llvm/test/Transforms/LoopVectorize/single-early-exit-deref-assumptions.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals none --version 4
; RUN: opt -S < %s -p loop-vectorize -enable-early-exit-vectorization -force-vector-width=4 | FileCheck %s

define i64 @early_exit_alignment_and_deref_known_via_assumption_with_constant_size(ptr noalias %p1, ptr noalias %p2) nofree nosync {
; CHECK-LABEL: define i64 @early_exit_alignment_and_deref_known_via_assumption_with_constant_size(
; CHECK-SAME: ptr noalias [[P1:%.*]], ptr noalias [[P2:%.*]]) #[[ATTR0:[0-9]+]] {
; CHECK-NEXT:  entry:
; CHECK-NEXT:    call void @llvm.assume(i1 true) [ "align"(ptr [[P1]], i64 4), "dereferenceable"(ptr [[P1]], i64 1024) ]
; CHECK-NEXT:    call void @llvm.assume(i1 true) [ "align"(ptr [[P2]], i64 4), "dereferenceable"(ptr [[P2]], i64 1024) ]
; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK:       vector.ph:
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       vector.body:
; CHECK-NEXT:    [[INDEX1:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT3:%.*]], [[LOOP]] ]
; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr inbounds i8, ptr [[P1]], i64 [[INDEX1]]
; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i8>, ptr [[TMP0]], align 1
; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i8, ptr [[P2]], i64 [[INDEX1]]
; CHECK-NEXT:    [[WIDE_LOAD2:%.*]] = load <4 x i8>, ptr [[TMP2]], align 1
; CHECK-NEXT:    [[TMP4:%.*]] = icmp ne <4 x i8> [[WIDE_LOAD]], [[WIDE_LOAD2]]
; CHECK-NEXT:    [[INDEX_NEXT3]] = add nuw i64 [[INDEX1]], 4
; CHECK-NEXT:    [[TMP5:%.*]] = call i1 @llvm.vector.reduce.or.v4i1(<4 x i1> [[TMP4]])
; CHECK-NEXT:    [[TMP6:%.*]] = icmp eq i64 [[INDEX_NEXT3]], 1024
; CHECK-NEXT:    [[TMP7:%.*]] = or i1 [[TMP5]], [[TMP6]]
; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_SPLIT:%.*]], label [[LOOP]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK:       middle.split:
; CHECK-NEXT:    br i1 [[TMP5]], label [[VECTOR_EARLY_EXIT:%.*]], label [[MIDDLE_BLOCK:%.*]]
; CHECK:       middle.block:
; CHECK-NEXT:    br label [[LOOP_END:%.*]]
; CHECK:       vector.early.exit:
; CHECK-NEXT:    [[TMP8:%.*]] = call i64 @llvm.experimental.cttz.elts.i64.v4i1(<4 x i1> [[TMP4]], i1 true)
; CHECK-NEXT:    [[TMP9:%.*]] = add i64 [[INDEX1]], [[TMP8]]
; CHECK-NEXT:    br label [[LOOP_END]]
; CHECK:       scalar.ph:
; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ]
; CHECK-NEXT:    br label [[LOOP1:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ [[INDEX_NEXT:%.*]], [[LOOP_INC:%.*]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i8, ptr [[P1]], i64 [[INDEX]]
; CHECK-NEXT:    [[LD1:%.*]] = load i8, ptr [[ARRAYIDX]], align 1
; CHECK-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[P2]], i64 [[INDEX]]
; CHECK-NEXT:    [[LD2:%.*]] = load i8, ptr [[ARRAYIDX1]], align 1
; CHECK-NEXT:    [[CMP3:%.*]] = icmp eq i8 [[LD1]], [[LD2]]
; CHECK-NEXT:    br i1 [[CMP3]], label [[LOOP_INC]], label [[LOOP_END]]
; CHECK:       loop.inc:
; CHECK-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], 1
; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDEX_NEXT]], 1024
; CHECK-NEXT:    br i1 [[EXITCOND]], label [[LOOP1]], label [[LOOP_END]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK:       loop.end:
; CHECK-NEXT:    [[RETVAL:%.*]] = phi i64 [ [[INDEX]], [[LOOP1]] ], [ -1, [[LOOP_INC]] ], [ -1, [[MIDDLE_BLOCK]] ], [ [[TMP9]], [[VECTOR_EARLY_EXIT]] ]
; CHECK-NEXT:    ret i64 [[RETVAL]]
;
entry:
  call void @llvm.assume(i1 true) [ "align"(ptr %p1, i64 4), "dereferenceable"(ptr %p1, i64 1024) ]
  call void @llvm.assume(i1 true) [ "align"(ptr %p2, i64 4), "dereferenceable"(ptr %p2, i64 1024) ]
  br label %loop

