL-NO-IC-NEXT: [[N_MOD_VF:%.*]] = urem i32 [[TMP0]], 4
; UNROLL-NO-IC-NEXT: [[N_VEC:%.*]] = sub i32 [[TMP0]], [[N_MOD_VF]]
-; UNROLL-NO-IC-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; UNROLL-NO-IC-NEXT: [[IND_END:%.*]] = add i8 [[T]], [[CAST_CRD]]
+; UNROLL-NO-IC-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; UNROLL-NO-IC-NEXT: [[IND_END:%.*]] = add i8 [[T]], [[CAST_VTC]]
; UNROLL-NO-IC-NEXT: [[TMP12:%.*]] = mul i32 [[N_VEC]], 4
; UNROLL-NO-IC-NEXT: [[IND_END2:%.*]] = add i32 [[EXT_MUL]], [[TMP12]]
-; UNROLL-NO-IC-NEXT: [[DOTSPLATINSERT5:%.*]] = insertelement <2 x i32> poison, i32 [[EXT_MUL]], i32 0
-; UNROLL-NO-IC-NEXT: [[DOTSPLAT6:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT5]], <2 x i32> poison, <2 x i32> zeroinitializer
-; UNROLL-NO-IC-NEXT: [[INDUCTION7:%.*]] = add <2 x i32> [[DOTSPLAT6]], <i32 0, i32 4>
+; UNROLL-NO-IC-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[EXT_MUL]], i32 0
+; UNROLL-NO-IC-NEXT: [[DOTSPLAT:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
+; UNROLL-NO-IC-NEXT: [[INDUCTION:%.*]] = add <2 x i32> [[DOTSPLAT]], <i32 0, i32 4>
; UNROLL-NO-IC-NEXT: br label [[VECTOR_BODY:%.*]]
; UNROLL-NO-IC: vector.body:
; UNROLL-NO-IC-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; UNROLL-NO-IC-NEXT: [[VEC_IND8:%.*]] = phi <2 x i32> [ [[INDUCTION7]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT11:%.*]], [[VECTOR_BODY]] ]
-; UNROLL-NO-IC-NEXT: [[STEP_ADD9:%.*]] = add <2 x i32> [[VEC_IND8]], <i32 8, i32 8>
+; UNROLL-NO-IC-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; UNROLL-NO-IC-NEXT: [[STEP_ADD:%.*]] = add <2 x i32> [[VEC_IND]], <i32 8, i32 8>
; UNROLL-NO-IC-NEXT: [[TMP13:%.*]] = trunc i32 [[INDEX]] to i8
; UNROLL-NO-IC-NEXT: [[OFFSET_IDX:%.*]] = add i8 [[T]], [[TMP13]]
; UNROLL-NO-IC-NEXT: [[TMP14:%.*]] = add i8 [[OFFSET_IDX]], 0
@@ -4189,12 +4189,12 @@ define void @wrappingindvars2(i8 %t, i32 %len, i32 *%A) {
; UNROLL-NO-IC-NEXT: [[TMP17:%.*]] = getelementptr inbounds i32, i32* [[A]], i8 [[TMP15]]
; UNROLL-NO-IC-NEXT: [[TMP18:%.*]] = getelementptr inbounds i32, i32* [[TMP16]], i32 0
; UNROLL-NO-IC-NEXT: [[TMP19:%.*]] = bitcast i32* [[TMP18]] to <2 x i32>*
-; UNROLL-NO-IC-NEXT: store <2 x i32> [[VEC_IND8]], <2 x i32>* [[TMP19]], align 4
+; UNROLL-NO-IC-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP19]], align 4
; UNROLL-NO-IC-NEXT: [[TMP20:%.*]] = getelementptr inbounds i32, i32* [[TMP16]], i32 2
; UNROLL-NO-IC-NEXT: [[TMP21:%.*]] = bitcast i32* [[TMP20]] to <2 x i32>*
-; UNROLL-NO-IC-NEXT: store <2 x i32> [[STEP_ADD9]], <2 x i32>* [[TMP21]], align 4
+; UNROLL-NO-IC-NEXT: store <2 x i32> [[STEP_ADD]], <2 x i32>* [[TMP21]], align 4
; UNROLL-NO-IC-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
-; UNROLL-NO-IC-NEXT: [[VEC_IND_NEXT11]] = add <2 x i32> [[STEP_ADD9]], <i32 8, i32 8>
+; UNROLL-NO-IC-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[STEP_ADD]], <i32 8, i32 8>
; UNROLL-NO-IC-NEXT: [[TMP22:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; UNROLL-NO-IC-NEXT: br i1 [[TMP22]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP38:![0-9]+]]
; UNROLL-NO-IC: middle.block:
@@ -4247,29 +4247,29 @@ define void @wrappingindvars2(i8 %t, i32 %len, i32 *%A) {
; INTERLEAVE-NEXT: br i1 [[TMP11]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
; INTERLEAVE: vector.ph:
; INTERLEAVE-NEXT: [[N_VEC:%.*]] = and i32 [[TMP0]], -8
-; INTERLEAVE-NEXT: [[CAST_CRD:%.*]] = trunc i32 [[N_VEC]] to i8
-; INTERLEAVE-NEXT: [[IND_END:%.*]] = add i8 [[CAST_CRD]], [[T]]
+; INTERLEAVE-NEXT: [[CAST_VTC:%.*]] = trunc i32 [[N_VEC]] to i8
