llvm/llvm/test/CodeGen/Thumb2/LowOverheadLoops/unrolled-and-vector.mir

# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
# RUN: llc -mtriple=thumbv8.1m.main -mattr=+mve -run-pass=arm-low-overhead-loops %s -o - | FileCheck %s

--- |
  define dso_local arm_aapcs_vfpcc void @unrolled_and_vector(ptr nocapture %res, ptr nocapture readonly %a, ptr nocapture readonly %b, i32 %N) {
  entry:
    %cmp10 = icmp eq i32 %N, 0
    br i1 %cmp10, label %for.cond.cleanup, label %vector.memcheck

  vector.memcheck:                                  ; preds = %entry
    %scevgep = getelementptr i8, ptr %res, i32 %N
    %scevgep12 = getelementptr i8, ptr %a, i32 %N
    %scevgep13 = getelementptr i8, ptr %b, i32 %N
    %bound0 = icmp ugt ptr %scevgep12, %res
    %bound1 = icmp ugt ptr %scevgep, %a
    %found.conflict = and i1 %bound0, %bound1
    %bound014 = icmp ugt ptr %scevgep13, %res
    %bound115 = icmp ugt ptr %scevgep, %b
    %found.conflict16 = and i1 %bound014, %bound115
    %conflict.rdx = or i1 %found.conflict, %found.conflict16
    %0 = add i32 %N, 15
    %1 = lshr i32 %0, 4
    %2 = shl nuw i32 %1, 4
    %3 = add i32 %2, -16
    %4 = lshr i32 %3, 4
    %5 = add nuw nsw i32 %4, 1
    br i1 %conflict.rdx, label %for.body.preheader, label %vector.ph

  for.body.preheader:                               ; preds = %vector.memcheck
    %6 = add i32 %N, -1
    %xtraiter = and i32 %N, 3
    %7 = icmp ult i32 %6, 3
    %8 = add i32 %N, -4
    %9 = sub i32 %8, %xtraiter
    %10 = lshr i32 %9, 2
    %11 = add nuw nsw i32 %10, 1
    br i1 %7, label %for.cond.cleanup.loopexit.unr-lcssa, label %for.body.preheader.new

  for.body.preheader.new:                           ; preds = %for.body.preheader
    %start1 = call i32 @llvm.start.loop.iterations.i32(i32 %11)
    br label %for.body

  vector.ph:                                        ; preds = %vector.memcheck
    %start2 = call i32 @llvm.start.loop.iterations.i32(i32 %5)
    br label %vector.body

  vector.body:                                      ; preds = %vector.body, %vector.ph
    %lsr.iv50 = phi ptr [ %scevgep51, %vector.body ], [ %res, %vector.ph ]
    %lsr.iv47 = phi ptr [ %scevgep48, %vector.body ], [ %b, %vector.ph ]
    %lsr.iv = phi ptr [ %scevgep45, %vector.body ], [ %a, %vector.ph ]
    %12 = phi i32 [ %start2, %vector.ph ], [ %17, %vector.body ]
    %13 = phi i32 [ %N, %vector.ph ], [ %15, %vector.body ]
    %lsr.iv5052 = bitcast ptr %lsr.iv50 to ptr
    %lsr.iv4749 = bitcast ptr %lsr.iv47 to ptr
    %lsr.iv46 = bitcast ptr %lsr.iv to ptr
    %14 = call <16 x i1> @llvm.arm.mve.vctp8(i32 %13)
    %15 = sub i32 %13, 16
    %wide.masked.load = call <16 x i8> @llvm.masked.load.v16i8.p0(ptr %lsr.iv46, i32 1, <16 x i1> %14, <16 x i8> undef)
    %wide.masked.load19 = call <16 x i8> @llvm.masked.load.v16i8.p0(ptr %lsr.iv4749, i32 1, <16 x i1> %14, <16 x i8> undef)
    %16 = add <16 x i8> %wide.masked.load19, %wide.masked.load
    call void @llvm.masked.store.v16i8.p0(<16 x i8> %16, ptr %lsr.iv5052, i32 1, <16 x i1> %14)
    %scevgep45 = getelementptr i8, ptr %lsr.iv, i32 16
    %scevgep48 = getelementptr i8, ptr %lsr.iv47, i32 16
    %scevgep51 = getelementptr i8, ptr %lsr.iv50, i32 16
    %17 = call i32 @llvm.loop.decrement.reg.i32.i32.i32(i32 %12, i32 1)
    %18 = icmp ne i32 %17, 0
    br i1 %18, label %vector.body, label %for.cond.cleanup

  for.cond.cleanup.loopexit.unr-lcssa:              ; preds = %for.body, %for.body.preheader
    %i.011.unr = phi i32 [ 0, %for.body.preheader ], [ %inc.3, %for.body ]
    %lcmp.mod = icmp eq i32 %xtraiter, 0
    br i1 %lcmp.mod, label %for.cond.cleanup, label %for.body.epil

  for.body.epil:                                    ; preds = %for.cond.cleanup.loopexit.unr-lcssa
    %arrayidx.epil = getelementptr inbounds i8, ptr %a, i32 %i.011.unr
    %19 = load i8, ptr %arrayidx.epil, align 1
    %arrayidx1.epil = getelementptr inbounds i8, ptr %b, i32 %i.011.unr
    %20 = load i8, ptr %arrayidx1.epil, align 1
    %add.epil = add i8 %20, %19
    %arrayidx4.epil = getelementptr inbounds i8, ptr %res, i32 %i.011.unr
    store i8 %add.epil, ptr %arrayidx4.epil, align 1
    %inc.epil = add nuw i32 %i.011.unr, 1
    %epil.iter.cmp = icmp eq i32 %xtraiter, 1
    br i1 %epil.iter.cmp, label %for.cond.cleanup, label %for.body.epil.1

