[AMDGPU] Wave and register controls

[• tstellarAMD]

Summary

• Implemented amdgpu-flat-work-group-size attribute
• Implemented amdgpu-num-active-waves-per-eu attribute
• Implemented amdgpu-num-sgpr attribute
• Implemented amdgpu-num-vgpr attribute
• Dynamic LDS constraints are in a separate patch

[kzhuravl]

lib/Target/AMDGPU/AMDGPUSubtarget.h

445

The reason I put this function here is because it either returns flat work group size for a given function or subtarget's defaults, but I am not sure if this is the right place.

455

The reason I put this function here is because it either returns number of active waves per execution unit for a given function or subtarget's defaults, but I am not sure if this is the right place.

lib/Target/AMDGPU/SIRegisterInfo.h

226

Is there a better place for this function?

261

Is there a better place for this function?

[arsenm]

Can you shrink the test cases? They are pretty big for not testing the content

lib/Target/AMDGPU/AMDGPUSubtarget.cpp

269–285

More DiagnosticInfos. These should also provide the function so clang will be able to provide a more useful debug location

294–309

Can you put blank lines after code before a comment, it's hard to read with it this dense

327–329

This should use DiagnosticInfoUnsupported. Although on second thought these should probably just clamp and diagnosis in clang

333–335

Ditto

342–345

Ditto

lib/Target/AMDGPU/AMDGPUSubtarget.h

445

We probably shouldn't be using IR classes in the subtarget, since there's a desire to someday be able to throw away the IR during codegen, but since there's already one thing already doing this here is fine for now

lib/Target/AMDGPU/SIInstrInfo.cpp

32

This doesn't look right

lib/Target/AMDGPU/SIMachineFunctionInfo.cpp

51–52

I think just 0, 0 or 0u, 0u should work

lib/Target/AMDGPU/SIMachineFunctionInfo.h

66

Separating line

lib/Target/AMDGPU/SIRegisterInfo.cpp

214

SIInstrInfo

236

You can remove the cast

lib/Target/AMDGPU/SIRegisterInfo.h

28

I don't think the RegisterInfo should have a reference to the Subtarget

test/CodeGen/AMDGPU/attr-amdgpu-flat-work-group-size-many-regs.ll

7–18

Can you run instnamer on this and use CHECK-LABEL

test/CodeGen/AMDGPU/attr-amdgpu-num-active-waves-per-eu-invalid.ll

8

Some tests with unparsable sizes would be good

[kzhuravl]

Review feedback (I will add more tests a bit later today)

[kzhuravl]

lib/Target/AMDGPU/AMDGPUSubtarget.cpp

269–285

Changed to clamp

[arsenm]

lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp

502

Line before comment

lib/Target/AMDGPU/AMDGPUAsmPrinter.h

77

Line before comment

lib/Target/AMDGPU/AMDGPUSubtarget.cpp

247

You can use those as the initializer value instead of the =

lib/Target/AMDGPU/SIMachineFunctionInfo.h

334–347

The Returns should be \returns to doxygenify, same for the rest of these

lib/Target/AMDGPU/SIRegisterInfo.h

190

More \returns

[kzhuravl]

Review feedback + Added tests

[kzhuravl]

Add missing tests

[kzhuravl]

Trim down large test cases

[• tstellarAMD]

lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp

128

Mesa is using this attribute, so we still need to support it. We should also leave some tests which use it.

[kzhuravl]

Review feedback

[• tstellarAMD]

LGTM. We just need to resolve the scheduler bug first, and then I think this can be committed.

lib/Target/AMDGPU/SIRegisterInfo.cpp

153–155

This actually uncovers a bug, because the scheduler will subtract the number of reserved registers from the value returned by this function, which in some cases gives us a negative pressure set limit. I think we need to remove this function before this patch can be committed. I will look into this.

[kzhuravl]

Ok, thanks for taking a look!

[kzhuravl]

Rebase

[• tstellarAMD]

lib/Target/AMDGPU/SIRegisterInfo.cpp

153–155

The fix for this is more complicated than I thought. I think we should fix this by maintaining the current behavior of this function.

This means that if the user does not specify any attributes that would limit the number of registers, then SGPRLimit and VGPRLimit should be set to the 10 waves per CU limits (24 VGPR and .48/80 SGPRS depending on the target).

If the user does specify attributes to limit the number of registers the function should return AMDGPURegisterInfo::getRegPressureSetLimit().

[kzhuravl]

Review feedback

[kzhuravl]

Ping

[kzhuravl]

Rebase + Fix WorkGroupSize -> WavesPerEU calculation

[yaxunl]

test/CodeGen/AMDGPU/attr-amdgpu-waves-per-eu.ll

75 ↗

(On Diff #69960)

Is this extra?

101 ↗

(On Diff #69960)

extra line?

[• tstellarAMD]

Fixed some bugs with mesa and integrate with the new scheduler.

[arsenm]

LGTM

lib/Target/AMDGPU/SIRegisterInfo.cpp

915–916

This should be narrowed to just if the xnack feature is enabled

[kzhuravl]

LGTM

[• tstellarAMD]

lib/Target/AMDGPU/SIRegisterInfo.cpp

915–916

We have to use 6 here for all VI, because FLAT_SCRATCH is located 2 sgprs below XNACK, so if you use FLAT_SCRATCH, you have to allocate space for XNACK too, even if the device does not support it.

lib/Target/AMDGPU/AMDGPUAsmPrinter.h

SIProgramInfo() :
       ScratchSize(0),
       ComputePGMRSrc1(0),
       LDSBlocks(0),
       ScratchBlocks(0),
       ComputePGMRSrc2(0),
       NumVGPR(0),
       NumSGPR(0),
       FlatUsed(false),
+      NumSGPRsForNumActiveWavesPerEU(0),
+      NumVGPRsForNumActiveWavesPerEU(0),
       ReservedVGPRFirst(0),
       ReservedVGPRCount(0),
       VCCUsed(false),
       CodeLen(0) {}
 
     // Fields set in PGM_RSRC1 pm4 packet.
     uint32_t VGPRBlocks;
     uint32_t SGPRBlocks;
struct SIProgramInfo {
 
     uint64_t ComputePGMRSrc2;
 
     uint32_t NumVGPR;
     uint32_t NumSGPR;
     uint32_t LDSSize;
     bool FlatUsed;
 
+    // Number of SGPRs that meets number of active waves per execution unit
+    // request.
+    uint32_t NumSGPRsForNumActiveWavesPerEU;
+    // Number of VGPRs that meets number of active waves per execution unit
+    // request.
+    uint32_t NumVGPRsForNumActiveWavesPerEU;
+
     // If ReservedVGPRCount is 0 then must be 0. Otherwise, this is the first
     // fixed VGPR number reserved.
     uint16_t ReservedVGPRFirst;
     // The number of consecutive VGPRs reserved.
     uint16_t ReservedVGPRCount;
 
     // Bonus information for debugging.
     bool VCCUsed;

lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp

if (STM.getGeneration() >= AMDGPUSubtarget::SOUTHERN_ISLANDS) {
                                   false);
       OutStreamer->emitRawComment(" IeeeMode: " + Twine(KernelInfo.IEEEMode),
                                   false);
       OutStreamer->emitRawComment(" ScratchSize: " + Twine(KernelInfo.ScratchSize),
                                   false);
       OutStreamer->emitRawComment(" LDSByteSize: " + Twine(KernelInfo.LDSSize) +
                                   " bytes/workgroup (compile time only)", false);
 
+      OutStreamer->emitRawComment(" SGPRBlocks: " +
+                                  Twine(KernelInfo.SGPRBlocks), false);
+      OutStreamer->emitRawComment(" VGPRBlocks: " +
+                                  Twine(KernelInfo.VGPRBlocks), false);
+
+      OutStreamer->emitRawComment(" NumSGPRsForNumActiveWavesPerEU: " +
+                                  Twine(KernelInfo.NumSGPRsForNumActiveWavesPerEU), false);
+      OutStreamer->emitRawComment(" NumVGPRsForNumActiveWavesPerEU: " +
+                                  Twine(KernelInfo.NumVGPRsForNumActiveWavesPerEU), false);
+
       OutStreamer->emitRawComment(" ReservedVGPRFirst: " + Twine(KernelInfo.ReservedVGPRFirst),
                                   false);
       OutStreamer->emitRawComment(" ReservedVGPRCount: " + Twine(KernelInfo.ReservedVGPRCount),
                                   false);
 
       OutStreamer->emitRawComment(" COMPUTE_PGM_RSRC2:USER_SGPR: " +
                                   Twine(G_00B84C_USER_SGPR(KernelInfo.ComputePGMRSrc2)),
                                   false);
void AMDGPUAsmPrinter::getSIProgramInfo(SIProgramInfo &ProgInfo,
   } else {
     if (STM.isXNACKEnabled())
       ExtraSGPRs = 4;
 
     if (FlatUsed)
       ExtraSGPRs = 6;
   }
 
-  MaxSGPR += ExtraSGPRs;
-
   // Record first reserved register and reserved register count fields, and
   // update max register counts if "amdgpu-debugger-reserve-regs" attribute was
-  // specified.
-  if (STM.debuggerReserveRegs()) {
-    ProgInfo.ReservedVGPRFirst = MaxVGPR + 1;
-    ProgInfo.ReservedVGPRCount = MFI->getDebuggerReservedVGPRCount();
-    MaxVGPR += MFI->getDebuggerReservedVGPRCount();
-  }
+  // requested.
+  ProgInfo.ReservedVGPRFirst = STM.debuggerReserveRegs() ? MaxVGPR + 1 : 0;
+  ProgInfo.ReservedVGPRCount = RI->getDebuggerReservedNumVGPRs();
+
+  MaxSGPR += ExtraSGPRs;
+  MaxVGPR += RI->getDebuggerReservedNumVGPRs();
 
   // We found the maximum register index. They start at 0, so add one to get the
   // number of registers.
   ProgInfo.NumVGPR = MaxVGPR + 1;
   ProgInfo.NumSGPR = MaxSGPR + 1;
 
+  // Adjust number of registers used to meet default/requested minimum/maximum
+  // number of active waves per execution unit request.
+  ProgInfo.NumSGPRsForNumActiveWavesPerEU = std::max(
+    ProgInfo.NumSGPR, RI->getMinNumSGPRs(MFI->getMaxNumActiveWavesPerEU()));
+  ProgInfo.NumVGPRsForNumActiveWavesPerEU = std::max(
+    ProgInfo.NumVGPR, RI->getMinNumVGPRs(MFI->getMaxNumActiveWavesPerEU()));
+
   if (STM.hasSGPRInitBug()) {
     if (ProgInfo.NumSGPR > AMDGPUSubtarget::FIXED_SGPR_COUNT_FOR_INIT_BUG) {
       LLVMContext &Ctx = MF.getFunction()->getContext();
       DiagnosticInfoResourceLimit Diag(*MF.getFunction(),
                                        "SGPRs with SGPR init bug",
                                        ProgInfo.NumSGPR, DS_Error);
       Ctx.diagnose(Diag);
     }
 
     ProgInfo.NumSGPR = AMDGPUSubtarget::FIXED_SGPR_COUNT_FOR_INIT_BUG;
+    ProgInfo.NumSGPRsForNumActiveWavesPerEU =
+      AMDGPUSubtarget::FIXED_SGPR_COUNT_FOR_INIT_BUG;
   }
 
   if (MFI->NumUserSGPRs > STM.getMaxNumUserSGPRs()) {
     LLVMContext &Ctx = MF.getFunction()->getContext();
     DiagnosticInfoResourceLimit Diag(*MF.getFunction(), "user SGPRs",
                                      MFI->NumUserSGPRs, DS_Error);
     Ctx.diagnose(Diag);
   }
 
   if (MFI->LDSSize > static_cast<unsigned>(STM.getLocalMemorySize())) {
     LLVMContext &Ctx = MF.getFunction()->getContext();
     DiagnosticInfoResourceLimit Diag(*MF.getFunction(), "local memory",
                                      MFI->LDSSize, DS_Error);
     Ctx.diagnose(Diag);
   }
 
-  ProgInfo.VGPRBlocks = (ProgInfo.NumVGPR - 1) / 4;
-  ProgInfo.SGPRBlocks = (ProgInfo.NumSGPR - 1) / 8;
+  // SGPRBlocks is actual number of SGPR blocks minus 1.
+  ProgInfo.SGPRBlocks = alignTo(ProgInfo.NumSGPRsForNumActiveWavesPerEU,
+                                RI->getSGPRAllocGranule());
+  ProgInfo.SGPRBlocks = ProgInfo.SGPRBlocks / RI->getSGPRAllocGranule() - 1;
+  // VGPRBlocks is actual number of VGPR blocks minus 1.
+  ProgInfo.VGPRBlocks = alignTo(ProgInfo.NumVGPRsForNumActiveWavesPerEU,
+                                RI->getVGPRAllocGranule());
+  ProgInfo.VGPRBlocks = ProgInfo.VGPRBlocks / RI->getVGPRAllocGranule() - 1;
+
   // Set the value to initialize FP_ROUND and FP_DENORM parts of the mode
   // register.
   ProgInfo.FloatMode = getFPMode(MF);
 
   ProgInfo.IEEEMode = 0;
 
   // Make clamp modifier on NaN input returns 0.
   ProgInfo.DX10Clamp = 1;
void AMDGPUAsmPrinter::getSIProgramInfo(SIProgramInfo &ProgInfo,
   if (STM.getGeneration() < AMDGPUSubtarget::SEA_ISLANDS) {
     // LDS is allocated in 64 dword blocks.
     LDSAlignShift = 8;
   } else {
     // LDS is allocated in 128 dword blocks.
     LDSAlignShift = 9;
   }
 
-  unsigned LDSSpillSize = MFI->LDSWaveSpillSize *
-                          MFI->getMaximumWorkGroupSize(MF);
+  unsigned LDSSpillSize =
+    MFI->LDSWaveSpillSize * MFI->getMaxFlatWorkGroupSize();
 
   ProgInfo.LDSSize = MFI->LDSSize + LDSSpillSize;
   ProgInfo.LDSBlocks =
       alignTo(ProgInfo.LDSSize, 1ULL << LDSAlignShift) >> LDSAlignShift;
 
   // Scratch is allocated in 256 dword blocks.
   unsigned ScratchAlignShift = 10;
   // We need to program the hardware with the amount of scratch memory that

lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp

bool AMDGPUPromoteAlloca::runOnFunction(Function &F) {
 
   unsigned MaxOccupancy = ST.getOccupancyWithLocalMemSize(CurrentLocalMemUsage);
 
   // Restrict local memory usage so that we don't drastically reduce occupancy,
   // unless it is already significantly reduced.
 
   // TODO: Have some sort of hint or other heuristics to guess occupancy based
   // on other factors..
-  unsigned OccupancyHint
-    = AMDGPU::getIntegerAttribute(F, "amdgpu-max-waves-per-eu", 0);
+  unsigned OccupancyHint = ST.getNumActiveWavesPerEU(F).second;
   if (OccupancyHint == 0)
     OccupancyHint = 7;
 
   // Clamp to max value.
-  OccupancyHint = std::min(OccupancyHint, ST.getMaxWavesPerCU());
+  OccupancyHint = std::min(OccupancyHint, ST.getMaxNumActiveWavesPerEU());
 
   // Check the hint but ignore it if it's obviously wrong from the existing LDS
   // usage.
   MaxOccupancy = std::min(OccupancyHint, MaxOccupancy);
 
 
   // Round up to the next tier of usage.
   unsigned MaxSizeWithWaveCount
void AMDGPUPromoteAlloca::handleAlloca(AllocaInst &I) {
   DEBUG(dbgs() << "Trying to promote " << I << '\n');
 
   if (tryPromoteAllocaToVector(&I)) {
     DEBUG(dbgs() << " alloca is not a candidate for vectorization.\n");
     return;
   }
 
   const Function &ContainingFunction = *I.getParent()->getParent();
+  const AMDGPUSubtarget &ST =
+    TM->getSubtarget<AMDGPUSubtarget>(ContainingFunction);
 
   // FIXME: We should also try to get this value from the reqd_work_group_size
   // function attribute if it is available.
-  unsigned WorkGroupSize = AMDGPU::getMaximumWorkGroupSize(ContainingFunction);
+  unsigned WorkGroupSize = ST.getFlatWorkGroupSizes(ContainingFunction).second;
 
   const DataLayout &DL = Mod->getDataLayout();
 
   unsigned Align = I.getAlignment();
   if (Align == 0)
     Align = DL.getABITypeAlignment(I.getAllocatedType());
 
   // FIXME: This computed padding is likely wrong since it depends on inverse

lib/Target/AMDGPU/AMDGPUSubtarget.h

bool isAmdHsaOS() const {
     return TargetTriple.getOS() == Triple::AMDHSA;
   }
   bool isVGPRSpillingEnabled(const Function& F) const;
 
   bool isXNACKEnabled() const {
     return EnableXNACK;
   }
 
-  unsigned getMaxWavesPerCU() const {
-    if (getGeneration() >= AMDGPUSubtarget::SOUTHERN_ISLANDS)
-      return 10;
-
-    // FIXME: Not sure what this is for other subtagets.
-    return 8;
-  }
-
   bool enableSubRegLiveness() const override {
     return true;
   }
 
   /// \brief Returns the offset in bytes from the start of the input buffer
   ///        of the first explicit kernel argument.
   unsigned getExplicitKernelArgOffset() const {
     return isAmdHsaOS() ? 0 : 36;
   }
 
   unsigned getMaxNumUserSGPRs() const {
     return 16;
   }
+
+  /// Returns number of execution units per compute unit supported by the
+  /// subtarget.
+  unsigned getNumEUsPerCU() const {
+    return 4;
+  }
+
+  /// Returns maximum number of work groups per compute unit supported by the
+  /// subtarget and limited by given flat work group size.
+  unsigned getMaxNumWorkGroupsPerCU(unsigned FlatWorkGroupSize) const {
+    if (getGeneration() < AMDGPUSubtarget::SOUTHERN_ISLANDS)
+      return 8;
+    return getNumWavesPerWorkGroup(FlatWorkGroupSize) == 1 ? 40 : 16;
+  }
+
+  /// Returns maximum number of active waves per compute unit supported by the
+  /// subtarget without any kind of limitation.
+  unsigned getMaxNumActiveWavesPerCU() const {
+    return getMaxNumActiveWavesPerEU() * getNumEUsPerCU();
+  }
+
+  /// Returns maximum number of active waves per compute unit supported by the
+  /// subtarget and limited by given flat work group size.
+  unsigned getMaxNumActiveWavesPerCU(unsigned FlatWorkGroupSize) const {
+    unsigned NumWavesPerWorkGroup =
+      getNumWavesPerWorkGroup(FlatWorkGroupSize);
+    unsigned MaxNumWorkGroupsPerCU =
+      getMaxNumWorkGroupsPerCU(FlatWorkGroupSize);
+    unsigned MaxNumActiveWavesPerCU =
+      NumWavesPerWorkGroup * MaxNumWorkGroupsPerCU;
+    MaxNumActiveWavesPerCU =
+      std::min(MaxNumActiveWavesPerCU, getMaxNumActiveWavesPerCU());
+    MaxNumActiveWavesPerCU =
+      alignDown(MaxNumActiveWavesPerCU, NumWavesPerWorkGroup);
+    MaxNumActiveWavesPerCU = MaxNumActiveWavesPerCU / NumWavesPerWorkGroup;
+    MaxNumActiveWavesPerCU = MaxNumActiveWavesPerCU * NumWavesPerWorkGroup;
+    return MaxNumActiveWavesPerCU;
+  }
+
+  /// Returns minimum number of active waves per execution unit supported by the
+  /// subtarget.
+  unsigned getMinNumActiveWavesPerEU() const {
+    return 1;
+  }
+
+  /// Returns maximum number of active waves per execution unit supported by the
+  /// subtarget without any kind of limitation.
+  unsigned getMaxNumActiveWavesPerEU() const {
+    // FIXME: Not sure what this is for subtargets below Southern Islands.
+    if (getGeneration() < AMDGPUSubtarget::SOUTHERN_ISLANDS)
+      return 8;
+    // FIXME: Need to take scratch memory into account.
+    return 10;
+  }
+
+  /// Returns maximum number of active waves per execution unit supported by the
+  /// subtarget and limited by given flat work group size.
+  unsigned getMaxNumActiveWavesPerEU(unsigned FlatWorkGroupSize) const {
+    unsigned MaxNumActiveWavesPerCU =
+      getMaxNumActiveWavesPerCU(FlatWorkGroupSize);
+    unsigned MaxNumActiveWavesPerEU =
+      alignDown(MaxNumActiveWavesPerCU, getNumEUsPerCU());
+    MaxNumActiveWavesPerEU = MaxNumActiveWavesPerEU / getNumEUsPerCU();
+    return MaxNumActiveWavesPerEU;
+  }
+
+  /// Returns minimum flat work group size supported by the subtarget.
+  unsigned getMinFlatWorkGroupSize() const {
+    return 1;
+  }
+
+  /// Returns maximum flat work group size supported by the subtarget.
+  unsigned getMaxFlatWorkGroupSize() const {
+    return 2048;
+  }
+
+  /// Returns number of waves per work group given the flat work group size.
+  unsigned getNumWavesPerWorkGroup(unsigned FlatWorkGroupSize) const {
+    return alignTo(FlatWorkGroupSize, getWavefrontSize()) / getWavefrontSize();
+  }
+
+  /// Returns subtarget's default pair of minimum/maximum flat work group sizes
+  /// for function \p F, or minimum/maximum flat work group sizes explicitly
+  /// requested using "amdgpu-flat-work-group-size" attribute attached to
+  /// function \p F. Returns subtarget's default values and emits error if
+  /// explicitly requested values cannot be converted to integer, or violate
+  /// subtarget's specifications.
+  std::pair<unsigned, unsigned> getFlatWorkGroupSizes(const Function &F) const;
+
+  /// Returns subtarget's default pair of minimum/maximum number of active waves
+  /// per execution unit for function \p F, or minimum/maximum number of active
+  /// waves per execution unit explicitly requested using
+  /// "amdgpu-num-active-waves-per-eu" attribute attached to function \p F.
+  /// Returns subtarget's default values and emits error if explicitly requested
+  /// values cannot be converted to integer, violate subtarget's specifications,
+  /// or are not compatible with minimum/maximum number of active waves limited
+  /// by flat work group size, register usage, and/or lds usage.
+  std::pair<unsigned, unsigned> getNumActiveWavesPerEU(const Function &F) const;
 };
 
 } // End namespace llvm
 
 #endif

lib/Target/AMDGPU/AMDGPUSubtarget.cpp

if (getGeneration() >= SOUTHERN_ISLANDS) {
     // pressure once register usage is above the threshold defined by
     // SIRegisterInfo::getRegPressureSetLimit()
     Policy.ShouldTrackPressure = true;
 
     // Enabling both top down and bottom up scheduling seems to give us less
     // register spills than just using one of these approaches on its own.
     Policy.OnlyTopDown = false;
     Policy.OnlyBottomUp = false;
 
     // Enabling ShouldTrackLaneMasks crashes the SI Machine Scheduler.
     if (!enableSIScheduler())
       Policy.ShouldTrackLaneMasks = true;
   }
 }
 
+std::pair<unsigned, unsigned> AMDGPUSubtarget::getFlatWorkGroupSizes(
+  const Function &F) const {
+  LLVMContext &Ctx = F.getContext();
+
+  // Default minimum/maximum flat work group sizes.
+  std::pair<unsigned, unsigned> Default =
+    AMDGPU::isCompute(F.getCallingConv()) ?
+      std::pair<unsigned, unsigned>(128, 256) :
+      std::pair<unsigned, unsigned>(1, getWavefrontSize());
+  // Requested minimum/maximum flat work group sizes.
+  std::pair<unsigned, unsigned> Requested =
+    AMDGPU::getIntegerPairAttribute(F, "amdgpu-flat-work-group-size", Default);
+
+  // Make sure requested minimum is less than requested maximum.
+  if (Requested.first > Requested.second) {
+    Ctx.emitError("invalid amdgpu-flat-work-group-size: "
+                  "minimum is greater than maximum");
+    return Default;
+  }
+  // Make sure requested values do not violate subtarget's specifications.
+  if (Requested.first < getMinFlatWorkGroupSize()) {
+    Ctx.emitError("invalid amdgpu-flat-work-group-size: "
+                  "requested minimum flat work group size is not supported by "
+                  "the subtarget");
+    return Default;
+  }
+  if (Requested.second > getMaxFlatWorkGroupSize()) {
+    Ctx.emitError("invalid amdgpu-flat-work-group-size: "
+                  "requested maximum flat work group size is not supported by "
+                  "the subtarget");
+    return Default;
+  }
+
+  return Requested;
+}
+
+std::pair<unsigned, unsigned> AMDGPUSubtarget::getNumActiveWavesPerEU(
+  const Function &F) const {
+  LLVMContext &Ctx = F.getContext();
+
+  // Default minimum/maximum number of active waves per execution unit.
+  std::pair<unsigned, unsigned> Default = std::pair<unsigned, unsigned>(1, 0);
+  // Default/requested minimum/maximum flat work group sizes.
+  std::pair<unsigned, unsigned> FlatWorkGroupSizes = getFlatWorkGroupSizes(F);
+  // If minimum/maximum flat work group sizes were explicitly requested using
+  // "amdgpu-flat-work-group-size" attribute, then set default minimum/maximum
+  // number of active waves per execution unit to values implied by requested
+  // minimum/maximum flat work group sizes.
+  unsigned ImpliedByMinFlatWorkGroupSize =
+    getMaxNumActiveWavesPerEU(FlatWorkGroupSizes.first);
+  unsigned ImpliedByMaxFlatWorkGroupSize =
+    getMaxNumActiveWavesPerEU(FlatWorkGroupSizes.second);
+  unsigned MinImpliedByFlatWorkGroupSize =
+    std::min(ImpliedByMinFlatWorkGroupSize, ImpliedByMaxFlatWorkGroupSize);
+  unsigned MaxImpliedByFlatWorkGroupSize =
+    std::max(ImpliedByMinFlatWorkGroupSize, ImpliedByMaxFlatWorkGroupSize);
+  if (F.hasFnAttribute("amdgpu-flat-work-group-size")) {
+    Default.first = MinImpliedByFlatWorkGroupSize;
+    Default.second = MaxImpliedByFlatWorkGroupSize;
+  }
+  // Requested minimum/maximum number of active waves per execution unit.
+  std::pair<unsigned, unsigned> Requested = AMDGPU::getIntegerPairAttribute(
+    F, "amdgpu-num-active-waves-per-eu", Default, true);
+
+  // Make sure requested minimum is less than requested maximum.
+  if (Requested.second && Requested.first > Requested.second) {
+    Ctx.emitError("invalid amdgpu-num-active-waves-per-eu: "
+                  "minimum is greater than maximum");
+    return Default;
+  }
+  // Make sure requested values do not violate subtarget's specifications.
+  if (Requested.first < getMinNumActiveWavesPerEU() ||
+      Requested.first > getMaxNumActiveWavesPerEU()) {
+    Ctx.emitError("invalid amdgpu-num-active-waves-per-eu: "
+                  "requested minimum number of active waves per execution unit "
+                  "is not supported by the subtarget");
+    return Default;
+  }
+  if (Requested.second > getMaxNumActiveWavesPerEU()) {
+    Ctx.emitError("invalid amdgpu-num-active-waves-per-eu: "
+                  "requested maximum number of active waves per execution unit "
+                  "is not supported by the subtarget");
+    return Default;
+  }
+  // Make sure requested values are compatible with values implied by requested
+  // minimum/maximum flat work group sizes.
+  if (Requested.first > MinImpliedByFlatWorkGroupSize ||
+      Requested.second > MaxImpliedByFlatWorkGroupSize) {
+    Ctx.emitError("invalid amdgpu-num-active-waves-per-eu: "
+                  "requested number of active waves per execution unit cannot "
+                  "be achieved with default or requested flat work group "
+                  "sizes");
+    return Default;
+  }
+
+  return Requested;
+}

lib/Target/AMDGPU/SIInstrInfo.cpp

 #include "llvm/IR/Function.h"
 #include "llvm/CodeGen/RegisterScavenging.h"
 #include "llvm/MC/MCInstrDesc.h"
 #include "llvm/Support/Debug.h"
 
 using namespace llvm;
 
 SIInstrInfo::SIInstrInfo(const AMDGPUSubtarget &st)
-    : AMDGPUInstrInfo(st), RI() {}
+    : AMDGPUInstrInfo(st), RI(st) {}
 
 //===----------------------------------------------------------------------===//
 // TargetInstrInfo callbacks
 //===----------------------------------------------------------------------===//
 
 static unsigned getNumOperandsNoGlue(SDNode *Node) {
   unsigned N = Node->getNumOperands();
   while (N && Node->getOperand(N - 1).getValueType() == MVT::Glue)
unsigned SIInstrInfo::calculateLDSSpillAddress(MachineBasicBlock &MBB,
                                                unsigned FrameOffset,
                                                unsigned Size) const {
   MachineFunction *MF = MBB.getParent();
   SIMachineFunctionInfo *MFI = MF->getInfo<SIMachineFunctionInfo>();
   const AMDGPUSubtarget &ST = MF->getSubtarget<AMDGPUSubtarget>();
   const SIRegisterInfo *TRI =
       static_cast<const SIRegisterInfo*>(ST.getRegisterInfo());
   DebugLoc DL = MBB.findDebugLoc(MI);
-  unsigned WorkGroupSize = MFI->getMaximumWorkGroupSize(*MF);
+  unsigned WorkGroupSize = MFI->getMaxFlatWorkGroupSize();
   unsigned WavefrontSize = ST.getWavefrontSize();
 
   unsigned TIDReg = MFI->getTIDReg();
   if (!MFI->hasCalculatedTID()) {
     MachineBasicBlock &Entry = MBB.getParent()->front();
     MachineBasicBlock::iterator Insert = Entry.front();
     DebugLoc DL = Insert->getDebugLoc();
 

lib/Target/AMDGPU/SIMachineFunctionInfo.h

class SIMachineFunctionInfo final : public AMDGPUMachineFunction {
   unsigned WorkGroupIDZSystemSGPR;
   unsigned WorkGroupInfoSystemSGPR;
   unsigned PrivateSegmentWaveByteOffsetSystemSGPR;
 
   // Graphics info.
   unsigned PSInputAddr;
   bool ReturnsVoid;
 
-  unsigned MaximumWorkGroupSize;
-
-  // Number of reserved VGPRs for debugger usage.
-  unsigned DebuggerReservedVGPRCount;
+  // A pair of default/requested minimum/maximum flat work group sizes.
+  // Minimum - first, maximum - second.
+  std::pair<unsigned, unsigned> FlatWorkGroupSizes;
+  // A pair of default/requested minimum/maximum number of active waves per
+  // execution unit. Minimum - first, maximum - second.
+  std::pair<unsigned, unsigned> NumActiveWavesPerEU;
 
 public:
   // FIXME: Make private
   unsigned LDSWaveSpillSize;
   unsigned PSInputEna;
   std::map<unsigned, unsigned> LaneVGPRs;
   unsigned ScratchOffsetReg;
   unsigned NumUserSGPRs;
public:
   bool returnsVoid() const {
     return ReturnsVoid;
   }
 
   void setIfReturnsVoid(bool Value) {
     ReturnsVoid = Value;
   }
 
-  /// \returns Number of reserved VGPRs for debugger usage.
-  unsigned getDebuggerReservedVGPRCount() const {
-    return DebuggerReservedVGPRCount;
+  /// Returns a pair of default/requested minimum/maximum flat work group sizes
+  /// for this function.
+  std::pair<unsigned, unsigned> getFlatWorkGroupSizes() const {
+    return FlatWorkGroupSizes;
+  }
+
+  /// Returns default/requested minimum flat work group size for this function.
+  unsigned getMinFlatWorkGroupSize() const {
+    return FlatWorkGroupSizes.first;
+  }
+
+  /// Returns default/requested maximum flat work group sizes for this function.
+  unsigned getMaxFlatWorkGroupSize() const {
+    return FlatWorkGroupSizes.second;
   }
 
-  unsigned getMaximumWorkGroupSize(const MachineFunction &MF) const;
+  /// Returns a pair of default/requested minimum/maximum number of active waves
+  /// per execution unit.
+  std::pair<unsigned, unsigned> getNumActiveWavesPerEU() const {
+    return NumActiveWavesPerEU;
+  }
+
+  /// Returns default/requested minimum number of active waves per execution
+  /// unit.
+  unsigned getMinNumActiveWavesPerEU() const {
+    return NumActiveWavesPerEU.first;
+  }
+
+  /// Returns default/requested maximum number of active waves per execution
+  /// unit.
+  unsigned getMaxNumActiveWavesPerEU() const {
+    return NumActiveWavesPerEU.second;
+  }
 };
 
 } // End namespace llvm
 
 #endif

lib/Target/AMDGPU/SIMachineFunctionInfo.cpp

: AMDGPUMachineFunction(MF),
     GridWorkGroupCountZUserSGPR(AMDGPU::NoRegister),
     WorkGroupIDXSystemSGPR(AMDGPU::NoRegister),
     WorkGroupIDYSystemSGPR(AMDGPU::NoRegister),
     WorkGroupIDZSystemSGPR(AMDGPU::NoRegister),
     WorkGroupInfoSystemSGPR(AMDGPU::NoRegister),
     PrivateSegmentWaveByteOffsetSystemSGPR(AMDGPU::NoRegister),
     PSInputAddr(0),
     ReturnsVoid(true),
-    MaximumWorkGroupSize(0),
-    DebuggerReservedVGPRCount(0),
+    FlatWorkGroupSizes(std::pair<unsigned, unsigned>(0, 0)),
+    NumActiveWavesPerEU(std::pair<unsigned, unsigned>(0, 0)),
     LDSWaveSpillSize(0),
     PSInputEna(0),
     NumUserSGPRs(0),
     NumSystemSGPRs(0),
     HasSpilledSGPRs(false),
     HasSpilledVGPRs(false),
     HasNonSpillStackObjects(false),
     HasFlatInstructions(false),
SIMachineFunctionInfo::SIMachineFunctionInfo(const MachineFunction &MF)
 
   // We don't need to worry about accessing spills with flat instructions.
   // TODO: On VI where we must use flat for global, we should be able to omit
   // this if it is never used for generic access.
   if (HasStackObjects && ST.getGeneration() >= AMDGPUSubtarget::SEA_ISLANDS &&
       ST.isAmdHsaOS())
     FlatScratchInit = true;
 
-  if (AMDGPU::isCompute(F->getCallingConv()))
-    MaximumWorkGroupSize = AMDGPU::getMaximumWorkGroupSize(*F);
-  else
-    MaximumWorkGroupSize = ST.getWavefrontSize();
-
-  if (ST.debuggerReserveRegs())
-    DebuggerReservedVGPRCount = 4;
+  FlatWorkGroupSizes = ST.getFlatWorkGroupSizes(*F);
+  NumActiveWavesPerEU = ST.getNumActiveWavesPerEU(*F);
 }
 
 unsigned SIMachineFunctionInfo::addPrivateSegmentBuffer(
   const SIRegisterInfo &TRI) {
   PrivateSegmentBufferUserSGPR = TRI.getMatchingSuperReg(
     getNextUserSGPR(), AMDGPU::sub0, &AMDGPU::SReg_128RegClass);
   NumUserSGPRs += 4;
   return PrivateSegmentBufferUserSGPR;
for (MachineFunction::iterator BI = MF->begin(), BE = MF->end();
          BI != BE; ++BI) {
       BI->addLiveIn(LaneVGPR);
     }
   }
 
   Spill.VGPR = LaneVGPRs[LaneVGPRIdx];
   return Spill;
 }
-
-unsigned SIMachineFunctionInfo::getMaximumWorkGroupSize(
-                                              const MachineFunction &MF) const {
-  return MaximumWorkGroupSize;
-}

lib/Target/AMDGPU/SIRegisterInfo.h

 #include "AMDGPUSubtarget.h"
 #include "llvm/CodeGen/MachineRegisterInfo.h"
 #include "llvm/Support/Debug.h"
 
 namespace llvm {
 
 struct SIRegisterInfo final : public AMDGPURegisterInfo {
 private:
+  const AMDGPUSubtarget &ST;
   unsigned SGPR32SetID;
   unsigned VGPR32SetID;
   BitVector SGPRPressureSets;
   BitVector VGPRPressureSets;
 
   void reserveRegisterTuples(BitVector &, unsigned Reg) const;
   void classifyPressureSet(unsigned PSetID, unsigned Reg,
                            BitVector &PressureSets) const;
 
 public:
-  SIRegisterInfo();
+  SIRegisterInfo(const AMDGPUSubtarget &st);
 
   /// Return the end register initially reserved for the scratch buffer in case
   /// spilling is needed.
   unsigned reservedPrivateSegmentBufferReg(const MachineFunction &MF) const;
 
   /// Return the end register initially reserved for the scratch wave offset in
   /// case spilling is needed.
   unsigned reservedPrivateSegmentWaveByteOffsetReg(
enum PreloadedValue {
     WORKITEM_ID_Y       = 16,
     WORKITEM_ID_Z       = 17
   };
 
   /// \brief Returns the physical register that \p Value is stored in.
   unsigned getPreloadedValue(const MachineFunction &MF,
                              enum PreloadedValue Value) const;
 
-  /// \brief Give the maximum number of VGPRs that can be used by \p WaveCount
-  ///        concurrent waves.
-  unsigned getNumVGPRsAllowed(unsigned WaveCount) const;
-
-  /// \brief Give the maximum number of SGPRs that can be used by \p WaveCount
-  ///        concurrent waves.
-  unsigned getNumSGPRsAllowed(AMDGPUSubtarget::Generation gen,
-                              unsigned WaveCount) const;
-
   unsigned findUnusedRegister(const MachineRegisterInfo &MRI,
                               const TargetRegisterClass *RC) const;
 
   unsigned getSGPR32PressureSet() const { return SGPR32SetID; };
   unsigned getVGPR32PressureSet() const { return VGPR32SetID; };
 
   bool isVGPR(const MachineRegisterInfo &MRI, unsigned Reg) const;
 
+  /// Returns SGPR allocation granularity supported by the subtarget.
+  unsigned getSGPRAllocGranule() const {
+    return 8;
+  }
+
+  /// Returns total number of SGPRs supported by the subtarget.
+  unsigned getTotalNumSGPRs() const {
+    if (ST.getGeneration() >= AMDGPUSubtarget::VOLCANIC_ISLANDS)
+      return 800;
+    return 512;
+  }
+
+  /// Returns addressable number of SGPRs supported by the subtarget.
+  unsigned getAddressableNumSGPRs() const {
+    if (ST.getGeneration() >= AMDGPUSubtarget::VOLCANIC_ISLANDS)
+      return 102;
+    return 104;
+  }
+
+  /// Returns reserved number of SGPRs supported by the subtarget.
+  unsigned getReservedNumSGPRs() const {
+    if (ST.getGeneration() >= AMDGPUSubtarget::VOLCANIC_ISLANDS)
+      return 6; // VCC, FLAT_SCRATCH, XNACK.
+    return 2; // VCC.
+  }
+
+  /// Returns minimum number of SGPRs that meets given number of active waves
+  /// per execution unit requirement.
+  unsigned getMinNumSGPRs(unsigned NumActiveWavesPerEU) const;
+
+  /// Returns maximum number of SGPRs that meets given number of active waves
+  /// per execution unit requirement.
+  unsigned getMaxNumSGPRs(unsigned NumActiveWavesPerEU) const;
+
+  /// Returns maximum number of SGPRs that meets number of active waves per
+  /// execution unit requirement for function \p MF, or number of SGPRs
+  /// explicitly requested using "amdgpu-num-sgpr" attribute attached to
+  /// function \p MF. Returns value that meets number of active waves per
+  /// execution unit requirement and emits error if explicitly requested value
+  /// cannot be converted to integer, violates subtarget's specifications, or
+  /// does not meet number of active waver per execution unit requirement.
+  unsigned getMaxNumSGPRs(const MachineFunction &MF) const;
+
+  /// Returns VGPR allocation granularity supported by the subtarget.
+  unsigned getVGPRAllocGranule() const {
+    return 4;
+  }
+
+  /// Returns total number of VGPRs supported by the subtarget.
+  unsigned getTotalNumVGPRs() const {
+    return 256;
+  }
+
+  /// Returns reserved number of VGPRs for debugger use supported by the
+  /// subtarget.
+  unsigned getDebuggerReservedNumVGPRs() const {
+    if (ST.debuggerReserveRegs())
+      return 4;
+    return 0;
+  }
+
+  /// Returns minimum number of SGPRs that meets given number of active waves
+  /// per execution unit requirement.
+  unsigned getMinNumVGPRs(unsigned NumActiveWavesPerEU) const;
+
+  /// Returns maximum number of VGPRs that meets given number of active waves
+  /// per execution unit requirement.
+  unsigned getMaxNumVGPRs(unsigned NumActiveWavesPerEU) const;
+
+  /// Returns maximum number of VGPRs that meets number of active waves per
+  /// execution unit requirement for function \p MF, or number of VGPRs
+  /// explicitly requested using "amdgpu-num-vgpr" attribute attached to
+  /// function \p MF. Returns value that meets number of active waves per
+  /// execution unit requirement and emits error if explicitly requested value
+  /// cannot be converted to integer, violates subtarget's specifications, or
+  /// does not meet number of active waves per execution unit requirement.
+  unsigned getMaxNumVGPRs(const MachineFunction &MF) const;
+
 private:
   void buildScratchLoadStore(MachineBasicBlock::iterator MI,
                              unsigned LoadStoreOp, const MachineOperand *SrcDst,
                              unsigned ScratchRsrcReg, unsigned ScratchOffset,
                              int64_t Offset,
                              RegScavenger *RS) const;
 };
 
 } // End namespace llvm
 
 #endif

lib/Target/AMDGPU/SIRegisterInfo.cpp

 #include "llvm/CodeGen/MachineFrameInfo.h"
 #include "llvm/CodeGen/MachineInstrBuilder.h"
 #include "llvm/CodeGen/RegisterScavenging.h"
 #include "llvm/IR/Function.h"
 #include "llvm/IR/LLVMContext.h"
 
 using namespace llvm;
 
-static unsigned getMaxWaveCountPerSIMD(const MachineFunction &MF) {
-  const SIMachineFunctionInfo& MFI = *MF.getInfo<SIMachineFunctionInfo>();
-  const AMDGPUSubtarget &ST = MF.getSubtarget<AMDGPUSubtarget>();
-  unsigned SIMDPerCU = 4;
-
-  unsigned MaxInvocationsPerWave = SIMDPerCU * ST.getWavefrontSize();
-  return alignTo(MFI.getMaximumWorkGroupSize(MF), MaxInvocationsPerWave) /
-           MaxInvocationsPerWave;
-}
-
-static unsigned getMaxWorkGroupSGPRCount(const MachineFunction &MF) {
-  const AMDGPUSubtarget &ST = MF.getSubtarget<AMDGPUSubtarget>();
-  unsigned MaxWaveCountPerSIMD = getMaxWaveCountPerSIMD(MF);
-
-  unsigned TotalSGPRCountPerSIMD, AddressableSGPRCount, SGPRUsageAlignment;
-  unsigned ReservedSGPRCount;
-
-  if (ST.getGeneration() >= AMDGPUSubtarget::VOLCANIC_ISLANDS) {
-    TotalSGPRCountPerSIMD = 800;
-    AddressableSGPRCount = 102;
-    SGPRUsageAlignment = 16;
-    ReservedSGPRCount = 6; // VCC, FLAT_SCRATCH, XNACK
-  } else {
-    TotalSGPRCountPerSIMD = 512;
-    AddressableSGPRCount = 104;
-    SGPRUsageAlignment = 8;
-    ReservedSGPRCount = 2; // VCC
-  }
-
-  unsigned MaxSGPRCount = (TotalSGPRCountPerSIMD / MaxWaveCountPerSIMD);
-  MaxSGPRCount = alignDown(MaxSGPRCount, SGPRUsageAlignment);
-
-  if (ST.hasSGPRInitBug())
-    MaxSGPRCount = AMDGPUSubtarget::FIXED_SGPR_COUNT_FOR_INIT_BUG;
-
-  return std::min(MaxSGPRCount - ReservedSGPRCount, AddressableSGPRCount);
-}
-
-static unsigned getMaxWorkGroupVGPRCount(const MachineFunction &MF) {
-  unsigned MaxWaveCountPerSIMD = getMaxWaveCountPerSIMD(MF);
-  unsigned TotalVGPRCountPerSIMD = 256;
-  unsigned VGPRUsageAlignment = 4;
-
-  return alignDown(TotalVGPRCountPerSIMD / MaxWaveCountPerSIMD,
-                   VGPRUsageAlignment);
-}
-
 static bool hasPressureSet(const int *PSets, unsigned PSetID) {
   for (unsigned i = 0; PSets[i] != -1; ++i) {
     if (PSets[i] == (int)PSetID)
       return true;
   }
   return false;
 }
 
 void SIRegisterInfo::classifyPressureSet(unsigned PSetID, unsigned Reg,
                                          BitVector &PressureSets) const {
   for (MCRegUnitIterator U(Reg, this); U.isValid(); ++U) {
     const int *PSets = getRegUnitPressureSets(*U);
     if (hasPressureSet(PSets, PSetID)) {
       PressureSets.set(PSetID);
       break;
     }
   }
 }
 
-SIRegisterInfo::SIRegisterInfo() : AMDGPURegisterInfo(),
-                                   SGPRPressureSets(getNumRegPressureSets()),
-                                   VGPRPressureSets(getNumRegPressureSets()) {
+SIRegisterInfo::SIRegisterInfo(const AMDGPUSubtarget &st)
+  :  AMDGPURegisterInfo(),
+     ST(st),
   unsigned NumRegPressureSets = getNumRegPressureSets();
 
   SGPR32SetID = NumRegPressureSets;
   VGPR32SetID = NumRegPressureSets;
   for (unsigned i = 0; i < NumRegPressureSets; ++i) {
     if (strncmp("SGPR_32", getRegPressureSetName(i), 7) == 0)
       SGPR32SetID = i;
     else if (strncmp("VGPR_32", getRegPressureSetName(i), 7) == 0)
void SIRegisterInfo::reserveRegisterTuples(BitVector &Reserved, unsigned Reg) const {
   MCRegAliasIterator R(Reg, this, true);
 
   for (; R.isValid(); ++R)
     Reserved.set(*R);
 }
 
 unsigned SIRegisterInfo::reservedPrivateSegmentBufferReg(
   const MachineFunction &MF) const {
-  unsigned BaseIdx = alignDown(getMaxWorkGroupSGPRCount(MF), 4) - 4;
+  unsigned BaseIdx = alignDown(getMaxNumSGPRs(MF), 4) - 4;
   unsigned BaseReg(AMDGPU::SGPR_32RegClass.getRegister(BaseIdx));
   return getMatchingSuperReg(BaseReg, AMDGPU::sub0, &AMDGPU::SReg_128RegClass);
 }
 
 unsigned SIRegisterInfo::reservedPrivateSegmentWaveByteOffsetReg(
   const MachineFunction &MF) const {
-  unsigned RegCount = getMaxWorkGroupSGPRCount(MF);
+  unsigned RegCount = getMaxNumSGPRs(MF);
   unsigned Reg;
 
   // Try to place it in a hole after PrivateSegmentbufferReg.
   if (RegCount & 3) {
     // We cannot put the segment buffer in (Idx - 4) ... (Idx - 1) due to
     // alignment constraints, so we have a hole where can put the wave offset.
     Reg = RegCount - 1;
   } else {
BitVector SIRegisterInfo::getReservedRegs(const MachineFunction &MF) const {
   reserveRegisterTuples(Reserved, AMDGPU::TMA);
   reserveRegisterTuples(Reserved, AMDGPU::TTMP0_TTMP1);
   reserveRegisterTuples(Reserved, AMDGPU::TTMP2_TTMP3);
   reserveRegisterTuples(Reserved, AMDGPU::TTMP4_TTMP5);
   reserveRegisterTuples(Reserved, AMDGPU::TTMP6_TTMP7);
   reserveRegisterTuples(Reserved, AMDGPU::TTMP8_TTMP9);
   reserveRegisterTuples(Reserved, AMDGPU::TTMP10_TTMP11);
 
-  unsigned MaxWorkGroupSGPRCount = getMaxWorkGroupSGPRCount(MF);
-  unsigned MaxWorkGroupVGPRCount = getMaxWorkGroupVGPRCount(MF);
-
-  unsigned NumSGPRs = AMDGPU::SGPR_32RegClass.getNumRegs();
-  unsigned NumVGPRs = AMDGPU::VGPR_32RegClass.getNumRegs();
-  for (unsigned i = MaxWorkGroupSGPRCount; i < NumSGPRs; ++i) {
+  unsigned MaxNumSGPRs = getMaxNumSGPRs(MF);
+  unsigned TotalNumSGPRs = AMDGPU::SGPR_32RegClass.getNumRegs();
+  for (unsigned i = MaxNumSGPRs; i < TotalNumSGPRs; ++i) {
     unsigned Reg = AMDGPU::SGPR_32RegClass.getRegister(i);
     reserveRegisterTuples(Reserved, Reg);
   }
 
-
-  for (unsigned i = MaxWorkGroupVGPRCount; i < NumVGPRs; ++i) {
+  unsigned MaxNumVGPRs = getMaxNumVGPRs(MF);
+  unsigned TotalNumVGPRs = AMDGPU::VGPR_32RegClass.getNumRegs();
+  for (unsigned i = MaxNumVGPRs; i < TotalNumVGPRs; ++i) {
     unsigned Reg = AMDGPU::VGPR_32RegClass.getRegister(i);
     reserveRegisterTuples(Reserved, Reg);
   }
 
   const SIMachineFunctionInfo *MFI = MF.getInfo<SIMachineFunctionInfo>();
 
   unsigned ScratchWaveOffsetReg = MFI->getScratchWaveOffsetReg();
   if (ScratchWaveOffsetReg != AMDGPU::NoRegister) {
     // Reserve 1 SGPR for scratch wave offset in case we need to spill.
     reserveRegisterTuples(Reserved, ScratchWaveOffsetReg);
   }
 
   unsigned ScratchRSrcReg = MFI->getScratchRSrcReg();
   if (ScratchRSrcReg != AMDGPU::NoRegister) {
     // Reserve 4 SGPRs for the scratch buffer resource descriptor in case we need
     // to spill.
     // TODO: May need to reserve a VGPR if doing LDS spilling.
     reserveRegisterTuples(Reserved, ScratchRSrcReg);
     assert(!isSubRegister(ScratchRSrcReg, ScratchWaveOffsetReg));
   }
 
