; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -passes=loop-vectorize,instcombine -S -mattr=avx512f < %s | FileCheck %s

target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"

; first test checks that loop with a reduction and a uniform store gets
; vectorized.

define i32 @inv_val_store_to_inv_address_with_reduction(ptr %a, i64 %n, ptr %b) {
; CHECK-LABEL: @inv_val_store_to_inv_address_with_reduction(
; CHECK-NEXT:  iter.check:
; CHECK-NEXT:    [[NTRUNC:%.*]] = trunc i64 [[N:%.*]] to i32
; CHECK-NEXT:    [[SMAX2:%.*]] = call i64 @llvm.smax.i64(i64 [[N]], i64 1)
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[SMAX2]], 8
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH:%.*]], label [[VECTOR_MEMCHECK:%.*]]
; CHECK:       vector.memcheck:
; CHECK-NEXT:    [[UGLYGEP:%.*]] = getelementptr i8, ptr [[A:%.*]], i64 4
; CHECK-NEXT:    [[SMAX:%.*]] = call i64 @llvm.smax.i64(i64 [[N]], i64 1)
; CHECK-NEXT:    [[TMP0:%.*]] = shl i64 [[SMAX]], 2
; CHECK-NEXT:    [[UGLYGEP1:%.*]] = getelementptr i8, ptr [[B:%.*]], i64 [[TMP0]]
; CHECK-NEXT:    [[BOUND0:%.*]] = icmp ugt ptr [[UGLYGEP1]], [[A]]
; CHECK-NEXT:    [[BOUND1:%.*]] = icmp ugt ptr [[UGLYGEP]], [[B]]
; CHECK-NEXT:    [[FOUND_CONFLICT:%.*]] = and i1 [[BOUND0]], [[BOUND1]]
; CHECK-NEXT:    br i1 [[FOUND_CONFLICT]], label [[VEC_EPILOG_SCALAR_PH]], label [[VECTOR_MAIN_LOOP_ITER_CHECK:%.*]]
; CHECK:       vector.main.loop.iter.check:
; CHECK-NEXT:    [[MIN_ITERS_CHECK3:%.*]] = icmp ult i64 [[SMAX2]], 64
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK3]], label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[N_VEC:%.*]] = and i64 [[SMAX2]], 9223372036854775744
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <16 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP5:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI4:%.*]] = phi <16 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP6:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI5:%.*]] = phi <16 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP7:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI6:%.*]] = phi <16 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP8:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <16 x i32>, ptr [[TMP1]], align 8, !alias.scope !0
; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i32, ptr [[TMP1]], i64 16
; CHECK-NEXT:    [[WIDE_LOAD7:%.*]] = load <16 x i32>, ptr [[TMP2]], align 8, !alias.scope !0
; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds i32, ptr [[TMP1]], i64 32
; CHECK-NEXT:    [[WIDE_LOAD8:%.*]] = load <16 x i32>, ptr [[TMP3]], align 8, !alias.scope !0
; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i32, ptr [[TMP1]], i64 48
; CHECK-NEXT:    [[WIDE_LOAD9:%.*]] = load <16 x i32>, ptr [[TMP4]], align 8, !alias.scope !0
; CHECK-NEXT:    [[TMP5]] = add <16 x i32> [[VEC_PHI]], [[WIDE_LOAD]]
; CHECK-NEXT:    [[TMP6]] = add <16 x i32> [[VEC_PHI4]], [[WIDE_LOAD7]]
; CHECK-NEXT:    [[TMP7]] = add <16 x i32> [[VEC_PHI5]], [[WIDE_LOAD8]]
; CHECK-NEXT:    [[TMP8]] = add <16 x i32> [[VEC_PHI6]], [[WIDE_LOAD9]]
; CHECK-NEXT:    store i32 [[NTRUNC]], ptr [[A]], align 4, !alias.scope !3, !noalias !0
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 64
; CHECK-NEXT:    [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    [[BIN_RDX:%.*]] = add <16 x i32> [[TMP6]], [[TMP5]]
; CHECK-NEXT:    [[BIN_RDX10:%.*]] = add <16 x i32> [[TMP7]], [[BIN_RDX]]
; CHECK-NEXT:    [[BIN_RDX11:%.*]] = add <16 x i32> [[TMP8]], [[BIN_RDX10]]
; CHECK-NEXT:    [[TMP10:%.*]] = call i32 @llvm.vector.reduce.add.v16i32(<16 x i32> [[BIN_RDX11]])