  for.cond.cleanup:                                 ; preds = %vector.body, %for.cond.cleanup.loopexit.unr-lcssa, %for.body.epil.1, %for.body.epil, %for.body.epil.2, %entry
    ret void

  for.body:                                         ; preds = %for.body, %for.body.preheader.new
    %i.011 = phi i32 [ 0, %for.body.preheader.new ], [ %inc.3, %for.body ]
    %21 = phi i32 [ %start1, %for.body.preheader.new ], [ %30, %for.body ]
    %scevgep23 = getelementptr i8, ptr %a, i32 %i.011
    %scevgep2453 = bitcast ptr %scevgep23 to ptr
    %22 = load i8, ptr %scevgep2453, align 1
    %scevgep27 = getelementptr i8, ptr %b, i32 %i.011
    %scevgep2854 = bitcast ptr %scevgep27 to ptr
    %23 = load i8, ptr %scevgep2854, align 1
    %add = add i8 %23, %22
    %scevgep31 = getelementptr i8, ptr %res, i32 %i.011
    %scevgep3255 = bitcast ptr %scevgep31 to ptr
    store i8 %add, ptr %scevgep3255, align 1
    %scevgep39 = getelementptr i8, ptr %a, i32 %i.011
    %scevgep40 = getelementptr i8, ptr %scevgep39, i32 1
    %24 = load i8, ptr %scevgep40, align 1
    %scevgep41 = getelementptr i8, ptr %b, i32 %i.011
    %scevgep42 = getelementptr i8, ptr %scevgep41, i32 1
    %25 = load i8, ptr %scevgep42, align 1
    %add.1 = add i8 %25, %24
    %scevgep43 = getelementptr i8, ptr %res, i32 %i.011
    %scevgep44 = getelementptr i8, ptr %scevgep43, i32 1
    store i8 %add.1, ptr %scevgep44, align 1
    %scevgep33 = getelementptr i8, ptr %a, i32 %i.011
    %scevgep34 = getelementptr i8, ptr %scevgep33, i32 2
    %26 = load i8, ptr %scevgep34, align 1
    %scevgep35 = getelementptr i8, ptr %b, i32 %i.011
    %scevgep36 = getelementptr i8, ptr %scevgep35, i32 2
    %27 = load i8, ptr %scevgep36, align 1
    %add.2 = add i8 %27, %26
    %scevgep37 = getelementptr i8, ptr %res, i32 %i.011
    %scevgep38 = getelementptr i8, ptr %scevgep37, i32 2
    store i8 %add.2, ptr %scevgep38, align 1
    %scevgep21 = getelementptr i8, ptr %a, i32 %i.011
    %scevgep22 = getelementptr i8, ptr %scevgep21, i32 3
    %28 = load i8, ptr %scevgep22, align 1
    %scevgep25 = getelementptr i8, ptr %b, i32 %i.011
    %scevgep26 = getelementptr i8, ptr %scevgep25, i32 3
    %29 = load i8, ptr %scevgep26, align 1
    %add.3 = add i8 %29, %28
    %scevgep29 = getelementptr i8, ptr %res, i32 %i.011
    %scevgep30 = getelementptr i8, ptr %scevgep29, i32 3
    store i8 %add.3, ptr %scevgep30, align 1
    %inc.3 = add nuw i32 %i.011, 4
    %30 = call i32 @llvm.loop.decrement.reg.i32.i32.i32(i32 %21, i32 1)
    %31 = icmp ne i32 %30, 0
    br i1 %31, label %for.body, label %for.cond.cleanup.loopexit.unr-lcssa

  for.body.epil.1:                                  ; preds = %for.body.epil
    %arrayidx.epil.1 = getelementptr inbounds i8, ptr %a, i32 %inc.epil
    %32 = load i8, ptr %arrayidx.epil.1, align 1
    %arrayidx1.epil.1 = getelementptr inbounds i8, ptr %b, i32 %inc.epil
    %33 = load i8, ptr %arrayidx1.epil.1, align 1
    %add.epil.1 = add i8 %33, %32
    %arrayidx4.epil.1 = getelementptr inbounds i8, ptr %res, i32 %inc.epil
    store i8 %add.epil.1, ptr %arrayidx4.epil.1, align 1
    %inc.epil.1 = add nuw i32 %i.011.unr, 2
    %epil.iter.cmp.1 = icmp eq i32 %xtraiter, 2
    br i1 %epil.iter.cmp.1, label %for.cond.cleanup, label %for.body.epil.2

  for.body.epil.2:                                  ; preds = %for.body.epil.1
    %arrayidx.epil.2 = getelementptr inbounds i8, ptr %a, i32 %inc.epil.1
    %34 = load i8, ptr %arrayidx.epil.2, align 1
    %arrayidx1.epil.2 = getelementptr inbounds i8, ptr %b, i32 %inc.epil.1
    %35 = load i8, ptr %arrayidx1.epil.2, align 1
    %add.epil.2 = add i8 %35, %34
    %arrayidx4.epil.2 = getelementptr inbounds i8, ptr %res, i32 %inc.epil.1
    store i8 %add.epil.2, ptr %arrayidx4.epil.2, align 1
    br label %for.cond.cleanup
  }

  declare <16 x i8> @llvm.masked.load.v16i8.p0(ptr, i32 immarg, <16 x i1>, <16 x i8>) #1
  declare void @llvm.masked.store.v16i8.p0(<16 x i8>, ptr, i32 immarg, <16 x i1>) #2
  declare i32 @llvm.start.loop.iterations.i32(i32) #3
  declare i32 @llvm.loop.decrement.reg.i32.i32.i32(i32, i32) #3
  declare <16 x i1> @llvm.arm.mve.vctp8(i32) #4