-  // Reserve registers for debugger usage if "amdgpu-debugger-reserve-trap-regs"
-  // attribute was specified.
-  const AMDGPUSubtarget &ST = MF.getSubtarget<AMDGPUSubtarget>();
-  if (ST.debuggerReserveRegs()) {
-    unsigned ReservedVGPRFirst =
-      MaxWorkGroupVGPRCount - MFI->getDebuggerReservedVGPRCount();
-    for (unsigned i = ReservedVGPRFirst; i < MaxWorkGroupVGPRCount; ++i) {
-      unsigned Reg = AMDGPU::VGPR_32RegClass.getRegister(i);
-      reserveRegisterTuples(Reserved, Reg);
-    }
-  }
-
   return Reserved;
 }
 
 unsigned SIRegisterInfo::getRegPressureSetLimit(const MachineFunction &MF,
                                                 unsigned Idx) const {
-  const AMDGPUSubtarget &STI = MF.getSubtarget<AMDGPUSubtarget>();
-  // FIXME: We should adjust the max number of waves based on LDS size.
-  unsigned SGPRLimit = getNumSGPRsAllowed(STI.getGeneration(),
-                                          STI.getMaxWavesPerCU());
-  unsigned VGPRLimit = getNumVGPRsAllowed(STI.getMaxWavesPerCU());
-
+  unsigned SGPRLimit = getMaxNumSGPRs(MF);
+  unsigned VGPRLimit = getMaxNumVGPRs(MF);
   unsigned VSLimit = SGPRLimit + VGPRLimit;
 
   if (SGPRPressureSets.test(Idx) && VGPRPressureSets.test(Idx)) {
     // FIXME: This is a hack. We should never be considering the pressure of
     // these since no virtual register should ever have this class.
     return VSLimit;
   }
 
void SIRegisterInfo::materializeFrameBaseRegister(MachineBasicBlock *MBB,
                                                   int64_t Offset) const {
   MachineBasicBlock::iterator Ins = MBB->begin();
   DebugLoc DL; // Defaults to "unknown"
 
   if (Ins != MBB->end())
     DL = Ins->getDebugLoc();
 
   MachineFunction *MF = MBB->getParent();
-  const AMDGPUSubtarget &Subtarget = MF->getSubtarget<AMDGPUSubtarget>();
-  const TargetInstrInfo *TII = Subtarget.getInstrInfo();
+  const TargetInstrInfo *TII = ST.getInstrInfo();
 
   if (Offset == 0) {
     BuildMI(*MBB, Ins, DL, TII->get(AMDGPU::V_MOV_B32_e32), BaseReg)
       .addFrameIndex(FrameIdx);
     return;
   }
 
   MachineRegisterInfo &MRI = MF->getRegInfo();
   unsigned UnusedCarry = MRI.createVirtualRegister(&AMDGPU::SReg_64RegClass);
 
   BuildMI(*MBB, Ins, DL, TII->get(AMDGPU::V_ADD_I32_e64), BaseReg)
     .addReg(UnusedCarry, RegState::Define | RegState::Dead)
     .addImm(Offset)
     .addFrameIndex(FrameIdx);
 }
 
 void SIRegisterInfo::resolveFrameIndex(MachineInstr &MI, unsigned BaseReg,
                                        int64_t Offset) const {
 
   MachineBasicBlock *MBB = MI.getParent();
   MachineFunction *MF = MBB->getParent();
-  const AMDGPUSubtarget &Subtarget = MF->getSubtarget<AMDGPUSubtarget>();
-  const SIInstrInfo *TII
-    = static_cast<const SIInstrInfo *>(Subtarget.getInstrInfo());
+  const SIInstrInfo *TII = static_cast<const SIInstrInfo *>(ST.getInstrInfo());
 
 #ifndef NDEBUG
   // FIXME: Is it possible to be storing a frame index to itself?
   bool SeenFI = false;
   for (const MachineOperand &MO: MI.operands()) {
     if (MO.isFI()) {
       if (SeenFI)
         llvm_unreachable("should not see multiple frame indices");
bool SIRegisterInfo::opCanUseInlineConstant(unsigned OpType) const {
   return OpType == AMDGPU::OPERAND_REG_INLINE_C;
 }
 
 // FIXME: Most of these are flexible with HSA and we don't need to reserve them
 // as input registers if unused. Whether the dispatch ptr is necessary should be
 // easy to detect from used intrinsics. Scratch setup is harder to know.
 unsigned SIRegisterInfo::getPreloadedValue(const MachineFunction &MF,
                                            enum PreloadedValue Value) const {
-
   const SIMachineFunctionInfo *MFI = MF.getInfo<SIMachineFunctionInfo>();
-  const AMDGPUSubtarget &ST = MF.getSubtarget<AMDGPUSubtarget>();
-  (void)ST;
+
   switch (Value) {
   case SIRegisterInfo::WORKGROUP_ID_X:
     assert(MFI->hasWorkGroupIDX());
     return MFI->WorkGroupIDXSystemSGPR;
   case SIRegisterInfo::WORKGROUP_ID_Y:
     assert(MFI->hasWorkGroupIDY());
     return MFI->WorkGroupIDYSystemSGPR;
   case SIRegisterInfo::WORKGROUP_ID_Z:

 unsigned SIRegisterInfo::findUnusedRegister(const MachineRegisterInfo &MRI,
                                            const TargetRegisterClass *RC) const {
   for (unsigned Reg : *RC)
     if (!MRI.isPhysRegUsed(Reg))
       return Reg;
   return AMDGPU::NoRegister;
 }
 
-unsigned SIRegisterInfo::getNumVGPRsAllowed(unsigned WaveCount) const {
-  switch(WaveCount) {
-    case 10: return 24;
-    case 9:  return 28;
-    case 8:  return 32;
-    case 7:  return 36;
-    case 6:  return 40;
-    case 5:  return 48;
-    case 4:  return 64;
-    case 3:  return 84;
-    case 2:  return 128;
-    default: return 256;
+bool SIRegisterInfo::isVGPR(const MachineRegisterInfo &MRI,
+                            unsigned Reg) const {
+  const TargetRegisterClass *RC;
+  if (TargetRegisterInfo::isVirtualRegister(Reg))
+    RC = MRI.getRegClass(Reg);
+  else
+    RC = getPhysRegClass(Reg);
+
+  return hasVGPRs(RC);
+}
+
+unsigned SIRegisterInfo::getMinNumSGPRs(unsigned NumActiveWavesPerEU) const {
+  if (ST.getGeneration() >= AMDGPUSubtarget::VOLCANIC_ISLANDS) {
+    switch (NumActiveWavesPerEU) {
+      case 0:  return 0;
+      case 10: return 0;
+      case 9:  return 0;
+      case 8:  return 81;
+      default: return 97;
+    }
+  } else {
+    switch (NumActiveWavesPerEU) {
+      case 0:  return 0;
+      case 10: return 0;
+      case 9:  return 49;
+      case 8:  return 57;
+      case 7:  return 65;
+      case 6:  return 73;
+      case 5:  return 81;
+      default: return 97;
+    }
   }
 }
 
-unsigned SIRegisterInfo::getNumSGPRsAllowed(AMDGPUSubtarget::Generation gen,
-                                            unsigned WaveCount) const {
-  if (gen >= AMDGPUSubtarget::VOLCANIC_ISLANDS) {
-    switch (WaveCount) {
+unsigned SIRegisterInfo::getMaxNumSGPRs(unsigned NumActiveWavesPerEU) const {
+  if (ST.getGeneration() >= AMDGPUSubtarget::VOLCANIC_ISLANDS) {
+    switch (NumActiveWavesPerEU) {
+      case 0:  return 80;
       case 10: return 80;
       case 9:  return 80;
       case 8:  return 96;
-      default: return 102;
+      default: return getAddressableNumSGPRs();
     }
   } else {
-    switch(WaveCount) {
+    switch (NumActiveWavesPerEU) {
+      case 0:  return 48;
       case 10: return 48;
       case 9:  return 56;
       case 8:  return 64;
       case 7:  return 72;
       case 6:  return 80;
       case 5:  return 96;
-      default: return 103;
+      default: return getAddressableNumSGPRs();
     }
   }
 }
 
-bool SIRegisterInfo::isVGPR(const MachineRegisterInfo &MRI,
-                            unsigned Reg) const {
-  const TargetRegisterClass *RC;
-  if (TargetRegisterInfo::isVirtualRegister(Reg))
-    RC = MRI.getRegClass(Reg);
-  else
-    RC = getPhysRegClass(Reg);
+unsigned SIRegisterInfo::getMaxNumSGPRs(const MachineFunction &MF) const {
+  const Function &F = *MF.getFunction();
+  LLVMContext &Ctx = F.getContext();
 
-  return hasVGPRs(RC);
+  const SIMachineFunctionInfo &MFI = *MF.getInfo<SIMachineFunctionInfo>();
+
+  // Compute maximum number of SGPRs function can use using default/requested
+  // minimum number of active waves per execution unit.
+  std::pair<unsigned, unsigned> NumActiveWavesPerEU =
+    MFI.getNumActiveWavesPerEU();
+  unsigned MaxNumSGPRs = getMaxNumSGPRs(NumActiveWavesPerEU.first);
+  // Check if maximum number of SGPRs was explicitly requested using
+  // "amdgpu-num-sgpr" attribute.
+  if (F.hasFnAttribute("amdgpu-num-sgpr")) {
+    unsigned Requested =
+      AMDGPU::getIntegerAttribute(F, "amdgpu-num-sgpr", MaxNumSGPRs);
+    // Make sure requested value does not violate subtarget's specifications.
+    if (Requested && Requested <= getReservedNumSGPRs()) {
+      Ctx.emitError("invalid amdgpu-num-sgpr: "
+                    "requested number of SGPRs is not supported by the "
+                    "subtarget (some number of SGPRs is required for internal "
+                    "use)");
+      Requested = 0;
+    }
+    // Make sure requested value is compatible with values implied by
+    // default/requested minimum/maximum number of active waves per execution
+    // unit.
+    if (Requested && Requested > getMaxNumSGPRs(NumActiveWavesPerEU.first)) {
+      Ctx.emitError("invalid amdgpu-num-sgpr: "
+                    "requested number of SGPRs cannot be achieved with default "
+                    "or requested minimum number of active waves per execution "
+                    "unit");
+      Requested = 0;
+    }
+    if (NumActiveWavesPerEU.second &&
+        Requested && Requested < getMinNumSGPRs(NumActiveWavesPerEU.second)) {
+      Ctx.emitError("invalid amdgpu-num-sgpr: "
+                    "requested number of SGPRs cannot be achieved with default "
+                    "or requested maximum number of active waves per execution "
+                    "unit");
+      Requested = 0;
+    }
+    if (Requested)
+      MaxNumSGPRs = Requested;
+  }
+
+  if (ST.hasSGPRInitBug())
+    MaxNumSGPRs = AMDGPUSubtarget::FIXED_SGPR_COUNT_FOR_INIT_BUG;
+
+  return MaxNumSGPRs - getReservedNumSGPRs();
+}
+
+unsigned SIRegisterInfo::getMinNumVGPRs(unsigned NumActiveWavesPerEU) const {
+  switch (NumActiveWavesPerEU) {
+    case 0:  return 0;
+    case 10: return 0;
+    case 9:  return 25;
+    case 8:  return 29;
+    case 7:  return 33;
+    case 6:  return 37;
+    case 5:  return 41;
+    case 4:  return 49;
+    case 3:  return 65;
+    case 2:  return 85;
+    default: return 129;
+  }
+}
+
+unsigned SIRegisterInfo::getMaxNumVGPRs(unsigned NumActiveWavesPerEU) const {
+  switch (NumActiveWavesPerEU) {
+    case 0:  return 24;
+    case 10: return 24;
+    case 9:  return 28;
+    case 8:  return 32;
+    case 7:  return 36;
+    case 6:  return 40;
+    case 5:  return 48;
+    case 4:  return 64;
+    case 3:  return 84;
+    case 2:  return 128;
+    default: return getTotalNumVGPRs();
+  }
+}
+
+unsigned SIRegisterInfo::getMaxNumVGPRs(const MachineFunction &MF) const {
+  const Function &F = *MF.getFunction();
+  LLVMContext &Ctx = F.getContext();
+
+  const SIMachineFunctionInfo &MFI = *MF.getInfo<SIMachineFunctionInfo>();
+
+  // Compute maximum number of VGPRs function can use using default/requested
+  // minimum number of active waves per execution unit.
+  std::pair<unsigned, unsigned> NumActiveWavesPerEU =
+    MFI.getNumActiveWavesPerEU();
+  unsigned MaxNumVGPRs = getMaxNumVGPRs(NumActiveWavesPerEU.first);
+  // Check if maximum number of VGPRs was explicitly requested using
+  // "amdgpu-num-vgpr" attribute.
+  if (F.hasFnAttribute("amdgpu-num-vgpr")) {
+    unsigned Requested =
+      AMDGPU::getIntegerAttribute(F, "amdgpu-num-vgpr", MaxNumVGPRs);
+    // Make sure requested value does not violate subtarget's specifications.
+    if (Requested && Requested <= getDebuggerReservedNumVGPRs()) {
+      Ctx.emitError("invalid amdgpu-num-vgpr: "
+                    "requested number of VGPRs is not supported by the "
+                    "subtarget (some number of VGPRs is required for debugger "
+                    "use)");
+      Requested = 0;
+    }
+    // Make sure requested value is compatible with values implied by
+    // default/requested minimum/maximum number of active waves per execution
+    // unit.
+    if (Requested && Requested > getMaxNumVGPRs(NumActiveWavesPerEU.first)) {
+      Ctx.emitError("invalid amdgpu-num-vgpr: "
+                    "requested number of VGPRs cannot be achieved with default "
+                    "or requested minimum number of active waves per execution "
+                    "unit");
+      Requested = 0;
+    }
+    if (NumActiveWavesPerEU.second &&
+        Requested && Requested < getMinNumVGPRs(NumActiveWavesPerEU.second)) {
+      Ctx.emitError("invalid amdgpu-num-vgpr: "
+                    "requested number of VGPRs cannot be achieved with default "
+                    "or requested maximum number of active waves per execution "
+                    "unit");
+      Requested = 0;
+    }
+    if (Requested)
+      MaxNumVGPRs = Requested;
+  }
+
+  return MaxNumVGPRs - getDebuggerReservedNumVGPRs();
 }

lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h

 MCSection *getHSADataGlobalProgramSection(MCContext &Ctx);
 
 MCSection *getHSARodataReadonlyAgentSection(MCContext &Ctx);
 
 bool isGroupSegment(const GlobalValue *GV);
 bool isGlobalSegment(const GlobalValue *GV);
 bool isReadOnlySegment(const GlobalValue *GV);
 
+/// Returns integer value requested using \p F's \p Name attribute. Returns \p
+/// Default if attribute is not present. Returns \p Default and emits error if
+/// requested value cannot be converted to integer.
 int getIntegerAttribute(const Function &F, StringRef Name, int Default);
 
-unsigned getMaximumWorkGroupSize(const Function &F);
+/// Returns a pair of integer values requested using \p F's \p Name attribute in
+/// "first[,second]" format ("second" is optional unless \p OnlyFirstRequired is
+/// false). Returns \p Default if attribute is not present. Returns \p Default
+/// and emits error if one of the requested values cannot be converted to
+/// integer, or \p OnlyFirstRequired is false and "second" value is not present.
+std::pair<int, int> getIntegerPairAttribute(const Function &F,
+                                            StringRef Name,
+                                            std::pair<int, int> Default,
+                                            bool OnlyFirstRequired = false);
+
 unsigned getInitialPSInputAddr(const Function &F);
 
 bool isShader(CallingConv::ID cc);
 bool isCompute(CallingConv::ID cc);
 
 bool isSI(const MCSubtargetInfo &STI);
 bool isCI(const MCSubtargetInfo &STI);
 bool isVI(const MCSubtargetInfo &STI);

lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp

if (Str.getAsInteger(0, Result)) {
       LLVMContext &Ctx = F.getContext();
       Ctx.emitError("can't parse integer attribute " + Name);
     }
   }
 
   return Result;
 }
 
-unsigned getMaximumWorkGroupSize(const Function &F) {
-  return getIntegerAttribute(F, "amdgpu-max-work-group-size", 256);
+std::pair<int, int> getIntegerPairAttribute(const Function &F,
+                                            StringRef Name,
+                                            std::pair<int, int> Default,
+                                            bool OnlyFirstRequired) {
+  Attribute A = F.getFnAttribute(Name);
+  if (!A.isStringAttribute())
+    return Default;
+
+  LLVMContext &Ctx = F.getContext();
+  std::pair<int, int> Ints = Default;
+  std::pair<StringRef, StringRef> Strs = A.getValueAsString().split(',');
+  if (Strs.first.trim().getAsInteger(0, Ints.first)) {
+    Ctx.emitError("can't parse first integer attribute " + Name);
+    return Default;
+  }
+  if (Strs.second.trim().getAsInteger(0, Ints.second)) {
+    if (!OnlyFirstRequired || Strs.second.trim().size()) {
+      Ctx.emitError("can't parse second integer attribute " + Name);
+      return Default;
+    }
+  }
+
+  return Ints;
 }
 
 unsigned getInitialPSInputAddr(const Function &F) {
   return getIntegerAttribute(F, "InitialPSInputAddr", 0);
 }
 
 bool isShader(CallingConv::ID cc) {
   switch(cc) {

test/CodeGen/AMDGPU/amdgpu.private-memory.ll

 define void @v2float_stack(<2 x float> addrspace(1)* %out, i32 %a) {
   %alloca = alloca [16 x <2 x float>]
   %tmp0 = getelementptr [16 x <2 x float>], [16 x <2 x float>]* %alloca, i32 0, i32 %a
   %tmp5 = load <2 x float>, <2 x float>* %tmp0
   store <2 x float> %tmp5, <2 x float> addrspace(1)* %out
   ret void
 }
 
-attributes #0 = { nounwind "amdgpu-max-waves-per-eu"="2" }
+attributes #0 = { nounwind "amdgpu-num-active-waves-per-eu"="1,2" }

test/CodeGen/AMDGPU/array-ptr-calc-i32.ll

define void @test_private_array_ptr_calc(i32 addrspace(1)* noalias %out, i32 addrspace(1)* noalias %inA, i32 addrspace(1)* noalias %inB) #0 {
   ; Dummy call
   call void @llvm.amdgcn.s.barrier()
   %reload = load i32, i32* %alloca_ptr, align 4
   %out_ptr = getelementptr inbounds i32, i32 addrspace(1)* %out, i32 %tid
   store i32 %reload, i32 addrspace(1)* %out_ptr, align 4
   ret void
 }
 
-attributes #0 = { nounwind "amdgpu-max-waves-per-eu"="1" }
+attributes #0 = { nounwind "amdgpu-num-active-waves-per-eu"="1,1" }
 attributes #1 = { nounwind readnone }
 attributes #2 = { nounwind convergent }

test/CodeGen/AMDGPU/attr-amdgpu-flat-work-group-size-invalid.ll

+; RUN: not llc -march=amdgcn -mcpu=fiji -verify-machineinstrs < %s 2>&1 | FileCheck %s
+
+; CHECK: invalid amdgpu-flat-work-group-size: minimum is greater than maximum
+define void @invalid_0() #0 {
+entry:
+  ret void
+}
+attributes #0 = {"amdgpu-flat-work-group-size"="128,64"}
+
+; CHECK: invalid amdgpu-flat-work-group-size: requested maximum flat work group size is not supported by the subtarget
+define void @invalid_1() #1 {
+entry:
+  ret void
+}
+attributes #1 = {"amdgpu-flat-work-group-size"="2048,4096"}

test/CodeGen/AMDGPU/attr-amdgpu-flat-work-group-size-many-regs.ll

+; RUN: llc -march=amdgcn -mcpu=fiji -verify-machineinstrs < %s | FileCheck %s
+
+; CHECK: SGPRBlocks: 1
+; CHECK: VGPRBlocks: 7
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 11
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 32
+define void @many_regs_128_256(i32 addrspace(1)* nocapture %A) #0 {
+entry:
+  %a = alloca [111 x i32], align 4
+  %0 = bitcast [111 x i32]* %a to i8*
+  call void @llvm.lifetime.start(i64 444, i8* %0) #2
+  %1 = load i32, i32 addrspace(1)* %A, align 4, !tbaa !7
+  %arrayidx1 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 0
+  store i32 %1, i32* %arrayidx1, align 4, !tbaa !7
+  %arrayidx.1 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 1
+  %2 = load i32, i32 addrspace(1)* %arrayidx.1, align 4, !tbaa !7
+  %arrayidx1.1 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 1
+  store i32 %2, i32* %arrayidx1.1, align 4, !tbaa !7
+  %arrayidx.2 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 2
+  %3 = load i32, i32 addrspace(1)* %arrayidx.2, align 4, !tbaa !7
+  %arrayidx1.2 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 2
+  store i32 %3, i32* %arrayidx1.2, align 4, !tbaa !7
+  %arrayidx.3 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 3
+  %4 = load i32, i32 addrspace(1)* %arrayidx.3, align 4, !tbaa !7
+  %arrayidx1.3 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 3
+  store i32 %4, i32* %arrayidx1.3, align 4, !tbaa !7
+  %arrayidx.4 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 4
+  %5 = load i32, i32 addrspace(1)* %arrayidx.4, align 4, !tbaa !7
+  %arrayidx1.4 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 4
+  store i32 %5, i32* %arrayidx1.4, align 4, !tbaa !7
+  %arrayidx.5 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 5
+  %6 = load i32, i32 addrspace(1)* %arrayidx.5, align 4, !tbaa !7
+  %arrayidx1.5 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 5
+  store i32 %6, i32* %arrayidx1.5, align 4, !tbaa !7
+  %arrayidx.6 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 6
+  %7 = load i32, i32 addrspace(1)* %arrayidx.6, align 4, !tbaa !7
+  %arrayidx1.6 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 6
+  store i32 %7, i32* %arrayidx1.6, align 4, !tbaa !7
+  %arrayidx.7 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 7
+  %8 = load i32, i32 addrspace(1)* %arrayidx.7, align 4, !tbaa !7
+  %arrayidx1.7 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 7
+  store i32 %8, i32* %arrayidx1.7, align 4, !tbaa !7
+  %arrayidx.8 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 8
+  %9 = load i32, i32 addrspace(1)* %arrayidx.8, align 4, !tbaa !7
+  %arrayidx1.8 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 8
+  store i32 %9, i32* %arrayidx1.8, align 4, !tbaa !7
+  %arrayidx.9 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 9
+  %10 = load i32, i32 addrspace(1)* %arrayidx.9, align 4, !tbaa !7
+  %arrayidx1.9 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 9
+  store i32 %10, i32* %arrayidx1.9, align 4, !tbaa !7
+  %arrayidx.10 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 10
+  %11 = load i32, i32 addrspace(1)* %arrayidx.10, align 4, !tbaa !7
+  %arrayidx1.10 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 10
+  store i32 %11, i32* %arrayidx1.10, align 4, !tbaa !7
+  %arrayidx.11 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 11
+  %12 = load i32, i32 addrspace(1)* %arrayidx.11, align 4, !tbaa !7
+  %arrayidx1.11 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 11
+  store i32 %12, i32* %arrayidx1.11, align 4, !tbaa !7
+  %arrayidx.12 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 12
+  %13 = load i32, i32 addrspace(1)* %arrayidx.12, align 4, !tbaa !7
+  %arrayidx1.12 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 12
+  store i32 %13, i32* %arrayidx1.12, align 4, !tbaa !7
+  %arrayidx.13 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 13
+  %14 = load i32, i32 addrspace(1)* %arrayidx.13, align 4, !tbaa !7
+  %arrayidx1.13 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 13
+  store i32 %14, i32* %arrayidx1.13, align 4, !tbaa !7
+  %arrayidx.14 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 14
+  %15 = load i32, i32 addrspace(1)* %arrayidx.14, align 4, !tbaa !7
+  %arrayidx1.14 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 14
+  store i32 %15, i32* %arrayidx1.14, align 4, !tbaa !7
+  %arrayidx.15 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 15
+  %16 = load i32, i32 addrspace(1)* %arrayidx.15, align 4, !tbaa !7
+  %arrayidx1.15 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 15
+  store i32 %16, i32* %arrayidx1.15, align 4, !tbaa !7
+  %arrayidx.16 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 16
+  %17 = load i32, i32 addrspace(1)* %arrayidx.16, align 4, !tbaa !7
+  %arrayidx1.16 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 16
+  store i32 %17, i32* %arrayidx1.16, align 4, !tbaa !7
+  %arrayidx.17 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 17
+  %18 = load i32, i32 addrspace(1)* %arrayidx.17, align 4, !tbaa !7
+  %arrayidx1.17 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 17
+  store i32 %18, i32* %arrayidx1.17, align 4, !tbaa !7
+  %arrayidx.18 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 18
+  %19 = load i32, i32 addrspace(1)* %arrayidx.18, align 4, !tbaa !7
+  %arrayidx1.18 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 18
+  store i32 %19, i32* %arrayidx1.18, align 4, !tbaa !7
+  %arrayidx.19 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 19
+  %20 = load i32, i32 addrspace(1)* %arrayidx.19, align 4, !tbaa !7
+  %arrayidx1.19 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 19
+  store i32 %20, i32* %arrayidx1.19, align 4, !tbaa !7
+  %arrayidx.20 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 20
+  %21 = load i32, i32 addrspace(1)* %arrayidx.20, align 4, !tbaa !7
+  %arrayidx1.20 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 20
+  store i32 %21, i32* %arrayidx1.20, align 4, !tbaa !7
+  %arrayidx.21 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 21
+  %22 = load i32, i32 addrspace(1)* %arrayidx.21, align 4, !tbaa !7
+  %arrayidx1.21 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 21
+  store i32 %22, i32* %arrayidx1.21, align 4, !tbaa !7
+  %arrayidx.22 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 22
+  %23 = load i32, i32 addrspace(1)* %arrayidx.22, align 4, !tbaa !7
+  %arrayidx1.22 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 22
+  store i32 %23, i32* %arrayidx1.22, align 4, !tbaa !7
+  %arrayidx.23 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 23
+  %24 = load i32, i32 addrspace(1)* %arrayidx.23, align 4, !tbaa !7
+  %arrayidx1.23 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 23
+  store i32 %24, i32* %arrayidx1.23, align 4, !tbaa !7
+  %arrayidx.24 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 24
+  %25 = load i32, i32 addrspace(1)* %arrayidx.24, align 4, !tbaa !7
+  %arrayidx1.24 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 24
+  store i32 %25, i32* %arrayidx1.24, align 4, !tbaa !7
+  %arrayidx.25 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 25
+  %26 = load i32, i32 addrspace(1)* %arrayidx.25, align 4, !tbaa !7
+  %arrayidx1.25 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 25
+  store i32 %26, i32* %arrayidx1.25, align 4, !tbaa !7
+  %arrayidx.26 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 26
+  %27 = load i32, i32 addrspace(1)* %arrayidx.26, align 4, !tbaa !7
+  %arrayidx1.26 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 26
+  store i32 %27, i32* %arrayidx1.26, align 4, !tbaa !7
+  %arrayidx.27 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 27
+  %28 = load i32, i32 addrspace(1)* %arrayidx.27, align 4, !tbaa !7
+  %arrayidx1.27 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 27
+  store i32 %28, i32* %arrayidx1.27, align 4, !tbaa !7
+  %arrayidx.28 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 28
+  %29 = load i32, i32 addrspace(1)* %arrayidx.28, align 4, !tbaa !7
+  %arrayidx1.28 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 28
+  store i32 %29, i32* %arrayidx1.28, align 4, !tbaa !7
+  %arrayidx.29 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 29
+  %30 = load i32, i32 addrspace(1)* %arrayidx.29, align 4, !tbaa !7
+  %arrayidx1.29 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 29
+  store i32 %30, i32* %arrayidx1.29, align 4, !tbaa !7
+  %arrayidx.30 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 30
+  %31 = load i32, i32 addrspace(1)* %arrayidx.30, align 4, !tbaa !7
+  %arrayidx1.30 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 30
+  store i32 %31, i32* %arrayidx1.30, align 4, !tbaa !7
+  %arrayidx.31 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 31
+  %32 = load i32, i32 addrspace(1)* %arrayidx.31, align 4, !tbaa !7
+  %arrayidx1.31 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 31
+  store i32 %32, i32* %arrayidx1.31, align 4, !tbaa !7
+  %arrayidx.32 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 32
+  %33 = load i32, i32 addrspace(1)* %arrayidx.32, align 4, !tbaa !7
+  %arrayidx1.32 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 32
+  store i32 %33, i32* %arrayidx1.32, align 4, !tbaa !7
+  %arrayidx.33 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 33
+  %34 = load i32, i32 addrspace(1)* %arrayidx.33, align 4, !tbaa !7
+  %arrayidx1.33 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 33
+  store i32 %34, i32* %arrayidx1.33, align 4, !tbaa !7
+  %arrayidx.34 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 34
+  %35 = load i32, i32 addrspace(1)* %arrayidx.34, align 4, !tbaa !7
+  %arrayidx1.34 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 34
+  store i32 %35, i32* %arrayidx1.34, align 4, !tbaa !7
+  %arrayidx.35 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 35
+  %36 = load i32, i32 addrspace(1)* %arrayidx.35, align 4, !tbaa !7
+  %arrayidx1.35 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 35
+  store i32 %36, i32* %arrayidx1.35, align 4, !tbaa !7
+  %arrayidx.36 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 36
+  %37 = load i32, i32 addrspace(1)* %arrayidx.36, align 4, !tbaa !7
+  %arrayidx1.36 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 36
+  store i32 %37, i32* %arrayidx1.36, align 4, !tbaa !7
+  %arrayidx.37 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 37
+  %38 = load i32, i32 addrspace(1)* %arrayidx.37, align 4, !tbaa !7
+  %arrayidx1.37 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 37
+  store i32 %38, i32* %arrayidx1.37, align 4, !tbaa !7
+  %arrayidx.38 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 38
+  %39 = load i32, i32 addrspace(1)* %arrayidx.38, align 4, !tbaa !7
+  %arrayidx1.38 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 38
+  store i32 %39, i32* %arrayidx1.38, align 4, !tbaa !7
+  %arrayidx.39 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 39
+  %40 = load i32, i32 addrspace(1)* %arrayidx.39, align 4, !tbaa !7
+  %arrayidx1.39 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 39
+  store i32 %40, i32* %arrayidx1.39, align 4, !tbaa !7
+  %arrayidx.40 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 40
+  %41 = load i32, i32 addrspace(1)* %arrayidx.40, align 4, !tbaa !7
+  %arrayidx1.40 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 40
+  store i32 %41, i32* %arrayidx1.40, align 4, !tbaa !7
+  %arrayidx.41 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 41
+  %42 = load i32, i32 addrspace(1)* %arrayidx.41, align 4, !tbaa !7
+  %arrayidx1.41 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 41
+  store i32 %42, i32* %arrayidx1.41, align 4, !tbaa !7
+  %arrayidx.42 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 42
+  %43 = load i32, i32 addrspace(1)* %arrayidx.42, align 4, !tbaa !7
+  %arrayidx1.42 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 42
+  store i32 %43, i32* %arrayidx1.42, align 4, !tbaa !7
+  %arrayidx.43 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 43
+  %44 = load i32, i32 addrspace(1)* %arrayidx.43, align 4, !tbaa !7
+  %arrayidx1.43 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 43
+  store i32 %44, i32* %arrayidx1.43, align 4, !tbaa !7
+  %arrayidx.44 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 44
+  %45 = load i32, i32 addrspace(1)* %arrayidx.44, align 4, !tbaa !7
+  %arrayidx1.44 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 44
+  store i32 %45, i32* %arrayidx1.44, align 4, !tbaa !7
+  %arrayidx.45 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 45
+  %46 = load i32, i32 addrspace(1)* %arrayidx.45, align 4, !tbaa !7
+  %arrayidx1.45 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 45
+  store i32 %46, i32* %arrayidx1.45, align 4, !tbaa !7
+  %arrayidx.46 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 46
+  %47 = load i32, i32 addrspace(1)* %arrayidx.46, align 4, !tbaa !7
+  %arrayidx1.46 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 46
+  store i32 %47, i32* %arrayidx1.46, align 4, !tbaa !7
+  %arrayidx.47 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 47
+  %48 = load i32, i32 addrspace(1)* %arrayidx.47, align 4, !tbaa !7
+  %arrayidx1.47 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 47
+  store i32 %48, i32* %arrayidx1.47, align 4, !tbaa !7
+  %arrayidx.48 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 48
+  %49 = load i32, i32 addrspace(1)* %arrayidx.48, align 4, !tbaa !7
+  %arrayidx1.48 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 48
+  store i32 %49, i32* %arrayidx1.48, align 4, !tbaa !7
+  %arrayidx.49 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 49
+  %50 = load i32, i32 addrspace(1)* %arrayidx.49, align 4, !tbaa !7
+  %arrayidx1.49 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 49
+  store i32 %50, i32* %arrayidx1.49, align 4, !tbaa !7
+  %arrayidx.50 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 50
+  %51 = load i32, i32 addrspace(1)* %arrayidx.50, align 4, !tbaa !7
+  %arrayidx1.50 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 50
+  store i32 %51, i32* %arrayidx1.50, align 4, !tbaa !7
+  %arrayidx.51 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 51
+  %52 = load i32, i32 addrspace(1)* %arrayidx.51, align 4, !tbaa !7
+  %arrayidx1.51 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 51
+  store i32 %52, i32* %arrayidx1.51, align 4, !tbaa !7
+  %arrayidx.52 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 52
+  %53 = load i32, i32 addrspace(1)* %arrayidx.52, align 4, !tbaa !7
+  %arrayidx1.52 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 52
+  store i32 %53, i32* %arrayidx1.52, align 4, !tbaa !7
+  %arrayidx.53 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 53
+  %54 = load i32, i32 addrspace(1)* %arrayidx.53, align 4, !tbaa !7
+  %arrayidx1.53 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 53
+  store i32 %54, i32* %arrayidx1.53, align 4, !tbaa !7
+  %arrayidx.54 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 54
+  %55 = load i32, i32 addrspace(1)* %arrayidx.54, align 4, !tbaa !7
+  %arrayidx1.54 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 54
+  store i32 %55, i32* %arrayidx1.54, align 4, !tbaa !7
+  %arrayidx.55 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 55
+  %56 = load i32, i32 addrspace(1)* %arrayidx.55, align 4, !tbaa !7
+  %arrayidx1.55 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 55
+  store i32 %56, i32* %arrayidx1.55, align 4, !tbaa !7
+  %arrayidx.56 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 56
+  %57 = load i32, i32 addrspace(1)* %arrayidx.56, align 4, !tbaa !7
+  %arrayidx1.56 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 56
+  store i32 %57, i32* %arrayidx1.56, align 4, !tbaa !7
+  %arrayidx.57 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 57
+  %58 = load i32, i32 addrspace(1)* %arrayidx.57, align 4, !tbaa !7
+  %arrayidx1.57 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 57
+  store i32 %58, i32* %arrayidx1.57, align 4, !tbaa !7
+  %arrayidx.58 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 58
+  %59 = load i32, i32 addrspace(1)* %arrayidx.58, align 4, !tbaa !7
+  %arrayidx1.58 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 58
+  store i32 %59, i32* %arrayidx1.58, align 4, !tbaa !7
+  %arrayidx.59 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 59
+  %60 = load i32, i32 addrspace(1)* %arrayidx.59, align 4, !tbaa !7
+  %arrayidx1.59 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 59
+  store i32 %60, i32* %arrayidx1.59, align 4, !tbaa !7
+  %arrayidx.60 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 60
+  %61 = load i32, i32 addrspace(1)* %arrayidx.60, align 4, !tbaa !7
+  %arrayidx1.60 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 60
+  store i32 %61, i32* %arrayidx1.60, align 4, !tbaa !7
+  %arrayidx.61 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 61
+  %62 = load i32, i32 addrspace(1)* %arrayidx.61, align 4, !tbaa !7
+  %arrayidx1.61 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 61
+  store i32 %62, i32* %arrayidx1.61, align 4, !tbaa !7
+  %arrayidx.62 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 62
+  %63 = load i32, i32 addrspace(1)* %arrayidx.62, align 4, !tbaa !7
+  %arrayidx1.62 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 62
+  store i32 %63, i32* %arrayidx1.62, align 4, !tbaa !7
+  %arrayidx.63 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 63
+  %64 = load i32, i32 addrspace(1)* %arrayidx.63, align 4, !tbaa !7
+  %arrayidx1.63 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 63
+  store i32 %64, i32* %arrayidx1.63, align 4, !tbaa !7
+  %arrayidx.64 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 64
+  %65 = load i32, i32 addrspace(1)* %arrayidx.64, align 4, !tbaa !7
+  %arrayidx1.64 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 64
+  store i32 %65, i32* %arrayidx1.64, align 4, !tbaa !7
+  %arrayidx.65 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 65
+  %66 = load i32, i32 addrspace(1)* %arrayidx.65, align 4, !tbaa !7
+  %arrayidx1.65 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 65
+  store i32 %66, i32* %arrayidx1.65, align 4, !tbaa !7
+  %arrayidx.66 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 66
+  %67 = load i32, i32 addrspace(1)* %arrayidx.66, align 4, !tbaa !7
+  %arrayidx1.66 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 66
+  store i32 %67, i32* %arrayidx1.66, align 4, !tbaa !7
+  %arrayidx.67 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 67
+  %68 = load i32, i32 addrspace(1)* %arrayidx.67, align 4, !tbaa !7
+  %arrayidx1.67 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 67
+  store i32 %68, i32* %arrayidx1.67, align 4, !tbaa !7
+  %arrayidx.68 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 68
+  %69 = load i32, i32 addrspace(1)* %arrayidx.68, align 4, !tbaa !7
+  %arrayidx1.68 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 68
+  store i32 %69, i32* %arrayidx1.68, align 4, !tbaa !7
+  %arrayidx.69 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 69
+  %70 = load i32, i32 addrspace(1)* %arrayidx.69, align 4, !tbaa !7
+  %arrayidx1.69 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 69
+  store i32 %70, i32* %arrayidx1.69, align 4, !tbaa !7
+  %arrayidx.70 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 70
+  %71 = load i32, i32 addrspace(1)* %arrayidx.70, align 4, !tbaa !7
+  %arrayidx1.70 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 70
+  store i32 %71, i32* %arrayidx1.70, align 4, !tbaa !7
+  %arrayidx.71 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 71
+  %72 = load i32, i32 addrspace(1)* %arrayidx.71, align 4, !tbaa !7
+  %arrayidx1.71 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 71
+  store i32 %72, i32* %arrayidx1.71, align 4, !tbaa !7
+  %arrayidx.72 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 72
+  %73 = load i32, i32 addrspace(1)* %arrayidx.72, align 4, !tbaa !7
+  %arrayidx1.72 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 72
+  store i32 %73, i32* %arrayidx1.72, align 4, !tbaa !7
+  %arrayidx.73 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 73
+  %74 = load i32, i32 addrspace(1)* %arrayidx.73, align 4, !tbaa !7
+  %arrayidx1.73 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 73
+  store i32 %74, i32* %arrayidx1.73, align 4, !tbaa !7
+  %arrayidx.74 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 74
+  %75 = load i32, i32 addrspace(1)* %arrayidx.74, align 4, !tbaa !7
+  %arrayidx1.74 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 74
+  store i32 %75, i32* %arrayidx1.74, align 4, !tbaa !7
+  %arrayidx.75 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 75
+  %76 = load i32, i32 addrspace(1)* %arrayidx.75, align 4, !tbaa !7
+  %arrayidx1.75 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 75
+  store i32 %76, i32* %arrayidx1.75, align 4, !tbaa !7
+  %arrayidx.76 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 76
+  %77 = load i32, i32 addrspace(1)* %arrayidx.76, align 4, !tbaa !7
+  %arrayidx1.76 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 76
+  store i32 %77, i32* %arrayidx1.76, align 4, !tbaa !7
+  %arrayidx.77 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 77
+  %78 = load i32, i32 addrspace(1)* %arrayidx.77, align 4, !tbaa !7
+  %arrayidx1.77 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 77
+  store i32 %78, i32* %arrayidx1.77, align 4, !tbaa !7
+  %arrayidx.78 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 78
+  %79 = load i32, i32 addrspace(1)* %arrayidx.78, align 4, !tbaa !7
+  %arrayidx1.78 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 78
+  store i32 %79, i32* %arrayidx1.78, align 4, !tbaa !7
+  %arrayidx.79 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 79
+  %80 = load i32, i32 addrspace(1)* %arrayidx.79, align 4, !tbaa !7
+  %arrayidx1.79 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 79
+  store i32 %80, i32* %arrayidx1.79, align 4, !tbaa !7
+  %arrayidx.80 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 80
+  %81 = load i32, i32 addrspace(1)* %arrayidx.80, align 4, !tbaa !7
+  %arrayidx1.80 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 80
+  store i32 %81, i32* %arrayidx1.80, align 4, !tbaa !7
+  %arrayidx.81 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 81
+  %82 = load i32, i32 addrspace(1)* %arrayidx.81, align 4, !tbaa !7
+  %arrayidx1.81 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 81
+  store i32 %82, i32* %arrayidx1.81, align 4, !tbaa !7
+  %arrayidx.82 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 82
+  %83 = load i32, i32 addrspace(1)* %arrayidx.82, align 4, !tbaa !7
+  %arrayidx1.82 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 82
+  store i32 %83, i32* %arrayidx1.82, align 4, !tbaa !7
+  %arrayidx.83 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 83
+  %84 = load i32, i32 addrspace(1)* %arrayidx.83, align 4, !tbaa !7
+  %arrayidx1.83 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 83
+  store i32 %84, i32* %arrayidx1.83, align 4, !tbaa !7
+  %arrayidx.84 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 84
+  %85 = load i32, i32 addrspace(1)* %arrayidx.84, align 4, !tbaa !7
+  %arrayidx1.84 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 84
+  store i32 %85, i32* %arrayidx1.84, align 4, !tbaa !7
+  %arrayidx.85 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 85
+  %86 = load i32, i32 addrspace(1)* %arrayidx.85, align 4, !tbaa !7
+  %arrayidx1.85 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 85
+  store i32 %86, i32* %arrayidx1.85, align 4, !tbaa !7
+  %arrayidx.86 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 86
+  %87 = load i32, i32 addrspace(1)* %arrayidx.86, align 4, !tbaa !7
+  %arrayidx1.86 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 86
+  store i32 %87, i32* %arrayidx1.86, align 4, !tbaa !7
+  %arrayidx.87 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 87
+  %88 = load i32, i32 addrspace(1)* %arrayidx.87, align 4, !tbaa !7
+  %arrayidx1.87 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 87
+  store i32 %88, i32* %arrayidx1.87, align 4, !tbaa !7
+  %arrayidx.88 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 88
+  %89 = load i32, i32 addrspace(1)* %arrayidx.88, align 4, !tbaa !7
+  %arrayidx1.88 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 88
+  store i32 %89, i32* %arrayidx1.88, align 4, !tbaa !7
+  %arrayidx.89 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 89
+  %90 = load i32, i32 addrspace(1)* %arrayidx.89, align 4, !tbaa !7
+  %arrayidx1.89 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 89
+  store i32 %90, i32* %arrayidx1.89, align 4, !tbaa !7
+  %arrayidx.90 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 90
+  %91 = load i32, i32 addrspace(1)* %arrayidx.90, align 4, !tbaa !7
+  %arrayidx1.90 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 90
+  store i32 %91, i32* %arrayidx1.90, align 4, !tbaa !7
+  %arrayidx.91 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 91
+  %92 = load i32, i32 addrspace(1)* %arrayidx.91, align 4, !tbaa !7
+  %arrayidx1.91 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 91
+  store i32 %92, i32* %arrayidx1.91, align 4, !tbaa !7
+  %arrayidx.92 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 92
+  %93 = load i32, i32 addrspace(1)* %arrayidx.92, align 4, !tbaa !7
+  %arrayidx1.92 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 92
+  store i32 %93, i32* %arrayidx1.92, align 4, !tbaa !7
+  %arrayidx.93 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 93
+  %94 = load i32, i32 addrspace(1)* %arrayidx.93, align 4, !tbaa !7
+  %arrayidx1.93 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 93
+  store i32 %94, i32* %arrayidx1.93, align 4, !tbaa !7
+  %arrayidx.94 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 94
+  %95 = load i32, i32 addrspace(1)* %arrayidx.94, align 4, !tbaa !7
+  %arrayidx1.94 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 94
+  store i32 %95, i32* %arrayidx1.94, align 4, !tbaa !7
+  %arrayidx.95 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 95
+  %96 = load i32, i32 addrspace(1)* %arrayidx.95, align 4, !tbaa !7
+  %arrayidx1.95 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 95
+  store i32 %96, i32* %arrayidx1.95, align 4, !tbaa !7
+  %arrayidx.96 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 96
+  %97 = load i32, i32 addrspace(1)* %arrayidx.96, align 4, !tbaa !7
+  %arrayidx1.96 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 96
+  store i32 %97, i32* %arrayidx1.96, align 4, !tbaa !7
+  %arrayidx.97 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 97
+  %98 = load i32, i32 addrspace(1)* %arrayidx.97, align 4, !tbaa !7
+  %arrayidx1.97 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 97
+  store i32 %98, i32* %arrayidx1.97, align 4, !tbaa !7
+  %arrayidx.98 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 98
+  %99 = load i32, i32 addrspace(1)* %arrayidx.98, align 4, !tbaa !7
+  %arrayidx1.98 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 98
+  store i32 %99, i32* %arrayidx1.98, align 4, !tbaa !7
+  %arrayidx.99 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 99
+  %100 = load i32, i32 addrspace(1)* %arrayidx.99, align 4, !tbaa !7
+  %arrayidx1.99 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 99
+  store i32 %100, i32* %arrayidx1.99, align 4, !tbaa !7
+  %arrayidx.100 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 100
+  %101 = load i32, i32 addrspace(1)* %arrayidx.100, align 4, !tbaa !7
+  %arrayidx1.100 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 100
+  store i32 %101, i32* %arrayidx1.100, align 4, !tbaa !7
+  %arrayidx.101 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 101
+  %102 = load i32, i32 addrspace(1)* %arrayidx.101, align 4, !tbaa !7
+  %arrayidx1.101 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 101
+  store i32 %102, i32* %arrayidx1.101, align 4, !tbaa !7
+  %arrayidx.102 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 102
+  %103 = load i32, i32 addrspace(1)* %arrayidx.102, align 4, !tbaa !7
+  %arrayidx1.102 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 102
+  store i32 %103, i32* %arrayidx1.102, align 4, !tbaa !7
+  %arrayidx.103 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 103
+  %104 = load i32, i32 addrspace(1)* %arrayidx.103, align 4, !tbaa !7
+  %arrayidx1.103 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 103
+  store i32 %104, i32* %arrayidx1.103, align 4, !tbaa !7
+  %arrayidx.104 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 104
+  %105 = load i32, i32 addrspace(1)* %arrayidx.104, align 4, !tbaa !7
+  %arrayidx1.104 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 104
+  store i32 %105, i32* %arrayidx1.104, align 4, !tbaa !7
+  %arrayidx.105 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 105
+  %106 = load i32, i32 addrspace(1)* %arrayidx.105, align 4, !tbaa !7
+  %arrayidx1.105 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 105
+  store i32 %106, i32* %arrayidx1.105, align 4, !tbaa !7
+  %arrayidx.106 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 106
+  %107 = load i32, i32 addrspace(1)* %arrayidx.106, align 4, !tbaa !7
+  %arrayidx1.106 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 106
+  store i32 %107, i32* %arrayidx1.106, align 4, !tbaa !7
+  %arrayidx.107 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 107
+  %108 = load i32, i32 addrspace(1)* %arrayidx.107, align 4, !tbaa !7
+  %arrayidx1.107 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 107
+  store i32 %108, i32* %arrayidx1.107, align 4, !tbaa !7
+  %arrayidx.108 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 108
+  %109 = load i32, i32 addrspace(1)* %arrayidx.108, align 4, !tbaa !7
+  %arrayidx1.108 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 108
+  store i32 %109, i32* %arrayidx1.108, align 4, !tbaa !7
+  %arrayidx.109 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 109
+  %110 = load i32, i32 addrspace(1)* %arrayidx.109, align 4, !tbaa !7
+  %arrayidx1.109 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 109
+  store i32 %110, i32* %arrayidx1.109, align 4, !tbaa !7
+  %arrayidx.110 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 110
+  %111 = load i32, i32 addrspace(1)* %arrayidx.110, align 4, !tbaa !7
+  %arrayidx1.110 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 110
+  store i32 %111, i32* %arrayidx1.110, align 4, !tbaa !7
+  %112 = load i32, i32* %arrayidx1, align 4, !tbaa !7
+  %mul = sub nsw i32 0, %112
+  store i32 %mul, i32 addrspace(1)* %A, align 4, !tbaa !7
+  %113 = load i32, i32* %arrayidx1.1, align 4, !tbaa !7
+  %mul.1 = sub nsw i32 0, %113
+  store i32 %mul.1, i32 addrspace(1)* %arrayidx.1, align 4, !tbaa !7
+  %114 = load i32, i32* %arrayidx1.2, align 4, !tbaa !7
+  %mul.2 = sub nsw i32 0, %114
+  store i32 %mul.2, i32 addrspace(1)* %arrayidx.2, align 4, !tbaa !7
+  %115 = load i32, i32* %arrayidx1.3, align 4, !tbaa !7
+  %mul.3 = sub nsw i32 0, %115
+  store i32 %mul.3, i32 addrspace(1)* %arrayidx.3, align 4, !tbaa !7
+  %116 = load i32, i32* %arrayidx1.4, align 4, !tbaa !7
+  %mul.4 = sub nsw i32 0, %116
+  store i32 %mul.4, i32 addrspace(1)* %arrayidx.4, align 4, !tbaa !7
+  %117 = load i32, i32* %arrayidx1.5, align 4, !tbaa !7
+  %mul.5 = sub nsw i32 0, %117
+  store i32 %mul.5, i32 addrspace(1)* %arrayidx.5, align 4, !tbaa !7
+  %118 = load i32, i32* %arrayidx1.6, align 4, !tbaa !7
+  %mul.6 = sub nsw i32 0, %118
+  store i32 %mul.6, i32 addrspace(1)* %arrayidx.6, align 4, !tbaa !7
+  %119 = load i32, i32* %arrayidx1.7, align 4, !tbaa !7
+  %mul.7 = sub nsw i32 0, %119
+  store i32 %mul.7, i32 addrspace(1)* %arrayidx.7, align 4, !tbaa !7
+  %120 = load i32, i32* %arrayidx1.8, align 4, !tbaa !7
+  %mul.8 = sub nsw i32 0, %120
+  store i32 %mul.8, i32 addrspace(1)* %arrayidx.8, align 4, !tbaa !7
+  %121 = load i32, i32* %arrayidx1.9, align 4, !tbaa !7
+  %mul.9 = sub nsw i32 0, %121
+  store i32 %mul.9, i32 addrspace(1)* %arrayidx.9, align 4, !tbaa !7
+  %122 = load i32, i32* %arrayidx1.10, align 4, !tbaa !7
+  %mul.10 = sub nsw i32 0, %122
+  store i32 %mul.10, i32 addrspace(1)* %arrayidx.10, align 4, !tbaa !7
+  %123 = load i32, i32* %arrayidx1.11, align 4, !tbaa !7
+  %mul.11 = sub nsw i32 0, %123
+  store i32 %mul.11, i32 addrspace(1)* %arrayidx.11, align 4, !tbaa !7
+  %124 = load i32, i32* %arrayidx1.12, align 4, !tbaa !7
+  %mul.12 = sub nsw i32 0, %124
+  store i32 %mul.12, i32 addrspace(1)* %arrayidx.12, align 4, !tbaa !7
+  %125 = load i32, i32* %arrayidx1.13, align 4, !tbaa !7
+  %mul.13 = sub nsw i32 0, %125
+  store i32 %mul.13, i32 addrspace(1)* %arrayidx.13, align 4, !tbaa !7
+  %126 = load i32, i32* %arrayidx1.14, align 4, !tbaa !7
+  %mul.14 = sub nsw i32 0, %126
+  store i32 %mul.14, i32 addrspace(1)* %arrayidx.14, align 4, !tbaa !7
+  %127 = load i32, i32* %arrayidx1.15, align 4, !tbaa !7
+  %mul.15 = sub nsw i32 0, %127
+  store i32 %mul.15, i32 addrspace(1)* %arrayidx.15, align 4, !tbaa !7
+  %128 = load i32, i32* %arrayidx1.16, align 4, !tbaa !7
+  %mul.16 = sub nsw i32 0, %128
+  store i32 %mul.16, i32 addrspace(1)* %arrayidx.16, align 4, !tbaa !7
+  %129 = load i32, i32* %arrayidx1.17, align 4, !tbaa !7
+  %mul.17 = sub nsw i32 0, %129
+  store i32 %mul.17, i32 addrspace(1)* %arrayidx.17, align 4, !tbaa !7
+  %130 = load i32, i32* %arrayidx1.18, align 4, !tbaa !7
+  %mul.18 = sub nsw i32 0, %130
+  store i32 %mul.18, i32 addrspace(1)* %arrayidx.18, align 4, !tbaa !7
+  %131 = load i32, i32* %arrayidx1.19, align 4, !tbaa !7
+  %mul.19 = sub nsw i32 0, %131
+  store i32 %mul.19, i32 addrspace(1)* %arrayidx.19, align 4, !tbaa !7
+  %132 = load i32, i32* %arrayidx1.20, align 4, !tbaa !7
+  %mul.20 = sub nsw i32 0, %132
+  store i32 %mul.20, i32 addrspace(1)* %arrayidx.20, align 4, !tbaa !7
+  %133 = load i32, i32* %arrayidx1.21, align 4, !tbaa !7
+  %mul.21 = sub nsw i32 0, %133
+  store i32 %mul.21, i32 addrspace(1)* %arrayidx.21, align 4, !tbaa !7
+  %134 = load i32, i32* %arrayidx1.22, align 4, !tbaa !7
+  %mul.22 = sub nsw i32 0, %134
+  store i32 %mul.22, i32 addrspace(1)* %arrayidx.22, align 4, !tbaa !7
+  %135 = load i32, i32* %arrayidx1.23, align 4, !tbaa !7
+  %mul.23 = sub nsw i32 0, %135
+  store i32 %mul.23, i32 addrspace(1)* %arrayidx.23, align 4, !tbaa !7
+  %136 = load i32, i32* %arrayidx1.24, align 4, !tbaa !7
+  %mul.24 = sub nsw i32 0, %136
+  store i32 %mul.24, i32 addrspace(1)* %arrayidx.24, align 4, !tbaa !7
+  %137 = load i32, i32* %arrayidx1.25, align 4, !tbaa !7
+  %mul.25 = sub nsw i32 0, %137
+  store i32 %mul.25, i32 addrspace(1)* %arrayidx.25, align 4, !tbaa !7
+  %138 = load i32, i32* %arrayidx1.26, align 4, !tbaa !7
+  %mul.26 = sub nsw i32 0, %138
+  store i32 %mul.26, i32 addrspace(1)* %arrayidx.26, align 4, !tbaa !7
+  %139 = load i32, i32* %arrayidx1.27, align 4, !tbaa !7
+  %mul.27 = sub nsw i32 0, %139
+  store i32 %mul.27, i32 addrspace(1)* %arrayidx.27, align 4, !tbaa !7
+  %140 = load i32, i32* %arrayidx1.28, align 4, !tbaa !7
+  %mul.28 = sub nsw i32 0, %140
+  store i32 %mul.28, i32 addrspace(1)* %arrayidx.28, align 4, !tbaa !7
+  %141 = load i32, i32* %arrayidx1.29, align 4, !tbaa !7
+  %mul.29 = sub nsw i32 0, %141
+  store i32 %mul.29, i32 addrspace(1)* %arrayidx.29, align 4, !tbaa !7
+  %142 = load i32, i32* %arrayidx1.30, align 4, !tbaa !7
+  %mul.30 = sub nsw i32 0, %142
+  store i32 %mul.30, i32 addrspace(1)* %arrayidx.30, align 4, !tbaa !7
+  %143 = load i32, i32* %arrayidx1.31, align 4, !tbaa !7
+  %mul.31 = sub nsw i32 0, %143
+  store i32 %mul.31, i32 addrspace(1)* %arrayidx.31, align 4, !tbaa !7
+  %144 = load i32, i32* %arrayidx1.32, align 4, !tbaa !7
+  %mul.32 = sub nsw i32 0, %144
+  store i32 %mul.32, i32 addrspace(1)* %arrayidx.32, align 4, !tbaa !7
+  %145 = load i32, i32* %arrayidx1.33, align 4, !tbaa !7
+  %mul.33 = sub nsw i32 0, %145
+  store i32 %mul.33, i32 addrspace(1)* %arrayidx.33, align 4, !tbaa !7
+  %146 = load i32, i32* %arrayidx1.34, align 4, !tbaa !7
+  %mul.34 = sub nsw i32 0, %146
+  store i32 %mul.34, i32 addrspace(1)* %arrayidx.34, align 4, !tbaa !7
+  %147 = load i32, i32* %arrayidx1.35, align 4, !tbaa !7
+  %mul.35 = sub nsw i32 0, %147
+  store i32 %mul.35, i32 addrspace(1)* %arrayidx.35, align 4, !tbaa !7
+  %148 = load i32, i32* %arrayidx1.36, align 4, !tbaa !7
+  %mul.36 = sub nsw i32 0, %148
+  store i32 %mul.36, i32 addrspace(1)* %arrayidx.36, align 4, !tbaa !7
+  %149 = load i32, i32* %arrayidx1.37, align 4, !tbaa !7
+  %mul.37 = sub nsw i32 0, %149
+  store i32 %mul.37, i32 addrspace(1)* %arrayidx.37, align 4, !tbaa !7
+  %150 = load i32, i32* %arrayidx1.38, align 4, !tbaa !7
+  %mul.38 = sub nsw i32 0, %150
+  store i32 %mul.38, i32 addrspace(1)* %arrayidx.38, align 4, !tbaa !7
+  %151 = load i32, i32* %arrayidx1.39, align 4, !tbaa !7
+  %mul.39 = sub nsw i32 0, %151
+  store i32 %mul.39, i32 addrspace(1)* %arrayidx.39, align 4, !tbaa !7
+  %152 = load i32, i32* %arrayidx1.40, align 4, !tbaa !7
+  %mul.40 = sub nsw i32 0, %152
+  store i32 %mul.40, i32 addrspace(1)* %arrayidx.40, align 4, !tbaa !7
+  %153 = load i32, i32* %arrayidx1.41, align 4, !tbaa !7
+  %mul.41 = sub nsw i32 0, %153
+  store i32 %mul.41, i32 addrspace(1)* %arrayidx.41, align 4, !tbaa !7
+  %154 = load i32, i32* %arrayidx1.42, align 4, !tbaa !7
+  %mul.42 = sub nsw i32 0, %154
+  store i32 %mul.42, i32 addrspace(1)* %arrayidx.42, align 4, !tbaa !7
+  %155 = load i32, i32* %arrayidx1.43, align 4, !tbaa !7
+  %mul.43 = sub nsw i32 0, %155
+  store i32 %mul.43, i32 addrspace(1)* %arrayidx.43, align 4, !tbaa !7
+  %156 = load i32, i32* %arrayidx1.44, align 4, !tbaa !7
+  %mul.44 = sub nsw i32 0, %156
+  store i32 %mul.44, i32 addrspace(1)* %arrayidx.44, align 4, !tbaa !7
+  %157 = load i32, i32* %arrayidx1.45, align 4, !tbaa !7
+  %mul.45 = sub nsw i32 0, %157
+  store i32 %mul.45, i32 addrspace(1)* %arrayidx.45, align 4, !tbaa !7
+  %158 = load i32, i32* %arrayidx1.46, align 4, !tbaa !7
+  %mul.46 = sub nsw i32 0, %158
+  store i32 %mul.46, i32 addrspace(1)* %arrayidx.46, align 4, !tbaa !7
+  %159 = load i32, i32* %arrayidx1.47, align 4, !tbaa !7
+  %mul.47 = sub nsw i32 0, %159
+  store i32 %mul.47, i32 addrspace(1)* %arrayidx.47, align 4, !tbaa !7
+  %160 = load i32, i32* %arrayidx1.48, align 4, !tbaa !7
+  %mul.48 = sub nsw i32 0, %160
+  store i32 %mul.48, i32 addrspace(1)* %arrayidx.48, align 4, !tbaa !7
+  %161 = load i32, i32* %arrayidx1.49, align 4, !tbaa !7
+  %mul.49 = sub nsw i32 0, %161
+  store i32 %mul.49, i32 addrspace(1)* %arrayidx.49, align 4, !tbaa !7
+  %162 = load i32, i32* %arrayidx1.50, align 4, !tbaa !7
+  %mul.50 = sub nsw i32 0, %162
+  store i32 %mul.50, i32 addrspace(1)* %arrayidx.50, align 4, !tbaa !7
+  %163 = load i32, i32* %arrayidx1.51, align 4, !tbaa !7
+  %mul.51 = sub nsw i32 0, %163
+  store i32 %mul.51, i32 addrspace(1)* %arrayidx.51, align 4, !tbaa !7
+  %164 = load i32, i32* %arrayidx1.52, align 4, !tbaa !7
+  %mul.52 = sub nsw i32 0, %164
+  store i32 %mul.52, i32 addrspace(1)* %arrayidx.52, align 4, !tbaa !7
+  %165 = load i32, i32* %arrayidx1.53, align 4, !tbaa !7
+  %mul.53 = sub nsw i32 0, %165
+  store i32 %mul.53, i32 addrspace(1)* %arrayidx.53, align 4, !tbaa !7
+  %166 = load i32, i32* %arrayidx1.54, align 4, !tbaa !7
+  %mul.54 = sub nsw i32 0, %166
+  store i32 %mul.54, i32 addrspace(1)* %arrayidx.54, align 4, !tbaa !7
+  %167 = load i32, i32* %arrayidx1.55, align 4, !tbaa !7
+  %mul.55 = sub nsw i32 0, %167
+  store i32 %mul.55, i32 addrspace(1)* %arrayidx.55, align 4, !tbaa !7
+  %168 = load i32, i32* %arrayidx1.56, align 4, !tbaa !7
+  %mul.56 = sub nsw i32 0, %168
+  store i32 %mul.56, i32 addrspace(1)* %arrayidx.56, align 4, !tbaa !7
+  %169 = load i32, i32* %arrayidx1.57, align 4, !tbaa !7
+  %mul.57 = sub nsw i32 0, %169
+  store i32 %mul.57, i32 addrspace(1)* %arrayidx.57, align 4, !tbaa !7
+  %170 = load i32, i32* %arrayidx1.58, align 4, !tbaa !7
+  %mul.58 = sub nsw i32 0, %170
+  store i32 %mul.58, i32 addrspace(1)* %arrayidx.58, align 4, !tbaa !7
+  %171 = load i32, i32* %arrayidx1.59, align 4, !tbaa !7
+  %mul.59 = sub nsw i32 0, %171
+  store i32 %mul.59, i32 addrspace(1)* %arrayidx.59, align 4, !tbaa !7
+  %172 = load i32, i32* %arrayidx1.60, align 4, !tbaa !7
+  %mul.60 = sub nsw i32 0, %172
+  store i32 %mul.60, i32 addrspace(1)* %arrayidx.60, align 4, !tbaa !7
+  %173 = load i32, i32* %arrayidx1.61, align 4, !tbaa !7
+  %mul.61 = sub nsw i32 0, %173
+  store i32 %mul.61, i32 addrspace(1)* %arrayidx.61, align 4, !tbaa !7
+  %174 = load i32, i32* %arrayidx1.62, align 4, !tbaa !7
+  %mul.62 = sub nsw i32 0, %174
+  store i32 %mul.62, i32 addrspace(1)* %arrayidx.62, align 4, !tbaa !7
+  %175 = load i32, i32* %arrayidx1.63, align 4, !tbaa !7
+  %mul.63 = sub nsw i32 0, %175
+  store i32 %mul.63, i32 addrspace(1)* %arrayidx.63, align 4, !tbaa !7
+  %176 = load i32, i32* %arrayidx1.64, align 4, !tbaa !7
+  %mul.64 = sub nsw i32 0, %176
+  store i32 %mul.64, i32 addrspace(1)* %arrayidx.64, align 4, !tbaa !7
+  %177 = load i32, i32* %arrayidx1.65, align 4, !tbaa !7
+  %mul.65 = sub nsw i32 0, %177
+  store i32 %mul.65, i32 addrspace(1)* %arrayidx.65, align 4, !tbaa !7
+  %178 = load i32, i32* %arrayidx1.66, align 4, !tbaa !7
+  %mul.66 = sub nsw i32 0, %178
+  store i32 %mul.66, i32 addrspace(1)* %arrayidx.66, align 4, !tbaa !7
+  %179 = load i32, i32* %arrayidx1.67, align 4, !tbaa !7
+  %mul.67 = sub nsw i32 0, %179
+  store i32 %mul.67, i32 addrspace(1)* %arrayidx.67, align 4, !tbaa !7
+  %180 = load i32, i32* %arrayidx1.68, align 4, !tbaa !7
+  %mul.68 = sub nsw i32 0, %180
+  store i32 %mul.68, i32 addrspace(1)* %arrayidx.68, align 4, !tbaa !7
+  %181 = load i32, i32* %arrayidx1.69, align 4, !tbaa !7
+  %mul.69 = sub nsw i32 0, %181
+  store i32 %mul.69, i32 addrspace(1)* %arrayidx.69, align 4, !tbaa !7
+  %182 = load i32, i32* %arrayidx1.70, align 4, !tbaa !7
+  %mul.70 = sub nsw i32 0, %182
+  store i32 %mul.70, i32 addrspace(1)* %arrayidx.70, align 4, !tbaa !7
+  %183 = load i32, i32* %arrayidx1.71, align 4, !tbaa !7
+  %mul.71 = sub nsw i32 0, %183
+  store i32 %mul.71, i32 addrspace(1)* %arrayidx.71, align 4, !tbaa !7
+  %184 = load i32, i32* %arrayidx1.72, align 4, !tbaa !7
+  %mul.72 = sub nsw i32 0, %184
+  store i32 %mul.72, i32 addrspace(1)* %arrayidx.72, align 4, !tbaa !7
+  %185 = load i32, i32* %arrayidx1.73, align 4, !tbaa !7
+  %mul.73 = sub nsw i32 0, %185
+  store i32 %mul.73, i32 addrspace(1)* %arrayidx.73, align 4, !tbaa !7
+  %186 = load i32, i32* %arrayidx1.74, align 4, !tbaa !7
+  %mul.74 = sub nsw i32 0, %186
+  store i32 %mul.74, i32 addrspace(1)* %arrayidx.74, align 4, !tbaa !7
+  %187 = load i32, i32* %arrayidx1.75, align 4, !tbaa !7
+  %mul.75 = sub nsw i32 0, %187
+  store i32 %mul.75, i32 addrspace(1)* %arrayidx.75, align 4, !tbaa !7
+  %188 = load i32, i32* %arrayidx1.76, align 4, !tbaa !7
+  %mul.76 = sub nsw i32 0, %188
+  store i32 %mul.76, i32 addrspace(1)* %arrayidx.76, align 4, !tbaa !7
+  %189 = load i32, i32* %arrayidx1.77, align 4, !tbaa !7
+  %mul.77 = sub nsw i32 0, %189
+  store i32 %mul.77, i32 addrspace(1)* %arrayidx.77, align 4, !tbaa !7
+  %190 = load i32, i32* %arrayidx1.78, align 4, !tbaa !7
+  %mul.78 = sub nsw i32 0, %190
+  store i32 %mul.78, i32 addrspace(1)* %arrayidx.78, align 4, !tbaa !7
+  %191 = load i32, i32* %arrayidx1.79, align 4, !tbaa !7
+  %mul.79 = sub nsw i32 0, %191
+  store i32 %mul.79, i32 addrspace(1)* %arrayidx.79, align 4, !tbaa !7
+  %192 = load i32, i32* %arrayidx1.80, align 4, !tbaa !7
+  %mul.80 = sub nsw i32 0, %192
+  store i32 %mul.80, i32 addrspace(1)* %arrayidx.80, align 4, !tbaa !7
+  %193 = load i32, i32* %arrayidx1.81, align 4, !tbaa !7
+  %mul.81 = sub nsw i32 0, %193
+  store i32 %mul.81, i32 addrspace(1)* %arrayidx.81, align 4, !tbaa !7
+  %194 = load i32, i32* %arrayidx1.82, align 4, !tbaa !7
+  %mul.82 = sub nsw i32 0, %194
+  store i32 %mul.82, i32 addrspace(1)* %arrayidx.82, align 4, !tbaa !7
+  %195 = load i32, i32* %arrayidx1.83, align 4, !tbaa !7
+  %mul.83 = sub nsw i32 0, %195
+  store i32 %mul.83, i32 addrspace(1)* %arrayidx.83, align 4, !tbaa !7
+  %196 = load i32, i32* %arrayidx1.84, align 4, !tbaa !7
+  %mul.84 = sub nsw i32 0, %196
+  store i32 %mul.84, i32 addrspace(1)* %arrayidx.84, align 4, !tbaa !7
+  %197 = load i32, i32* %arrayidx1.85, align 4, !tbaa !7
+  %mul.85 = sub nsw i32 0, %197
+  store i32 %mul.85, i32 addrspace(1)* %arrayidx.85, align 4, !tbaa !7
+  %198 = load i32, i32* %arrayidx1.86, align 4, !tbaa !7
+  %mul.86 = sub nsw i32 0, %198
+  store i32 %mul.86, i32 addrspace(1)* %arrayidx.86, align 4, !tbaa !7
+  %199 = load i32, i32* %arrayidx1.87, align 4, !tbaa !7
+  %mul.87 = sub nsw i32 0, %199
+  store i32 %mul.87, i32 addrspace(1)* %arrayidx.87, align 4, !tbaa !7
+  %200 = load i32, i32* %arrayidx1.88, align 4, !tbaa !7
+  %mul.88 = sub nsw i32 0, %200
+  store i32 %mul.88, i32 addrspace(1)* %arrayidx.88, align 4, !tbaa !7
+  %201 = load i32, i32* %arrayidx1.89, align 4, !tbaa !7
+  %mul.89 = sub nsw i32 0, %201
+  store i32 %mul.89, i32 addrspace(1)* %arrayidx.89, align 4, !tbaa !7
+  %202 = load i32, i32* %arrayidx1.90, align 4, !tbaa !7
+  %mul.90 = sub nsw i32 0, %202
+  store i32 %mul.90, i32 addrspace(1)* %arrayidx.90, align 4, !tbaa !7
+  %203 = load i32, i32* %arrayidx1.91, align 4, !tbaa !7
+  %mul.91 = sub nsw i32 0, %203
+  store i32 %mul.91, i32 addrspace(1)* %arrayidx.91, align 4, !tbaa !7
+  %204 = load i32, i32* %arrayidx1.92, align 4, !tbaa !7
+  %mul.92 = sub nsw i32 0, %204
+  store i32 %mul.92, i32 addrspace(1)* %arrayidx.92, align 4, !tbaa !7
+  %205 = load i32, i32* %arrayidx1.93, align 4, !tbaa !7
+  %mul.93 = sub nsw i32 0, %205
+  store i32 %mul.93, i32 addrspace(1)* %arrayidx.93, align 4, !tbaa !7
+  %206 = load i32, i32* %arrayidx1.94, align 4, !tbaa !7
+  %mul.94 = sub nsw i32 0, %206
+  store i32 %mul.94, i32 addrspace(1)* %arrayidx.94, align 4, !tbaa !7
+  %207 = load i32, i32* %arrayidx1.95, align 4, !tbaa !7
+  %mul.95 = sub nsw i32 0, %207
+  store i32 %mul.95, i32 addrspace(1)* %arrayidx.95, align 4, !tbaa !7
+  %208 = load i32, i32* %arrayidx1.96, align 4, !tbaa !7
+  %mul.96 = sub nsw i32 0, %208
+  store i32 %mul.96, i32 addrspace(1)* %arrayidx.96, align 4, !tbaa !7
+  %209 = load i32, i32* %arrayidx1.97, align 4, !tbaa !7
+  %mul.97 = sub nsw i32 0, %209
+  store i32 %mul.97, i32 addrspace(1)* %arrayidx.97, align 4, !tbaa !7
+  %210 = load i32, i32* %arrayidx1.98, align 4, !tbaa !7
+  %mul.98 = sub nsw i32 0, %210
+  store i32 %mul.98, i32 addrspace(1)* %arrayidx.98, align 4, !tbaa !7
+  %211 = load i32, i32* %arrayidx1.99, align 4, !tbaa !7
+  %mul.99 = sub nsw i32 0, %211
+  store i32 %mul.99, i32 addrspace(1)* %arrayidx.99, align 4, !tbaa !7
+  %212 = load i32, i32* %arrayidx1.100, align 4, !tbaa !7
+  %mul.100 = sub nsw i32 0, %212
+  store i32 %mul.100, i32 addrspace(1)* %arrayidx.100, align 4, !tbaa !7
+  %213 = load i32, i32* %arrayidx1.101, align 4, !tbaa !7
+  %mul.101 = sub nsw i32 0, %213
+  store i32 %mul.101, i32 addrspace(1)* %arrayidx.101, align 4, !tbaa !7
+  %214 = load i32, i32* %arrayidx1.102, align 4, !tbaa !7
+  %mul.102 = sub nsw i32 0, %214
+  store i32 %mul.102, i32 addrspace(1)* %arrayidx.102, align 4, !tbaa !7
+  %215 = load i32, i32* %arrayidx1.103, align 4, !tbaa !7
+  %mul.103 = sub nsw i32 0, %215
+  store i32 %mul.103, i32 addrspace(1)* %arrayidx.103, align 4, !tbaa !7
+  %216 = load i32, i32* %arrayidx1.104, align 4, !tbaa !7
+  %mul.104 = sub nsw i32 0, %216
+  store i32 %mul.104, i32 addrspace(1)* %arrayidx.104, align 4, !tbaa !7
+  %217 = load i32, i32* %arrayidx1.105, align 4, !tbaa !7
+  %mul.105 = sub nsw i32 0, %217
+  store i32 %mul.105, i32 addrspace(1)* %arrayidx.105, align 4, !tbaa !7
+  %218 = load i32, i32* %arrayidx1.106, align 4, !tbaa !7
+  %mul.106 = sub nsw i32 0, %218
+  store i32 %mul.106, i32 addrspace(1)* %arrayidx.106, align 4, !tbaa !7
+  %219 = load i32, i32* %arrayidx1.107, align 4, !tbaa !7
+  %mul.107 = sub nsw i32 0, %219
+  store i32 %mul.107, i32 addrspace(1)* %arrayidx.107, align 4, !tbaa !7
+  %220 = load i32, i32* %arrayidx1.108, align 4, !tbaa !7
+  %mul.108 = sub nsw i32 0, %220
+  store i32 %mul.108, i32 addrspace(1)* %arrayidx.108, align 4, !tbaa !7
+  %221 = load i32, i32* %arrayidx1.109, align 4, !tbaa !7
+  %mul.109 = sub nsw i32 0, %221
+  store i32 %mul.109, i32 addrspace(1)* %arrayidx.109, align 4, !tbaa !7
+  %222 = load i32, i32* %arrayidx1.110, align 4, !tbaa !7
+  %mul.110 = sub nsw i32 0, %222
+  store i32 %mul.110, i32 addrspace(1)* %arrayidx.110, align 4, !tbaa !7
+  call void @llvm.lifetime.end(i64 444, i8* %0) #2
+  ret void
+}
+
+; Function Attrs: argmemonly nounwind
+declare void @llvm.lifetime.start(i64, i8* nocapture) #1
+
+; Function Attrs: argmemonly nounwind
+declare void @llvm.lifetime.end(i64, i8* nocapture) #1
+
+attributes #0 = { nounwind "amdgpu-flat-work-group-size"="128,256" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="fiji" "target-features"="+16-bit-insts,+s-memrealtime" "unsafe-fp-math"="false" "use-soft-float"="false" }
+attributes #1 = { argmemonly nounwind }
+attributes #2 = { nounwind }
+
+!opencl.kernels = !{!0}
+!llvm.ident = !{!6}
+
+!0 = !{void (i32 addrspace(1)*)* @many_regs_128_256, !1, !2, !3, !4, !5}
+!1 = !{!"kernel_arg_addr_space", i32 1}
+!2 = !{!"kernel_arg_access_qual", !"none"}
+!3 = !{!"kernel_arg_type", !"int*"}
+!4 = !{!"kernel_arg_base_type", !"int*"}
+!5 = !{!"kernel_arg_type_qual", !""}
+!6 = !{!"clang version 3.9.0 (trunk 272684)"}
+!7 = !{!8, !8, i64 0}
+!8 = !{!"int", !9, i64 0}
+!9 = !{!"omnipotent char", !10, i64 0}
+!10 = !{!"Simple C/C++ TBAA"}

