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:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 4, i1 true)
; CHECK-NEXT:    [[TMP14:%.*]] = shl i64 [[INDEX]], 1
; CHECK-NEXT:    [[TMP15:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP14]]
; CHECK-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 2
; CHECK-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.vp.load.nxv8i32.p0(ptr align 4 [[TMP15]], <vscale x 8 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; 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_MASKED_VEC]])
; CHECK-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
; CHECK-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP10:%.*]] = add <vscale x 4 x i32> [[TMP8]], splat (i32 1)
; CHECK-NEXT:    [[TMP11:%.*]] = add <vscale x 4 x i32> [[TMP9]], splat (i32 2)
; CHECK-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 2
; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> [[TMP10]], <vscale x 4 x i32> [[TMP11]])
; CHECK-NEXT:    call void @llvm.vp.store.nxv8i32.p0(<vscale x 8 x i32> [[INTERLEAVED_VEC]], ptr align 4 [[TMP15]], <vscale x 8 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; CHECK-NEXT:    [[TMP16:%.*]] = zext i32 [[TMP7]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP16]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP16]]
; CHECK-NEXT:    [[TMP12:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @load_store_factor2_i32(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br 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:%.*]] = shl i64 [[INDEX]], 1
; FIXED-NEXT:    [[TMP1:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP0]]
; FIXED-NEXT:    [[WIDE_VEC:%.*]] = load <16 x i32>, ptr [[TMP1]], 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:    [[TMP2:%.*]] = add <8 x i32> [[STRIDED_VEC]], splat (i32 1)
; FIXED-NEXT:    [[TMP3:%.*]] = add <8 x i32> [[STRIDED_VEC1]], splat (i32 2)
; FIXED-NEXT:    [[TMP4:%.*]] = shufflevector <8 x i32> [[TMP2]], <8 x i32> [[TMP3]], <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> [[TMP4]], <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 [[TMP1]], align 4
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
; FIXED-NEXT:    [[TMP5:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP5]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @load_store_factor2_i32(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 4, i1 true)
; SCALABLE-NEXT:    [[TMP14:%.*]] = shl i64 [[INDEX]], 1
; SCALABLE-NEXT:    [[TMP15:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP14]]
; SCALABLE-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 2
; SCALABLE-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.vp.load.nxv8i32.p0(ptr align 4 [[TMP15]], <vscale x 8 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; 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_MASKED_VEC]])
; SCALABLE-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
; SCALABLE-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP10:%.*]] = add <vscale x 4 x i32> [[TMP8]], splat (i32 1)
; SCALABLE-NEXT:    [[TMP11:%.*]] = add <vscale x 4 x i32> [[TMP9]], splat (i32 2)
; SCALABLE-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 2
; SCALABLE-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> [[TMP10]], <vscale x 4 x i32> [[TMP11]])
; SCALABLE-NEXT:    call void @llvm.vp.store.nxv8i32.p0(<vscale x 8 x i32> [[INTERLEAVED_VEC]], ptr align 4 [[TMP15]], <vscale x 8 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; SCALABLE-NEXT:    [[TMP16:%.*]] = zext i32 [[TMP7]] to i64
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP16]], [[INDEX]]
; SCALABLE-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP16]]
; SCALABLE-NEXT:    [[TMP12:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; SCALABLE-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; 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:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 2, i1 true)
; CHECK-NEXT:    [[TMP8:%.*]] = shl i64 [[INDEX]], 1
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP8]]
; CHECK-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 2
; CHECK-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 4 x i64> @llvm.vp.load.nxv4i64.p0(ptr align 8 [[TMP14]], <vscale x 4 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; 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_MASKED_VEC]])
; CHECK-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 0
; CHECK-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP13:%.*]] = add <vscale x 2 x i64> [[TMP20]], splat (i64 1)
; CHECK-NEXT:    [[TMP11:%.*]] = add <vscale x 2 x i64> [[TMP9]], splat (i64 2)
; CHECK-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 2
; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64> [[TMP13]], <vscale x 2 x i64> [[TMP11]])
; CHECK-NEXT:    call void @llvm.vp.store.nxv4i64.p0(<vscale x 4 x i64> [[INTERLEAVED_VEC]], ptr align 8 [[TMP14]], <vscale x 4 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; CHECK-NEXT:    [[TMP16:%.*]] = zext i32 [[TMP7]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP16]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP16]]
; CHECK-NEXT:    [[TMP12:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @load_store_factor2_i64(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br 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:%.*]] = shl i64 [[INDEX]], 1
; FIXED-NEXT:    [[TMP1:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP0]]
; FIXED-NEXT:    [[WIDE_VEC:%.*]] = load <8 x i64>, ptr [[TMP1]], 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:    [[TMP2:%.*]] = add <4 x i64> [[STRIDED_VEC]], splat (i64 1)
; FIXED-NEXT:    [[TMP3:%.*]] = add <4 x i64> [[STRIDED_VEC1]], splat (i64 2)
; FIXED-NEXT:    [[TMP4:%.*]] = shufflevector <4 x i64> [[TMP2]], <4 x i64> [[TMP3]], <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> [[TMP4]], <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 [[TMP1]], align 8
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; FIXED-NEXT:    [[TMP5:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP5]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @load_store_factor2_i64(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 2, i1 true)
; SCALABLE-NEXT:    [[TMP8:%.*]] = shl i64 [[INDEX]], 1
; SCALABLE-NEXT:    [[TMP14:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP8]]
; SCALABLE-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 2
; SCALABLE-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 4 x i64> @llvm.vp.load.nxv4i64.p0(ptr align 8 [[TMP14]], <vscale x 4 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; 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_MASKED_VEC]])
; SCALABLE-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 0
; SCALABLE-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP13:%.*]] = add <vscale x 2 x i64> [[TMP20]], splat (i64 1)
; SCALABLE-NEXT:    [[TMP11:%.*]] = add <vscale x 2 x i64> [[TMP9]], splat (i64 2)
; SCALABLE-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 2
; SCALABLE-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64> [[TMP13]], <vscale x 2 x i64> [[TMP11]])
; SCALABLE-NEXT:    call void @llvm.vp.store.nxv4i64.p0(<vscale x 4 x i64> [[INTERLEAVED_VEC]], ptr align 8 [[TMP14]], <vscale x 4 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; SCALABLE-NEXT:    [[TMP16:%.*]] = zext i32 [[TMP7]] to i64
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP16]], [[INDEX]]
; SCALABLE-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP16]]
; SCALABLE-NEXT:    [[TMP12:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; SCALABLE-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; 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 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 4, i1 true)
; CHECK-NEXT:    [[TMP16:%.*]] = mul i64 [[INDEX]], 3
; CHECK-NEXT:    [[TMP17:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP16]]