...
---
name:            unrolled_and_vector
alignment:       2
exposesReturnsTwice: false
legalized:       false
regBankSelected: false
selected:        false
failedISel:      false
tracksRegLiveness: true
hasWinCFI:       false
registers:       []
liveins:
  - { reg: '$r0', virtual-reg: '' }
  - { reg: '$r1', virtual-reg: '' }
  - { reg: '$r2', virtual-reg: '' }
  - { reg: '$r3', virtual-reg: '' }
frameInfo:
  isFrameAddressTaken: false
  isReturnAddressTaken: false
  hasStackMap:     false
  hasPatchPoint:   false
  stackSize:       32
  offsetAdjustment: -24
  maxAlignment:    4
  adjustsStack:    false
  hasCalls:        false
  stackProtector:  ''
  maxCallFrameSize: 0
  cvBytesOfCalleeSavedRegisters: 0
  hasOpaqueSPAdjustment: false
  hasVAStart:      false
  hasMustTailInVarArgFunc: false
  localFrameSize:  0
  savePoint:       ''
  restorePoint:    ''
fixedStack:      []
stack:
  - { id: 0, name: '', type: spill-slot, offset: -4, size: 4, alignment: 4,
      stack-id: default, callee-saved-register: '$lr', callee-saved-restored: false,
      debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
  - { id: 1, name: '', type: spill-slot, offset: -8, size: 4, alignment: 4,
      stack-id: default, callee-saved-register: '$r7', callee-saved-restored: true,
      debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
  - { id: 2, name: '', type: spill-slot, offset: -12, size: 4, alignment: 4,
      stack-id: default, callee-saved-register: '$r6', callee-saved-restored: true,
      debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
  - { id: 3, name: '', type: spill-slot, offset: -16, size: 4, alignment: 4,
      stack-id: default, callee-saved-register: '$r5', callee-saved-restored: true,
      debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
  - { id: 4, name: '', type: spill-slot, offset: -20, size: 4, alignment: 4,
      stack-id: default, callee-saved-register: '$r4', callee-saved-restored: true,
      debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
  - { id: 5, name: '', type: spill-slot, offset: -24, size: 4, alignment: 4,
      stack-id: default, callee-saved-register: '$r11', callee-saved-restored: true,
      debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
  - { id: 6, name: '', type: spill-slot, offset: -28, size: 4, alignment: 4,
      stack-id: default, callee-saved-register: '$r9', callee-saved-restored: true,
      debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
  - { id: 7, name: '', type: spill-slot, offset: -32, size: 4, alignment: 4,
      stack-id: default, callee-saved-register: '$r8', callee-saved-restored: true,
      debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
callSites:       []
constants:       []
machineFunctionInfo: {}
body:             |
  ; CHECK-LABEL: name: unrolled_and_vector
  ; CHECK: bb.0.entry:
  ; CHECK-NEXT:   successors: %bb.11(0x30000000), %bb.1(0x50000000)
  ; CHECK-NEXT:   liveins: $lr, $r0, $r1, $r2, $r3, $r4, $r5, $r6, $r8, $r9, $r11
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT:   frame-setup tPUSH 14 /* CC::al */, $noreg, killed $r4, killed $r5, killed $r6, killed $lr, implicit-def $sp, implicit $sp
  ; CHECK-NEXT:   frame-setup CFI_INSTRUCTION def_cfa_offset 20
  ; CHECK-NEXT:   frame-setup CFI_INSTRUCTION offset $lr, -4
  ; CHECK-NEXT:   frame-setup CFI_INSTRUCTION offset $r7, -8
  ; CHECK-NEXT:   frame-setup CFI_INSTRUCTION offset $r6, -12
  ; CHECK-NEXT:   frame-setup CFI_INSTRUCTION offset $r5, -16
  ; CHECK-NEXT:   frame-setup CFI_INSTRUCTION offset $r4, -20
  ; CHECK-NEXT:   dead $r7 = frame-setup tADDrSPi $sp, 3, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   frame-setup CFI_INSTRUCTION def_cfa $r7, 8
  ; CHECK-NEXT:   $sp = frame-setup t2STMDB_UPD $sp, 14 /* CC::al */, $noreg, killed $r8, killed $r9, killed $r11
  ; CHECK-NEXT:   frame-setup CFI_INSTRUCTION offset $r11, -24
  ; CHECK-NEXT:   frame-setup CFI_INSTRUCTION offset $r9, -28
  ; CHECK-NEXT:   frame-setup CFI_INSTRUCTION offset $r8, -32
  ; CHECK-NEXT:   tCMPi8 renamable $r3, 0, 14 /* CC::al */, $noreg, implicit-def $cpsr
  ; CHECK-NEXT:   tBcc %bb.11, 0 /* CC::eq */, killed $cpsr
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT: bb.1.vector.memcheck:
  ; CHECK-NEXT:   successors: %bb.2(0x40000000), %bb.4(0x40000000)
  ; CHECK-NEXT:   liveins: $r0, $r1, $r2, $r3
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT:   renamable $r4, dead $cpsr = tADDrr renamable $r0, renamable $r3, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   renamable $r5, dead $cpsr = tADDrr renamable $r2, renamable $r3, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   tCMPr renamable $r4, renamable $r2, 14 /* CC::al */, $noreg, implicit-def $cpsr
  ; CHECK-NEXT:   renamable $lr = t2MOVi 1, 14 /* CC::al */, $noreg, $noreg
  ; CHECK-NEXT:   renamable $r12 = t2CSINC $zr, $zr, 9, implicit killed $cpsr
  ; CHECK-NEXT:   tCMPr killed renamable $r5, renamable $r0, 14 /* CC::al */, $noreg, implicit-def $cpsr
  ; CHECK-NEXT:   renamable $r6 = t2CSINC $zr, $zr, 9, implicit killed $cpsr
  ; CHECK-NEXT:   tCMPr killed renamable $r4, renamable $r1, 14 /* CC::al */, $noreg, implicit-def $cpsr
  ; CHECK-NEXT:   renamable $r5 = t2ADDrr renamable $r1, renamable $r3, 14 /* CC::al */, $noreg, $noreg
  ; CHECK-NEXT:   renamable $r4 = t2CSINC $zr, $zr, 9, implicit killed $cpsr
  ; CHECK-NEXT:   tCMPr killed renamable $r5, renamable $r0, 14 /* CC::al */, $noreg, implicit-def $cpsr
  ; CHECK-NEXT:   renamable $r5 = t2CSINC $zr, $zr, 9, implicit killed $cpsr
  ; CHECK-NEXT:   renamable $r5, dead $cpsr = tAND killed renamable $r5, killed renamable $r4, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   dead renamable $r5, $cpsr = tLSLri killed renamable $r5, 31, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   t2IT 0, 4, implicit-def $itstate
  ; CHECK-NEXT:   renamable $r6 = t2ANDrr killed renamable $r6, killed renamable $r12, 0 /* CC::eq */, $cpsr, $noreg, implicit killed $r6, implicit $itstate
  ; CHECK-NEXT:   dead renamable $r6 = t2LSLri killed renamable $r6, 31, 0 /* CC::eq */, killed $cpsr, def $cpsr, implicit killed $r6, implicit killed $itstate
  ; CHECK-NEXT:   tBcc %bb.4, 0 /* CC::eq */, killed $cpsr
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT: bb.2.for.body.preheader:
  ; CHECK-NEXT:   successors: %bb.3(0x40000000), %bb.6(0x40000000)
  ; CHECK-NEXT:   liveins: $lr, $r0, $r1, $r2, $r3
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT:   renamable $r4, dead $cpsr = tSUBi3 renamable $r3, 1, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   renamable $r12 = t2ANDri renamable $r3, 3, 14 /* CC::al */, $noreg, $noreg
  ; CHECK-NEXT:   tCMPi8 killed renamable $r4, 3, 14 /* CC::al */, $noreg, implicit-def $cpsr
  ; CHECK-NEXT:   tBcc %bb.6, 2 /* CC::hs */, killed $cpsr
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT: bb.3:
  ; CHECK-NEXT:   successors: %bb.8(0x80000000)
  ; CHECK-NEXT:   liveins: $r0, $r1, $r2, $r12
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT:   renamable $r3, dead $cpsr = tMOVi8 0, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   tB %bb.8, 14 /* CC::al */, $noreg
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT: bb.4.vector.ph:
  ; CHECK-NEXT:   successors: %bb.5(0x80000000)
  ; CHECK-NEXT:   liveins: $r0, $r1, $r2, $r3
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT:   $lr = MVE_DLSTP_8 killed renamable $r3
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT: bb.5.vector.body:
  ; CHECK-NEXT:   successors: %bb.5(0x7c000000), %bb.11(0x04000000)
  ; CHECK-NEXT:   liveins: $lr, $r0, $r1, $r2
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT:   renamable $r1, renamable $q0 = MVE_VLDRBU8_post killed renamable $r1, 16, 0, $noreg, $noreg :: (load (s128) from %ir.lsr.iv46, align 1)