+; INTERLEAVE-NEXT: [[IND_END:%.*]] = add i8 [[CAST_VTC]], [[T]]
; INTERLEAVE-NEXT: [[TMP12:%.*]] = add i32 [[N_VEC]], [[EXT]]
; INTERLEAVE-NEXT: [[IND_END2:%.*]] = shl i32 [[TMP12]], 2
-; INTERLEAVE-NEXT: [[DOTSPLATINSERT5:%.*]] = insertelement <4 x i32> poison, i32 [[EXT_MUL]], i64 0
-; INTERLEAVE-NEXT: [[DOTSPLAT6:%.*]] = shufflevector <4 x i32> [[DOTSPLATINSERT5]], <4 x i32> poison, <4 x i32> zeroinitializer
-; INTERLEAVE-NEXT: [[INDUCTION7:%.*]] = add nuw nsw <4 x i32> [[DOTSPLAT6]], <i32 0, i32 4, i32 8, i32 12>
+; INTERLEAVE-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[EXT_MUL]], i64 0
+; INTERLEAVE-NEXT: [[DOTSPLAT:%.*]] = shufflevector <4 x i32> [[DOTSPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
+; INTERLEAVE-NEXT: [[INDUCTION:%.*]] = add nuw nsw <4 x i32> [[DOTSPLAT]], <i32 0, i32 4, i32 8, i32 12>
; INTERLEAVE-NEXT: br label [[VECTOR_BODY:%.*]]
; INTERLEAVE: vector.body:
; INTERLEAVE-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; INTERLEAVE-NEXT: [[VEC_IND8:%.*]] = phi <4 x i32> [ [[INDUCTION7]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT11:%.*]], [[VECTOR_BODY]] ]
-; INTERLEAVE-NEXT: [[STEP_ADD9:%.*]] = add <4 x i32> [[VEC_IND8]], <i32 16, i32 16, i32 16, i32 16>
+; INTERLEAVE-NEXT: [[VEC_IND:%.*]] = phi <4 x i32> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; INTERLEAVE-NEXT: [[STEP_ADD:%.*]] = add <4 x i32> [[VEC_IND]], <i32 16, i32 16, i32 16, i32 16>
; INTERLEAVE-NEXT: [[TMP13:%.*]] = trunc i32 [[INDEX]] to i8
; INTERLEAVE-NEXT: [[OFFSET_IDX:%.*]] = add i8 [[TMP13]], [[T]]
; INTERLEAVE-NEXT: [[TMP14:%.*]] = sext i8 [[OFFSET_IDX]] to i64
; INTERLEAVE-NEXT: [[TMP15:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i64 [[TMP14]]
; INTERLEAVE-NEXT: [[TMP16:%.*]] = bitcast i32* [[TMP15]] to <4 x i32>*
-; INTERLEAVE-NEXT: store <4 x i32> [[VEC_IND8]], <4 x i32>* [[TMP16]], align 4
+; INTERLEAVE-NEXT: store <4 x i32> [[VEC_IND]], <4 x i32>* [[TMP16]], align 4
; INTERLEAVE-NEXT: [[TMP17:%.*]] = getelementptr inbounds i32, i32* [[TMP15]], i64 4
; INTERLEAVE-NEXT: [[TMP18:%.*]] = bitcast i32* [[TMP17]] to <4 x i32>*
-; INTERLEAVE-NEXT: store <4 x i32> [[STEP_ADD9]], <4 x i32>* [[TMP18]], align 4
+; INTERLEAVE-NEXT: store <4 x i32> [[STEP_ADD]], <4 x i32>* [[TMP18]], align 4
; INTERLEAVE-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 8
-; INTERLEAVE-NEXT: [[VEC_IND_NEXT11]] = add <4 x i32> [[VEC_IND8]], <i32 32, i32 32, i32 32, i32 32>
+; INTERLEAVE-NEXT: [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], <i32 32, i32 32, i32 32, i32 32>
; INTERLEAVE-NEXT: [[TMP19:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; INTERLEAVE-NEXT: br i1 [[TMP19]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP38:![0-9]+]]
; INTERLEAVE: middle.block:
@@ -4343,14 +4343,14 @@ define void @veciv(i32* nocapture %a, i32 %start, i32 %k) {
; CHECK-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ <i32 0, i32 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP0:%.*]] = add i32 [[INDEX]], 0
-; CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[TMP0]]
-; CHECK-NEXT: [[TMP3:%.*]] = getelementptr inbounds i32, i32* [[TMP2]], i32 0
-; CHECK-NEXT: [[TMP4:%.*]] = bitcast i32* [[TMP3]] to <2 x i32>*
-; CHECK-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP4]], align 4
+; CHECK-NEXT: [[TMP1:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[TMP0]]
+; CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds i32, i32* [[TMP1]], i32 0
+; CHECK-NEXT: [[TMP3:%.*]] = bitcast i32* [[TMP2]] to <2 x i32>*