test/CodeGen/AMDGPU/attr-amdgpu-flat-work-group-size.ll

+; RUN: llc -march=amdgcn -mcpu=fiji -verify-machineinstrs < %s | FileCheck %s
+
+; CHECK: SGPRBlocks: 0
+; CHECK: VGPRBlocks: 0
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 1
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 1
+define void @empty_64_64() #0 {
+entry:
+  ret void
+}
+attributes #0 = {"amdgpu-flat-work-group-size"="64,64"}
+
+; CHECK: SGPRBlocks: 0
+; CHECK: VGPRBlocks: 0
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 1
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 1
+define void @empty_64_128() #1 {
+entry:
+  ret void
+}
+attributes #1 = {"amdgpu-flat-work-group-size"="64,128"}
+
+; CHECK: SGPRBlocks: 10
+; CHECK: VGPRBlocks: 7
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 81
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 29
+define void @empty_128_128() #2 {
+entry:
+  ret void
+}
+attributes #2 = {"amdgpu-flat-work-group-size"="128,128"}

test/CodeGen/AMDGPU/attr-amdgpu-num-active-waves-per-eu-invalid.ll

+; RUN: not llc -march=amdgcn -mcpu=fiji -verify-machineinstrs < %s 2>&1 | FileCheck %s
+
+; CHECK: invalid amdgpu-num-active-waves-per-eu: minimum is greater than maximum
+define void @invalid_0() #0 {
+entry:
+  ret void
+}
+attributes #0 = {"amdgpu-num-active-waves-per-eu"="10,1"}
+
+; CHECK: invalid amdgpu-num-active-waves-per-eu: requested minimum number of active waves per execution unit is not supported by the subtarget
+define void @invalid_1() #1 {
+entry:
+  ret void
+}
+attributes #1 = {"amdgpu-num-active-waves-per-eu"="20"}
+
+; CHECK: invalid amdgpu-num-active-waves-per-eu: requested maximum number of active waves per execution unit is not supported by the subtarget
+define void @invalid_2() #2 {
+entry:
+  ret void
+}
+attributes #2 = {"amdgpu-num-active-waves-per-eu"="1,20"}
+
+; CHECK: invalid amdgpu-num-active-waves-per-eu: requested number of active waves per execution unit cannot be achieved with default or requested flat work group sizes
+define void @invalid_3() #3 {
+entry:
+  ret void
+}
+attributes #3 = {"amdgpu-num-active-waves-per-eu"="10,10" "amdgpu-flat-work-group-size"="128,128"}