  ; CHECK-NEXT:   renamable $r2, renamable $q1 = MVE_VLDRBU8_post killed renamable $r2, 16, 0, $noreg, $noreg :: (load (s128) from %ir.lsr.iv4749, align 1)
  ; CHECK-NEXT:   renamable $q0 = MVE_VADDi8 killed renamable $q1, killed renamable $q0, 0, $noreg, $noreg, undef renamable $q0
  ; CHECK-NEXT:   renamable $r0 = MVE_VSTRBU8_post killed renamable $q0, killed renamable $r0, 16, 0, killed $noreg, $noreg :: (store (s128) into %ir.lsr.iv5052, align 1)
  ; CHECK-NEXT:   $lr = MVE_LETP killed renamable $lr, %bb.5
  ; CHECK-NEXT:   tB %bb.11, 14 /* CC::al */, $noreg
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT: bb.6.for.body.preheader.new:
  ; CHECK-NEXT:   successors: %bb.7(0x80000000)
  ; CHECK-NEXT:   liveins: $lr, $r0, $r1, $r2, $r3, $r12
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT:   renamable $r3 = t2BICri killed renamable $r3, 3, 14 /* CC::al */, $noreg, $noreg
  ; CHECK-NEXT:   renamable $r3, dead $cpsr = tSUBi8 killed renamable $r3, 4, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   renamable $lr = nuw nsw t2ADDrs killed renamable $lr, killed renamable $r3, 19, 14 /* CC::al */, $noreg, $noreg
  ; CHECK-NEXT:   renamable $r3, dead $cpsr = tMOVi8 0, 14 /* CC::al */, $noreg
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT: bb.7.for.body:
  ; CHECK-NEXT:   successors: %bb.7(0x7c000000), %bb.8(0x04000000)
  ; CHECK-NEXT:   liveins: $lr, $r0, $r1, $r2, $r3, $r12
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT:   renamable $r4 = tLDRBr renamable $r1, $r3, 14 /* CC::al */, $noreg :: (load (s8) from %ir.scevgep2453)
  ; CHECK-NEXT:   renamable $r9 = t2ADDrr renamable $r1, renamable $r3, 14 /* CC::al */, $noreg, $noreg
  ; CHECK-NEXT:   renamable $r5 = tLDRBr renamable $r2, $r3, 14 /* CC::al */, $noreg :: (load (s8) from %ir.scevgep2854)
  ; CHECK-NEXT:   renamable $r6, dead $cpsr = tADDrr renamable $r2, renamable $r3, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   renamable $r4 = tADDhirr killed renamable $r4, killed renamable $r5, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   tSTRBr killed renamable $r4, renamable $r0, $r3, 14 /* CC::al */, $noreg :: (store (s8) into %ir.scevgep3255)
  ; CHECK-NEXT:   renamable $r8 = t2LDRBi12 renamable $r9, 1, 14 /* CC::al */, $noreg :: (load (s8) from %ir.scevgep40)
  ; CHECK-NEXT:   renamable $r5 = tLDRBi renamable $r6, 1, 14 /* CC::al */, $noreg :: (load (s8) from %ir.scevgep42)
  ; CHECK-NEXT:   renamable $r8 = tADDhirr killed renamable $r8, killed renamable $r5, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   renamable $r5, dead $cpsr = tADDrr renamable $r0, renamable $r3, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   renamable $r3, dead $cpsr = nuw tADDi8 killed renamable $r3, 4, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   t2STRBi12 killed renamable $r8, renamable $r5, 1, 14 /* CC::al */, $noreg :: (store (s8) into %ir.scevgep44)
  ; CHECK-NEXT:   renamable $r8 = t2LDRBi12 renamable $r9, 2, 14 /* CC::al */, $noreg :: (load (s8) from %ir.scevgep34)
  ; CHECK-NEXT:   renamable $r4 = tLDRBi renamable $r6, 2, 14 /* CC::al */, $noreg :: (load (s8) from %ir.scevgep36)
  ; CHECK-NEXT:   renamable $r4 = tADDhirr killed renamable $r4, killed renamable $r8, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   tSTRBi killed renamable $r4, renamable $r5, 2, 14 /* CC::al */, $noreg :: (store (s8) into %ir.scevgep38)
  ; CHECK-NEXT:   renamable $r4 = t2LDRBi12 killed renamable $r9, 3, 14 /* CC::al */, $noreg :: (load (s8) from %ir.scevgep22)
  ; CHECK-NEXT:   renamable $r6 = tLDRBi killed renamable $r6, 3, 14 /* CC::al */, $noreg :: (load (s8) from %ir.scevgep26)
  ; CHECK-NEXT:   renamable $r4 = tADDhirr killed renamable $r4, killed renamable $r6, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   tSTRBi killed renamable $r4, killed renamable $r5, 3, 14 /* CC::al */, $noreg :: (store (s8) into %ir.scevgep30)
  ; CHECK-NEXT:   $lr = t2LEUpdate killed renamable $lr, %bb.7
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT: bb.8.for.cond.cleanup.loopexit.unr-lcssa:
  ; CHECK-NEXT:   successors: %bb.11(0x30000000), %bb.9(0x50000000)
  ; CHECK-NEXT:   liveins: $r0, $r1, $r2, $r3, $r12
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT:   t2CMPri renamable $r12, 0, 14 /* CC::al */, $noreg, implicit-def $cpsr
  ; CHECK-NEXT:   tBcc %bb.11, 0 /* CC::eq */, killed $cpsr
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT: bb.9.for.body.epil:
  ; CHECK-NEXT:   successors: %bb.11(0x40000000), %bb.10(0x40000000)
  ; CHECK-NEXT:   liveins: $r0, $r1, $r2, $r3, $r12
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT:   renamable $r6 = tLDRBr renamable $r1, $r3, 14 /* CC::al */, $noreg :: (load (s8) from %ir.arrayidx.epil)
  ; CHECK-NEXT:   t2CMPri renamable $r12, 1, 14 /* CC::al */, $noreg, implicit-def $cpsr
  ; CHECK-NEXT:   renamable $r5 = tLDRBr renamable $r2, $r3, 14 /* CC::al */, $noreg :: (load (s8) from %ir.arrayidx1.epil)
  ; CHECK-NEXT:   renamable $r6 = tADDhirr killed renamable $r6, killed renamable $r5, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   tSTRBr killed renamable $r6, renamable $r0, $r3, 14 /* CC::al */, $noreg :: (store (s8) into %ir.arrayidx4.epil)
  ; CHECK-NEXT:   tBcc %bb.11, 0 /* CC::eq */, killed $cpsr
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT: bb.10.for.body.epil.1:
  ; CHECK-NEXT:   successors: %bb.11(0x40000000), %bb.12(0x40000000)
  ; CHECK-NEXT:   liveins: $r0, $r1, $r2, $r3, $r12
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT:   renamable $r6, dead $cpsr = nuw tADDi3 renamable $r3, 1, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   t2CMPri killed renamable $r12, 2, 14 /* CC::al */, $noreg, implicit-def $cpsr
  ; CHECK-NEXT:   renamable $r5 = tLDRBr renamable $r1, $r6, 14 /* CC::al */, $noreg :: (load (s8) from %ir.arrayidx.epil.1)
  ; CHECK-NEXT:   renamable $r4 = tLDRBr renamable $r2, $r6, 14 /* CC::al */, $noreg :: (load (s8) from %ir.arrayidx1.epil.1)
  ; CHECK-NEXT:   renamable $r5 = tADDhirr killed renamable $r5, killed renamable $r4, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   tSTRBr killed renamable $r5, renamable $r0, killed $r6, 14 /* CC::al */, $noreg :: (store (s8) into %ir.arrayidx4.epil.1)
  ; CHECK-NEXT:   tBcc %bb.12, 1 /* CC::ne */, killed $cpsr
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT: bb.11.for.cond.cleanup:
  ; CHECK-NEXT:   $sp = t2LDMIA_UPD $sp, 14 /* CC::al */, $noreg, def $r8, def $r9, def $r11
  ; CHECK-NEXT:   tPOP_RET 14 /* CC::al */, $noreg, def $r4, def $r5, def $r6, def $r7, def $pc
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT: bb.12.for.body.epil.2:
  ; CHECK-NEXT:   liveins: $r0, $r1, $r2, $r3
  ; CHECK-NEXT: {{  $}}
  ; CHECK-NEXT:   renamable $r3, dead $cpsr = nuw tADDi8 killed renamable $r3, 2, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   renamable $r1 = tLDRBr killed renamable $r1, $r3, 14 /* CC::al */, $noreg :: (load (s8) from %ir.arrayidx.epil.2)
  ; CHECK-NEXT:   renamable $r2 = tLDRBr killed renamable $r2, $r3, 14 /* CC::al */, $noreg :: (load (s8) from %ir.arrayidx1.epil.2)
  ; CHECK-NEXT:   renamable $r1 = tADDhirr killed renamable $r1, killed renamable $r2, 14 /* CC::al */, $noreg
  ; CHECK-NEXT:   tSTRBr killed renamable $r1, killed renamable $r0, killed $r3, 14 /* CC::al */, $noreg :: (store (s8) into %ir.arrayidx4.epil.2)
  ; CHECK-NEXT:   $sp = t2LDMIA_UPD $sp, 14 /* CC::al */, $noreg, def $r8, def $r9, def $r11
  ; CHECK-NEXT:   tPOP_RET 14 /* CC::al */, $noreg, def $r4, def $r5, def $r6, def $r7, def $pc
  bb.0.entry:
    successors: %bb.11(0x30000000), %bb.1(0x50000000)
    liveins: $r0, $r1, $r2, $r3, $r4, $r5, $r6, $lr, $r8, $r9, $r11

