llvm/test/Transforms/LoopVectorize/RISCV/interleaved-accesses.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -passes=loop-vectorize -mtriple=riscv64 -mattr=+v -S | FileCheck %s
; RUN: opt < %s -passes=loop-vectorize -scalable-vectorization=off -mtriple=riscv64 -mattr=+v -S | FileCheck %s --check-prefix=FIXED
; RUN: opt < %s -passes=loop-vectorize -scalable-vectorization=on -mtriple=riscv64 -mattr=+v -S | FileCheck %s --check-prefix=SCALABLE

define void @load_store_factor2_i32(ptr %p) {
; CHECK-LABEL: @load_store_factor2_i32(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1024, [[TMP1]]
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP3]]
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
; CHECK-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 4
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP6:%.*]] = add i64 [[INDEX]], 0
; CHECK-NEXT:    [[TMP7:%.*]] = shl i64 [[TMP6]], 1
; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP7]]
; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr i32, ptr [[TMP8]], i32 0
; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP9]], align 4
; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
; CHECK-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
; CHECK-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP12:%.*]] = add <vscale x 4 x i32> [[TMP10]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP13:%.*]] = add i64 [[TMP7]], 1
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr i32, ptr [[P]], i64 [[TMP13]]
; CHECK-NEXT:    [[TMP15:%.*]] = add <vscale x 4 x i32> [[TMP11]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 2, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP16:%.*]] = getelementptr i32, ptr [[TMP14]], i32 -1
; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> [[TMP12]], <vscale x 4 x i32> [[TMP15]])
; CHECK-NEXT:    store <vscale x 8 x i32> [[INTERLEAVED_VEC]], ptr [[TMP16]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; CHECK-NEXT:    [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
; CHECK:       scalar.ph:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; CHECK-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 1
; CHECK-NEXT:    [[Q0:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET0]]
; CHECK-NEXT:    [[X0:%.*]] = load i32, ptr [[Q0]], align 4
; CHECK-NEXT:    [[Y0:%.*]] = add i32 [[X0]], 1
; CHECK-NEXT:    store i32 [[Y0]], ptr [[Q0]], align 4
; CHECK-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; CHECK-NEXT:    [[Q1:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET1]]
; CHECK-NEXT:    [[X1:%.*]] = load i32, ptr [[Q1]], align 4
; CHECK-NEXT:    [[Y1:%.*]] = add i32 [[X1]], 2
; CHECK-NEXT:    store i32 [[Y1]], ptr [[Q1]], align 4
; CHECK-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; CHECK-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; CHECK-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @load_store_factor2_i32(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; FIXED:       vector.ph:
; FIXED-NEXT:    br label [[VECTOR_BODY:%.*]]
; FIXED:       vector.body:
; FIXED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; FIXED-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
; FIXED-NEXT:    [[TMP1:%.*]] = shl i64 [[TMP0]], 1
; FIXED-NEXT:    [[TMP2:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP1]]
; FIXED-NEXT:    [[TMP3:%.*]] = getelementptr i32, ptr [[TMP2]], i32 0
; FIXED-NEXT:    [[WIDE_VEC:%.*]] = load <16 x i32>, ptr [[TMP3]], align 4
; FIXED-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <16 x i32> [[WIDE_VEC]], <16 x i32> poison, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
; FIXED-NEXT:    [[STRIDED_VEC1:%.*]] = shufflevector <16 x i32> [[WIDE_VEC]], <16 x i32> poison, <8 x i32> <i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15>
; FIXED-NEXT:    [[TMP4:%.*]] = add <8 x i32> [[STRIDED_VEC]], <i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1, i32 1>
; FIXED-NEXT:    [[TMP5:%.*]] = add i64 [[TMP1]], 1
; FIXED-NEXT:    [[TMP6:%.*]] = getelementptr i32, ptr [[P]], i64 [[TMP5]]
; FIXED-NEXT:    [[TMP7:%.*]] = add <8 x i32> [[STRIDED_VEC1]], <i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2, i32 2>
; FIXED-NEXT:    [[TMP8:%.*]] = getelementptr i32, ptr [[TMP6]], i32 -1
; FIXED-NEXT:    [[TMP9:%.*]] = shufflevector <8 x i32> [[TMP4]], <8 x i32> [[TMP7]], <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
; FIXED-NEXT:    [[INTERLEAVED_VEC:%.*]] = shufflevector <16 x i32> [[TMP9]], <16 x i32> poison, <16 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11, i32 4, i32 12, i32 5, i32 13, i32 6, i32 14, i32 7, i32 15>
; FIXED-NEXT:    store <16 x i32> [[INTERLEAVED_VEC]], ptr [[TMP8]], align 4
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
; FIXED-NEXT:    [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
; FIXED:       scalar.ph:
; FIXED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1024, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       loop:
; FIXED-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; FIXED-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 1
; FIXED-NEXT:    [[Q0:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET0]]
; FIXED-NEXT:    [[X0:%.*]] = load i32, ptr [[Q0]], align 4
; FIXED-NEXT:    [[Y0:%.*]] = add i32 [[X0]], 1
; FIXED-NEXT:    store i32 [[Y0]], ptr [[Q0]], align 4
; FIXED-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; FIXED-NEXT:    [[Q1:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET1]]
; FIXED-NEXT:    [[X1:%.*]] = load i32, ptr [[Q1]], align 4
; FIXED-NEXT:    [[Y1:%.*]] = add i32 [[X1]], 2
; FIXED-NEXT:    store i32 [[Y1]], ptr [[Q1]], align 4
; FIXED-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; FIXED-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; FIXED-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @load_store_factor2_i32(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
; SCALABLE-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1024, [[TMP1]]
; SCALABLE-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; SCALABLE:       vector.ph:
; SCALABLE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
; SCALABLE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP3]]
; SCALABLE-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
; SCALABLE-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 4
; SCALABLE-NEXT:    br label [[VECTOR_BODY:%.*]]
; SCALABLE:       vector.body:
; SCALABLE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP6:%.*]] = add i64 [[INDEX]], 0
; SCALABLE-NEXT:    [[TMP7:%.*]] = shl i64 [[TMP6]], 1
; SCALABLE-NEXT:    [[TMP8:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP7]]
; SCALABLE-NEXT:    [[TMP9:%.*]] = getelementptr i32, ptr [[TMP8]], i32 0
; SCALABLE-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP9]], align 4
; SCALABLE-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
; SCALABLE-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
; SCALABLE-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP12:%.*]] = add <vscale x 4 x i32> [[TMP10]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP13:%.*]] = add i64 [[TMP7]], 1
; SCALABLE-NEXT:    [[TMP14:%.*]] = getelementptr i32, ptr [[P]], i64 [[TMP13]]
; SCALABLE-NEXT:    [[TMP15:%.*]] = add <vscale x 4 x i32> [[TMP11]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 2, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP16:%.*]] = getelementptr i32, ptr [[TMP14]], i32 -1
; SCALABLE-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> [[TMP12]], <vscale x 4 x i32> [[TMP15]])
; SCALABLE-NEXT:    store <vscale x 8 x i32> [[INTERLEAVED_VEC]], ptr [[TMP16]], align 4
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; SCALABLE-NEXT:    [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; SCALABLE-NEXT:    br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
; SCALABLE-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
; SCALABLE:       scalar.ph:
; SCALABLE-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; SCALABLE:       loop:
; SCALABLE-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; SCALABLE-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 1
; SCALABLE-NEXT:    [[Q0:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET0]]
; SCALABLE-NEXT:    [[X0:%.*]] = load i32, ptr [[Q0]], align 4
; SCALABLE-NEXT:    [[Y0:%.*]] = add i32 [[X0]], 1
; SCALABLE-NEXT:    store i32 [[Y0]], ptr [[Q0]], align 4
; SCALABLE-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; SCALABLE-NEXT:    [[Q1:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET1]]
; SCALABLE-NEXT:    [[X1:%.*]] = load i32, ptr [[Q1]], align 4
; SCALABLE-NEXT:    [[Y1:%.*]] = add i32 [[X1]], 2
; SCALABLE-NEXT:    store i32 [[Y1]], ptr [[Q1]], align 4
; SCALABLE-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; SCALABLE-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; SCALABLE-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
; SCALABLE:       exit:
; SCALABLE-NEXT:    ret void
;
entry:
  br label %loop
loop:
  %i = phi i64 [0, %entry], [%nexti, %loop]

  %offset0 = shl i64 %i, 1
  %q0 = getelementptr i32, ptr %p, i64 %offset0
  %x0 = load i32, ptr %q0
  %y0 = add i32 %x0, 1
  store i32 %y0, ptr %q0

  %offset1 = add i64 %offset0, 1
  %q1 = getelementptr i32, ptr %p, i64 %offset1
  %x1 = load i32, ptr %q1
  %y1 = add i32 %x1, 2
  store i32 %y1, ptr %q1

  %nexti = add i64 %i, 1
  %done = icmp eq i64 %nexti, 1024
  br i1 %done, label %exit, label %loop
exit:
  ret void
}

