llvm/test/Transforms/LoopVectorize/trunc-extended-icmps.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
; RUN: opt -p loop-vectorize -force-vector-width=4 -force-vector-interleave=1 -S %s | FileCheck %s

target datalayout = "e-m:e-p:64:64-i64:64-i128:128-n32:64-S128"

define i32 @test_icmp_constant_op_zext(ptr %dst) {
; CHECK-LABEL: define i32 @test_icmp_constant_op_zext(
; CHECK-SAME: ptr [[DST:%.*]]) {
; 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 i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[DOTCAST:%.*]] = trunc i32 [[INDEX]] to i16
; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i16 1, [[DOTCAST]]
; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr i8, ptr [[DST]], i16 [[OFFSET_IDX]]
; CHECK-NEXT:    store <4 x i8> splat (i8 109), ptr [[TMP1]], align 1
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
; CHECK-NEXT:    [[TMP3:%.*]] = icmp eq i32 [[INDEX_NEXT]], 996
; CHECK-NEXT:    br i1 [[TMP3]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[SCALAR_PH:%.*]]
; CHECK:       scalar.ph:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[IV:%.*]] = phi i16 [ 997, [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
; CHECK-NEXT:    [[C:%.*]] = icmp ne i64 7304878031173690989, 0
; CHECK-NEXT:    [[C_EXT:%.*]] = zext i1 [[C]] to i64
; CHECK-NEXT:    [[OR:%.*]] = or i64 7304878031173690989, [[C_EXT]]
; CHECK-NEXT:    [[OR_TRUNC:%.*]] = trunc i64 [[OR]] to i8
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i8, ptr [[DST]], i16 [[IV]]
; CHECK-NEXT:    store i8 [[OR_TRUNC]], ptr [[GEP]], align 1
; CHECK-NEXT:    [[IV_NEXT]] = add i16 [[IV]], 1
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i16 [[IV_NEXT]], 1000
; CHECK-NEXT:    br i1 [[EC]], label [[EXIT:%.*]], label [[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK:       exit:
; CHECK-NEXT:    ret i32 0
;
entry:
  br label %loop

loop:
  %iv = phi i16 [ 1, %entry ], [ %iv.next, %loop ]
  %c = icmp ne i64 7304878031173690989, 0
  %c.ext = zext i1 %c to i64
  %or = or i64 7304878031173690989, %c.ext
  %or.trunc = trunc i64 %or to i8
  %gep = getelementptr i8, ptr %dst, i16 %iv
  store i8 %or.trunc, ptr %gep, align 1
  %iv.next = add i16 %iv, 1
  %ec = icmp eq i16 %iv.next, 1000
  br i1 %ec, label %exit, label %loop

exit:
  ret i32 0
}


define i32 @test_icmp_and_op_zext(ptr %dst, i64 %a) {
; CHECK-LABEL: define i32 @test_icmp_and_op_zext(
; CHECK-SAME: ptr [[DST:%.*]], i64 [[A:%.*]]) {
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[AND:%.*]] = and i64 [[A]], 7304878031173690989
; CHECK-NEXT:    br label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i64> poison, i64 [[AND]], i64 0
; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i64> [[BROADCAST_SPLATINSERT]], <4 x i64> poison, <4 x i32> zeroinitializer
; CHECK-NEXT:    [[TMP1:%.*]] = icmp ne <4 x i64> [[BROADCAST_SPLAT]], zeroinitializer
; CHECK-NEXT:    [[TMP2:%.*]] = zext <4 x i1> [[TMP1]] to <4 x i64>
; CHECK-NEXT:    [[TMP3:%.*]] = or <4 x i64> [[BROADCAST_SPLAT]], [[TMP2]]
; CHECK-NEXT:    [[TMP4:%.*]] = trunc <4 x i64> [[TMP3]] to <4 x i8>
; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT:    [[DOTCAST:%.*]] = trunc i32 [[INDEX]] to i16
; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i16 1, [[DOTCAST]]
; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr i8, ptr [[DST]], i16 [[OFFSET_IDX]]
; CHECK-NEXT:    store <4 x i8> [[TMP4]], ptr [[TMP5]], align 1
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i32 [[INDEX_NEXT]], 996
; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[SCALAR_PH:%.*]]
; CHECK:       scalar.ph:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[IV:%.*]] = phi i16 [ 997, [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
; CHECK-NEXT:    [[C:%.*]] = icmp ne i64 [[AND]], 0
; CHECK-NEXT:    [[C_EXT:%.*]] = zext i1 [[C]] to i64
; CHECK-NEXT:    [[OR:%.*]] = or i64 [[AND]], [[C_EXT]]
; CHECK-NEXT:    [[OR_TRUNC:%.*]] = trunc i64 [[OR]] to i8
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i8, ptr [[DST]], i16 [[IV]]
; CHECK-NEXT:    store i8 [[OR_TRUNC]], ptr [[GEP]], align 1
; CHECK-NEXT:    [[IV_NEXT]] = add i16 [[IV]], 1
; CHECK-NEXT:    [[EC:%.*]] = icmp eq i16 [[IV_NEXT]], 1000
; CHECK-NEXT:    br i1 [[EC]], label [[EXIT:%.*]], label [[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK:       exit:
; CHECK-NEXT:    ret i32 0
;
entry:
  %and = and i64 %a, 7304878031173690989
  br label %loop