    frame-setup tPUSH 14, $noreg, killed $r4, killed $r5, killed $r6, killed $lr, implicit-def $sp, implicit $sp
    frame-setup CFI_INSTRUCTION def_cfa_offset 20
    frame-setup CFI_INSTRUCTION offset $lr, -4
    frame-setup CFI_INSTRUCTION offset $r7, -8
    frame-setup CFI_INSTRUCTION offset $r6, -12
    frame-setup CFI_INSTRUCTION offset $r5, -16
    frame-setup CFI_INSTRUCTION offset $r4, -20
    $r7 = frame-setup tADDrSPi $sp, 3, 14, $noreg
    frame-setup CFI_INSTRUCTION def_cfa $r7, 8
    $sp = frame-setup t2STMDB_UPD $sp, 14, $noreg, killed $r8, killed $r9, killed $r11
    frame-setup CFI_INSTRUCTION offset $r11, -24
    frame-setup CFI_INSTRUCTION offset $r9, -28
    frame-setup CFI_INSTRUCTION offset $r8, -32
    tCMPi8 renamable $r3, 0, 14, $noreg, implicit-def $cpsr
    tBcc %bb.11, 0, killed $cpsr

  bb.1.vector.memcheck:
    successors: %bb.2(0x40000000), %bb.4(0x40000000)
    liveins: $r0, $r1, $r2, $r3