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[SMAX2]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label [[FOR_END:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]]
; CHECK:       vec.epilog.iter.check:
; CHECK-NEXT:    [[N_VEC_REMAINING:%.*]] = and i64 [[SMAX2]], 56
; CHECK-NEXT:    [[MIN_EPILOG_ITERS_CHECK:%.*]] = icmp eq i64 [[N_VEC_REMAINING]], 0
; CHECK-NEXT:    br i1 [[MIN_EPILOG_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]]
; CHECK:       vec.epilog.ph:
; CHECK-NEXT:    [[BC_MERGE_RDX:%.*]] = phi i32 [ 0, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ], [ [[TMP10]], [[VEC_EPILOG_ITER_CHECK]] ]
; CHECK-NEXT:    [[VEC_EPILOG_RESUME_VAL:%.*]] = phi i64 [ 0, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ], [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ]
; CHECK-NEXT:    [[N_VEC13:%.*]] = and i64 [[SMAX2]], 9223372036854775800
; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <8 x i32> <i32 poison, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>, i32 [[BC_MERGE_RDX]], i64 0
; CHECK-NEXT:    br label [[VEC_EPILOG_VECTOR_BODY:%.*]]
; CHECK:       vec.epilog.vector.body:
; CHECK-NEXT:    [[INDEX15:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT18:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
; CHECK-NEXT:    [[VEC_PHI16:%.*]] = phi <8 x i32> [ [[TMP11]], [[VEC_EPILOG_PH]] ], [ [[TMP13:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[INDEX15]]
; CHECK-NEXT:    [[WIDE_LOAD17:%.*]] = load <8 x i32>, ptr [[TMP12]], align 8, !alias.scope !8
; CHECK-NEXT:    [[TMP13]] = add <8 x i32> [[VEC_PHI16]], [[WIDE_LOAD17]]
; CHECK-NEXT:    store i32 [[NTRUNC]], ptr [[A]], align 4, !alias.scope !11, !noalias !8
; CHECK-NEXT:    [[INDEX_NEXT18]] = add nuw i64 [[INDEX15]], 8
; CHECK-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT18]], [[N_VEC13]]
; CHECK-NEXT:    br i1 [[TMP14]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP13:![0-9]+]]
; CHECK:       vec.epilog.middle.block:
; CHECK-NEXT:    [[TMP15:%.*]] = call i32 @llvm.vector.reduce.add.v8i32(<8 x i32> [[TMP13]])
; CHECK-NEXT:    [[CMP_N14:%.*]] = icmp eq i64 [[SMAX2]], [[N_VEC13]]
; CHECK-NEXT:    br i1 [[CMP_N14]], label [[FOR_END]], label [[VEC_EPILOG_SCALAR_PH]]
; CHECK:       vec.epilog.scalar.ph:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC13]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MEMCHECK]] ], [ 0, [[ITER_CHECK:%.*]] ]
; CHECK-NEXT:    [[BC_MERGE_RDX19:%.*]] = phi i32 [ [[TMP15]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[TMP10]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MEMCHECK]] ], [ 0, [[ITER_CHECK]] ]
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.body:
; CHECK-NEXT:    [[I:%.*]] = phi i64 [ [[I_NEXT:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[VEC_EPILOG_SCALAR_PH]] ]
; CHECK-NEXT:    [[T0:%.*]] = phi i32 [ [[T3:%.*]], [[FOR_BODY]] ], [ [[BC_MERGE_RDX19]], [[VEC_EPILOG_SCALAR_PH]] ]
; CHECK-NEXT:    [[T1:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[I]]
; CHECK-NEXT:    [[T2:%.*]] = load i32, ptr [[T1]], align 8
; CHECK-NEXT:    [[T3]] = add i32 [[T0]], [[T2]]
; CHECK-NEXT:    store i32 [[NTRUNC]], ptr [[A]], align 4
; CHECK-NEXT:    [[I_NEXT]] = add nuw nsw i64 [[I]], 1
; CHECK-NEXT:    [[COND:%.*]] = icmp slt i64 [[I_NEXT]], [[N]]
; CHECK-NEXT:    br i1 [[COND]], label [[FOR_BODY]], label [[FOR_END]], !llvm.loop [[LOOP14:![0-9]+]]
; CHECK:       for.end:
; CHECK-NEXT:    [[T4:%.*]] = phi i32 [ [[T3]], [[FOR_BODY]] ], [ [[TMP10]], [[MIDDLE_BLOCK]] ], [ [[TMP15]], [[VEC_EPILOG_MIDDLE_BLOCK]] ]
; CHECK-NEXT:    ret i32 [[T4]]
;
entry:
  %ntrunc = trunc i64 %n to i32
  br label %for.body

