; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -passes='require<profile-summary>,function(codegenprepare)' -S %s | FileCheck %s

target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
target triple = "arm64-apple-ios"

; It's profitable to convert the zext to a shuffle, which in turn will be
; lowered to 4 tbl instructions. The masks are materialized outside the loop.
define void @zext_v16i8_to_v16i32_in_loop(ptr %src, ptr %dst) {
; CHECK-LABEL: @zext_v16i8_to_v16i32_in_loop(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
; CHECK-NEXT:    [[SRC_GEP:%.*]] = getelementptr i8, ptr [[SRC:%.*]], i64 [[IV]]
; CHECK-NEXT:    [[LOAD:%.*]] = load <16 x i8>, ptr [[SRC_GEP]], align 16
; CHECK-NEXT:    [[TMP0:%.*]] = shufflevector <16 x i8> [[LOAD]], <16 x i8> <i8 0, i8 poison, i8 poison, i8 poison, i8 poison, i8 poison, i8 poison, i8 poison, i8 poison, i8 poison, i8 poison, i8 poison, i8 poison, i8 poison, i8 poison, i8 poison>, <64 x i32> <i32 0, i32 16, i32 16, i32 16, i32 1, i32 16, i32 16, i32 16, i32 2, i32 16, i32 16, i32 16, i32 3, i32 16, i32 16, i32 16, i32 4, i32 16, i32 16, i32 16, i32 5, i32 16, i32 16, i32 16, i32 6, i32 16, i32 16, i32 16, i32 7, i32 16, i32 16, i32 16, i32 8, i32 16, i32 16, i32 16, i32 9, i32 16, i32 16, i32 16, i32 10, i32 16, i32 16, i32 16, i32 11, i32 16, i32 16, i32 16, i32 12, i32 16, i32 16, i32 16, i32 13, i32 16, i32 16, i32 16, i32 14, i32 16, i32 16, i32 16, i32 15, i32 16, i32 16, i32 16>
; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <64 x i8> [[TMP0]] to <16 x i32>
; CHECK-NEXT:    [[DST_GEP:%.*]] = getelementptr i32, ptr [[DST:%.*]], i64 [[IV]]
; CHECK-NEXT:    store <16 x i32> [[TMP1]], ptr [[DST_GEP]], align 64
; CHECK-NEXT:    [[IV_NEXT]] = add nuw i64 [[IV]], 16
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], 128
; CHECK-NEXT:    br i1 [[EC]], label [[EXIT:%.*]], label [[LOOP]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
  %src.gep = getelementptr i8, ptr %src, i64 %iv
  %load = load <16 x i8>, ptr %src.gep
  %ext = zext <16 x i8> %load to <16 x i32>
  %dst.gep = getelementptr i32, ptr %dst, i64 %iv
  store <16 x i32> %ext, ptr %dst.gep
  %iv.next = add nuw i64 %iv, 16
  %ec = icmp eq i64 %iv.next, 128
  br i1 %ec, label %exit, label %loop

exit:
  ret void
}

; Not profitable to use shuffle/tbl, as 4 tbls + materializing the masks
; require more instructions than lowering zext directly.
define void @zext_v16i8_to_v16i32_no_loop(ptr %src, ptr %dst) {
; CHECK-LABEL: @zext_v16i8_to_v16i32_no_loop(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[LOAD:%.*]] = load <16 x i8>, ptr [[SRC:%.*]], align 16
; CHECK-NEXT:    [[EXT:%.*]] = zext <16 x i8> [[LOAD]] to <16 x i32>
; CHECK-NEXT:    store <16 x i32> [[EXT]], ptr [[DST:%.*]], align 64
; CHECK-NEXT:    ret void
;
entry:
  %load = load <16 x i8>, ptr %src
  %ext = zext <16 x i8> %load to <16 x i32>
  store <16 x i32> %ext, ptr %dst
  ret void
}