    renamable $r4, dead $cpsr = tADDrr renamable $r0, renamable $r3, 14, $noreg
    renamable $r5, dead $cpsr = tADDrr renamable $r2, renamable $r3, 14, $noreg
    tCMPr renamable $r4, renamable $r2, 14, $noreg, implicit-def $cpsr
    renamable $lr = t2MOVi 1, 14, $noreg, $noreg
    renamable $r12 = t2CSINC $zr, $zr, 9, implicit killed $cpsr
    tCMPr killed renamable $r5, renamable $r0, 14, $noreg, implicit-def $cpsr
    renamable $r6 = t2CSINC $zr, $zr, 9, implicit killed $cpsr
    tCMPr killed renamable $r4, renamable $r1, 14, $noreg, implicit-def $cpsr
    renamable $r5 = t2ADDrr renamable $r1, renamable $r3, 14, $noreg, $noreg
    renamable $r4 = t2CSINC $zr, $zr, 9, implicit killed $cpsr
    tCMPr killed renamable $r5, renamable $r0, 14, $noreg, implicit-def $cpsr
    renamable $r5 = t2CSINC $zr, $zr, 9, implicit killed $cpsr
    renamable $r5, dead $cpsr = tAND killed renamable $r5, killed renamable $r4, 14, $noreg
    dead renamable $r5, $cpsr = tLSLri killed renamable $r5, 31, 14, $noreg
    t2IT 0, 4, implicit-def $itstate
    renamable $r6 = t2ANDrr killed renamable $r6, killed renamable $r12, 0, $cpsr, $noreg, implicit $r6, implicit $itstate
    dead renamable $r6 = t2LSLri killed renamable $r6, 31, 0, killed $cpsr, def $cpsr, implicit killed $r6, implicit killed $itstate
    tBcc %bb.4, 0, killed $cpsr