for.body:                                         ; preds = %for.body, %entry
  %i = phi i64 [ %i.next, %for.body ], [ 0, %entry ]
  %t0 = phi i32 [ %t3, %for.body ], [ 0, %entry ]
  %t1 = getelementptr inbounds i32, ptr %b, i64 %i
  %t2 = load i32, ptr %t1, align 8
  %t3 = add i32 %t0, %t2
  store i32 %ntrunc, ptr %a
  %i.next = add nuw nsw i64 %i, 1
  %cond = icmp slt i64 %i.next, %n
  br i1 %cond, label %for.body, label %for.end

for.end:                                          ; preds = %for.body
  %t4 = phi i32 [ %t3, %for.body ]
  ret i32 %t4
}

; Conditional store
; if (b[i] == k) a = ntrunc
define void @inv_val_store_to_inv_address_conditional(ptr %a, i64 %n, ptr %b, i32 %k) {
; CHECK-LABEL: @inv_val_store_to_inv_address_conditional(
; CHECK-NEXT:  iter.check:
; CHECK-NEXT:    [[NTRUNC:%.*]] = trunc i64 [[N:%.*]] to i32
; CHECK-NEXT:    [[SMAX2:%.*]] = call i64 @llvm.smax.i64(i64 [[N]], i64 1)
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[SMAX2]], 8
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH:%.*]], label [[VECTOR_MEMCHECK:%.*]]
; CHECK:       vector.memcheck:
; CHECK-NEXT:    [[SMAX:%.*]] = call i64 @llvm.smax.i64(i64 [[N]], i64 1)
; CHECK-NEXT:    [[TMP0:%.*]] = shl i64 [[SMAX]], 2
; CHECK-NEXT:    [[UGLYGEP:%.*]] = getelementptr i8, ptr [[B:%.*]], i64 [[TMP0]]
; CHECK-NEXT:    [[UGLYGEP1:%.*]] = getelementptr i8, ptr [[A:%.*]], i64 4
; CHECK-NEXT:    [[BOUND0:%.*]] = icmp ugt ptr [[UGLYGEP1]], [[B]]
; CHECK-NEXT:    [[BOUND1:%.*]] = icmp ugt ptr [[UGLYGEP]], [[A]]
; CHECK-NEXT:    [[FOUND_CONFLICT:%.*]] = and i1 [[BOUND0]], [[BOUND1]]
; CHECK-NEXT:    br i1 [[FOUND_CONFLICT]], label [[VEC_EPILOG_SCALAR_PH]], label [[VECTOR_MAIN_LOOP_ITER_CHECK:%.*]]
; CHECK:       vector.main.loop.iter.check:
; CHECK-NEXT:    [[MIN_ITERS_CHECK3:%.*]] = icmp ult i64 [[SMAX2]], 16
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK3]], label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[N_VEC:%.*]] = and i64 [[SMAX2]], 9223372036854775792
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <16 x i32> poison, i32 [[K:%.*]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <16 x i32> [[BROADCAST_SPLATINSERT]], <16 x i32> poison, <16 x i32> zeroinitializer
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT4:%.*]] = insertelement <16 x i32> poison, i32 [[NTRUNC]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT5:%.*]] = shufflevector <16 x i32> [[BROADCAST_SPLATINSERT4]], <16 x i32> poison, <16 x i32> zeroinitializer
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT6:%.*]] = insertelement <16 x ptr> poison, ptr [[A]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT7:%.*]] = shufflevector <16 x ptr> [[BROADCAST_SPLATINSERT6]], <16 x ptr> poison, <16 x i32> zeroinitializer
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <16 x i32>, ptr [[TMP1]], align 8, !alias.scope !15, !noalias !18
; CHECK-NEXT:    [[TMP2:%.*]] = icmp eq <16 x i32> [[WIDE_LOAD]], [[BROADCAST_SPLAT]]
; CHECK-NEXT:    store <16 x i32> [[BROADCAST_SPLAT5]], ptr [[TMP1]], align 4, !alias.scope !15, !noalias !18
; CHECK-NEXT:    call void @llvm.masked.scatter.v16i32.v16p0(<16 x i32> [[BROADCAST_SPLAT5]], <16 x ptr> [[BROADCAST_SPLAT7]], i32 4, <16 x i1> [[TMP2]]), !alias.scope !18
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16
; CHECK-NEXT:    [[TMP3:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP3]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP20:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[SMAX2]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label [[FOR_END:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]]