; CHECK-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 3
; CHECK-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 12 x i32> @llvm.vp.load.nxv12i32.p0(ptr align 4 [[TMP17]], <vscale x 12 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave3.nxv12i32(<vscale x 12 x i32> [[WIDE_MASKED_VEC]])
; CHECK-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
; CHECK-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 2
; CHECK-NEXT:    [[TMP11:%.*]] = add <vscale x 4 x i32> [[TMP8]], splat (i32 1)
; CHECK-NEXT:    [[TMP12:%.*]] = add <vscale x 4 x i32> [[TMP9]], splat (i32 2)
; CHECK-NEXT:    [[TMP13:%.*]] = add <vscale x 4 x i32> [[TMP10]], splat (i32 3)
; CHECK-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 3
; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 12 x i32> @llvm.vector.interleave3.nxv12i32(<vscale x 4 x i32> [[TMP11]], <vscale x 4 x i32> [[TMP12]], <vscale x 4 x i32> [[TMP13]])
; CHECK-NEXT:    call void @llvm.vp.store.nxv12i32.p0(<vscale x 12 x i32> [[INTERLEAVED_VEC]], ptr align 4 [[TMP17]], <vscale x 12 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; CHECK-NEXT:    [[TMP19:%.*]] = zext i32 [[TMP7]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP19]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP19]]
; CHECK-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @load_store_factor3_i32(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br 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:%.*]] = mul i64 [[INDEX]], 3
; FIXED-NEXT:    [[TMP1:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP0]]
; FIXED-NEXT:    [[WIDE_VEC:%.*]] = load <24 x i32>, ptr [[TMP1]], align 4
; FIXED-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <24 x i32> [[WIDE_VEC]], <24 x i32> poison, <8 x i32> <i32 0, i32 3, i32 6, i32 9, i32 12, i32 15, i32 18, i32 21>
; FIXED-NEXT:    [[STRIDED_VEC1:%.*]] = shufflevector <24 x i32> [[WIDE_VEC]], <24 x i32> poison, <8 x i32> <i32 1, i32 4, i32 7, i32 10, i32 13, i32 16, i32 19, i32 22>
; FIXED-NEXT:    [[STRIDED_VEC2:%.*]] = shufflevector <24 x i32> [[WIDE_VEC]], <24 x i32> poison, <8 x i32> <i32 2, i32 5, i32 8, i32 11, i32 14, i32 17, i32 20, i32 23>
; FIXED-NEXT:    [[TMP2:%.*]] = add <8 x i32> [[STRIDED_VEC]], splat (i32 1)
; FIXED-NEXT:    [[TMP3:%.*]] = add <8 x i32> [[STRIDED_VEC1]], splat (i32 2)
; FIXED-NEXT:    [[TMP4:%.*]] = add <8 x i32> [[STRIDED_VEC2]], splat (i32 3)
; FIXED-NEXT:    [[TMP5:%.*]] = shufflevector <8 x i32> [[TMP2]], <8 x i32> [[TMP3]], <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:    [[TMP6:%.*]] = shufflevector <8 x i32> [[TMP4]], <8 x i32> poison, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
; FIXED-NEXT:    [[TMP7:%.*]] = shufflevector <16 x i32> [[TMP5]], <16 x i32> [[TMP6]], <24 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, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23>
; FIXED-NEXT:    [[INTERLEAVED_VEC:%.*]] = shufflevector <24 x i32> [[TMP7]], <24 x i32> poison, <24 x i32> <i32 0, i32 8, i32 16, i32 1, i32 9, i32 17, i32 2, i32 10, i32 18, i32 3, i32 11, i32 19, i32 4, i32 12, i32 20, i32 5, i32 13, i32 21, i32 6, i32 14, i32 22, i32 7, i32 15, i32 23>
; FIXED-NEXT:    store <24 x i32> [[INTERLEAVED_VEC]], ptr [[TMP1]], align 4
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
; FIXED-NEXT:    [[TMP8:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP8]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @load_store_factor3_i32(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 4, i1 true)
; SCALABLE-NEXT:    [[TMP16:%.*]] = mul i64 [[INDEX]], 3
; SCALABLE-NEXT:    [[TMP17:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP16]]
; SCALABLE-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 3
; SCALABLE-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 12 x i32> @llvm.vp.load.nxv12i32.p0(ptr align 4 [[TMP17]], <vscale x 12 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; SCALABLE-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave3.nxv12i32(<vscale x 12 x i32> [[WIDE_MASKED_VEC]])
; SCALABLE-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
; SCALABLE-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 2
; SCALABLE-NEXT:    [[TMP11:%.*]] = add <vscale x 4 x i32> [[TMP8]], splat (i32 1)
; SCALABLE-NEXT:    [[TMP12:%.*]] = add <vscale x 4 x i32> [[TMP9]], splat (i32 2)
; SCALABLE-NEXT:    [[TMP13:%.*]] = add <vscale x 4 x i32> [[TMP10]], splat (i32 3)
; SCALABLE-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 3
; SCALABLE-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 12 x i32> @llvm.vector.interleave3.nxv12i32(<vscale x 4 x i32> [[TMP11]], <vscale x 4 x i32> [[TMP12]], <vscale x 4 x i32> [[TMP13]])
; SCALABLE-NEXT:    call void @llvm.vp.store.nxv12i32.p0(<vscale x 12 x i32> [[INTERLEAVED_VEC]], ptr align 4 [[TMP17]], <vscale x 12 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; SCALABLE-NEXT:    [[TMP19:%.*]] = zext i32 [[TMP7]] to i64
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP19]], [[INDEX]]
; SCALABLE-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP19]]
; SCALABLE-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; SCALABLE-NEXT:    br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; 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:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 2, i1 true)
; CHECK-NEXT:    [[TMP8:%.*]] = mul i64 [[INDEX]], 3
; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP8]]
; CHECK-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 3
; CHECK-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 6 x i64> @llvm.vp.load.nxv6i64.p0(ptr align 8 [[TMP14]], <vscale x 6 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave3.nxv6i64(<vscale x 6 x i64> [[WIDE_MASKED_VEC]])
; CHECK-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 0
; CHECK-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 2
; CHECK-NEXT:    [[TMP25:%.*]] = add <vscale x 2 x i64> [[TMP23]], splat (i64 1)
; CHECK-NEXT:    [[TMP12:%.*]] = add <vscale x 2 x i64> [[TMP9]], splat (i64 2)
; CHECK-NEXT:    [[TMP13:%.*]] = add <vscale x 2 x i64> [[TMP10]], splat (i64 3)
; CHECK-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 3
; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 6 x i64> @llvm.vector.interleave3.nxv6i64(<vscale x 2 x i64> [[TMP25]], <vscale x 2 x i64> [[TMP12]], <vscale x 2 x i64> [[TMP13]])
; CHECK-NEXT:    call void @llvm.vp.store.nxv6i64.p0(<vscale x 6 x i64> [[INTERLEAVED_VEC]], ptr align 8 [[TMP14]], <vscale x 6 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; CHECK-NEXT:    [[TMP19:%.*]] = zext i32 [[TMP7]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP19]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP19]]
; CHECK-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @load_store_factor3_i64(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br 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:%.*]] = mul i64 [[INDEX]], 3
; FIXED-NEXT:    [[TMP1:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP0]]
; FIXED-NEXT:    [[WIDE_VEC:%.*]] = load <12 x i64>, ptr [[TMP1]], align 8
; FIXED-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <12 x i64> [[WIDE_VEC]], <12 x i64> poison, <4 x i32> <i32 0, i32 3, i32 6, i32 9>
; FIXED-NEXT:    [[STRIDED_VEC1:%.*]] = shufflevector <12 x i64> [[WIDE_VEC]], <12 x i64> poison, <4 x i32> <i32 1, i32 4, i32 7, i32 10>
; FIXED-NEXT:    [[STRIDED_VEC2:%.*]] = shufflevector <12 x i64> [[WIDE_VEC]], <12 x i64> poison, <4 x i32> <i32 2, i32 5, i32 8, i32 11>
; FIXED-NEXT:    [[TMP2:%.*]] = add <4 x i64> [[STRIDED_VEC]], splat (i64 1)
; FIXED-NEXT:    [[TMP3:%.*]] = add <4 x i64> [[STRIDED_VEC1]], splat (i64 2)
; FIXED-NEXT:    [[TMP4:%.*]] = add <4 x i64> [[STRIDED_VEC2]], splat (i64 3)
; FIXED-NEXT:    [[TMP5:%.*]] = shufflevector <4 x i64> [[TMP2]], <4 x i64> [[TMP3]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
; FIXED-NEXT:    [[TMP6:%.*]] = shufflevector <4 x i64> [[TMP4]], <4 x i64> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison>