  bb.2.for.body.preheader:
    successors: %bb.3(0x40000000), %bb.6(0x40000000)
    liveins: $lr, $r0, $r1, $r2, $r3

    renamable $r4, dead $cpsr = tSUBi3 renamable $r3, 1, 14, $noreg
    renamable $r12 = t2ANDri renamable $r3, 3, 14, $noreg, $noreg
    tCMPi8 killed renamable $r4, 3, 14, $noreg, implicit-def $cpsr
    tBcc %bb.6, 2, killed $cpsr

  bb.3:
    successors: %bb.8(0x80000000)
    liveins: $r0, $r1, $r2, $r12

    renamable $r3, dead $cpsr = tMOVi8 0, 14, $noreg
    tB %bb.8, 14, $noreg

  bb.4.vector.ph:
    successors: %bb.5(0x80000000)
    liveins: $lr, $r0, $r1, $r2, $r3

    renamable $r6 = t2ADDri renamable $r3, 15, 14, $noreg, $noreg
    renamable $r6 = t2BICri killed renamable $r6, 15, 14, $noreg, $noreg
    renamable $r6, dead $cpsr = tSUBi8 killed renamable $r6, 16, 14, $noreg
    renamable $lr = nuw nsw t2ADDrs killed renamable $lr, killed renamable $r6, 35, 14, $noreg, $noreg
    $lr = t2DoLoopStart renamable $lr

  bb.5.vector.body:
    successors: %bb.5(0x7c000000), %bb.11(0x04000000)
    liveins: $lr, $r0, $r1, $r2, $r3

    renamable $vpr = MVE_VCTP8 renamable $r3, 0, $noreg, $noreg
    renamable $r3, dead $cpsr = tSUBi8 killed renamable $r3, 16, 14, $noreg
    MVE_VPST 4, implicit $vpr
    renamable $r1, renamable $q0 = MVE_VLDRBU8_post killed renamable $r1, 16, 1, renamable $vpr, $noreg :: (load (s128) from %ir.lsr.iv46, align 1)
    renamable $r2, renamable $q1 = MVE_VLDRBU8_post killed renamable $r2, 16, 1, renamable $vpr, $noreg :: (load (s128) from %ir.lsr.iv4749, align 1)
    renamable $lr = t2LoopDec killed renamable $lr, 1
    renamable $q0 = MVE_VADDi8 killed renamable $q1, killed renamable $q0, 0, $noreg, $noreg, undef renamable $q0
    MVE_VPST 8, implicit $vpr
    renamable $r0 = MVE_VSTRBU8_post killed renamable $q0, killed renamable $r0, 16, 1, killed renamable $vpr, $noreg :: (store (s128) into %ir.lsr.iv5052, align 1)
    t2LoopEnd renamable $lr, %bb.5, implicit-def dead $cpsr
    tB %bb.11, 14, $noreg

  bb.6.for.body.preheader.new:
    successors: %bb.7(0x80000000)
    liveins: $lr, $r0, $r1, $r2, $r3, $r12

    renamable $r3 = t2BICri killed renamable $r3, 3, 14, $noreg, $noreg
    renamable $r3, dead $cpsr = tSUBi8 killed renamable $r3, 4, 14, $noreg
    renamable $lr = nuw nsw t2ADDrs killed renamable $lr, killed renamable $r3, 19, 14, $noreg, $noreg
    renamable $r3, dead $cpsr = tMOVi8 0, 14, $noreg
    $lr = t2DoLoopStart renamable $lr

  bb.7.for.body:
    successors: %bb.7(0x7c000000), %bb.8(0x04000000)
    liveins: $lr, $r0, $r1, $r2, $r3, $r12