loop:
  %index = phi i64 [ %index.next, %loop.inc ], [ 0, %entry ]
  %arrayidx = getelementptr inbounds i8, ptr %p1, i64 %index
  %ld1 = load i8, ptr %arrayidx, align 1
  %arrayidx1 = getelementptr inbounds i8, ptr %p2, i64 %index
  %ld2 = load i8, ptr %arrayidx1, align 1
  %cmp3 = icmp eq i8 %ld1, %ld2
  br i1 %cmp3, label %loop.inc, label %loop.end

loop.inc:
  %index.next = add i64 %index, 1
  %exitcond = icmp ne i64 %index.next, 1024
  br i1 %exitcond, label %loop, label %loop.end

loop.end:
  %retval = phi i64 [ %index, %loop ], [ -1, %loop.inc ]
  ret i64 %retval
}

define i64 @early_exit_alignment_and_deref_via_assumption_with_constant_size_too_small(ptr noalias %p1, ptr noalias %p2) nofree nosync {
; CHECK-LABEL: define i64 @early_exit_alignment_and_deref_via_assumption_with_constant_size_too_small(
; CHECK-SAME: ptr noalias [[P1:%.*]], ptr noalias [[P2:%.*]]) #[[ATTR0]] {
; CHECK-NEXT:  entry:
; CHECK-NEXT:    call void @llvm.assume(i1 true) [ "align"(ptr [[P1]], i64 4), "dereferenceable"(ptr [[P1]], i64 1024) ]
; CHECK-NEXT:    call void @llvm.assume(i1 true) [ "align"(ptr [[P2]], i64 4), "dereferenceable"(ptr [[P2]], i64 1024) ]
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ [[INDEX_NEXT:%.*]], [[LOOP_INC:%.*]] ], [ 0, [[ENTRY:%.*]] ]
; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i8, ptr [[P1]], i64 [[INDEX]]
; CHECK-NEXT:    [[LD1:%.*]] = load i8, ptr [[ARRAYIDX]], align 1
; CHECK-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[P2]], i64 [[INDEX]]
; CHECK-NEXT:    [[LD2:%.*]] = load i8, ptr [[ARRAYIDX1]], align 1
; CHECK-NEXT:    [[CMP3:%.*]] = icmp eq i8 [[LD1]], [[LD2]]
; CHECK-NEXT:    br i1 [[CMP3]], label [[LOOP_INC]], label [[LOOP_END:%.*]]
; CHECK:       loop.inc:
; CHECK-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], 1
; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDEX_NEXT]], 1025
; CHECK-NEXT:    br i1 [[EXITCOND]], label [[LOOP]], label [[LOOP_END]]
; CHECK:       loop.end:
; CHECK-NEXT:    [[RETVAL:%.*]] = phi i64 [ [[INDEX]], [[LOOP]] ], [ -1, [[LOOP_INC]] ]
; CHECK-NEXT:    ret i64 [[RETVAL]]
;
entry:
  call void @llvm.assume(i1 true) [ "align"(ptr %p1, i64 4), "dereferenceable"(ptr %p1, i64 1024) ]
  call void @llvm.assume(i1 true) [ "align"(ptr %p2, i64 4), "dereferenceable"(ptr %p2, i64 1024) ]
  br label %loop