; FIXED-NEXT:    [[TMP7:%.*]] = shufflevector <8 x i64> [[TMP5]], <8 x i64> [[TMP6]], <12 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>
; FIXED-NEXT:    [[INTERLEAVED_VEC:%.*]] = shufflevector <12 x i64> [[TMP7]], <12 x i64> poison, <12 x i32> <i32 0, i32 4, i32 8, i32 1, i32 5, i32 9, i32 2, i32 6, i32 10, i32 3, i32 7, i32 11>
; FIXED-NEXT:    store <12 x i64> [[INTERLEAVED_VEC]], ptr [[TMP1]], align 8
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; FIXED-NEXT:    [[TMP8:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP8]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @load_store_factor3_i64(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 2, i1 true)
; SCALABLE-NEXT:    [[TMP8:%.*]] = mul i64 [[INDEX]], 3
; SCALABLE-NEXT:    [[TMP14:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP8]]
; SCALABLE-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 3
; SCALABLE-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 6 x i64> @llvm.vp.load.nxv6i64.p0(ptr align 8 [[TMP14]], <vscale x 6 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; SCALABLE-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave3.nxv6i64(<vscale x 6 x i64> [[WIDE_MASKED_VEC]])
; SCALABLE-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 0
; SCALABLE-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 2
; SCALABLE-NEXT:    [[TMP25:%.*]] = add <vscale x 2 x i64> [[TMP23]], splat (i64 1)
; SCALABLE-NEXT:    [[TMP12:%.*]] = add <vscale x 2 x i64> [[TMP9]], splat (i64 2)
; SCALABLE-NEXT:    [[TMP13:%.*]] = add <vscale x 2 x i64> [[TMP10]], splat (i64 3)
; SCALABLE-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 3
; SCALABLE-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 6 x i64> @llvm.vector.interleave3.nxv6i64(<vscale x 2 x i64> [[TMP25]], <vscale x 2 x i64> [[TMP12]], <vscale x 2 x i64> [[TMP13]])
; SCALABLE-NEXT:    call void @llvm.vp.store.nxv6i64.p0(<vscale x 6 x i64> [[INTERLEAVED_VEC]], ptr align 8 [[TMP14]], <vscale x 6 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; SCALABLE-NEXT:    [[TMP19:%.*]] = zext i32 [[TMP7]] to i64
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP19]], [[INDEX]]
; SCALABLE-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP19]]
; SCALABLE-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; SCALABLE-NEXT:    br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; 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_factor4(ptr %p) {
; CHECK-LABEL: @load_store_factor4(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 2, i1 true)
; CHECK-NEXT:    [[TMP8:%.*]] = mul i64 [[INDEX]], 4
; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP8]]
; CHECK-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 4
; CHECK-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 8 x i64> @llvm.vp.load.nxv8i64.p0(ptr align 8 [[TMP9]], <vscale x 8 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave4.nxv8i64(<vscale x 8 x i64> [[WIDE_MASKED_VEC]])
; CHECK-NEXT:    [[TMP24:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <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>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 2
; CHECK-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 3
; CHECK-NEXT:    [[TMP26:%.*]] = add <vscale x 2 x i64> [[TMP24]], splat (i64 1)
; CHECK-NEXT:    [[TMP15:%.*]] = add <vscale x 2 x i64> [[TMP11]], splat (i64 2)
; CHECK-NEXT:    [[TMP16:%.*]] = add <vscale x 2 x i64> [[TMP12]], splat (i64 3)
; CHECK-NEXT:    [[TMP17:%.*]] = add <vscale x 2 x i64> [[TMP13]], splat (i64 4)
; CHECK-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 4
; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i64> @llvm.vector.interleave4.nxv8i64(<vscale x 2 x i64> [[TMP26]], <vscale x 2 x i64> [[TMP15]], <vscale x 2 x i64> [[TMP16]], <vscale x 2 x i64> [[TMP17]])
; CHECK-NEXT:    call void @llvm.vp.store.nxv8i64.p0(<vscale x 8 x i64> [[INTERLEAVED_VEC]], ptr align 8 [[TMP9]], <vscale x 8 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; CHECK-NEXT:    [[TMP22:%.*]] = zext i32 [[TMP7]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP22]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP22]]
; CHECK-NEXT:    [[TMP18:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @load_store_factor4(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br 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:%.*]] = mul i64 [[INDEX]], 4
; FIXED-NEXT:    [[TMP1:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP0]]
; FIXED-NEXT:    [[WIDE_VEC:%.*]] = load <16 x i64>, ptr [[TMP1]], align 8
; FIXED-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <16 x i64> [[WIDE_VEC]], <16 x i64> poison, <4 x i32> <i32 0, i32 4, i32 8, i32 12>
; FIXED-NEXT:    [[STRIDED_VEC1:%.*]] = shufflevector <16 x i64> [[WIDE_VEC]], <16 x i64> poison, <4 x i32> <i32 1, i32 5, i32 9, i32 13>
; FIXED-NEXT:    [[STRIDED_VEC2:%.*]] = shufflevector <16 x i64> [[WIDE_VEC]], <16 x i64> poison, <4 x i32> <i32 2, i32 6, i32 10, i32 14>
; FIXED-NEXT:    [[STRIDED_VEC3:%.*]] = shufflevector <16 x i64> [[WIDE_VEC]], <16 x i64> poison, <4 x i32> <i32 3, i32 7, i32 11, i32 15>
; FIXED-NEXT:    [[TMP2:%.*]] = add <4 x i64> [[STRIDED_VEC]], splat (i64 1)
; FIXED-NEXT:    [[TMP3:%.*]] = add <4 x i64> [[STRIDED_VEC1]], splat (i64 2)
; FIXED-NEXT:    [[TMP4:%.*]] = add <4 x i64> [[STRIDED_VEC2]], splat (i64 3)
; FIXED-NEXT:    [[TMP5:%.*]] = add <4 x i64> [[STRIDED_VEC3]], splat (i64 4)
; FIXED-NEXT:    [[TMP6:%.*]] = shufflevector <4 x i64> [[TMP2]], <4 x i64> [[TMP3]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
; FIXED-NEXT:    [[TMP7:%.*]] = shufflevector <4 x i64> [[TMP4]], <4 x i64> [[TMP5]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
; FIXED-NEXT:    [[TMP8:%.*]] = shufflevector <8 x i64> [[TMP6]], <8 x i64> [[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 i64> [[TMP8]], <16 x i64> poison, <16 x i32> <i32 0, i32 4, i32 8, i32 12, i32 1, i32 5, i32 9, i32 13, i32 2, i32 6, i32 10, i32 14, i32 3, i32 7, i32 11, i32 15>
; FIXED-NEXT:    store <16 x i64> [[INTERLEAVED_VEC]], ptr [[TMP1]], align 8
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; FIXED-NEXT:    [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @load_store_factor4(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 2, i1 true)
; SCALABLE-NEXT:    [[TMP8:%.*]] = mul i64 [[INDEX]], 4
; SCALABLE-NEXT:    [[TMP9:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP8]]
; SCALABLE-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 4
; SCALABLE-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 8 x i64> @llvm.vp.load.nxv8i64.p0(ptr align 8 [[TMP9]], <vscale x 8 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; SCALABLE-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave4.nxv8i64(<vscale x 8 x i64> [[WIDE_MASKED_VEC]])
; SCALABLE-NEXT:    [[TMP24:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <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>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 2
; SCALABLE-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 3
; SCALABLE-NEXT:    [[TMP26:%.*]] = add <vscale x 2 x i64> [[TMP24]], splat (i64 1)
; SCALABLE-NEXT:    [[TMP15:%.*]] = add <vscale x 2 x i64> [[TMP11]], splat (i64 2)
; SCALABLE-NEXT:    [[TMP16:%.*]] = add <vscale x 2 x i64> [[TMP12]], splat (i64 3)
; SCALABLE-NEXT:    [[TMP17:%.*]] = add <vscale x 2 x i64> [[TMP13]], splat (i64 4)
; SCALABLE-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 4
; SCALABLE-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i64> @llvm.vector.interleave4.nxv8i64(<vscale x 2 x i64> [[TMP26]], <vscale x 2 x i64> [[TMP15]], <vscale x 2 x i64> [[TMP16]], <vscale x 2 x i64> [[TMP17]])
; SCALABLE-NEXT:    call void @llvm.vp.store.nxv8i64.p0(<vscale x 8 x i64> [[INTERLEAVED_VEC]], ptr align 8 [[TMP9]], <vscale x 8 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; SCALABLE-NEXT:    [[TMP22:%.*]] = zext i32 [[TMP7]] to i64