loop:
  %iv = phi i16 [ 1, %entry ], [ %iv.next, %loop ]
  %c = icmp ne i64 %and, 0
  %c.ext = zext i1 %c to i64
  %or = or i64 %and, %c.ext
  %or.trunc = trunc i64 %or to i8
  %gep = getelementptr i8, ptr %dst, i16 %iv
  store i8 %or.trunc, ptr %gep, align 1
  %iv.next = add i16 %iv, 1
  %ec = icmp eq i16 %iv.next, 1000
  br i1 %ec, label %exit, label %loop

exit:
  ret i32 0
}

define void @ext_cmp(ptr %src.1, ptr %src.2, ptr noalias %dst) {
; CHECK-LABEL: define void @ext_cmp(
; CHECK-SAME: ptr [[SRC_1:%.*]], ptr [[SRC_2:%.*]], ptr noalias [[DST:%.*]]) {
; 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:    [[TMP0:%.*]] = getelementptr inbounds i16, ptr [[DST]], i64 [[INDEX]]
; CHECK-NEXT:    store <4 x i16> zeroinitializer, ptr [[TMP0]], align 2
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT:    [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1000
; CHECK-NEXT:    br i1 [[TMP11]], 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
;
entry:
  br label %loop

loop:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
  %gep.src.1 = getelementptr inbounds i16, ptr %src.1, i64 %iv
  %i2 = load i16, ptr %gep.src.1, align 2
  %i3 = sext i16 %i2 to i32
  %c.1 = icmp sgt i32 0, %i3
  %gep.src.2 = getelementptr inbounds i8, ptr %src.2, i64 %iv
  %i4 = load i8, ptr %gep.src.2, align 2
  %i5 = zext i8 %i4 to i32
  %i6 = select i1 %c.1, i32 0, i32 %i5
  %i7 = and i32 %i6, 0
  %i8 = trunc nuw nsw i32 %i7 to i16
  %gep.dst = getelementptr inbounds i16, ptr %dst, i64 %iv
  store i16 %i8, ptr %gep.dst, align 2
  %iv.next = add nsw i64 %iv, 1
  %ec = icmp eq i64 %iv.next, 1000
  br i1 %ec, label %exit, label %loop

exit:
  ret void
}

; The narrowed %cmp keeps its i1 result type when truncating to minimal bitwidth.
define void @narrowed_icmp_used_by_ext(ptr noalias %src, ptr noalias %dst) {
; CHECK-LABEL: define void @narrowed_icmp_used_by_ext(
; CHECK-SAME: ptr noalias [[SRC:%.*]], ptr noalias [[DST:%.*]]) {
; 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:    [[TMP0:%.*]] = getelementptr inbounds i64, ptr [[SRC]], i64 [[INDEX]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i64>, ptr [[TMP0]], align 8
; CHECK-NEXT:    [[TMP1:%.*]] = trunc <4 x i64> [[WIDE_LOAD]] to <4 x i32>
; CHECK-NEXT:    [[TMP2:%.*]] = icmp sle <4 x i32> [[TMP1]], splat (i32 1)
; CHECK-NEXT:    [[TMP3:%.*]] = zext <4 x i1> [[TMP2]] to <4 x i32>
; CHECK-NEXT:    [[TMP4:%.*]] = or <4 x i32> splat (i32 1), [[TMP3]]
; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i32, ptr [[DST]], i64 [[INDEX]]
; CHECK-NEXT:    store <4 x i32> [[TMP4]], ptr [[TMP5]], align 4
; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT:    [[TMP6:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1024
; CHECK-NEXT:    br i1 [[TMP6]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[EXIT:%.*]]
; CHECK:       exit:
; CHECK-NEXT:    ret void
;
entry:
  br label %loop

loop:
  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
  %gep.src = getelementptr inbounds i64, ptr %src, i64 %iv
  %l = load i64, ptr %gep.src
  %cmp = icmp sle i64 %l, 1
  %ext = zext i1 %cmp to i64
  %or = or i64 1, %ext
  %or.trunc = trunc i64 %or to i32
  %gep.dst = getelementptr inbounds i32, ptr %dst, i64 %iv
  store i32 %or.trunc, ptr %gep.dst, align 4
  %iv.next = add i64 %iv, 1
  %ec = icmp eq i64 %iv.next, 1024
  br i1 %ec, label %exit, label %loop

exit:
  ret void
}
;.
; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
; CHECK: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
; CHECK: [[LOOP4]] = distinct !{[[LOOP4]], [[META1]], [[META2]]}
; CHECK: [[LOOP5]] = distinct !{[[LOOP5]], [[META2]], [[META1]]}
; CHECK: [[LOOP6]] = distinct !{[[LOOP6]], [[META1]], [[META2]]}
; CHECK: [[LOOP7]] = distinct !{[[LOOP7]], [[META1]], [[META2]]}
;.