+; CHECK-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP3]], align 4
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 2
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
-; CHECK-NEXT: [[TMP5:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT: br i1 [[TMP5]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP40:![0-9]+]]
+; CHECK-NEXT: [[TMP4:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT: br i1 [[TMP4]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP40:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i32 [[K]], [[N_VEC]]
; CHECK-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
@@ -4454,19 +4454,19 @@ define void @veciv(i32* nocapture %a, i32 %start, i32 %k) {
; UNROLL-NO-IC-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ <i32 0, i32 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; UNROLL-NO-IC-NEXT: [[STEP_ADD:%.*]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
; UNROLL-NO-IC-NEXT: [[TMP0:%.*]] = add i32 [[INDEX]], 0
-; UNROLL-NO-IC-NEXT: [[TMP2:%.*]] = add i32 [[INDEX]], 2
-; UNROLL-NO-IC-NEXT: [[TMP4:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[TMP0]]
-; UNROLL-NO-IC-NEXT: [[TMP5:%.*]] = getelementptr inbounds i32, i32* [[A]], i32 [[TMP2]]
-; UNROLL-NO-IC-NEXT: [[TMP6:%.*]] = getelementptr inbounds i32, i32* [[TMP4]], i32 0
+; UNROLL-NO-IC-NEXT: [[TMP1:%.*]] = add i32 [[INDEX]], 2
+; UNROLL-NO-IC-NEXT: [[TMP2:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[TMP0]]
+; UNROLL-NO-IC-NEXT: [[TMP3:%.*]] = getelementptr inbounds i32, i32* [[A]], i32 [[TMP1]]
+; UNROLL-NO-IC-NEXT: [[TMP4:%.*]] = getelementptr inbounds i32, i32* [[TMP2]], i32 0
+; UNROLL-NO-IC-NEXT: [[TMP5:%.*]] = bitcast i32* [[TMP4]] to <2 x i32>*
+; UNROLL-NO-IC-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP5]], align 4
+; UNROLL-NO-IC-NEXT: [[TMP6:%.*]] = getelementptr inbounds i32, i32* [[TMP2]], i32 2
; UNROLL-NO-IC-NEXT: [[TMP7:%.*]] = bitcast i32* [[TMP6]] to <2 x i32>*
-; UNROLL-NO-IC-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP7]], align 4
-; UNROLL-NO-IC-NEXT: [[TMP8:%.*]] = getelementptr inbounds i32, i32* [[TMP4]], i32 2
-; UNROLL-NO-IC-NEXT: [[TMP9:%.*]] = bitcast i32* [[TMP8]] to <2 x i32>*
-; UNROLL-NO-IC-NEXT: store <2 x i32> [[STEP_ADD]], <2 x i32>* [[TMP9]], align 4
+; UNROLL-NO-IC-NEXT: store <2 x i32> [[STEP_ADD]], <2 x i32>* [[TMP7]], align 4
; UNROLL-NO-IC-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
; UNROLL-NO-IC-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[STEP_ADD]], <i32 2, i32 2>
-; UNROLL-NO-IC-NEXT: [[TMP10:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
-; UNROLL-NO-IC-NEXT: br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP40:![0-9]+]]
+; UNROLL-NO-IC-NEXT: [[TMP8:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; UNROLL-NO-IC-NEXT: br i1 [[TMP8]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP40:![0-9]+]]
; UNROLL-NO-IC: middle.block:
; UNROLL-NO-IC-NEXT: [[CMP_N:%.*]] = icmp eq i32 [[K]], [[N_VEC]]
; UNROLL-NO-IC-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
@@ -4558,14 +4558,14 @@ define void @trunciv(i32* nocapture %a, i32 %start, i64 %k) {
; CHECK-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ <i32 0, i32 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP5:%.*]] = trunc i64 [[INDEX]] to i32
; CHECK-NEXT: [[TMP6:%.*]] = add i32 [[TMP5]], 0
-; CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[TMP6]]
-; CHECK-NEXT: [[TMP9:%.*]] = getelementptr inbounds i32, i32* [[TMP8]], i32 0
-; CHECK-NEXT: [[TMP10:%.*]] = bitcast i32* [[TMP9]] to <2 x i32>*