define void @load_store_factor2_i64(ptr %p) {
; CHECK-LABEL: @load_store_factor2_i64(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1024, [[TMP1]]
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP3]]
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
; CHECK-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP6:%.*]] = add i64 [[INDEX]], 0
; CHECK-NEXT:    [[TMP7:%.*]] = shl i64 [[TMP6]], 1
; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP7]]
; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr i64, ptr [[TMP8]], i32 0
; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 4 x i64>, ptr [[TMP9]], align 8
; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> [[WIDE_VEC]])
; CHECK-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 0
; CHECK-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP12:%.*]] = add <vscale x 2 x i64> [[TMP10]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP13:%.*]] = add i64 [[TMP7]], 1
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr i64, ptr [[P]], i64 [[TMP13]]
; CHECK-NEXT:    [[TMP15:%.*]] = add <vscale x 2 x i64> [[TMP11]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 2, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP16:%.*]] = getelementptr i64, ptr [[TMP14]], i32 -1
; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64> [[TMP12]], <vscale x 2 x i64> [[TMP15]])
; CHECK-NEXT:    store <vscale x 4 x i64> [[INTERLEAVED_VEC]], ptr [[TMP16]], align 8
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; CHECK-NEXT:    [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
; CHECK:       scalar.ph:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; CHECK-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 1
; CHECK-NEXT:    [[Q0:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET0]]
; CHECK-NEXT:    [[X0:%.*]] = load i64, ptr [[Q0]], align 8
; CHECK-NEXT:    [[Y0:%.*]] = add i64 [[X0]], 1
; CHECK-NEXT:    store i64 [[Y0]], ptr [[Q0]], align 8
; CHECK-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; CHECK-NEXT:    [[Q1:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET1]]
; CHECK-NEXT:    [[X1:%.*]] = load i64, ptr [[Q1]], align 8
; CHECK-NEXT:    [[Y1:%.*]] = add i64 [[X1]], 2
; CHECK-NEXT:    store i64 [[Y1]], ptr [[Q1]], align 8
; CHECK-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; CHECK-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; CHECK-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @load_store_factor2_i64(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; FIXED:       vector.ph:
; FIXED-NEXT:    br label [[VECTOR_BODY:%.*]]
; FIXED:       vector.body:
; FIXED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; FIXED-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
; FIXED-NEXT:    [[TMP1:%.*]] = shl i64 [[TMP0]], 1
; FIXED-NEXT:    [[TMP2:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP1]]
; FIXED-NEXT:    [[TMP3:%.*]] = getelementptr i64, ptr [[TMP2]], i32 0
; FIXED-NEXT:    [[WIDE_VEC:%.*]] = load <8 x i64>, ptr [[TMP3]], align 8
; FIXED-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <8 x i64> [[WIDE_VEC]], <8 x i64> poison, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
; FIXED-NEXT:    [[STRIDED_VEC1:%.*]] = shufflevector <8 x i64> [[WIDE_VEC]], <8 x i64> poison, <4 x i32> <i32 1, i32 3, i32 5, i32 7>
; FIXED-NEXT:    [[TMP4:%.*]] = add <4 x i64> [[STRIDED_VEC]], <i64 1, i64 1, i64 1, i64 1>
; FIXED-NEXT:    [[TMP5:%.*]] = add i64 [[TMP1]], 1
; FIXED-NEXT:    [[TMP6:%.*]] = getelementptr i64, ptr [[P]], i64 [[TMP5]]
; FIXED-NEXT:    [[TMP7:%.*]] = add <4 x i64> [[STRIDED_VEC1]], <i64 2, i64 2, i64 2, i64 2>
; FIXED-NEXT:    [[TMP8:%.*]] = getelementptr i64, ptr [[TMP6]], i32 -1
; FIXED-NEXT:    [[TMP9:%.*]] = shufflevector <4 x i64> [[TMP4]], <4 x i64> [[TMP7]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
; FIXED-NEXT:    [[INTERLEAVED_VEC:%.*]] = shufflevector <8 x i64> [[TMP9]], <8 x i64> poison, <8 x i32> <i32 0, i32 4, i32 1, i32 5, i32 2, i32 6, i32 3, i32 7>
; FIXED-NEXT:    store <8 x i64> [[INTERLEAVED_VEC]], ptr [[TMP8]], align 8
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; FIXED-NEXT:    [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
; FIXED:       scalar.ph:
; FIXED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1024, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       loop:
; FIXED-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; FIXED-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 1
; FIXED-NEXT:    [[Q0:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET0]]
; FIXED-NEXT:    [[X0:%.*]] = load i64, ptr [[Q0]], align 8
; FIXED-NEXT:    [[Y0:%.*]] = add i64 [[X0]], 1
; FIXED-NEXT:    store i64 [[Y0]], ptr [[Q0]], align 8
; FIXED-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; FIXED-NEXT:    [[Q1:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET1]]
; FIXED-NEXT:    [[X1:%.*]] = load i64, ptr [[Q1]], align 8
; FIXED-NEXT:    [[Y1:%.*]] = add i64 [[X1]], 2
; FIXED-NEXT:    store i64 [[Y1]], ptr [[Q1]], align 8
; FIXED-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; FIXED-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; FIXED-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @load_store_factor2_i64(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; SCALABLE-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1024, [[TMP1]]
; SCALABLE-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; SCALABLE:       vector.ph:
; SCALABLE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; SCALABLE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP3]]
; SCALABLE-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
; SCALABLE-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; SCALABLE-NEXT:    br label [[VECTOR_BODY:%.*]]
; SCALABLE:       vector.body:
; SCALABLE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP6:%.*]] = add i64 [[INDEX]], 0
; SCALABLE-NEXT:    [[TMP7:%.*]] = shl i64 [[TMP6]], 1
; SCALABLE-NEXT:    [[TMP8:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP7]]
; SCALABLE-NEXT:    [[TMP9:%.*]] = getelementptr i64, ptr [[TMP8]], i32 0
; SCALABLE-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 4 x i64>, ptr [[TMP9]], align 8
; SCALABLE-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> [[WIDE_VEC]])
; SCALABLE-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 0
; SCALABLE-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP12:%.*]] = add <vscale x 2 x i64> [[TMP10]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP13:%.*]] = add i64 [[TMP7]], 1
; SCALABLE-NEXT:    [[TMP14:%.*]] = getelementptr i64, ptr [[P]], i64 [[TMP13]]
; SCALABLE-NEXT:    [[TMP15:%.*]] = add <vscale x 2 x i64> [[TMP11]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 2, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP16:%.*]] = getelementptr i64, ptr [[TMP14]], i32 -1
; SCALABLE-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64> [[TMP12]], <vscale x 2 x i64> [[TMP15]])
; SCALABLE-NEXT:    store <vscale x 4 x i64> [[INTERLEAVED_VEC]], ptr [[TMP16]], align 8
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; SCALABLE-NEXT:    [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; SCALABLE-NEXT:    br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
; SCALABLE-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
; SCALABLE:       scalar.ph:
; SCALABLE-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; SCALABLE:       loop:
; SCALABLE-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; SCALABLE-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 1
; SCALABLE-NEXT:    [[Q0:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET0]]
; SCALABLE-NEXT:    [[X0:%.*]] = load i64, ptr [[Q0]], align 8
; SCALABLE-NEXT:    [[Y0:%.*]] = add i64 [[X0]], 1
; SCALABLE-NEXT:    store i64 [[Y0]], ptr [[Q0]], align 8
; SCALABLE-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; SCALABLE-NEXT:    [[Q1:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET1]]
; SCALABLE-NEXT:    [[X1:%.*]] = load i64, ptr [[Q1]], align 8
; SCALABLE-NEXT:    [[Y1:%.*]] = add i64 [[X1]], 2
; SCALABLE-NEXT:    store i64 [[Y1]], ptr [[Q1]], align 8
; SCALABLE-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; SCALABLE-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; SCALABLE-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
; SCALABLE:       exit:
; SCALABLE-NEXT:    ret void
;
entry:
  br label %loop
loop:
  %i = phi i64 [0, %entry], [%nexti, %loop]

  %offset0 = shl i64 %i, 1
  %q0 = getelementptr i64, ptr %p, i64 %offset0
  %x0 = load i64, ptr %q0
  %y0 = add i64 %x0, 1
  store i64 %y0, ptr %q0

  %offset1 = add i64 %offset0, 1
  %q1 = getelementptr i64, ptr %p, i64 %offset1
  %x1 = load i64, ptr %q1
  %y1 = add i64 %x1, 2
  store i64 %y1, ptr %q1

  %nexti = add i64 %i, 1
  %done = icmp eq i64 %nexti, 1024
  br i1 %done, label %exit, label %loop
exit:
  ret void
}