test/CodeGen/AMDGPU/attr-amdgpu-num-active-waves-per-eu-many-regs.ll

+; RUN: llc -march=amdgcn -mcpu=fiji -verify-machineinstrs < %s | FileCheck %s
+
+; Exactly 1 active wave per execution unit.
+; CHECK: SGPRBlocks: 12
+; CHECK: VGPRBlocks: 32
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 97
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 129
+define void @many_regs_exactly_1(i32 addrspace(1)* nocapture %A) #0 {
+entry:
+  %a = alloca [111 x i32], align 4
+  %0 = bitcast [111 x i32]* %a to i8*
+  call void @llvm.lifetime.start(i64 444, i8* %0) #2
+  %1 = load i32, i32 addrspace(1)* %A, align 4, !tbaa !7
+  %arrayidx1 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 0
+  store i32 %1, i32* %arrayidx1, align 4, !tbaa !7
+  %arrayidx.1 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 1
+  %2 = load i32, i32 addrspace(1)* %arrayidx.1, align 4, !tbaa !7
+  %arrayidx1.1 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 1
+  store i32 %2, i32* %arrayidx1.1, align 4, !tbaa !7
+  %arrayidx.2 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 2
+  %3 = load i32, i32 addrspace(1)* %arrayidx.2, align 4, !tbaa !7
+  %arrayidx1.2 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 2
+  store i32 %3, i32* %arrayidx1.2, align 4, !tbaa !7
+  %arrayidx.3 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 3
+  %4 = load i32, i32 addrspace(1)* %arrayidx.3, align 4, !tbaa !7
+  %arrayidx1.3 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 3
+  store i32 %4, i32* %arrayidx1.3, align 4, !tbaa !7
+  %arrayidx.4 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 4
+  %5 = load i32, i32 addrspace(1)* %arrayidx.4, align 4, !tbaa !7
+  %arrayidx1.4 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 4
+  store i32 %5, i32* %arrayidx1.4, align 4, !tbaa !7
+  %arrayidx.5 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 5
+  %6 = load i32, i32 addrspace(1)* %arrayidx.5, align 4, !tbaa !7
+  %arrayidx1.5 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 5
+  store i32 %6, i32* %arrayidx1.5, align 4, !tbaa !7
+  %arrayidx.6 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 6
+  %7 = load i32, i32 addrspace(1)* %arrayidx.6, align 4, !tbaa !7
+  %arrayidx1.6 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 6
+  store i32 %7, i32* %arrayidx1.6, align 4, !tbaa !7
+  %arrayidx.7 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 7
+  %8 = load i32, i32 addrspace(1)* %arrayidx.7, align 4, !tbaa !7
+  %arrayidx1.7 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 7
+  store i32 %8, i32* %arrayidx1.7, align 4, !tbaa !7
+  %arrayidx.8 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 8
+  %9 = load i32, i32 addrspace(1)* %arrayidx.8, align 4, !tbaa !7
+  %arrayidx1.8 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 8
+  store i32 %9, i32* %arrayidx1.8, align 4, !tbaa !7
+  %arrayidx.9 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 9
+  %10 = load i32, i32 addrspace(1)* %arrayidx.9, align 4, !tbaa !7
+  %arrayidx1.9 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 9
+  store i32 %10, i32* %arrayidx1.9, align 4, !tbaa !7
+  %arrayidx.10 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 10
+  %11 = load i32, i32 addrspace(1)* %arrayidx.10, align 4, !tbaa !7
+  %arrayidx1.10 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 10
+  store i32 %11, i32* %arrayidx1.10, align 4, !tbaa !7
+  %arrayidx.11 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 11
+  %12 = load i32, i32 addrspace(1)* %arrayidx.11, align 4, !tbaa !7
+  %arrayidx1.11 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 11
+  store i32 %12, i32* %arrayidx1.11, align 4, !tbaa !7
+  %arrayidx.12 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 12
+  %13 = load i32, i32 addrspace(1)* %arrayidx.12, align 4, !tbaa !7
+  %arrayidx1.12 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 12
+  store i32 %13, i32* %arrayidx1.12, align 4, !tbaa !7
+  %arrayidx.13 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 13
+  %14 = load i32, i32 addrspace(1)* %arrayidx.13, align 4, !tbaa !7
+  %arrayidx1.13 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 13
+  store i32 %14, i32* %arrayidx1.13, align 4, !tbaa !7
+  %arrayidx.14 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 14
+  %15 = load i32, i32 addrspace(1)* %arrayidx.14, align 4, !tbaa !7
+  %arrayidx1.14 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 14
+  store i32 %15, i32* %arrayidx1.14, align 4, !tbaa !7
+  %arrayidx.15 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 15
+  %16 = load i32, i32 addrspace(1)* %arrayidx.15, align 4, !tbaa !7
+  %arrayidx1.15 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 15
+  store i32 %16, i32* %arrayidx1.15, align 4, !tbaa !7
+  %arrayidx.16 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 16
+  %17 = load i32, i32 addrspace(1)* %arrayidx.16, align 4, !tbaa !7
+  %arrayidx1.16 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 16
+  store i32 %17, i32* %arrayidx1.16, align 4, !tbaa !7
+  %arrayidx.17 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 17
+  %18 = load i32, i32 addrspace(1)* %arrayidx.17, align 4, !tbaa !7
+  %arrayidx1.17 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 17
+  store i32 %18, i32* %arrayidx1.17, align 4, !tbaa !7
+  %arrayidx.18 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 18
+  %19 = load i32, i32 addrspace(1)* %arrayidx.18, align 4, !tbaa !7
+  %arrayidx1.18 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 18
+  store i32 %19, i32* %arrayidx1.18, align 4, !tbaa !7
+  %arrayidx.19 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 19
+  %20 = load i32, i32 addrspace(1)* %arrayidx.19, align 4, !tbaa !7
+  %arrayidx1.19 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 19
+  store i32 %20, i32* %arrayidx1.19, align 4, !tbaa !7
+  %arrayidx.20 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 20
+  %21 = load i32, i32 addrspace(1)* %arrayidx.20, align 4, !tbaa !7
+  %arrayidx1.20 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 20
+  store i32 %21, i32* %arrayidx1.20, align 4, !tbaa !7
+  %arrayidx.21 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 21
+  %22 = load i32, i32 addrspace(1)* %arrayidx.21, align 4, !tbaa !7
+  %arrayidx1.21 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 21
+  store i32 %22, i32* %arrayidx1.21, align 4, !tbaa !7
+  %arrayidx.22 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 22
+  %23 = load i32, i32 addrspace(1)* %arrayidx.22, align 4, !tbaa !7
+  %arrayidx1.22 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 22
+  store i32 %23, i32* %arrayidx1.22, align 4, !tbaa !7
+  %arrayidx.23 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 23
+  %24 = load i32, i32 addrspace(1)* %arrayidx.23, align 4, !tbaa !7
+  %arrayidx1.23 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 23
+  store i32 %24, i32* %arrayidx1.23, align 4, !tbaa !7
+  %arrayidx.24 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 24
+  %25 = load i32, i32 addrspace(1)* %arrayidx.24, align 4, !tbaa !7
+  %arrayidx1.24 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 24
+  store i32 %25, i32* %arrayidx1.24, align 4, !tbaa !7
+  %arrayidx.25 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 25
+  %26 = load i32, i32 addrspace(1)* %arrayidx.25, align 4, !tbaa !7
+  %arrayidx1.25 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 25
+  store i32 %26, i32* %arrayidx1.25, align 4, !tbaa !7
+  %arrayidx.26 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 26
+  %27 = load i32, i32 addrspace(1)* %arrayidx.26, align 4, !tbaa !7
+  %arrayidx1.26 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 26
+  store i32 %27, i32* %arrayidx1.26, align 4, !tbaa !7
+  %arrayidx.27 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 27
+  %28 = load i32, i32 addrspace(1)* %arrayidx.27, align 4, !tbaa !7
+  %arrayidx1.27 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 27
+  store i32 %28, i32* %arrayidx1.27, align 4, !tbaa !7
+  %arrayidx.28 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 28
+  %29 = load i32, i32 addrspace(1)* %arrayidx.28, align 4, !tbaa !7
+  %arrayidx1.28 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 28
+  store i32 %29, i32* %arrayidx1.28, align 4, !tbaa !7
+  %arrayidx.29 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 29
+  %30 = load i32, i32 addrspace(1)* %arrayidx.29, align 4, !tbaa !7
+  %arrayidx1.29 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 29
+  store i32 %30, i32* %arrayidx1.29, align 4, !tbaa !7
+  %arrayidx.30 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 30
+  %31 = load i32, i32 addrspace(1)* %arrayidx.30, align 4, !tbaa !7
+  %arrayidx1.30 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 30
+  store i32 %31, i32* %arrayidx1.30, align 4, !tbaa !7
+  %arrayidx.31 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 31
+  %32 = load i32, i32 addrspace(1)* %arrayidx.31, align 4, !tbaa !7
+  %arrayidx1.31 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 31
+  store i32 %32, i32* %arrayidx1.31, align 4, !tbaa !7
+  %arrayidx.32 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 32
+  %33 = load i32, i32 addrspace(1)* %arrayidx.32, align 4, !tbaa !7
+  %arrayidx1.32 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 32
+  store i32 %33, i32* %arrayidx1.32, align 4, !tbaa !7
+  %arrayidx.33 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 33
+  %34 = load i32, i32 addrspace(1)* %arrayidx.33, align 4, !tbaa !7
+  %arrayidx1.33 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 33
+  store i32 %34, i32* %arrayidx1.33, align 4, !tbaa !7
+  %arrayidx.34 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 34
+  %35 = load i32, i32 addrspace(1)* %arrayidx.34, align 4, !tbaa !7
+  %arrayidx1.34 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 34
+  store i32 %35, i32* %arrayidx1.34, align 4, !tbaa !7
+  %arrayidx.35 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 35
+  %36 = load i32, i32 addrspace(1)* %arrayidx.35, align 4, !tbaa !7
+  %arrayidx1.35 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 35
+  store i32 %36, i32* %arrayidx1.35, align 4, !tbaa !7
+  %arrayidx.36 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 36
+  %37 = load i32, i32 addrspace(1)* %arrayidx.36, align 4, !tbaa !7
+  %arrayidx1.36 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 36
+  store i32 %37, i32* %arrayidx1.36, align 4, !tbaa !7
+  %arrayidx.37 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 37
+  %38 = load i32, i32 addrspace(1)* %arrayidx.37, align 4, !tbaa !7
+  %arrayidx1.37 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 37
+  store i32 %38, i32* %arrayidx1.37, align 4, !tbaa !7
+  %arrayidx.38 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 38
+  %39 = load i32, i32 addrspace(1)* %arrayidx.38, align 4, !tbaa !7
+  %arrayidx1.38 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 38
+  store i32 %39, i32* %arrayidx1.38, align 4, !tbaa !7
+  %arrayidx.39 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 39
+  %40 = load i32, i32 addrspace(1)* %arrayidx.39, align 4, !tbaa !7
+  %arrayidx1.39 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 39
+  store i32 %40, i32* %arrayidx1.39, align 4, !tbaa !7
+  %arrayidx.40 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 40
+  %41 = load i32, i32 addrspace(1)* %arrayidx.40, align 4, !tbaa !7
+  %arrayidx1.40 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 40
+  store i32 %41, i32* %arrayidx1.40, align 4, !tbaa !7
+  %arrayidx.41 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 41
+  %42 = load i32, i32 addrspace(1)* %arrayidx.41, align 4, !tbaa !7
+  %arrayidx1.41 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 41
+  store i32 %42, i32* %arrayidx1.41, align 4, !tbaa !7
+  %arrayidx.42 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 42
+  %43 = load i32, i32 addrspace(1)* %arrayidx.42, align 4, !tbaa !7
+  %arrayidx1.42 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 42
+  store i32 %43, i32* %arrayidx1.42, align 4, !tbaa !7
+  %arrayidx.43 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 43
+  %44 = load i32, i32 addrspace(1)* %arrayidx.43, align 4, !tbaa !7
+  %arrayidx1.43 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 43
+  store i32 %44, i32* %arrayidx1.43, align 4, !tbaa !7
+  %arrayidx.44 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 44
+  %45 = load i32, i32 addrspace(1)* %arrayidx.44, align 4, !tbaa !7
+  %arrayidx1.44 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 44
+  store i32 %45, i32* %arrayidx1.44, align 4, !tbaa !7
+  %arrayidx.45 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 45
+  %46 = load i32, i32 addrspace(1)* %arrayidx.45, align 4, !tbaa !7
+  %arrayidx1.45 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 45
+  store i32 %46, i32* %arrayidx1.45, align 4, !tbaa !7
+  %arrayidx.46 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 46
+  %47 = load i32, i32 addrspace(1)* %arrayidx.46, align 4, !tbaa !7
+  %arrayidx1.46 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 46
+  store i32 %47, i32* %arrayidx1.46, align 4, !tbaa !7
+  %arrayidx.47 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 47
+  %48 = load i32, i32 addrspace(1)* %arrayidx.47, align 4, !tbaa !7
+  %arrayidx1.47 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 47
+  store i32 %48, i32* %arrayidx1.47, align 4, !tbaa !7
+  %arrayidx.48 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 48
+  %49 = load i32, i32 addrspace(1)* %arrayidx.48, align 4, !tbaa !7
+  %arrayidx1.48 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 48
+  store i32 %49, i32* %arrayidx1.48, align 4, !tbaa !7
+  %arrayidx.49 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 49
+  %50 = load i32, i32 addrspace(1)* %arrayidx.49, align 4, !tbaa !7
+  %arrayidx1.49 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 49
+  store i32 %50, i32* %arrayidx1.49, align 4, !tbaa !7
+  %arrayidx.50 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 50
+  %51 = load i32, i32 addrspace(1)* %arrayidx.50, align 4, !tbaa !7
+  %arrayidx1.50 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 50
+  store i32 %51, i32* %arrayidx1.50, align 4, !tbaa !7
+  %arrayidx.51 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 51
+  %52 = load i32, i32 addrspace(1)* %arrayidx.51, align 4, !tbaa !7
+  %arrayidx1.51 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 51
+  store i32 %52, i32* %arrayidx1.51, align 4, !tbaa !7
+  %arrayidx.52 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 52
+  %53 = load i32, i32 addrspace(1)* %arrayidx.52, align 4, !tbaa !7
+  %arrayidx1.52 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 52
+  store i32 %53, i32* %arrayidx1.52, align 4, !tbaa !7
+  %arrayidx.53 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 53
+  %54 = load i32, i32 addrspace(1)* %arrayidx.53, align 4, !tbaa !7
+  %arrayidx1.53 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 53
+  store i32 %54, i32* %arrayidx1.53, align 4, !tbaa !7
+  %arrayidx.54 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 54
+  %55 = load i32, i32 addrspace(1)* %arrayidx.54, align 4, !tbaa !7
+  %arrayidx1.54 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 54
+  store i32 %55, i32* %arrayidx1.54, align 4, !tbaa !7
+  %arrayidx.55 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 55
+  %56 = load i32, i32 addrspace(1)* %arrayidx.55, align 4, !tbaa !7
+  %arrayidx1.55 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 55
+  store i32 %56, i32* %arrayidx1.55, align 4, !tbaa !7
+  %arrayidx.56 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 56
+  %57 = load i32, i32 addrspace(1)* %arrayidx.56, align 4, !tbaa !7
+  %arrayidx1.56 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 56
+  store i32 %57, i32* %arrayidx1.56, align 4, !tbaa !7
+  %arrayidx.57 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 57
+  %58 = load i32, i32 addrspace(1)* %arrayidx.57, align 4, !tbaa !7
+  %arrayidx1.57 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 57
+  store i32 %58, i32* %arrayidx1.57, align 4, !tbaa !7
+  %arrayidx.58 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 58
+  %59 = load i32, i32 addrspace(1)* %arrayidx.58, align 4, !tbaa !7
+  %arrayidx1.58 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 58
+  store i32 %59, i32* %arrayidx1.58, align 4, !tbaa !7
+  %arrayidx.59 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 59
+  %60 = load i32, i32 addrspace(1)* %arrayidx.59, align 4, !tbaa !7
+  %arrayidx1.59 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 59
+  store i32 %60, i32* %arrayidx1.59, align 4, !tbaa !7
+  %arrayidx.60 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 60
+  %61 = load i32, i32 addrspace(1)* %arrayidx.60, align 4, !tbaa !7
+  %arrayidx1.60 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 60
+  store i32 %61, i32* %arrayidx1.60, align 4, !tbaa !7
+  %arrayidx.61 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 61
+  %62 = load i32, i32 addrspace(1)* %arrayidx.61, align 4, !tbaa !7
+  %arrayidx1.61 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 61
+  store i32 %62, i32* %arrayidx1.61, align 4, !tbaa !7
+  %arrayidx.62 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 62
+  %63 = load i32, i32 addrspace(1)* %arrayidx.62, align 4, !tbaa !7
+  %arrayidx1.62 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 62
+  store i32 %63, i32* %arrayidx1.62, align 4, !tbaa !7
+  %arrayidx.63 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 63
+  %64 = load i32, i32 addrspace(1)* %arrayidx.63, align 4, !tbaa !7
+  %arrayidx1.63 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 63
+  store i32 %64, i32* %arrayidx1.63, align 4, !tbaa !7
+  %arrayidx.64 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 64
+  %65 = load i32, i32 addrspace(1)* %arrayidx.64, align 4, !tbaa !7
+  %arrayidx1.64 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 64
+  store i32 %65, i32* %arrayidx1.64, align 4, !tbaa !7
+  %arrayidx.65 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 65
+  %66 = load i32, i32 addrspace(1)* %arrayidx.65, align 4, !tbaa !7
+  %arrayidx1.65 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 65
+  store i32 %66, i32* %arrayidx1.65, align 4, !tbaa !7
+  %arrayidx.66 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 66
+  %67 = load i32, i32 addrspace(1)* %arrayidx.66, align 4, !tbaa !7
+  %arrayidx1.66 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 66
+  store i32 %67, i32* %arrayidx1.66, align 4, !tbaa !7
+  %arrayidx.67 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 67
+  %68 = load i32, i32 addrspace(1)* %arrayidx.67, align 4, !tbaa !7
+  %arrayidx1.67 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 67
+  store i32 %68, i32* %arrayidx1.67, align 4, !tbaa !7
+  %arrayidx.68 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 68
+  %69 = load i32, i32 addrspace(1)* %arrayidx.68, align 4, !tbaa !7
+  %arrayidx1.68 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 68
+  store i32 %69, i32* %arrayidx1.68, align 4, !tbaa !7
+  %arrayidx.69 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 69
+  %70 = load i32, i32 addrspace(1)* %arrayidx.69, align 4, !tbaa !7
+  %arrayidx1.69 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 69
+  store i32 %70, i32* %arrayidx1.69, align 4, !tbaa !7
+  %arrayidx.70 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 70
+  %71 = load i32, i32 addrspace(1)* %arrayidx.70, align 4, !tbaa !7
+  %arrayidx1.70 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 70
+  store i32 %71, i32* %arrayidx1.70, align 4, !tbaa !7
+  %arrayidx.71 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 71
+  %72 = load i32, i32 addrspace(1)* %arrayidx.71, align 4, !tbaa !7
+  %arrayidx1.71 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 71
+  store i32 %72, i32* %arrayidx1.71, align 4, !tbaa !7
+  %arrayidx.72 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 72
+  %73 = load i32, i32 addrspace(1)* %arrayidx.72, align 4, !tbaa !7
+  %arrayidx1.72 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 72
+  store i32 %73, i32* %arrayidx1.72, align 4, !tbaa !7
+  %arrayidx.73 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 73
+  %74 = load i32, i32 addrspace(1)* %arrayidx.73, align 4, !tbaa !7
+  %arrayidx1.73 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 73
+  store i32 %74, i32* %arrayidx1.73, align 4, !tbaa !7
+  %arrayidx.74 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 74
+  %75 = load i32, i32 addrspace(1)* %arrayidx.74, align 4, !tbaa !7
+  %arrayidx1.74 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 74
+  store i32 %75, i32* %arrayidx1.74, align 4, !tbaa !7
+  %arrayidx.75 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 75
+  %76 = load i32, i32 addrspace(1)* %arrayidx.75, align 4, !tbaa !7
+  %arrayidx1.75 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 75
+  store i32 %76, i32* %arrayidx1.75, align 4, !tbaa !7
+  %arrayidx.76 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 76
+  %77 = load i32, i32 addrspace(1)* %arrayidx.76, align 4, !tbaa !7
+  %arrayidx1.76 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 76
+  store i32 %77, i32* %arrayidx1.76, align 4, !tbaa !7
+  %arrayidx.77 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 77
+  %78 = load i32, i32 addrspace(1)* %arrayidx.77, align 4, !tbaa !7
+  %arrayidx1.77 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 77
+  store i32 %78, i32* %arrayidx1.77, align 4, !tbaa !7
+  %arrayidx.78 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 78
+  %79 = load i32, i32 addrspace(1)* %arrayidx.78, align 4, !tbaa !7
+  %arrayidx1.78 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 78
+  store i32 %79, i32* %arrayidx1.78, align 4, !tbaa !7
+  %arrayidx.79 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 79
+  %80 = load i32, i32 addrspace(1)* %arrayidx.79, align 4, !tbaa !7
+  %arrayidx1.79 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 79
+  store i32 %80, i32* %arrayidx1.79, align 4, !tbaa !7
+  %arrayidx.80 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 80
+  %81 = load i32, i32 addrspace(1)* %arrayidx.80, align 4, !tbaa !7
+  %arrayidx1.80 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 80
+  store i32 %81, i32* %arrayidx1.80, align 4, !tbaa !7
+  %arrayidx.81 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 81
+  %82 = load i32, i32 addrspace(1)* %arrayidx.81, align 4, !tbaa !7
+  %arrayidx1.81 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 81
+  store i32 %82, i32* %arrayidx1.81, align 4, !tbaa !7
+  %arrayidx.82 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 82
+  %83 = load i32, i32 addrspace(1)* %arrayidx.82, align 4, !tbaa !7
+  %arrayidx1.82 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 82
+  store i32 %83, i32* %arrayidx1.82, align 4, !tbaa !7
+  %arrayidx.83 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 83
+  %84 = load i32, i32 addrspace(1)* %arrayidx.83, align 4, !tbaa !7
+  %arrayidx1.83 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 83
+  store i32 %84, i32* %arrayidx1.83, align 4, !tbaa !7
+  %arrayidx.84 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 84
+  %85 = load i32, i32 addrspace(1)* %arrayidx.84, align 4, !tbaa !7
+  %arrayidx1.84 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 84
+  store i32 %85, i32* %arrayidx1.84, align 4, !tbaa !7
+  %arrayidx.85 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 85
+  %86 = load i32, i32 addrspace(1)* %arrayidx.85, align 4, !tbaa !7
+  %arrayidx1.85 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 85
+  store i32 %86, i32* %arrayidx1.85, align 4, !tbaa !7
+  %arrayidx.86 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 86
+  %87 = load i32, i32 addrspace(1)* %arrayidx.86, align 4, !tbaa !7
+  %arrayidx1.86 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 86
+  store i32 %87, i32* %arrayidx1.86, align 4, !tbaa !7
+  %arrayidx.87 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 87
+  %88 = load i32, i32 addrspace(1)* %arrayidx.87, align 4, !tbaa !7
+  %arrayidx1.87 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 87
+  store i32 %88, i32* %arrayidx1.87, align 4, !tbaa !7
+  %arrayidx.88 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 88
+  %89 = load i32, i32 addrspace(1)* %arrayidx.88, align 4, !tbaa !7
+  %arrayidx1.88 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 88
+  store i32 %89, i32* %arrayidx1.88, align 4, !tbaa !7
+  %arrayidx.89 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 89
+  %90 = load i32, i32 addrspace(1)* %arrayidx.89, align 4, !tbaa !7
+  %arrayidx1.89 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 89
+  store i32 %90, i32* %arrayidx1.89, align 4, !tbaa !7
+  %arrayidx.90 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 90
+  %91 = load i32, i32 addrspace(1)* %arrayidx.90, align 4, !tbaa !7
+  %arrayidx1.90 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 90
+  store i32 %91, i32* %arrayidx1.90, align 4, !tbaa !7
+  %arrayidx.91 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 91
+  %92 = load i32, i32 addrspace(1)* %arrayidx.91, align 4, !tbaa !7
+  %arrayidx1.91 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 91
+  store i32 %92, i32* %arrayidx1.91, align 4, !tbaa !7
+  %arrayidx.92 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 92
+  %93 = load i32, i32 addrspace(1)* %arrayidx.92, align 4, !tbaa !7
+  %arrayidx1.92 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 92
+  store i32 %93, i32* %arrayidx1.92, align 4, !tbaa !7
+  %arrayidx.93 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 93
+  %94 = load i32, i32 addrspace(1)* %arrayidx.93, align 4, !tbaa !7
+  %arrayidx1.93 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 93
+  store i32 %94, i32* %arrayidx1.93, align 4, !tbaa !7
+  %arrayidx.94 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 94
+  %95 = load i32, i32 addrspace(1)* %arrayidx.94, align 4, !tbaa !7
+  %arrayidx1.94 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 94
+  store i32 %95, i32* %arrayidx1.94, align 4, !tbaa !7
+  %arrayidx.95 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 95
+  %96 = load i32, i32 addrspace(1)* %arrayidx.95, align 4, !tbaa !7
+  %arrayidx1.95 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 95
+  store i32 %96, i32* %arrayidx1.95, align 4, !tbaa !7
+  %arrayidx.96 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 96
+  %97 = load i32, i32 addrspace(1)* %arrayidx.96, align 4, !tbaa !7
+  %arrayidx1.96 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 96
+  store i32 %97, i32* %arrayidx1.96, align 4, !tbaa !7
+  %arrayidx.97 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 97
+  %98 = load i32, i32 addrspace(1)* %arrayidx.97, align 4, !tbaa !7
+  %arrayidx1.97 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 97
+  store i32 %98, i32* %arrayidx1.97, align 4, !tbaa !7
+  %arrayidx.98 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 98
+  %99 = load i32, i32 addrspace(1)* %arrayidx.98, align 4, !tbaa !7
+  %arrayidx1.98 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 98
+  store i32 %99, i32* %arrayidx1.98, align 4, !tbaa !7
+  %arrayidx.99 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 99
+  %100 = load i32, i32 addrspace(1)* %arrayidx.99, align 4, !tbaa !7
+  %arrayidx1.99 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 99
+  store i32 %100, i32* %arrayidx1.99, align 4, !tbaa !7
+  %arrayidx.100 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 100
+  %101 = load i32, i32 addrspace(1)* %arrayidx.100, align 4, !tbaa !7
+  %arrayidx1.100 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 100
+  store i32 %101, i32* %arrayidx1.100, align 4, !tbaa !7
+  %arrayidx.101 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 101
+  %102 = load i32, i32 addrspace(1)* %arrayidx.101, align 4, !tbaa !7
+  %arrayidx1.101 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 101
+  store i32 %102, i32* %arrayidx1.101, align 4, !tbaa !7
+  %arrayidx.102 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 102
+  %103 = load i32, i32 addrspace(1)* %arrayidx.102, align 4, !tbaa !7
+  %arrayidx1.102 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 102
+  store i32 %103, i32* %arrayidx1.102, align 4, !tbaa !7
+  %arrayidx.103 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 103
+  %104 = load i32, i32 addrspace(1)* %arrayidx.103, align 4, !tbaa !7
+  %arrayidx1.103 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 103
+  store i32 %104, i32* %arrayidx1.103, align 4, !tbaa !7
+  %arrayidx.104 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 104
+  %105 = load i32, i32 addrspace(1)* %arrayidx.104, align 4, !tbaa !7
+  %arrayidx1.104 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 104
+  store i32 %105, i32* %arrayidx1.104, align 4, !tbaa !7
+  %arrayidx.105 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 105
+  %106 = load i32, i32 addrspace(1)* %arrayidx.105, align 4, !tbaa !7
+  %arrayidx1.105 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 105
+  store i32 %106, i32* %arrayidx1.105, align 4, !tbaa !7
+  %arrayidx.106 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 106
+  %107 = load i32, i32 addrspace(1)* %arrayidx.106, align 4, !tbaa !7
+  %arrayidx1.106 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 106
+  store i32 %107, i32* %arrayidx1.106, align 4, !tbaa !7
+  %arrayidx.107 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 107
+  %108 = load i32, i32 addrspace(1)* %arrayidx.107, align 4, !tbaa !7
+  %arrayidx1.107 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 107
+  store i32 %108, i32* %arrayidx1.107, align 4, !tbaa !7
+  %arrayidx.108 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 108
+  %109 = load i32, i32 addrspace(1)* %arrayidx.108, align 4, !tbaa !7
+  %arrayidx1.108 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 108
+  store i32 %109, i32* %arrayidx1.108, align 4, !tbaa !7
+  %arrayidx.109 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 109
+  %110 = load i32, i32 addrspace(1)* %arrayidx.109, align 4, !tbaa !7
+  %arrayidx1.109 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 109
+  store i32 %110, i32* %arrayidx1.109, align 4, !tbaa !7
+  %arrayidx.110 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 110
+  %111 = load i32, i32 addrspace(1)* %arrayidx.110, align 4, !tbaa !7
+  %arrayidx1.110 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 110
+  store i32 %111, i32* %arrayidx1.110, align 4, !tbaa !7
+  %112 = load i32, i32* %arrayidx1, align 4, !tbaa !7
+  %mul = sub nsw i32 0, %112
+  store i32 %mul, i32 addrspace(1)* %A, align 4, !tbaa !7
+  %113 = load i32, i32* %arrayidx1.1, align 4, !tbaa !7
+  %mul.1 = sub nsw i32 0, %113
+  store i32 %mul.1, i32 addrspace(1)* %arrayidx.1, align 4, !tbaa !7
+  %114 = load i32, i32* %arrayidx1.2, align 4, !tbaa !7
+  %mul.2 = sub nsw i32 0, %114
+  store i32 %mul.2, i32 addrspace(1)* %arrayidx.2, align 4, !tbaa !7
+  %115 = load i32, i32* %arrayidx1.3, align 4, !tbaa !7
+  %mul.3 = sub nsw i32 0, %115
+  store i32 %mul.3, i32 addrspace(1)* %arrayidx.3, align 4, !tbaa !7
+  %116 = load i32, i32* %arrayidx1.4, align 4, !tbaa !7
+  %mul.4 = sub nsw i32 0, %116
+  store i32 %mul.4, i32 addrspace(1)* %arrayidx.4, align 4, !tbaa !7
+  %117 = load i32, i32* %arrayidx1.5, align 4, !tbaa !7
+  %mul.5 = sub nsw i32 0, %117
+  store i32 %mul.5, i32 addrspace(1)* %arrayidx.5, align 4, !tbaa !7
+  %118 = load i32, i32* %arrayidx1.6, align 4, !tbaa !7
+  %mul.6 = sub nsw i32 0, %118
+  store i32 %mul.6, i32 addrspace(1)* %arrayidx.6, align 4, !tbaa !7
+  %119 = load i32, i32* %arrayidx1.7, align 4, !tbaa !7
+  %mul.7 = sub nsw i32 0, %119
+  store i32 %mul.7, i32 addrspace(1)* %arrayidx.7, align 4, !tbaa !7
+  %120 = load i32, i32* %arrayidx1.8, align 4, !tbaa !7
+  %mul.8 = sub nsw i32 0, %120
+  store i32 %mul.8, i32 addrspace(1)* %arrayidx.8, align 4, !tbaa !7
+  %121 = load i32, i32* %arrayidx1.9, align 4, !tbaa !7
+  %mul.9 = sub nsw i32 0, %121
+  store i32 %mul.9, i32 addrspace(1)* %arrayidx.9, align 4, !tbaa !7
+  %122 = load i32, i32* %arrayidx1.10, align 4, !tbaa !7
+  %mul.10 = sub nsw i32 0, %122
+  store i32 %mul.10, i32 addrspace(1)* %arrayidx.10, align 4, !tbaa !7
+  %123 = load i32, i32* %arrayidx1.11, align 4, !tbaa !7
+  %mul.11 = sub nsw i32 0, %123
+  store i32 %mul.11, i32 addrspace(1)* %arrayidx.11, align 4, !tbaa !7
+  %124 = load i32, i32* %arrayidx1.12, align 4, !tbaa !7
+  %mul.12 = sub nsw i32 0, %124
+  store i32 %mul.12, i32 addrspace(1)* %arrayidx.12, align 4, !tbaa !7
+  %125 = load i32, i32* %arrayidx1.13, align 4, !tbaa !7
+  %mul.13 = sub nsw i32 0, %125
+  store i32 %mul.13, i32 addrspace(1)* %arrayidx.13, align 4, !tbaa !7
+  %126 = load i32, i32* %arrayidx1.14, align 4, !tbaa !7
+  %mul.14 = sub nsw i32 0, %126
+  store i32 %mul.14, i32 addrspace(1)* %arrayidx.14, align 4, !tbaa !7
+  %127 = load i32, i32* %arrayidx1.15, align 4, !tbaa !7
+  %mul.15 = sub nsw i32 0, %127
+  store i32 %mul.15, i32 addrspace(1)* %arrayidx.15, align 4, !tbaa !7
+  %128 = load i32, i32* %arrayidx1.16, align 4, !tbaa !7
+  %mul.16 = sub nsw i32 0, %128
+  store i32 %mul.16, i32 addrspace(1)* %arrayidx.16, align 4, !tbaa !7
+  %129 = load i32, i32* %arrayidx1.17, align 4, !tbaa !7
+  %mul.17 = sub nsw i32 0, %129
+  store i32 %mul.17, i32 addrspace(1)* %arrayidx.17, align 4, !tbaa !7
+  %130 = load i32, i32* %arrayidx1.18, align 4, !tbaa !7
+  %mul.18 = sub nsw i32 0, %130
+  store i32 %mul.18, i32 addrspace(1)* %arrayidx.18, align 4, !tbaa !7
+  %131 = load i32, i32* %arrayidx1.19, align 4, !tbaa !7
+  %mul.19 = sub nsw i32 0, %131
+  store i32 %mul.19, i32 addrspace(1)* %arrayidx.19, align 4, !tbaa !7
+  %132 = load i32, i32* %arrayidx1.20, align 4, !tbaa !7
+  %mul.20 = sub nsw i32 0, %132
+  store i32 %mul.20, i32 addrspace(1)* %arrayidx.20, align 4, !tbaa !7
+  %133 = load i32, i32* %arrayidx1.21, align 4, !tbaa !7
+  %mul.21 = sub nsw i32 0, %133
+  store i32 %mul.21, i32 addrspace(1)* %arrayidx.21, align 4, !tbaa !7
+  %134 = load i32, i32* %arrayidx1.22, align 4, !tbaa !7
+  %mul.22 = sub nsw i32 0, %134
+  store i32 %mul.22, i32 addrspace(1)* %arrayidx.22, align 4, !tbaa !7
+  %135 = load i32, i32* %arrayidx1.23, align 4, !tbaa !7
+  %mul.23 = sub nsw i32 0, %135
+  store i32 %mul.23, i32 addrspace(1)* %arrayidx.23, align 4, !tbaa !7
+  %136 = load i32, i32* %arrayidx1.24, align 4, !tbaa !7
+  %mul.24 = sub nsw i32 0, %136
+  store i32 %mul.24, i32 addrspace(1)* %arrayidx.24, align 4, !tbaa !7
+  %137 = load i32, i32* %arrayidx1.25, align 4, !tbaa !7
+  %mul.25 = sub nsw i32 0, %137
+  store i32 %mul.25, i32 addrspace(1)* %arrayidx.25, align 4, !tbaa !7
+  %138 = load i32, i32* %arrayidx1.26, align 4, !tbaa !7
+  %mul.26 = sub nsw i32 0, %138
+  store i32 %mul.26, i32 addrspace(1)* %arrayidx.26, align 4, !tbaa !7
+  %139 = load i32, i32* %arrayidx1.27, align 4, !tbaa !7
+  %mul.27 = sub nsw i32 0, %139
+  store i32 %mul.27, i32 addrspace(1)* %arrayidx.27, align 4, !tbaa !7
+  %140 = load i32, i32* %arrayidx1.28, align 4, !tbaa !7
+  %mul.28 = sub nsw i32 0, %140
+  store i32 %mul.28, i32 addrspace(1)* %arrayidx.28, align 4, !tbaa !7
+  %141 = load i32, i32* %arrayidx1.29, align 4, !tbaa !7
+  %mul.29 = sub nsw i32 0, %141
+  store i32 %mul.29, i32 addrspace(1)* %arrayidx.29, align 4, !tbaa !7
+  %142 = load i32, i32* %arrayidx1.30, align 4, !tbaa !7
+  %mul.30 = sub nsw i32 0, %142
+  store i32 %mul.30, i32 addrspace(1)* %arrayidx.30, align 4, !tbaa !7
+  %143 = load i32, i32* %arrayidx1.31, align 4, !tbaa !7
+  %mul.31 = sub nsw i32 0, %143
+  store i32 %mul.31, i32 addrspace(1)* %arrayidx.31, align 4, !tbaa !7
+  %144 = load i32, i32* %arrayidx1.32, align 4, !tbaa !7
+  %mul.32 = sub nsw i32 0, %144
+  store i32 %mul.32, i32 addrspace(1)* %arrayidx.32, align 4, !tbaa !7
+  %145 = load i32, i32* %arrayidx1.33, align 4, !tbaa !7
+  %mul.33 = sub nsw i32 0, %145
+  store i32 %mul.33, i32 addrspace(1)* %arrayidx.33, align 4, !tbaa !7
+  %146 = load i32, i32* %arrayidx1.34, align 4, !tbaa !7
+  %mul.34 = sub nsw i32 0, %146
+  store i32 %mul.34, i32 addrspace(1)* %arrayidx.34, align 4, !tbaa !7
+  %147 = load i32, i32* %arrayidx1.35, align 4, !tbaa !7
+  %mul.35 = sub nsw i32 0, %147
+  store i32 %mul.35, i32 addrspace(1)* %arrayidx.35, align 4, !tbaa !7
+  %148 = load i32, i32* %arrayidx1.36, align 4, !tbaa !7
+  %mul.36 = sub nsw i32 0, %148
+  store i32 %mul.36, i32 addrspace(1)* %arrayidx.36, align 4, !tbaa !7
+  %149 = load i32, i32* %arrayidx1.37, align 4, !tbaa !7
+  %mul.37 = sub nsw i32 0, %149
+  store i32 %mul.37, i32 addrspace(1)* %arrayidx.37, align 4, !tbaa !7
+  %150 = load i32, i32* %arrayidx1.38, align 4, !tbaa !7
+  %mul.38 = sub nsw i32 0, %150
+  store i32 %mul.38, i32 addrspace(1)* %arrayidx.38, align 4, !tbaa !7
+  %151 = load i32, i32* %arrayidx1.39, align 4, !tbaa !7
+  %mul.39 = sub nsw i32 0, %151
+  store i32 %mul.39, i32 addrspace(1)* %arrayidx.39, align 4, !tbaa !7
+  %152 = load i32, i32* %arrayidx1.40, align 4, !tbaa !7
+  %mul.40 = sub nsw i32 0, %152
+  store i32 %mul.40, i32 addrspace(1)* %arrayidx.40, align 4, !tbaa !7
+  %153 = load i32, i32* %arrayidx1.41, align 4, !tbaa !7
+  %mul.41 = sub nsw i32 0, %153
+  store i32 %mul.41, i32 addrspace(1)* %arrayidx.41, align 4, !tbaa !7
+  %154 = load i32, i32* %arrayidx1.42, align 4, !tbaa !7
+  %mul.42 = sub nsw i32 0, %154
+  store i32 %mul.42, i32 addrspace(1)* %arrayidx.42, align 4, !tbaa !7
+  %155 = load i32, i32* %arrayidx1.43, align 4, !tbaa !7
+  %mul.43 = sub nsw i32 0, %155
+  store i32 %mul.43, i32 addrspace(1)* %arrayidx.43, align 4, !tbaa !7
+  %156 = load i32, i32* %arrayidx1.44, align 4, !tbaa !7
+  %mul.44 = sub nsw i32 0, %156
+  store i32 %mul.44, i32 addrspace(1)* %arrayidx.44, align 4, !tbaa !7
+  %157 = load i32, i32* %arrayidx1.45, align 4, !tbaa !7
+  %mul.45 = sub nsw i32 0, %157
+  store i32 %mul.45, i32 addrspace(1)* %arrayidx.45, align 4, !tbaa !7
+  %158 = load i32, i32* %arrayidx1.46, align 4, !tbaa !7
+  %mul.46 = sub nsw i32 0, %158
+  store i32 %mul.46, i32 addrspace(1)* %arrayidx.46, align 4, !tbaa !7
+  %159 = load i32, i32* %arrayidx1.47, align 4, !tbaa !7
+  %mul.47 = sub nsw i32 0, %159
+  store i32 %mul.47, i32 addrspace(1)* %arrayidx.47, align 4, !tbaa !7
+  %160 = load i32, i32* %arrayidx1.48, align 4, !tbaa !7
+  %mul.48 = sub nsw i32 0, %160
+  store i32 %mul.48, i32 addrspace(1)* %arrayidx.48, align 4, !tbaa !7
+  %161 = load i32, i32* %arrayidx1.49, align 4, !tbaa !7
+  %mul.49 = sub nsw i32 0, %161
+  store i32 %mul.49, i32 addrspace(1)* %arrayidx.49, align 4, !tbaa !7
+  %162 = load i32, i32* %arrayidx1.50, align 4, !tbaa !7
+  %mul.50 = sub nsw i32 0, %162
+  store i32 %mul.50, i32 addrspace(1)* %arrayidx.50, align 4, !tbaa !7
+  %163 = load i32, i32* %arrayidx1.51, align 4, !tbaa !7
+  %mul.51 = sub nsw i32 0, %163
+  store i32 %mul.51, i32 addrspace(1)* %arrayidx.51, align 4, !tbaa !7
+  %164 = load i32, i32* %arrayidx1.52, align 4, !tbaa !7
+  %mul.52 = sub nsw i32 0, %164
+  store i32 %mul.52, i32 addrspace(1)* %arrayidx.52, align 4, !tbaa !7
+  %165 = load i32, i32* %arrayidx1.53, align 4, !tbaa !7
+  %mul.53 = sub nsw i32 0, %165
+  store i32 %mul.53, i32 addrspace(1)* %arrayidx.53, align 4, !tbaa !7
+  %166 = load i32, i32* %arrayidx1.54, align 4, !tbaa !7
+  %mul.54 = sub nsw i32 0, %166
+  store i32 %mul.54, i32 addrspace(1)* %arrayidx.54, align 4, !tbaa !7
+  %167 = load i32, i32* %arrayidx1.55, align 4, !tbaa !7
+  %mul.55 = sub nsw i32 0, %167
+  store i32 %mul.55, i32 addrspace(1)* %arrayidx.55, align 4, !tbaa !7
+  %168 = load i32, i32* %arrayidx1.56, align 4, !tbaa !7
+  %mul.56 = sub nsw i32 0, %168
+  store i32 %mul.56, i32 addrspace(1)* %arrayidx.56, align 4, !tbaa !7
+  %169 = load i32, i32* %arrayidx1.57, align 4, !tbaa !7
+  %mul.57 = sub nsw i32 0, %169
+  store i32 %mul.57, i32 addrspace(1)* %arrayidx.57, align 4, !tbaa !7
+  %170 = load i32, i32* %arrayidx1.58, align 4, !tbaa !7
+  %mul.58 = sub nsw i32 0, %170
+  store i32 %mul.58, i32 addrspace(1)* %arrayidx.58, align 4, !tbaa !7
+  %171 = load i32, i32* %arrayidx1.59, align 4, !tbaa !7
+  %mul.59 = sub nsw i32 0, %171
+  store i32 %mul.59, i32 addrspace(1)* %arrayidx.59, align 4, !tbaa !7
+  %172 = load i32, i32* %arrayidx1.60, align 4, !tbaa !7
+  %mul.60 = sub nsw i32 0, %172
+  store i32 %mul.60, i32 addrspace(1)* %arrayidx.60, align 4, !tbaa !7
+  %173 = load i32, i32* %arrayidx1.61, align 4, !tbaa !7
+  %mul.61 = sub nsw i32 0, %173
+  store i32 %mul.61, i32 addrspace(1)* %arrayidx.61, align 4, !tbaa !7
+  %174 = load i32, i32* %arrayidx1.62, align 4, !tbaa !7
+  %mul.62 = sub nsw i32 0, %174
+  store i32 %mul.62, i32 addrspace(1)* %arrayidx.62, align 4, !tbaa !7
+  %175 = load i32, i32* %arrayidx1.63, align 4, !tbaa !7
+  %mul.63 = sub nsw i32 0, %175
+  store i32 %mul.63, i32 addrspace(1)* %arrayidx.63, align 4, !tbaa !7
+  %176 = load i32, i32* %arrayidx1.64, align 4, !tbaa !7
+  %mul.64 = sub nsw i32 0, %176
+  store i32 %mul.64, i32 addrspace(1)* %arrayidx.64, align 4, !tbaa !7
+  %177 = load i32, i32* %arrayidx1.65, align 4, !tbaa !7
+  %mul.65 = sub nsw i32 0, %177
+  store i32 %mul.65, i32 addrspace(1)* %arrayidx.65, align 4, !tbaa !7
+  %178 = load i32, i32* %arrayidx1.66, align 4, !tbaa !7
+  %mul.66 = sub nsw i32 0, %178
+  store i32 %mul.66, i32 addrspace(1)* %arrayidx.66, align 4, !tbaa !7
+  %179 = load i32, i32* %arrayidx1.67, align 4, !tbaa !7
+  %mul.67 = sub nsw i32 0, %179
+  store i32 %mul.67, i32 addrspace(1)* %arrayidx.67, align 4, !tbaa !7
+  %180 = load i32, i32* %arrayidx1.68, align 4, !tbaa !7
+  %mul.68 = sub nsw i32 0, %180
+  store i32 %mul.68, i32 addrspace(1)* %arrayidx.68, align 4, !tbaa !7
+  %181 = load i32, i32* %arrayidx1.69, align 4, !tbaa !7
+  %mul.69 = sub nsw i32 0, %181
+  store i32 %mul.69, i32 addrspace(1)* %arrayidx.69, align 4, !tbaa !7
+  %182 = load i32, i32* %arrayidx1.70, align 4, !tbaa !7
+  %mul.70 = sub nsw i32 0, %182
+  store i32 %mul.70, i32 addrspace(1)* %arrayidx.70, align 4, !tbaa !7
+  %183 = load i32, i32* %arrayidx1.71, align 4, !tbaa !7
+  %mul.71 = sub nsw i32 0, %183
+  store i32 %mul.71, i32 addrspace(1)* %arrayidx.71, align 4, !tbaa !7
+  %184 = load i32, i32* %arrayidx1.72, align 4, !tbaa !7
+  %mul.72 = sub nsw i32 0, %184
+  store i32 %mul.72, i32 addrspace(1)* %arrayidx.72, align 4, !tbaa !7
+  %185 = load i32, i32* %arrayidx1.73, align 4, !tbaa !7
+  %mul.73 = sub nsw i32 0, %185
+  store i32 %mul.73, i32 addrspace(1)* %arrayidx.73, align 4, !tbaa !7
+  %186 = load i32, i32* %arrayidx1.74, align 4, !tbaa !7
+  %mul.74 = sub nsw i32 0, %186
+  store i32 %mul.74, i32 addrspace(1)* %arrayidx.74, align 4, !tbaa !7
+  %187 = load i32, i32* %arrayidx1.75, align 4, !tbaa !7
+  %mul.75 = sub nsw i32 0, %187
+  store i32 %mul.75, i32 addrspace(1)* %arrayidx.75, align 4, !tbaa !7
+  %188 = load i32, i32* %arrayidx1.76, align 4, !tbaa !7
+  %mul.76 = sub nsw i32 0, %188
+  store i32 %mul.76, i32 addrspace(1)* %arrayidx.76, align 4, !tbaa !7
+  %189 = load i32, i32* %arrayidx1.77, align 4, !tbaa !7
+  %mul.77 = sub nsw i32 0, %189
+  store i32 %mul.77, i32 addrspace(1)* %arrayidx.77, align 4, !tbaa !7
+  %190 = load i32, i32* %arrayidx1.78, align 4, !tbaa !7
+  %mul.78 = sub nsw i32 0, %190
+  store i32 %mul.78, i32 addrspace(1)* %arrayidx.78, align 4, !tbaa !7
+  %191 = load i32, i32* %arrayidx1.79, align 4, !tbaa !7
+  %mul.79 = sub nsw i32 0, %191
+  store i32 %mul.79, i32 addrspace(1)* %arrayidx.79, align 4, !tbaa !7
+  %192 = load i32, i32* %arrayidx1.80, align 4, !tbaa !7
+  %mul.80 = sub nsw i32 0, %192
+  store i32 %mul.80, i32 addrspace(1)* %arrayidx.80, align 4, !tbaa !7
+  %193 = load i32, i32* %arrayidx1.81, align 4, !tbaa !7
+  %mul.81 = sub nsw i32 0, %193
+  store i32 %mul.81, i32 addrspace(1)* %arrayidx.81, align 4, !tbaa !7
+  %194 = load i32, i32* %arrayidx1.82, align 4, !tbaa !7
+  %mul.82 = sub nsw i32 0, %194
+  store i32 %mul.82, i32 addrspace(1)* %arrayidx.82, align 4, !tbaa !7
+  %195 = load i32, i32* %arrayidx1.83, align 4, !tbaa !7
+  %mul.83 = sub nsw i32 0, %195
+  store i32 %mul.83, i32 addrspace(1)* %arrayidx.83, align 4, !tbaa !7
+  %196 = load i32, i32* %arrayidx1.84, align 4, !tbaa !7
+  %mul.84 = sub nsw i32 0, %196
+  store i32 %mul.84, i32 addrspace(1)* %arrayidx.84, align 4, !tbaa !7
+  %197 = load i32, i32* %arrayidx1.85, align 4, !tbaa !7
+  %mul.85 = sub nsw i32 0, %197
+  store i32 %mul.85, i32 addrspace(1)* %arrayidx.85, align 4, !tbaa !7
+  %198 = load i32, i32* %arrayidx1.86, align 4, !tbaa !7
+  %mul.86 = sub nsw i32 0, %198
+  store i32 %mul.86, i32 addrspace(1)* %arrayidx.86, align 4, !tbaa !7
+  %199 = load i32, i32* %arrayidx1.87, align 4, !tbaa !7
+  %mul.87 = sub nsw i32 0, %199
+  store i32 %mul.87, i32 addrspace(1)* %arrayidx.87, align 4, !tbaa !7
+  %200 = load i32, i32* %arrayidx1.88, align 4, !tbaa !7
+  %mul.88 = sub nsw i32 0, %200
+  store i32 %mul.88, i32 addrspace(1)* %arrayidx.88, align 4, !tbaa !7
+  %201 = load i32, i32* %arrayidx1.89, align 4, !tbaa !7
+  %mul.89 = sub nsw i32 0, %201
+  store i32 %mul.89, i32 addrspace(1)* %arrayidx.89, align 4, !tbaa !7
+  %202 = load i32, i32* %arrayidx1.90, align 4, !tbaa !7
+  %mul.90 = sub nsw i32 0, %202
+  store i32 %mul.90, i32 addrspace(1)* %arrayidx.90, align 4, !tbaa !7
+  %203 = load i32, i32* %arrayidx1.91, align 4, !tbaa !7
+  %mul.91 = sub nsw i32 0, %203
+  store i32 %mul.91, i32 addrspace(1)* %arrayidx.91, align 4, !tbaa !7
+  %204 = load i32, i32* %arrayidx1.92, align 4, !tbaa !7
+  %mul.92 = sub nsw i32 0, %204
+  store i32 %mul.92, i32 addrspace(1)* %arrayidx.92, align 4, !tbaa !7
+  %205 = load i32, i32* %arrayidx1.93, align 4, !tbaa !7
+  %mul.93 = sub nsw i32 0, %205
+  store i32 %mul.93, i32 addrspace(1)* %arrayidx.93, align 4, !tbaa !7
+  %206 = load i32, i32* %arrayidx1.94, align 4, !tbaa !7
+  %mul.94 = sub nsw i32 0, %206
+  store i32 %mul.94, i32 addrspace(1)* %arrayidx.94, align 4, !tbaa !7
+  %207 = load i32, i32* %arrayidx1.95, align 4, !tbaa !7
+  %mul.95 = sub nsw i32 0, %207
+  store i32 %mul.95, i32 addrspace(1)* %arrayidx.95, align 4, !tbaa !7
+  %208 = load i32, i32* %arrayidx1.96, align 4, !tbaa !7
+  %mul.96 = sub nsw i32 0, %208
+  store i32 %mul.96, i32 addrspace(1)* %arrayidx.96, align 4, !tbaa !7
+  %209 = load i32, i32* %arrayidx1.97, align 4, !tbaa !7
+  %mul.97 = sub nsw i32 0, %209
+  store i32 %mul.97, i32 addrspace(1)* %arrayidx.97, align 4, !tbaa !7
+  %210 = load i32, i32* %arrayidx1.98, align 4, !tbaa !7
+  %mul.98 = sub nsw i32 0, %210
+  store i32 %mul.98, i32 addrspace(1)* %arrayidx.98, align 4, !tbaa !7
+  %211 = load i32, i32* %arrayidx1.99, align 4, !tbaa !7
+  %mul.99 = sub nsw i32 0, %211
+  store i32 %mul.99, i32 addrspace(1)* %arrayidx.99, align 4, !tbaa !7
+  %212 = load i32, i32* %arrayidx1.100, align 4, !tbaa !7
+  %mul.100 = sub nsw i32 0, %212
+  store i32 %mul.100, i32 addrspace(1)* %arrayidx.100, align 4, !tbaa !7
+  %213 = load i32, i32* %arrayidx1.101, align 4, !tbaa !7
+  %mul.101 = sub nsw i32 0, %213
+  store i32 %mul.101, i32 addrspace(1)* %arrayidx.101, align 4, !tbaa !7
+  %214 = load i32, i32* %arrayidx1.102, align 4, !tbaa !7
+  %mul.102 = sub nsw i32 0, %214
+  store i32 %mul.102, i32 addrspace(1)* %arrayidx.102, align 4, !tbaa !7
+  %215 = load i32, i32* %arrayidx1.103, align 4, !tbaa !7
+  %mul.103 = sub nsw i32 0, %215
+  store i32 %mul.103, i32 addrspace(1)* %arrayidx.103, align 4, !tbaa !7
+  %216 = load i32, i32* %arrayidx1.104, align 4, !tbaa !7
+  %mul.104 = sub nsw i32 0, %216
+  store i32 %mul.104, i32 addrspace(1)* %arrayidx.104, align 4, !tbaa !7
+  %217 = load i32, i32* %arrayidx1.105, align 4, !tbaa !7
+  %mul.105 = sub nsw i32 0, %217
+  store i32 %mul.105, i32 addrspace(1)* %arrayidx.105, align 4, !tbaa !7
+  %218 = load i32, i32* %arrayidx1.106, align 4, !tbaa !7
+  %mul.106 = sub nsw i32 0, %218
+  store i32 %mul.106, i32 addrspace(1)* %arrayidx.106, align 4, !tbaa !7
+  %219 = load i32, i32* %arrayidx1.107, align 4, !tbaa !7
+  %mul.107 = sub nsw i32 0, %219
+  store i32 %mul.107, i32 addrspace(1)* %arrayidx.107, align 4, !tbaa !7
+  %220 = load i32, i32* %arrayidx1.108, align 4, !tbaa !7
+  %mul.108 = sub nsw i32 0, %220
+  store i32 %mul.108, i32 addrspace(1)* %arrayidx.108, align 4, !tbaa !7
+  %221 = load i32, i32* %arrayidx1.109, align 4, !tbaa !7
+  %mul.109 = sub nsw i32 0, %221
+  store i32 %mul.109, i32 addrspace(1)* %arrayidx.109, align 4, !tbaa !7
+  %222 = load i32, i32* %arrayidx1.110, align 4, !tbaa !7
+  %mul.110 = sub nsw i32 0, %222
+  store i32 %mul.110, i32 addrspace(1)* %arrayidx.110, align 4, !tbaa !7
+  call void @llvm.lifetime.end(i64 444, i8* %0) #2
+  ret void
+}
+attributes #0 = { nounwind "amdgpu-num-active-waves-per-eu"="1,1" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="fiji" "target-features"="+16-bit-insts,+s-memrealtime" "unsafe-fp-math"="false" "use-soft-float"="false" }
+
+; Exactly 5 active waves per execution unit.
+; CHECK: SGPRBlocks: 12
+; CHECK: VGPRBlocks: 11
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 97
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 48
+define void @many_regs_exactly_5(i32 addrspace(1)* nocapture %A) #3 {
+entry:
+  %a = alloca [111 x i32], align 4
+  %0 = bitcast [111 x i32]* %a to i8*
+  call void @llvm.lifetime.start(i64 444, i8* %0) #2
+  %1 = load i32, i32 addrspace(1)* %A, align 4, !tbaa !7
+  %arrayidx1 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 0
+  store i32 %1, i32* %arrayidx1, align 4, !tbaa !7
+  %arrayidx.1 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 1
+  %2 = load i32, i32 addrspace(1)* %arrayidx.1, align 4, !tbaa !7
+  %arrayidx1.1 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 1
+  store i32 %2, i32* %arrayidx1.1, align 4, !tbaa !7
+  %arrayidx.2 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 2
+  %3 = load i32, i32 addrspace(1)* %arrayidx.2, align 4, !tbaa !7
+  %arrayidx1.2 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 2
+  store i32 %3, i32* %arrayidx1.2, align 4, !tbaa !7
+  %arrayidx.3 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 3
+  %4 = load i32, i32 addrspace(1)* %arrayidx.3, align 4, !tbaa !7
+  %arrayidx1.3 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 3
+  store i32 %4, i32* %arrayidx1.3, align 4, !tbaa !7
+  %arrayidx.4 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 4
+  %5 = load i32, i32 addrspace(1)* %arrayidx.4, align 4, !tbaa !7
+  %arrayidx1.4 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 4
+  store i32 %5, i32* %arrayidx1.4, align 4, !tbaa !7
+  %arrayidx.5 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 5
+  %6 = load i32, i32 addrspace(1)* %arrayidx.5, align 4, !tbaa !7
+  %arrayidx1.5 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 5
+  store i32 %6, i32* %arrayidx1.5, align 4, !tbaa !7
+  %arrayidx.6 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 6
+  %7 = load i32, i32 addrspace(1)* %arrayidx.6, align 4, !tbaa !7
+  %arrayidx1.6 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 6
+  store i32 %7, i32* %arrayidx1.6, align 4, !tbaa !7
+  %arrayidx.7 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 7
+  %8 = load i32, i32 addrspace(1)* %arrayidx.7, align 4, !tbaa !7
+  %arrayidx1.7 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 7
+  store i32 %8, i32* %arrayidx1.7, align 4, !tbaa !7
+  %arrayidx.8 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 8
+  %9 = load i32, i32 addrspace(1)* %arrayidx.8, align 4, !tbaa !7
+  %arrayidx1.8 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 8
+  store i32 %9, i32* %arrayidx1.8, align 4, !tbaa !7
+  %arrayidx.9 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 9
+  %10 = load i32, i32 addrspace(1)* %arrayidx.9, align 4, !tbaa !7
+  %arrayidx1.9 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 9
+  store i32 %10, i32* %arrayidx1.9, align 4, !tbaa !7
+  %arrayidx.10 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 10
+  %11 = load i32, i32 addrspace(1)* %arrayidx.10, align 4, !tbaa !7
+  %arrayidx1.10 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 10
+  store i32 %11, i32* %arrayidx1.10, align 4, !tbaa !7
+  %arrayidx.11 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 11
+  %12 = load i32, i32 addrspace(1)* %arrayidx.11, align 4, !tbaa !7
+  %arrayidx1.11 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 11
+  store i32 %12, i32* %arrayidx1.11, align 4, !tbaa !7
+  %arrayidx.12 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 12
+  %13 = load i32, i32 addrspace(1)* %arrayidx.12, align 4, !tbaa !7
+  %arrayidx1.12 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 12
+  store i32 %13, i32* %arrayidx1.12, align 4, !tbaa !7
+  %arrayidx.13 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 13
+  %14 = load i32, i32 addrspace(1)* %arrayidx.13, align 4, !tbaa !7
+  %arrayidx1.13 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 13
+  store i32 %14, i32* %arrayidx1.13, align 4, !tbaa !7
+  %arrayidx.14 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 14
+  %15 = load i32, i32 addrspace(1)* %arrayidx.14, align 4, !tbaa !7
+  %arrayidx1.14 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 14
+  store i32 %15, i32* %arrayidx1.14, align 4, !tbaa !7
+  %arrayidx.15 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 15
+  %16 = load i32, i32 addrspace(1)* %arrayidx.15, align 4, !tbaa !7
+  %arrayidx1.15 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 15
+  store i32 %16, i32* %arrayidx1.15, align 4, !tbaa !7
+  %arrayidx.16 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 16
+  %17 = load i32, i32 addrspace(1)* %arrayidx.16, align 4, !tbaa !7
+  %arrayidx1.16 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 16
+  store i32 %17, i32* %arrayidx1.16, align 4, !tbaa !7
+  %arrayidx.17 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 17
+  %18 = load i32, i32 addrspace(1)* %arrayidx.17, align 4, !tbaa !7
+  %arrayidx1.17 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 17
+  store i32 %18, i32* %arrayidx1.17, align 4, !tbaa !7
+  %arrayidx.18 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 18
+  %19 = load i32, i32 addrspace(1)* %arrayidx.18, align 4, !tbaa !7
+  %arrayidx1.18 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 18
+  store i32 %19, i32* %arrayidx1.18, align 4, !tbaa !7
+  %arrayidx.19 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 19
+  %20 = load i32, i32 addrspace(1)* %arrayidx.19, align 4, !tbaa !7
+  %arrayidx1.19 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 19
+  store i32 %20, i32* %arrayidx1.19, align 4, !tbaa !7
+  %arrayidx.20 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 20
+  %21 = load i32, i32 addrspace(1)* %arrayidx.20, align 4, !tbaa !7
+  %arrayidx1.20 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 20
+  store i32 %21, i32* %arrayidx1.20, align 4, !tbaa !7
+  %arrayidx.21 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 21
+  %22 = load i32, i32 addrspace(1)* %arrayidx.21, align 4, !tbaa !7
+  %arrayidx1.21 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 21
+  store i32 %22, i32* %arrayidx1.21, align 4, !tbaa !7
+  %arrayidx.22 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 22
+  %23 = load i32, i32 addrspace(1)* %arrayidx.22, align 4, !tbaa !7
+  %arrayidx1.22 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 22
+  store i32 %23, i32* %arrayidx1.22, align 4, !tbaa !7
+  %arrayidx.23 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 23
+  %24 = load i32, i32 addrspace(1)* %arrayidx.23, align 4, !tbaa !7
+  %arrayidx1.23 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 23
+  store i32 %24, i32* %arrayidx1.23, align 4, !tbaa !7
+  %arrayidx.24 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 24
+  %25 = load i32, i32 addrspace(1)* %arrayidx.24, align 4, !tbaa !7
+  %arrayidx1.24 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 24
+  store i32 %25, i32* %arrayidx1.24, align 4, !tbaa !7
+  %arrayidx.25 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 25
+  %26 = load i32, i32 addrspace(1)* %arrayidx.25, align 4, !tbaa !7
+  %arrayidx1.25 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 25
+  store i32 %26, i32* %arrayidx1.25, align 4, !tbaa !7
+  %arrayidx.26 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 26
+  %27 = load i32, i32 addrspace(1)* %arrayidx.26, align 4, !tbaa !7
+  %arrayidx1.26 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 26
+  store i32 %27, i32* %arrayidx1.26, align 4, !tbaa !7
+  %arrayidx.27 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 27
+  %28 = load i32, i32 addrspace(1)* %arrayidx.27, align 4, !tbaa !7
+  %arrayidx1.27 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 27
+  store i32 %28, i32* %arrayidx1.27, align 4, !tbaa !7
+  %arrayidx.28 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 28
+  %29 = load i32, i32 addrspace(1)* %arrayidx.28, align 4, !tbaa !7
+  %arrayidx1.28 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 28
+  store i32 %29, i32* %arrayidx1.28, align 4, !tbaa !7
+  %arrayidx.29 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 29
+  %30 = load i32, i32 addrspace(1)* %arrayidx.29, align 4, !tbaa !7
+  %arrayidx1.29 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 29
+  store i32 %30, i32* %arrayidx1.29, align 4, !tbaa !7
+  %arrayidx.30 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 30
+  %31 = load i32, i32 addrspace(1)* %arrayidx.30, align 4, !tbaa !7
+  %arrayidx1.30 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 30
+  store i32 %31, i32* %arrayidx1.30, align 4, !tbaa !7
+  %arrayidx.31 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 31
+  %32 = load i32, i32 addrspace(1)* %arrayidx.31, align 4, !tbaa !7
+  %arrayidx1.31 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 31
+  store i32 %32, i32* %arrayidx1.31, align 4, !tbaa !7
+  %arrayidx.32 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 32
+  %33 = load i32, i32 addrspace(1)* %arrayidx.32, align 4, !tbaa !7
+  %arrayidx1.32 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 32
+  store i32 %33, i32* %arrayidx1.32, align 4, !tbaa !7
+  %arrayidx.33 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 33
+  %34 = load i32, i32 addrspace(1)* %arrayidx.33, align 4, !tbaa !7
+  %arrayidx1.33 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 33
+  store i32 %34, i32* %arrayidx1.33, align 4, !tbaa !7
+  %arrayidx.34 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 34
+  %35 = load i32, i32 addrspace(1)* %arrayidx.34, align 4, !tbaa !7
+  %arrayidx1.34 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 34
+  store i32 %35, i32* %arrayidx1.34, align 4, !tbaa !7
+  %arrayidx.35 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 35
+  %36 = load i32, i32 addrspace(1)* %arrayidx.35, align 4, !tbaa !7
+  %arrayidx1.35 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 35
+  store i32 %36, i32* %arrayidx1.35, align 4, !tbaa !7
+  %arrayidx.36 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 36
+  %37 = load i32, i32 addrspace(1)* %arrayidx.36, align 4, !tbaa !7
+  %arrayidx1.36 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 36
+  store i32 %37, i32* %arrayidx1.36, align 4, !tbaa !7
+  %arrayidx.37 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 37
+  %38 = load i32, i32 addrspace(1)* %arrayidx.37, align 4, !tbaa !7
+  %arrayidx1.37 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 37
+  store i32 %38, i32* %arrayidx1.37, align 4, !tbaa !7
+  %arrayidx.38 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 38
+  %39 = load i32, i32 addrspace(1)* %arrayidx.38, align 4, !tbaa !7
+  %arrayidx1.38 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 38
+  store i32 %39, i32* %arrayidx1.38, align 4, !tbaa !7
+  %arrayidx.39 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 39
+  %40 = load i32, i32 addrspace(1)* %arrayidx.39, align 4, !tbaa !7
+  %arrayidx1.39 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 39
+  store i32 %40, i32* %arrayidx1.39, align 4, !tbaa !7
+  %arrayidx.40 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 40
+  %41 = load i32, i32 addrspace(1)* %arrayidx.40, align 4, !tbaa !7
+  %arrayidx1.40 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 40
+  store i32 %41, i32* %arrayidx1.40, align 4, !tbaa !7
+  %arrayidx.41 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 41
+  %42 = load i32, i32 addrspace(1)* %arrayidx.41, align 4, !tbaa !7
+  %arrayidx1.41 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 41
+  store i32 %42, i32* %arrayidx1.41, align 4, !tbaa !7
+  %arrayidx.42 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 42
+  %43 = load i32, i32 addrspace(1)* %arrayidx.42, align 4, !tbaa !7
+  %arrayidx1.42 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 42
+  store i32 %43, i32* %arrayidx1.42, align 4, !tbaa !7
+  %arrayidx.43 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 43
+  %44 = load i32, i32 addrspace(1)* %arrayidx.43, align 4, !tbaa !7
+  %arrayidx1.43 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 43
+  store i32 %44, i32* %arrayidx1.43, align 4, !tbaa !7
+  %arrayidx.44 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 44
+  %45 = load i32, i32 addrspace(1)* %arrayidx.44, align 4, !tbaa !7
+  %arrayidx1.44 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 44
+  store i32 %45, i32* %arrayidx1.44, align 4, !tbaa !7
+  %arrayidx.45 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 45
+  %46 = load i32, i32 addrspace(1)* %arrayidx.45, align 4, !tbaa !7
+  %arrayidx1.45 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 45
+  store i32 %46, i32* %arrayidx1.45, align 4, !tbaa !7
+  %arrayidx.46 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 46
+  %47 = load i32, i32 addrspace(1)* %arrayidx.46, align 4, !tbaa !7
+  %arrayidx1.46 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 46
+  store i32 %47, i32* %arrayidx1.46, align 4, !tbaa !7
+  %arrayidx.47 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 47
+  %48 = load i32, i32 addrspace(1)* %arrayidx.47, align 4, !tbaa !7
+  %arrayidx1.47 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 47
+  store i32 %48, i32* %arrayidx1.47, align 4, !tbaa !7
+  %arrayidx.48 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 48
+  %49 = load i32, i32 addrspace(1)* %arrayidx.48, align 4, !tbaa !7
+  %arrayidx1.48 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 48
+  store i32 %49, i32* %arrayidx1.48, align 4, !tbaa !7
+  %arrayidx.49 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 49
+  %50 = load i32, i32 addrspace(1)* %arrayidx.49, align 4, !tbaa !7
+  %arrayidx1.49 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 49
+  store i32 %50, i32* %arrayidx1.49, align 4, !tbaa !7
+  %arrayidx.50 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 50
+  %51 = load i32, i32 addrspace(1)* %arrayidx.50, align 4, !tbaa !7
+  %arrayidx1.50 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 50
+  store i32 %51, i32* %arrayidx1.50, align 4, !tbaa !7
+  %arrayidx.51 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 51
+  %52 = load i32, i32 addrspace(1)* %arrayidx.51, align 4, !tbaa !7
+  %arrayidx1.51 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 51
+  store i32 %52, i32* %arrayidx1.51, align 4, !tbaa !7
+  %arrayidx.52 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 52
+  %53 = load i32, i32 addrspace(1)* %arrayidx.52, align 4, !tbaa !7
+  %arrayidx1.52 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 52
+  store i32 %53, i32* %arrayidx1.52, align 4, !tbaa !7
+  %arrayidx.53 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 53
+  %54 = load i32, i32 addrspace(1)* %arrayidx.53, align 4, !tbaa !7
+  %arrayidx1.53 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 53
+  store i32 %54, i32* %arrayidx1.53, align 4, !tbaa !7
+  %arrayidx.54 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 54
+  %55 = load i32, i32 addrspace(1)* %arrayidx.54, align 4, !tbaa !7
+  %arrayidx1.54 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 54
+  store i32 %55, i32* %arrayidx1.54, align 4, !tbaa !7
+  %arrayidx.55 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 55
+  %56 = load i32, i32 addrspace(1)* %arrayidx.55, align 4, !tbaa !7
+  %arrayidx1.55 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 55
+  store i32 %56, i32* %arrayidx1.55, align 4, !tbaa !7
+  %arrayidx.56 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 56
+  %57 = load i32, i32 addrspace(1)* %arrayidx.56, align 4, !tbaa !7
+  %arrayidx1.56 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 56
+  store i32 %57, i32* %arrayidx1.56, align 4, !tbaa !7
+  %arrayidx.57 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 57
+  %58 = load i32, i32 addrspace(1)* %arrayidx.57, align 4, !tbaa !7
+  %arrayidx1.57 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 57
+  store i32 %58, i32* %arrayidx1.57, align 4, !tbaa !7
+  %arrayidx.58 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 58
+  %59 = load i32, i32 addrspace(1)* %arrayidx.58, align 4, !tbaa !7
+  %arrayidx1.58 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 58
+  store i32 %59, i32* %arrayidx1.58, align 4, !tbaa !7
+  %arrayidx.59 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 59
+  %60 = load i32, i32 addrspace(1)* %arrayidx.59, align 4, !tbaa !7
+  %arrayidx1.59 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 59
+  store i32 %60, i32* %arrayidx1.59, align 4, !tbaa !7
+  %arrayidx.60 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 60
+  %61 = load i32, i32 addrspace(1)* %arrayidx.60, align 4, !tbaa !7
+  %arrayidx1.60 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 60
+  store i32 %61, i32* %arrayidx1.60, align 4, !tbaa !7
+  %arrayidx.61 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 61
+  %62 = load i32, i32 addrspace(1)* %arrayidx.61, align 4, !tbaa !7
+  %arrayidx1.61 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 61
+  store i32 %62, i32* %arrayidx1.61, align 4, !tbaa !7
+  %arrayidx.62 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 62
+  %63 = load i32, i32 addrspace(1)* %arrayidx.62, align 4, !tbaa !7
+  %arrayidx1.62 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 62
+  store i32 %63, i32* %arrayidx1.62, align 4, !tbaa !7
+  %arrayidx.63 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 63
+  %64 = load i32, i32 addrspace(1)* %arrayidx.63, align 4, !tbaa !7
+  %arrayidx1.63 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 63
+  store i32 %64, i32* %arrayidx1.63, align 4, !tbaa !7
+  %arrayidx.64 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 64
+  %65 = load i32, i32 addrspace(1)* %arrayidx.64, align 4, !tbaa !7
+  %arrayidx1.64 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 64
+  store i32 %65, i32* %arrayidx1.64, align 4, !tbaa !7
+  %arrayidx.65 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 65
+  %66 = load i32, i32 addrspace(1)* %arrayidx.65, align 4, !tbaa !7
+  %arrayidx1.65 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 65
+  store i32 %66, i32* %arrayidx1.65, align 4, !tbaa !7
+  %arrayidx.66 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 66
+  %67 = load i32, i32 addrspace(1)* %arrayidx.66, align 4, !tbaa !7
+  %arrayidx1.66 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 66
+  store i32 %67, i32* %arrayidx1.66, align 4, !tbaa !7
+  %arrayidx.67 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 67
+  %68 = load i32, i32 addrspace(1)* %arrayidx.67, align 4, !tbaa !7
+  %arrayidx1.67 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 67
+  store i32 %68, i32* %arrayidx1.67, align 4, !tbaa !7
+  %arrayidx.68 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 68
+  %69 = load i32, i32 addrspace(1)* %arrayidx.68, align 4, !tbaa !7
+  %arrayidx1.68 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 68
+  store i32 %69, i32* %arrayidx1.68, align 4, !tbaa !7
+  %arrayidx.69 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 69
+  %70 = load i32, i32 addrspace(1)* %arrayidx.69, align 4, !tbaa !7
+  %arrayidx1.69 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 69
+  store i32 %70, i32* %arrayidx1.69, align 4, !tbaa !7
+  %arrayidx.70 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 70
+  %71 = load i32, i32 addrspace(1)* %arrayidx.70, align 4, !tbaa !7
+  %arrayidx1.70 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 70
+  store i32 %71, i32* %arrayidx1.70, align 4, !tbaa !7
+  %arrayidx.71 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 71
+  %72 = load i32, i32 addrspace(1)* %arrayidx.71, align 4, !tbaa !7
+  %arrayidx1.71 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 71
+  store i32 %72, i32* %arrayidx1.71, align 4, !tbaa !7
+  %arrayidx.72 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 72
+  %73 = load i32, i32 addrspace(1)* %arrayidx.72, align 4, !tbaa !7
+  %arrayidx1.72 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 72
+  store i32 %73, i32* %arrayidx1.72, align 4, !tbaa !7
+  %arrayidx.73 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 73
+  %74 = load i32, i32 addrspace(1)* %arrayidx.73, align 4, !tbaa !7
+  %arrayidx1.73 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 73
+  store i32 %74, i32* %arrayidx1.73, align 4, !tbaa !7
+  %arrayidx.74 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 74
+  %75 = load i32, i32 addrspace(1)* %arrayidx.74, align 4, !tbaa !7
+  %arrayidx1.74 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 74
+  store i32 %75, i32* %arrayidx1.74, align 4, !tbaa !7
+  %arrayidx.75 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 75
+  %76 = load i32, i32 addrspace(1)* %arrayidx.75, align 4, !tbaa !7
+  %arrayidx1.75 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 75
+  store i32 %76, i32* %arrayidx1.75, align 4, !tbaa !7
+  %arrayidx.76 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 76
+  %77 = load i32, i32 addrspace(1)* %arrayidx.76, align 4, !tbaa !7
+  %arrayidx1.76 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 76
+  store i32 %77, i32* %arrayidx1.76, align 4, !tbaa !7
+  %arrayidx.77 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 77
+  %78 = load i32, i32 addrspace(1)* %arrayidx.77, align 4, !tbaa !7
+  %arrayidx1.77 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 77
+  store i32 %78, i32* %arrayidx1.77, align 4, !tbaa !7
+  %arrayidx.78 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 78
+  %79 = load i32, i32 addrspace(1)* %arrayidx.78, align 4, !tbaa !7
+  %arrayidx1.78 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 78
+  store i32 %79, i32* %arrayidx1.78, align 4, !tbaa !7
+  %arrayidx.79 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 79
+  %80 = load i32, i32 addrspace(1)* %arrayidx.79, align 4, !tbaa !7
+  %arrayidx1.79 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 79
+  store i32 %80, i32* %arrayidx1.79, align 4, !tbaa !7
+  %arrayidx.80 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 80
+  %81 = load i32, i32 addrspace(1)* %arrayidx.80, align 4, !tbaa !7
+  %arrayidx1.80 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 80
+  store i32 %81, i32* %arrayidx1.80, align 4, !tbaa !7
+  %arrayidx.81 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 81
+  %82 = load i32, i32 addrspace(1)* %arrayidx.81, align 4, !tbaa !7
+  %arrayidx1.81 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 81
+  store i32 %82, i32* %arrayidx1.81, align 4, !tbaa !7
+  %arrayidx.82 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 82
+  %83 = load i32, i32 addrspace(1)* %arrayidx.82, align 4, !tbaa !7
+  %arrayidx1.82 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 82
+  store i32 %83, i32* %arrayidx1.82, align 4, !tbaa !7
+  %arrayidx.83 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 83
+  %84 = load i32, i32 addrspace(1)* %arrayidx.83, align 4, !tbaa !7
+  %arrayidx1.83 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 83
+  store i32 %84, i32* %arrayidx1.83, align 4, !tbaa !7
+  %arrayidx.84 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 84
+  %85 = load i32, i32 addrspace(1)* %arrayidx.84, align 4, !tbaa !7
+  %arrayidx1.84 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 84
+  store i32 %85, i32* %arrayidx1.84, align 4, !tbaa !7
+  %arrayidx.85 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 85
+  %86 = load i32, i32 addrspace(1)* %arrayidx.85, align 4, !tbaa !7
+  %arrayidx1.85 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 85
+  store i32 %86, i32* %arrayidx1.85, align 4, !tbaa !7
+  %arrayidx.86 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 86
+  %87 = load i32, i32 addrspace(1)* %arrayidx.86, align 4, !tbaa !7
+  %arrayidx1.86 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 86
+  store i32 %87, i32* %arrayidx1.86, align 4, !tbaa !7
+  %arrayidx.87 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 87
+  %88 = load i32, i32 addrspace(1)* %arrayidx.87, align 4, !tbaa !7
+  %arrayidx1.87 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 87
+  store i32 %88, i32* %arrayidx1.87, align 4, !tbaa !7
+  %arrayidx.88 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 88
+  %89 = load i32, i32 addrspace(1)* %arrayidx.88, align 4, !tbaa !7
+  %arrayidx1.88 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 88
+  store i32 %89, i32* %arrayidx1.88, align 4, !tbaa !7
+  %arrayidx.89 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 89
+  %90 = load i32, i32 addrspace(1)* %arrayidx.89, align 4, !tbaa !7
+  %arrayidx1.89 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 89
+  store i32 %90, i32* %arrayidx1.89, align 4, !tbaa !7