-; CHECK-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP10]], align 4
+; CHECK-NEXT: [[TMP7:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[TMP6]]
+; CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds i32, i32* [[TMP7]], i32 0
+; CHECK-NEXT: [[TMP9:%.*]] = bitcast i32* [[TMP8]] to <2 x i32>*
+; CHECK-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP9]], align 4
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
-; CHECK-NEXT: [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT: br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP42:![0-9]+]]
+; CHECK-NEXT: [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT: br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP42:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[K]], [[N_VEC]]
; CHECK-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
@@ -4692,19 +4692,19 @@ define void @trunciv(i32* nocapture %a, i32 %start, i64 %k) {
; UNROLL-NO-IC-NEXT: [[STEP_ADD:%.*]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
; UNROLL-NO-IC-NEXT: [[TMP5:%.*]] = trunc i64 [[INDEX]] to i32
; UNROLL-NO-IC-NEXT: [[TMP6:%.*]] = add i32 [[TMP5]], 0
-; UNROLL-NO-IC-NEXT: [[TMP8:%.*]] = add i32 [[TMP5]], 2
-; UNROLL-NO-IC-NEXT: [[TMP10:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[TMP6]]
-; UNROLL-NO-IC-NEXT: [[TMP11:%.*]] = getelementptr inbounds i32, i32* [[A]], i32 [[TMP8]]
-; UNROLL-NO-IC-NEXT: [[TMP12:%.*]] = getelementptr inbounds i32, i32* [[TMP10]], i32 0
+; UNROLL-NO-IC-NEXT: [[TMP7:%.*]] = add i32 [[TMP5]], 2
+; UNROLL-NO-IC-NEXT: [[TMP8:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[TMP6]]
+; UNROLL-NO-IC-NEXT: [[TMP9:%.*]] = getelementptr inbounds i32, i32* [[A]], i32 [[TMP7]]
+; UNROLL-NO-IC-NEXT: [[TMP10:%.*]] = getelementptr inbounds i32, i32* [[TMP8]], i32 0
+; UNROLL-NO-IC-NEXT: [[TMP11:%.*]] = bitcast i32* [[TMP10]] to <2 x i32>*
+; UNROLL-NO-IC-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP11]], align 4
+; UNROLL-NO-IC-NEXT: [[TMP12:%.*]] = getelementptr inbounds i32, i32* [[TMP8]], i32 2
; UNROLL-NO-IC-NEXT: [[TMP13:%.*]] = bitcast i32* [[TMP12]] to <2 x i32>*
-; UNROLL-NO-IC-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP13]], align 4
-; UNROLL-NO-IC-NEXT: [[TMP14:%.*]] = getelementptr inbounds i32, i32* [[TMP10]], i32 2
-; UNROLL-NO-IC-NEXT: [[TMP15:%.*]] = bitcast i32* [[TMP14]] to <2 x i32>*
-; UNROLL-NO-IC-NEXT: store <2 x i32> [[STEP_ADD]], <2 x i32>* [[TMP15]], align 4
+; UNROLL-NO-IC-NEXT: store <2 x i32> [[STEP_ADD]], <2 x i32>* [[TMP13]], align 4
; UNROLL-NO-IC-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; UNROLL-NO-IC-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[STEP_ADD]], <i32 2, i32 2>
-; UNROLL-NO-IC-NEXT: [[TMP16:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
-; UNROLL-NO-IC-NEXT: br i1 [[TMP16]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP42:![0-9]+]]
+; UNROLL-NO-IC-NEXT: [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; UNROLL-NO-IC-NEXT: br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP42:![0-9]+]]
; UNROLL-NO-IC: middle.block:
; UNROLL-NO-IC-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[K]], [[N_VEC]]
; UNROLL-NO-IC-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
@@ -4805,14 +4805,14 @@ define void @nonprimary(i32* nocapture %a, i32 %start, i32 %i, i32 %k) {
; CHECK-NEXT: [[VEC_IND:%.*]] = phi <2 x i32> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[OFFSET_IDX:%.*]] = add i32 [[I]], [[INDEX]]
; CHECK-NEXT: [[TMP1:%.*]] = add i32 [[OFFSET_IDX]], 0