define void @load_store_factor3_i32(ptr %p) {
; CHECK-LABEL: @load_store_factor3_i32(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 3
; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP1]]
; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr i32, ptr [[TMP2]], i32 0
; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <6 x i32>, ptr [[TMP3]], align 4
; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <6 x i32> [[WIDE_VEC]], <6 x i32> poison, <2 x i32> <i32 0, i32 3>
; CHECK-NEXT:    [[STRIDED_VEC1:%.*]] = shufflevector <6 x i32> [[WIDE_VEC]], <6 x i32> poison, <2 x i32> <i32 1, i32 4>
; CHECK-NEXT:    [[STRIDED_VEC2:%.*]] = shufflevector <6 x i32> [[WIDE_VEC]], <6 x i32> poison, <2 x i32> <i32 2, i32 5>
; CHECK-NEXT:    [[TMP4:%.*]] = add <2 x i32> [[STRIDED_VEC]], <i32 1, i32 1>
; CHECK-NEXT:    [[TMP5:%.*]] = add i64 [[TMP1]], 1
; CHECK-NEXT:    [[TMP6:%.*]] = add <2 x i32> [[STRIDED_VEC1]], <i32 2, i32 2>
; CHECK-NEXT:    [[TMP7:%.*]] = add i64 [[TMP5]], 1
; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr i32, ptr [[P]], i64 [[TMP7]]
; CHECK-NEXT:    [[TMP9:%.*]] = add <2 x i32> [[STRIDED_VEC2]], <i32 3, i32 3>
; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr i32, ptr [[TMP8]], i32 -2
; CHECK-NEXT:    [[TMP11:%.*]] = shufflevector <2 x i32> [[TMP4]], <2 x i32> [[TMP6]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; CHECK-NEXT:    [[TMP12:%.*]] = shufflevector <2 x i32> [[TMP9]], <2 x i32> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
; CHECK-NEXT:    [[TMP13:%.*]] = shufflevector <4 x i32> [[TMP11]], <4 x i32> [[TMP12]], <6 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5>
; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = shufflevector <6 x i32> [[TMP13]], <6 x i32> poison, <6 x i32> <i32 0, i32 2, i32 4, i32 1, i32 3, i32 5>
; CHECK-NEXT:    store <6 x i32> [[INTERLEAVED_VEC]], ptr [[TMP10]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; CHECK-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; CHECK-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
; CHECK:       scalar.ph:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1024, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; CHECK-NEXT:    [[OFFSET0:%.*]] = mul i64 [[I]], 3
; CHECK-NEXT:    [[Q0:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET0]]
; CHECK-NEXT:    [[X0:%.*]] = load i32, ptr [[Q0]], align 4
; CHECK-NEXT:    [[Y0:%.*]] = add i32 [[X0]], 1
; CHECK-NEXT:    store i32 [[Y0]], ptr [[Q0]], align 4
; CHECK-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; CHECK-NEXT:    [[Q1:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET1]]
; CHECK-NEXT:    [[X1:%.*]] = load i32, ptr [[Q1]], align 4
; CHECK-NEXT:    [[Y1:%.*]] = add i32 [[X1]], 2
; CHECK-NEXT:    store i32 [[Y1]], ptr [[Q1]], align 4
; CHECK-NEXT:    [[OFFSET2:%.*]] = add i64 [[OFFSET1]], 1
; CHECK-NEXT:    [[Q2:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET2]]
; CHECK-NEXT:    [[X2:%.*]] = load i32, ptr [[Q2]], align 4
; CHECK-NEXT:    [[Y2:%.*]] = add i32 [[X2]], 3
; CHECK-NEXT:    store i32 [[Y2]], ptr [[Q2]], align 4
; CHECK-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; CHECK-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; CHECK-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP7:![0-9]+]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @load_store_factor3_i32(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; FIXED:       vector.ph:
; FIXED-NEXT:    br label [[VECTOR_BODY:%.*]]
; FIXED:       vector.body:
; FIXED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; FIXED-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
; FIXED-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 3
; FIXED-NEXT:    [[TMP2:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP1]]
; FIXED-NEXT:    [[TMP3:%.*]] = getelementptr i32, ptr [[TMP2]], i32 0
; FIXED-NEXT:    [[WIDE_VEC:%.*]] = load <6 x i32>, ptr [[TMP3]], align 4
; FIXED-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <6 x i32> [[WIDE_VEC]], <6 x i32> poison, <2 x i32> <i32 0, i32 3>
; FIXED-NEXT:    [[STRIDED_VEC1:%.*]] = shufflevector <6 x i32> [[WIDE_VEC]], <6 x i32> poison, <2 x i32> <i32 1, i32 4>
; FIXED-NEXT:    [[STRIDED_VEC2:%.*]] = shufflevector <6 x i32> [[WIDE_VEC]], <6 x i32> poison, <2 x i32> <i32 2, i32 5>
; FIXED-NEXT:    [[TMP4:%.*]] = add <2 x i32> [[STRIDED_VEC]], <i32 1, i32 1>
; FIXED-NEXT:    [[TMP5:%.*]] = add i64 [[TMP1]], 1
; FIXED-NEXT:    [[TMP6:%.*]] = add <2 x i32> [[STRIDED_VEC1]], <i32 2, i32 2>
; FIXED-NEXT:    [[TMP7:%.*]] = add i64 [[TMP5]], 1
; FIXED-NEXT:    [[TMP8:%.*]] = getelementptr i32, ptr [[P]], i64 [[TMP7]]
; FIXED-NEXT:    [[TMP9:%.*]] = add <2 x i32> [[STRIDED_VEC2]], <i32 3, i32 3>
; FIXED-NEXT:    [[TMP10:%.*]] = getelementptr i32, ptr [[TMP8]], i32 -2
; FIXED-NEXT:    [[TMP11:%.*]] = shufflevector <2 x i32> [[TMP4]], <2 x i32> [[TMP6]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; FIXED-NEXT:    [[TMP12:%.*]] = shufflevector <2 x i32> [[TMP9]], <2 x i32> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
; FIXED-NEXT:    [[TMP13:%.*]] = shufflevector <4 x i32> [[TMP11]], <4 x i32> [[TMP12]], <6 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5>
; FIXED-NEXT:    [[INTERLEAVED_VEC:%.*]] = shufflevector <6 x i32> [[TMP13]], <6 x i32> poison, <6 x i32> <i32 0, i32 2, i32 4, i32 1, i32 3, i32 5>
; FIXED-NEXT:    store <6 x i32> [[INTERLEAVED_VEC]], ptr [[TMP10]], align 4
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; FIXED-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
; FIXED:       scalar.ph:
; FIXED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1024, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       loop:
; FIXED-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; FIXED-NEXT:    [[OFFSET0:%.*]] = mul i64 [[I]], 3
; FIXED-NEXT:    [[Q0:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET0]]
; FIXED-NEXT:    [[X0:%.*]] = load i32, ptr [[Q0]], align 4
; FIXED-NEXT:    [[Y0:%.*]] = add i32 [[X0]], 1
; FIXED-NEXT:    store i32 [[Y0]], ptr [[Q0]], align 4
; FIXED-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; FIXED-NEXT:    [[Q1:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET1]]
; FIXED-NEXT:    [[X1:%.*]] = load i32, ptr [[Q1]], align 4
; FIXED-NEXT:    [[Y1:%.*]] = add i32 [[X1]], 2
; FIXED-NEXT:    store i32 [[Y1]], ptr [[Q1]], align 4
; FIXED-NEXT:    [[OFFSET2:%.*]] = add i64 [[OFFSET1]], 1
; FIXED-NEXT:    [[Q2:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET2]]
; FIXED-NEXT:    [[X2:%.*]] = load i32, ptr [[Q2]], align 4
; FIXED-NEXT:    [[Y2:%.*]] = add i32 [[X2]], 3
; FIXED-NEXT:    store i32 [[Y2]], ptr [[Q2]], align 4
; FIXED-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; FIXED-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; FIXED-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP7:![0-9]+]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @load_store_factor3_i32(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; SCALABLE:       vector.ph:
; SCALABLE-NEXT:    br label [[VECTOR_BODY:%.*]]
; SCALABLE:       vector.body:
; SCALABLE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
; SCALABLE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 3
; SCALABLE-NEXT:    [[TMP2:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP1]]
; SCALABLE-NEXT:    [[TMP3:%.*]] = getelementptr i32, ptr [[TMP2]], i32 0
; SCALABLE-NEXT:    [[WIDE_VEC:%.*]] = load <6 x i32>, ptr [[TMP3]], align 4
; SCALABLE-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <6 x i32> [[WIDE_VEC]], <6 x i32> poison, <2 x i32> <i32 0, i32 3>
; SCALABLE-NEXT:    [[STRIDED_VEC1:%.*]] = shufflevector <6 x i32> [[WIDE_VEC]], <6 x i32> poison, <2 x i32> <i32 1, i32 4>
; SCALABLE-NEXT:    [[STRIDED_VEC2:%.*]] = shufflevector <6 x i32> [[WIDE_VEC]], <6 x i32> poison, <2 x i32> <i32 2, i32 5>
; SCALABLE-NEXT:    [[TMP4:%.*]] = add <2 x i32> [[STRIDED_VEC]], <i32 1, i32 1>
; SCALABLE-NEXT:    [[TMP5:%.*]] = add i64 [[TMP1]], 1
; SCALABLE-NEXT:    [[TMP6:%.*]] = add <2 x i32> [[STRIDED_VEC1]], <i32 2, i32 2>
; SCALABLE-NEXT:    [[TMP7:%.*]] = add i64 [[TMP5]], 1
; SCALABLE-NEXT:    [[TMP8:%.*]] = getelementptr i32, ptr [[P]], i64 [[TMP7]]
; SCALABLE-NEXT:    [[TMP9:%.*]] = add <2 x i32> [[STRIDED_VEC2]], <i32 3, i32 3>
; SCALABLE-NEXT:    [[TMP10:%.*]] = getelementptr i32, ptr [[TMP8]], i32 -2
; SCALABLE-NEXT:    [[TMP11:%.*]] = shufflevector <2 x i32> [[TMP4]], <2 x i32> [[TMP6]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; SCALABLE-NEXT:    [[TMP12:%.*]] = shufflevector <2 x i32> [[TMP9]], <2 x i32> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
; SCALABLE-NEXT:    [[TMP13:%.*]] = shufflevector <4 x i32> [[TMP11]], <4 x i32> [[TMP12]], <6 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5>
; SCALABLE-NEXT:    [[INTERLEAVED_VEC:%.*]] = shufflevector <6 x i32> [[TMP13]], <6 x i32> poison, <6 x i32> <i32 0, i32 2, i32 4, i32 1, i32 3, i32 5>
; SCALABLE-NEXT:    store <6 x i32> [[INTERLEAVED_VEC]], ptr [[TMP10]], align 4
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; SCALABLE-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; SCALABLE-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
; SCALABLE:       scalar.ph:
; SCALABLE-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1024, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; SCALABLE:       loop:
; SCALABLE-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; SCALABLE-NEXT:    [[OFFSET0:%.*]] = mul i64 [[I]], 3
; SCALABLE-NEXT:    [[Q0:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET0]]
; SCALABLE-NEXT:    [[X0:%.*]] = load i32, ptr [[Q0]], align 4
; SCALABLE-NEXT:    [[Y0:%.*]] = add i32 [[X0]], 1
; SCALABLE-NEXT:    store i32 [[Y0]], ptr [[Q0]], align 4
; SCALABLE-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; SCALABLE-NEXT:    [[Q1:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET1]]
; SCALABLE-NEXT:    [[X1:%.*]] = load i32, ptr [[Q1]], align 4
; SCALABLE-NEXT:    [[Y1:%.*]] = add i32 [[X1]], 2
; SCALABLE-NEXT:    store i32 [[Y1]], ptr [[Q1]], align 4
; SCALABLE-NEXT:    [[OFFSET2:%.*]] = add i64 [[OFFSET1]], 1
; SCALABLE-NEXT:    [[Q2:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET2]]
; SCALABLE-NEXT:    [[X2:%.*]] = load i32, ptr [[Q2]], align 4
; SCALABLE-NEXT:    [[Y2:%.*]] = add i32 [[X2]], 3
; SCALABLE-NEXT:    store i32 [[Y2]], ptr [[Q2]], align 4
; SCALABLE-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; SCALABLE-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; SCALABLE-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP7:![0-9]+]]
; SCALABLE:       exit:
; SCALABLE-NEXT:    ret void
;
entry:
  br label %loop
loop:
  %i = phi i64 [0, %entry], [%nexti, %loop]

  %offset0 = mul i64 %i, 3
  %q0 = getelementptr i32, ptr %p, i64 %offset0
  %x0 = load i32, ptr %q0
  %y0 = add i32 %x0, 1
  store i32 %y0, ptr %q0

  %offset1 = add i64 %offset0, 1
  %q1 = getelementptr i32, ptr %p, i64 %offset1
  %x1 = load i32, ptr %q1
  %y1 = add i32 %x1, 2
  store i32 %y1, ptr %q1

  %offset2 = add i64 %offset1, 1
  %q2 = getelementptr i32, ptr %p, i64 %offset2
  %x2 = load i32, ptr %q2
  %y2 = add i32 %x2, 3
  store i32 %y2, ptr %q2

  %nexti = add i64 %i, 1
  %done = icmp eq i64 %nexti, 1024
  br i1 %done, label %exit, label %loop
exit:
  ret void
}