+  %arrayidx.90 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 90
+  %91 = load i32, i32 addrspace(1)* %arrayidx.90, align 4, !tbaa !7
+  %arrayidx1.90 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 90
+  store i32 %91, i32* %arrayidx1.90, align 4, !tbaa !7
+  %arrayidx.91 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 91
+  %92 = load i32, i32 addrspace(1)* %arrayidx.91, align 4, !tbaa !7
+  %arrayidx1.91 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 91
+  store i32 %92, i32* %arrayidx1.91, align 4, !tbaa !7
+  %arrayidx.92 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 92
+  %93 = load i32, i32 addrspace(1)* %arrayidx.92, align 4, !tbaa !7
+  %arrayidx1.92 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 92
+  store i32 %93, i32* %arrayidx1.92, align 4, !tbaa !7
+  %arrayidx.93 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 93
+  %94 = load i32, i32 addrspace(1)* %arrayidx.93, align 4, !tbaa !7
+  %arrayidx1.93 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 93
+  store i32 %94, i32* %arrayidx1.93, align 4, !tbaa !7
+  %arrayidx.94 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 94
+  %95 = load i32, i32 addrspace(1)* %arrayidx.94, align 4, !tbaa !7
+  %arrayidx1.94 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 94
+  store i32 %95, i32* %arrayidx1.94, align 4, !tbaa !7
+  %arrayidx.95 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 95
+  %96 = load i32, i32 addrspace(1)* %arrayidx.95, align 4, !tbaa !7
+  %arrayidx1.95 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 95
+  store i32 %96, i32* %arrayidx1.95, align 4, !tbaa !7
+  %arrayidx.96 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 96
+  %97 = load i32, i32 addrspace(1)* %arrayidx.96, align 4, !tbaa !7
+  %arrayidx1.96 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 96
+  store i32 %97, i32* %arrayidx1.96, align 4, !tbaa !7
+  %arrayidx.97 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 97
+  %98 = load i32, i32 addrspace(1)* %arrayidx.97, align 4, !tbaa !7
+  %arrayidx1.97 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 97
+  store i32 %98, i32* %arrayidx1.97, align 4, !tbaa !7
+  %arrayidx.98 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 98
+  %99 = load i32, i32 addrspace(1)* %arrayidx.98, align 4, !tbaa !7
+  %arrayidx1.98 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 98
+  store i32 %99, i32* %arrayidx1.98, align 4, !tbaa !7
+  %arrayidx.99 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 99
+  %100 = load i32, i32 addrspace(1)* %arrayidx.99, align 4, !tbaa !7
+  %arrayidx1.99 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 99
+  store i32 %100, i32* %arrayidx1.99, align 4, !tbaa !7
+  %arrayidx.100 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 100
+  %101 = load i32, i32 addrspace(1)* %arrayidx.100, align 4, !tbaa !7
+  %arrayidx1.100 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 100
+  store i32 %101, i32* %arrayidx1.100, align 4, !tbaa !7
+  %arrayidx.101 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 101
+  %102 = load i32, i32 addrspace(1)* %arrayidx.101, align 4, !tbaa !7
+  %arrayidx1.101 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 101
+  store i32 %102, i32* %arrayidx1.101, align 4, !tbaa !7
+  %arrayidx.102 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 102
+  %103 = load i32, i32 addrspace(1)* %arrayidx.102, align 4, !tbaa !7
+  %arrayidx1.102 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 102
+  store i32 %103, i32* %arrayidx1.102, align 4, !tbaa !7
+  %arrayidx.103 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 103
+  %104 = load i32, i32 addrspace(1)* %arrayidx.103, align 4, !tbaa !7
+  %arrayidx1.103 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 103
+  store i32 %104, i32* %arrayidx1.103, align 4, !tbaa !7
+  %arrayidx.104 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 104
+  %105 = load i32, i32 addrspace(1)* %arrayidx.104, align 4, !tbaa !7
+  %arrayidx1.104 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 104
+  store i32 %105, i32* %arrayidx1.104, align 4, !tbaa !7
+  %arrayidx.105 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 105
+  %106 = load i32, i32 addrspace(1)* %arrayidx.105, align 4, !tbaa !7
+  %arrayidx1.105 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 105
+  store i32 %106, i32* %arrayidx1.105, align 4, !tbaa !7
+  %arrayidx.106 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 106
+  %107 = load i32, i32 addrspace(1)* %arrayidx.106, align 4, !tbaa !7
+  %arrayidx1.106 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 106
+  store i32 %107, i32* %arrayidx1.106, align 4, !tbaa !7
+  %arrayidx.107 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 107
+  %108 = load i32, i32 addrspace(1)* %arrayidx.107, align 4, !tbaa !7
+  %arrayidx1.107 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 107
+  store i32 %108, i32* %arrayidx1.107, align 4, !tbaa !7
+  %arrayidx.108 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 108
+  %109 = load i32, i32 addrspace(1)* %arrayidx.108, align 4, !tbaa !7
+  %arrayidx1.108 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 108
+  store i32 %109, i32* %arrayidx1.108, align 4, !tbaa !7
+  %arrayidx.109 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 109
+  %110 = load i32, i32 addrspace(1)* %arrayidx.109, align 4, !tbaa !7
+  %arrayidx1.109 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 109
+  store i32 %110, i32* %arrayidx1.109, align 4, !tbaa !7
+  %arrayidx.110 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 110
+  %111 = load i32, i32 addrspace(1)* %arrayidx.110, align 4, !tbaa !7
+  %arrayidx1.110 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 110
+  store i32 %111, i32* %arrayidx1.110, align 4, !tbaa !7
+  %112 = load i32, i32* %arrayidx1, align 4, !tbaa !7
+  %mul = sub nsw i32 0, %112
+  store i32 %mul, i32 addrspace(1)* %A, align 4, !tbaa !7
+  %113 = load i32, i32* %arrayidx1.1, align 4, !tbaa !7
+  %mul.1 = sub nsw i32 0, %113
+  store i32 %mul.1, i32 addrspace(1)* %arrayidx.1, align 4, !tbaa !7
+  %114 = load i32, i32* %arrayidx1.2, align 4, !tbaa !7
+  %mul.2 = sub nsw i32 0, %114
+  store i32 %mul.2, i32 addrspace(1)* %arrayidx.2, align 4, !tbaa !7
+  %115 = load i32, i32* %arrayidx1.3, align 4, !tbaa !7
+  %mul.3 = sub nsw i32 0, %115
+  store i32 %mul.3, i32 addrspace(1)* %arrayidx.3, align 4, !tbaa !7
+  %116 = load i32, i32* %arrayidx1.4, align 4, !tbaa !7
+  %mul.4 = sub nsw i32 0, %116
+  store i32 %mul.4, i32 addrspace(1)* %arrayidx.4, align 4, !tbaa !7
+  %117 = load i32, i32* %arrayidx1.5, align 4, !tbaa !7
+  %mul.5 = sub nsw i32 0, %117
+  store i32 %mul.5, i32 addrspace(1)* %arrayidx.5, align 4, !tbaa !7
+  %118 = load i32, i32* %arrayidx1.6, align 4, !tbaa !7
+  %mul.6 = sub nsw i32 0, %118
+  store i32 %mul.6, i32 addrspace(1)* %arrayidx.6, align 4, !tbaa !7
+  %119 = load i32, i32* %arrayidx1.7, align 4, !tbaa !7
+  %mul.7 = sub nsw i32 0, %119
+  store i32 %mul.7, i32 addrspace(1)* %arrayidx.7, align 4, !tbaa !7
+  %120 = load i32, i32* %arrayidx1.8, align 4, !tbaa !7
+  %mul.8 = sub nsw i32 0, %120
+  store i32 %mul.8, i32 addrspace(1)* %arrayidx.8, align 4, !tbaa !7
+  %121 = load i32, i32* %arrayidx1.9, align 4, !tbaa !7
+  %mul.9 = sub nsw i32 0, %121
+  store i32 %mul.9, i32 addrspace(1)* %arrayidx.9, align 4, !tbaa !7
+  %122 = load i32, i32* %arrayidx1.10, align 4, !tbaa !7
+  %mul.10 = sub nsw i32 0, %122
+  store i32 %mul.10, i32 addrspace(1)* %arrayidx.10, align 4, !tbaa !7
+  %123 = load i32, i32* %arrayidx1.11, align 4, !tbaa !7
+  %mul.11 = sub nsw i32 0, %123
+  store i32 %mul.11, i32 addrspace(1)* %arrayidx.11, align 4, !tbaa !7
+  %124 = load i32, i32* %arrayidx1.12, align 4, !tbaa !7
+  %mul.12 = sub nsw i32 0, %124
+  store i32 %mul.12, i32 addrspace(1)* %arrayidx.12, align 4, !tbaa !7
+  %125 = load i32, i32* %arrayidx1.13, align 4, !tbaa !7
+  %mul.13 = sub nsw i32 0, %125
+  store i32 %mul.13, i32 addrspace(1)* %arrayidx.13, align 4, !tbaa !7
+  %126 = load i32, i32* %arrayidx1.14, align 4, !tbaa !7
+  %mul.14 = sub nsw i32 0, %126
+  store i32 %mul.14, i32 addrspace(1)* %arrayidx.14, align 4, !tbaa !7
+  %127 = load i32, i32* %arrayidx1.15, align 4, !tbaa !7
+  %mul.15 = sub nsw i32 0, %127
+  store i32 %mul.15, i32 addrspace(1)* %arrayidx.15, align 4, !tbaa !7
+  %128 = load i32, i32* %arrayidx1.16, align 4, !tbaa !7
+  %mul.16 = sub nsw i32 0, %128
+  store i32 %mul.16, i32 addrspace(1)* %arrayidx.16, align 4, !tbaa !7
+  %129 = load i32, i32* %arrayidx1.17, align 4, !tbaa !7
+  %mul.17 = sub nsw i32 0, %129
+  store i32 %mul.17, i32 addrspace(1)* %arrayidx.17, align 4, !tbaa !7
+  %130 = load i32, i32* %arrayidx1.18, align 4, !tbaa !7
+  %mul.18 = sub nsw i32 0, %130
+  store i32 %mul.18, i32 addrspace(1)* %arrayidx.18, align 4, !tbaa !7
+  %131 = load i32, i32* %arrayidx1.19, align 4, !tbaa !7
+  %mul.19 = sub nsw i32 0, %131
+  store i32 %mul.19, i32 addrspace(1)* %arrayidx.19, align 4, !tbaa !7
+  %132 = load i32, i32* %arrayidx1.20, align 4, !tbaa !7
+  %mul.20 = sub nsw i32 0, %132
+  store i32 %mul.20, i32 addrspace(1)* %arrayidx.20, align 4, !tbaa !7
+  %133 = load i32, i32* %arrayidx1.21, align 4, !tbaa !7
+  %mul.21 = sub nsw i32 0, %133
+  store i32 %mul.21, i32 addrspace(1)* %arrayidx.21, align 4, !tbaa !7
+  %134 = load i32, i32* %arrayidx1.22, align 4, !tbaa !7
+  %mul.22 = sub nsw i32 0, %134
+  store i32 %mul.22, i32 addrspace(1)* %arrayidx.22, align 4, !tbaa !7
+  %135 = load i32, i32* %arrayidx1.23, align 4, !tbaa !7
+  %mul.23 = sub nsw i32 0, %135
+  store i32 %mul.23, i32 addrspace(1)* %arrayidx.23, align 4, !tbaa !7
+  %136 = load i32, i32* %arrayidx1.24, align 4, !tbaa !7
+  %mul.24 = sub nsw i32 0, %136
+  store i32 %mul.24, i32 addrspace(1)* %arrayidx.24, align 4, !tbaa !7
+  %137 = load i32, i32* %arrayidx1.25, align 4, !tbaa !7
+  %mul.25 = sub nsw i32 0, %137
+  store i32 %mul.25, i32 addrspace(1)* %arrayidx.25, align 4, !tbaa !7
+  %138 = load i32, i32* %arrayidx1.26, align 4, !tbaa !7
+  %mul.26 = sub nsw i32 0, %138
+  store i32 %mul.26, i32 addrspace(1)* %arrayidx.26, align 4, !tbaa !7
+  %139 = load i32, i32* %arrayidx1.27, align 4, !tbaa !7
+  %mul.27 = sub nsw i32 0, %139
+  store i32 %mul.27, i32 addrspace(1)* %arrayidx.27, align 4, !tbaa !7
+  %140 = load i32, i32* %arrayidx1.28, align 4, !tbaa !7
+  %mul.28 = sub nsw i32 0, %140
+  store i32 %mul.28, i32 addrspace(1)* %arrayidx.28, align 4, !tbaa !7
+  %141 = load i32, i32* %arrayidx1.29, align 4, !tbaa !7
+  %mul.29 = sub nsw i32 0, %141
+  store i32 %mul.29, i32 addrspace(1)* %arrayidx.29, align 4, !tbaa !7
+  %142 = load i32, i32* %arrayidx1.30, align 4, !tbaa !7
+  %mul.30 = sub nsw i32 0, %142
+  store i32 %mul.30, i32 addrspace(1)* %arrayidx.30, align 4, !tbaa !7
+  %143 = load i32, i32* %arrayidx1.31, align 4, !tbaa !7
+  %mul.31 = sub nsw i32 0, %143
+  store i32 %mul.31, i32 addrspace(1)* %arrayidx.31, align 4, !tbaa !7
+  %144 = load i32, i32* %arrayidx1.32, align 4, !tbaa !7
+  %mul.32 = sub nsw i32 0, %144
+  store i32 %mul.32, i32 addrspace(1)* %arrayidx.32, align 4, !tbaa !7
+  %145 = load i32, i32* %arrayidx1.33, align 4, !tbaa !7
+  %mul.33 = sub nsw i32 0, %145
+  store i32 %mul.33, i32 addrspace(1)* %arrayidx.33, align 4, !tbaa !7
+  %146 = load i32, i32* %arrayidx1.34, align 4, !tbaa !7
+  %mul.34 = sub nsw i32 0, %146
+  store i32 %mul.34, i32 addrspace(1)* %arrayidx.34, align 4, !tbaa !7
+  %147 = load i32, i32* %arrayidx1.35, align 4, !tbaa !7
+  %mul.35 = sub nsw i32 0, %147
+  store i32 %mul.35, i32 addrspace(1)* %arrayidx.35, align 4, !tbaa !7
+  %148 = load i32, i32* %arrayidx1.36, align 4, !tbaa !7
+  %mul.36 = sub nsw i32 0, %148
+  store i32 %mul.36, i32 addrspace(1)* %arrayidx.36, align 4, !tbaa !7
+  %149 = load i32, i32* %arrayidx1.37, align 4, !tbaa !7
+  %mul.37 = sub nsw i32 0, %149
+  store i32 %mul.37, i32 addrspace(1)* %arrayidx.37, align 4, !tbaa !7
+  %150 = load i32, i32* %arrayidx1.38, align 4, !tbaa !7
+  %mul.38 = sub nsw i32 0, %150
+  store i32 %mul.38, i32 addrspace(1)* %arrayidx.38, align 4, !tbaa !7
+  %151 = load i32, i32* %arrayidx1.39, align 4, !tbaa !7
+  %mul.39 = sub nsw i32 0, %151
+  store i32 %mul.39, i32 addrspace(1)* %arrayidx.39, align 4, !tbaa !7
+  %152 = load i32, i32* %arrayidx1.40, align 4, !tbaa !7
+  %mul.40 = sub nsw i32 0, %152
+  store i32 %mul.40, i32 addrspace(1)* %arrayidx.40, align 4, !tbaa !7
+  %153 = load i32, i32* %arrayidx1.41, align 4, !tbaa !7
+  %mul.41 = sub nsw i32 0, %153
+  store i32 %mul.41, i32 addrspace(1)* %arrayidx.41, align 4, !tbaa !7
+  %154 = load i32, i32* %arrayidx1.42, align 4, !tbaa !7
+  %mul.42 = sub nsw i32 0, %154
+  store i32 %mul.42, i32 addrspace(1)* %arrayidx.42, align 4, !tbaa !7
+  %155 = load i32, i32* %arrayidx1.43, align 4, !tbaa !7
+  %mul.43 = sub nsw i32 0, %155
+  store i32 %mul.43, i32 addrspace(1)* %arrayidx.43, align 4, !tbaa !7
+  %156 = load i32, i32* %arrayidx1.44, align 4, !tbaa !7
+  %mul.44 = sub nsw i32 0, %156
+  store i32 %mul.44, i32 addrspace(1)* %arrayidx.44, align 4, !tbaa !7
+  %157 = load i32, i32* %arrayidx1.45, align 4, !tbaa !7
+  %mul.45 = sub nsw i32 0, %157
+  store i32 %mul.45, i32 addrspace(1)* %arrayidx.45, align 4, !tbaa !7
+  %158 = load i32, i32* %arrayidx1.46, align 4, !tbaa !7
+  %mul.46 = sub nsw i32 0, %158
+  store i32 %mul.46, i32 addrspace(1)* %arrayidx.46, align 4, !tbaa !7
+  %159 = load i32, i32* %arrayidx1.47, align 4, !tbaa !7
+  %mul.47 = sub nsw i32 0, %159
+  store i32 %mul.47, i32 addrspace(1)* %arrayidx.47, align 4, !tbaa !7
+  %160 = load i32, i32* %arrayidx1.48, align 4, !tbaa !7
+  %mul.48 = sub nsw i32 0, %160
+  store i32 %mul.48, i32 addrspace(1)* %arrayidx.48, align 4, !tbaa !7
+  %161 = load i32, i32* %arrayidx1.49, align 4, !tbaa !7
+  %mul.49 = sub nsw i32 0, %161
+  store i32 %mul.49, i32 addrspace(1)* %arrayidx.49, align 4, !tbaa !7
+  %162 = load i32, i32* %arrayidx1.50, align 4, !tbaa !7
+  %mul.50 = sub nsw i32 0, %162
+  store i32 %mul.50, i32 addrspace(1)* %arrayidx.50, align 4, !tbaa !7
+  %163 = load i32, i32* %arrayidx1.51, align 4, !tbaa !7
+  %mul.51 = sub nsw i32 0, %163
+  store i32 %mul.51, i32 addrspace(1)* %arrayidx.51, align 4, !tbaa !7
+  %164 = load i32, i32* %arrayidx1.52, align 4, !tbaa !7
+  %mul.52 = sub nsw i32 0, %164
+  store i32 %mul.52, i32 addrspace(1)* %arrayidx.52, align 4, !tbaa !7
+  %165 = load i32, i32* %arrayidx1.53, align 4, !tbaa !7
+  %mul.53 = sub nsw i32 0, %165
+  store i32 %mul.53, i32 addrspace(1)* %arrayidx.53, align 4, !tbaa !7
+  %166 = load i32, i32* %arrayidx1.54, align 4, !tbaa !7
+  %mul.54 = sub nsw i32 0, %166
+  store i32 %mul.54, i32 addrspace(1)* %arrayidx.54, align 4, !tbaa !7
+  %167 = load i32, i32* %arrayidx1.55, align 4, !tbaa !7
+  %mul.55 = sub nsw i32 0, %167
+  store i32 %mul.55, i32 addrspace(1)* %arrayidx.55, align 4, !tbaa !7
+  %168 = load i32, i32* %arrayidx1.56, align 4, !tbaa !7
+  %mul.56 = sub nsw i32 0, %168
+  store i32 %mul.56, i32 addrspace(1)* %arrayidx.56, align 4, !tbaa !7
+  %169 = load i32, i32* %arrayidx1.57, align 4, !tbaa !7
+  %mul.57 = sub nsw i32 0, %169
+  store i32 %mul.57, i32 addrspace(1)* %arrayidx.57, align 4, !tbaa !7
+  %170 = load i32, i32* %arrayidx1.58, align 4, !tbaa !7
+  %mul.58 = sub nsw i32 0, %170
+  store i32 %mul.58, i32 addrspace(1)* %arrayidx.58, align 4, !tbaa !7
+  %171 = load i32, i32* %arrayidx1.59, align 4, !tbaa !7
+  %mul.59 = sub nsw i32 0, %171
+  store i32 %mul.59, i32 addrspace(1)* %arrayidx.59, align 4, !tbaa !7
+  %172 = load i32, i32* %arrayidx1.60, align 4, !tbaa !7
+  %mul.60 = sub nsw i32 0, %172
+  store i32 %mul.60, i32 addrspace(1)* %arrayidx.60, align 4, !tbaa !7
+  %173 = load i32, i32* %arrayidx1.61, align 4, !tbaa !7
+  %mul.61 = sub nsw i32 0, %173
+  store i32 %mul.61, i32 addrspace(1)* %arrayidx.61, align 4, !tbaa !7
+  %174 = load i32, i32* %arrayidx1.62, align 4, !tbaa !7
+  %mul.62 = sub nsw i32 0, %174
+  store i32 %mul.62, i32 addrspace(1)* %arrayidx.62, align 4, !tbaa !7
+  %175 = load i32, i32* %arrayidx1.63, align 4, !tbaa !7
+  %mul.63 = sub nsw i32 0, %175
+  store i32 %mul.63, i32 addrspace(1)* %arrayidx.63, align 4, !tbaa !7
+  %176 = load i32, i32* %arrayidx1.64, align 4, !tbaa !7
+  %mul.64 = sub nsw i32 0, %176
+  store i32 %mul.64, i32 addrspace(1)* %arrayidx.64, align 4, !tbaa !7
+  %177 = load i32, i32* %arrayidx1.65, align 4, !tbaa !7
+  %mul.65 = sub nsw i32 0, %177
+  store i32 %mul.65, i32 addrspace(1)* %arrayidx.65, align 4, !tbaa !7
+  %178 = load i32, i32* %arrayidx1.66, align 4, !tbaa !7
+  %mul.66 = sub nsw i32 0, %178
+  store i32 %mul.66, i32 addrspace(1)* %arrayidx.66, align 4, !tbaa !7
+  %179 = load i32, i32* %arrayidx1.67, align 4, !tbaa !7
+  %mul.67 = sub nsw i32 0, %179
+  store i32 %mul.67, i32 addrspace(1)* %arrayidx.67, align 4, !tbaa !7
+  %180 = load i32, i32* %arrayidx1.68, align 4, !tbaa !7
+  %mul.68 = sub nsw i32 0, %180
+  store i32 %mul.68, i32 addrspace(1)* %arrayidx.68, align 4, !tbaa !7
+  %181 = load i32, i32* %arrayidx1.69, align 4, !tbaa !7
+  %mul.69 = sub nsw i32 0, %181
+  store i32 %mul.69, i32 addrspace(1)* %arrayidx.69, align 4, !tbaa !7
+  %182 = load i32, i32* %arrayidx1.70, align 4, !tbaa !7
+  %mul.70 = sub nsw i32 0, %182
+  store i32 %mul.70, i32 addrspace(1)* %arrayidx.70, align 4, !tbaa !7
+  %183 = load i32, i32* %arrayidx1.71, align 4, !tbaa !7
+  %mul.71 = sub nsw i32 0, %183
+  store i32 %mul.71, i32 addrspace(1)* %arrayidx.71, align 4, !tbaa !7
+  %184 = load i32, i32* %arrayidx1.72, align 4, !tbaa !7
+  %mul.72 = sub nsw i32 0, %184
+  store i32 %mul.72, i32 addrspace(1)* %arrayidx.72, align 4, !tbaa !7
+  %185 = load i32, i32* %arrayidx1.73, align 4, !tbaa !7
+  %mul.73 = sub nsw i32 0, %185
+  store i32 %mul.73, i32 addrspace(1)* %arrayidx.73, align 4, !tbaa !7
+  %186 = load i32, i32* %arrayidx1.74, align 4, !tbaa !7
+  %mul.74 = sub nsw i32 0, %186
+  store i32 %mul.74, i32 addrspace(1)* %arrayidx.74, align 4, !tbaa !7
+  %187 = load i32, i32* %arrayidx1.75, align 4, !tbaa !7
+  %mul.75 = sub nsw i32 0, %187
+  store i32 %mul.75, i32 addrspace(1)* %arrayidx.75, align 4, !tbaa !7
+  %188 = load i32, i32* %arrayidx1.76, align 4, !tbaa !7
+  %mul.76 = sub nsw i32 0, %188
+  store i32 %mul.76, i32 addrspace(1)* %arrayidx.76, align 4, !tbaa !7
+  %189 = load i32, i32* %arrayidx1.77, align 4, !tbaa !7
+  %mul.77 = sub nsw i32 0, %189
+  store i32 %mul.77, i32 addrspace(1)* %arrayidx.77, align 4, !tbaa !7
+  %190 = load i32, i32* %arrayidx1.78, align 4, !tbaa !7
+  %mul.78 = sub nsw i32 0, %190
+  store i32 %mul.78, i32 addrspace(1)* %arrayidx.78, align 4, !tbaa !7
+  %191 = load i32, i32* %arrayidx1.79, align 4, !tbaa !7
+  %mul.79 = sub nsw i32 0, %191
+  store i32 %mul.79, i32 addrspace(1)* %arrayidx.79, align 4, !tbaa !7
+  %192 = load i32, i32* %arrayidx1.80, align 4, !tbaa !7
+  %mul.80 = sub nsw i32 0, %192
+  store i32 %mul.80, i32 addrspace(1)* %arrayidx.80, align 4, !tbaa !7
+  %193 = load i32, i32* %arrayidx1.81, align 4, !tbaa !7
+  %mul.81 = sub nsw i32 0, %193
+  store i32 %mul.81, i32 addrspace(1)* %arrayidx.81, align 4, !tbaa !7
+  %194 = load i32, i32* %arrayidx1.82, align 4, !tbaa !7
+  %mul.82 = sub nsw i32 0, %194
+  store i32 %mul.82, i32 addrspace(1)* %arrayidx.82, align 4, !tbaa !7
+  %195 = load i32, i32* %arrayidx1.83, align 4, !tbaa !7
+  %mul.83 = sub nsw i32 0, %195
+  store i32 %mul.83, i32 addrspace(1)* %arrayidx.83, align 4, !tbaa !7
+  %196 = load i32, i32* %arrayidx1.84, align 4, !tbaa !7
+  %mul.84 = sub nsw i32 0, %196
+  store i32 %mul.84, i32 addrspace(1)* %arrayidx.84, align 4, !tbaa !7
+  %197 = load i32, i32* %arrayidx1.85, align 4, !tbaa !7
+  %mul.85 = sub nsw i32 0, %197
+  store i32 %mul.85, i32 addrspace(1)* %arrayidx.85, align 4, !tbaa !7
+  %198 = load i32, i32* %arrayidx1.86, align 4, !tbaa !7
+  %mul.86 = sub nsw i32 0, %198
+  store i32 %mul.86, i32 addrspace(1)* %arrayidx.86, align 4, !tbaa !7
+  %199 = load i32, i32* %arrayidx1.87, align 4, !tbaa !7
+  %mul.87 = sub nsw i32 0, %199
+  store i32 %mul.87, i32 addrspace(1)* %arrayidx.87, align 4, !tbaa !7
+  %200 = load i32, i32* %arrayidx1.88, align 4, !tbaa !7
+  %mul.88 = sub nsw i32 0, %200
+  store i32 %mul.88, i32 addrspace(1)* %arrayidx.88, align 4, !tbaa !7
+  %201 = load i32, i32* %arrayidx1.89, align 4, !tbaa !7
+  %mul.89 = sub nsw i32 0, %201
+  store i32 %mul.89, i32 addrspace(1)* %arrayidx.89, align 4, !tbaa !7
+  %202 = load i32, i32* %arrayidx1.90, align 4, !tbaa !7
+  %mul.90 = sub nsw i32 0, %202
+  store i32 %mul.90, i32 addrspace(1)* %arrayidx.90, align 4, !tbaa !7
+  %203 = load i32, i32* %arrayidx1.91, align 4, !tbaa !7
+  %mul.91 = sub nsw i32 0, %203
+  store i32 %mul.91, i32 addrspace(1)* %arrayidx.91, align 4, !tbaa !7
+  %204 = load i32, i32* %arrayidx1.92, align 4, !tbaa !7
+  %mul.92 = sub nsw i32 0, %204
+  store i32 %mul.92, i32 addrspace(1)* %arrayidx.92, align 4, !tbaa !7
+  %205 = load i32, i32* %arrayidx1.93, align 4, !tbaa !7
+  %mul.93 = sub nsw i32 0, %205
+  store i32 %mul.93, i32 addrspace(1)* %arrayidx.93, align 4, !tbaa !7
+  %206 = load i32, i32* %arrayidx1.94, align 4, !tbaa !7
+  %mul.94 = sub nsw i32 0, %206
+  store i32 %mul.94, i32 addrspace(1)* %arrayidx.94, align 4, !tbaa !7
+  %207 = load i32, i32* %arrayidx1.95, align 4, !tbaa !7
+  %mul.95 = sub nsw i32 0, %207
+  store i32 %mul.95, i32 addrspace(1)* %arrayidx.95, align 4, !tbaa !7
+  %208 = load i32, i32* %arrayidx1.96, align 4, !tbaa !7
+  %mul.96 = sub nsw i32 0, %208
+  store i32 %mul.96, i32 addrspace(1)* %arrayidx.96, align 4, !tbaa !7
+  %209 = load i32, i32* %arrayidx1.97, align 4, !tbaa !7
+  %mul.97 = sub nsw i32 0, %209
+  store i32 %mul.97, i32 addrspace(1)* %arrayidx.97, align 4, !tbaa !7
+  %210 = load i32, i32* %arrayidx1.98, align 4, !tbaa !7
+  %mul.98 = sub nsw i32 0, %210
+  store i32 %mul.98, i32 addrspace(1)* %arrayidx.98, align 4, !tbaa !7
+  %211 = load i32, i32* %arrayidx1.99, align 4, !tbaa !7
+  %mul.99 = sub nsw i32 0, %211
+  store i32 %mul.99, i32 addrspace(1)* %arrayidx.99, align 4, !tbaa !7
+  %212 = load i32, i32* %arrayidx1.100, align 4, !tbaa !7
+  %mul.100 = sub nsw i32 0, %212
+  store i32 %mul.100, i32 addrspace(1)* %arrayidx.100, align 4, !tbaa !7
+  %213 = load i32, i32* %arrayidx1.101, align 4, !tbaa !7
+  %mul.101 = sub nsw i32 0, %213
+  store i32 %mul.101, i32 addrspace(1)* %arrayidx.101, align 4, !tbaa !7
+  %214 = load i32, i32* %arrayidx1.102, align 4, !tbaa !7
+  %mul.102 = sub nsw i32 0, %214
+  store i32 %mul.102, i32 addrspace(1)* %arrayidx.102, align 4, !tbaa !7
+  %215 = load i32, i32* %arrayidx1.103, align 4, !tbaa !7
+  %mul.103 = sub nsw i32 0, %215
+  store i32 %mul.103, i32 addrspace(1)* %arrayidx.103, align 4, !tbaa !7
+  %216 = load i32, i32* %arrayidx1.104, align 4, !tbaa !7
+  %mul.104 = sub nsw i32 0, %216
+  store i32 %mul.104, i32 addrspace(1)* %arrayidx.104, align 4, !tbaa !7
+  %217 = load i32, i32* %arrayidx1.105, align 4, !tbaa !7
+  %mul.105 = sub nsw i32 0, %217
+  store i32 %mul.105, i32 addrspace(1)* %arrayidx.105, align 4, !tbaa !7
+  %218 = load i32, i32* %arrayidx1.106, align 4, !tbaa !7
+  %mul.106 = sub nsw i32 0, %218
+  store i32 %mul.106, i32 addrspace(1)* %arrayidx.106, align 4, !tbaa !7
+  %219 = load i32, i32* %arrayidx1.107, align 4, !tbaa !7
+  %mul.107 = sub nsw i32 0, %219
+  store i32 %mul.107, i32 addrspace(1)* %arrayidx.107, align 4, !tbaa !7
+  %220 = load i32, i32* %arrayidx1.108, align 4, !tbaa !7
+  %mul.108 = sub nsw i32 0, %220
+  store i32 %mul.108, i32 addrspace(1)* %arrayidx.108, align 4, !tbaa !7
+  %221 = load i32, i32* %arrayidx1.109, align 4, !tbaa !7
+  %mul.109 = sub nsw i32 0, %221
+  store i32 %mul.109, i32 addrspace(1)* %arrayidx.109, align 4, !tbaa !7
+  %222 = load i32, i32* %arrayidx1.110, align 4, !tbaa !7
+  %mul.110 = sub nsw i32 0, %222
+  store i32 %mul.110, i32 addrspace(1)* %arrayidx.110, align 4, !tbaa !7
+  call void @llvm.lifetime.end(i64 444, i8* %0) #2
+  ret void
+}
+attributes #3 = { nounwind "amdgpu-num-active-waves-per-eu"="5,5" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="fiji" "target-features"="+16-bit-insts,+s-memrealtime" "unsafe-fp-math"="false" "use-soft-float"="false" }
+
+; Exactly 10 active waves per execution unit.
+; CHECK: SGPRBlocks: 1
+; CHECK: VGPRBlocks: 5
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 11
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 24
+define void @many_regs_exactly_10(i32 addrspace(1)* nocapture %A) #4 {
+entry:
+  %a = alloca [111 x i32], align 4
+  %0 = bitcast [111 x i32]* %a to i8*
+  call void @llvm.lifetime.start(i64 444, i8* %0) #2
+  %1 = load i32, i32 addrspace(1)* %A, align 4, !tbaa !7
+  %arrayidx1 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 0
+  store i32 %1, i32* %arrayidx1, align 4, !tbaa !7
+  %arrayidx.1 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 1
+  %2 = load i32, i32 addrspace(1)* %arrayidx.1, align 4, !tbaa !7
+  %arrayidx1.1 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 1
+  store i32 %2, i32* %arrayidx1.1, align 4, !tbaa !7
+  %arrayidx.2 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 2
+  %3 = load i32, i32 addrspace(1)* %arrayidx.2, align 4, !tbaa !7
+  %arrayidx1.2 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 2
+  store i32 %3, i32* %arrayidx1.2, align 4, !tbaa !7
+  %arrayidx.3 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 3
+  %4 = load i32, i32 addrspace(1)* %arrayidx.3, align 4, !tbaa !7
+  %arrayidx1.3 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 3
+  store i32 %4, i32* %arrayidx1.3, align 4, !tbaa !7
+  %arrayidx.4 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 4
+  %5 = load i32, i32 addrspace(1)* %arrayidx.4, align 4, !tbaa !7
+  %arrayidx1.4 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 4
+  store i32 %5, i32* %arrayidx1.4, align 4, !tbaa !7
+  %arrayidx.5 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 5
+  %6 = load i32, i32 addrspace(1)* %arrayidx.5, align 4, !tbaa !7
+  %arrayidx1.5 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 5
+  store i32 %6, i32* %arrayidx1.5, align 4, !tbaa !7
+  %arrayidx.6 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 6
+  %7 = load i32, i32 addrspace(1)* %arrayidx.6, align 4, !tbaa !7
+  %arrayidx1.6 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 6
+  store i32 %7, i32* %arrayidx1.6, align 4, !tbaa !7
+  %arrayidx.7 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 7
+  %8 = load i32, i32 addrspace(1)* %arrayidx.7, align 4, !tbaa !7
+  %arrayidx1.7 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 7
+  store i32 %8, i32* %arrayidx1.7, align 4, !tbaa !7
+  %arrayidx.8 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 8
+  %9 = load i32, i32 addrspace(1)* %arrayidx.8, align 4, !tbaa !7
+  %arrayidx1.8 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 8
+  store i32 %9, i32* %arrayidx1.8, align 4, !tbaa !7
+  %arrayidx.9 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 9
+  %10 = load i32, i32 addrspace(1)* %arrayidx.9, align 4, !tbaa !7
+  %arrayidx1.9 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 9
+  store i32 %10, i32* %arrayidx1.9, align 4, !tbaa !7
+  %arrayidx.10 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 10
+  %11 = load i32, i32 addrspace(1)* %arrayidx.10, align 4, !tbaa !7
+  %arrayidx1.10 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 10
+  store i32 %11, i32* %arrayidx1.10, align 4, !tbaa !7
+  %arrayidx.11 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 11
+  %12 = load i32, i32 addrspace(1)* %arrayidx.11, align 4, !tbaa !7
+  %arrayidx1.11 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 11
+  store i32 %12, i32* %arrayidx1.11, align 4, !tbaa !7
+  %arrayidx.12 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 12
+  %13 = load i32, i32 addrspace(1)* %arrayidx.12, align 4, !tbaa !7
+  %arrayidx1.12 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 12
+  store i32 %13, i32* %arrayidx1.12, align 4, !tbaa !7
+  %arrayidx.13 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 13
+  %14 = load i32, i32 addrspace(1)* %arrayidx.13, align 4, !tbaa !7
+  %arrayidx1.13 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 13
+  store i32 %14, i32* %arrayidx1.13, align 4, !tbaa !7
+  %arrayidx.14 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 14
+  %15 = load i32, i32 addrspace(1)* %arrayidx.14, align 4, !tbaa !7
+  %arrayidx1.14 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 14
+  store i32 %15, i32* %arrayidx1.14, align 4, !tbaa !7
+  %arrayidx.15 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 15
+  %16 = load i32, i32 addrspace(1)* %arrayidx.15, align 4, !tbaa !7
+  %arrayidx1.15 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 15
+  store i32 %16, i32* %arrayidx1.15, align 4, !tbaa !7
+  %arrayidx.16 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 16
+  %17 = load i32, i32 addrspace(1)* %arrayidx.16, align 4, !tbaa !7
+  %arrayidx1.16 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 16
+  store i32 %17, i32* %arrayidx1.16, align 4, !tbaa !7
+  %arrayidx.17 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 17
+  %18 = load i32, i32 addrspace(1)* %arrayidx.17, align 4, !tbaa !7
+  %arrayidx1.17 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 17
+  store i32 %18, i32* %arrayidx1.17, align 4, !tbaa !7
+  %arrayidx.18 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 18
+  %19 = load i32, i32 addrspace(1)* %arrayidx.18, align 4, !tbaa !7
+  %arrayidx1.18 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 18
+  store i32 %19, i32* %arrayidx1.18, align 4, !tbaa !7
+  %arrayidx.19 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 19
+  %20 = load i32, i32 addrspace(1)* %arrayidx.19, align 4, !tbaa !7
+  %arrayidx1.19 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 19
+  store i32 %20, i32* %arrayidx1.19, align 4, !tbaa !7
+  %arrayidx.20 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 20
+  %21 = load i32, i32 addrspace(1)* %arrayidx.20, align 4, !tbaa !7
+  %arrayidx1.20 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 20
+  store i32 %21, i32* %arrayidx1.20, align 4, !tbaa !7
+  %arrayidx.21 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 21
+  %22 = load i32, i32 addrspace(1)* %arrayidx.21, align 4, !tbaa !7
+  %arrayidx1.21 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 21
+  store i32 %22, i32* %arrayidx1.21, align 4, !tbaa !7
+  %arrayidx.22 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 22
+  %23 = load i32, i32 addrspace(1)* %arrayidx.22, align 4, !tbaa !7
+  %arrayidx1.22 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 22
+  store i32 %23, i32* %arrayidx1.22, align 4, !tbaa !7
+  %arrayidx.23 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 23
+  %24 = load i32, i32 addrspace(1)* %arrayidx.23, align 4, !tbaa !7
+  %arrayidx1.23 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 23
+  store i32 %24, i32* %arrayidx1.23, align 4, !tbaa !7
+  %arrayidx.24 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 24
+  %25 = load i32, i32 addrspace(1)* %arrayidx.24, align 4, !tbaa !7
+  %arrayidx1.24 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 24
+  store i32 %25, i32* %arrayidx1.24, align 4, !tbaa !7
+  %arrayidx.25 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 25
+  %26 = load i32, i32 addrspace(1)* %arrayidx.25, align 4, !tbaa !7
+  %arrayidx1.25 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 25
+  store i32 %26, i32* %arrayidx1.25, align 4, !tbaa !7
+  %arrayidx.26 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 26
+  %27 = load i32, i32 addrspace(1)* %arrayidx.26, align 4, !tbaa !7
+  %arrayidx1.26 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 26
+  store i32 %27, i32* %arrayidx1.26, align 4, !tbaa !7
+  %arrayidx.27 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 27
+  %28 = load i32, i32 addrspace(1)* %arrayidx.27, align 4, !tbaa !7
+  %arrayidx1.27 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 27
+  store i32 %28, i32* %arrayidx1.27, align 4, !tbaa !7
+  %arrayidx.28 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 28
+  %29 = load i32, i32 addrspace(1)* %arrayidx.28, align 4, !tbaa !7
+  %arrayidx1.28 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 28
+  store i32 %29, i32* %arrayidx1.28, align 4, !tbaa !7
+  %arrayidx.29 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 29
+  %30 = load i32, i32 addrspace(1)* %arrayidx.29, align 4, !tbaa !7
+  %arrayidx1.29 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 29
+  store i32 %30, i32* %arrayidx1.29, align 4, !tbaa !7
+  %arrayidx.30 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 30
+  %31 = load i32, i32 addrspace(1)* %arrayidx.30, align 4, !tbaa !7
+  %arrayidx1.30 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 30
+  store i32 %31, i32* %arrayidx1.30, align 4, !tbaa !7
+  %arrayidx.31 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 31
+  %32 = load i32, i32 addrspace(1)* %arrayidx.31, align 4, !tbaa !7
+  %arrayidx1.31 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 31
+  store i32 %32, i32* %arrayidx1.31, align 4, !tbaa !7
+  %arrayidx.32 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 32
+  %33 = load i32, i32 addrspace(1)* %arrayidx.32, align 4, !tbaa !7
+  %arrayidx1.32 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 32
+  store i32 %33, i32* %arrayidx1.32, align 4, !tbaa !7
+  %arrayidx.33 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 33
+  %34 = load i32, i32 addrspace(1)* %arrayidx.33, align 4, !tbaa !7
+  %arrayidx1.33 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 33
+  store i32 %34, i32* %arrayidx1.33, align 4, !tbaa !7
+  %arrayidx.34 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 34
+  %35 = load i32, i32 addrspace(1)* %arrayidx.34, align 4, !tbaa !7
+  %arrayidx1.34 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 34
+  store i32 %35, i32* %arrayidx1.34, align 4, !tbaa !7
+  %arrayidx.35 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 35
+  %36 = load i32, i32 addrspace(1)* %arrayidx.35, align 4, !tbaa !7
+  %arrayidx1.35 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 35
+  store i32 %36, i32* %arrayidx1.35, align 4, !tbaa !7
+  %arrayidx.36 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 36
+  %37 = load i32, i32 addrspace(1)* %arrayidx.36, align 4, !tbaa !7
+  %arrayidx1.36 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 36
+  store i32 %37, i32* %arrayidx1.36, align 4, !tbaa !7
+  %arrayidx.37 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 37
+  %38 = load i32, i32 addrspace(1)* %arrayidx.37, align 4, !tbaa !7
+  %arrayidx1.37 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 37
+  store i32 %38, i32* %arrayidx1.37, align 4, !tbaa !7
+  %arrayidx.38 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 38
+  %39 = load i32, i32 addrspace(1)* %arrayidx.38, align 4, !tbaa !7
+  %arrayidx1.38 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 38
+  store i32 %39, i32* %arrayidx1.38, align 4, !tbaa !7
+  %arrayidx.39 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 39
+  %40 = load i32, i32 addrspace(1)* %arrayidx.39, align 4, !tbaa !7
+  %arrayidx1.39 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 39
+  store i32 %40, i32* %arrayidx1.39, align 4, !tbaa !7
+  %arrayidx.40 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 40
+  %41 = load i32, i32 addrspace(1)* %arrayidx.40, align 4, !tbaa !7
+  %arrayidx1.40 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 40
+  store i32 %41, i32* %arrayidx1.40, align 4, !tbaa !7
+  %arrayidx.41 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 41
+  %42 = load i32, i32 addrspace(1)* %arrayidx.41, align 4, !tbaa !7
+  %arrayidx1.41 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 41
+  store i32 %42, i32* %arrayidx1.41, align 4, !tbaa !7
+  %arrayidx.42 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 42
+  %43 = load i32, i32 addrspace(1)* %arrayidx.42, align 4, !tbaa !7
+  %arrayidx1.42 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 42
+  store i32 %43, i32* %arrayidx1.42, align 4, !tbaa !7
+  %arrayidx.43 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 43
+  %44 = load i32, i32 addrspace(1)* %arrayidx.43, align 4, !tbaa !7
+  %arrayidx1.43 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 43
+  store i32 %44, i32* %arrayidx1.43, align 4, !tbaa !7
+  %arrayidx.44 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 44
+  %45 = load i32, i32 addrspace(1)* %arrayidx.44, align 4, !tbaa !7
+  %arrayidx1.44 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 44
+  store i32 %45, i32* %arrayidx1.44, align 4, !tbaa !7
+  %arrayidx.45 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 45
+  %46 = load i32, i32 addrspace(1)* %arrayidx.45, align 4, !tbaa !7
+  %arrayidx1.45 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 45
+  store i32 %46, i32* %arrayidx1.45, align 4, !tbaa !7
+  %arrayidx.46 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 46
+  %47 = load i32, i32 addrspace(1)* %arrayidx.46, align 4, !tbaa !7
+  %arrayidx1.46 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 46
+  store i32 %47, i32* %arrayidx1.46, align 4, !tbaa !7
+  %arrayidx.47 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 47
+  %48 = load i32, i32 addrspace(1)* %arrayidx.47, align 4, !tbaa !7
+  %arrayidx1.47 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 47
+  store i32 %48, i32* %arrayidx1.47, align 4, !tbaa !7
+  %arrayidx.48 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 48
+  %49 = load i32, i32 addrspace(1)* %arrayidx.48, align 4, !tbaa !7
+  %arrayidx1.48 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 48
+  store i32 %49, i32* %arrayidx1.48, align 4, !tbaa !7
+  %arrayidx.49 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 49
+  %50 = load i32, i32 addrspace(1)* %arrayidx.49, align 4, !tbaa !7
+  %arrayidx1.49 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 49
+  store i32 %50, i32* %arrayidx1.49, align 4, !tbaa !7
+  %arrayidx.50 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 50
+  %51 = load i32, i32 addrspace(1)* %arrayidx.50, align 4, !tbaa !7
+  %arrayidx1.50 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 50
+  store i32 %51, i32* %arrayidx1.50, align 4, !tbaa !7
+  %arrayidx.51 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 51
+  %52 = load i32, i32 addrspace(1)* %arrayidx.51, align 4, !tbaa !7
+  %arrayidx1.51 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 51
+  store i32 %52, i32* %arrayidx1.51, align 4, !tbaa !7
+  %arrayidx.52 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 52
+  %53 = load i32, i32 addrspace(1)* %arrayidx.52, align 4, !tbaa !7
+  %arrayidx1.52 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 52
+  store i32 %53, i32* %arrayidx1.52, align 4, !tbaa !7
+  %arrayidx.53 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 53
+  %54 = load i32, i32 addrspace(1)* %arrayidx.53, align 4, !tbaa !7
+  %arrayidx1.53 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 53
+  store i32 %54, i32* %arrayidx1.53, align 4, !tbaa !7
+  %arrayidx.54 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 54
+  %55 = load i32, i32 addrspace(1)* %arrayidx.54, align 4, !tbaa !7
+  %arrayidx1.54 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 54
+  store i32 %55, i32* %arrayidx1.54, align 4, !tbaa !7
+  %arrayidx.55 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 55
+  %56 = load i32, i32 addrspace(1)* %arrayidx.55, align 4, !tbaa !7
+  %arrayidx1.55 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 55
+  store i32 %56, i32* %arrayidx1.55, align 4, !tbaa !7
+  %arrayidx.56 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 56
+  %57 = load i32, i32 addrspace(1)* %arrayidx.56, align 4, !tbaa !7
+  %arrayidx1.56 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 56
+  store i32 %57, i32* %arrayidx1.56, align 4, !tbaa !7
+  %arrayidx.57 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 57
+  %58 = load i32, i32 addrspace(1)* %arrayidx.57, align 4, !tbaa !7
+  %arrayidx1.57 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 57
+  store i32 %58, i32* %arrayidx1.57, align 4, !tbaa !7
+  %arrayidx.58 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 58
+  %59 = load i32, i32 addrspace(1)* %arrayidx.58, align 4, !tbaa !7
+  %arrayidx1.58 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 58
+  store i32 %59, i32* %arrayidx1.58, align 4, !tbaa !7
+  %arrayidx.59 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 59
+  %60 = load i32, i32 addrspace(1)* %arrayidx.59, align 4, !tbaa !7
+  %arrayidx1.59 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 59
+  store i32 %60, i32* %arrayidx1.59, align 4, !tbaa !7
+  %arrayidx.60 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 60
+  %61 = load i32, i32 addrspace(1)* %arrayidx.60, align 4, !tbaa !7
+  %arrayidx1.60 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 60
+  store i32 %61, i32* %arrayidx1.60, align 4, !tbaa !7
+  %arrayidx.61 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 61
+  %62 = load i32, i32 addrspace(1)* %arrayidx.61, align 4, !tbaa !7
+  %arrayidx1.61 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 61
+  store i32 %62, i32* %arrayidx1.61, align 4, !tbaa !7
+  %arrayidx.62 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 62
+  %63 = load i32, i32 addrspace(1)* %arrayidx.62, align 4, !tbaa !7
+  %arrayidx1.62 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 62
+  store i32 %63, i32* %arrayidx1.62, align 4, !tbaa !7
+  %arrayidx.63 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 63
+  %64 = load i32, i32 addrspace(1)* %arrayidx.63, align 4, !tbaa !7
+  %arrayidx1.63 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 63
+  store i32 %64, i32* %arrayidx1.63, align 4, !tbaa !7
+  %arrayidx.64 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 64
+  %65 = load i32, i32 addrspace(1)* %arrayidx.64, align 4, !tbaa !7
+  %arrayidx1.64 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 64
+  store i32 %65, i32* %arrayidx1.64, align 4, !tbaa !7
+  %arrayidx.65 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 65
+  %66 = load i32, i32 addrspace(1)* %arrayidx.65, align 4, !tbaa !7
+  %arrayidx1.65 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 65
+  store i32 %66, i32* %arrayidx1.65, align 4, !tbaa !7
+  %arrayidx.66 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 66
+  %67 = load i32, i32 addrspace(1)* %arrayidx.66, align 4, !tbaa !7
+  %arrayidx1.66 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 66
+  store i32 %67, i32* %arrayidx1.66, align 4, !tbaa !7
+  %arrayidx.67 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 67
+  %68 = load i32, i32 addrspace(1)* %arrayidx.67, align 4, !tbaa !7
+  %arrayidx1.67 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 67
+  store i32 %68, i32* %arrayidx1.67, align 4, !tbaa !7
+  %arrayidx.68 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 68
+  %69 = load i32, i32 addrspace(1)* %arrayidx.68, align 4, !tbaa !7
+  %arrayidx1.68 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 68
+  store i32 %69, i32* %arrayidx1.68, align 4, !tbaa !7
+  %arrayidx.69 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 69
+  %70 = load i32, i32 addrspace(1)* %arrayidx.69, align 4, !tbaa !7
+  %arrayidx1.69 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 69
+  store i32 %70, i32* %arrayidx1.69, align 4, !tbaa !7
+  %arrayidx.70 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 70
+  %71 = load i32, i32 addrspace(1)* %arrayidx.70, align 4, !tbaa !7
+  %arrayidx1.70 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 70
+  store i32 %71, i32* %arrayidx1.70, align 4, !tbaa !7
+  %arrayidx.71 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 71
+  %72 = load i32, i32 addrspace(1)* %arrayidx.71, align 4, !tbaa !7
+  %arrayidx1.71 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 71
+  store i32 %72, i32* %arrayidx1.71, align 4, !tbaa !7
+  %arrayidx.72 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 72
+  %73 = load i32, i32 addrspace(1)* %arrayidx.72, align 4, !tbaa !7
+  %arrayidx1.72 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 72
+  store i32 %73, i32* %arrayidx1.72, align 4, !tbaa !7
+  %arrayidx.73 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 73
+  %74 = load i32, i32 addrspace(1)* %arrayidx.73, align 4, !tbaa !7
+  %arrayidx1.73 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 73
+  store i32 %74, i32* %arrayidx1.73, align 4, !tbaa !7
+  %arrayidx.74 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 74
+  %75 = load i32, i32 addrspace(1)* %arrayidx.74, align 4, !tbaa !7
+  %arrayidx1.74 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 74
+  store i32 %75, i32* %arrayidx1.74, align 4, !tbaa !7
+  %arrayidx.75 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 75
+  %76 = load i32, i32 addrspace(1)* %arrayidx.75, align 4, !tbaa !7
+  %arrayidx1.75 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 75
+  store i32 %76, i32* %arrayidx1.75, align 4, !tbaa !7
+  %arrayidx.76 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 76
+  %77 = load i32, i32 addrspace(1)* %arrayidx.76, align 4, !tbaa !7
+  %arrayidx1.76 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 76
+  store i32 %77, i32* %arrayidx1.76, align 4, !tbaa !7
+  %arrayidx.77 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 77
+  %78 = load i32, i32 addrspace(1)* %arrayidx.77, align 4, !tbaa !7
+  %arrayidx1.77 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 77
+  store i32 %78, i32* %arrayidx1.77, align 4, !tbaa !7
+  %arrayidx.78 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 78
+  %79 = load i32, i32 addrspace(1)* %arrayidx.78, align 4, !tbaa !7
+  %arrayidx1.78 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 78
+  store i32 %79, i32* %arrayidx1.78, align 4, !tbaa !7
+  %arrayidx.79 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 79
+  %80 = load i32, i32 addrspace(1)* %arrayidx.79, align 4, !tbaa !7
+  %arrayidx1.79 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 79
+  store i32 %80, i32* %arrayidx1.79, align 4, !tbaa !7
+  %arrayidx.80 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 80
+  %81 = load i32, i32 addrspace(1)* %arrayidx.80, align 4, !tbaa !7
+  %arrayidx1.80 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 80
+  store i32 %81, i32* %arrayidx1.80, align 4, !tbaa !7
+  %arrayidx.81 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 81
+  %82 = load i32, i32 addrspace(1)* %arrayidx.81, align 4, !tbaa !7
+  %arrayidx1.81 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 81
+  store i32 %82, i32* %arrayidx1.81, align 4, !tbaa !7
+  %arrayidx.82 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 82
+  %83 = load i32, i32 addrspace(1)* %arrayidx.82, align 4, !tbaa !7
+  %arrayidx1.82 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 82
+  store i32 %83, i32* %arrayidx1.82, align 4, !tbaa !7
+  %arrayidx.83 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 83
+  %84 = load i32, i32 addrspace(1)* %arrayidx.83, align 4, !tbaa !7
+  %arrayidx1.83 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 83
+  store i32 %84, i32* %arrayidx1.83, align 4, !tbaa !7
+  %arrayidx.84 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 84
+  %85 = load i32, i32 addrspace(1)* %arrayidx.84, align 4, !tbaa !7
+  %arrayidx1.84 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 84
+  store i32 %85, i32* %arrayidx1.84, align 4, !tbaa !7
+  %arrayidx.85 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 85
+  %86 = load i32, i32 addrspace(1)* %arrayidx.85, align 4, !tbaa !7
+  %arrayidx1.85 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 85
+  store i32 %86, i32* %arrayidx1.85, align 4, !tbaa !7
+  %arrayidx.86 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 86
+  %87 = load i32, i32 addrspace(1)* %arrayidx.86, align 4, !tbaa !7
+  %arrayidx1.86 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 86
+  store i32 %87, i32* %arrayidx1.86, align 4, !tbaa !7
+  %arrayidx.87 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 87
+  %88 = load i32, i32 addrspace(1)* %arrayidx.87, align 4, !tbaa !7
+  %arrayidx1.87 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 87
+  store i32 %88, i32* %arrayidx1.87, align 4, !tbaa !7
+  %arrayidx.88 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 88
+  %89 = load i32, i32 addrspace(1)* %arrayidx.88, align 4, !tbaa !7
+  %arrayidx1.88 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 88
+  store i32 %89, i32* %arrayidx1.88, align 4, !tbaa !7
+  %arrayidx.89 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 89
+  %90 = load i32, i32 addrspace(1)* %arrayidx.89, align 4, !tbaa !7
+  %arrayidx1.89 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 89
+  store i32 %90, i32* %arrayidx1.89, align 4, !tbaa !7
+  %arrayidx.90 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 90
+  %91 = load i32, i32 addrspace(1)* %arrayidx.90, align 4, !tbaa !7
+  %arrayidx1.90 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 90
+  store i32 %91, i32* %arrayidx1.90, align 4, !tbaa !7
+  %arrayidx.91 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 91
+  %92 = load i32, i32 addrspace(1)* %arrayidx.91, align 4, !tbaa !7
+  %arrayidx1.91 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 91
+  store i32 %92, i32* %arrayidx1.91, align 4, !tbaa !7
+  %arrayidx.92 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 92
+  %93 = load i32, i32 addrspace(1)* %arrayidx.92, align 4, !tbaa !7
+  %arrayidx1.92 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 92
+  store i32 %93, i32* %arrayidx1.92, align 4, !tbaa !7
+  %arrayidx.93 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 93
+  %94 = load i32, i32 addrspace(1)* %arrayidx.93, align 4, !tbaa !7
+  %arrayidx1.93 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 93
+  store i32 %94, i32* %arrayidx1.93, align 4, !tbaa !7
+  %arrayidx.94 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 94
+  %95 = load i32, i32 addrspace(1)* %arrayidx.94, align 4, !tbaa !7
+  %arrayidx1.94 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 94
+  store i32 %95, i32* %arrayidx1.94, align 4, !tbaa !7
+  %arrayidx.95 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 95
+  %96 = load i32, i32 addrspace(1)* %arrayidx.95, align 4, !tbaa !7
+  %arrayidx1.95 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 95
+  store i32 %96, i32* %arrayidx1.95, align 4, !tbaa !7
+  %arrayidx.96 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 96
+  %97 = load i32, i32 addrspace(1)* %arrayidx.96, align 4, !tbaa !7
+  %arrayidx1.96 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 96
+  store i32 %97, i32* %arrayidx1.96, align 4, !tbaa !7
+  %arrayidx.97 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 97
+  %98 = load i32, i32 addrspace(1)* %arrayidx.97, align 4, !tbaa !7
+  %arrayidx1.97 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 97
+  store i32 %98, i32* %arrayidx1.97, align 4, !tbaa !7
+  %arrayidx.98 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 98
+  %99 = load i32, i32 addrspace(1)* %arrayidx.98, align 4, !tbaa !7
+  %arrayidx1.98 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 98
+  store i32 %99, i32* %arrayidx1.98, align 4, !tbaa !7
+  %arrayidx.99 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 99
+  %100 = load i32, i32 addrspace(1)* %arrayidx.99, align 4, !tbaa !7
+  %arrayidx1.99 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 99
+  store i32 %100, i32* %arrayidx1.99, align 4, !tbaa !7
+  %arrayidx.100 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 100
+  %101 = load i32, i32 addrspace(1)* %arrayidx.100, align 4, !tbaa !7
+  %arrayidx1.100 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 100
+  store i32 %101, i32* %arrayidx1.100, align 4, !tbaa !7
+  %arrayidx.101 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 101
+  %102 = load i32, i32 addrspace(1)* %arrayidx.101, align 4, !tbaa !7
+  %arrayidx1.101 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 101
+  store i32 %102, i32* %arrayidx1.101, align 4, !tbaa !7
+  %arrayidx.102 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 102
+  %103 = load i32, i32 addrspace(1)* %arrayidx.102, align 4, !tbaa !7
+  %arrayidx1.102 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 102
+  store i32 %103, i32* %arrayidx1.102, align 4, !tbaa !7
+  %arrayidx.103 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 103
+  %104 = load i32, i32 addrspace(1)* %arrayidx.103, align 4, !tbaa !7
+  %arrayidx1.103 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 103
+  store i32 %104, i32* %arrayidx1.103, align 4, !tbaa !7
+  %arrayidx.104 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 104
+  %105 = load i32, i32 addrspace(1)* %arrayidx.104, align 4, !tbaa !7
+  %arrayidx1.104 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 104
+  store i32 %105, i32* %arrayidx1.104, align 4, !tbaa !7
+  %arrayidx.105 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 105
+  %106 = load i32, i32 addrspace(1)* %arrayidx.105, align 4, !tbaa !7
+  %arrayidx1.105 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 105
+  store i32 %106, i32* %arrayidx1.105, align 4, !tbaa !7
+  %arrayidx.106 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 106
+  %107 = load i32, i32 addrspace(1)* %arrayidx.106, align 4, !tbaa !7
+  %arrayidx1.106 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 106
+  store i32 %107, i32* %arrayidx1.106, align 4, !tbaa !7
+  %arrayidx.107 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 107
+  %108 = load i32, i32 addrspace(1)* %arrayidx.107, align 4, !tbaa !7
+  %arrayidx1.107 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 107
+  store i32 %108, i32* %arrayidx1.107, align 4, !tbaa !7
+  %arrayidx.108 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 108
+  %109 = load i32, i32 addrspace(1)* %arrayidx.108, align 4, !tbaa !7
+  %arrayidx1.108 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 108
+  store i32 %109, i32* %arrayidx1.108, align 4, !tbaa !7
+  %arrayidx.109 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 109
+  %110 = load i32, i32 addrspace(1)* %arrayidx.109, align 4, !tbaa !7
+  %arrayidx1.109 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 109
+  store i32 %110, i32* %arrayidx1.109, align 4, !tbaa !7
+  %arrayidx.110 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 110
+  %111 = load i32, i32 addrspace(1)* %arrayidx.110, align 4, !tbaa !7
+  %arrayidx1.110 = getelementptr inbounds [111 x i32], [111 x i32]* %a, i32 0, i32 110
+  store i32 %111, i32* %arrayidx1.110, align 4, !tbaa !7
+  %112 = load i32, i32* %arrayidx1, align 4, !tbaa !7
+  %mul = sub nsw i32 0, %112
+  store i32 %mul, i32 addrspace(1)* %A, align 4, !tbaa !7
+  %113 = load i32, i32* %arrayidx1.1, align 4, !tbaa !7
+  %mul.1 = sub nsw i32 0, %113
+  store i32 %mul.1, i32 addrspace(1)* %arrayidx.1, align 4, !tbaa !7
+  %114 = load i32, i32* %arrayidx1.2, align 4, !tbaa !7
+  %mul.2 = sub nsw i32 0, %114
+  store i32 %mul.2, i32 addrspace(1)* %arrayidx.2, align 4, !tbaa !7
+  %115 = load i32, i32* %arrayidx1.3, align 4, !tbaa !7
+  %mul.3 = sub nsw i32 0, %115
+  store i32 %mul.3, i32 addrspace(1)* %arrayidx.3, align 4, !tbaa !7
+  %116 = load i32, i32* %arrayidx1.4, align 4, !tbaa !7
+  %mul.4 = sub nsw i32 0, %116
+  store i32 %mul.4, i32 addrspace(1)* %arrayidx.4, align 4, !tbaa !7
+  %117 = load i32, i32* %arrayidx1.5, align 4, !tbaa !7
+  %mul.5 = sub nsw i32 0, %117
+  store i32 %mul.5, i32 addrspace(1)* %arrayidx.5, align 4, !tbaa !7
+  %118 = load i32, i32* %arrayidx1.6, align 4, !tbaa !7
+  %mul.6 = sub nsw i32 0, %118
+  store i32 %mul.6, i32 addrspace(1)* %arrayidx.6, align 4, !tbaa !7
+  %119 = load i32, i32* %arrayidx1.7, align 4, !tbaa !7
+  %mul.7 = sub nsw i32 0, %119
+  store i32 %mul.7, i32 addrspace(1)* %arrayidx.7, align 4, !tbaa !7
+  %120 = load i32, i32* %arrayidx1.8, align 4, !tbaa !7
+  %mul.8 = sub nsw i32 0, %120
+  store i32 %mul.8, i32 addrspace(1)* %arrayidx.8, align 4, !tbaa !7
+  %121 = load i32, i32* %arrayidx1.9, align 4, !tbaa !7
+  %mul.9 = sub nsw i32 0, %121
+  store i32 %mul.9, i32 addrspace(1)* %arrayidx.9, align 4, !tbaa !7
+  %122 = load i32, i32* %arrayidx1.10, align 4, !tbaa !7
+  %mul.10 = sub nsw i32 0, %122
+  store i32 %mul.10, i32 addrspace(1)* %arrayidx.10, align 4, !tbaa !7
+  %123 = load i32, i32* %arrayidx1.11, align 4, !tbaa !7
+  %mul.11 = sub nsw i32 0, %123
+  store i32 %mul.11, i32 addrspace(1)* %arrayidx.11, align 4, !tbaa !7
+  %124 = load i32, i32* %arrayidx1.12, align 4, !tbaa !7
+  %mul.12 = sub nsw i32 0, %124
+  store i32 %mul.12, i32 addrspace(1)* %arrayidx.12, align 4, !tbaa !7
+  %125 = load i32, i32* %arrayidx1.13, align 4, !tbaa !7
+  %mul.13 = sub nsw i32 0, %125
+  store i32 %mul.13, i32 addrspace(1)* %arrayidx.13, align 4, !tbaa !7
+  %126 = load i32, i32* %arrayidx1.14, align 4, !tbaa !7
+  %mul.14 = sub nsw i32 0, %126
+  store i32 %mul.14, i32 addrspace(1)* %arrayidx.14, align 4, !tbaa !7
+  %127 = load i32, i32* %arrayidx1.15, align 4, !tbaa !7
+  %mul.15 = sub nsw i32 0, %127
+  store i32 %mul.15, i32 addrspace(1)* %arrayidx.15, align 4, !tbaa !7
+  %128 = load i32, i32* %arrayidx1.16, align 4, !tbaa !7
+  %mul.16 = sub nsw i32 0, %128
+  store i32 %mul.16, i32 addrspace(1)* %arrayidx.16, align 4, !tbaa !7
+  %129 = load i32, i32* %arrayidx1.17, align 4, !tbaa !7
+  %mul.17 = sub nsw i32 0, %129
+  store i32 %mul.17, i32 addrspace(1)* %arrayidx.17, align 4, !tbaa !7
+  %130 = load i32, i32* %arrayidx1.18, align 4, !tbaa !7
+  %mul.18 = sub nsw i32 0, %130
+  store i32 %mul.18, i32 addrspace(1)* %arrayidx.18, align 4, !tbaa !7
+  %131 = load i32, i32* %arrayidx1.19, align 4, !tbaa !7
+  %mul.19 = sub nsw i32 0, %131
+  store i32 %mul.19, i32 addrspace(1)* %arrayidx.19, align 4, !tbaa !7
+  %132 = load i32, i32* %arrayidx1.20, align 4, !tbaa !7
+  %mul.20 = sub nsw i32 0, %132
+  store i32 %mul.20, i32 addrspace(1)* %arrayidx.20, align 4, !tbaa !7
+  %133 = load i32, i32* %arrayidx1.21, align 4, !tbaa !7
+  %mul.21 = sub nsw i32 0, %133
+  store i32 %mul.21, i32 addrspace(1)* %arrayidx.21, align 4, !tbaa !7
+  %134 = load i32, i32* %arrayidx1.22, align 4, !tbaa !7
+  %mul.22 = sub nsw i32 0, %134
+  store i32 %mul.22, i32 addrspace(1)* %arrayidx.22, align 4, !tbaa !7
+  %135 = load i32, i32* %arrayidx1.23, align 4, !tbaa !7
+  %mul.23 = sub nsw i32 0, %135
+  store i32 %mul.23, i32 addrspace(1)* %arrayidx.23, align 4, !tbaa !7
+  %136 = load i32, i32* %arrayidx1.24, align 4, !tbaa !7
+  %mul.24 = sub nsw i32 0, %136
+  store i32 %mul.24, i32 addrspace(1)* %arrayidx.24, align 4, !tbaa !7
+  %137 = load i32, i32* %arrayidx1.25, align 4, !tbaa !7
+  %mul.25 = sub nsw i32 0, %137
+  store i32 %mul.25, i32 addrspace(1)* %arrayidx.25, align 4, !tbaa !7
+  %138 = load i32, i32* %arrayidx1.26, align 4, !tbaa !7
+  %mul.26 = sub nsw i32 0, %138
+  store i32 %mul.26, i32 addrspace(1)* %arrayidx.26, align 4, !tbaa !7
+  %139 = load i32, i32* %arrayidx1.27, align 4, !tbaa !7
+  %mul.27 = sub nsw i32 0, %139
+  store i32 %mul.27, i32 addrspace(1)* %arrayidx.27, align 4, !tbaa !7
+  %140 = load i32, i32* %arrayidx1.28, align 4, !tbaa !7
+  %mul.28 = sub nsw i32 0, %140
+  store i32 %mul.28, i32 addrspace(1)* %arrayidx.28, align 4, !tbaa !7
+  %141 = load i32, i32* %arrayidx1.29, align 4, !tbaa !7
+  %mul.29 = sub nsw i32 0, %141
+  store i32 %mul.29, i32 addrspace(1)* %arrayidx.29, align 4, !tbaa !7
+  %142 = load i32, i32* %arrayidx1.30, align 4, !tbaa !7
+  %mul.30 = sub nsw i32 0, %142
+  store i32 %mul.30, i32 addrspace(1)* %arrayidx.30, align 4, !tbaa !7
+  %143 = load i32, i32* %arrayidx1.31, align 4, !tbaa !7
+  %mul.31 = sub nsw i32 0, %143
+  store i32 %mul.31, i32 addrspace(1)* %arrayidx.31, align 4, !tbaa !7
+  %144 = load i32, i32* %arrayidx1.32, align 4, !tbaa !7
+  %mul.32 = sub nsw i32 0, %144
+  store i32 %mul.32, i32 addrspace(1)* %arrayidx.32, align 4, !tbaa !7
+  %145 = load i32, i32* %arrayidx1.33, align 4, !tbaa !7
+  %mul.33 = sub nsw i32 0, %145
+  store i32 %mul.33, i32 addrspace(1)* %arrayidx.33, align 4, !tbaa !7
+  %146 = load i32, i32* %arrayidx1.34, align 4, !tbaa !7
+  %mul.34 = sub nsw i32 0, %146
+  store i32 %mul.34, i32 addrspace(1)* %arrayidx.34, align 4, !tbaa !7
+  %147 = load i32, i32* %arrayidx1.35, align 4, !tbaa !7
+  %mul.35 = sub nsw i32 0, %147
+  store i32 %mul.35, i32 addrspace(1)* %arrayidx.35, align 4, !tbaa !7
+  %148 = load i32, i32* %arrayidx1.36, align 4, !tbaa !7
+  %mul.36 = sub nsw i32 0, %148
+  store i32 %mul.36, i32 addrspace(1)* %arrayidx.36, align 4, !tbaa !7
+  %149 = load i32, i32* %arrayidx1.37, align 4, !tbaa !7
+  %mul.37 = sub nsw i32 0, %149
+  store i32 %mul.37, i32 addrspace(1)* %arrayidx.37, align 4, !tbaa !7
+  %150 = load i32, i32* %arrayidx1.38, align 4, !tbaa !7
+  %mul.38 = sub nsw i32 0, %150
+  store i32 %mul.38, i32 addrspace(1)* %arrayidx.38, align 4, !tbaa !7
+  %151 = load i32, i32* %arrayidx1.39, align 4, !tbaa !7
+  %mul.39 = sub nsw i32 0, %151
+  store i32 %mul.39, i32 addrspace(1)* %arrayidx.39, align 4, !tbaa !7
+  %152 = load i32, i32* %arrayidx1.40, align 4, !tbaa !7
+  %mul.40 = sub nsw i32 0, %152
+  store i32 %mul.40, i32 addrspace(1)* %arrayidx.40, align 4, !tbaa !7
+  %153 = load i32, i32* %arrayidx1.41, align 4, !tbaa !7
+  %mul.41 = sub nsw i32 0, %153
+  store i32 %mul.41, i32 addrspace(1)* %arrayidx.41, align 4, !tbaa !7
+  %154 = load i32, i32* %arrayidx1.42, align 4, !tbaa !7
+  %mul.42 = sub nsw i32 0, %154
+  store i32 %mul.42, i32 addrspace(1)* %arrayidx.42, align 4, !tbaa !7
+  %155 = load i32, i32* %arrayidx1.43, align 4, !tbaa !7
+  %mul.43 = sub nsw i32 0, %155
+  store i32 %mul.43, i32 addrspace(1)* %arrayidx.43, align 4, !tbaa !7
+  %156 = load i32, i32* %arrayidx1.44, align 4, !tbaa !7
+  %mul.44 = sub nsw i32 0, %156
+  store i32 %mul.44, i32 addrspace(1)* %arrayidx.44, align 4, !tbaa !7
+  %157 = load i32, i32* %arrayidx1.45, align 4, !tbaa !7
+  %mul.45 = sub nsw i32 0, %157
+  store i32 %mul.45, i32 addrspace(1)* %arrayidx.45, align 4, !tbaa !7
+  %158 = load i32, i32* %arrayidx1.46, align 4, !tbaa !7
+  %mul.46 = sub nsw i32 0, %158
+  store i32 %mul.46, i32 addrspace(1)* %arrayidx.46, align 4, !tbaa !7
+  %159 = load i32, i32* %arrayidx1.47, align 4, !tbaa !7
+  %mul.47 = sub nsw i32 0, %159
+  store i32 %mul.47, i32 addrspace(1)* %arrayidx.47, align 4, !tbaa !7
+  %160 = load i32, i32* %arrayidx1.48, align 4, !tbaa !7
+  %mul.48 = sub nsw i32 0, %160
+  store i32 %mul.48, i32 addrspace(1)* %arrayidx.48, align 4, !tbaa !7
+  %161 = load i32, i32* %arrayidx1.49, align 4, !tbaa !7
+  %mul.49 = sub nsw i32 0, %161
+  store i32 %mul.49, i32 addrspace(1)* %arrayidx.49, align 4, !tbaa !7
+  %162 = load i32, i32* %arrayidx1.50, align 4, !tbaa !7
+  %mul.50 = sub nsw i32 0, %162
+  store i32 %mul.50, i32 addrspace(1)* %arrayidx.50, align 4, !tbaa !7
+  %163 = load i32, i32* %arrayidx1.51, align 4, !tbaa !7
+  %mul.51 = sub nsw i32 0, %163
+  store i32 %mul.51, i32 addrspace(1)* %arrayidx.51, align 4, !tbaa !7
+  %164 = load i32, i32* %arrayidx1.52, align 4, !tbaa !7
+  %mul.52 = sub nsw i32 0, %164
+  store i32 %mul.52, i32 addrspace(1)* %arrayidx.52, align 4, !tbaa !7
+  %165 = load i32, i32* %arrayidx1.53, align 4, !tbaa !7
+  %mul.53 = sub nsw i32 0, %165
+  store i32 %mul.53, i32 addrspace(1)* %arrayidx.53, align 4, !tbaa !7
+  %166 = load i32, i32* %arrayidx1.54, align 4, !tbaa !7
+  %mul.54 = sub nsw i32 0, %166
+  store i32 %mul.54, i32 addrspace(1)* %arrayidx.54, align 4, !tbaa !7
+  %167 = load i32, i32* %arrayidx1.55, align 4, !tbaa !7
+  %mul.55 = sub nsw i32 0, %167
+  store i32 %mul.55, i32 addrspace(1)* %arrayidx.55, align 4, !tbaa !7
+  %168 = load i32, i32* %arrayidx1.56, align 4, !tbaa !7
+  %mul.56 = sub nsw i32 0, %168
+  store i32 %mul.56, i32 addrspace(1)* %arrayidx.56, align 4, !tbaa !7
+  %169 = load i32, i32* %arrayidx1.57, align 4, !tbaa !7
+  %mul.57 = sub nsw i32 0, %169
+  store i32 %mul.57, i32 addrspace(1)* %arrayidx.57, align 4, !tbaa !7
+  %170 = load i32, i32* %arrayidx1.58, align 4, !tbaa !7
+  %mul.58 = sub nsw i32 0, %170
+  store i32 %mul.58, i32 addrspace(1)* %arrayidx.58, align 4, !tbaa !7
+  %171 = load i32, i32* %arrayidx1.59, align 4, !tbaa !7
+  %mul.59 = sub nsw i32 0, %171
+  store i32 %mul.59, i32 addrspace(1)* %arrayidx.59, align 4, !tbaa !7
+  %172 = load i32, i32* %arrayidx1.60, align 4, !tbaa !7
+  %mul.60 = sub nsw i32 0, %172
+  store i32 %mul.60, i32 addrspace(1)* %arrayidx.60, align 4, !tbaa !7
+  %173 = load i32, i32* %arrayidx1.61, align 4, !tbaa !7
+  %mul.61 = sub nsw i32 0, %173
+  store i32 %mul.61, i32 addrspace(1)* %arrayidx.61, align 4, !tbaa !7
+  %174 = load i32, i32* %arrayidx1.62, align 4, !tbaa !7
+  %mul.62 = sub nsw i32 0, %174
+  store i32 %mul.62, i32 addrspace(1)* %arrayidx.62, align 4, !tbaa !7
+  %175 = load i32, i32* %arrayidx1.63, align 4, !tbaa !7
+  %mul.63 = sub nsw i32 0, %175
+  store i32 %mul.63, i32 addrspace(1)* %arrayidx.63, align 4, !tbaa !7
+  %176 = load i32, i32* %arrayidx1.64, align 4, !tbaa !7
+  %mul.64 = sub nsw i32 0, %176
+  store i32 %mul.64, i32 addrspace(1)* %arrayidx.64, align 4, !tbaa !7
+  %177 = load i32, i32* %arrayidx1.65, align 4, !tbaa !7
+  %mul.65 = sub nsw i32 0, %177
+  store i32 %mul.65, i32 addrspace(1)* %arrayidx.65, align 4, !tbaa !7
+  %178 = load i32, i32* %arrayidx1.66, align 4, !tbaa !7
+  %mul.66 = sub nsw i32 0, %178
+  store i32 %mul.66, i32 addrspace(1)* %arrayidx.66, align 4, !tbaa !7
+  %179 = load i32, i32* %arrayidx1.67, align 4, !tbaa !7
+  %mul.67 = sub nsw i32 0, %179
+  store i32 %mul.67, i32 addrspace(1)* %arrayidx.67, align 4, !tbaa !7
+  %180 = load i32, i32* %arrayidx1.68, align 4, !tbaa !7
+  %mul.68 = sub nsw i32 0, %180
+  store i32 %mul.68, i32 addrspace(1)* %arrayidx.68, align 4, !tbaa !7
+  %181 = load i32, i32* %arrayidx1.69, align 4, !tbaa !7
+  %mul.69 = sub nsw i32 0, %181
+  store i32 %mul.69, i32 addrspace(1)* %arrayidx.69, align 4, !tbaa !7
+  %182 = load i32, i32* %arrayidx1.70, align 4, !tbaa !7
+  %mul.70 = sub nsw i32 0, %182
+  store i32 %mul.70, i32 addrspace(1)* %arrayidx.70, align 4, !tbaa !7
+  %183 = load i32, i32* %arrayidx1.71, align 4, !tbaa !7
+  %mul.71 = sub nsw i32 0, %183
+  store i32 %mul.71, i32 addrspace(1)* %arrayidx.71, align 4, !tbaa !7
+  %184 = load i32, i32* %arrayidx1.72, align 4, !tbaa !7
+  %mul.72 = sub nsw i32 0, %184
+  store i32 %mul.72, i32 addrspace(1)* %arrayidx.72, align 4, !tbaa !7
+  %185 = load i32, i32* %arrayidx1.73, align 4, !tbaa !7
+  %mul.73 = sub nsw i32 0, %185
+  store i32 %mul.73, i32 addrspace(1)* %arrayidx.73, align 4, !tbaa !7
+  %186 = load i32, i32* %arrayidx1.74, align 4, !tbaa !7
+  %mul.74 = sub nsw i32 0, %186
+  store i32 %mul.74, i32 addrspace(1)* %arrayidx.74, align 4, !tbaa !7
+  %187 = load i32, i32* %arrayidx1.75, align 4, !tbaa !7
+  %mul.75 = sub nsw i32 0, %187
+  store i32 %mul.75, i32 addrspace(1)* %arrayidx.75, align 4, !tbaa !7
+  %188 = load i32, i32* %arrayidx1.76, align 4, !tbaa !7
+  %mul.76 = sub nsw i32 0, %188
+  store i32 %mul.76, i32 addrspace(1)* %arrayidx.76, align 4, !tbaa !7
+  %189 = load i32, i32* %arrayidx1.77, align 4, !tbaa !7
+  %mul.77 = sub nsw i32 0, %189
+  store i32 %mul.77, i32 addrspace(1)* %arrayidx.77, align 4, !tbaa !7
+  %190 = load i32, i32* %arrayidx1.78, align 4, !tbaa !7
+  %mul.78 = sub nsw i32 0, %190
+  store i32 %mul.78, i32 addrspace(1)* %arrayidx.78, align 4, !tbaa !7
+  %191 = load i32, i32* %arrayidx1.79, align 4, !tbaa !7
+  %mul.79 = sub nsw i32 0, %191
+  store i32 %mul.79, i32 addrspace(1)* %arrayidx.79, align 4, !tbaa !7
+  %192 = load i32, i32* %arrayidx1.80, align 4, !tbaa !7
+  %mul.80 = sub nsw i32 0, %192
+  store i32 %mul.80, i32 addrspace(1)* %arrayidx.80, align 4, !tbaa !7
+  %193 = load i32, i32* %arrayidx1.81, align 4, !tbaa !7
+  %mul.81 = sub nsw i32 0, %193
+  store i32 %mul.81, i32 addrspace(1)* %arrayidx.81, align 4, !tbaa !7
+  %194 = load i32, i32* %arrayidx1.82, align 4, !tbaa !7
+  %mul.82 = sub nsw i32 0, %194
+  store i32 %mul.82, i32 addrspace(1)* %arrayidx.82, align 4, !tbaa !7
+  %195 = load i32, i32* %arrayidx1.83, align 4, !tbaa !7
+  %mul.83 = sub nsw i32 0, %195
+  store i32 %mul.83, i32 addrspace(1)* %arrayidx.83, align 4, !tbaa !7
+  %196 = load i32, i32* %arrayidx1.84, align 4, !tbaa !7
+  %mul.84 = sub nsw i32 0, %196
+  store i32 %mul.84, i32 addrspace(1)* %arrayidx.84, align 4, !tbaa !7
+  %197 = load i32, i32* %arrayidx1.85, align 4, !tbaa !7
+  %mul.85 = sub nsw i32 0, %197
+  store i32 %mul.85, i32 addrspace(1)* %arrayidx.85, align 4, !tbaa !7
+  %198 = load i32, i32* %arrayidx1.86, align 4, !tbaa !7
+  %mul.86 = sub nsw i32 0, %198
+  store i32 %mul.86, i32 addrspace(1)* %arrayidx.86, align 4, !tbaa !7
+  %199 = load i32, i32* %arrayidx1.87, align 4, !tbaa !7
+  %mul.87 = sub nsw i32 0, %199
+  store i32 %mul.87, i32 addrspace(1)* %arrayidx.87, align 4, !tbaa !7
+  %200 = load i32, i32* %arrayidx1.88, align 4, !tbaa !7
+  %mul.88 = sub nsw i32 0, %200
+  store i32 %mul.88, i32 addrspace(1)* %arrayidx.88, align 4, !tbaa !7
+  %201 = load i32, i32* %arrayidx1.89, align 4, !tbaa !7
+  %mul.89 = sub nsw i32 0, %201
+  store i32 %mul.89, i32 addrspace(1)* %arrayidx.89, align 4, !tbaa !7
+  %202 = load i32, i32* %arrayidx1.90, align 4, !tbaa !7
+  %mul.90 = sub nsw i32 0, %202
+  store i32 %mul.90, i32 addrspace(1)* %arrayidx.90, align 4, !tbaa !7
+  %203 = load i32, i32* %arrayidx1.91, align 4, !tbaa !7
+  %mul.91 = sub nsw i32 0, %203
+  store i32 %mul.91, i32 addrspace(1)* %arrayidx.91, align 4, !tbaa !7
+  %204 = load i32, i32* %arrayidx1.92, align 4, !tbaa !7
+  %mul.92 = sub nsw i32 0, %204
+  store i32 %mul.92, i32 addrspace(1)* %arrayidx.92, align 4, !tbaa !7
+  %205 = load i32, i32* %arrayidx1.93, align 4, !tbaa !7
+  %mul.93 = sub nsw i32 0, %205
+  store i32 %mul.93, i32 addrspace(1)* %arrayidx.93, align 4, !tbaa !7
+  %206 = load i32, i32* %arrayidx1.94, align 4, !tbaa !7
+  %mul.94 = sub nsw i32 0, %206
+  store i32 %mul.94, i32 addrspace(1)* %arrayidx.94, align 4, !tbaa !7
+  %207 = load i32, i32* %arrayidx1.95, align 4, !tbaa !7
+  %mul.95 = sub nsw i32 0, %207
+  store i32 %mul.95, i32 addrspace(1)* %arrayidx.95, align 4, !tbaa !7
+  %208 = load i32, i32* %arrayidx1.96, align 4, !tbaa !7
+  %mul.96 = sub nsw i32 0, %208
+  store i32 %mul.96, i32 addrspace(1)* %arrayidx.96, align 4, !tbaa !7
+  %209 = load i32, i32* %arrayidx1.97, align 4, !tbaa !7
+  %mul.97 = sub nsw i32 0, %209
+  store i32 %mul.97, i32 addrspace(1)* %arrayidx.97, align 4, !tbaa !7
+  %210 = load i32, i32* %arrayidx1.98, align 4, !tbaa !7
+  %mul.98 = sub nsw i32 0, %210
+  store i32 %mul.98, i32 addrspace(1)* %arrayidx.98, align 4, !tbaa !7
+  %211 = load i32, i32* %arrayidx1.99, align 4, !tbaa !7
+  %mul.99 = sub nsw i32 0, %211
+  store i32 %mul.99, i32 addrspace(1)* %arrayidx.99, align 4, !tbaa !7
+  %212 = load i32, i32* %arrayidx1.100, align 4, !tbaa !7
+  %mul.100 = sub nsw i32 0, %212
+  store i32 %mul.100, i32 addrspace(1)* %arrayidx.100, align 4, !tbaa !7
+  %213 = load i32, i32* %arrayidx1.101, align 4, !tbaa !7
+  %mul.101 = sub nsw i32 0, %213
+  store i32 %mul.101, i32 addrspace(1)* %arrayidx.101, align 4, !tbaa !7
+  %214 = load i32, i32* %arrayidx1.102, align 4, !tbaa !7
+  %mul.102 = sub nsw i32 0, %214
+  store i32 %mul.102, i32 addrspace(1)* %arrayidx.102, align 4, !tbaa !7
+  %215 = load i32, i32* %arrayidx1.103, align 4, !tbaa !7
+  %mul.103 = sub nsw i32 0, %215
+  store i32 %mul.103, i32 addrspace(1)* %arrayidx.103, align 4, !tbaa !7
+  %216 = load i32, i32* %arrayidx1.104, align 4, !tbaa !7
+  %mul.104 = sub nsw i32 0, %216
+  store i32 %mul.104, i32 addrspace(1)* %arrayidx.104, align 4, !tbaa !7
+  %217 = load i32, i32* %arrayidx1.105, align 4, !tbaa !7
+  %mul.105 = sub nsw i32 0, %217
+  store i32 %mul.105, i32 addrspace(1)* %arrayidx.105, align 4, !tbaa !7
+  %218 = load i32, i32* %arrayidx1.106, align 4, !tbaa !7
+  %mul.106 = sub nsw i32 0, %218
+  store i32 %mul.106, i32 addrspace(1)* %arrayidx.106, align 4, !tbaa !7
+  %219 = load i32, i32* %arrayidx1.107, align 4, !tbaa !7
+  %mul.107 = sub nsw i32 0, %219
+  store i32 %mul.107, i32 addrspace(1)* %arrayidx.107, align 4, !tbaa !7
+  %220 = load i32, i32* %arrayidx1.108, align 4, !tbaa !7
+  %mul.108 = sub nsw i32 0, %220
+  store i32 %mul.108, i32 addrspace(1)* %arrayidx.108, align 4, !tbaa !7
+  %221 = load i32, i32* %arrayidx1.109, align 4, !tbaa !7
+  %mul.109 = sub nsw i32 0, %221
+  store i32 %mul.109, i32 addrspace(1)* %arrayidx.109, align 4, !tbaa !7
+  %222 = load i32, i32* %arrayidx1.110, align 4, !tbaa !7
+  %mul.110 = sub nsw i32 0, %222
+  store i32 %mul.110, i32 addrspace(1)* %arrayidx.110, align 4, !tbaa !7
+  call void @llvm.lifetime.end(i64 444, i8* %0) #2
+  ret void
+}
+attributes #4 = { nounwind "amdgpu-num-active-waves-per-eu"="10,10" "amdgpu-flat-work-group-size"="256,256" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="fiji" "target-features"="+16-bit-insts,+s-memrealtime" "unsafe-fp-math"="false" "use-soft-float"="false" }
+
+; Function Attrs: argmemonly nounwind
+declare void @llvm.lifetime.start(i64, i8* nocapture) #1
+
+; Function Attrs: argmemonly nounwind
+declare void @llvm.lifetime.end(i64, i8* nocapture) #1
+
+attributes #1 = { argmemonly nounwind }
+attributes #2 = { nounwind }
+
+!llvm.ident = !{!6}
+
+!1 = !{!"kernel_arg_addr_space", i32 1}
+!2 = !{!"kernel_arg_access_qual", !"none"}
+!3 = !{!"kernel_arg_type", !"int*"}
+!4 = !{!"kernel_arg_base_type", !"int*"}
+!5 = !{!"kernel_arg_type_qual", !""}
+!6 = !{!"clang version 3.9.0 (trunk 272684)"}
+!7 = !{!8, !8, i64 0}
+!8 = !{!"int", !9, i64 0}
+!9 = !{!"omnipotent char", !10, i64 0}
+!10 = !{!"Simple C/C++ TBAA"}