-; CHECK-NEXT: [[TMP3:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[TMP1]]
-; CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds i32, i32* [[TMP3]], i32 0
-; CHECK-NEXT: [[TMP5:%.*]] = bitcast i32* [[TMP4]] to <2 x i32>*
-; CHECK-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP5]], align 4
+; CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[TMP1]]
+; CHECK-NEXT: [[TMP3:%.*]] = getelementptr inbounds i32, i32* [[TMP2]], i32 0
+; CHECK-NEXT: [[TMP4:%.*]] = bitcast i32* [[TMP3]] to <2 x i32>*
+; CHECK-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP4]], align 4
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 2
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
-; CHECK-NEXT: [[TMP6:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
-; CHECK-NEXT: br i1 [[TMP6]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP44:![0-9]+]]
+; CHECK-NEXT: [[TMP5:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT: br i1 [[TMP5]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP44:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i32 [[TMP0]], [[N_VEC]]
; CHECK-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
@@ -4934,19 +4934,19 @@ define void @nonprimary(i32* nocapture %a, i32 %start, i32 %i, i32 %k) {
; UNROLL-NO-IC-NEXT: [[STEP_ADD:%.*]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
; UNROLL-NO-IC-NEXT: [[OFFSET_IDX:%.*]] = add i32 [[I]], [[INDEX]]
; UNROLL-NO-IC-NEXT: [[TMP1:%.*]] = add i32 [[OFFSET_IDX]], 0
-; UNROLL-NO-IC-NEXT: [[TMP3:%.*]] = add i32 [[OFFSET_IDX]], 2
-; UNROLL-NO-IC-NEXT: [[TMP5:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[TMP1]]
-; UNROLL-NO-IC-NEXT: [[TMP6:%.*]] = getelementptr inbounds i32, i32* [[A]], i32 [[TMP3]]
-; UNROLL-NO-IC-NEXT: [[TMP7:%.*]] = getelementptr inbounds i32, i32* [[TMP5]], i32 0
+; UNROLL-NO-IC-NEXT: [[TMP2:%.*]] = add i32 [[OFFSET_IDX]], 2
+; UNROLL-NO-IC-NEXT: [[TMP3:%.*]] = getelementptr inbounds i32, i32* [[A:%.*]], i32 [[TMP1]]
+; UNROLL-NO-IC-NEXT: [[TMP4:%.*]] = getelementptr inbounds i32, i32* [[A]], i32 [[TMP2]]
+; UNROLL-NO-IC-NEXT: [[TMP5:%.*]] = getelementptr inbounds i32, i32* [[TMP3]], i32 0
+; UNROLL-NO-IC-NEXT: [[TMP6:%.*]] = bitcast i32* [[TMP5]] to <2 x i32>*
+; UNROLL-NO-IC-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP6]], align 4
+; UNROLL-NO-IC-NEXT: [[TMP7:%.*]] = getelementptr inbounds i32, i32* [[TMP3]], i32 2
; UNROLL-NO-IC-NEXT: [[TMP8:%.*]] = bitcast i32* [[TMP7]] to <2 x i32>*
-; UNROLL-NO-IC-NEXT: store <2 x i32> [[VEC_IND]], <2 x i32>* [[TMP8]], align 4
-; UNROLL-NO-IC-NEXT: [[TMP9:%.*]] = getelementptr inbounds i32, i32* [[TMP5]], i32 2
-; UNROLL-NO-IC-NEXT: [[TMP10:%.*]] = bitcast i32* [[TMP9]] to <2 x i32>*
-; UNROLL-NO-IC-NEXT: store <2 x i32> [[STEP_ADD]], <2 x i32>* [[TMP10]], align 4
+; UNROLL-NO-IC-NEXT: store <2 x i32> [[STEP_ADD]], <2 x i32>* [[TMP8]], align 4
; UNROLL-NO-IC-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
; UNROLL-NO-IC-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[STEP_ADD]], <i32 2, i32 2>
-; UNROLL-NO-IC-NEXT: [[TMP11:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
-; UNROLL-NO-IC-NEXT: br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP44:![0-9]+]]
+; UNROLL-NO-IC-NEXT: [[TMP9:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
+; UNROLL-NO-IC-NEXT: br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP44:![0-9]+]]
; UNROLL-NO-IC: middle.block:
; UNROLL-NO-IC-NEXT: [[CMP_N:%.*]] = icmp eq i32 [[TMP0]], [[N_VEC]]
; UNROLL-NO-IC-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