define void @load_store_factor3_i64(ptr %p) {
; CHECK-LABEL: @load_store_factor3_i64(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1024, [[TMP1]]
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP3]]
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
; CHECK-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; CHECK-NEXT:    [[TMP6:%.*]] = call <vscale x 2 x i64> @llvm.experimental.stepvector.nxv2i64()
; CHECK-NEXT:    [[TMP7:%.*]] = add <vscale x 2 x i64> [[TMP6]], zeroinitializer
; CHECK-NEXT:    [[TMP8:%.*]] = mul <vscale x 2 x i64> [[TMP7]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[INDUCTION:%.*]] = add <vscale x 2 x i64> zeroinitializer, [[TMP8]]
; CHECK-NEXT:    [[TMP9:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP10:%.*]] = mul i64 [[TMP9]], 2
; CHECK-NEXT:    [[TMP11:%.*]] = mul i64 1, [[TMP10]]
; CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[TMP11]], i64 0
; CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 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:    [[VEC_IND:%.*]] = phi <vscale x 2 x i64> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP12:%.*]] = mul <vscale x 2 x i64> [[VEC_IND]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 3, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr i64, ptr [[P:%.*]], <vscale x 2 x i64> [[TMP12]]
; CHECK-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP13]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; CHECK-NEXT:    [[TMP14:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP14]], <vscale x 2 x ptr> [[TMP13]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; CHECK-NEXT:    [[TMP15:%.*]] = add <vscale x 2 x i64> [[TMP12]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP16:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP15]]
; CHECK-NEXT:    [[WIDE_MASKED_GATHER1:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP16]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; CHECK-NEXT:    [[TMP17:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER1]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 2, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP17]], <vscale x 2 x ptr> [[TMP16]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; CHECK-NEXT:    [[TMP18:%.*]] = add <vscale x 2 x i64> [[TMP15]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP19:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP18]]
; CHECK-NEXT:    [[WIDE_MASKED_GATHER2:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP19]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; CHECK-NEXT:    [[TMP20:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER2]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 3, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP20]], <vscale x 2 x ptr> [[TMP19]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 2 x i64> [[VEC_IND]], [[DOTSPLAT]]
; CHECK-NEXT:    [[TMP21:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP21]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
; CHECK:       scalar.ph:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; CHECK-NEXT:    [[OFFSET0:%.*]] = mul i64 [[I]], 3
; CHECK-NEXT:    [[Q0:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET0]]
; CHECK-NEXT:    [[X0:%.*]] = load i64, ptr [[Q0]], align 8
; CHECK-NEXT:    [[Y0:%.*]] = add i64 [[X0]], 1
; CHECK-NEXT:    store i64 [[Y0]], ptr [[Q0]], align 8
; CHECK-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; CHECK-NEXT:    [[Q1:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET1]]
; CHECK-NEXT:    [[X1:%.*]] = load i64, ptr [[Q1]], align 8
; CHECK-NEXT:    [[Y1:%.*]] = add i64 [[X1]], 2
; CHECK-NEXT:    store i64 [[Y1]], ptr [[Q1]], align 8
; CHECK-NEXT:    [[OFFSET2:%.*]] = add i64 [[OFFSET1]], 1
; CHECK-NEXT:    [[Q2:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET2]]
; CHECK-NEXT:    [[X2:%.*]] = load i64, ptr [[Q2]], align 8
; CHECK-NEXT:    [[Y2:%.*]] = add i64 [[X2]], 3
; CHECK-NEXT:    store i64 [[Y2]], ptr [[Q2]], align 8
; CHECK-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; CHECK-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; CHECK-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP9:![0-9]+]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @load_store_factor3_i64(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; FIXED:       vector.ph:
; FIXED-NEXT:    br label [[VECTOR_BODY:%.*]]
; FIXED:       vector.body:
; FIXED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; FIXED-NEXT:    [[VEC_IND:%.*]] = phi <4 x i64> [ <i64 0, i64 1, i64 2, i64 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; FIXED-NEXT:    [[TMP0:%.*]] = mul <4 x i64> [[VEC_IND]], <i64 3, i64 3, i64 3, i64 3>
; FIXED-NEXT:    [[TMP1:%.*]] = getelementptr i64, ptr [[P:%.*]], <4 x i64> [[TMP0]]
; FIXED-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <4 x i64> @llvm.masked.gather.v4i64.v4p0(<4 x ptr> [[TMP1]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i64> poison)
; FIXED-NEXT:    [[TMP2:%.*]] = add <4 x i64> [[WIDE_MASKED_GATHER]], <i64 1, i64 1, i64 1, i64 1>
; FIXED-NEXT:    call void @llvm.masked.scatter.v4i64.v4p0(<4 x i64> [[TMP2]], <4 x ptr> [[TMP1]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>)
; FIXED-NEXT:    [[TMP3:%.*]] = add <4 x i64> [[TMP0]], <i64 1, i64 1, i64 1, i64 1>
; FIXED-NEXT:    [[TMP4:%.*]] = getelementptr i64, ptr [[P]], <4 x i64> [[TMP3]]
; FIXED-NEXT:    [[WIDE_MASKED_GATHER1:%.*]] = call <4 x i64> @llvm.masked.gather.v4i64.v4p0(<4 x ptr> [[TMP4]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i64> poison)
; FIXED-NEXT:    [[TMP5:%.*]] = add <4 x i64> [[WIDE_MASKED_GATHER1]], <i64 2, i64 2, i64 2, i64 2>
; FIXED-NEXT:    call void @llvm.masked.scatter.v4i64.v4p0(<4 x i64> [[TMP5]], <4 x ptr> [[TMP4]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>)
; FIXED-NEXT:    [[TMP6:%.*]] = add <4 x i64> [[TMP3]], <i64 1, i64 1, i64 1, i64 1>
; FIXED-NEXT:    [[TMP7:%.*]] = getelementptr i64, ptr [[P]], <4 x i64> [[TMP6]]
; FIXED-NEXT:    [[WIDE_MASKED_GATHER2:%.*]] = call <4 x i64> @llvm.masked.gather.v4i64.v4p0(<4 x ptr> [[TMP7]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i64> poison)
; FIXED-NEXT:    [[TMP8:%.*]] = add <4 x i64> [[WIDE_MASKED_GATHER2]], <i64 3, i64 3, i64 3, i64 3>
; FIXED-NEXT:    call void @llvm.masked.scatter.v4i64.v4p0(<4 x i64> [[TMP8]], <4 x ptr> [[TMP7]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>)
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; FIXED-NEXT:    [[VEC_IND_NEXT]] = add <4 x i64> [[VEC_IND]], <i64 4, i64 4, i64 4, i64 4>
; FIXED-NEXT:    [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
; FIXED:       scalar.ph:
; FIXED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1024, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       loop:
; FIXED-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; FIXED-NEXT:    [[OFFSET0:%.*]] = mul i64 [[I]], 3
; FIXED-NEXT:    [[Q0:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET0]]
; FIXED-NEXT:    [[X0:%.*]] = load i64, ptr [[Q0]], align 8
; FIXED-NEXT:    [[Y0:%.*]] = add i64 [[X0]], 1
; FIXED-NEXT:    store i64 [[Y0]], ptr [[Q0]], align 8
; FIXED-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; FIXED-NEXT:    [[Q1:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET1]]
; FIXED-NEXT:    [[X1:%.*]] = load i64, ptr [[Q1]], align 8
; FIXED-NEXT:    [[Y1:%.*]] = add i64 [[X1]], 2
; FIXED-NEXT:    store i64 [[Y1]], ptr [[Q1]], align 8
; FIXED-NEXT:    [[OFFSET2:%.*]] = add i64 [[OFFSET1]], 1
; FIXED-NEXT:    [[Q2:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET2]]
; FIXED-NEXT:    [[X2:%.*]] = load i64, ptr [[Q2]], align 8
; FIXED-NEXT:    [[Y2:%.*]] = add i64 [[X2]], 3
; FIXED-NEXT:    store i64 [[Y2]], ptr [[Q2]], align 8
; FIXED-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; FIXED-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; FIXED-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP9:![0-9]+]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @load_store_factor3_i64(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; SCALABLE-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1024, [[TMP1]]
; SCALABLE-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; SCALABLE:       vector.ph:
; SCALABLE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; SCALABLE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP3]]
; SCALABLE-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
; SCALABLE-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; SCALABLE-NEXT:    [[TMP6:%.*]] = call <vscale x 2 x i64> @llvm.experimental.stepvector.nxv2i64()
; SCALABLE-NEXT:    [[TMP7:%.*]] = add <vscale x 2 x i64> [[TMP6]], zeroinitializer
; SCALABLE-NEXT:    [[TMP8:%.*]] = mul <vscale x 2 x i64> [[TMP7]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[INDUCTION:%.*]] = add <vscale x 2 x i64> zeroinitializer, [[TMP8]]
; SCALABLE-NEXT:    [[TMP9:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP10:%.*]] = mul i64 [[TMP9]], 2
; SCALABLE-NEXT:    [[TMP11:%.*]] = mul i64 1, [[TMP10]]
; SCALABLE-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[TMP11]], i64 0
; SCALABLE-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
; SCALABLE-NEXT:    br label [[VECTOR_BODY:%.*]]
; SCALABLE:       vector.body:
; SCALABLE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[VEC_IND:%.*]] = phi <vscale x 2 x i64> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP12:%.*]] = mul <vscale x 2 x i64> [[VEC_IND]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 3, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP13:%.*]] = getelementptr i64, ptr [[P:%.*]], <vscale x 2 x i64> [[TMP12]]
; SCALABLE-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP13]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; SCALABLE-NEXT:    [[TMP14:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP14]], <vscale x 2 x ptr> [[TMP13]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; SCALABLE-NEXT:    [[TMP15:%.*]] = add <vscale x 2 x i64> [[TMP12]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP16:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP15]]
; SCALABLE-NEXT:    [[WIDE_MASKED_GATHER1:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP16]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; SCALABLE-NEXT:    [[TMP17:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER1]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 2, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP17]], <vscale x 2 x ptr> [[TMP16]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; SCALABLE-NEXT:    [[TMP18:%.*]] = add <vscale x 2 x i64> [[TMP15]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP19:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP18]]
; SCALABLE-NEXT:    [[WIDE_MASKED_GATHER2:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP19]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; SCALABLE-NEXT:    [[TMP20:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER2]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 3, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP20]], <vscale x 2 x ptr> [[TMP19]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; SCALABLE-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 2 x i64> [[VEC_IND]], [[DOTSPLAT]]
; SCALABLE-NEXT:    [[TMP21:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; SCALABLE-NEXT:    br i1 [[TMP21]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
; SCALABLE-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
; SCALABLE:       scalar.ph:
; SCALABLE-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; SCALABLE:       loop:
; SCALABLE-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; SCALABLE-NEXT:    [[OFFSET0:%.*]] = mul i64 [[I]], 3
; SCALABLE-NEXT:    [[Q0:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET0]]
; SCALABLE-NEXT:    [[X0:%.*]] = load i64, ptr [[Q0]], align 8
; SCALABLE-NEXT:    [[Y0:%.*]] = add i64 [[X0]], 1
; SCALABLE-NEXT:    store i64 [[Y0]], ptr [[Q0]], align 8
; SCALABLE-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; SCALABLE-NEXT:    [[Q1:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET1]]
; SCALABLE-NEXT:    [[X1:%.*]] = load i64, ptr [[Q1]], align 8
; SCALABLE-NEXT:    [[Y1:%.*]] = add i64 [[X1]], 2
; SCALABLE-NEXT:    store i64 [[Y1]], ptr [[Q1]], align 8
; SCALABLE-NEXT:    [[OFFSET2:%.*]] = add i64 [[OFFSET1]], 1
; SCALABLE-NEXT:    [[Q2:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET2]]
; SCALABLE-NEXT:    [[X2:%.*]] = load i64, ptr [[Q2]], align 8
; SCALABLE-NEXT:    [[Y2:%.*]] = add i64 [[X2]], 3
; SCALABLE-NEXT:    store i64 [[Y2]], ptr [[Q2]], align 8
; SCALABLE-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; SCALABLE-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; SCALABLE-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP9:![0-9]+]]
; SCALABLE:       exit:
; SCALABLE-NEXT:    ret void
;
entry:
  br label %loop
loop:
  %i = phi i64 [0, %entry], [%nexti, %loop]

  %offset0 = mul i64 %i, 3
  %q0 = getelementptr i64, ptr %p, i64 %offset0
  %x0 = load i64, ptr %q0
  %y0 = add i64 %x0, 1
  store i64 %y0, ptr %q0

  %offset1 = add i64 %offset0, 1
  %q1 = getelementptr i64, ptr %p, i64 %offset1
  %x1 = load i64, ptr %q1
  %y1 = add i64 %x1, 2
  store i64 %y1, ptr %q1

  %offset2 = add i64 %offset1, 1
  %q2 = getelementptr i64, ptr %p, i64 %offset2
  %x2 = load i64, ptr %q2
  %y2 = add i64 %x2, 3
  store i64 %y2, ptr %q2

  %nexti = add i64 %i, 1
  %done = icmp eq i64 %nexti, 1024
  br i1 %done, label %exit, label %loop
exit:
  ret void
}