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP22]], [[INDEX]]
; SCALABLE-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP22]]
; SCALABLE-NEXT:    [[TMP18:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; SCALABLE-NEXT:    br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; SCALABLE:       exit:
; SCALABLE-NEXT:    ret void
;
entry:
  br label %loop
loop:
  %i = phi i64 [0, %entry], [%nexti, %loop]

  %offset0 = mul i64 %i, 4
  %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

  %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_factor5(ptr %p) {
; CHECK-LABEL: @load_store_factor5(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 1, i1 true)
; CHECK-NEXT:    [[TMP6:%.*]] = mul i64 [[INDEX]], 5
; CHECK-NEXT:    [[TMP19:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP6]]
; CHECK-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 5
; CHECK-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 5 x i64> @llvm.vp.load.nxv5i64.p0(ptr align 8 [[TMP19]], <vscale x 5 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.vector.deinterleave5.nxv5i64(<vscale x 5 x i64> [[WIDE_MASKED_VEC]])
; CHECK-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 0
; CHECK-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 2
; CHECK-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 3
; CHECK-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 4
; CHECK-NEXT:    [[TMP13:%.*]] = add <vscale x 1 x i64> [[TMP8]], splat (i64 1)
; CHECK-NEXT:    [[TMP14:%.*]] = add <vscale x 1 x i64> [[TMP9]], splat (i64 2)
; CHECK-NEXT:    [[TMP15:%.*]] = add <vscale x 1 x i64> [[TMP10]], splat (i64 3)
; CHECK-NEXT:    [[TMP16:%.*]] = add <vscale x 1 x i64> [[TMP11]], splat (i64 4)
; CHECK-NEXT:    [[TMP17:%.*]] = add <vscale x 1 x i64> [[TMP12]], splat (i64 5)
; CHECK-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 5
; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 5 x i64> @llvm.vector.interleave5.nxv5i64(<vscale x 1 x i64> [[TMP13]], <vscale x 1 x i64> [[TMP14]], <vscale x 1 x i64> [[TMP15]], <vscale x 1 x i64> [[TMP16]], <vscale x 1 x i64> [[TMP17]])
; CHECK-NEXT:    call void @llvm.vp.store.nxv5i64.p0(<vscale x 5 x i64> [[INTERLEAVED_VEC]], ptr align 8 [[TMP19]], <vscale x 5 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; CHECK-NEXT:    [[TMP25:%.*]] = zext i32 [[TMP7]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP25]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP25]]
; CHECK-NEXT:    [[TMP20:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP20]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @load_store_factor5(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br 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:%.*]] = mul i64 [[INDEX]], 5
; FIXED-NEXT:    [[TMP1:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP0]]
; FIXED-NEXT:    [[WIDE_VEC:%.*]] = load <10 x i64>, ptr [[TMP1]], align 8
; FIXED-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <10 x i64> [[WIDE_VEC]], <10 x i64> poison, <2 x i32> <i32 0, i32 5>
; FIXED-NEXT:    [[STRIDED_VEC1:%.*]] = shufflevector <10 x i64> [[WIDE_VEC]], <10 x i64> poison, <2 x i32> <i32 1, i32 6>
; FIXED-NEXT:    [[STRIDED_VEC2:%.*]] = shufflevector <10 x i64> [[WIDE_VEC]], <10 x i64> poison, <2 x i32> <i32 2, i32 7>
; FIXED-NEXT:    [[STRIDED_VEC3:%.*]] = shufflevector <10 x i64> [[WIDE_VEC]], <10 x i64> poison, <2 x i32> <i32 3, i32 8>
; FIXED-NEXT:    [[STRIDED_VEC4:%.*]] = shufflevector <10 x i64> [[WIDE_VEC]], <10 x i64> poison, <2 x i32> <i32 4, i32 9>
; FIXED-NEXT:    [[TMP2:%.*]] = add <2 x i64> [[STRIDED_VEC]], splat (i64 1)
; FIXED-NEXT:    [[TMP3:%.*]] = add <2 x i64> [[STRIDED_VEC1]], splat (i64 2)
; FIXED-NEXT:    [[TMP4:%.*]] = add <2 x i64> [[STRIDED_VEC2]], splat (i64 3)
; FIXED-NEXT:    [[TMP5:%.*]] = add <2 x i64> [[STRIDED_VEC3]], splat (i64 4)
; FIXED-NEXT:    [[TMP6:%.*]] = add <2 x i64> [[STRIDED_VEC4]], splat (i64 5)
; FIXED-NEXT:    [[TMP7:%.*]] = shufflevector <2 x i64> [[TMP2]], <2 x i64> [[TMP3]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; FIXED-NEXT:    [[TMP8:%.*]] = shufflevector <2 x i64> [[TMP4]], <2 x i64> [[TMP5]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; FIXED-NEXT:    [[TMP9:%.*]] = shufflevector <4 x i64> [[TMP7]], <4 x i64> [[TMP8]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
; FIXED-NEXT:    [[TMP10:%.*]] = shufflevector <2 x i64> [[TMP6]], <2 x i64> poison, <8 x i32> <i32 0, i32 1, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
; FIXED-NEXT:    [[TMP11:%.*]] = shufflevector <8 x i64> [[TMP9]], <8 x i64> [[TMP10]], <10 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9>
; FIXED-NEXT:    [[INTERLEAVED_VEC:%.*]] = shufflevector <10 x i64> [[TMP11]], <10 x i64> poison, <10 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 1, i32 3, i32 5, i32 7, i32 9>
; FIXED-NEXT:    store <10 x i64> [[INTERLEAVED_VEC]], ptr [[TMP1]], align 8
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; FIXED-NEXT:    [[TMP12:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @load_store_factor5(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 1, i1 true)
; SCALABLE-NEXT:    [[TMP6:%.*]] = mul i64 [[INDEX]], 5
; SCALABLE-NEXT:    [[TMP19:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP6]]
; SCALABLE-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 5
; SCALABLE-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 5 x i64> @llvm.vp.load.nxv5i64.p0(ptr align 8 [[TMP19]], <vscale x 5 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; SCALABLE-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.vector.deinterleave5.nxv5i64(<vscale x 5 x i64> [[WIDE_MASKED_VEC]])
; SCALABLE-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 0
; SCALABLE-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 2
; SCALABLE-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 3
; SCALABLE-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 4
; SCALABLE-NEXT:    [[TMP13:%.*]] = add <vscale x 1 x i64> [[TMP8]], splat (i64 1)
; SCALABLE-NEXT:    [[TMP14:%.*]] = add <vscale x 1 x i64> [[TMP9]], splat (i64 2)
; SCALABLE-NEXT:    [[TMP15:%.*]] = add <vscale x 1 x i64> [[TMP10]], splat (i64 3)
; SCALABLE-NEXT:    [[TMP16:%.*]] = add <vscale x 1 x i64> [[TMP11]], splat (i64 4)
; SCALABLE-NEXT:    [[TMP17:%.*]] = add <vscale x 1 x i64> [[TMP12]], splat (i64 5)
; SCALABLE-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 5
; SCALABLE-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 5 x i64> @llvm.vector.interleave5.nxv5i64(<vscale x 1 x i64> [[TMP13]], <vscale x 1 x i64> [[TMP14]], <vscale x 1 x i64> [[TMP15]], <vscale x 1 x i64> [[TMP16]], <vscale x 1 x i64> [[TMP17]])
; SCALABLE-NEXT:    call void @llvm.vp.store.nxv5i64.p0(<vscale x 5 x i64> [[INTERLEAVED_VEC]], ptr align 8 [[TMP19]], <vscale x 5 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; SCALABLE-NEXT:    [[TMP25:%.*]] = zext i32 [[TMP7]] to i64
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP25]], [[INDEX]]
; SCALABLE-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP25]]
; SCALABLE-NEXT:    [[TMP20:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; SCALABLE-NEXT:    br i1 [[TMP20]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; SCALABLE:       exit:
; SCALABLE-NEXT:    ret void
;
entry:
  br label %loop
loop:
  %i = phi i64 [0, %entry], [%nexti, %loop]

  %offset0 = mul i64 %i, 5
  %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

  %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_factor6(ptr %p) {
; CHECK-LABEL: @load_store_factor6(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 1, i1 true)
; CHECK-NEXT:    [[TMP6:%.*]] = mul i64 [[INDEX]], 6
; CHECK-NEXT:    [[TMP21:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP6]]
; CHECK-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 6
; CHECK-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 6 x i64> @llvm.vp.load.nxv6i64.p0(ptr align 8 [[TMP21]], <vscale x 6 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.vector.deinterleave6.nxv6i64(<vscale x 6 x i64> [[WIDE_MASKED_VEC]])