; CHECK:       vec.epilog.iter.check:
; CHECK-NEXT:    [[N_VEC_REMAINING:%.*]] = and i64 [[SMAX2]], 8
; CHECK-NEXT:    [[MIN_EPILOG_ITERS_CHECK_NOT_NOT:%.*]] = icmp eq i64 [[N_VEC_REMAINING]], 0
; CHECK-NEXT:    br i1 [[MIN_EPILOG_ITERS_CHECK_NOT_NOT]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]]
; CHECK:       vec.epilog.ph:
; CHECK-NEXT:    [[VEC_EPILOG_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ]
; CHECK-NEXT:    [[N_VEC9:%.*]] = and i64 [[SMAX2]], 9223372036854775800
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT13:%.*]] = insertelement <8 x i32> poison, i32 [[K]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT14:%.*]] = shufflevector <8 x i32> [[BROADCAST_SPLATINSERT13]], <8 x i32> poison, <8 x i32> zeroinitializer
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT15:%.*]] = insertelement <8 x i32> poison, i32 [[NTRUNC]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT16:%.*]] = shufflevector <8 x i32> [[BROADCAST_SPLATINSERT15]], <8 x i32> poison, <8 x i32> zeroinitializer
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT17:%.*]] = insertelement <8 x ptr> poison, ptr [[A]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT18:%.*]] = shufflevector <8 x ptr> [[BROADCAST_SPLATINSERT17]], <8 x ptr> poison, <8 x i32> zeroinitializer
; CHECK-NEXT:    br label [[VEC_EPILOG_VECTOR_BODY:%.*]]
; CHECK:       vec.epilog.vector.body:
; CHECK-NEXT:    [[INDEX11:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT19:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[INDEX11]]
; CHECK-NEXT:    [[WIDE_LOAD12:%.*]] = load <8 x i32>, ptr [[TMP4]], align 8, !alias.scope !21, !noalias !24
; CHECK-NEXT:    [[TMP5:%.*]] = icmp eq <8 x i32> [[WIDE_LOAD12]], [[BROADCAST_SPLAT14]]
; CHECK-NEXT:    store <8 x i32> [[BROADCAST_SPLAT16]], ptr [[TMP4]], align 4, !alias.scope !21, !noalias !24
; CHECK-NEXT:    call void @llvm.masked.scatter.v8i32.v8p0(<8 x i32> [[BROADCAST_SPLAT16]], <8 x ptr> [[BROADCAST_SPLAT18]], i32 4, <8 x i1> [[TMP5]]), !alias.scope !24
; CHECK-NEXT:    [[INDEX_NEXT19]] = add nuw i64 [[INDEX11]], 8
; CHECK-NEXT:    [[TMP6:%.*]] = icmp eq i64 [[INDEX_NEXT19]], [[N_VEC9]]
; CHECK-NEXT:    br i1 [[TMP6]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP26:![0-9]+]]
; CHECK:       vec.epilog.middle.block:
; CHECK-NEXT:    [[CMP_N10:%.*]] = icmp eq i64 [[SMAX2]], [[N_VEC9]]
; CHECK-NEXT:    br i1 [[CMP_N10]], label [[FOR_END]], label [[VEC_EPILOG_SCALAR_PH]]
; CHECK:       vec.epilog.scalar.ph:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC9]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MEMCHECK]] ], [ 0, [[ITER_CHECK:%.*]] ]
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.body:
; CHECK-NEXT:    [[I:%.*]] = phi i64 [ [[I_NEXT:%.*]], [[LATCH:%.*]] ], [ [[BC_RESUME_VAL]], [[VEC_EPILOG_SCALAR_PH]] ]
; CHECK-NEXT:    [[T1:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[I]]
; CHECK-NEXT:    [[T2:%.*]] = load i32, ptr [[T1]], align 8
; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[T2]], [[K]]
; CHECK-NEXT:    store i32 [[NTRUNC]], ptr [[T1]], align 4
; CHECK-NEXT:    br i1 [[CMP]], label [[COND_STORE:%.*]], label [[LATCH]]
; CHECK:       cond_store:
; CHECK-NEXT:    store i32 [[NTRUNC]], ptr [[A]], align 4
; CHECK-NEXT:    br label [[LATCH]]
; CHECK:       latch:
; CHECK-NEXT:    [[I_NEXT]] = add nuw nsw i64 [[I]], 1
; CHECK-NEXT:    [[COND:%.*]] = icmp slt i64 [[I_NEXT]], [[N]]
; CHECK-NEXT:    br i1 [[COND]], label [[FOR_BODY]], label [[FOR_END]], !llvm.loop [[LOOP27:![0-9]+]]
; CHECK:       for.end:
; CHECK-NEXT:    ret void
;
entry:
  %ntrunc = trunc i64 %n to i32
  br label %for.body