define void @load_store_factor8(ptr %p) {
; CHECK-LABEL: @load_store_factor8(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1024, [[TMP1]]
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP3]]
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
; CHECK-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; CHECK-NEXT:    [[TMP6:%.*]] = call <vscale x 2 x i64> @llvm.experimental.stepvector.nxv2i64()
; CHECK-NEXT:    [[TMP7:%.*]] = add <vscale x 2 x i64> [[TMP6]], zeroinitializer
; CHECK-NEXT:    [[TMP8:%.*]] = mul <vscale x 2 x i64> [[TMP7]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[INDUCTION:%.*]] = add <vscale x 2 x i64> zeroinitializer, [[TMP8]]
; CHECK-NEXT:    [[TMP9:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP10:%.*]] = mul i64 [[TMP9]], 2
; CHECK-NEXT:    [[TMP11:%.*]] = mul i64 1, [[TMP10]]
; CHECK-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[TMP11]], i64 0
; CHECK-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 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:    [[VEC_IND:%.*]] = phi <vscale x 2 x i64> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP12:%.*]] = shl <vscale x 2 x i64> [[VEC_IND]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 3, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr i64, ptr [[P:%.*]], <vscale x 2 x i64> [[TMP12]]
; CHECK-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP13]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; CHECK-NEXT:    [[TMP14:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP14]], <vscale x 2 x ptr> [[TMP13]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; CHECK-NEXT:    [[TMP15:%.*]] = add <vscale x 2 x i64> [[TMP12]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP16:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP15]]
; CHECK-NEXT:    [[WIDE_MASKED_GATHER1:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP16]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; CHECK-NEXT:    [[TMP17:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER1]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 2, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP17]], <vscale x 2 x ptr> [[TMP16]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; CHECK-NEXT:    [[TMP18:%.*]] = add <vscale x 2 x i64> [[TMP15]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP19:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP18]]
; CHECK-NEXT:    [[WIDE_MASKED_GATHER2:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP19]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; CHECK-NEXT:    [[TMP20:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER2]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 3, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP20]], <vscale x 2 x ptr> [[TMP19]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; CHECK-NEXT:    [[TMP21:%.*]] = add <vscale x 2 x i64> [[TMP18]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP22:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP21]]
; CHECK-NEXT:    [[WIDE_MASKED_GATHER3:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP22]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; CHECK-NEXT:    [[TMP23:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER3]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 4, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP23]], <vscale x 2 x ptr> [[TMP22]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; CHECK-NEXT:    [[TMP24:%.*]] = add <vscale x 2 x i64> [[TMP21]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP25:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP24]]
; CHECK-NEXT:    [[WIDE_MASKED_GATHER4:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP25]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; CHECK-NEXT:    [[TMP26:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER4]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 5, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP26]], <vscale x 2 x ptr> [[TMP25]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; CHECK-NEXT:    [[TMP27:%.*]] = add <vscale x 2 x i64> [[TMP24]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP28:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP27]]
; CHECK-NEXT:    [[WIDE_MASKED_GATHER5:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP28]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; CHECK-NEXT:    [[TMP29:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER5]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 6, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP29]], <vscale x 2 x ptr> [[TMP28]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; CHECK-NEXT:    [[TMP30:%.*]] = add <vscale x 2 x i64> [[TMP27]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP31:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP30]]
; CHECK-NEXT:    [[WIDE_MASKED_GATHER6:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP31]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; CHECK-NEXT:    [[TMP32:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER6]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 7, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP32]], <vscale x 2 x ptr> [[TMP31]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; CHECK-NEXT:    [[TMP33:%.*]] = add <vscale x 2 x i64> [[TMP30]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    [[TMP34:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP33]]
; CHECK-NEXT:    [[WIDE_MASKED_GATHER7:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP34]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; CHECK-NEXT:    [[TMP35:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER7]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 8, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; CHECK-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP35]], <vscale x 2 x ptr> [[TMP34]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 2 x i64> [[VEC_IND]], [[DOTSPLAT]]
; CHECK-NEXT:    [[TMP36:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP36]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
; CHECK:       scalar.ph:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; CHECK-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 3
; CHECK-NEXT:    [[Q0:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET0]]
; CHECK-NEXT:    [[X0:%.*]] = load i64, ptr [[Q0]], align 8
; CHECK-NEXT:    [[Y0:%.*]] = add i64 [[X0]], 1
; CHECK-NEXT:    store i64 [[Y0]], ptr [[Q0]], align 8
; CHECK-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; CHECK-NEXT:    [[Q1:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET1]]
; CHECK-NEXT:    [[X1:%.*]] = load i64, ptr [[Q1]], align 8
; CHECK-NEXT:    [[Y1:%.*]] = add i64 [[X1]], 2
; CHECK-NEXT:    store i64 [[Y1]], ptr [[Q1]], align 8
; CHECK-NEXT:    [[OFFSET2:%.*]] = add i64 [[OFFSET1]], 1
; CHECK-NEXT:    [[Q2:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET2]]
; CHECK-NEXT:    [[X2:%.*]] = load i64, ptr [[Q2]], align 8
; CHECK-NEXT:    [[Y2:%.*]] = add i64 [[X2]], 3
; CHECK-NEXT:    store i64 [[Y2]], ptr [[Q2]], align 8
; CHECK-NEXT:    [[OFFSET3:%.*]] = add i64 [[OFFSET2]], 1
; CHECK-NEXT:    [[Q3:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET3]]
; CHECK-NEXT:    [[X3:%.*]] = load i64, ptr [[Q3]], align 8
; CHECK-NEXT:    [[Y3:%.*]] = add i64 [[X3]], 4
; CHECK-NEXT:    store i64 [[Y3]], ptr [[Q3]], align 8
; CHECK-NEXT:    [[OFFSET4:%.*]] = add i64 [[OFFSET3]], 1
; CHECK-NEXT:    [[Q4:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET4]]
; CHECK-NEXT:    [[X4:%.*]] = load i64, ptr [[Q4]], align 8
; CHECK-NEXT:    [[Y4:%.*]] = add i64 [[X4]], 5
; CHECK-NEXT:    store i64 [[Y4]], ptr [[Q4]], align 8
; CHECK-NEXT:    [[OFFSET5:%.*]] = add i64 [[OFFSET4]], 1
; CHECK-NEXT:    [[Q5:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET5]]
; CHECK-NEXT:    [[X5:%.*]] = load i64, ptr [[Q5]], align 8
; CHECK-NEXT:    [[Y5:%.*]] = add i64 [[X5]], 6
; CHECK-NEXT:    store i64 [[Y5]], ptr [[Q5]], align 8
; CHECK-NEXT:    [[OFFSET6:%.*]] = add i64 [[OFFSET5]], 1
; CHECK-NEXT:    [[Q6:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET6]]
; CHECK-NEXT:    [[X6:%.*]] = load i64, ptr [[Q6]], align 8
; CHECK-NEXT:    [[Y6:%.*]] = add i64 [[X6]], 7
; CHECK-NEXT:    store i64 [[Y6]], ptr [[Q6]], align 8
; CHECK-NEXT:    [[OFFSET7:%.*]] = add i64 [[OFFSET6]], 1
; CHECK-NEXT:    [[Q7:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET7]]
; CHECK-NEXT:    [[X7:%.*]] = load i64, ptr [[Q7]], align 8
; CHECK-NEXT:    [[Y7:%.*]] = add i64 [[X7]], 8
; CHECK-NEXT:    store i64 [[Y7]], ptr [[Q7]], align 8
; CHECK-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; CHECK-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; CHECK-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP11:![0-9]+]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @load_store_factor8(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; FIXED:       vector.ph:
; FIXED-NEXT:    br label [[VECTOR_BODY:%.*]]
; FIXED:       vector.body:
; FIXED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; FIXED-NEXT:    [[VEC_IND:%.*]] = phi <4 x i64> [ <i64 0, i64 1, i64 2, i64 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; FIXED-NEXT:    [[TMP0:%.*]] = shl <4 x i64> [[VEC_IND]], <i64 3, i64 3, i64 3, i64 3>
; FIXED-NEXT:    [[TMP1:%.*]] = getelementptr i64, ptr [[P:%.*]], <4 x i64> [[TMP0]]
; FIXED-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <4 x i64> @llvm.masked.gather.v4i64.v4p0(<4 x ptr> [[TMP1]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i64> poison)
; FIXED-NEXT:    [[TMP2:%.*]] = add <4 x i64> [[WIDE_MASKED_GATHER]], <i64 1, i64 1, i64 1, i64 1>
; FIXED-NEXT:    call void @llvm.masked.scatter.v4i64.v4p0(<4 x i64> [[TMP2]], <4 x ptr> [[TMP1]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>)
; FIXED-NEXT:    [[TMP3:%.*]] = add <4 x i64> [[TMP0]], <i64 1, i64 1, i64 1, i64 1>
; FIXED-NEXT:    [[TMP4:%.*]] = getelementptr i64, ptr [[P]], <4 x i64> [[TMP3]]
; FIXED-NEXT:    [[WIDE_MASKED_GATHER1:%.*]] = call <4 x i64> @llvm.masked.gather.v4i64.v4p0(<4 x ptr> [[TMP4]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i64> poison)
; FIXED-NEXT:    [[TMP5:%.*]] = add <4 x i64> [[WIDE_MASKED_GATHER1]], <i64 2, i64 2, i64 2, i64 2>
; FIXED-NEXT:    call void @llvm.masked.scatter.v4i64.v4p0(<4 x i64> [[TMP5]], <4 x ptr> [[TMP4]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>)
; FIXED-NEXT:    [[TMP6:%.*]] = add <4 x i64> [[TMP3]], <i64 1, i64 1, i64 1, i64 1>
; FIXED-NEXT:    [[TMP7:%.*]] = getelementptr i64, ptr [[P]], <4 x i64> [[TMP6]]
; FIXED-NEXT:    [[WIDE_MASKED_GATHER2:%.*]] = call <4 x i64> @llvm.masked.gather.v4i64.v4p0(<4 x ptr> [[TMP7]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i64> poison)
; FIXED-NEXT:    [[TMP8:%.*]] = add <4 x i64> [[WIDE_MASKED_GATHER2]], <i64 3, i64 3, i64 3, i64 3>
; FIXED-NEXT:    call void @llvm.masked.scatter.v4i64.v4p0(<4 x i64> [[TMP8]], <4 x ptr> [[TMP7]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>)
; FIXED-NEXT:    [[TMP9:%.*]] = add <4 x i64> [[TMP6]], <i64 1, i64 1, i64 1, i64 1>
; FIXED-NEXT:    [[TMP10:%.*]] = getelementptr i64, ptr [[P]], <4 x i64> [[TMP9]]
; FIXED-NEXT:    [[WIDE_MASKED_GATHER3:%.*]] = call <4 x i64> @llvm.masked.gather.v4i64.v4p0(<4 x ptr> [[TMP10]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i64> poison)
; FIXED-NEXT:    [[TMP11:%.*]] = add <4 x i64> [[WIDE_MASKED_GATHER3]], <i64 4, i64 4, i64 4, i64 4>
; FIXED-NEXT:    call void @llvm.masked.scatter.v4i64.v4p0(<4 x i64> [[TMP11]], <4 x ptr> [[TMP10]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>)
; FIXED-NEXT:    [[TMP12:%.*]] = add <4 x i64> [[TMP9]], <i64 1, i64 1, i64 1, i64 1>
; FIXED-NEXT:    [[TMP13:%.*]] = getelementptr i64, ptr [[P]], <4 x i64> [[TMP12]]
; FIXED-NEXT:    [[WIDE_MASKED_GATHER4:%.*]] = call <4 x i64> @llvm.masked.gather.v4i64.v4p0(<4 x ptr> [[TMP13]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i64> poison)
; FIXED-NEXT:    [[TMP14:%.*]] = add <4 x i64> [[WIDE_MASKED_GATHER4]], <i64 5, i64 5, i64 5, i64 5>
; FIXED-NEXT:    call void @llvm.masked.scatter.v4i64.v4p0(<4 x i64> [[TMP14]], <4 x ptr> [[TMP13]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>)
; FIXED-NEXT:    [[TMP15:%.*]] = add <4 x i64> [[TMP12]], <i64 1, i64 1, i64 1, i64 1>
; FIXED-NEXT:    [[TMP16:%.*]] = getelementptr i64, ptr [[P]], <4 x i64> [[TMP15]]
; FIXED-NEXT:    [[WIDE_MASKED_GATHER5:%.*]] = call <4 x i64> @llvm.masked.gather.v4i64.v4p0(<4 x ptr> [[TMP16]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i64> poison)
; FIXED-NEXT:    [[TMP17:%.*]] = add <4 x i64> [[WIDE_MASKED_GATHER5]], <i64 6, i64 6, i64 6, i64 6>
; FIXED-NEXT:    call void @llvm.masked.scatter.v4i64.v4p0(<4 x i64> [[TMP17]], <4 x ptr> [[TMP16]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>)
; FIXED-NEXT:    [[TMP18:%.*]] = add <4 x i64> [[TMP15]], <i64 1, i64 1, i64 1, i64 1>
; FIXED-NEXT:    [[TMP19:%.*]] = getelementptr i64, ptr [[P]], <4 x i64> [[TMP18]]
; FIXED-NEXT:    [[WIDE_MASKED_GATHER6:%.*]] = call <4 x i64> @llvm.masked.gather.v4i64.v4p0(<4 x ptr> [[TMP19]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i64> poison)
; FIXED-NEXT:    [[TMP20:%.*]] = add <4 x i64> [[WIDE_MASKED_GATHER6]], <i64 7, i64 7, i64 7, i64 7>
; FIXED-NEXT:    call void @llvm.masked.scatter.v4i64.v4p0(<4 x i64> [[TMP20]], <4 x ptr> [[TMP19]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>)
; FIXED-NEXT:    [[TMP21:%.*]] = add <4 x i64> [[TMP18]], <i64 1, i64 1, i64 1, i64 1>
; FIXED-NEXT:    [[TMP22:%.*]] = getelementptr i64, ptr [[P]], <4 x i64> [[TMP21]]