test/CodeGen/AMDGPU/attr-amdgpu-num-active-waves-per-eu.ll

+; RUN: llc -march=amdgcn -mcpu=fiji -verify-machineinstrs < %s | FileCheck %s
+
+; Exactly 1 active wave per execution unit.
+; CHECK: SGPRBlocks: 12
+; CHECK: VGPRBlocks: 32
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 97
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 129
+define void @empty_exactly_1() #0 {
+entry:
+  ret void
+}
+attributes #0 = {"amdgpu-num-active-waves-per-eu"="1,1"}
+
+; Exactly 5 active waves per execution unit.
+; CHECK: SGPRBlocks: 12
+; CHECK: VGPRBlocks: 10
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 97
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 41
+define void @empty_exactly_5() #1 {
+entry:
+  ret void
+}
+attributes #1 = {"amdgpu-num-active-waves-per-eu"="5,5"}
+
+; Exactly 10 active waves per execution unit.
+; CHECK: SGPRBlocks: 0
+; CHECK: VGPRBlocks: 0
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 1
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 1
+define void @empty_exactly_10() #2 {
+entry:
+  ret void
+}
+attributes #2 = {"amdgpu-num-active-waves-per-eu"="10,10" "amdgpu-flat-work-group-size"="256,256"}
+
+; At least 1 active wave per execution unit.
+; CHECK: SGPRBlocks: 0
+; CHECK: VGPRBlocks: 0
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 1
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 1
+define void @empty_at_least_1() #3 {
+entry:
+  ret void
+}
+attributes #3 = {"amdgpu-num-active-waves-per-eu"="1"}
+
+; At least 5 active waves per execution unit.
+; CHECK: SGPRBlocks: 0
+; CHECK: VGPRBlocks: 0
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 1
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 1
+define void @empty_at_least_5() #4 {
+entry:
+  ret void
+}
+attributes #4 = {"amdgpu-num-active-waves-per-eu"="5"}
+
+; At least 10 active waves per execution unit.
+; CHECK: SGPRBlocks: 0
+; CHECK: VGPRBlocks: 0
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 1
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 1
+define void @empty_at_least_10() #5 {
+entry:
+  ret void
+}
+attributes #5 = {"amdgpu-num-active-waves-per-eu"="10" "amdgpu-flat-work-group-size"="256,256"}
+
+; At most 1 active wave per execution unit (same as @empty_exactly_1).
+
+; At most 5 active waves per execution unit.
+; CHECK: SGPRBlocks: 12
+; CHECK: VGPRBlocks: 10
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 97
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 41
+define void @empty_at_most_5() #6 {
+entry:
+  ret void
+}
+attributes #6 = {"amdgpu-num-active-waves-per-eu"="1,5"}
+
+; At most 10 active waves per execution unit.
+; CHECK: SGPRBlocks: 0
+; CHECK: VGPRBlocks: 0
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 1
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 1
+define void @empty_at_most_10() #7 {
+entry:
+  ret void
+}
+attributes #7 = {"amdgpu-num-active-waves-per-eu"="1,10"}
+
+; Between 1 and 5 active waves per execution unit (same as @empty_at_most_5).
+
+; Between 5 and 10 active waves per execution unit.
+; CHECK: SGPRBlocks: 0
+; CHECK: VGPRBlocks: 0
+; CHECK: NumSGPRsForNumActiveWavesPerEU: 1
+; CHECK: NumVGPRsForNumActiveWavesPerEU: 1
+define void @empty_between_5_and_10() #8 {
+entry:
+  ret void
+}
+attributes #8 = {"amdgpu-num-active-waves-per-eu"="5,10"}