for.body:                                         ; preds = %for.body, %entry
  %i = phi i64 [ %i.next, %latch ], [ 0, %entry ]
  %t1 = getelementptr inbounds i32, ptr %b, i64 %i
  %t2 = load i32, ptr %t1, align 8
  %cmp = icmp eq i32 %t2, %k
  store i32 %ntrunc, ptr %t1
  br i1 %cmp, label %cond_store, label %latch

cond_store:
  store i32 %ntrunc, ptr %a
  br label %latch

latch:
  %i.next = add nuw nsw i64 %i, 1
  %cond = icmp slt i64 %i.next, %n
  br i1 %cond, label %for.body, label %for.end

for.end:                                          ; preds = %for.body
  ret void
}

define void @variant_val_store_to_inv_address_conditional(ptr %a, i64 %n, ptr %b, ptr %c, i32 %k) {
; CHECK-LABEL: @variant_val_store_to_inv_address_conditional(
; CHECK-NEXT:  iter.check:
; CHECK-NEXT:    [[NTRUNC:%.*]] = trunc i64 [[N:%.*]] to i32
; CHECK-NEXT:    [[SMAX10:%.*]] = call i64 @llvm.smax.i64(i64 [[N]], i64 1)
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[SMAX10]], 8
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH:%.*]], label [[VECTOR_MEMCHECK:%.*]]
; CHECK:       vector.memcheck:
; CHECK-NEXT:    [[SMAX:%.*]] = call i64 @llvm.smax.i64(i64 [[N]], i64 1)
; CHECK-NEXT:    [[TMP0:%.*]] = shl i64 [[SMAX]], 2
; CHECK-NEXT:    [[UGLYGEP:%.*]] = getelementptr i8, ptr [[B:%.*]], i64 [[TMP0]]
; CHECK-NEXT:    [[UGLYGEP1:%.*]] = getelementptr i8, ptr [[A:%.*]], i64 4
; CHECK-NEXT:    [[UGLYGEP2:%.*]] = getelementptr i8, ptr [[C:%.*]], i64 [[TMP0]]
; CHECK-NEXT:    [[BOUND0:%.*]] = icmp ugt ptr [[UGLYGEP1]], [[B]]
; CHECK-NEXT:    [[BOUND1:%.*]] = icmp ugt ptr [[UGLYGEP]], [[A]]
; CHECK-NEXT:    [[FOUND_CONFLICT:%.*]] = and i1 [[BOUND0]], [[BOUND1]]
; CHECK-NEXT:    [[BOUND03:%.*]] = icmp ugt ptr [[UGLYGEP2]], [[B]]
; CHECK-NEXT:    [[BOUND14:%.*]] = icmp ugt ptr [[UGLYGEP]], [[C]]
; CHECK-NEXT:    [[FOUND_CONFLICT5:%.*]] = and i1 [[BOUND03]], [[BOUND14]]
; CHECK-NEXT:    [[CONFLICT_RDX:%.*]] = or i1 [[FOUND_CONFLICT]], [[FOUND_CONFLICT5]]
; CHECK-NEXT:    [[BOUND06:%.*]] = icmp ugt ptr [[UGLYGEP2]], [[A]]
; CHECK-NEXT:    [[BOUND17:%.*]] = icmp ugt ptr [[UGLYGEP1]], [[C]]
; CHECK-NEXT:    [[FOUND_CONFLICT8:%.*]] = and i1 [[BOUND06]], [[BOUND17]]
; CHECK-NEXT:    [[CONFLICT_RDX9:%.*]] = or i1 [[CONFLICT_RDX]], [[FOUND_CONFLICT8]]
; CHECK-NEXT:    br i1 [[CONFLICT_RDX9]], label [[VEC_EPILOG_SCALAR_PH]], label [[VECTOR_MAIN_LOOP_ITER_CHECK:%.*]]
; CHECK:       vector.main.loop.iter.check:
; CHECK-NEXT:    [[MIN_ITERS_CHECK11:%.*]] = icmp ult i64 [[SMAX10]], 16
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK11]], label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[N_VEC:%.*]] = and i64 [[SMAX10]], 9223372036854775792
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <16 x i32> poison, i32 [[K:%.*]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <16 x i32> [[BROADCAST_SPLATINSERT]], <16 x i32> poison, <16 x i32> zeroinitializer
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT12:%.*]] = insertelement <16 x i32> poison, i32 [[NTRUNC]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT13:%.*]] = shufflevector <16 x i32> [[BROADCAST_SPLATINSERT12]], <16 x i32> poison, <16 x i32> zeroinitializer