; CHECK-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 0
; CHECK-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 2
; CHECK-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 3
; CHECK-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 4
; CHECK-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 5
; CHECK-NEXT:    [[TMP14:%.*]] = add <vscale x 1 x i64> [[TMP8]], splat (i64 1)
; CHECK-NEXT:    [[TMP15:%.*]] = add <vscale x 1 x i64> [[TMP9]], splat (i64 2)
; CHECK-NEXT:    [[TMP16:%.*]] = add <vscale x 1 x i64> [[TMP10]], splat (i64 3)
; CHECK-NEXT:    [[TMP17:%.*]] = add <vscale x 1 x i64> [[TMP11]], splat (i64 4)
; CHECK-NEXT:    [[TMP18:%.*]] = add <vscale x 1 x i64> [[TMP12]], splat (i64 5)
; CHECK-NEXT:    [[TMP19:%.*]] = add <vscale x 1 x i64> [[TMP13]], splat (i64 6)
; CHECK-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 6
; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 6 x i64> @llvm.vector.interleave6.nxv6i64(<vscale x 1 x i64> [[TMP14]], <vscale x 1 x i64> [[TMP15]], <vscale x 1 x i64> [[TMP16]], <vscale x 1 x i64> [[TMP17]], <vscale x 1 x i64> [[TMP18]], <vscale x 1 x i64> [[TMP19]])
; CHECK-NEXT:    call void @llvm.vp.store.nxv6i64.p0(<vscale x 6 x i64> [[INTERLEAVED_VEC]], ptr align 8 [[TMP21]], <vscale x 6 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; CHECK-NEXT:    [[TMP28:%.*]] = zext i32 [[TMP7]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP28]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP28]]
; CHECK-NEXT:    [[TMP22:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP22]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP9:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @load_store_factor6(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br 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:%.*]] = mul i64 [[INDEX]], 6
; FIXED-NEXT:    [[TMP1:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP0]]
; FIXED-NEXT:    [[WIDE_VEC:%.*]] = load <12 x i64>, ptr [[TMP1]], align 8
; FIXED-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <12 x i64> [[WIDE_VEC]], <12 x i64> poison, <2 x i32> <i32 0, i32 6>
; FIXED-NEXT:    [[STRIDED_VEC1:%.*]] = shufflevector <12 x i64> [[WIDE_VEC]], <12 x i64> poison, <2 x i32> <i32 1, i32 7>
; FIXED-NEXT:    [[STRIDED_VEC2:%.*]] = shufflevector <12 x i64> [[WIDE_VEC]], <12 x i64> poison, <2 x i32> <i32 2, i32 8>
; FIXED-NEXT:    [[STRIDED_VEC3:%.*]] = shufflevector <12 x i64> [[WIDE_VEC]], <12 x i64> poison, <2 x i32> <i32 3, i32 9>
; FIXED-NEXT:    [[STRIDED_VEC4:%.*]] = shufflevector <12 x i64> [[WIDE_VEC]], <12 x i64> poison, <2 x i32> <i32 4, i32 10>
; FIXED-NEXT:    [[STRIDED_VEC5:%.*]] = shufflevector <12 x i64> [[WIDE_VEC]], <12 x i64> poison, <2 x i32> <i32 5, i32 11>
; FIXED-NEXT:    [[TMP2:%.*]] = add <2 x i64> [[STRIDED_VEC]], splat (i64 1)
; FIXED-NEXT:    [[TMP3:%.*]] = add <2 x i64> [[STRIDED_VEC1]], splat (i64 2)
; FIXED-NEXT:    [[TMP4:%.*]] = add <2 x i64> [[STRIDED_VEC2]], splat (i64 3)
; FIXED-NEXT:    [[TMP5:%.*]] = add <2 x i64> [[STRIDED_VEC3]], splat (i64 4)
; FIXED-NEXT:    [[TMP6:%.*]] = add <2 x i64> [[STRIDED_VEC4]], splat (i64 5)
; FIXED-NEXT:    [[TMP7:%.*]] = add <2 x i64> [[STRIDED_VEC5]], splat (i64 6)
; FIXED-NEXT:    [[TMP8:%.*]] = shufflevector <2 x i64> [[TMP2]], <2 x i64> [[TMP3]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; FIXED-NEXT:    [[TMP9:%.*]] = shufflevector <2 x i64> [[TMP4]], <2 x i64> [[TMP5]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; FIXED-NEXT:    [[TMP10:%.*]] = shufflevector <2 x i64> [[TMP6]], <2 x i64> [[TMP7]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; FIXED-NEXT:    [[TMP11:%.*]] = shufflevector <4 x i64> [[TMP8]], <4 x i64> [[TMP9]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
; FIXED-NEXT:    [[TMP12:%.*]] = shufflevector <4 x i64> [[TMP10]], <4 x i64> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison>
; FIXED-NEXT:    [[TMP13:%.*]] = shufflevector <8 x i64> [[TMP11]], <8 x i64> [[TMP12]], <12 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>
; FIXED-NEXT:    [[INTERLEAVED_VEC:%.*]] = shufflevector <12 x i64> [[TMP13]], <12 x i64> poison, <12 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 1, i32 3, i32 5, i32 7, i32 9, i32 11>
; FIXED-NEXT:    store <12 x i64> [[INTERLEAVED_VEC]], ptr [[TMP1]], align 8
; 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 [[LOOP8:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @load_store_factor6(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 1, i1 true)
; SCALABLE-NEXT:    [[TMP6:%.*]] = mul i64 [[INDEX]], 6
; SCALABLE-NEXT:    [[TMP21:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP6]]
; SCALABLE-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 6
; SCALABLE-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 6 x i64> @llvm.vp.load.nxv6i64.p0(ptr align 8 [[TMP21]], <vscale x 6 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; SCALABLE-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.vector.deinterleave6.nxv6i64(<vscale x 6 x i64> [[WIDE_MASKED_VEC]])
; SCALABLE-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 0
; SCALABLE-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 2
; SCALABLE-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 3
; SCALABLE-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 4
; SCALABLE-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 5
; SCALABLE-NEXT:    [[TMP14:%.*]] = add <vscale x 1 x i64> [[TMP8]], splat (i64 1)
; SCALABLE-NEXT:    [[TMP15:%.*]] = add <vscale x 1 x i64> [[TMP9]], splat (i64 2)
; SCALABLE-NEXT:    [[TMP16:%.*]] = add <vscale x 1 x i64> [[TMP10]], splat (i64 3)
; SCALABLE-NEXT:    [[TMP17:%.*]] = add <vscale x 1 x i64> [[TMP11]], splat (i64 4)
; SCALABLE-NEXT:    [[TMP18:%.*]] = add <vscale x 1 x i64> [[TMP12]], splat (i64 5)
; SCALABLE-NEXT:    [[TMP19:%.*]] = add <vscale x 1 x i64> [[TMP13]], splat (i64 6)
; SCALABLE-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 6
; SCALABLE-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 6 x i64> @llvm.vector.interleave6.nxv6i64(<vscale x 1 x i64> [[TMP14]], <vscale x 1 x i64> [[TMP15]], <vscale x 1 x i64> [[TMP16]], <vscale x 1 x i64> [[TMP17]], <vscale x 1 x i64> [[TMP18]], <vscale x 1 x i64> [[TMP19]])
; SCALABLE-NEXT:    call void @llvm.vp.store.nxv6i64.p0(<vscale x 6 x i64> [[INTERLEAVED_VEC]], ptr align 8 [[TMP21]], <vscale x 6 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; SCALABLE-NEXT:    [[TMP28:%.*]] = zext i32 [[TMP7]] to i64
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP28]], [[INDEX]]
; SCALABLE-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP28]]
; SCALABLE-NEXT:    [[TMP22:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; SCALABLE-NEXT:    br i1 [[TMP22]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP9:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; SCALABLE:       exit:
; SCALABLE-NEXT:    ret void
;
entry:
  br label %loop
loop:
  %i = phi i64 [0, %entry], [%nexti, %loop]

  %offset0 = mul i64 %i, 6
  %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

  %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_factor7(ptr %p) {
; CHECK-LABEL: @load_store_factor7(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 1, i1 true)
; CHECK-NEXT:    [[TMP6:%.*]] = mul i64 [[INDEX]], 7
; CHECK-NEXT:    [[TMP23:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP6]]
; CHECK-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 7
; CHECK-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 7 x i64> @llvm.vp.load.nxv7i64.p0(ptr align 8 [[TMP23]], <vscale x 7 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.vector.deinterleave7.nxv7i64(<vscale x 7 x i64> [[WIDE_MASKED_VEC]])
; CHECK-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 0
; CHECK-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 2
; CHECK-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 3
; CHECK-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 4
