Compiler Explorer

Source code

; ModuleID = '/tmp/file.c'
source_filename = "/tmp/file.c"
target datalayout = "e-m:e-p:32:32-Fi8-i64:64-v128:64:128-a:0:32-n32-S64"
target triple = "thumbv8.1m.main-arm-none-eabi"

; Function Attrs: nofree norecurse nounwind
define dso_local void @arm_abs_q15(i16* nocapture readonly %pSrc, i16* nocapture %pDst, i32 %blockSize) local_unnamed_addr #0 {
entry:
  %cmp19 = icmp eq i32 %blockSize, 0
  br i1 %cmp19, label %while.end, label %while.body.preheader

while.body.preheader:                             ; preds = %entry
  %min.iters.check = icmp ult i32 %blockSize, 8
  br i1 %min.iters.check, label %while.body.preheader33, label %vector.memcheck

vector.memcheck:                                  ; preds = %while.body.preheader
  %scevgep = getelementptr i16, i16* %pDst, i32 %blockSize
  %scevgep26 = getelementptr i16, i16* %pSrc, i32 %blockSize
  %bound0 = icmp ugt i16* %scevgep26, %pDst
  %bound1 = icmp ugt i16* %scevgep, %pSrc
  %found.conflict = and i1 %bound0, %bound1
  br i1 %found.conflict, label %while.body.preheader33, label %vector.ph

vector.ph:                                        ; preds = %vector.memcheck
  %n.vec = and i32 %blockSize, -8
  %ind.end = getelementptr i16, i16* %pSrc, i32 %n.vec
  %ind.end29 = sub i32 %blockSize, %n.vec
  %ind.end31 = getelementptr i16, i16* %pDst, i32 %n.vec
  br label %vector.body

vector.body:                                      ; preds = %vector.body, %vector.ph
  %index = phi i32 [ 0, %vector.ph ], [ %index.next, %vector.body ]
  %next.gep = getelementptr i16, i16* %pSrc, i32 %index
  %next.gep32 = getelementptr i16, i16* %pDst, i32 %index
  %0 = bitcast i16* %next.gep to <8 x i16>*
  %wide.load = load <8 x i16>, <8 x i16>* %0, align 2, !tbaa !3, !alias.scope !7
  %1 = icmp sgt <8 x i16> %wide.load, zeroinitializer
  %2 = icmp eq <8 x i16> %wide.load, <i16 -32768, i16 -32768, i16 -32768, i16 -32768, i16 -32768, i16 -32768, i16 -32768, i16 -32768>
  %3 = sub <8 x i16> zeroinitializer, %wide.load
  %4 = select <8 x i1> %2, <8 x i16> <i16 32767, i16 32767, i16 32767, i16 32767, i16 32767, i16 32767, i16 32767, i16 32767>, <8 x i16> %3
  %5 = select <8 x i1> %1, <8 x i16> %wide.load, <8 x i16> %4
  %6 = bitcast i16* %next.gep32 to <8 x i16>*
  store <8 x i16> %5, <8 x i16>* %6, align 2, !tbaa !3, !alias.scope !10, !noalias !7
  %index.next = add i32 %index, 8
  %7 = icmp eq i32 %index.next, %n.vec
  br i1 %7, label %middle.block, label %vector.body, !llvm.loop !12

middle.block:                                     ; preds = %vector.body
  %cmp.n = icmp eq i32 %n.vec, %blockSize
  br i1 %cmp.n, label %while.end, label %while.body.preheader33

while.body.preheader33:                           ; preds = %middle.block, %vector.memcheck, %while.body.preheader
  %pSrc.addr.022.ph = phi i16* [ %pSrc, %vector.memcheck ], [ %pSrc, %while.body.preheader ], [ %ind.end, %middle.block ]
  %blkCnt.021.ph = phi i32 [ %blockSize, %vector.memcheck ], [ %blockSize, %while.body.preheader ], [ %ind.end29, %middle.block ]
  %pDst.addr.020.ph = phi i16* [ %pDst, %vector.memcheck ], [ %pDst, %while.body.preheader ], [ %ind.end31, %middle.block ]
  br label %while.body

while.body:                                       ; preds = %while.body.preheader33, %while.body
  %pSrc.addr.022 = phi i16* [ %incdec.ptr, %while.body ], [ %pSrc.addr.022.ph, %while.body.preheader33 ]
  %blkCnt.021 = phi i32 [ %dec, %while.body ], [ %blkCnt.021.ph, %while.body.preheader33 ]
  %pDst.addr.020 = phi i16* [ %incdec.ptr13, %while.body ], [ %pDst.addr.020.ph, %while.body.preheader33 ]
  %incdec.ptr = getelementptr inbounds i16, i16* %pSrc.addr.022, i32 1
  %8 = load i16, i16* %pSrc.addr.022, align 2, !tbaa !3
  %cmp1 = icmp sgt i16 %8, 0
  %cmp5 = icmp eq i16 %8, -32768
  %sub = sub i16 0, %8
  %cond = select i1 %cmp5, i16 32767, i16 %sub
  %cond11 = select i1 %cmp1, i16 %8, i16 %cond
  %incdec.ptr13 = getelementptr inbounds i16, i16* %pDst.addr.020, i32 1
  store i16 %cond11, i16* %pDst.addr.020, align 2, !tbaa !3
  %dec = add i32 %blkCnt.021, -1
  %cmp = icmp eq i32 %dec, 0
  br i1 %cmp, label %while.end, label %while.body, !llvm.loop !14

while.end:                                        ; preds = %while.body, %middle.block, %entry
  ret void
}

attributes #0 = { nofree norecurse nounwind "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "frame-pointer"="all" "less-precise-fpmad"="false" "min-legal-vector-width"="0" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="generic" "target-features"="+armv8.1-m.main,+dsp,+fp-armv8d16sp,+fp16,+fullfp16,+hwdiv,+lob,+mve,+mve.fp,+ras,+strict-align,+thumb-mode,+vfp2sp,+vfp3d16sp,+vfp4d16sp" "unsafe-fp-math"="false" "use-soft-float"="false" }

!llvm.module.flags = !{!0, !1}
!llvm.ident = !{!2}

!0 = !{i32 1, !"wchar_size", i32 4}
!1 = !{i32 1, !"min_enum_size", i32 4}
!2 = !{!"clang version 11.0.0 (https://github.com/llvm/llvm-project 9d1dbd2e16e21c01933739f16e988239d5dcca04)"}
!3 = !{!4, !4, i64 0}
!4 = !{!"short", !5, i64 0}
!5 = !{!"omnipotent char", !6, i64 0}
!6 = !{!"Simple C/C++ TBAA"}
!7 = !{!8}
!8 = distinct !{!8, !9}
!9 = distinct !{!9, !"LVerDomain"}
!10 = !{!11}
!11 = distinct !{!11, !9}
!12 = distinct !{!12, !13}
!13 = !{!"llvm.loop.isvectorized", i32 1}
!14 = distinct !{!14, !13}