define void @zext_v16i8_to_v16i16_in_loop(ptr %src, ptr %dst) {
; CHECK-LABEL: @zext_v16i8_to_v16i16_in_loop(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
; CHECK-NEXT:    [[SRC_GEP:%.*]] = getelementptr i8, ptr [[SRC:%.*]], i64 [[IV]]
; CHECK-NEXT:    [[LOAD:%.*]] = load <16 x i8>, ptr [[SRC_GEP]], align 16
; CHECK-NEXT:    [[EXT:%.*]] = zext <16 x i8> [[LOAD]] to <16 x i16>
; CHECK-NEXT:    [[DST_GEP:%.*]] = getelementptr i16, ptr [[DST:%.*]], i64 [[IV]]
; CHECK-NEXT:    store <16 x i16> [[EXT]], ptr [[DST_GEP]], align 32
; CHECK-NEXT:    [[IV_NEXT]] = add nuw i64 [[IV]], 16
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], 128
; CHECK-NEXT:    br i1 [[EC]], label [[EXIT:%.*]], label [[LOOP]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
  %src.gep = getelementptr i8, ptr %src, i64 %iv
  %load = load <16 x i8>, ptr %src.gep
  %ext = zext <16 x i8> %load to <16 x i16>
  %dst.gep = getelementptr i16, ptr %dst, i64 %iv
  store <16 x i16> %ext, ptr %dst.gep
  %iv.next = add nuw i64 %iv, 16
  %ec = icmp eq i64 %iv.next, 128
  br i1 %ec, label %exit, label %loop

exit:
  ret void
}

define void @zext_v8i8_to_v8i32_in_loop(ptr %src, ptr %dst) {
; CHECK-LABEL: @zext_v8i8_to_v8i32_in_loop(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
; CHECK-NEXT:    [[SRC_GEP:%.*]] = getelementptr i8, ptr [[SRC:%.*]], i64 [[IV]]
; CHECK-NEXT:    [[LOAD:%.*]] = load <8 x i8>, ptr [[SRC_GEP]], align 8
; CHECK-NEXT:    [[TMP0:%.*]] = shufflevector <8 x i8> [[LOAD]], <8 x i8> <i8 0, i8 poison, i8 poison, i8 poison, i8 poison, i8 poison, i8 poison, i8 poison>, <32 x i32> <i32 0, i32 8, i32 8, i32 8, i32 1, i32 8, i32 8, i32 8, i32 2, i32 8, i32 8, i32 8, i32 3, i32 8, i32 8, i32 8, i32 4, i32 8, i32 8, i32 8, i32 5, i32 8, i32 8, i32 8, i32 6, i32 8, i32 8, i32 8, i32 7, i32 8, i32 8, i32 8>
; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <32 x i8> [[TMP0]] to <8 x i32>
; CHECK-NEXT:    [[DST_GEP:%.*]] = getelementptr i32, ptr [[DST:%.*]], i64 [[IV]]
; CHECK-NEXT:    store <8 x i32> [[TMP1]], ptr [[DST_GEP]], align 32
; CHECK-NEXT:    [[IV_NEXT]] = add nuw i64 [[IV]], 16
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], 128
; CHECK-NEXT:    br i1 [[EC]], label [[EXIT:%.*]], label [[LOOP]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
  %src.gep = getelementptr i8, ptr %src, i64 %iv
  %load = load <8 x i8>, ptr %src.gep
  %ext = zext <8 x i8> %load to <8 x i32>
  %dst.gep = getelementptr i32, ptr %dst, i64 %iv
  store <8 x i32> %ext, ptr %dst.gep
  %iv.next = add nuw i64 %iv, 16
  %ec = icmp eq i64 %iv.next, 128
  br i1 %ec, label %exit, label %loop

exit:
  ret void
}

define void @zext_v16i8_to_v16i64_in_loop(ptr %src, ptr %dst) {
; CHECK-LABEL: @zext_v16i8_to_v16i64_in_loop(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
; CHECK-NEXT:    [[SRC_GEP:%.*]] = getelementptr i8, ptr [[SRC:%.*]], i64 [[IV]]
; CHECK-NEXT:    [[LOAD:%.*]] = load <16 x i8>, ptr [[SRC_GEP]], align 16
; CHECK-NEXT:    [[EXT:%.*]] = zext <16 x i8> [[LOAD]] to <16 x i64>
; CHECK-NEXT:    [[DST_GEP:%.*]] = getelementptr i64, ptr [[DST:%.*]], i64 [[IV]]
; CHECK-NEXT:    store <16 x i64> [[EXT]], ptr [[DST_GEP]], align 128
; CHECK-NEXT:    [[IV_NEXT]] = add nuw i64 [[IV]], 16
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], 128
; CHECK-NEXT:    br i1 [[EC]], label [[EXIT:%.*]], label [[LOOP]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
  %src.gep = getelementptr i8, ptr %src, i64 %iv
  %load = load <16 x i8>, ptr %src.gep
  %ext = zext <16 x i8> %load to <16 x i64>
  %dst.gep = getelementptr i64, ptr %dst, i64 %iv
  store <16 x i64> %ext, ptr %dst.gep
  %iv.next = add nuw i64 %iv, 16
  %ec = icmp eq i64 %iv.next, 128
  br i1 %ec, label %exit, label %loop

exit:
  ret void
}