; FIXED-NEXT:    [[WIDE_MASKED_GATHER7:%.*]] = call <4 x i64> @llvm.masked.gather.v4i64.v4p0(<4 x ptr> [[TMP22]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i64> poison)
; FIXED-NEXT:    [[TMP23:%.*]] = add <4 x i64> [[WIDE_MASKED_GATHER7]], <i64 8, i64 8, i64 8, i64 8>
; FIXED-NEXT:    call void @llvm.masked.scatter.v4i64.v4p0(<4 x i64> [[TMP23]], <4 x ptr> [[TMP22]], i32 8, <4 x i1> <i1 true, i1 true, i1 true, i1 true>)
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; FIXED-NEXT:    [[VEC_IND_NEXT]] = add <4 x i64> [[VEC_IND]], <i64 4, i64 4, i64 4, i64 4>
; FIXED-NEXT:    [[TMP24:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP24]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
; FIXED:       scalar.ph:
; FIXED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1024, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       loop:
; FIXED-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; FIXED-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 3
; FIXED-NEXT:    [[Q0:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET0]]
; FIXED-NEXT:    [[X0:%.*]] = load i64, ptr [[Q0]], align 8
; FIXED-NEXT:    [[Y0:%.*]] = add i64 [[X0]], 1
; FIXED-NEXT:    store i64 [[Y0]], ptr [[Q0]], align 8
; FIXED-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; FIXED-NEXT:    [[Q1:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET1]]
; FIXED-NEXT:    [[X1:%.*]] = load i64, ptr [[Q1]], align 8
; FIXED-NEXT:    [[Y1:%.*]] = add i64 [[X1]], 2
; FIXED-NEXT:    store i64 [[Y1]], ptr [[Q1]], align 8
; FIXED-NEXT:    [[OFFSET2:%.*]] = add i64 [[OFFSET1]], 1
; FIXED-NEXT:    [[Q2:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET2]]
; FIXED-NEXT:    [[X2:%.*]] = load i64, ptr [[Q2]], align 8
; FIXED-NEXT:    [[Y2:%.*]] = add i64 [[X2]], 3
; FIXED-NEXT:    store i64 [[Y2]], ptr [[Q2]], align 8
; FIXED-NEXT:    [[OFFSET3:%.*]] = add i64 [[OFFSET2]], 1
; FIXED-NEXT:    [[Q3:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET3]]
; FIXED-NEXT:    [[X3:%.*]] = load i64, ptr [[Q3]], align 8
; FIXED-NEXT:    [[Y3:%.*]] = add i64 [[X3]], 4
; FIXED-NEXT:    store i64 [[Y3]], ptr [[Q3]], align 8
; FIXED-NEXT:    [[OFFSET4:%.*]] = add i64 [[OFFSET3]], 1
; FIXED-NEXT:    [[Q4:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET4]]
; FIXED-NEXT:    [[X4:%.*]] = load i64, ptr [[Q4]], align 8
; FIXED-NEXT:    [[Y4:%.*]] = add i64 [[X4]], 5
; FIXED-NEXT:    store i64 [[Y4]], ptr [[Q4]], align 8
; FIXED-NEXT:    [[OFFSET5:%.*]] = add i64 [[OFFSET4]], 1
; FIXED-NEXT:    [[Q5:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET5]]
; FIXED-NEXT:    [[X5:%.*]] = load i64, ptr [[Q5]], align 8
; FIXED-NEXT:    [[Y5:%.*]] = add i64 [[X5]], 6
; FIXED-NEXT:    store i64 [[Y5]], ptr [[Q5]], align 8
; FIXED-NEXT:    [[OFFSET6:%.*]] = add i64 [[OFFSET5]], 1
; FIXED-NEXT:    [[Q6:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET6]]
; FIXED-NEXT:    [[X6:%.*]] = load i64, ptr [[Q6]], align 8
; FIXED-NEXT:    [[Y6:%.*]] = add i64 [[X6]], 7
; FIXED-NEXT:    store i64 [[Y6]], ptr [[Q6]], align 8
; FIXED-NEXT:    [[OFFSET7:%.*]] = add i64 [[OFFSET6]], 1
; FIXED-NEXT:    [[Q7:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET7]]
; FIXED-NEXT:    [[X7:%.*]] = load i64, ptr [[Q7]], align 8
; FIXED-NEXT:    [[Y7:%.*]] = add i64 [[X7]], 8
; FIXED-NEXT:    store i64 [[Y7]], ptr [[Q7]], align 8
; FIXED-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; FIXED-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; FIXED-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP11:![0-9]+]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @load_store_factor8(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; SCALABLE-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1024, [[TMP1]]
; SCALABLE-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; SCALABLE:       vector.ph:
; SCALABLE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; SCALABLE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP3]]
; SCALABLE-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
; SCALABLE-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; SCALABLE-NEXT:    [[TMP6:%.*]] = call <vscale x 2 x i64> @llvm.experimental.stepvector.nxv2i64()
; SCALABLE-NEXT:    [[TMP7:%.*]] = add <vscale x 2 x i64> [[TMP6]], zeroinitializer
; SCALABLE-NEXT:    [[TMP8:%.*]] = mul <vscale x 2 x i64> [[TMP7]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[INDUCTION:%.*]] = add <vscale x 2 x i64> zeroinitializer, [[TMP8]]
; SCALABLE-NEXT:    [[TMP9:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP10:%.*]] = mul i64 [[TMP9]], 2
; SCALABLE-NEXT:    [[TMP11:%.*]] = mul i64 1, [[TMP10]]
; SCALABLE-NEXT:    [[DOTSPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[TMP11]], i64 0
; SCALABLE-NEXT:    [[DOTSPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[DOTSPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
; SCALABLE-NEXT:    br label [[VECTOR_BODY:%.*]]
; SCALABLE:       vector.body:
; SCALABLE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[VEC_IND:%.*]] = phi <vscale x 2 x i64> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP12:%.*]] = shl <vscale x 2 x i64> [[VEC_IND]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 3, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP13:%.*]] = getelementptr i64, ptr [[P:%.*]], <vscale x 2 x i64> [[TMP12]]
; SCALABLE-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP13]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; SCALABLE-NEXT:    [[TMP14:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP14]], <vscale x 2 x ptr> [[TMP13]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; SCALABLE-NEXT:    [[TMP15:%.*]] = add <vscale x 2 x i64> [[TMP12]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP16:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP15]]
; SCALABLE-NEXT:    [[WIDE_MASKED_GATHER1:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP16]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; SCALABLE-NEXT:    [[TMP17:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER1]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 2, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP17]], <vscale x 2 x ptr> [[TMP16]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; SCALABLE-NEXT:    [[TMP18:%.*]] = add <vscale x 2 x i64> [[TMP15]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP19:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP18]]
; SCALABLE-NEXT:    [[WIDE_MASKED_GATHER2:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP19]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; SCALABLE-NEXT:    [[TMP20:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER2]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 3, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP20]], <vscale x 2 x ptr> [[TMP19]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; SCALABLE-NEXT:    [[TMP21:%.*]] = add <vscale x 2 x i64> [[TMP18]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP22:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP21]]
; SCALABLE-NEXT:    [[WIDE_MASKED_GATHER3:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP22]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; SCALABLE-NEXT:    [[TMP23:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER3]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 4, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP23]], <vscale x 2 x ptr> [[TMP22]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; SCALABLE-NEXT:    [[TMP24:%.*]] = add <vscale x 2 x i64> [[TMP21]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP25:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP24]]
; SCALABLE-NEXT:    [[WIDE_MASKED_GATHER4:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP25]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; SCALABLE-NEXT:    [[TMP26:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER4]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 5, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP26]], <vscale x 2 x ptr> [[TMP25]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; SCALABLE-NEXT:    [[TMP27:%.*]] = add <vscale x 2 x i64> [[TMP24]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP28:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP27]]
; SCALABLE-NEXT:    [[WIDE_MASKED_GATHER5:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP28]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; SCALABLE-NEXT:    [[TMP29:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER5]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 6, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP29]], <vscale x 2 x ptr> [[TMP28]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; SCALABLE-NEXT:    [[TMP30:%.*]] = add <vscale x 2 x i64> [[TMP27]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP31:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP30]]
; SCALABLE-NEXT:    [[WIDE_MASKED_GATHER6:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP31]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; SCALABLE-NEXT:    [[TMP32:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER6]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 7, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP32]], <vscale x 2 x ptr> [[TMP31]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; SCALABLE-NEXT:    [[TMP33:%.*]] = add <vscale x 2 x i64> [[TMP30]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 1, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    [[TMP34:%.*]] = getelementptr i64, ptr [[P]], <vscale x 2 x i64> [[TMP33]]
; SCALABLE-NEXT:    [[WIDE_MASKED_GATHER7:%.*]] = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> [[TMP34]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> poison)
; SCALABLE-NEXT:    [[TMP35:%.*]] = add <vscale x 2 x i64> [[WIDE_MASKED_GATHER7]], shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> poison, i64 8, i64 0), <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer)
; SCALABLE-NEXT:    call void @llvm.masked.scatter.nxv2i64.nxv2p0(<vscale x 2 x i64> [[TMP35]], <vscale x 2 x ptr> [[TMP34]], i32 8, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer))
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; SCALABLE-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 2 x i64> [[VEC_IND]], [[DOTSPLAT]]
; SCALABLE-NEXT:    [[TMP36:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; SCALABLE-NEXT:    br i1 [[TMP36]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
; SCALABLE-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
; SCALABLE:       scalar.ph:
; SCALABLE-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; SCALABLE:       loop:
; SCALABLE-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; SCALABLE-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 3
; SCALABLE-NEXT:    [[Q0:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET0]]
; SCALABLE-NEXT:    [[X0:%.*]] = load i64, ptr [[Q0]], align 8
; SCALABLE-NEXT:    [[Y0:%.*]] = add i64 [[X0]], 1
; SCALABLE-NEXT:    store i64 [[Y0]], ptr [[Q0]], align 8
; SCALABLE-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; SCALABLE-NEXT:    [[Q1:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET1]]
; SCALABLE-NEXT:    [[X1:%.*]] = load i64, ptr [[Q1]], align 8
; SCALABLE-NEXT:    [[Y1:%.*]] = add i64 [[X1]], 2
; SCALABLE-NEXT:    store i64 [[Y1]], ptr [[Q1]], align 8
; SCALABLE-NEXT:    [[OFFSET2:%.*]] = add i64 [[OFFSET1]], 1
; SCALABLE-NEXT:    [[Q2:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET2]]
; SCALABLE-NEXT:    [[X2:%.*]] = load i64, ptr [[Q2]], align 8
; SCALABLE-NEXT:    [[Y2:%.*]] = add i64 [[X2]], 3
; SCALABLE-NEXT:    store i64 [[Y2]], ptr [[Q2]], align 8
; SCALABLE-NEXT:    [[OFFSET3:%.*]] = add i64 [[OFFSET2]], 1
; SCALABLE-NEXT:    [[Q3:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET3]]
; SCALABLE-NEXT:    [[X3:%.*]] = load i64, ptr [[Q3]], align 8
; SCALABLE-NEXT:    [[Y3:%.*]] = add i64 [[X3]], 4
; SCALABLE-NEXT:    store i64 [[Y3]], ptr [[Q3]], align 8
; SCALABLE-NEXT:    [[OFFSET4:%.*]] = add i64 [[OFFSET3]], 1
; SCALABLE-NEXT:    [[Q4:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET4]]
; SCALABLE-NEXT:    [[X4:%.*]] = load i64, ptr [[Q4]], align 8
; SCALABLE-NEXT:    [[Y4:%.*]] = add i64 [[X4]], 5
; SCALABLE-NEXT:    store i64 [[Y4]], ptr [[Q4]], align 8
; SCALABLE-NEXT:    [[OFFSET5:%.*]] = add i64 [[OFFSET4]], 1
; SCALABLE-NEXT:    [[Q5:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET5]]
; SCALABLE-NEXT:    [[X5:%.*]] = load i64, ptr [[Q5]], align 8
; SCALABLE-NEXT:    [[Y5:%.*]] = add i64 [[X5]], 6
; SCALABLE-NEXT:    store i64 [[Y5]], ptr [[Q5]], align 8
; SCALABLE-NEXT:    [[OFFSET6:%.*]] = add i64 [[OFFSET5]], 1
; SCALABLE-NEXT:    [[Q6:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET6]]
; SCALABLE-NEXT:    [[X6:%.*]] = load i64, ptr [[Q6]], align 8
; SCALABLE-NEXT:    [[Y6:%.*]] = add i64 [[X6]], 7
; SCALABLE-NEXT:    store i64 [[Y6]], ptr [[Q6]], align 8
; SCALABLE-NEXT:    [[OFFSET7:%.*]] = add i64 [[OFFSET6]], 1
; SCALABLE-NEXT:    [[Q7:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET7]]
; SCALABLE-NEXT:    [[X7:%.*]] = load i64, ptr [[Q7]], align 8
; SCALABLE-NEXT:    [[Y7:%.*]] = add i64 [[X7]], 8
; SCALABLE-NEXT:    store i64 [[Y7]], ptr [[Q7]], align 8
; SCALABLE-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; SCALABLE-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; SCALABLE-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP11:![0-9]+]]
; SCALABLE:       exit:
; SCALABLE-NEXT:    ret void
;
entry:
  br label %loop
loop:
  %i = phi i64 [0, %entry], [%nexti, %loop]