test/CodeGen/AMDGPU/attr-amdgpu-num-sgpr-invalid.ll

+; RUN: not llc -march=amdgcn -mcpu=fiji -verify-machineinstrs < %s 2>&1 | FileCheck %s
+
+; For fiji, 8 active waves per execution unit -> [81, 96] SGPRs can be used.
+
+; CHECK: invalid amdgpu-num-sgpr: requested number of SGPRs cannot be achieved with default or requested maximum number of active waves per execution unit
+define void @invalid_0() #0 {
+entry:
+  ret void
+}
+attributes #0 = {"amdgpu-num-sgpr"="80" "amdgpu-num-active-waves-per-eu"="8,8"}
+
+; CHECK: invalid amdgpu-num-sgpr: requested number of SGPRs cannot be achieved with default or requested minimum number of active waves per execution unit
+define void @invalid_1() #1 {
+entry:
+  ret void
+}
+attributes #1 = {"amdgpu-num-sgpr"="97" "amdgpu-num-active-waves-per-eu"="8,8"}

test/CodeGen/AMDGPU/attr-amdgpu-num-sgpr.ll

+; RUN: llc -march=amdgcn -mcpu=fiji -verify-machineinstrs < %s | FileCheck %s
+
+source_filename = "attr-amdgpu-num-sgpr.cl"
+target datalayout = "e-p:32:32-p1:64:64-p2:64:64-p3:32:32-p4:64:64-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64"
+target triple = "amdgcn--amdhsa"
+
+; CHECK: NumSgprs: 16
+define void @attr_amdgpu_num_sgpr_16(i32 addrspace(1)* nocapture %A, i32 addrspace(1)* nocapture %B) local_unnamed_addr #0 {
+entry:
+  %arr = alloca [222 x i32], align 4
+  %0 = bitcast [222 x i32]* %arr to i8*
+  call void @llvm.lifetime.start(i64 888, i8* %0) #2
+  %1 = load i32, i32 addrspace(1)* %A, align 4, !tbaa !7
+  %add = add nsw i32 %1, 123
+  %arrayidx1 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 0
+  store i32 %add, i32* %arrayidx1, align 4, !tbaa !7
+  %arrayidx.1 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 1
+  %2 = load i32, i32 addrspace(1)* %arrayidx.1, align 4, !tbaa !7
+  %add.1 = add nsw i32 %2, 123
+  %arrayidx1.1 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 1
+  store i32 %add.1, i32* %arrayidx1.1, align 4, !tbaa !7
+  %arrayidx.2 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 2
+  %3 = load i32, i32 addrspace(1)* %arrayidx.2, align 4, !tbaa !7
+  %add.2 = add nsw i32 %3, 123
+  %arrayidx1.2 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 2
+  store i32 %add.2, i32* %arrayidx1.2, align 4, !tbaa !7
+  %arrayidx.3 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 3
+  %4 = load i32, i32 addrspace(1)* %arrayidx.3, align 4, !tbaa !7
+  %add.3 = add nsw i32 %4, 123
+  %arrayidx1.3 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 3
+  store i32 %add.3, i32* %arrayidx1.3, align 4, !tbaa !7
+  %arrayidx.4 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 4
+  %5 = load i32, i32 addrspace(1)* %arrayidx.4, align 4, !tbaa !7
+  %add.4 = add nsw i32 %5, 123
+  %arrayidx1.4 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 4
+  store i32 %add.4, i32* %arrayidx1.4, align 4, !tbaa !7
+  %arrayidx.5 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 5
+  %6 = load i32, i32 addrspace(1)* %arrayidx.5, align 4, !tbaa !7
+  %add.5 = add nsw i32 %6, 123
+  %arrayidx1.5 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 5
+  store i32 %add.5, i32* %arrayidx1.5, align 4, !tbaa !7
+  %arrayidx.6 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 6
+  %7 = load i32, i32 addrspace(1)* %arrayidx.6, align 4, !tbaa !7
+  %add.6 = add nsw i32 %7, 123
+  %arrayidx1.6 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 6
+  store i32 %add.6, i32* %arrayidx1.6, align 4, !tbaa !7
+  %arrayidx.7 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 7
+  %8 = load i32, i32 addrspace(1)* %arrayidx.7, align 4, !tbaa !7
+  %add.7 = add nsw i32 %8, 123
+  %arrayidx1.7 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 7
+  store i32 %add.7, i32* %arrayidx1.7, align 4, !tbaa !7
+  %arrayidx.8 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 8
+  %9 = load i32, i32 addrspace(1)* %arrayidx.8, align 4, !tbaa !7
+  %add.8 = add nsw i32 %9, 123
+  %arrayidx1.8 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 8
+  store i32 %add.8, i32* %arrayidx1.8, align 4, !tbaa !7
+  %arrayidx.9 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 9
+  %10 = load i32, i32 addrspace(1)* %arrayidx.9, align 4, !tbaa !7
+  %add.9 = add nsw i32 %10, 123
+  %arrayidx1.9 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 9
+  store i32 %add.9, i32* %arrayidx1.9, align 4, !tbaa !7
+  %arrayidx.10 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 10
+  %11 = load i32, i32 addrspace(1)* %arrayidx.10, align 4, !tbaa !7
+  %add.10 = add nsw i32 %11, 123
+  %arrayidx1.10 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 10
+  store i32 %add.10, i32* %arrayidx1.10, align 4, !tbaa !7
+  %arrayidx.11 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 11
+  %12 = load i32, i32 addrspace(1)* %arrayidx.11, align 4, !tbaa !7
+  %add.11 = add nsw i32 %12, 123
+  %arrayidx1.11 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 11
+  store i32 %add.11, i32* %arrayidx1.11, align 4, !tbaa !7
+  %arrayidx.12 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 12
+  %13 = load i32, i32 addrspace(1)* %arrayidx.12, align 4, !tbaa !7
+  %add.12 = add nsw i32 %13, 123
+  %arrayidx1.12 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 12
+  store i32 %add.12, i32* %arrayidx1.12, align 4, !tbaa !7
+  %arrayidx.13 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 13
+  %14 = load i32, i32 addrspace(1)* %arrayidx.13, align 4, !tbaa !7
+  %add.13 = add nsw i32 %14, 123
+  %arrayidx1.13 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 13
+  store i32 %add.13, i32* %arrayidx1.13, align 4, !tbaa !7
+  %arrayidx.14 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 14
+  %15 = load i32, i32 addrspace(1)* %arrayidx.14, align 4, !tbaa !7
+  %add.14 = add nsw i32 %15, 123
+  %arrayidx1.14 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 14
+  store i32 %add.14, i32* %arrayidx1.14, align 4, !tbaa !7
+  %arrayidx.15 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 15
+  %16 = load i32, i32 addrspace(1)* %arrayidx.15, align 4, !tbaa !7
+  %add.15 = add nsw i32 %16, 123
+  %arrayidx1.15 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 15
+  store i32 %add.15, i32* %arrayidx1.15, align 4, !tbaa !7
+  %arrayidx.16 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 16
+  %17 = load i32, i32 addrspace(1)* %arrayidx.16, align 4, !tbaa !7
+  %add.16 = add nsw i32 %17, 123
+  %arrayidx1.16 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 16
+  store i32 %add.16, i32* %arrayidx1.16, align 4, !tbaa !7
+  %arrayidx.17 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 17
+  %18 = load i32, i32 addrspace(1)* %arrayidx.17, align 4, !tbaa !7
+  %add.17 = add nsw i32 %18, 123
+  %arrayidx1.17 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 17
+  store i32 %add.17, i32* %arrayidx1.17, align 4, !tbaa !7
+  %arrayidx.18 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 18
+  %19 = load i32, i32 addrspace(1)* %arrayidx.18, align 4, !tbaa !7
+  %add.18 = add nsw i32 %19, 123
+  %arrayidx1.18 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 18
+  store i32 %add.18, i32* %arrayidx1.18, align 4, !tbaa !7
+  %arrayidx.19 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 19
+  %20 = load i32, i32 addrspace(1)* %arrayidx.19, align 4, !tbaa !7
+  %add.19 = add nsw i32 %20, 123
+  %arrayidx1.19 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 19
+  store i32 %add.19, i32* %arrayidx1.19, align 4, !tbaa !7
+  %arrayidx.20 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 20
+  %21 = load i32, i32 addrspace(1)* %arrayidx.20, align 4, !tbaa !7
+  %add.20 = add nsw i32 %21, 123
+  %arrayidx1.20 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 20
+  store i32 %add.20, i32* %arrayidx1.20, align 4, !tbaa !7
+  %arrayidx.21 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 21
+  %22 = load i32, i32 addrspace(1)* %arrayidx.21, align 4, !tbaa !7
+  %add.21 = add nsw i32 %22, 123
+  %arrayidx1.21 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 21
+  store i32 %add.21, i32* %arrayidx1.21, align 4, !tbaa !7
+  %arrayidx.22 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 22
+  %23 = load i32, i32 addrspace(1)* %arrayidx.22, align 4, !tbaa !7
+  %add.22 = add nsw i32 %23, 123
+  %arrayidx1.22 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 22
+  store i32 %add.22, i32* %arrayidx1.22, align 4, !tbaa !7
+  %arrayidx.23 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 23
+  %24 = load i32, i32 addrspace(1)* %arrayidx.23, align 4, !tbaa !7
+  %add.23 = add nsw i32 %24, 123
+  %arrayidx1.23 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 23
+  store i32 %add.23, i32* %arrayidx1.23, align 4, !tbaa !7
+  %arrayidx.24 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 24
+  %25 = load i32, i32 addrspace(1)* %arrayidx.24, align 4, !tbaa !7
+  %add.24 = add nsw i32 %25, 123
+  %arrayidx1.24 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 24
+  store i32 %add.24, i32* %arrayidx1.24, align 4, !tbaa !7
+  %arrayidx.25 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 25
+  %26 = load i32, i32 addrspace(1)* %arrayidx.25, align 4, !tbaa !7
+  %add.25 = add nsw i32 %26, 123
+  %arrayidx1.25 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 25
+  store i32 %add.25, i32* %arrayidx1.25, align 4, !tbaa !7
+  %arrayidx.26 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 26
+  %27 = load i32, i32 addrspace(1)* %arrayidx.26, align 4, !tbaa !7
+  %add.26 = add nsw i32 %27, 123
+  %arrayidx1.26 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 26
+  store i32 %add.26, i32* %arrayidx1.26, align 4, !tbaa !7
+  %arrayidx.27 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 27
+  %28 = load i32, i32 addrspace(1)* %arrayidx.27, align 4, !tbaa !7
+  %add.27 = add nsw i32 %28, 123
+  %arrayidx1.27 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 27
+  store i32 %add.27, i32* %arrayidx1.27, align 4, !tbaa !7
+  %arrayidx.28 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 28
+  %29 = load i32, i32 addrspace(1)* %arrayidx.28, align 4, !tbaa !7
+  %add.28 = add nsw i32 %29, 123
+  %arrayidx1.28 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 28
+  store i32 %add.28, i32* %arrayidx1.28, align 4, !tbaa !7
+  %arrayidx.29 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 29
+  %30 = load i32, i32 addrspace(1)* %arrayidx.29, align 4, !tbaa !7
+  %add.29 = add nsw i32 %30, 123
+  %arrayidx1.29 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 29
+  store i32 %add.29, i32* %arrayidx1.29, align 4, !tbaa !7
+  %arrayidx.30 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 30
+  %31 = load i32, i32 addrspace(1)* %arrayidx.30, align 4, !tbaa !7
+  %add.30 = add nsw i32 %31, 123
+  %arrayidx1.30 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 30
+  store i32 %add.30, i32* %arrayidx1.30, align 4, !tbaa !7
+  %arrayidx.31 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 31
+  %32 = load i32, i32 addrspace(1)* %arrayidx.31, align 4, !tbaa !7
+  %add.31 = add nsw i32 %32, 123
+  %arrayidx1.31 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 31
+  store i32 %add.31, i32* %arrayidx1.31, align 4, !tbaa !7
+  %arrayidx.32 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 32
+  %33 = load i32, i32 addrspace(1)* %arrayidx.32, align 4, !tbaa !7
+  %add.32 = add nsw i32 %33, 123
+  %arrayidx1.32 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 32
+  store i32 %add.32, i32* %arrayidx1.32, align 4, !tbaa !7
+  %arrayidx.33 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 33
+  %34 = load i32, i32 addrspace(1)* %arrayidx.33, align 4, !tbaa !7
+  %add.33 = add nsw i32 %34, 123
+  %arrayidx1.33 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 33
+  store i32 %add.33, i32* %arrayidx1.33, align 4, !tbaa !7
+  %arrayidx.34 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 34
+  %35 = load i32, i32 addrspace(1)* %arrayidx.34, align 4, !tbaa !7
+  %add.34 = add nsw i32 %35, 123
+  %arrayidx1.34 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 34
+  store i32 %add.34, i32* %arrayidx1.34, align 4, !tbaa !7
+  %arrayidx.35 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 35
+  %36 = load i32, i32 addrspace(1)* %arrayidx.35, align 4, !tbaa !7
+  %add.35 = add nsw i32 %36, 123
+  %arrayidx1.35 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 35
+  store i32 %add.35, i32* %arrayidx1.35, align 4, !tbaa !7
+  %arrayidx.36 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 36
+  %37 = load i32, i32 addrspace(1)* %arrayidx.36, align 4, !tbaa !7
+  %add.36 = add nsw i32 %37, 123
+  %arrayidx1.36 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 36
+  store i32 %add.36, i32* %arrayidx1.36, align 4, !tbaa !7
+  %arrayidx.37 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 37
+  %38 = load i32, i32 addrspace(1)* %arrayidx.37, align 4, !tbaa !7
+  %add.37 = add nsw i32 %38, 123
+  %arrayidx1.37 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 37
+  store i32 %add.37, i32* %arrayidx1.37, align 4, !tbaa !7
+  %arrayidx.38 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 38
+  %39 = load i32, i32 addrspace(1)* %arrayidx.38, align 4, !tbaa !7
+  %add.38 = add nsw i32 %39, 123
+  %arrayidx1.38 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 38
+  store i32 %add.38, i32* %arrayidx1.38, align 4, !tbaa !7
+  %arrayidx.39 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 39
+  %40 = load i32, i32 addrspace(1)* %arrayidx.39, align 4, !tbaa !7
+  %add.39 = add nsw i32 %40, 123
+  %arrayidx1.39 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 39
+  store i32 %add.39, i32* %arrayidx1.39, align 4, !tbaa !7
+  %arrayidx.40 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 40
+  %41 = load i32, i32 addrspace(1)* %arrayidx.40, align 4, !tbaa !7
+  %add.40 = add nsw i32 %41, 123
+  %arrayidx1.40 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 40
+  store i32 %add.40, i32* %arrayidx1.40, align 4, !tbaa !7
+  %arrayidx.41 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 41
+  %42 = load i32, i32 addrspace(1)* %arrayidx.41, align 4, !tbaa !7
+  %add.41 = add nsw i32 %42, 123
+  %arrayidx1.41 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 41
+  store i32 %add.41, i32* %arrayidx1.41, align 4, !tbaa !7
+  %arrayidx.42 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 42
+  %43 = load i32, i32 addrspace(1)* %arrayidx.42, align 4, !tbaa !7
+  %add.42 = add nsw i32 %43, 123
+  %arrayidx1.42 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 42
+  store i32 %add.42, i32* %arrayidx1.42, align 4, !tbaa !7
+  %arrayidx.43 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 43
+  %44 = load i32, i32 addrspace(1)* %arrayidx.43, align 4, !tbaa !7
+  %add.43 = add nsw i32 %44, 123
+  %arrayidx1.43 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 43
+  store i32 %add.43, i32* %arrayidx1.43, align 4, !tbaa !7
+  %arrayidx.44 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 44
+  %45 = load i32, i32 addrspace(1)* %arrayidx.44, align 4, !tbaa !7
+  %add.44 = add nsw i32 %45, 123
+  %arrayidx1.44 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 44
+  store i32 %add.44, i32* %arrayidx1.44, align 4, !tbaa !7
+  %arrayidx.45 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 45
+  %46 = load i32, i32 addrspace(1)* %arrayidx.45, align 4, !tbaa !7
+  %add.45 = add nsw i32 %46, 123
+  %arrayidx1.45 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 45
+  store i32 %add.45, i32* %arrayidx1.45, align 4, !tbaa !7
+  %arrayidx.46 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 46
+  %47 = load i32, i32 addrspace(1)* %arrayidx.46, align 4, !tbaa !7
+  %add.46 = add nsw i32 %47, 123
+  %arrayidx1.46 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 46
+  store i32 %add.46, i32* %arrayidx1.46, align 4, !tbaa !7
+  %arrayidx.47 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 47
+  %48 = load i32, i32 addrspace(1)* %arrayidx.47, align 4, !tbaa !7
+  %add.47 = add nsw i32 %48, 123
+  %arrayidx1.47 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 47
+  store i32 %add.47, i32* %arrayidx1.47, align 4, !tbaa !7
+  %arrayidx.48 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 48
+  %49 = load i32, i32 addrspace(1)* %arrayidx.48, align 4, !tbaa !7
+  %add.48 = add nsw i32 %49, 123
+  %arrayidx1.48 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 48
+  store i32 %add.48, i32* %arrayidx1.48, align 4, !tbaa !7
+  %arrayidx.49 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 49
+  %50 = load i32, i32 addrspace(1)* %arrayidx.49, align 4, !tbaa !7
+  %add.49 = add nsw i32 %50, 123
+  %arrayidx1.49 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 49
+  store i32 %add.49, i32* %arrayidx1.49, align 4, !tbaa !7
+  %arrayidx.50 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 50
+  %51 = load i32, i32 addrspace(1)* %arrayidx.50, align 4, !tbaa !7
+  %add.50 = add nsw i32 %51, 123
+  %arrayidx1.50 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 50
+  store i32 %add.50, i32* %arrayidx1.50, align 4, !tbaa !7
+  %arrayidx.51 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 51
+  %52 = load i32, i32 addrspace(1)* %arrayidx.51, align 4, !tbaa !7
+  %add.51 = add nsw i32 %52, 123
+  %arrayidx1.51 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 51
+  store i32 %add.51, i32* %arrayidx1.51, align 4, !tbaa !7
+  %arrayidx.52 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 52
+  %53 = load i32, i32 addrspace(1)* %arrayidx.52, align 4, !tbaa !7
+  %add.52 = add nsw i32 %53, 123
+  %arrayidx1.52 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 52
+  store i32 %add.52, i32* %arrayidx1.52, align 4, !tbaa !7
+  %arrayidx.53 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 53
+  %54 = load i32, i32 addrspace(1)* %arrayidx.53, align 4, !tbaa !7
+  %add.53 = add nsw i32 %54, 123
+  %arrayidx1.53 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 53
+  store i32 %add.53, i32* %arrayidx1.53, align 4, !tbaa !7
+  %arrayidx.54 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 54
+  %55 = load i32, i32 addrspace(1)* %arrayidx.54, align 4, !tbaa !7
+  %add.54 = add nsw i32 %55, 123
+  %arrayidx1.54 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 54
+  store i32 %add.54, i32* %arrayidx1.54, align 4, !tbaa !7
+  %arrayidx.55 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 55
+  %56 = load i32, i32 addrspace(1)* %arrayidx.55, align 4, !tbaa !7
+  %add.55 = add nsw i32 %56, 123
+  %arrayidx1.55 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 55
+  store i32 %add.55, i32* %arrayidx1.55, align 4, !tbaa !7
+  %arrayidx.56 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 56
+  %57 = load i32, i32 addrspace(1)* %arrayidx.56, align 4, !tbaa !7
+  %add.56 = add nsw i32 %57, 123
+  %arrayidx1.56 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 56
+  store i32 %add.56, i32* %arrayidx1.56, align 4, !tbaa !7
+  %arrayidx.57 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 57
+  %58 = load i32, i32 addrspace(1)* %arrayidx.57, align 4, !tbaa !7
+  %add.57 = add nsw i32 %58, 123
+  %arrayidx1.57 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 57
+  store i32 %add.57, i32* %arrayidx1.57, align 4, !tbaa !7
+  %arrayidx.58 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 58
+  %59 = load i32, i32 addrspace(1)* %arrayidx.58, align 4, !tbaa !7
+  %add.58 = add nsw i32 %59, 123
+  %arrayidx1.58 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 58
+  store i32 %add.58, i32* %arrayidx1.58, align 4, !tbaa !7
+  %arrayidx.59 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 59
+  %60 = load i32, i32 addrspace(1)* %arrayidx.59, align 4, !tbaa !7
+  %add.59 = add nsw i32 %60, 123
+  %arrayidx1.59 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 59
+  store i32 %add.59, i32* %arrayidx1.59, align 4, !tbaa !7
+  %arrayidx.60 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 60
+  %61 = load i32, i32 addrspace(1)* %arrayidx.60, align 4, !tbaa !7
+  %add.60 = add nsw i32 %61, 123
+  %arrayidx1.60 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 60
+  store i32 %add.60, i32* %arrayidx1.60, align 4, !tbaa !7
+  %arrayidx.61 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 61
+  %62 = load i32, i32 addrspace(1)* %arrayidx.61, align 4, !tbaa !7
+  %add.61 = add nsw i32 %62, 123
+  %arrayidx1.61 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 61
+  store i32 %add.61, i32* %arrayidx1.61, align 4, !tbaa !7
+  %arrayidx.62 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 62
+  %63 = load i32, i32 addrspace(1)* %arrayidx.62, align 4, !tbaa !7
+  %add.62 = add nsw i32 %63, 123
+  %arrayidx1.62 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 62
+  store i32 %add.62, i32* %arrayidx1.62, align 4, !tbaa !7
+  %arrayidx.63 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 63
+  %64 = load i32, i32 addrspace(1)* %arrayidx.63, align 4, !tbaa !7
+  %add.63 = add nsw i32 %64, 123
+  %arrayidx1.63 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 63
+  store i32 %add.63, i32* %arrayidx1.63, align 4, !tbaa !7
+  %arrayidx.64 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 64
+  %65 = load i32, i32 addrspace(1)* %arrayidx.64, align 4, !tbaa !7
+  %add.64 = add nsw i32 %65, 123
+  %arrayidx1.64 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 64
+  store i32 %add.64, i32* %arrayidx1.64, align 4, !tbaa !7
+  %arrayidx.65 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 65
+  %66 = load i32, i32 addrspace(1)* %arrayidx.65, align 4, !tbaa !7
+  %add.65 = add nsw i32 %66, 123
+  %arrayidx1.65 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 65
+  store i32 %add.65, i32* %arrayidx1.65, align 4, !tbaa !7
+  %arrayidx.66 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 66
+  %67 = load i32, i32 addrspace(1)* %arrayidx.66, align 4, !tbaa !7
+  %add.66 = add nsw i32 %67, 123
+  %arrayidx1.66 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 66
+  store i32 %add.66, i32* %arrayidx1.66, align 4, !tbaa !7
+  %arrayidx.67 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 67
+  %68 = load i32, i32 addrspace(1)* %arrayidx.67, align 4, !tbaa !7
+  %add.67 = add nsw i32 %68, 123
+  %arrayidx1.67 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 67
+  store i32 %add.67, i32* %arrayidx1.67, align 4, !tbaa !7
+  %arrayidx.68 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 68
+  %69 = load i32, i32 addrspace(1)* %arrayidx.68, align 4, !tbaa !7
+  %add.68 = add nsw i32 %69, 123
+  %arrayidx1.68 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 68
+  store i32 %add.68, i32* %arrayidx1.68, align 4, !tbaa !7
+  %arrayidx.69 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 69
+  %70 = load i32, i32 addrspace(1)* %arrayidx.69, align 4, !tbaa !7
+  %add.69 = add nsw i32 %70, 123
+  %arrayidx1.69 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 69
+  store i32 %add.69, i32* %arrayidx1.69, align 4, !tbaa !7
+  %arrayidx.70 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 70
+  %71 = load i32, i32 addrspace(1)* %arrayidx.70, align 4, !tbaa !7
+  %add.70 = add nsw i32 %71, 123
+  %arrayidx1.70 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 70
+  store i32 %add.70, i32* %arrayidx1.70, align 4, !tbaa !7
+  %arrayidx.71 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 71
+  %72 = load i32, i32 addrspace(1)* %arrayidx.71, align 4, !tbaa !7
+  %add.71 = add nsw i32 %72, 123
+  %arrayidx1.71 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 71
+  store i32 %add.71, i32* %arrayidx1.71, align 4, !tbaa !7
+  %arrayidx.72 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 72
+  %73 = load i32, i32 addrspace(1)* %arrayidx.72, align 4, !tbaa !7
+  %add.72 = add nsw i32 %73, 123
+  %arrayidx1.72 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 72
+  store i32 %add.72, i32* %arrayidx1.72, align 4, !tbaa !7
+  %arrayidx.73 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 73
+  %74 = load i32, i32 addrspace(1)* %arrayidx.73, align 4, !tbaa !7
+  %add.73 = add nsw i32 %74, 123
+  %arrayidx1.73 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 73
+  store i32 %add.73, i32* %arrayidx1.73, align 4, !tbaa !7
+  %arrayidx.74 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 74
+  %75 = load i32, i32 addrspace(1)* %arrayidx.74, align 4, !tbaa !7
+  %add.74 = add nsw i32 %75, 123
+  %arrayidx1.74 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 74
+  store i32 %add.74, i32* %arrayidx1.74, align 4, !tbaa !7
+  %arrayidx.75 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 75
+  %76 = load i32, i32 addrspace(1)* %arrayidx.75, align 4, !tbaa !7
+  %add.75 = add nsw i32 %76, 123
+  %arrayidx1.75 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 75
+  store i32 %add.75, i32* %arrayidx1.75, align 4, !tbaa !7
+  %arrayidx.76 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 76
+  %77 = load i32, i32 addrspace(1)* %arrayidx.76, align 4, !tbaa !7
+  %add.76 = add nsw i32 %77, 123
+  %arrayidx1.76 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 76
+  store i32 %add.76, i32* %arrayidx1.76, align 4, !tbaa !7
+  %arrayidx.77 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 77
+  %78 = load i32, i32 addrspace(1)* %arrayidx.77, align 4, !tbaa !7
+  %add.77 = add nsw i32 %78, 123
+  %arrayidx1.77 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 77
+  store i32 %add.77, i32* %arrayidx1.77, align 4, !tbaa !7
+  %arrayidx.78 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 78
+  %79 = load i32, i32 addrspace(1)* %arrayidx.78, align 4, !tbaa !7
+  %add.78 = add nsw i32 %79, 123
+  %arrayidx1.78 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 78
+  store i32 %add.78, i32* %arrayidx1.78, align 4, !tbaa !7
+  %arrayidx.79 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 79
+  %80 = load i32, i32 addrspace(1)* %arrayidx.79, align 4, !tbaa !7
+  %add.79 = add nsw i32 %80, 123
+  %arrayidx1.79 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 79
+  store i32 %add.79, i32* %arrayidx1.79, align 4, !tbaa !7
+  %arrayidx.80 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 80
+  %81 = load i32, i32 addrspace(1)* %arrayidx.80, align 4, !tbaa !7
+  %add.80 = add nsw i32 %81, 123
+  %arrayidx1.80 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 80
+  store i32 %add.80, i32* %arrayidx1.80, align 4, !tbaa !7
+  %arrayidx.81 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 81
+  %82 = load i32, i32 addrspace(1)* %arrayidx.81, align 4, !tbaa !7
+  %add.81 = add nsw i32 %82, 123
+  %arrayidx1.81 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 81
+  store i32 %add.81, i32* %arrayidx1.81, align 4, !tbaa !7
+  %arrayidx.82 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 82
+  %83 = load i32, i32 addrspace(1)* %arrayidx.82, align 4, !tbaa !7
+  %add.82 = add nsw i32 %83, 123
+  %arrayidx1.82 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 82
+  store i32 %add.82, i32* %arrayidx1.82, align 4, !tbaa !7
+  %arrayidx.83 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 83
+  %84 = load i32, i32 addrspace(1)* %arrayidx.83, align 4, !tbaa !7
+  %add.83 = add nsw i32 %84, 123
+  %arrayidx1.83 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 83
+  store i32 %add.83, i32* %arrayidx1.83, align 4, !tbaa !7
+  %arrayidx.84 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 84
+  %85 = load i32, i32 addrspace(1)* %arrayidx.84, align 4, !tbaa !7
+  %add.84 = add nsw i32 %85, 123
+  %arrayidx1.84 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 84
+  store i32 %add.84, i32* %arrayidx1.84, align 4, !tbaa !7
+  %arrayidx.85 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 85
+  %86 = load i32, i32 addrspace(1)* %arrayidx.85, align 4, !tbaa !7
+  %add.85 = add nsw i32 %86, 123
+  %arrayidx1.85 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 85
+  store i32 %add.85, i32* %arrayidx1.85, align 4, !tbaa !7
+  %arrayidx.86 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 86
+  %87 = load i32, i32 addrspace(1)* %arrayidx.86, align 4, !tbaa !7
+  %add.86 = add nsw i32 %87, 123
+  %arrayidx1.86 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 86
+  store i32 %add.86, i32* %arrayidx1.86, align 4, !tbaa !7
+  %arrayidx.87 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 87
+  %88 = load i32, i32 addrspace(1)* %arrayidx.87, align 4, !tbaa !7
+  %add.87 = add nsw i32 %88, 123
+  %arrayidx1.87 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 87
+  store i32 %add.87, i32* %arrayidx1.87, align 4, !tbaa !7
+  %arrayidx.88 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 88
+  %89 = load i32, i32 addrspace(1)* %arrayidx.88, align 4, !tbaa !7
+  %add.88 = add nsw i32 %89, 123
+  %arrayidx1.88 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 88
+  store i32 %add.88, i32* %arrayidx1.88, align 4, !tbaa !7
+  %arrayidx.89 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 89
+  %90 = load i32, i32 addrspace(1)* %arrayidx.89, align 4, !tbaa !7
+  %add.89 = add nsw i32 %90, 123
+  %arrayidx1.89 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 89
+  store i32 %add.89, i32* %arrayidx1.89, align 4, !tbaa !7
+  %arrayidx.90 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 90
+  %91 = load i32, i32 addrspace(1)* %arrayidx.90, align 4, !tbaa !7
+  %add.90 = add nsw i32 %91, 123
+  %arrayidx1.90 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 90
+  store i32 %add.90, i32* %arrayidx1.90, align 4, !tbaa !7
+  %arrayidx.91 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 91
+  %92 = load i32, i32 addrspace(1)* %arrayidx.91, align 4, !tbaa !7
+  %add.91 = add nsw i32 %92, 123
+  %arrayidx1.91 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 91
+  store i32 %add.91, i32* %arrayidx1.91, align 4, !tbaa !7
+  %arrayidx.92 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 92
+  %93 = load i32, i32 addrspace(1)* %arrayidx.92, align 4, !tbaa !7
+  %add.92 = add nsw i32 %93, 123
+  %arrayidx1.92 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 92
+  store i32 %add.92, i32* %arrayidx1.92, align 4, !tbaa !7
+  %arrayidx.93 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 93
+  %94 = load i32, i32 addrspace(1)* %arrayidx.93, align 4, !tbaa !7
+  %add.93 = add nsw i32 %94, 123
+  %arrayidx1.93 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 93
+  store i32 %add.93, i32* %arrayidx1.93, align 4, !tbaa !7
+  %arrayidx.94 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 94
+  %95 = load i32, i32 addrspace(1)* %arrayidx.94, align 4, !tbaa !7
+  %add.94 = add nsw i32 %95, 123
+  %arrayidx1.94 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 94
+  store i32 %add.94, i32* %arrayidx1.94, align 4, !tbaa !7
+  %arrayidx.95 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 95
+  %96 = load i32, i32 addrspace(1)* %arrayidx.95, align 4, !tbaa !7
+  %add.95 = add nsw i32 %96, 123
+  %arrayidx1.95 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 95
+  store i32 %add.95, i32* %arrayidx1.95, align 4, !tbaa !7
+  %arrayidx.96 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 96
+  %97 = load i32, i32 addrspace(1)* %arrayidx.96, align 4, !tbaa !7
+  %add.96 = add nsw i32 %97, 123
+  %arrayidx1.96 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 96
+  store i32 %add.96, i32* %arrayidx1.96, align 4, !tbaa !7
+  %arrayidx.97 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 97
+  %98 = load i32, i32 addrspace(1)* %arrayidx.97, align 4, !tbaa !7
+  %add.97 = add nsw i32 %98, 123
+  %arrayidx1.97 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 97
+  store i32 %add.97, i32* %arrayidx1.97, align 4, !tbaa !7
+  %arrayidx.98 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 98
+  %99 = load i32, i32 addrspace(1)* %arrayidx.98, align 4, !tbaa !7
+  %add.98 = add nsw i32 %99, 123
+  %arrayidx1.98 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 98
+  store i32 %add.98, i32* %arrayidx1.98, align 4, !tbaa !7
+  %arrayidx.99 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 99
+  %100 = load i32, i32 addrspace(1)* %arrayidx.99, align 4, !tbaa !7
+  %add.99 = add nsw i32 %100, 123
+  %arrayidx1.99 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 99
+  store i32 %add.99, i32* %arrayidx1.99, align 4, !tbaa !7
+  %arrayidx.100 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 100
+  %101 = load i32, i32 addrspace(1)* %arrayidx.100, align 4, !tbaa !7
+  %add.100 = add nsw i32 %101, 123
+  %arrayidx1.100 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 100
+  store i32 %add.100, i32* %arrayidx1.100, align 4, !tbaa !7
+  %arrayidx.101 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 101
+  %102 = load i32, i32 addrspace(1)* %arrayidx.101, align 4, !tbaa !7
+  %add.101 = add nsw i32 %102, 123
+  %arrayidx1.101 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 101
+  store i32 %add.101, i32* %arrayidx1.101, align 4, !tbaa !7
+  %arrayidx.102 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 102
+  %103 = load i32, i32 addrspace(1)* %arrayidx.102, align 4, !tbaa !7
+  %add.102 = add nsw i32 %103, 123
+  %arrayidx1.102 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 102
+  store i32 %add.102, i32* %arrayidx1.102, align 4, !tbaa !7
+  %arrayidx.103 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 103
+  %104 = load i32, i32 addrspace(1)* %arrayidx.103, align 4, !tbaa !7
+  %add.103 = add nsw i32 %104, 123
+  %arrayidx1.103 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 103
+  store i32 %add.103, i32* %arrayidx1.103, align 4, !tbaa !7
+  %arrayidx.104 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 104
+  %105 = load i32, i32 addrspace(1)* %arrayidx.104, align 4, !tbaa !7
+  %add.104 = add nsw i32 %105, 123
+  %arrayidx1.104 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 104
+  store i32 %add.104, i32* %arrayidx1.104, align 4, !tbaa !7
+  %arrayidx.105 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 105
+  %106 = load i32, i32 addrspace(1)* %arrayidx.105, align 4, !tbaa !7
+  %add.105 = add nsw i32 %106, 123
+  %arrayidx1.105 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 105
+  store i32 %add.105, i32* %arrayidx1.105, align 4, !tbaa !7
+  %arrayidx.106 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 106
+  %107 = load i32, i32 addrspace(1)* %arrayidx.106, align 4, !tbaa !7
+  %add.106 = add nsw i32 %107, 123
+  %arrayidx1.106 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 106
+  store i32 %add.106, i32* %arrayidx1.106, align 4, !tbaa !7
+  %arrayidx.107 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 107
+  %108 = load i32, i32 addrspace(1)* %arrayidx.107, align 4, !tbaa !7
+  %add.107 = add nsw i32 %108, 123
+  %arrayidx1.107 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 107
+  store i32 %add.107, i32* %arrayidx1.107, align 4, !tbaa !7
+  %arrayidx.108 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 108
+  %109 = load i32, i32 addrspace(1)* %arrayidx.108, align 4, !tbaa !7
+  %add.108 = add nsw i32 %109, 123
+  %arrayidx1.108 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 108
+  store i32 %add.108, i32* %arrayidx1.108, align 4, !tbaa !7
+  %arrayidx.109 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 109
+  %110 = load i32, i32 addrspace(1)* %arrayidx.109, align 4, !tbaa !7
+  %add.109 = add nsw i32 %110, 123
+  %arrayidx1.109 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 109
+  store i32 %add.109, i32* %arrayidx1.109, align 4, !tbaa !7
+  %arrayidx.110 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 110
+  %111 = load i32, i32 addrspace(1)* %arrayidx.110, align 4, !tbaa !7
+  %add.110 = add nsw i32 %111, 123
+  %arrayidx1.110 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 110
+  store i32 %add.110, i32* %arrayidx1.110, align 4, !tbaa !7
+  %112 = load i32, i32 addrspace(1)* %A, align 4, !tbaa !7
+  %cmp3 = icmp slt i32 %112, 50
+  br i1 %cmp3, label %for.body8.preheader, label %for.body18.preheader
+
+for.body18.preheader:                             ; preds = %entry
+  %mul2080 = mul nsw i32 %112, -2
+  store i32 %mul2080, i32 addrspace(1)* %A, align 4, !tbaa !7
+  br label %for.body18.for.body18_crit_edge
+
+for.body8.preheader:                              ; preds = %entry
+  %mul78 = sub nsw i32 0, %112
+  store i32 %mul78, i32 addrspace(1)* %A, align 4, !tbaa !7
+  br label %for.body8.for.body8_crit_edge
+
+for.body8.for.body8_crit_edge:                    ; preds = %for.body8.for.body8_crit_edge, %for.body8.preheader
+  %inc1279 = phi i32 [ 1, %for.body8.preheader ], [ %inc12.10, %for.body8.for.body8_crit_edge ]
+  %.phi.trans.insert74 = sext i32 %inc1279 to i64
+  %arrayidx9.phi.trans.insert = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert74
+  %.pre75 = load i32, i32 addrspace(1)* %arrayidx9.phi.trans.insert, align 4, !tbaa !7
+  %113 = sext i32 %inc1279 to i64
+  %arrayidx9 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %113
+  %mul = sub nsw i32 0, %.pre75
+  store i32 %mul, i32 addrspace(1)* %arrayidx9, align 4, !tbaa !7
+  %inc12 = add nuw nsw i32 %inc1279, 1
+  %.phi.trans.insert74.1 = sext i32 %inc12 to i64
+  %arrayidx9.phi.trans.insert.1 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert74.1
+  %.pre75.1 = load i32, i32 addrspace(1)* %arrayidx9.phi.trans.insert.1, align 4, !tbaa !7
+  %114 = sext i32 %inc12 to i64
+  %arrayidx9.1 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %114
+  %mul.1 = sub nsw i32 0, %.pre75.1
+  store i32 %mul.1, i32 addrspace(1)* %arrayidx9.1, align 4, !tbaa !7
+  %inc12.1 = add nsw i32 %inc1279, 2
+  %.phi.trans.insert74.2 = sext i32 %inc12.1 to i64
+  %arrayidx9.phi.trans.insert.2 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert74.2
+  %.pre75.2 = load i32, i32 addrspace(1)* %arrayidx9.phi.trans.insert.2, align 4, !tbaa !7
+  %115 = sext i32 %inc12.1 to i64
+  %arrayidx9.2 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %115
+  %mul.2 = sub nsw i32 0, %.pre75.2
+  store i32 %mul.2, i32 addrspace(1)* %arrayidx9.2, align 4, !tbaa !7
+  %inc12.2 = add nsw i32 %inc1279, 3
+  %.phi.trans.insert74.3 = sext i32 %inc12.2 to i64
+  %arrayidx9.phi.trans.insert.3 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert74.3
+  %.pre75.3 = load i32, i32 addrspace(1)* %arrayidx9.phi.trans.insert.3, align 4, !tbaa !7
+  %116 = sext i32 %inc12.2 to i64
+  %arrayidx9.3 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %116
+  %mul.3 = sub nsw i32 0, %.pre75.3
+  store i32 %mul.3, i32 addrspace(1)* %arrayidx9.3, align 4, !tbaa !7
+  %inc12.3 = add nsw i32 %inc1279, 4
+  %.phi.trans.insert74.4 = sext i32 %inc12.3 to i64
+  %arrayidx9.phi.trans.insert.4 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert74.4
+  %.pre75.4 = load i32, i32 addrspace(1)* %arrayidx9.phi.trans.insert.4, align 4, !tbaa !7
+  %117 = sext i32 %inc12.3 to i64
+  %arrayidx9.4 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %117
+  %mul.4 = sub nsw i32 0, %.pre75.4
+  store i32 %mul.4, i32 addrspace(1)* %arrayidx9.4, align 4, !tbaa !7
+  %inc12.4 = add nsw i32 %inc1279, 5
+  %.phi.trans.insert74.5 = sext i32 %inc12.4 to i64
+  %arrayidx9.phi.trans.insert.5 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert74.5
+  %.pre75.5 = load i32, i32 addrspace(1)* %arrayidx9.phi.trans.insert.5, align 4, !tbaa !7
+  %118 = sext i32 %inc12.4 to i64
+  %arrayidx9.5 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %118
+  %mul.5 = sub nsw i32 0, %.pre75.5
+  store i32 %mul.5, i32 addrspace(1)* %arrayidx9.5, align 4, !tbaa !7
+  %inc12.5 = add nsw i32 %inc1279, 6
+  %.phi.trans.insert74.6 = sext i32 %inc12.5 to i64
+  %arrayidx9.phi.trans.insert.6 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert74.6
+  %.pre75.6 = load i32, i32 addrspace(1)* %arrayidx9.phi.trans.insert.6, align 4, !tbaa !7
+  %119 = sext i32 %inc12.5 to i64
+  %arrayidx9.6 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %119
+  %mul.6 = sub nsw i32 0, %.pre75.6
+  store i32 %mul.6, i32 addrspace(1)* %arrayidx9.6, align 4, !tbaa !7
+  %inc12.6 = add nsw i32 %inc1279, 7
+  %.phi.trans.insert74.7 = sext i32 %inc12.6 to i64
+  %arrayidx9.phi.trans.insert.7 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert74.7
+  %.pre75.7 = load i32, i32 addrspace(1)* %arrayidx9.phi.trans.insert.7, align 4, !tbaa !7
+  %120 = sext i32 %inc12.6 to i64
+  %arrayidx9.7 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %120
+  %mul.7 = sub nsw i32 0, %.pre75.7
+  store i32 %mul.7, i32 addrspace(1)* %arrayidx9.7, align 4, !tbaa !7
+  %inc12.7 = add nsw i32 %inc1279, 8
+  %.phi.trans.insert74.8 = sext i32 %inc12.7 to i64
+  %arrayidx9.phi.trans.insert.8 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert74.8
+  %.pre75.8 = load i32, i32 addrspace(1)* %arrayidx9.phi.trans.insert.8, align 4, !tbaa !7
+  %121 = sext i32 %inc12.7 to i64
+  %arrayidx9.8 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %121
+  %mul.8 = sub nsw i32 0, %.pre75.8
+  store i32 %mul.8, i32 addrspace(1)* %arrayidx9.8, align 4, !tbaa !7
+  %inc12.8 = add nsw i32 %inc1279, 9
+  %.phi.trans.insert74.9 = sext i32 %inc12.8 to i64
+  %arrayidx9.phi.trans.insert.9 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert74.9
+  %.pre75.9 = load i32, i32 addrspace(1)* %arrayidx9.phi.trans.insert.9, align 4, !tbaa !7
+  %122 = sext i32 %inc12.8 to i64
+  %arrayidx9.9 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %122
+  %mul.9 = sub nsw i32 0, %.pre75.9
+  store i32 %mul.9, i32 addrspace(1)* %arrayidx9.9, align 4, !tbaa !7
+  %inc12.9 = add nsw i32 %inc1279, 10
+  %.phi.trans.insert74.10 = sext i32 %inc12.9 to i64
+  %arrayidx9.phi.trans.insert.10 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert74.10
+  %.pre75.10 = load i32, i32 addrspace(1)* %arrayidx9.phi.trans.insert.10, align 4, !tbaa !7
+  %123 = sext i32 %inc12.9 to i64
+  %arrayidx9.10 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %123
+  %mul.10 = sub nsw i32 0, %.pre75.10
+  store i32 %mul.10, i32 addrspace(1)* %arrayidx9.10, align 4, !tbaa !7
+  %inc12.10 = add nsw i32 %inc1279, 11
+  %exitcond71.10 = icmp eq i32 %inc12.10, 111
+  br i1 %exitcond71.10, label %if.end.loopexit, label %for.body8.for.body8_crit_edge
+
+for.body18.for.body18_crit_edge:                  ; preds = %for.body18.for.body18_crit_edge, %for.body18.preheader
+  %inc2381 = phi i32 [ 1, %for.body18.preheader ], [ %inc23.10, %for.body18.for.body18_crit_edge ]
+  %.phi.trans.insert = sext i32 %inc2381 to i64
+  %arrayidx19.phi.trans.insert = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert
+  %.pre = load i32, i32 addrspace(1)* %arrayidx19.phi.trans.insert, align 4, !tbaa !7
+  %124 = sext i32 %inc2381 to i64
+  %arrayidx19 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %124
+  %mul20 = mul nsw i32 %.pre, -2
+  store i32 %mul20, i32 addrspace(1)* %arrayidx19, align 4, !tbaa !7
+  %inc23 = add nuw nsw i32 %inc2381, 1
+  %.phi.trans.insert.1 = sext i32 %inc23 to i64
+  %arrayidx19.phi.trans.insert.1 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert.1
+  %.pre.1 = load i32, i32 addrspace(1)* %arrayidx19.phi.trans.insert.1, align 4, !tbaa !7
+  %125 = sext i32 %inc23 to i64
+  %arrayidx19.1 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %125
+  %mul20.1 = mul nsw i32 %.pre.1, -2
+  store i32 %mul20.1, i32 addrspace(1)* %arrayidx19.1, align 4, !tbaa !7
+  %inc23.1 = add nsw i32 %inc2381, 2
+  %.phi.trans.insert.2 = sext i32 %inc23.1 to i64
+  %arrayidx19.phi.trans.insert.2 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert.2
+  %.pre.2 = load i32, i32 addrspace(1)* %arrayidx19.phi.trans.insert.2, align 4, !tbaa !7
+  %126 = sext i32 %inc23.1 to i64
+  %arrayidx19.2 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %126
+  %mul20.2 = mul nsw i32 %.pre.2, -2
+  store i32 %mul20.2, i32 addrspace(1)* %arrayidx19.2, align 4, !tbaa !7
+  %inc23.2 = add nsw i32 %inc2381, 3
+  %.phi.trans.insert.3 = sext i32 %inc23.2 to i64
+  %arrayidx19.phi.trans.insert.3 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert.3
+  %.pre.3 = load i32, i32 addrspace(1)* %arrayidx19.phi.trans.insert.3, align 4, !tbaa !7
+  %127 = sext i32 %inc23.2 to i64
+  %arrayidx19.3 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %127
+  %mul20.3 = mul nsw i32 %.pre.3, -2
+  store i32 %mul20.3, i32 addrspace(1)* %arrayidx19.3, align 4, !tbaa !7
+  %inc23.3 = add nsw i32 %inc2381, 4
+  %.phi.trans.insert.4 = sext i32 %inc23.3 to i64
+  %arrayidx19.phi.trans.insert.4 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert.4
+  %.pre.4 = load i32, i32 addrspace(1)* %arrayidx19.phi.trans.insert.4, align 4, !tbaa !7
+  %128 = sext i32 %inc23.3 to i64
+  %arrayidx19.4 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %128
+  %mul20.4 = mul nsw i32 %.pre.4, -2
+  store i32 %mul20.4, i32 addrspace(1)* %arrayidx19.4, align 4, !tbaa !7
+  %inc23.4 = add nsw i32 %inc2381, 5
+  %.phi.trans.insert.5 = sext i32 %inc23.4 to i64
+  %arrayidx19.phi.trans.insert.5 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert.5
+  %.pre.5 = load i32, i32 addrspace(1)* %arrayidx19.phi.trans.insert.5, align 4, !tbaa !7
+  %129 = sext i32 %inc23.4 to i64
+  %arrayidx19.5 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %129
+  %mul20.5 = mul nsw i32 %.pre.5, -2
+  store i32 %mul20.5, i32 addrspace(1)* %arrayidx19.5, align 4, !tbaa !7
+  %inc23.5 = add nsw i32 %inc2381, 6
+  %.phi.trans.insert.6 = sext i32 %inc23.5 to i64
+  %arrayidx19.phi.trans.insert.6 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert.6
+  %.pre.6 = load i32, i32 addrspace(1)* %arrayidx19.phi.trans.insert.6, align 4, !tbaa !7
+  %130 = sext i32 %inc23.5 to i64
+  %arrayidx19.6 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %130
+  %mul20.6 = mul nsw i32 %.pre.6, -2
+  store i32 %mul20.6, i32 addrspace(1)* %arrayidx19.6, align 4, !tbaa !7
+  %inc23.6 = add nsw i32 %inc2381, 7
+  %.phi.trans.insert.7 = sext i32 %inc23.6 to i64
+  %arrayidx19.phi.trans.insert.7 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert.7
+  %.pre.7 = load i32, i32 addrspace(1)* %arrayidx19.phi.trans.insert.7, align 4, !tbaa !7
+  %131 = sext i32 %inc23.6 to i64
+  %arrayidx19.7 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %131
+  %mul20.7 = mul nsw i32 %.pre.7, -2
+  store i32 %mul20.7, i32 addrspace(1)* %arrayidx19.7, align 4, !tbaa !7
+  %inc23.7 = add nsw i32 %inc2381, 8
+  %.phi.trans.insert.8 = sext i32 %inc23.7 to i64
+  %arrayidx19.phi.trans.insert.8 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert.8
+  %.pre.8 = load i32, i32 addrspace(1)* %arrayidx19.phi.trans.insert.8, align 4, !tbaa !7
+  %132 = sext i32 %inc23.7 to i64
+  %arrayidx19.8 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %132
+  %mul20.8 = mul nsw i32 %.pre.8, -2
+  store i32 %mul20.8, i32 addrspace(1)* %arrayidx19.8, align 4, !tbaa !7
+  %inc23.8 = add nsw i32 %inc2381, 9
+  %.phi.trans.insert.9 = sext i32 %inc23.8 to i64
+  %arrayidx19.phi.trans.insert.9 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert.9
+  %.pre.9 = load i32, i32 addrspace(1)* %arrayidx19.phi.trans.insert.9, align 4, !tbaa !7
+  %133 = sext i32 %inc23.8 to i64
+  %arrayidx19.9 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %133
+  %mul20.9 = mul nsw i32 %.pre.9, -2
+  store i32 %mul20.9, i32 addrspace(1)* %arrayidx19.9, align 4, !tbaa !7
+  %inc23.9 = add nsw i32 %inc2381, 10
+  %.phi.trans.insert.10 = sext i32 %inc23.9 to i64
+  %arrayidx19.phi.trans.insert.10 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %.phi.trans.insert.10
+  %.pre.10 = load i32, i32 addrspace(1)* %arrayidx19.phi.trans.insert.10, align 4, !tbaa !7
+  %134 = sext i32 %inc23.9 to i64
+  %arrayidx19.10 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %134
+  %mul20.10 = mul nsw i32 %.pre.10, -2
+  store i32 %mul20.10, i32 addrspace(1)* %arrayidx19.10, align 4, !tbaa !7
+  %inc23.10 = add nsw i32 %inc2381, 11
+  %exitcond72.10 = icmp eq i32 %inc23.10, 111
+  br i1 %exitcond72.10, label %if.end.loopexit82, label %for.body18.for.body18_crit_edge
+
+if.end.loopexit:                                  ; preds = %for.body8.for.body8_crit_edge
+  br label %if.end
+
+if.end.loopexit82:                                ; preds = %for.body18.for.body18_crit_edge
+  br label %if.end
+
+if.end:                                           ; preds = %if.end.loopexit82, %if.end.loopexit
+  %val.0 = phi i32 [ 779, %if.end.loopexit ], [ -775, %if.end.loopexit82 ]
+  br label %for.body30
+
+for.cond.cleanup29:                               ; preds = %for.body30
+  call void @llvm.lifetime.end(i64 888, i8* nonnull %0) #2
+  ret void
+
+for.body30:                                       ; preds = %for.body30, %if.end
+  %i26.066 = phi i32 [ 0, %if.end ], [ %inc38.2, %for.body30 ]
+  %arrayidx31 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 %i26.066
+  %135 = load i32, i32* %arrayidx31, align 4, !tbaa !7
+  %136 = sext i32 %i26.066 to i64
+  %arrayidx32 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %136
+  %137 = load i32, i32 addrspace(1)* %arrayidx32, align 4, !tbaa !7
+  %mul33 = mul nsw i32 %137, %135
+  %add35 = add i32 %val.0, %mul33
+  %arrayidx36 = getelementptr inbounds i32, i32 addrspace(1)* %B, i64 %136
+  store i32 %add35, i32 addrspace(1)* %arrayidx36, align 4, !tbaa !7
+  %inc38 = add nuw nsw i32 %i26.066, 1
+  %arrayidx31.1 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 %inc38
+  %138 = load i32, i32* %arrayidx31.1, align 4, !tbaa !7
+  %139 = sext i32 %inc38 to i64
+  %arrayidx32.1 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %139
+  %140 = load i32, i32 addrspace(1)* %arrayidx32.1, align 4, !tbaa !7
+  %mul33.1 = mul nsw i32 %140, %138
+  %add35.1 = add i32 %val.0, %mul33.1
+  %arrayidx36.1 = getelementptr inbounds i32, i32 addrspace(1)* %B, i64 %139
+  store i32 %add35.1, i32 addrspace(1)* %arrayidx36.1, align 4, !tbaa !7
+  %inc38.1 = add nsw i32 %i26.066, 2
+  %arrayidx31.2 = getelementptr inbounds [222 x i32], [222 x i32]* %arr, i32 0, i32 %inc38.1
+  %141 = load i32, i32* %arrayidx31.2, align 4, !tbaa !7
+  %142 = sext i32 %inc38.1 to i64
+  %arrayidx32.2 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %142
+  %143 = load i32, i32 addrspace(1)* %arrayidx32.2, align 4, !tbaa !7
+  %mul33.2 = mul nsw i32 %143, %141
+  %add35.2 = add i32 %val.0, %mul33.2
+  %arrayidx36.2 = getelementptr inbounds i32, i32 addrspace(1)* %B, i64 %142
+  store i32 %add35.2, i32 addrspace(1)* %arrayidx36.2, align 4, !tbaa !7
+  %inc38.2 = add nsw i32 %i26.066, 3
+  %exitcond.2 = icmp eq i32 %inc38.2, 111
+  br i1 %exitcond.2, label %for.cond.cleanup29, label %for.body30
+}
+
+; Function Attrs: argmemonly nounwind
+declare void @llvm.lifetime.start(i64, i8* nocapture) #1
+
+; Function Attrs: argmemonly nounwind
+declare void @llvm.lifetime.end(i64, i8* nocapture) #1
+
+attributes #0 = { nounwind "amdgpu-num-sgpr"="16" "amdgpu-num-active-waves-per-eu"="1,10" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="fiji" "target-features"="+16-bit-insts,+s-memrealtime" "unsafe-fp-math"="false" "use-soft-float"="false" }
+attributes #1 = { argmemonly nounwind }
+attributes #2 = { nounwind }
+
+!opencl.kernels = !{!0}
+!llvm.ident = !{!6}
+
+!0 = !{void (i32 addrspace(1)*, i32 addrspace(1)*)* @attr_amdgpu_num_sgpr_16, !1, !2, !3, !4, !5}
+!1 = !{!"kernel_arg_addr_space", i32 1, i32 1}
+!2 = !{!"kernel_arg_access_qual", !"none", !"none"}
+!3 = !{!"kernel_arg_type", !"int*", !"int*"}
+!4 = !{!"kernel_arg_base_type", !"int*", !"int*"}
+!5 = !{!"kernel_arg_type_qual", !"", !""}
+!6 = !{!"clang version 3.9.0 (trunk 272917)"}
+!7 = !{!8, !8, i64 0}
+!8 = !{!"int", !9, i64 0}
+!9 = !{!"omnipotent char", !10, i64 0}
+!10 = !{!"Simple C/C++ TBAA"}