; CHECK-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 5
; CHECK-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 6
; CHECK-NEXT:    [[TMP15:%.*]] = add <vscale x 1 x i64> [[TMP8]], splat (i64 1)
; CHECK-NEXT:    [[TMP16:%.*]] = add <vscale x 1 x i64> [[TMP9]], splat (i64 2)
; CHECK-NEXT:    [[TMP17:%.*]] = add <vscale x 1 x i64> [[TMP10]], splat (i64 3)
; CHECK-NEXT:    [[TMP18:%.*]] = add <vscale x 1 x i64> [[TMP11]], splat (i64 4)
; CHECK-NEXT:    [[TMP19:%.*]] = add <vscale x 1 x i64> [[TMP12]], splat (i64 5)
; CHECK-NEXT:    [[TMP20:%.*]] = add <vscale x 1 x i64> [[TMP13]], splat (i64 6)
; CHECK-NEXT:    [[TMP21:%.*]] = add <vscale x 1 x i64> [[TMP14]], splat (i64 7)
; CHECK-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 7
; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 7 x i64> @llvm.vector.interleave7.nxv7i64(<vscale x 1 x i64> [[TMP15]], <vscale x 1 x i64> [[TMP16]], <vscale x 1 x i64> [[TMP17]], <vscale x 1 x i64> [[TMP18]], <vscale x 1 x i64> [[TMP19]], <vscale x 1 x i64> [[TMP20]], <vscale x 1 x i64> [[TMP21]])
; CHECK-NEXT:    call void @llvm.vp.store.nxv7i64.p0(<vscale x 7 x i64> [[INTERLEAVED_VEC]], ptr align 8 [[TMP23]], <vscale x 7 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; CHECK-NEXT:    [[TMP31:%.*]] = zext i32 [[TMP7]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP31]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP31]]
; CHECK-NEXT:    [[TMP24:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP24]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @load_store_factor7(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br 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:%.*]] = mul i64 [[INDEX]], 7
; FIXED-NEXT:    [[TMP1:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP0]]
; FIXED-NEXT:    [[WIDE_VEC:%.*]] = load <14 x i64>, ptr [[TMP1]], align 8
; FIXED-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <14 x i64> [[WIDE_VEC]], <14 x i64> poison, <2 x i32> <i32 0, i32 7>
; FIXED-NEXT:    [[STRIDED_VEC1:%.*]] = shufflevector <14 x i64> [[WIDE_VEC]], <14 x i64> poison, <2 x i32> <i32 1, i32 8>
; FIXED-NEXT:    [[STRIDED_VEC2:%.*]] = shufflevector <14 x i64> [[WIDE_VEC]], <14 x i64> poison, <2 x i32> <i32 2, i32 9>
; FIXED-NEXT:    [[STRIDED_VEC3:%.*]] = shufflevector <14 x i64> [[WIDE_VEC]], <14 x i64> poison, <2 x i32> <i32 3, i32 10>
; FIXED-NEXT:    [[STRIDED_VEC4:%.*]] = shufflevector <14 x i64> [[WIDE_VEC]], <14 x i64> poison, <2 x i32> <i32 4, i32 11>
; FIXED-NEXT:    [[STRIDED_VEC5:%.*]] = shufflevector <14 x i64> [[WIDE_VEC]], <14 x i64> poison, <2 x i32> <i32 5, i32 12>
; FIXED-NEXT:    [[STRIDED_VEC6:%.*]] = shufflevector <14 x i64> [[WIDE_VEC]], <14 x i64> poison, <2 x i32> <i32 6, i32 13>
; FIXED-NEXT:    [[TMP2:%.*]] = add <2 x i64> [[STRIDED_VEC]], splat (i64 1)
; FIXED-NEXT:    [[TMP3:%.*]] = add <2 x i64> [[STRIDED_VEC1]], splat (i64 2)
; FIXED-NEXT:    [[TMP4:%.*]] = add <2 x i64> [[STRIDED_VEC2]], splat (i64 3)
; FIXED-NEXT:    [[TMP5:%.*]] = add <2 x i64> [[STRIDED_VEC3]], splat (i64 4)
; FIXED-NEXT:    [[TMP6:%.*]] = add <2 x i64> [[STRIDED_VEC4]], splat (i64 5)
; FIXED-NEXT:    [[TMP7:%.*]] = add <2 x i64> [[STRIDED_VEC5]], splat (i64 6)
; FIXED-NEXT:    [[TMP8:%.*]] = add <2 x i64> [[STRIDED_VEC6]], splat (i64 7)
; FIXED-NEXT:    [[TMP9:%.*]] = shufflevector <2 x i64> [[TMP2]], <2 x i64> [[TMP3]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; FIXED-NEXT:    [[TMP10:%.*]] = shufflevector <2 x i64> [[TMP4]], <2 x i64> [[TMP5]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; FIXED-NEXT:    [[TMP11:%.*]] = shufflevector <2 x i64> [[TMP6]], <2 x i64> [[TMP7]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; FIXED-NEXT:    [[TMP12:%.*]] = shufflevector <4 x i64> [[TMP9]], <4 x i64> [[TMP10]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
; FIXED-NEXT:    [[TMP13:%.*]] = shufflevector <2 x i64> [[TMP8]], <2 x i64> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
; FIXED-NEXT:    [[TMP14:%.*]] = shufflevector <4 x i64> [[TMP11]], <4 x i64> [[TMP13]], <6 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5>
; FIXED-NEXT:    [[TMP15:%.*]] = shufflevector <6 x i64> [[TMP14]], <6 x i64> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 poison, i32 poison>
; FIXED-NEXT:    [[TMP16:%.*]] = shufflevector <8 x i64> [[TMP12]], <8 x i64> [[TMP15]], <14 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>
; FIXED-NEXT:    [[INTERLEAVED_VEC:%.*]] = shufflevector <14 x i64> [[TMP16]], <14 x i64> poison, <14 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13>
; FIXED-NEXT:    store <14 x i64> [[INTERLEAVED_VEC]], ptr [[TMP1]], align 8
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; FIXED-NEXT:    [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP9:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @load_store_factor7(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 1, i1 true)
; SCALABLE-NEXT:    [[TMP6:%.*]] = mul i64 [[INDEX]], 7
; SCALABLE-NEXT:    [[TMP23:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP6]]
; SCALABLE-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 7
; SCALABLE-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 7 x i64> @llvm.vp.load.nxv7i64.p0(ptr align 8 [[TMP23]], <vscale x 7 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; SCALABLE-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.vector.deinterleave7.nxv7i64(<vscale x 7 x i64> [[WIDE_MASKED_VEC]])
; SCALABLE-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 0
; SCALABLE-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 2
; SCALABLE-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 3
; SCALABLE-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 4
; SCALABLE-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 5
; SCALABLE-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 6
; SCALABLE-NEXT:    [[TMP15:%.*]] = add <vscale x 1 x i64> [[TMP8]], splat (i64 1)
; SCALABLE-NEXT:    [[TMP16:%.*]] = add <vscale x 1 x i64> [[TMP9]], splat (i64 2)
; SCALABLE-NEXT:    [[TMP17:%.*]] = add <vscale x 1 x i64> [[TMP10]], splat (i64 3)
; SCALABLE-NEXT:    [[TMP18:%.*]] = add <vscale x 1 x i64> [[TMP11]], splat (i64 4)
; SCALABLE-NEXT:    [[TMP19:%.*]] = add <vscale x 1 x i64> [[TMP12]], splat (i64 5)
; SCALABLE-NEXT:    [[TMP20:%.*]] = add <vscale x 1 x i64> [[TMP13]], splat (i64 6)
; SCALABLE-NEXT:    [[TMP21:%.*]] = add <vscale x 1 x i64> [[TMP14]], splat (i64 7)
; SCALABLE-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 7
; SCALABLE-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 7 x i64> @llvm.vector.interleave7.nxv7i64(<vscale x 1 x i64> [[TMP15]], <vscale x 1 x i64> [[TMP16]], <vscale x 1 x i64> [[TMP17]], <vscale x 1 x i64> [[TMP18]], <vscale x 1 x i64> [[TMP19]], <vscale x 1 x i64> [[TMP20]], <vscale x 1 x i64> [[TMP21]])
; SCALABLE-NEXT:    call void @llvm.vp.store.nxv7i64.p0(<vscale x 7 x i64> [[INTERLEAVED_VEC]], ptr align 8 [[TMP23]], <vscale x 7 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; SCALABLE-NEXT:    [[TMP31:%.*]] = zext i32 [[TMP7]] to i64
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP31]], [[INDEX]]
; SCALABLE-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP31]]
; SCALABLE-NEXT:    [[TMP24:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; SCALABLE-NEXT:    br i1 [[TMP24]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; SCALABLE:       exit:
; SCALABLE-NEXT:    ret void
;
entry:
  br label %loop
loop:
  %i = phi i64 [0, %entry], [%nexti, %loop]

  %offset0 = mul i64 %i, 7
  %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

  %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:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 1, i1 true)
; CHECK-NEXT:    [[TMP6:%.*]] = shl i64 [[INDEX]], 3
; CHECK-NEXT:    [[TMP24:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP6]]
; CHECK-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 8