@@ -6131,20 +6131,20 @@ define void @pr52460_first_order_recurrence_truncated_iv(i32* noalias %src, i32*
; CHECK-NEXT: [[VEC_IND]] = phi <2 x i32> [ <i32 0, i32 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP0:%.*]] = trunc i64 [[INDEX]] to i32
; CHECK-NEXT: [[TMP1:%.*]] = add i32 [[TMP0]], 0
-; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <2 x i32> [[VECTOR_RECUR]], <2 x i32> [[VEC_IND]], <2 x i32> <i32 1, i32 2>
-; CHECK-NEXT: [[TMP4:%.*]] = load i32, i32* [[SRC:%.*]], align 4
-; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[TMP4]], i32 0
+; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <2 x i32> [[VECTOR_RECUR]], <2 x i32> [[VEC_IND]], <2 x i32> <i32 1, i32 2>
+; CHECK-NEXT: [[TMP3:%.*]] = load i32, i32* [[SRC:%.*]], align 4
+; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[TMP3]], i32 0
; CHECK-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x i32> [[BROADCAST_SPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
-; CHECK-NEXT: [[TMP5:%.*]] = mul nsw <2 x i32> [[BROADCAST_SPLAT]], [[TMP3]]
-; CHECK-NEXT: [[TMP6:%.*]] = getelementptr i32, i32* [[DST:%.*]], i32 [[TMP1]]
-; CHECK-NEXT: [[TMP7:%.*]] = add <2 x i32> [[VEC_IND]], [[TMP5]]
-; CHECK-NEXT: [[TMP8:%.*]] = getelementptr i32, i32* [[TMP6]], i32 0
-; CHECK-NEXT: [[TMP9:%.*]] = bitcast i32* [[TMP8]] to <2 x i32>*
-; CHECK-NEXT: store <2 x i32> [[TMP7]], <2 x i32>* [[TMP9]], align 4
+; CHECK-NEXT: [[TMP4:%.*]] = mul nsw <2 x i32> [[BROADCAST_SPLAT]], [[TMP2]]
+; CHECK-NEXT: [[TMP5:%.*]] = getelementptr i32, i32* [[DST:%.*]], i32 [[TMP1]]
+; CHECK-NEXT: [[TMP6:%.*]] = add <2 x i32> [[VEC_IND]], [[TMP4]]
+; CHECK-NEXT: [[TMP7:%.*]] = getelementptr i32, i32* [[TMP5]], i32 0
+; CHECK-NEXT: [[TMP8:%.*]] = bitcast i32* [[TMP7]] to <2 x i32>*
+; CHECK-NEXT: store <2 x i32> [[TMP6]], <2 x i32>* [[TMP8]], align 4
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
-; CHECK-NEXT: [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT]], 100
-; CHECK-NEXT: br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP52:![0-9]+]]
+; CHECK-NEXT: [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], 100
+; CHECK-NEXT: br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP52:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 100, 100
; CHECK-NEXT: [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <2 x i32> [[VEC_IND]], i32 1
@@ -6259,31 +6259,31 @@ define void @pr52460_first_order_recurrence_truncated_iv(i32* noalias %src, i32*
; UNROLL-NO-IC-NEXT: [[STEP_ADD]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
; UNROLL-NO-IC-NEXT: [[TMP0:%.*]] = trunc i64 [[INDEX]] to i32
; UNROLL-NO-IC-NEXT: [[TMP1:%.*]] = add i32 [[TMP0]], 0
-; UNROLL-NO-IC-NEXT: [[TMP3:%.*]] = add i32 [[TMP0]], 2
-; UNROLL-NO-IC-NEXT: [[TMP5:%.*]] = shufflevector <2 x i32> [[VECTOR_RECUR]], <2 x i32> [[VEC_IND]], <2 x i32> <i32 1, i32 2>
-; UNROLL-NO-IC-NEXT: [[TMP6:%.*]] = shufflevector <2 x i32> [[VEC_IND]], <2 x i32> [[STEP_ADD]], <2 x i32> <i32 1, i32 2>
-; UNROLL-NO-IC-NEXT: [[TMP7:%.*]] = load i32, i32* [[SRC:%.*]], align 4
-; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[TMP7]], i32 0
+; UNROLL-NO-IC-NEXT: [[TMP2:%.*]] = add i32 [[TMP0]], 2
+; UNROLL-NO-IC-NEXT: [[TMP3:%.*]] = shufflevector <2 x i32> [[VECTOR_RECUR]], <2 x i32> [[VEC_IND]], <2 x i32> <i32 1, i32 2>
+; UNROLL-NO-IC-NEXT: [[TMP4:%.*]] = shufflevector <2 x i32> [[VEC_IND]], <2 x i32> [[STEP_ADD]], <2 x i32> <i32 1, i32 2>
+; UNROLL-NO-IC-NEXT: [[TMP5:%.*]] = load i32, i32* [[SRC:%.*]], align 4
+; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[TMP5]], i32 0
; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x i32> [[BROADCAST_SPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
-; UNROLL-NO-IC-NEXT: [[TMP8:%.*]] = load i32, i32* [[SRC]], align 4
-; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLATINSERT3:%.*]] = insertelement <2 x i32> poison, i32 [[TMP8]], i32 0
+; UNROLL-NO-IC-NEXT: [[TMP6:%.*]] = load i32, i32* [[SRC]], align 4
+; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLATINSERT3:%.*]] = insertelement <2 x i32> poison, i32 [[TMP6]], i32 0
; UNROLL-NO-IC-NEXT: [[BROADCAST_SPLAT4:%.*]] = shufflevector <2 x i32> [[BROADCAST_SPLATINSERT3]], <2 x i32> poison, <2 x i32> zeroinitializer
-; UNROLL-NO-IC-NEXT: [[TMP9:%.*]] = mul nsw <2 x i32> [[BROADCAST_SPLAT]], [[TMP5]]
-; UNROLL-NO-IC-NEXT: [[TMP10:%.*]] = mul nsw <2 x i32> [[BROADCAST_SPLAT4]], [[TMP6]]
-; UNROLL-NO-IC-NEXT: [[TMP11:%.*]] = getelementptr i32, i32* [[DST:%.*]], i32 [[TMP1]]
-; UNROLL-NO-IC-NEXT: [[TMP12:%.*]] = getelementptr i32, i32* [[DST]], i32 [[TMP3]]
-; UNROLL-NO-IC-NEXT: [[TMP13:%.*]] = add <2 x i32> [[VEC_IND]], [[TMP9]]
-; UNROLL-NO-IC-NEXT: [[TMP14:%.*]] = add <2 x i32> [[STEP_ADD]], [[TMP10]]
-; UNROLL-NO-IC-NEXT: [[TMP15:%.*]] = getelementptr i32, i32* [[TMP11]], i32 0
+; UNROLL-NO-IC-NEXT: [[TMP7:%.*]] = mul nsw <2 x i32> [[BROADCAST_SPLAT]], [[TMP3]]
+; UNROLL-NO-IC-NEXT: [[TMP8:%.*]] = mul nsw <2 x i32> [[BROADCAST_SPLAT4]], [[TMP4]]
+; UNROLL-NO-IC-NEXT: [[TMP9:%.*]] = getelementptr i32, i32* [[DST:%.*]], i32 [[TMP1]]
+; UNROLL-NO-IC-NEXT: [[TMP10:%.*]] = getelementptr i32, i32* [[DST]], i32 [[TMP2]]
+; UNROLL-NO-IC-NEXT: [[TMP11:%.*]] = add <2 x i32> [[VEC_IND]], [[TMP7]]
+; UNROLL-NO-IC-NEXT: [[TMP12:%.*]] = add <2 x i32> [[STEP_ADD]], [[TMP8]]
+; UNROLL-NO-IC-NEXT: [[TMP13:%.*]] = getelementptr i32, i32* [[TMP9]], i32 0
+; UNROLL-NO-IC-NEXT: [[TMP14:%.*]] = bitcast i32* [[TMP13]] to <2 x i32>*
+; UNROLL-NO-IC-NEXT: store <2 x i32> [[TMP11]], <2 x i32>* [[TMP14]], align 4
+; UNROLL-NO-IC-NEXT: [[TMP15:%.*]] = getelementptr i32, i32* [[TMP9]], i32 2
; UNROLL-NO-IC-NEXT: [[TMP16:%.*]] = bitcast i32* [[TMP15]] to <2 x i32>*
-; UNROLL-NO-IC-NEXT: store <2 x i32> [[TMP13]], <2 x i32>* [[TMP16]], align 4
-; UNROLL-NO-IC-NEXT: [[TMP17:%.*]] = getelementptr i32, i32* [[TMP11]], i32 2
-; UNROLL-NO-IC-NEXT: [[TMP18:%.*]] = bitcast i32* [[TMP17]] to <2 x i32>*
-; UNROLL-NO-IC-NEXT: store <2 x i32> [[TMP14]], <2 x i32>* [[TMP18]], align 4
+; UNROLL-NO-IC-NEXT: store <2 x i32> [[TMP12]], <2 x i32>* [[TMP16]], align 4
; UNROLL-NO-IC-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; UNROLL-NO-IC-NEXT: [[VEC_IND_NEXT]] = add <2 x i32> [[STEP_ADD]], <i32 2, i32 2>
-; UNROLL-NO-IC-NEXT: [[TMP19:%.*]] = icmp eq i64 [[INDEX_NEXT]], 100
-; UNROLL-NO-IC-NEXT: br i1 [[TMP19]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP52:![0-9]+]]
+; UNROLL-NO-IC-NEXT: [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], 100
+; UNROLL-NO-IC-NEXT: br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP52:![0-9]+]]
; UNROLL-NO-IC: middle.block:
; UNROLL-NO-IC-NEXT: [[CMP_N:%.*]] = icmp eq i64 100, 100
; UNROLL-NO-IC-NEXT: [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <2 x i32> [[STEP_ADD]], i32 1
@@ -6428,8 +6428,8 @@ define void @test_optimized_cast_induction_feeding_first_order_recurrence(i64 %n
; CHECK: vector.ph:
; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], 2
; CHECK-NEXT: [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]]
-; CHECK-NEXT: [[CAST_CRD:%.*]] = trunc i64 [[N_VEC]] to i32
-; CHECK-NEXT: [[IND_END:%.*]] = mul i32 [[CAST_CRD]], [[STEP]]