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT14:%.*]] = insertelement <16 x ptr> poison, ptr [[A]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT15:%.*]] = shufflevector <16 x ptr> [[BROADCAST_SPLATINSERT14]], <16 x ptr> poison, <16 x i32> zeroinitializer
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <16 x i32>, ptr [[TMP1]], align 8, !alias.scope !28, !noalias !31
; CHECK-NEXT:    [[TMP2:%.*]] = icmp eq <16 x i32> [[WIDE_LOAD]], [[BROADCAST_SPLAT]]
; CHECK-NEXT:    store <16 x i32> [[BROADCAST_SPLAT13]], ptr [[TMP1]], align 4, !alias.scope !28, !noalias !31
; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr i32, ptr [[C]], i64 [[INDEX]]
; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <16 x i32> @llvm.masked.load.v16i32.p0(ptr [[TMP3]], i32 8, <16 x i1> [[TMP2]], <16 x i32> poison), !alias.scope !34
; CHECK-NEXT:    call void @llvm.masked.scatter.v16i32.v16p0(<16 x i32> [[WIDE_MASKED_LOAD]], <16 x ptr> [[BROADCAST_SPLAT15]], i32 4, <16 x i1> [[TMP2]]), !alias.scope !35, !noalias !34
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16
; CHECK-NEXT:    [[TMP4:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP4]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP36:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[SMAX10]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label [[FOR_END:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]]
; CHECK:       vec.epilog.iter.check:
; CHECK-NEXT:    [[N_VEC_REMAINING:%.*]] = and i64 [[SMAX10]], 8
; CHECK-NEXT:    [[MIN_EPILOG_ITERS_CHECK_NOT_NOT:%.*]] = icmp eq i64 [[N_VEC_REMAINING]], 0
; CHECK-NEXT:    br i1 [[MIN_EPILOG_ITERS_CHECK_NOT_NOT]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]]
; CHECK:       vec.epilog.ph:
; CHECK-NEXT:    [[VEC_EPILOG_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ]
; CHECK-NEXT:    [[N_VEC17:%.*]] = and i64 [[SMAX10]], 9223372036854775800
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT21:%.*]] = insertelement <8 x i32> poison, i32 [[K]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT22:%.*]] = shufflevector <8 x i32> [[BROADCAST_SPLATINSERT21]], <8 x i32> poison, <8 x i32> zeroinitializer
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT23:%.*]] = insertelement <8 x i32> poison, i32 [[NTRUNC]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT24:%.*]] = shufflevector <8 x i32> [[BROADCAST_SPLATINSERT23]], <8 x i32> poison, <8 x i32> zeroinitializer
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT26:%.*]] = insertelement <8 x ptr> poison, ptr [[A]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT27:%.*]] = shufflevector <8 x ptr> [[BROADCAST_SPLATINSERT26]], <8 x ptr> poison, <8 x i32> zeroinitializer
; CHECK-NEXT:    br label [[VEC_EPILOG_VECTOR_BODY:%.*]]
; CHECK:       vec.epilog.vector.body:
; CHECK-NEXT:    [[INDEX19:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT28:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[INDEX19]]
; CHECK-NEXT:    [[WIDE_LOAD20:%.*]] = load <8 x i32>, ptr [[TMP5]], align 8, !alias.scope !37, !noalias !40