; CHECK-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 8 x i64> @llvm.vp.load.nxv8i64.p0(ptr align 8 [[TMP24]], <vscale x 8 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.vector.deinterleave8.nxv8i64(<vscale x 8 x i64> [[WIDE_MASKED_VEC]])
; CHECK-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 0
; CHECK-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 2
; CHECK-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 3
; CHECK-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 4
; CHECK-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 5
; CHECK-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 6
; CHECK-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 7
; CHECK-NEXT:    [[TMP16:%.*]] = add <vscale x 1 x i64> [[TMP8]], splat (i64 1)
; CHECK-NEXT:    [[TMP17:%.*]] = add <vscale x 1 x i64> [[TMP9]], splat (i64 2)
; CHECK-NEXT:    [[TMP18:%.*]] = add <vscale x 1 x i64> [[TMP10]], splat (i64 3)
; CHECK-NEXT:    [[TMP19:%.*]] = add <vscale x 1 x i64> [[TMP11]], splat (i64 4)
; CHECK-NEXT:    [[TMP20:%.*]] = add <vscale x 1 x i64> [[TMP12]], splat (i64 5)
; CHECK-NEXT:    [[TMP21:%.*]] = add <vscale x 1 x i64> [[TMP13]], splat (i64 6)
; CHECK-NEXT:    [[TMP22:%.*]] = add <vscale x 1 x i64> [[TMP14]], splat (i64 7)
; CHECK-NEXT:    [[TMP23:%.*]] = add <vscale x 1 x i64> [[TMP15]], splat (i64 8)
; CHECK-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 8
; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i64> @llvm.vector.interleave8.nxv8i64(<vscale x 1 x i64> [[TMP16]], <vscale x 1 x i64> [[TMP17]], <vscale x 1 x i64> [[TMP18]], <vscale x 1 x i64> [[TMP19]], <vscale x 1 x i64> [[TMP20]], <vscale x 1 x i64> [[TMP21]], <vscale x 1 x i64> [[TMP22]], <vscale x 1 x i64> [[TMP23]])
; CHECK-NEXT:    call void @llvm.vp.store.nxv8i64.p0(<vscale x 8 x i64> [[INTERLEAVED_VEC]], ptr align 8 [[TMP24]], <vscale x 8 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; CHECK-NEXT:    [[TMP34:%.*]] = zext i32 [[TMP7]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP34]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP34]]
; CHECK-NEXT:    [[TMP25:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP25]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP11:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @load_store_factor8(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br 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:%.*]] = shl i64 [[INDEX]], 3
; FIXED-NEXT:    [[TMP1:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP0]]
; FIXED-NEXT:    [[WIDE_VEC:%.*]] = load <16 x i64>, ptr [[TMP1]], align 8
; FIXED-NEXT:    [[STRIDED_VEC:%.*]] = shufflevector <16 x i64> [[WIDE_VEC]], <16 x i64> poison, <2 x i32> <i32 0, i32 8>
; FIXED-NEXT:    [[STRIDED_VEC1:%.*]] = shufflevector <16 x i64> [[WIDE_VEC]], <16 x i64> poison, <2 x i32> <i32 1, i32 9>
; FIXED-NEXT:    [[STRIDED_VEC2:%.*]] = shufflevector <16 x i64> [[WIDE_VEC]], <16 x i64> poison, <2 x i32> <i32 2, i32 10>
; FIXED-NEXT:    [[STRIDED_VEC3:%.*]] = shufflevector <16 x i64> [[WIDE_VEC]], <16 x i64> poison, <2 x i32> <i32 3, i32 11>
; FIXED-NEXT:    [[STRIDED_VEC4:%.*]] = shufflevector <16 x i64> [[WIDE_VEC]], <16 x i64> poison, <2 x i32> <i32 4, i32 12>
; FIXED-NEXT:    [[STRIDED_VEC5:%.*]] = shufflevector <16 x i64> [[WIDE_VEC]], <16 x i64> poison, <2 x i32> <i32 5, i32 13>
; FIXED-NEXT:    [[STRIDED_VEC6:%.*]] = shufflevector <16 x i64> [[WIDE_VEC]], <16 x i64> poison, <2 x i32> <i32 6, i32 14>
; FIXED-NEXT:    [[STRIDED_VEC7:%.*]] = shufflevector <16 x i64> [[WIDE_VEC]], <16 x i64> poison, <2 x i32> <i32 7, i32 15>
; FIXED-NEXT:    [[TMP2:%.*]] = add <2 x i64> [[STRIDED_VEC]], splat (i64 1)
; FIXED-NEXT:    [[TMP3:%.*]] = add <2 x i64> [[STRIDED_VEC1]], splat (i64 2)
; FIXED-NEXT:    [[TMP4:%.*]] = add <2 x i64> [[STRIDED_VEC2]], splat (i64 3)
; FIXED-NEXT:    [[TMP5:%.*]] = add <2 x i64> [[STRIDED_VEC3]], splat (i64 4)
; FIXED-NEXT:    [[TMP6:%.*]] = add <2 x i64> [[STRIDED_VEC4]], splat (i64 5)
; FIXED-NEXT:    [[TMP7:%.*]] = add <2 x i64> [[STRIDED_VEC5]], splat (i64 6)
; FIXED-NEXT:    [[TMP8:%.*]] = add <2 x i64> [[STRIDED_VEC6]], splat (i64 7)
; FIXED-NEXT:    [[TMP9:%.*]] = add <2 x i64> [[STRIDED_VEC7]], splat (i64 8)
; FIXED-NEXT:    [[TMP10:%.*]] = shufflevector <2 x i64> [[TMP2]], <2 x i64> [[TMP3]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; FIXED-NEXT:    [[TMP11:%.*]] = shufflevector <2 x i64> [[TMP4]], <2 x i64> [[TMP5]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; FIXED-NEXT:    [[TMP12:%.*]] = shufflevector <2 x i64> [[TMP6]], <2 x i64> [[TMP7]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; FIXED-NEXT:    [[TMP13:%.*]] = shufflevector <2 x i64> [[TMP8]], <2 x i64> [[TMP9]], <4 x i32> <i32 0, i32 1, i32 2, i32 3>
; FIXED-NEXT:    [[TMP14:%.*]] = shufflevector <4 x i64> [[TMP10]], <4 x i64> [[TMP11]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
; FIXED-NEXT:    [[TMP15:%.*]] = shufflevector <4 x i64> [[TMP12]], <4 x i64> [[TMP13]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
; FIXED-NEXT:    [[TMP16:%.*]] = shufflevector <8 x i64> [[TMP14]], <8 x i64> [[TMP15]], <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 i64> [[TMP16]], <16 x i64> poison, <16 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14, i32 1, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15>
; FIXED-NEXT:    store <16 x i64> [[INTERLEAVED_VEC]], ptr [[TMP1]], align 8
; FIXED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; FIXED-NEXT:    [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; FIXED-NEXT:    br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @load_store_factor8(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 1, i1 true)
; SCALABLE-NEXT:    [[TMP6:%.*]] = shl i64 [[INDEX]], 3
; SCALABLE-NEXT:    [[TMP24:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP6]]
; SCALABLE-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 8
; SCALABLE-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 8 x i64> @llvm.vp.load.nxv8i64.p0(ptr align 8 [[TMP24]], <vscale x 8 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; SCALABLE-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.vector.deinterleave8.nxv8i64(<vscale x 8 x i64> [[WIDE_MASKED_VEC]])
; SCALABLE-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 0
; SCALABLE-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 2
; SCALABLE-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 3
; SCALABLE-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 4
; SCALABLE-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 5
; SCALABLE-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 6
; SCALABLE-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64> } [[STRIDED_VEC]], 7
; SCALABLE-NEXT:    [[TMP16:%.*]] = add <vscale x 1 x i64> [[TMP8]], splat (i64 1)
; SCALABLE-NEXT:    [[TMP17:%.*]] = add <vscale x 1 x i64> [[TMP9]], splat (i64 2)
; SCALABLE-NEXT:    [[TMP18:%.*]] = add <vscale x 1 x i64> [[TMP10]], splat (i64 3)
; SCALABLE-NEXT:    [[TMP19:%.*]] = add <vscale x 1 x i64> [[TMP11]], splat (i64 4)
; SCALABLE-NEXT:    [[TMP20:%.*]] = add <vscale x 1 x i64> [[TMP12]], splat (i64 5)
; SCALABLE-NEXT:    [[TMP21:%.*]] = add <vscale x 1 x i64> [[TMP13]], splat (i64 6)
; SCALABLE-NEXT:    [[TMP22:%.*]] = add <vscale x 1 x i64> [[TMP14]], splat (i64 7)
; SCALABLE-NEXT:    [[TMP23:%.*]] = add <vscale x 1 x i64> [[TMP15]], splat (i64 8)
; SCALABLE-NEXT:    [[INTERLEAVE_EVL1:%.*]] = mul nuw nsw i32 [[TMP7]], 8
; SCALABLE-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i64> @llvm.vector.interleave8.nxv8i64(<vscale x 1 x i64> [[TMP16]], <vscale x 1 x i64> [[TMP17]], <vscale x 1 x i64> [[TMP18]], <vscale x 1 x i64> [[TMP19]], <vscale x 1 x i64> [[TMP20]], <vscale x 1 x i64> [[TMP21]], <vscale x 1 x i64> [[TMP22]], <vscale x 1 x i64> [[TMP23]])
; SCALABLE-NEXT:    call void @llvm.vp.store.nxv8i64.p0(<vscale x 8 x i64> [[INTERLEAVED_VEC]], ptr align 8 [[TMP24]], <vscale x 8 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL1]])
; SCALABLE-NEXT:    [[TMP34:%.*]] = zext i32 [[TMP7]] to i64