loop:
  %index = phi i64 [ %index.next, %loop.inc ], [ 0, %entry ]
  %arrayidx = getelementptr inbounds i8, ptr %p1, i64 %index
  %ld1 = load i8, ptr %arrayidx, align 1
  %arrayidx1 = getelementptr inbounds i8, ptr %p2, i64 %index
  %ld2 = load i8, ptr %arrayidx1, align 1
  %cmp3 = icmp eq i8 %ld1, %ld2
  br i1 %cmp3, label %loop.inc, label %loop.end

loop.inc:
  %index.next = add i64 %index, 1
  %exitcond = icmp ne i64 %index.next, 1025
  br i1 %exitcond, label %loop, label %loop.end

loop.end:
  %retval = phi i64 [ %index, %loop ], [ -1, %loop.inc ]
  ret i64 %retval
}

define i64 @early_exit_alignment_and_deref_known_via_assumption(ptr noalias %p1, ptr noalias %p2, i64 %n) nofree nosync {
; CHECK-LABEL: define i64 @early_exit_alignment_and_deref_known_via_assumption(
; CHECK-SAME: ptr noalias [[P1:%.*]], ptr noalias [[P2:%.*]], i64 [[N:%.*]]) #[[ATTR0]] {
; CHECK-NEXT:  entry:
; CHECK-NEXT:    call void @llvm.assume(i1 true) [ "align"(ptr [[P1]], i64 4), "dereferenceable"(ptr [[P1]], i64 [[N]]) ]
; CHECK-NEXT:    call void @llvm.assume(i1 true) [ "align"(ptr [[P2]], i64 4), "dereferenceable"(ptr [[P2]], i64 [[N]]) ]
; CHECK-NEXT:    br label [[LOOP:%.*]]
; CHECK:       loop:
; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ [[INDEX_NEXT:%.*]], [[LOOP_INC:%.*]] ], [ 0, [[ENTRY:%.*]] ]
; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i8, ptr [[P1]], i64 [[INDEX]]
; CHECK-NEXT:    [[LD1:%.*]] = load i8, ptr [[ARRAYIDX]], align 1
; CHECK-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, ptr [[P2]], i64 [[INDEX]]
; CHECK-NEXT:    [[LD2:%.*]] = load i8, ptr [[ARRAYIDX1]], align 1
; CHECK-NEXT:    [[CMP3:%.*]] = icmp eq i8 [[LD1]], [[LD2]]
; CHECK-NEXT:    br i1 [[CMP3]], label [[LOOP_INC]], label [[LOOP_END:%.*]]
; CHECK:       loop.inc:
; CHECK-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], 1
; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp ne i64 [[INDEX_NEXT]], [[N]]
; CHECK-NEXT:    br i1 [[EXITCOND]], label [[LOOP]], label [[LOOP_END]]
; CHECK:       loop.end:
; CHECK-NEXT:    [[RETVAL:%.*]] = phi i64 [ [[INDEX]], [[LOOP]] ], [ -1, [[LOOP_INC]] ]
; CHECK-NEXT:    ret i64 [[RETVAL]]
;
entry:
  call void @llvm.assume(i1 true) [ "align"(ptr %p1, i64 4), "dereferenceable"(ptr %p1, i64 %n) ]
  call void @llvm.assume(i1 true) [ "align"(ptr %p2, i64 4), "dereferenceable"(ptr %p2, i64 %n) ]
  br label %loop

loop:
  %index = phi i64 [ %index.next, %loop.inc ], [ 0, %entry ]
  %arrayidx = getelementptr inbounds i8, ptr %p1, i64 %index
  %ld1 = load i8, ptr %arrayidx, align 1
  %arrayidx1 = getelementptr inbounds i8, ptr %p2, i64 %index
  %ld2 = load i8, ptr %arrayidx1, align 1
  %cmp3 = icmp eq i8 %ld1, %ld2
  br i1 %cmp3, label %loop.inc, label %loop.end

loop.inc:
  %index.next = add i64 %index, 1
  %exitcond = icmp ne i64 %index.next, %n
  br i1 %exitcond, label %loop, label %loop.end

loop.end:
  %retval = phi i64 [ %index, %loop ], [ -1, %loop.inc ]
  ret i64 %retval
}