+; CHECK-NEXT: [[CAST_VTC:%.*]] = trunc i64 [[N_VEC]] to i32
+; CHECK-NEXT: [[IND_END:%.*]] = mul i32 [[CAST_VTC]], [[STEP]]
; CHECK-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[STEP]], i32 0
; CHECK-NEXT: [[DOTSPLAT:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
; CHECK-NEXT: [[TMP17:%.*]] = mul <2 x i32> <i32 0, i32 1>, [[DOTSPLAT]]
@@ -6504,8 +6504,8 @@ define void @test_optimized_cast_induction_feeding_first_order_recurrence(i64 %n
; IND-NEXT: br i1 [[TMP15]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
; IND: vector.ph:
; IND-NEXT: [[N_VEC:%.*]] = and i64 [[N]], -2
-; IND-NEXT: [[CAST_CRD:%.*]] = trunc i64 [[N_VEC]] to i32
-; IND-NEXT: [[IND_END:%.*]] = mul i32 [[CAST_CRD]], [[STEP]]
+; IND-NEXT: [[CAST_VTC:%.*]] = trunc i64 [[N_VEC]] to i32
+; IND-NEXT: [[IND_END:%.*]] = mul i32 [[CAST_VTC]], [[STEP]]
; IND-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[STEP]], i64 0
; IND-NEXT: [[DOTSPLAT:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
; IND-NEXT: [[TMP16:%.*]] = mul nuw <2 x i32> [[DOTSPLAT]], <i32 0, i32 1>
@@ -6576,8 +6576,8 @@ define void @test_optimized_cast_induction_feeding_first_order_recurrence(i64 %n
; UNROLL-NEXT: br i1 [[TMP15]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
; UNROLL: vector.ph:
; UNROLL-NEXT: [[N_VEC:%.*]] = and i64 [[N]], -4
-; UNROLL-NEXT: [[CAST_CRD:%.*]] = trunc i64 [[N_VEC]] to i32
-; UNROLL-NEXT: [[IND_END:%.*]] = mul i32 [[CAST_CRD]], [[STEP]]
+; UNROLL-NEXT: [[CAST_VTC:%.*]] = trunc i64 [[N_VEC]] to i32
+; UNROLL-NEXT: [[IND_END:%.*]] = mul i32 [[CAST_VTC]], [[STEP]]
; UNROLL-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[STEP]], i64 0
; UNROLL-NEXT: [[DOTSPLAT:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
; UNROLL-NEXT: [[TMP16:%.*]] = mul nuw <2 x i32> [[DOTSPLAT]], <i32 0, i32 1>
@@ -6656,8 +6656,8 @@ define void @test_optimized_cast_induction_feeding_first_order_recurrence(i64 %n
; UNROLL-NO-IC: vector.ph:
; UNROLL-NO-IC-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], 4
; UNROLL-NO-IC-NEXT: [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]]
-; UNROLL-NO-IC-NEXT: [[CAST_CRD:%.*]] = trunc i64 [[N_VEC]] to i32
-; UNROLL-NO-IC-NEXT: [[IND_END:%.*]] = mul i32 [[CAST_CRD]], [[STEP]]
+; UNROLL-NO-IC-NEXT: [[CAST_VTC:%.*]] = trunc i64 [[N_VEC]] to i32
+; UNROLL-NO-IC-NEXT: [[IND_END:%.*]] = mul i32 [[CAST_VTC]], [[STEP]]
; UNROLL-NO-IC-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[STEP]], i32 0
; UNROLL-NO-IC-NEXT: [[DOTSPLAT:%.*]] = shufflevector <2 x i32> [[DOTSPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
; UNROLL-NO-IC-NEXT: [[TMP17:%.*]] = mul <2 x i32> <i32 0, i32 1>, [[DOTSPLAT]]
@@ -6739,8 +6739,8 @@ define void @test_optimized_cast_induction_feeding_first_order_recurrence(i64 %n
; INTERLEAVE-NEXT: br i1 [[TMP15]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
; INTERLEAVE: vector.ph:
; INTERLEAVE-NEXT: [[N_VEC:%.*]] = and i64 [[N]], -8
-; INTERLEAVE-NEXT: [[CAST_CRD:%.*]] = trunc i64 [[N_VEC]] to i32
-; INTERLEAVE-NEXT: [[IND_END:%.*]] = mul i32 [[CAST_CRD]], [[STEP]]
+; INTERLEAVE-NEXT: [[CAST_VTC:%.*]] = trunc i64 [[N_VEC]] to i32
+; INTERLEAVE-NEXT: [[IND_END:%.*]] = mul i32 [[CAST_VTC]], [[STEP]]
; INTERLEAVE-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[STEP]], i64 0
; INTERLEAVE-NEXT: [[DOTSPLAT:%.*]] = shufflevector <4 x i32> [[DOTSPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
; INTERLEAVE-NEXT: [[TMP16:%.*]] = mul <4 x i32> [[DOTSPLAT]], <i32 0, i32 1, i32 2, i32 3>
More information about the llvm-commits
mailing list