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP34]], [[INDEX]]
; SCALABLE-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP34]]
; SCALABLE-NEXT:    [[TMP25:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; SCALABLE-NEXT:    br i1 [[TMP25]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP11:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; 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:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 4, i1 true)
; CHECK-NEXT:    [[TMP13:%.*]] = shl i64 [[INDEX]], 1
; CHECK-NEXT:    [[TMP15:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP13]]
; CHECK-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 2
; CHECK-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.vp.load.nxv8i32.p0(ptr align 4 [[TMP15]], <vscale x 8 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; 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_MASKED_VEC]])
; CHECK-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
; CHECK-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP10:%.*]] = add <vscale x 4 x i32> [[TMP8]], [[TMP9]]
; CHECK-NEXT:    [[TMP11:%.*]] = getelementptr i32, ptr [[Q:%.*]], i64 [[INDEX]]
; CHECK-NEXT:    call void @llvm.vp.store.nxv4i32.p0(<vscale x 4 x i32> [[TMP10]], ptr align 4 [[TMP11]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP7]])
; CHECK-NEXT:    [[TMP16:%.*]] = zext i32 [[TMP7]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP16]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP16]]
; CHECK-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @combine_load_factor2_i32(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br 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:    [[TMP2:%.*]] = shl i64 [[INDEX]], 1
; FIXED-NEXT:    [[TMP4:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP2]]
; FIXED-NEXT:    [[WIDE_VEC2:%.*]] = load <16 x i32>, ptr [[TMP4]], align 4
; FIXED-NEXT:    [[STRIDED_VEC3:%.*]] = shufflevector <16 x i32> [[WIDE_VEC2]], <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_VEC4:%.*]] = shufflevector <16 x i32> [[WIDE_VEC2]], <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:    [[TMP6:%.*]] = add <8 x i32> [[STRIDED_VEC3]], [[STRIDED_VEC4]]
; FIXED-NEXT:    [[TMP7:%.*]] = getelementptr i32, ptr [[Q:%.*]], i64 [[INDEX]]
; FIXED-NEXT:    store <8 x i32> [[TMP6]], ptr [[TMP7]], 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 [[LOOP11:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @combine_load_factor2_i32(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 4, i1 true)
; SCALABLE-NEXT:    [[TMP13:%.*]] = shl i64 [[INDEX]], 1
; SCALABLE-NEXT:    [[TMP15:%.*]] = getelementptr i32, ptr [[P:%.*]], i64 [[TMP13]]
; SCALABLE-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 2
; SCALABLE-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.vp.load.nxv8i32.p0(ptr align 4 [[TMP15]], <vscale x 8 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; 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_MASKED_VEC]])
; SCALABLE-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
; SCALABLE-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP10:%.*]] = add <vscale x 4 x i32> [[TMP8]], [[TMP9]]
; SCALABLE-NEXT:    [[TMP11:%.*]] = getelementptr i32, ptr [[Q:%.*]], i64 [[INDEX]]
; SCALABLE-NEXT:    call void @llvm.vp.store.nxv4i32.p0(<vscale x 4 x i32> [[TMP10]], ptr align 4 [[TMP11]], <vscale x 4 x i1> splat (i1 true), i32 [[TMP7]])
; SCALABLE-NEXT:    [[TMP16:%.*]] = zext i32 [[TMP7]] to i64
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP16]], [[INDEX]]
; SCALABLE-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP16]]
; SCALABLE-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; SCALABLE-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; 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:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 2, i1 true)
; CHECK-NEXT:    [[TMP13:%.*]] = shl i64 [[INDEX]], 1
; CHECK-NEXT:    [[TMP15:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP13]]
; CHECK-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 2
; CHECK-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 4 x i64> @llvm.vp.load.nxv4i64.p0(ptr align 8 [[TMP15]], <vscale x 4 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; 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_MASKED_VEC]])
; CHECK-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 0
; CHECK-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 1
; CHECK-NEXT:    [[TMP10:%.*]] = add <vscale x 2 x i64> [[TMP8]], [[TMP9]]
; CHECK-NEXT:    [[TMP11:%.*]] = getelementptr i64, ptr [[Q:%.*]], i64 [[INDEX]]
; CHECK-NEXT:    call void @llvm.vp.store.nxv2i64.p0(<vscale x 2 x i64> [[TMP10]], ptr align 8 [[TMP11]], <vscale x 2 x i1> splat (i1 true), i32 [[TMP7]])
; CHECK-NEXT:    [[TMP16:%.*]] = zext i32 [[TMP7]] to i64
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP16]], [[INDEX]]
; CHECK-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP16]]
; CHECK-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; CHECK-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP13:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
; FIXED-LABEL: @combine_load_factor2_i64(
; FIXED-NEXT:  entry:
; FIXED-NEXT:    br 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:    [[TMP2:%.*]] = shl i64 [[INDEX]], 1
; FIXED-NEXT:    [[TMP4:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP2]]
; FIXED-NEXT:    [[WIDE_VEC2:%.*]] = load <8 x i64>, ptr [[TMP4]], align 8
; FIXED-NEXT:    [[STRIDED_VEC3:%.*]] = shufflevector <8 x i64> [[WIDE_VEC2]], <8 x i64> poison, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
; FIXED-NEXT:    [[STRIDED_VEC4:%.*]] = shufflevector <8 x i64> [[WIDE_VEC2]], <8 x i64> poison, <4 x i32> <i32 1, i32 3, i32 5, i32 7>
; FIXED-NEXT:    [[TMP6:%.*]] = add <4 x i64> [[STRIDED_VEC3]], [[STRIDED_VEC4]]
; FIXED-NEXT:    [[TMP7:%.*]] = getelementptr i64, ptr [[Q:%.*]], i64 [[INDEX]]
; FIXED-NEXT:    store <4 x i64> [[TMP6]], ptr [[TMP7]], 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 [[LOOP12:![0-9]+]]
; FIXED:       middle.block:
; FIXED-NEXT:    br label [[LOOP:%.*]]
; FIXED:       exit:
; FIXED-NEXT:    ret void
;
; SCALABLE-LABEL: @combine_load_factor2_i64(
; SCALABLE-NEXT:  entry:
; SCALABLE-NEXT:    br 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:    [[AVL:%.*]] = phi i64 [ 1024, [[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], [[VECTOR_BODY]] ]
; SCALABLE-NEXT:    [[TMP7:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 2, i1 true)
; SCALABLE-NEXT:    [[TMP13:%.*]] = shl i64 [[INDEX]], 1
; SCALABLE-NEXT:    [[TMP15:%.*]] = getelementptr i64, ptr [[P:%.*]], i64 [[TMP13]]
; SCALABLE-NEXT:    [[INTERLEAVE_EVL:%.*]] = mul nuw nsw i32 [[TMP7]], 2
; SCALABLE-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 4 x i64> @llvm.vp.load.nxv4i64.p0(ptr align 8 [[TMP15]], <vscale x 4 x i1> splat (i1 true), i32 [[INTERLEAVE_EVL]])
; 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_MASKED_VEC]])
; SCALABLE-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 0
; SCALABLE-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } [[STRIDED_VEC]], 1
; SCALABLE-NEXT:    [[TMP10:%.*]] = add <vscale x 2 x i64> [[TMP8]], [[TMP9]]
; SCALABLE-NEXT:    [[TMP11:%.*]] = getelementptr i64, ptr [[Q:%.*]], i64 [[INDEX]]
; SCALABLE-NEXT:    call void @llvm.vp.store.nxv2i64.p0(<vscale x 2 x i64> [[TMP10]], ptr align 8 [[TMP11]], <vscale x 2 x i1> splat (i1 true), i32 [[TMP7]])
; SCALABLE-NEXT:    [[TMP16:%.*]] = zext i32 [[TMP7]] to i64
; SCALABLE-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[TMP16]], [[INDEX]]
; SCALABLE-NEXT:    [[AVL_NEXT]] = sub nuw i64 [[AVL]], [[TMP16]]
; SCALABLE-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[AVL_NEXT]], 0
; SCALABLE-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP13:![0-9]+]]
; SCALABLE:       middle.block:
; SCALABLE-NEXT:    br label [[LOOP:%.*]]
; 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
}