    renamable $r4 = tLDRBr renamable $r1, $r3, 14, $noreg :: (load (s8) from %ir.scevgep2453)
    renamable $r9 = t2ADDrr renamable $r1, renamable $r3, 14, $noreg, $noreg
    renamable $r5 = tLDRBr renamable $r2, $r3, 14, $noreg :: (load (s8) from %ir.scevgep2854)
    renamable $r6, dead $cpsr = tADDrr renamable $r2, renamable $r3, 14, $noreg
    renamable $lr = t2LoopDec killed renamable $lr, 1
    renamable $r4 = tADDhirr killed renamable $r4, killed renamable $r5, 14, $noreg
    tSTRBr killed renamable $r4, renamable $r0, $r3, 14, $noreg :: (store (s8) into %ir.scevgep3255)
    renamable $r8 = t2LDRBi12 renamable $r9, 1, 14, $noreg :: (load (s8) from %ir.scevgep40)
    renamable $r5 = tLDRBi renamable $r6, 1, 14, $noreg :: (load (s8) from %ir.scevgep42)
    renamable $r8 = tADDhirr killed renamable $r8, killed renamable $r5, 14, $noreg
    renamable $r5, dead $cpsr = tADDrr renamable $r0, renamable $r3, 14, $noreg
    renamable $r3, dead $cpsr = nuw tADDi8 killed renamable $r3, 4, 14, $noreg
    t2STRBi12 killed renamable $r8, renamable $r5, 1, 14, $noreg :: (store (s8) into %ir.scevgep44)
    renamable $r8 = t2LDRBi12 renamable $r9, 2, 14, $noreg :: (load (s8) from %ir.scevgep34)
    renamable $r4 = tLDRBi renamable $r6, 2, 14, $noreg :: (load (s8) from %ir.scevgep36)
    renamable $r4 = tADDhirr killed renamable $r4, killed renamable $r8, 14, $noreg
    tSTRBi killed renamable $r4, renamable $r5, 2, 14, $noreg :: (store (s8) into %ir.scevgep38)
    renamable $r4 = t2LDRBi12 killed renamable $r9, 3, 14, $noreg :: (load (s8) from %ir.scevgep22)
    renamable $r6 = tLDRBi killed renamable $r6, 3, 14, $noreg :: (load (s8) from %ir.scevgep26)
    renamable $r4 = tADDhirr killed renamable $r4, killed renamable $r6, 14, $noreg
    tSTRBi killed renamable $r4, killed renamable $r5, 3, 14, $noreg :: (store (s8) into %ir.scevgep30)
    t2LoopEnd renamable $lr, %bb.7, implicit-def dead $cpsr
    tB %bb.8, 14, $noreg

  bb.8.for.cond.cleanup.loopexit.unr-lcssa:
    successors: %bb.11(0x30000000), %bb.9(0x50000000)
    liveins: $r0, $r1, $r2, $r3, $r12

    t2CMPri renamable $r12, 0, 14, $noreg, implicit-def $cpsr
    tBcc %bb.11, 0, killed $cpsr

  bb.9.for.body.epil:
    successors: %bb.11(0x40000000), %bb.10(0x40000000)
    liveins: $r0, $r1, $r2, $r3, $r12

    renamable $r6 = tLDRBr renamable $r1, $r3, 14, $noreg :: (load (s8) from %ir.arrayidx.epil)
    t2CMPri renamable $r12, 1, 14, $noreg, implicit-def $cpsr
    renamable $r5 = tLDRBr renamable $r2, $r3, 14, $noreg :: (load (s8) from %ir.arrayidx1.epil)
    renamable $r6 = tADDhirr killed renamable $r6, killed renamable $r5, 14, $noreg
    tSTRBr killed renamable $r6, renamable $r0, $r3, 14, $noreg :: (store (s8) into %ir.arrayidx4.epil)
    tBcc %bb.11, 0, killed $cpsr

  bb.10.for.body.epil.1:
    successors: %bb.11(0x40000000), %bb.12(0x40000000)
    liveins: $r0, $r1, $r2, $r3, $r12

    renamable $r6, dead $cpsr = nuw tADDi3 renamable $r3, 1, 14, $noreg
    t2CMPri killed renamable $r12, 2, 14, $noreg, implicit-def $cpsr
    renamable $r5 = tLDRBr renamable $r1, $r6, 14, $noreg :: (load (s8) from %ir.arrayidx.epil.1)
    renamable $r4 = tLDRBr renamable $r2, $r6, 14, $noreg :: (load (s8) from %ir.arrayidx1.epil.1)
    renamable $r5 = tADDhirr killed renamable $r5, killed renamable $r4, 14, $noreg
    tSTRBr killed renamable $r5, renamable $r0, killed $r6, 14, $noreg :: (store (s8) into %ir.arrayidx4.epil.1)
    tBcc %bb.12, 1, killed $cpsr

  bb.11.for.cond.cleanup:
    $sp = t2LDMIA_UPD $sp, 14, $noreg, def $r8, def $r9, def $r11
    tPOP_RET 14, $noreg, def $r4, def $r5, def $r6, def $r7, def $pc

  bb.12.for.body.epil.2:
    liveins: $r0, $r1, $r2, $r3

    renamable $r3, dead $cpsr = nuw tADDi8 killed renamable $r3, 2, 14, $noreg
    renamable $r1 = tLDRBr killed renamable $r1, $r3, 14, $noreg :: (load (s8) from %ir.arrayidx.epil.2)
    renamable $r2 = tLDRBr killed renamable $r2, $r3, 14, $noreg :: (load (s8) from %ir.arrayidx1.epil.2)
    renamable $r1 = tADDhirr killed renamable $r1, killed renamable $r2, 14, $noreg
    tSTRBr killed renamable $r1, killed renamable $r0, killed $r3, 14, $noreg :: (store (s8) into %ir.arrayidx4.epil.2)
    $sp = t2LDMIA_UPD $sp, 14, $noreg, def $r8, def $r9, def $r11
    tPOP_RET 14, $noreg, def $r4, def $r5, def $r6, def $r7, def $pc

...