  %offset0 = shl i64 %i, 3
  %q0 = getelementptr i64, ptr %p, i64 %offset0
  %x0 = load i64, ptr %q0
  %y0 = add i64 %x0, 1
  store i64 %y0, ptr %q0

  %offset1 = add i64 %offset0, 1
  %q1 = getelementptr i64, ptr %p, i64 %offset1
  %x1 = load i64, ptr %q1
  %y1 = add i64 %x1, 2
  store i64 %y1, ptr %q1

  %offset2 = add i64 %offset1, 1
  %q2 = getelementptr i64, ptr %p, i64 %offset2
  %x2 = load i64, ptr %q2
  %y2 = add i64 %x2, 3
  store i64 %y2, ptr %q2

  %offset3 = add i64 %offset2, 1
  %q3 = getelementptr i64, ptr %p, i64 %offset3
  %x3 = load i64, ptr %q3
  %y3 = add i64 %x3, 4
  store i64 %y3, ptr %q3

  %offset4 = add i64 %offset3, 1
  %q4 = getelementptr i64, ptr %p, i64 %offset4
  %x4 = load i64, ptr %q4
  %y4 = add i64 %x4, 5
  store i64 %y4, ptr %q4

  %offset5 = add i64 %offset4, 1
  %q5 = getelementptr i64, ptr %p, i64 %offset5
  %x5 = load i64, ptr %q5
  %y5 = add i64 %x5, 6
  store i64 %y5, ptr %q5

  %offset6 = add i64 %offset5, 1
  %q6 = getelementptr i64, ptr %p, i64 %offset6
  %x6 = load i64, ptr %q6
  %y6 = add i64 %x6, 7
  store i64 %y6, ptr %q6

  %offset7 = add i64 %offset6, 1
  %q7 = getelementptr i64, ptr %p, i64 %offset7
  %x7 = load i64, ptr %q7
  %y7 = add i64 %x7, 8
  store i64 %y7, ptr %q7

  %nexti = add i64 %i, 1
  %done = icmp eq i64 %nexti, 1024
  br i1 %done, label %exit, label %loop
exit:
  ret void
}

define void @combine_load_factor2_i32(ptr noalias %p, ptr noalias %q) {
; CHECK-LABEL: @combine_load_factor2_i32(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1024, [[TMP1]]
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP3]]
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
; CHECK-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 4
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP6:%.*]] = add i64 [[INDEX]], 0
; CHECK-NEXT:    [[TMP7:%.*]] = shl i64 [[TMP6]], 1
; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP7]]
; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr i32, ptr [[TMP8]], i32 0
; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP9]], align 4
; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
; CHECK-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
; CHECK-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP12:%.*]] = add <vscale x 4 x i32> [[TMP10]], [[TMP11]]
; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr i32, ptr [[Q:%.*]], i64 [[TMP6]]
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr i32, ptr [[TMP13]], i32 0
; CHECK-NEXT:    store <vscale x 4 x i32> [[TMP12]], ptr [[TMP14]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; CHECK-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
; CHECK:       scalar.ph:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; CHECK-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 1
; CHECK-NEXT:    [[Q0:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET0]]
; CHECK-NEXT:    [[X0:%.*]] = load i32, ptr [[Q0]], align 4
; CHECK-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; CHECK-NEXT:    [[Q1:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET1]]
; CHECK-NEXT:    [[X1:%.*]] = load i32, ptr [[Q1]], align 4
; CHECK-NEXT:    [[RES:%.*]] = add i32 [[X0]], [[X1]]
; CHECK-NEXT:    [[DST:%.*]] = getelementptr i32, ptr [[Q]], i64 [[I]]
; CHECK-NEXT:    store i32 [[RES]], ptr [[DST]], align 4
; CHECK-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; CHECK-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; CHECK-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP13:![0-9]+]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @combine_load_factor2_i32(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; FIXED:       vector.ph:
; FIXED-NEXT:    br label [[VECTOR_BODY:%.*]]
; FIXED:       vector.body:
; FIXED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; FIXED-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
; FIXED-NEXT:    [[TMP1:%.*]] = add i64 [[INDEX]], 8
; FIXED-NEXT:    [[TMP2:%.*]] = shl i64 [[TMP0]], 1
; FIXED-NEXT:    [[TMP3:%.*]] = shl i64 [[TMP1]], 1
; FIXED-NEXT:    [[TMP4:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP2]]
; FIXED-NEXT:    [[TMP5:%.*]] = getelementptr i32, ptr [[P]], i64 [[TMP3]]
; FIXED-NEXT:    [[TMP6:%.*]] = getelementptr i32, ptr [[TMP4]], i32 0
; FIXED-NEXT:    [[TMP7:%.*]] = getelementptr i32, ptr [[TMP5]], i32 0
; FIXED-NEXT:    [[WIDE_VEC:%.*]] = load <16 x i32>, ptr [[TMP6]], align 4
; FIXED-NEXT:    [[WIDE_VEC1:%.*]] = load <16 x i32>, ptr [[TMP7]], align 4
; FIXED-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <16 x i32> [[WIDE_VEC]], <16 x i32> poison, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
; FIXED-NEXT:    [[STRIDED_VEC2:%.*]] = shufflevector <16 x i32> [[WIDE_VEC1]], <16 x i32> poison, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
; FIXED-NEXT:    [[STRIDED_VEC3:%.*]] = shufflevector <16 x i32> [[WIDE_VEC]], <16 x i32> poison, <8 x i32> <i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15>
; FIXED-NEXT:    [[STRIDED_VEC4:%.*]] = shufflevector <16 x i32> [[WIDE_VEC1]], <16 x i32> poison, <8 x i32> <i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15>
; FIXED-NEXT:    [[TMP8:%.*]] = add <8 x i32> [[STRIDED_VEC]], [[STRIDED_VEC3]]
; FIXED-NEXT:    [[TMP9:%.*]] = add <8 x i32> [[STRIDED_VEC2]], [[STRIDED_VEC4]]
; FIXED-NEXT:    [[TMP10:%.*]] = getelementptr i32, ptr [[Q:%.*]], i64 [[TMP0]]
; FIXED-NEXT:    [[TMP11:%.*]] = getelementptr i32, ptr [[Q]], i64 [[TMP1]]
; FIXED-NEXT:    [[TMP12:%.*]] = getelementptr i32, ptr [[TMP10]], i32 0
; FIXED-NEXT:    [[TMP13:%.*]] = getelementptr i32, ptr [[TMP10]], i32 8
; FIXED-NEXT:    store <8 x i32> [[TMP8]], ptr [[TMP12]], align 4
; FIXED-NEXT:    store <8 x i32> [[TMP9]], ptr [[TMP13]], align 4
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16
; FIXED-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
; FIXED:       scalar.ph:
; FIXED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1024, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       loop:
; FIXED-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; FIXED-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 1
; FIXED-NEXT:    [[Q0:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET0]]
; FIXED-NEXT:    [[X0:%.*]] = load i32, ptr [[Q0]], align 4
; FIXED-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; FIXED-NEXT:    [[Q1:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET1]]
; FIXED-NEXT:    [[X1:%.*]] = load i32, ptr [[Q1]], align 4
; FIXED-NEXT:    [[RES:%.*]] = add i32 [[X0]], [[X1]]
; FIXED-NEXT:    [[DST:%.*]] = getelementptr i32, ptr [[Q]], i64 [[I]]
; FIXED-NEXT:    store i32 [[RES]], ptr [[DST]], align 4
; FIXED-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; FIXED-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; FIXED-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP13:![0-9]+]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @combine_load_factor2_i32(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
; SCALABLE-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1024, [[TMP1]]
; SCALABLE-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; SCALABLE:       vector.ph:
; SCALABLE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
; SCALABLE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP3]]
; SCALABLE-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
; SCALABLE-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 4
; SCALABLE-NEXT:    br label [[VECTOR_BODY:%.*]]
; SCALABLE:       vector.body:
; SCALABLE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP6:%.*]] = add i64 [[INDEX]], 0
; SCALABLE-NEXT:    [[TMP7:%.*]] = shl i64 [[TMP6]], 1
; SCALABLE-NEXT:    [[TMP8:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP7]]
; SCALABLE-NEXT:    [[TMP9:%.*]] = getelementptr i32, ptr [[TMP8]], i32 0
; SCALABLE-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP9]], align 4
; SCALABLE-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
; SCALABLE-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
; SCALABLE-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP12:%.*]] = add <vscale x 4 x i32> [[TMP10]], [[TMP11]]
; SCALABLE-NEXT:    [[TMP13:%.*]] = getelementptr i32, ptr [[Q:%.*]], i64 [[TMP6]]
; SCALABLE-NEXT:    [[TMP14:%.*]] = getelementptr i32, ptr [[TMP13]], i32 0
; SCALABLE-NEXT:    store <vscale x 4 x i32> [[TMP12]], ptr [[TMP14]], align 4
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; SCALABLE-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; SCALABLE-NEXT:    br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
; SCALABLE-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
; SCALABLE:       scalar.ph:
; SCALABLE-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; SCALABLE:       loop:
; SCALABLE-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; SCALABLE-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 1
; SCALABLE-NEXT:    [[Q0:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET0]]
; SCALABLE-NEXT:    [[X0:%.*]] = load i32, ptr [[Q0]], align 4
; SCALABLE-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; SCALABLE-NEXT:    [[Q1:%.*]] = getelementptr i32, ptr [[P]], i64 [[OFFSET1]]
; SCALABLE-NEXT:    [[X1:%.*]] = load i32, ptr [[Q1]], align 4
; SCALABLE-NEXT:    [[RES:%.*]] = add i32 [[X0]], [[X1]]
; SCALABLE-NEXT:    [[DST:%.*]] = getelementptr i32, ptr [[Q]], i64 [[I]]
; SCALABLE-NEXT:    store i32 [[RES]], ptr [[DST]], align 4
; SCALABLE-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; SCALABLE-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; SCALABLE-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP13:![0-9]+]]
; SCALABLE:       exit:
; SCALABLE-NEXT:    ret void
;
entry:
  br label %loop
loop:
  %i = phi i64 [0, %entry], [%nexti, %loop]