; CHECK-NEXT:    [[TMP6:%.*]] = icmp eq <8 x i32> [[WIDE_LOAD20]], [[BROADCAST_SPLAT22]]
; CHECK-NEXT:    store <8 x i32> [[BROADCAST_SPLAT24]], ptr [[TMP5]], align 4, !alias.scope !37, !noalias !40
; CHECK-NEXT:    [[TMP7:%.*]] = getelementptr i32, ptr [[C]], i64 [[INDEX19]]
; CHECK-NEXT:    [[WIDE_MASKED_LOAD25:%.*]] = call <8 x i32> @llvm.masked.load.v8i32.p0(ptr [[TMP7]], i32 8, <8 x i1> [[TMP6]], <8 x i32> poison), !alias.scope !43
; CHECK-NEXT:    call void @llvm.masked.scatter.v8i32.v8p0(<8 x i32> [[WIDE_MASKED_LOAD25]], <8 x ptr> [[BROADCAST_SPLAT27]], i32 4, <8 x i1> [[TMP6]]), !alias.scope !44, !noalias !43
; CHECK-NEXT:    [[INDEX_NEXT28]] = add nuw i64 [[INDEX19]], 8
; CHECK-NEXT:    [[TMP8:%.*]] = icmp eq i64 [[INDEX_NEXT28]], [[N_VEC17]]
; CHECK-NEXT:    br i1 [[TMP8]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP45:![0-9]+]]
; CHECK:       vec.epilog.middle.block:
; CHECK-NEXT:    [[CMP_N18:%.*]] = icmp eq i64 [[SMAX10]], [[N_VEC17]]
; CHECK-NEXT:    br i1 [[CMP_N18]], label [[FOR_END]], label [[VEC_EPILOG_SCALAR_PH]]
; CHECK:       vec.epilog.scalar.ph:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC17]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MEMCHECK]] ], [ 0, [[ITER_CHECK:%.*]] ]
; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
; CHECK:       for.body:
; CHECK-NEXT:    [[I:%.*]] = phi i64 [ [[I_NEXT:%.*]], [[LATCH:%.*]] ], [ [[BC_RESUME_VAL]], [[VEC_EPILOG_SCALAR_PH]] ]
; CHECK-NEXT:    [[T1:%.*]] = getelementptr inbounds i32, ptr [[B]], i64 [[I]]
; CHECK-NEXT:    [[T2:%.*]] = load i32, ptr [[T1]], align 8
; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[T2]], [[K]]
; CHECK-NEXT:    store i32 [[NTRUNC]], ptr [[T1]], align 4
; CHECK-NEXT:    br i1 [[CMP]], label [[COND_STORE:%.*]], label [[LATCH]]
; CHECK:       cond_store:
; CHECK-NEXT:    [[T3:%.*]] = getelementptr inbounds i32, ptr [[C]], i64 [[I]]
; CHECK-NEXT:    [[T4:%.*]] = load i32, ptr [[T3]], align 8
; CHECK-NEXT:    store i32 [[T4]], ptr [[A]], align 4
; CHECK-NEXT:    br label [[LATCH]]
; CHECK:       latch:
; CHECK-NEXT:    [[I_NEXT]] = add nuw nsw i64 [[I]], 1
; CHECK-NEXT:    [[COND:%.*]] = icmp slt i64 [[I_NEXT]], [[N]]
; CHECK-NEXT:    br i1 [[COND]], label [[FOR_BODY]], label [[FOR_END]], !llvm.loop [[LOOP46:![0-9]+]]
; CHECK:       for.end:
; CHECK-NEXT:    ret void
;
entry:
  %ntrunc = trunc i64 %n to i32
  br label %for.body

for.body:                                         ; preds = %for.body, %entry
  %i = phi i64 [ %i.next, %latch ], [ 0, %entry ]
  %t1 = getelementptr inbounds i32, ptr %b, i64 %i
  %t2 = load i32, ptr %t1, align 8
  %cmp = icmp eq i32 %t2, %k
  store i32 %ntrunc, ptr %t1
  br i1 %cmp, label %cond_store, label %latch

cond_store:
  %t3 = getelementptr inbounds i32, ptr %c, i64 %i
  %t4 = load i32, ptr %t3, align 8
  store i32 %t4, ptr %a
  br label %latch

latch:
  %i.next = add nuw nsw i64 %i, 1
  %cond = icmp slt i64 %i.next, %n
  br i1 %cond, label %for.body, label %for.end

for.end:                                          ; preds = %for.body
  ret void
}