  %offset0 = shl i64 %i, 1
  %q0 = getelementptr i32, ptr %p, i64 %offset0
  %x0 = load i32, ptr %q0

  %offset1 = add i64 %offset0, 1
  %q1 = getelementptr i32, ptr %p, i64 %offset1
  %x1 = load i32, ptr %q1

  %res = add i32 %x0, %x1

  %dst = getelementptr i32, ptr %q, i64 %i
  store i32 %res, ptr %dst

  %nexti = add i64 %i, 1
  %done = icmp eq i64 %nexti, 1024
  br i1 %done, label %exit, label %loop
exit:
  ret void
}

define void @combine_load_factor2_i64(ptr noalias %p, ptr noalias %q) {
; CHECK-LABEL: @combine_load_factor2_i64(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1024, [[TMP1]]
; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP3]]
; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
; CHECK-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; CHECK-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP6:%.*]] = add i64 [[INDEX]], 0
; CHECK-NEXT:    [[TMP7:%.*]] = shl i64 [[TMP6]], 1
; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP7]]
; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr i64, ptr [[TMP8]], i32 0
; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 4 x i64>, ptr [[TMP9]], align 8
; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> [[WIDE_VEC]])
; CHECK-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 0
; CHECK-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP12:%.*]] = add <vscale x 2 x i64> [[TMP10]], [[TMP11]]
; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr i64, ptr [[Q:%.*]], i64 [[TMP6]]
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr i64, ptr [[TMP13]], i32 0
; CHECK-NEXT:    store <vscale x 2 x i64> [[TMP12]], ptr [[TMP14]], align 8
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; CHECK-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT:    br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP14:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
; CHECK-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
; CHECK:       scalar.ph:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; CHECK-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 1
; CHECK-NEXT:    [[Q0:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET0]]
; CHECK-NEXT:    [[X0:%.*]] = load i64, ptr [[Q0]], align 8
; CHECK-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; CHECK-NEXT:    [[Q1:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET1]]
; CHECK-NEXT:    [[X1:%.*]] = load i64, ptr [[Q1]], align 8
; CHECK-NEXT:    [[RES:%.*]] = add i64 [[X0]], [[X1]]
; CHECK-NEXT:    [[DST:%.*]] = getelementptr i64, ptr [[Q]], i64 [[I]]
; CHECK-NEXT:    store i64 [[RES]], ptr [[DST]], align 8
; CHECK-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; CHECK-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; CHECK-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP15:![0-9]+]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @combine_load_factor2_i64(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; FIXED:       vector.ph:
; FIXED-NEXT:    br label [[VECTOR_BODY:%.*]]
; FIXED:       vector.body:
; FIXED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; FIXED-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
; FIXED-NEXT:    [[TMP1:%.*]] = add i64 [[INDEX]], 4
; FIXED-NEXT:    [[TMP2:%.*]] = shl i64 [[TMP0]], 1
; FIXED-NEXT:    [[TMP3:%.*]] = shl i64 [[TMP1]], 1
; FIXED-NEXT:    [[TMP4:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP2]]
; FIXED-NEXT:    [[TMP5:%.*]] = getelementptr i64, ptr [[P]], i64 [[TMP3]]
; FIXED-NEXT:    [[TMP6:%.*]] = getelementptr i64, ptr [[TMP4]], i32 0
; FIXED-NEXT:    [[TMP7:%.*]] = getelementptr i64, ptr [[TMP5]], i32 0
; FIXED-NEXT:    [[WIDE_VEC:%.*]] = load <8 x i64>, ptr [[TMP6]], align 8
; FIXED-NEXT:    [[WIDE_VEC1:%.*]] = load <8 x i64>, ptr [[TMP7]], align 8
; FIXED-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <8 x i64> [[WIDE_VEC]], <8 x i64> poison, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
; FIXED-NEXT:    [[STRIDED_VEC2:%.*]] = shufflevector <8 x i64> [[WIDE_VEC1]], <8 x i64> poison, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
; FIXED-NEXT:    [[STRIDED_VEC3:%.*]] = shufflevector <8 x i64> [[WIDE_VEC]], <8 x i64> poison, <4 x i32> <i32 1, i32 3, i32 5, i32 7>
; FIXED-NEXT:    [[STRIDED_VEC4:%.*]] = shufflevector <8 x i64> [[WIDE_VEC1]], <8 x i64> poison, <4 x i32> <i32 1, i32 3, i32 5, i32 7>
; FIXED-NEXT:    [[TMP8:%.*]] = add <4 x i64> [[STRIDED_VEC]], [[STRIDED_VEC3]]
; FIXED-NEXT:    [[TMP9:%.*]] = add <4 x i64> [[STRIDED_VEC2]], [[STRIDED_VEC4]]
; FIXED-NEXT:    [[TMP10:%.*]] = getelementptr i64, ptr [[Q:%.*]], i64 [[TMP0]]
; FIXED-NEXT:    [[TMP11:%.*]] = getelementptr i64, ptr [[Q]], i64 [[TMP1]]
; FIXED-NEXT:    [[TMP12:%.*]] = getelementptr i64, ptr [[TMP10]], i32 0
; FIXED-NEXT:    [[TMP13:%.*]] = getelementptr i64, ptr [[TMP10]], i32 4
; FIXED-NEXT:    store <4 x i64> [[TMP8]], ptr [[TMP12]], align 8
; FIXED-NEXT:    store <4 x i64> [[TMP9]], ptr [[TMP13]], align 8
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
; FIXED-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP14:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
; FIXED:       scalar.ph:
; FIXED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1024, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       loop:
; FIXED-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; FIXED-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 1
; FIXED-NEXT:    [[Q0:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET0]]
; FIXED-NEXT:    [[X0:%.*]] = load i64, ptr [[Q0]], align 8
; FIXED-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; FIXED-NEXT:    [[Q1:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET1]]
; FIXED-NEXT:    [[X1:%.*]] = load i64, ptr [[Q1]], align 8
; FIXED-NEXT:    [[RES:%.*]] = add i64 [[X0]], [[X1]]
; FIXED-NEXT:    [[DST:%.*]] = getelementptr i64, ptr [[Q]], i64 [[I]]
; FIXED-NEXT:    store i64 [[RES]], ptr [[DST]], align 8
; FIXED-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; FIXED-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; FIXED-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP15:![0-9]+]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @combine_load_factor2_i64(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
; SCALABLE-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1024, [[TMP1]]
; SCALABLE-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; SCALABLE:       vector.ph:
; SCALABLE-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 2
; SCALABLE-NEXT:    [[N_MOD_VF:%.*]] = urem i64 1024, [[TMP3]]
; SCALABLE-NEXT:    [[N_VEC:%.*]] = sub i64 1024, [[N_MOD_VF]]
; SCALABLE-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
; SCALABLE-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
; SCALABLE-NEXT:    br label [[VECTOR_BODY:%.*]]
; SCALABLE:       vector.body:
; SCALABLE-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP6:%.*]] = add i64 [[INDEX]], 0
; SCALABLE-NEXT:    [[TMP7:%.*]] = shl i64 [[TMP6]], 1
; SCALABLE-NEXT:    [[TMP8:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP7]]
; SCALABLE-NEXT:    [[TMP9:%.*]] = getelementptr i64, ptr [[TMP8]], i32 0
; SCALABLE-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 4 x i64>, ptr [[TMP9]], align 8
; SCALABLE-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> [[WIDE_VEC]])
; SCALABLE-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 0
; SCALABLE-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP12:%.*]] = add <vscale x 2 x i64> [[TMP10]], [[TMP11]]
; SCALABLE-NEXT:    [[TMP13:%.*]] = getelementptr i64, ptr [[Q:%.*]], i64 [[TMP6]]
; SCALABLE-NEXT:    [[TMP14:%.*]] = getelementptr i64, ptr [[TMP13]], i32 0
; SCALABLE-NEXT:    store <vscale x 2 x i64> [[TMP12]], ptr [[TMP14]], align 8
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]]
; SCALABLE-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; SCALABLE-NEXT:    br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP14:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]]
; SCALABLE-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
; SCALABLE:       scalar.ph:
; SCALABLE-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; SCALABLE:       loop:
; SCALABLE-NEXT:    [[I:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[NEXTI:%.*]], [[LOOP]] ]
; SCALABLE-NEXT:    [[OFFSET0:%.*]] = shl i64 [[I]], 1
; SCALABLE-NEXT:    [[Q0:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET0]]
; SCALABLE-NEXT:    [[X0:%.*]] = load i64, ptr [[Q0]], align 8
; SCALABLE-NEXT:    [[OFFSET1:%.*]] = add i64 [[OFFSET0]], 1
; SCALABLE-NEXT:    [[Q1:%.*]] = getelementptr i64, ptr [[P]], i64 [[OFFSET1]]
; SCALABLE-NEXT:    [[X1:%.*]] = load i64, ptr [[Q1]], align 8
; SCALABLE-NEXT:    [[RES:%.*]] = add i64 [[X0]], [[X1]]
; SCALABLE-NEXT:    [[DST:%.*]] = getelementptr i64, ptr [[Q]], i64 [[I]]
; SCALABLE-NEXT:    store i64 [[RES]], ptr [[DST]], align 8
; SCALABLE-NEXT:    [[NEXTI]] = add i64 [[I]], 1
; SCALABLE-NEXT:    [[DONE:%.*]] = icmp eq i64 [[NEXTI]], 1024
; SCALABLE-NEXT:    br i1 [[DONE]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP15:![0-9]+]]
; SCALABLE:       exit:
; SCALABLE-NEXT:    ret void
;
entry:
  br label %loop
loop:
  %i = phi i64 [0, %entry], [%nexti, %loop]

  %offset0 = shl i64 %i, 1
  %q0 = getelementptr i64, ptr %p, i64 %offset0
  %x0 = load i64, ptr %q0

  %offset1 = add i64 %offset0, 1
  %q1 = getelementptr i64, ptr %p, i64 %offset1
  %x1 = load i64, ptr %q1

  %res = add i64 %x0, %x1

  %dst = getelementptr i64, ptr %q, i64 %i
  store i64 %res, ptr %dst

  %nexti = add i64 %i, 1
  %done = icmp eq i64 %nexti, 1024
  br i1 %done, label %exit, label %loop
exit:
  ret void
}