test/CodeGen/AMDGPU/attr-amdgpu-num-vgpr-invalid.ll

+; RUN: not llc -march=amdgcn -mcpu=fiji -verify-machineinstrs < %s 2>&1 | FileCheck %s
+
+; For fiji, 8 active waves per execution unit -> [29, 32] VGPRs can be used.
+
+; CHECK: invalid amdgpu-num-vgpr: requested number of VGPRs cannot be achieved with default or requested maximum number of active waves per execution unit
+define void @invalid_0() #0 {
+entry:
+  ret void
+}
+attributes #0 = {"amdgpu-num-vgpr"="28" "amdgpu-num-active-waves-per-eu"="8,8"}
+
+; CHECK: invalid amdgpu-num-vgpr: requested number of VGPRs cannot be achieved with default or requested minimum number of active waves per execution unit
+define void @invalid_1() #1 {
+entry:
+  ret void
+}
+attributes #1 = {"amdgpu-num-vgpr"="33" "amdgpu-num-active-waves-per-eu"="8,8"}

test/CodeGen/AMDGPU/attr-amdgpu-num-vgpr.ll

+; RUN: llc -march=amdgcn -mcpu=fiji -verify-machineinstrs < %s | FileCheck %s
+
+; ModuleID = 'attr-amdgpu-num-vgpr.cl'
+source_filename = "attr-amdgpu-num-vgpr.cl"
+target datalayout = "e-p:32:32-p1:64:64-p2:64:64-p3:32:32-p4:64:64-p5:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64"
+target triple = "amdgcn--amdhsa"
+
+; CHECK: NumVgprs: 16
+define void @attr_amdgpu_num_vgpr_16(i32 addrspace(1)* nocapture %A) local_unnamed_addr #0 {
+entry:
+  %arr = alloca [111 x i32], align 4
+  %0 = bitcast [111 x i32]* %arr to i8*
+  call void @llvm.lifetime.start(i64 444, i8* %0) #2
+  %1 = load i32, i32 addrspace(1)* %A, align 4, !tbaa !7
+  %arrayidx1 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 0
+  store i32 %1, i32* %arrayidx1, align 4, !tbaa !7
+  %arrayidx.1 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 1
+  %2 = load i32, i32 addrspace(1)* %arrayidx.1, align 4, !tbaa !7
+  %arrayidx1.1 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 1
+  store i32 %2, i32* %arrayidx1.1, align 4, !tbaa !7
+  %arrayidx.2 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 2
+  %3 = load i32, i32 addrspace(1)* %arrayidx.2, align 4, !tbaa !7
+  %arrayidx1.2 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 2
+  store i32 %3, i32* %arrayidx1.2, align 4, !tbaa !7
+  %arrayidx.3 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 3
+  %4 = load i32, i32 addrspace(1)* %arrayidx.3, align 4, !tbaa !7
+  %arrayidx1.3 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 3
+  store i32 %4, i32* %arrayidx1.3, align 4, !tbaa !7
+  %arrayidx.4 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 4
+  %5 = load i32, i32 addrspace(1)* %arrayidx.4, align 4, !tbaa !7
+  %arrayidx1.4 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 4
+  store i32 %5, i32* %arrayidx1.4, align 4, !tbaa !7
+  %arrayidx.5 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 5
+  %6 = load i32, i32 addrspace(1)* %arrayidx.5, align 4, !tbaa !7
+  %arrayidx1.5 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 5
+  store i32 %6, i32* %arrayidx1.5, align 4, !tbaa !7
+  %arrayidx.6 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 6
+  %7 = load i32, i32 addrspace(1)* %arrayidx.6, align 4, !tbaa !7
+  %arrayidx1.6 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 6
+  store i32 %7, i32* %arrayidx1.6, align 4, !tbaa !7
+  %arrayidx.7 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 7
+  %8 = load i32, i32 addrspace(1)* %arrayidx.7, align 4, !tbaa !7
+  %arrayidx1.7 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 7
+  store i32 %8, i32* %arrayidx1.7, align 4, !tbaa !7
+  %arrayidx.8 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 8
+  %9 = load i32, i32 addrspace(1)* %arrayidx.8, align 4, !tbaa !7
+  %arrayidx1.8 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 8
+  store i32 %9, i32* %arrayidx1.8, align 4, !tbaa !7
+  %arrayidx.9 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 9
+  %10 = load i32, i32 addrspace(1)* %arrayidx.9, align 4, !tbaa !7
+  %arrayidx1.9 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 9
+  store i32 %10, i32* %arrayidx1.9, align 4, !tbaa !7
+  %arrayidx.10 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 10
+  %11 = load i32, i32 addrspace(1)* %arrayidx.10, align 4, !tbaa !7
+  %arrayidx1.10 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 10
+  store i32 %11, i32* %arrayidx1.10, align 4, !tbaa !7
+  %arrayidx.11 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 11
+  %12 = load i32, i32 addrspace(1)* %arrayidx.11, align 4, !tbaa !7
+  %arrayidx1.11 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 11
+  store i32 %12, i32* %arrayidx1.11, align 4, !tbaa !7
+  %arrayidx.12 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 12
+  %13 = load i32, i32 addrspace(1)* %arrayidx.12, align 4, !tbaa !7
+  %arrayidx1.12 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 12
+  store i32 %13, i32* %arrayidx1.12, align 4, !tbaa !7
+  %arrayidx.13 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 13
+  %14 = load i32, i32 addrspace(1)* %arrayidx.13, align 4, !tbaa !7
+  %arrayidx1.13 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 13
+  store i32 %14, i32* %arrayidx1.13, align 4, !tbaa !7
+  %arrayidx.14 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 14
+  %15 = load i32, i32 addrspace(1)* %arrayidx.14, align 4, !tbaa !7
+  %arrayidx1.14 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 14
+  store i32 %15, i32* %arrayidx1.14, align 4, !tbaa !7
+  %arrayidx.15 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 15
+  %16 = load i32, i32 addrspace(1)* %arrayidx.15, align 4, !tbaa !7
+  %arrayidx1.15 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 15
+  store i32 %16, i32* %arrayidx1.15, align 4, !tbaa !7
+  %arrayidx.16 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 16
+  %17 = load i32, i32 addrspace(1)* %arrayidx.16, align 4, !tbaa !7
+  %arrayidx1.16 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 16
+  store i32 %17, i32* %arrayidx1.16, align 4, !tbaa !7
+  %arrayidx.17 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 17
+  %18 = load i32, i32 addrspace(1)* %arrayidx.17, align 4, !tbaa !7
+  %arrayidx1.17 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 17
+  store i32 %18, i32* %arrayidx1.17, align 4, !tbaa !7
+  %arrayidx.18 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 18
+  %19 = load i32, i32 addrspace(1)* %arrayidx.18, align 4, !tbaa !7
+  %arrayidx1.18 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 18
+  store i32 %19, i32* %arrayidx1.18, align 4, !tbaa !7
+  %arrayidx.19 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 19
+  %20 = load i32, i32 addrspace(1)* %arrayidx.19, align 4, !tbaa !7
+  %arrayidx1.19 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 19
+  store i32 %20, i32* %arrayidx1.19, align 4, !tbaa !7
+  %arrayidx.20 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 20
+  %21 = load i32, i32 addrspace(1)* %arrayidx.20, align 4, !tbaa !7
+  %arrayidx1.20 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 20
+  store i32 %21, i32* %arrayidx1.20, align 4, !tbaa !7
+  %arrayidx.21 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 21
+  %22 = load i32, i32 addrspace(1)* %arrayidx.21, align 4, !tbaa !7
+  %arrayidx1.21 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 21
+  store i32 %22, i32* %arrayidx1.21, align 4, !tbaa !7
+  %arrayidx.22 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 22
+  %23 = load i32, i32 addrspace(1)* %arrayidx.22, align 4, !tbaa !7
+  %arrayidx1.22 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 22
+  store i32 %23, i32* %arrayidx1.22, align 4, !tbaa !7
+  %arrayidx.23 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 23
+  %24 = load i32, i32 addrspace(1)* %arrayidx.23, align 4, !tbaa !7
+  %arrayidx1.23 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 23
+  store i32 %24, i32* %arrayidx1.23, align 4, !tbaa !7
+  %arrayidx.24 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 24
+  %25 = load i32, i32 addrspace(1)* %arrayidx.24, align 4, !tbaa !7
+  %arrayidx1.24 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 24
+  store i32 %25, i32* %arrayidx1.24, align 4, !tbaa !7
+  %arrayidx.25 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 25
+  %26 = load i32, i32 addrspace(1)* %arrayidx.25, align 4, !tbaa !7
+  %arrayidx1.25 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 25
+  store i32 %26, i32* %arrayidx1.25, align 4, !tbaa !7
+  %arrayidx.26 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 26
+  %27 = load i32, i32 addrspace(1)* %arrayidx.26, align 4, !tbaa !7
+  %arrayidx1.26 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 26
+  store i32 %27, i32* %arrayidx1.26, align 4, !tbaa !7
+  %arrayidx.27 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 27
+  %28 = load i32, i32 addrspace(1)* %arrayidx.27, align 4, !tbaa !7
+  %arrayidx1.27 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 27
+  store i32 %28, i32* %arrayidx1.27, align 4, !tbaa !7
+  %arrayidx.28 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 28
+  %29 = load i32, i32 addrspace(1)* %arrayidx.28, align 4, !tbaa !7
+  %arrayidx1.28 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 28
+  store i32 %29, i32* %arrayidx1.28, align 4, !tbaa !7
+  %arrayidx.29 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 29
+  %30 = load i32, i32 addrspace(1)* %arrayidx.29, align 4, !tbaa !7
+  %arrayidx1.29 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 29
+  store i32 %30, i32* %arrayidx1.29, align 4, !tbaa !7
+  %arrayidx.30 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 30
+  %31 = load i32, i32 addrspace(1)* %arrayidx.30, align 4, !tbaa !7
+  %arrayidx1.30 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 30
+  store i32 %31, i32* %arrayidx1.30, align 4, !tbaa !7
+  %arrayidx.31 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 31
+  %32 = load i32, i32 addrspace(1)* %arrayidx.31, align 4, !tbaa !7
+  %arrayidx1.31 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 31
+  store i32 %32, i32* %arrayidx1.31, align 4, !tbaa !7
+  %arrayidx.32 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 32
+  %33 = load i32, i32 addrspace(1)* %arrayidx.32, align 4, !tbaa !7
+  %arrayidx1.32 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 32
+  store i32 %33, i32* %arrayidx1.32, align 4, !tbaa !7
+  %arrayidx.33 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 33
+  %34 = load i32, i32 addrspace(1)* %arrayidx.33, align 4, !tbaa !7
+  %arrayidx1.33 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 33
+  store i32 %34, i32* %arrayidx1.33, align 4, !tbaa !7
+  %arrayidx.34 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 34
+  %35 = load i32, i32 addrspace(1)* %arrayidx.34, align 4, !tbaa !7
+  %arrayidx1.34 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 34
+  store i32 %35, i32* %arrayidx1.34, align 4, !tbaa !7
+  %arrayidx.35 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 35
+  %36 = load i32, i32 addrspace(1)* %arrayidx.35, align 4, !tbaa !7
+  %arrayidx1.35 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 35
+  store i32 %36, i32* %arrayidx1.35, align 4, !tbaa !7
+  %arrayidx.36 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 36
+  %37 = load i32, i32 addrspace(1)* %arrayidx.36, align 4, !tbaa !7
+  %arrayidx1.36 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 36
+  store i32 %37, i32* %arrayidx1.36, align 4, !tbaa !7
+  %arrayidx.37 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 37
+  %38 = load i32, i32 addrspace(1)* %arrayidx.37, align 4, !tbaa !7
+  %arrayidx1.37 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 37
+  store i32 %38, i32* %arrayidx1.37, align 4, !tbaa !7
+  %arrayidx.38 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 38
+  %39 = load i32, i32 addrspace(1)* %arrayidx.38, align 4, !tbaa !7
+  %arrayidx1.38 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 38
+  store i32 %39, i32* %arrayidx1.38, align 4, !tbaa !7
+  %arrayidx.39 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 39
+  %40 = load i32, i32 addrspace(1)* %arrayidx.39, align 4, !tbaa !7
+  %arrayidx1.39 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 39
+  store i32 %40, i32* %arrayidx1.39, align 4, !tbaa !7
+  %arrayidx.40 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 40
+  %41 = load i32, i32 addrspace(1)* %arrayidx.40, align 4, !tbaa !7
+  %arrayidx1.40 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 40
+  store i32 %41, i32* %arrayidx1.40, align 4, !tbaa !7
+  %arrayidx.41 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 41
+  %42 = load i32, i32 addrspace(1)* %arrayidx.41, align 4, !tbaa !7
+  %arrayidx1.41 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 41
+  store i32 %42, i32* %arrayidx1.41, align 4, !tbaa !7
+  %arrayidx.42 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 42
+  %43 = load i32, i32 addrspace(1)* %arrayidx.42, align 4, !tbaa !7
+  %arrayidx1.42 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 42
+  store i32 %43, i32* %arrayidx1.42, align 4, !tbaa !7
+  %arrayidx.43 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 43
+  %44 = load i32, i32 addrspace(1)* %arrayidx.43, align 4, !tbaa !7
+  %arrayidx1.43 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 43
+  store i32 %44, i32* %arrayidx1.43, align 4, !tbaa !7
+  %arrayidx.44 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 44
+  %45 = load i32, i32 addrspace(1)* %arrayidx.44, align 4, !tbaa !7
+  %arrayidx1.44 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 44
+  store i32 %45, i32* %arrayidx1.44, align 4, !tbaa !7
+  %arrayidx.45 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 45
+  %46 = load i32, i32 addrspace(1)* %arrayidx.45, align 4, !tbaa !7
+  %arrayidx1.45 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 45
+  store i32 %46, i32* %arrayidx1.45, align 4, !tbaa !7
+  %arrayidx.46 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 46
+  %47 = load i32, i32 addrspace(1)* %arrayidx.46, align 4, !tbaa !7
+  %arrayidx1.46 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 46
+  store i32 %47, i32* %arrayidx1.46, align 4, !tbaa !7
+  %arrayidx.47 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 47
+  %48 = load i32, i32 addrspace(1)* %arrayidx.47, align 4, !tbaa !7
+  %arrayidx1.47 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 47
+  store i32 %48, i32* %arrayidx1.47, align 4, !tbaa !7
+  %arrayidx.48 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 48
+  %49 = load i32, i32 addrspace(1)* %arrayidx.48, align 4, !tbaa !7
+  %arrayidx1.48 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 48
+  store i32 %49, i32* %arrayidx1.48, align 4, !tbaa !7
+  %arrayidx.49 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 49
+  %50 = load i32, i32 addrspace(1)* %arrayidx.49, align 4, !tbaa !7
+  %arrayidx1.49 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 49
+  store i32 %50, i32* %arrayidx1.49, align 4, !tbaa !7
+  %arrayidx.50 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 50
+  %51 = load i32, i32 addrspace(1)* %arrayidx.50, align 4, !tbaa !7
+  %arrayidx1.50 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 50
+  store i32 %51, i32* %arrayidx1.50, align 4, !tbaa !7
+  %arrayidx.51 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 51
+  %52 = load i32, i32 addrspace(1)* %arrayidx.51, align 4, !tbaa !7
+  %arrayidx1.51 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 51
+  store i32 %52, i32* %arrayidx1.51, align 4, !tbaa !7
+  %arrayidx.52 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 52
+  %53 = load i32, i32 addrspace(1)* %arrayidx.52, align 4, !tbaa !7
+  %arrayidx1.52 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 52
+  store i32 %53, i32* %arrayidx1.52, align 4, !tbaa !7
+  %arrayidx.53 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 53
+  %54 = load i32, i32 addrspace(1)* %arrayidx.53, align 4, !tbaa !7
+  %arrayidx1.53 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 53
+  store i32 %54, i32* %arrayidx1.53, align 4, !tbaa !7
+  %arrayidx.54 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 54
+  %55 = load i32, i32 addrspace(1)* %arrayidx.54, align 4, !tbaa !7
+  %arrayidx1.54 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 54
+  store i32 %55, i32* %arrayidx1.54, align 4, !tbaa !7
+  %arrayidx.55 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 55
+  %56 = load i32, i32 addrspace(1)* %arrayidx.55, align 4, !tbaa !7
+  %arrayidx1.55 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 55
+  store i32 %56, i32* %arrayidx1.55, align 4, !tbaa !7
+  %arrayidx.56 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 56
+  %57 = load i32, i32 addrspace(1)* %arrayidx.56, align 4, !tbaa !7
+  %arrayidx1.56 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 56
+  store i32 %57, i32* %arrayidx1.56, align 4, !tbaa !7
+  %arrayidx.57 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 57
+  %58 = load i32, i32 addrspace(1)* %arrayidx.57, align 4, !tbaa !7
+  %arrayidx1.57 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 57
+  store i32 %58, i32* %arrayidx1.57, align 4, !tbaa !7
+  %arrayidx.58 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 58
+  %59 = load i32, i32 addrspace(1)* %arrayidx.58, align 4, !tbaa !7
+  %arrayidx1.58 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 58
+  store i32 %59, i32* %arrayidx1.58, align 4, !tbaa !7
+  %arrayidx.59 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 59
+  %60 = load i32, i32 addrspace(1)* %arrayidx.59, align 4, !tbaa !7
+  %arrayidx1.59 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 59
+  store i32 %60, i32* %arrayidx1.59, align 4, !tbaa !7
+  %arrayidx.60 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 60
+  %61 = load i32, i32 addrspace(1)* %arrayidx.60, align 4, !tbaa !7
+  %arrayidx1.60 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 60
+  store i32 %61, i32* %arrayidx1.60, align 4, !tbaa !7
+  %arrayidx.61 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 61
+  %62 = load i32, i32 addrspace(1)* %arrayidx.61, align 4, !tbaa !7
+  %arrayidx1.61 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 61
+  store i32 %62, i32* %arrayidx1.61, align 4, !tbaa !7
+  %arrayidx.62 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 62
+  %63 = load i32, i32 addrspace(1)* %arrayidx.62, align 4, !tbaa !7
+  %arrayidx1.62 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 62
+  store i32 %63, i32* %arrayidx1.62, align 4, !tbaa !7
+  %arrayidx.63 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 63
+  %64 = load i32, i32 addrspace(1)* %arrayidx.63, align 4, !tbaa !7
+  %arrayidx1.63 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 63
+  store i32 %64, i32* %arrayidx1.63, align 4, !tbaa !7
+  %arrayidx.64 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 64
+  %65 = load i32, i32 addrspace(1)* %arrayidx.64, align 4, !tbaa !7
+  %arrayidx1.64 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 64
+  store i32 %65, i32* %arrayidx1.64, align 4, !tbaa !7
+  %arrayidx.65 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 65
+  %66 = load i32, i32 addrspace(1)* %arrayidx.65, align 4, !tbaa !7
+  %arrayidx1.65 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 65
+  store i32 %66, i32* %arrayidx1.65, align 4, !tbaa !7
+  %arrayidx.66 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 66
+  %67 = load i32, i32 addrspace(1)* %arrayidx.66, align 4, !tbaa !7
+  %arrayidx1.66 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 66
+  store i32 %67, i32* %arrayidx1.66, align 4, !tbaa !7
+  %arrayidx.67 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 67
+  %68 = load i32, i32 addrspace(1)* %arrayidx.67, align 4, !tbaa !7
+  %arrayidx1.67 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 67
+  store i32 %68, i32* %arrayidx1.67, align 4, !tbaa !7
+  %arrayidx.68 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 68
+  %69 = load i32, i32 addrspace(1)* %arrayidx.68, align 4, !tbaa !7
+  %arrayidx1.68 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 68
+  store i32 %69, i32* %arrayidx1.68, align 4, !tbaa !7
+  %arrayidx.69 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 69
+  %70 = load i32, i32 addrspace(1)* %arrayidx.69, align 4, !tbaa !7
+  %arrayidx1.69 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 69
+  store i32 %70, i32* %arrayidx1.69, align 4, !tbaa !7
+  %arrayidx.70 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 70
+  %71 = load i32, i32 addrspace(1)* %arrayidx.70, align 4, !tbaa !7
+  %arrayidx1.70 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 70
+  store i32 %71, i32* %arrayidx1.70, align 4, !tbaa !7
+  %arrayidx.71 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 71
+  %72 = load i32, i32 addrspace(1)* %arrayidx.71, align 4, !tbaa !7
+  %arrayidx1.71 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 71
+  store i32 %72, i32* %arrayidx1.71, align 4, !tbaa !7
+  %arrayidx.72 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 72
+  %73 = load i32, i32 addrspace(1)* %arrayidx.72, align 4, !tbaa !7
+  %arrayidx1.72 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 72
+  store i32 %73, i32* %arrayidx1.72, align 4, !tbaa !7
+  %arrayidx.73 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 73
+  %74 = load i32, i32 addrspace(1)* %arrayidx.73, align 4, !tbaa !7
+  %arrayidx1.73 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 73
+  store i32 %74, i32* %arrayidx1.73, align 4, !tbaa !7
+  %arrayidx.74 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 74
+  %75 = load i32, i32 addrspace(1)* %arrayidx.74, align 4, !tbaa !7
+  %arrayidx1.74 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 74
+  store i32 %75, i32* %arrayidx1.74, align 4, !tbaa !7
+  %arrayidx.75 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 75
+  %76 = load i32, i32 addrspace(1)* %arrayidx.75, align 4, !tbaa !7
+  %arrayidx1.75 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 75
+  store i32 %76, i32* %arrayidx1.75, align 4, !tbaa !7
+  %arrayidx.76 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 76
+  %77 = load i32, i32 addrspace(1)* %arrayidx.76, align 4, !tbaa !7
+  %arrayidx1.76 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 76
+  store i32 %77, i32* %arrayidx1.76, align 4, !tbaa !7
+  %arrayidx.77 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 77
+  %78 = load i32, i32 addrspace(1)* %arrayidx.77, align 4, !tbaa !7
+  %arrayidx1.77 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 77
+  store i32 %78, i32* %arrayidx1.77, align 4, !tbaa !7
+  %arrayidx.78 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 78
+  %79 = load i32, i32 addrspace(1)* %arrayidx.78, align 4, !tbaa !7
+  %arrayidx1.78 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 78
+  store i32 %79, i32* %arrayidx1.78, align 4, !tbaa !7
+  %arrayidx.79 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 79
+  %80 = load i32, i32 addrspace(1)* %arrayidx.79, align 4, !tbaa !7
+  %arrayidx1.79 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 79
+  store i32 %80, i32* %arrayidx1.79, align 4, !tbaa !7
+  %arrayidx.80 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 80
+  %81 = load i32, i32 addrspace(1)* %arrayidx.80, align 4, !tbaa !7
+  %arrayidx1.80 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 80
+  store i32 %81, i32* %arrayidx1.80, align 4, !tbaa !7
+  %arrayidx.81 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 81
+  %82 = load i32, i32 addrspace(1)* %arrayidx.81, align 4, !tbaa !7
+  %arrayidx1.81 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 81
+  store i32 %82, i32* %arrayidx1.81, align 4, !tbaa !7
+  %arrayidx.82 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 82
+  %83 = load i32, i32 addrspace(1)* %arrayidx.82, align 4, !tbaa !7
+  %arrayidx1.82 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 82
+  store i32 %83, i32* %arrayidx1.82, align 4, !tbaa !7
+  %arrayidx.83 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 83
+  %84 = load i32, i32 addrspace(1)* %arrayidx.83, align 4, !tbaa !7
+  %arrayidx1.83 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 83
+  store i32 %84, i32* %arrayidx1.83, align 4, !tbaa !7
+  %arrayidx.84 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 84
+  %85 = load i32, i32 addrspace(1)* %arrayidx.84, align 4, !tbaa !7
+  %arrayidx1.84 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 84
+  store i32 %85, i32* %arrayidx1.84, align 4, !tbaa !7
+  %arrayidx.85 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 85
+  %86 = load i32, i32 addrspace(1)* %arrayidx.85, align 4, !tbaa !7
+  %arrayidx1.85 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 85
+  store i32 %86, i32* %arrayidx1.85, align 4, !tbaa !7
+  %arrayidx.86 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 86
+  %87 = load i32, i32 addrspace(1)* %arrayidx.86, align 4, !tbaa !7
+  %arrayidx1.86 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 86
+  store i32 %87, i32* %arrayidx1.86, align 4, !tbaa !7
+  %arrayidx.87 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 87
+  %88 = load i32, i32 addrspace(1)* %arrayidx.87, align 4, !tbaa !7
+  %arrayidx1.87 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 87
+  store i32 %88, i32* %arrayidx1.87, align 4, !tbaa !7
+  %arrayidx.88 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 88
+  %89 = load i32, i32 addrspace(1)* %arrayidx.88, align 4, !tbaa !7
+  %arrayidx1.88 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 88
+  store i32 %89, i32* %arrayidx1.88, align 4, !tbaa !7
+  %arrayidx.89 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 89
+  %90 = load i32, i32 addrspace(1)* %arrayidx.89, align 4, !tbaa !7
+  %arrayidx1.89 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 89
+  store i32 %90, i32* %arrayidx1.89, align 4, !tbaa !7
+  %arrayidx.90 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 90
+  %91 = load i32, i32 addrspace(1)* %arrayidx.90, align 4, !tbaa !7
+  %arrayidx1.90 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 90
+  store i32 %91, i32* %arrayidx1.90, align 4, !tbaa !7
+  %arrayidx.91 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 91
+  %92 = load i32, i32 addrspace(1)* %arrayidx.91, align 4, !tbaa !7
+  %arrayidx1.91 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 91
+  store i32 %92, i32* %arrayidx1.91, align 4, !tbaa !7
+  %arrayidx.92 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 92
+  %93 = load i32, i32 addrspace(1)* %arrayidx.92, align 4, !tbaa !7
+  %arrayidx1.92 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 92
+  store i32 %93, i32* %arrayidx1.92, align 4, !tbaa !7
+  %arrayidx.93 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 93
+  %94 = load i32, i32 addrspace(1)* %arrayidx.93, align 4, !tbaa !7
+  %arrayidx1.93 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 93
+  store i32 %94, i32* %arrayidx1.93, align 4, !tbaa !7
+  %arrayidx.94 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 94
+  %95 = load i32, i32 addrspace(1)* %arrayidx.94, align 4, !tbaa !7
+  %arrayidx1.94 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 94
+  store i32 %95, i32* %arrayidx1.94, align 4, !tbaa !7
+  %arrayidx.95 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 95
+  %96 = load i32, i32 addrspace(1)* %arrayidx.95, align 4, !tbaa !7
+  %arrayidx1.95 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 95
+  store i32 %96, i32* %arrayidx1.95, align 4, !tbaa !7
+  %arrayidx.96 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 96
+  %97 = load i32, i32 addrspace(1)* %arrayidx.96, align 4, !tbaa !7
+  %arrayidx1.96 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 96
+  store i32 %97, i32* %arrayidx1.96, align 4, !tbaa !7
+  %arrayidx.97 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 97
+  %98 = load i32, i32 addrspace(1)* %arrayidx.97, align 4, !tbaa !7
+  %arrayidx1.97 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 97
+  store i32 %98, i32* %arrayidx1.97, align 4, !tbaa !7
+  %arrayidx.98 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 98
+  %99 = load i32, i32 addrspace(1)* %arrayidx.98, align 4, !tbaa !7
+  %arrayidx1.98 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 98
+  store i32 %99, i32* %arrayidx1.98, align 4, !tbaa !7
+  %arrayidx.99 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 99
+  %100 = load i32, i32 addrspace(1)* %arrayidx.99, align 4, !tbaa !7
+  %arrayidx1.99 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 99
+  store i32 %100, i32* %arrayidx1.99, align 4, !tbaa !7
+  %arrayidx.100 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 100
+  %101 = load i32, i32 addrspace(1)* %arrayidx.100, align 4, !tbaa !7
+  %arrayidx1.100 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 100
+  store i32 %101, i32* %arrayidx1.100, align 4, !tbaa !7
+  %arrayidx.101 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 101
+  %102 = load i32, i32 addrspace(1)* %arrayidx.101, align 4, !tbaa !7
+  %arrayidx1.101 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 101
+  store i32 %102, i32* %arrayidx1.101, align 4, !tbaa !7
+  %arrayidx.102 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 102
+  %103 = load i32, i32 addrspace(1)* %arrayidx.102, align 4, !tbaa !7
+  %arrayidx1.102 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 102
+  store i32 %103, i32* %arrayidx1.102, align 4, !tbaa !7
+  %arrayidx.103 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 103
+  %104 = load i32, i32 addrspace(1)* %arrayidx.103, align 4, !tbaa !7
+  %arrayidx1.103 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 103
+  store i32 %104, i32* %arrayidx1.103, align 4, !tbaa !7
+  %arrayidx.104 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 104
+  %105 = load i32, i32 addrspace(1)* %arrayidx.104, align 4, !tbaa !7
+  %arrayidx1.104 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 104
+  store i32 %105, i32* %arrayidx1.104, align 4, !tbaa !7
+  %arrayidx.105 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 105
+  %106 = load i32, i32 addrspace(1)* %arrayidx.105, align 4, !tbaa !7
+  %arrayidx1.105 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 105
+  store i32 %106, i32* %arrayidx1.105, align 4, !tbaa !7
+  %arrayidx.106 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 106
+  %107 = load i32, i32 addrspace(1)* %arrayidx.106, align 4, !tbaa !7
+  %arrayidx1.106 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 106
+  store i32 %107, i32* %arrayidx1.106, align 4, !tbaa !7
+  %arrayidx.107 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 107
+  %108 = load i32, i32 addrspace(1)* %arrayidx.107, align 4, !tbaa !7
+  %arrayidx1.107 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 107
+  store i32 %108, i32* %arrayidx1.107, align 4, !tbaa !7
+  %arrayidx.108 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 108
+  %109 = load i32, i32 addrspace(1)* %arrayidx.108, align 4, !tbaa !7
+  %arrayidx1.108 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 108
+  store i32 %109, i32* %arrayidx1.108, align 4, !tbaa !7
+  %arrayidx.109 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 109
+  %110 = load i32, i32 addrspace(1)* %arrayidx.109, align 4, !tbaa !7
+  %arrayidx1.109 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 109
+  store i32 %110, i32* %arrayidx1.109, align 4, !tbaa !7
+  %arrayidx.110 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 110
+  %111 = load i32, i32 addrspace(1)* %arrayidx.110, align 4, !tbaa !7
+  %arrayidx1.110 = getelementptr inbounds [111 x i32], [111 x i32]* %arr, i32 0, i32 110
+  store i32 %111, i32* %arrayidx1.110, align 4, !tbaa !7
+  %112 = load i32, i32* %arrayidx1, align 4, !tbaa !7
+  %mul = sub nsw i32 0, %112
+  store i32 %mul, i32 addrspace(1)* %A, align 4, !tbaa !7
+  %113 = load i32, i32* %arrayidx1.1, align 4, !tbaa !7
+  %mul.1 = sub nsw i32 0, %113
+  store i32 %mul.1, i32 addrspace(1)* %arrayidx.1, align 4, !tbaa !7
+  %114 = load i32, i32* %arrayidx1.2, align 4, !tbaa !7
+  %mul.2 = sub nsw i32 0, %114
+  store i32 %mul.2, i32 addrspace(1)* %arrayidx.2, align 4, !tbaa !7
+  %115 = load i32, i32* %arrayidx1.3, align 4, !tbaa !7
+  %mul.3 = sub nsw i32 0, %115
+  store i32 %mul.3, i32 addrspace(1)* %arrayidx.3, align 4, !tbaa !7
+  %116 = load i32, i32* %arrayidx1.4, align 4, !tbaa !7
+  %mul.4 = sub nsw i32 0, %116
+  store i32 %mul.4, i32 addrspace(1)* %arrayidx.4, align 4, !tbaa !7
+  %117 = load i32, i32* %arrayidx1.5, align 4, !tbaa !7
+  %mul.5 = sub nsw i32 0, %117
+  store i32 %mul.5, i32 addrspace(1)* %arrayidx.5, align 4, !tbaa !7
+  %118 = load i32, i32* %arrayidx1.6, align 4, !tbaa !7
+  %mul.6 = sub nsw i32 0, %118
+  store i32 %mul.6, i32 addrspace(1)* %arrayidx.6, align 4, !tbaa !7
+  %119 = load i32, i32* %arrayidx1.7, align 4, !tbaa !7
+  %mul.7 = sub nsw i32 0, %119
+  store i32 %mul.7, i32 addrspace(1)* %arrayidx.7, align 4, !tbaa !7
+  %120 = load i32, i32* %arrayidx1.8, align 4, !tbaa !7
+  %mul.8 = sub nsw i32 0, %120
+  store i32 %mul.8, i32 addrspace(1)* %arrayidx.8, align 4, !tbaa !7
+  %121 = load i32, i32* %arrayidx1.9, align 4, !tbaa !7
+  %mul.9 = sub nsw i32 0, %121
+  store i32 %mul.9, i32 addrspace(1)* %arrayidx.9, align 4, !tbaa !7
+  %122 = load i32, i32* %arrayidx1.10, align 4, !tbaa !7
+  %mul.10 = sub nsw i32 0, %122
+  store i32 %mul.10, i32 addrspace(1)* %arrayidx.10, align 4, !tbaa !7
+  %123 = load i32, i32* %arrayidx1.11, align 4, !tbaa !7
+  %mul.11 = sub nsw i32 0, %123
+  store i32 %mul.11, i32 addrspace(1)* %arrayidx.11, align 4, !tbaa !7
+  %124 = load i32, i32* %arrayidx1.12, align 4, !tbaa !7
+  %mul.12 = sub nsw i32 0, %124
+  store i32 %mul.12, i32 addrspace(1)* %arrayidx.12, align 4, !tbaa !7
+  %125 = load i32, i32* %arrayidx1.13, align 4, !tbaa !7
+  %mul.13 = sub nsw i32 0, %125
+  store i32 %mul.13, i32 addrspace(1)* %arrayidx.13, align 4, !tbaa !7
+  %126 = load i32, i32* %arrayidx1.14, align 4, !tbaa !7
+  %mul.14 = sub nsw i32 0, %126
+  store i32 %mul.14, i32 addrspace(1)* %arrayidx.14, align 4, !tbaa !7
+  %127 = load i32, i32* %arrayidx1.15, align 4, !tbaa !7
+  %mul.15 = sub nsw i32 0, %127
+  store i32 %mul.15, i32 addrspace(1)* %arrayidx.15, align 4, !tbaa !7
+  %128 = load i32, i32* %arrayidx1.16, align 4, !tbaa !7
+  %mul.16 = sub nsw i32 0, %128
+  store i32 %mul.16, i32 addrspace(1)* %arrayidx.16, align 4, !tbaa !7
+  %129 = load i32, i32* %arrayidx1.17, align 4, !tbaa !7
+  %mul.17 = sub nsw i32 0, %129
+  store i32 %mul.17, i32 addrspace(1)* %arrayidx.17, align 4, !tbaa !7
+  %130 = load i32, i32* %arrayidx1.18, align 4, !tbaa !7
+  %mul.18 = sub nsw i32 0, %130
+  store i32 %mul.18, i32 addrspace(1)* %arrayidx.18, align 4, !tbaa !7
+  %131 = load i32, i32* %arrayidx1.19, align 4, !tbaa !7
+  %mul.19 = sub nsw i32 0, %131
+  store i32 %mul.19, i32 addrspace(1)* %arrayidx.19, align 4, !tbaa !7
+  %132 = load i32, i32* %arrayidx1.20, align 4, !tbaa !7
+  %mul.20 = sub nsw i32 0, %132
+  store i32 %mul.20, i32 addrspace(1)* %arrayidx.20, align 4, !tbaa !7
+  %133 = load i32, i32* %arrayidx1.21, align 4, !tbaa !7
+  %mul.21 = sub nsw i32 0, %133
+  store i32 %mul.21, i32 addrspace(1)* %arrayidx.21, align 4, !tbaa !7
+  %134 = load i32, i32* %arrayidx1.22, align 4, !tbaa !7
+  %mul.22 = sub nsw i32 0, %134
+  store i32 %mul.22, i32 addrspace(1)* %arrayidx.22, align 4, !tbaa !7
+  %135 = load i32, i32* %arrayidx1.23, align 4, !tbaa !7
+  %mul.23 = sub nsw i32 0, %135
+  store i32 %mul.23, i32 addrspace(1)* %arrayidx.23, align 4, !tbaa !7
+  %136 = load i32, i32* %arrayidx1.24, align 4, !tbaa !7
+  %mul.24 = sub nsw i32 0, %136
+  store i32 %mul.24, i32 addrspace(1)* %arrayidx.24, align 4, !tbaa !7
+  %137 = load i32, i32* %arrayidx1.25, align 4, !tbaa !7
+  %mul.25 = sub nsw i32 0, %137
+  store i32 %mul.25, i32 addrspace(1)* %arrayidx.25, align 4, !tbaa !7
+  %138 = load i32, i32* %arrayidx1.26, align 4, !tbaa !7
+  %mul.26 = sub nsw i32 0, %138
+  store i32 %mul.26, i32 addrspace(1)* %arrayidx.26, align 4, !tbaa !7
+  %139 = load i32, i32* %arrayidx1.27, align 4, !tbaa !7
+  %mul.27 = sub nsw i32 0, %139
+  store i32 %mul.27, i32 addrspace(1)* %arrayidx.27, align 4, !tbaa !7
+  %140 = load i32, i32* %arrayidx1.28, align 4, !tbaa !7
+  %mul.28 = sub nsw i32 0, %140
+  store i32 %mul.28, i32 addrspace(1)* %arrayidx.28, align 4, !tbaa !7
+  %141 = load i32, i32* %arrayidx1.29, align 4, !tbaa !7
+  %mul.29 = sub nsw i32 0, %141
+  store i32 %mul.29, i32 addrspace(1)* %arrayidx.29, align 4, !tbaa !7
+  %142 = load i32, i32* %arrayidx1.30, align 4, !tbaa !7
+  %mul.30 = sub nsw i32 0, %142
+  store i32 %mul.30, i32 addrspace(1)* %arrayidx.30, align 4, !tbaa !7
+  %143 = load i32, i32* %arrayidx1.31, align 4, !tbaa !7
+  %mul.31 = sub nsw i32 0, %143
+  store i32 %mul.31, i32 addrspace(1)* %arrayidx.31, align 4, !tbaa !7
+  %144 = load i32, i32* %arrayidx1.32, align 4, !tbaa !7
+  %mul.32 = sub nsw i32 0, %144
+  store i32 %mul.32, i32 addrspace(1)* %arrayidx.32, align 4, !tbaa !7
+  %145 = load i32, i32* %arrayidx1.33, align 4, !tbaa !7
+  %mul.33 = sub nsw i32 0, %145
+  store i32 %mul.33, i32 addrspace(1)* %arrayidx.33, align 4, !tbaa !7
+  %146 = load i32, i32* %arrayidx1.34, align 4, !tbaa !7
+  %mul.34 = sub nsw i32 0, %146
+  store i32 %mul.34, i32 addrspace(1)* %arrayidx.34, align 4, !tbaa !7
+  %147 = load i32, i32* %arrayidx1.35, align 4, !tbaa !7
+  %mul.35 = sub nsw i32 0, %147
+  store i32 %mul.35, i32 addrspace(1)* %arrayidx.35, align 4, !tbaa !7
+  %148 = load i32, i32* %arrayidx1.36, align 4, !tbaa !7
+  %mul.36 = sub nsw i32 0, %148
+  store i32 %mul.36, i32 addrspace(1)* %arrayidx.36, align 4, !tbaa !7
+  %149 = load i32, i32* %arrayidx1.37, align 4, !tbaa !7
+  %mul.37 = sub nsw i32 0, %149
+  store i32 %mul.37, i32 addrspace(1)* %arrayidx.37, align 4, !tbaa !7
+  %150 = load i32, i32* %arrayidx1.38, align 4, !tbaa !7
+  %mul.38 = sub nsw i32 0, %150
+  store i32 %mul.38, i32 addrspace(1)* %arrayidx.38, align 4, !tbaa !7
+  %151 = load i32, i32* %arrayidx1.39, align 4, !tbaa !7
+  %mul.39 = sub nsw i32 0, %151
+  store i32 %mul.39, i32 addrspace(1)* %arrayidx.39, align 4, !tbaa !7
+  %152 = load i32, i32* %arrayidx1.40, align 4, !tbaa !7
+  %mul.40 = sub nsw i32 0, %152
+  store i32 %mul.40, i32 addrspace(1)* %arrayidx.40, align 4, !tbaa !7
+  %153 = load i32, i32* %arrayidx1.41, align 4, !tbaa !7
+  %mul.41 = sub nsw i32 0, %153
+  store i32 %mul.41, i32 addrspace(1)* %arrayidx.41, align 4, !tbaa !7
+  %154 = load i32, i32* %arrayidx1.42, align 4, !tbaa !7
+  %mul.42 = sub nsw i32 0, %154
+  store i32 %mul.42, i32 addrspace(1)* %arrayidx.42, align 4, !tbaa !7
+  %155 = load i32, i32* %arrayidx1.43, align 4, !tbaa !7
+  %mul.43 = sub nsw i32 0, %155
+  store i32 %mul.43, i32 addrspace(1)* %arrayidx.43, align 4, !tbaa !7
+  %156 = load i32, i32* %arrayidx1.44, align 4, !tbaa !7
+  %mul.44 = sub nsw i32 0, %156
+  store i32 %mul.44, i32 addrspace(1)* %arrayidx.44, align 4, !tbaa !7
+  %157 = load i32, i32* %arrayidx1.45, align 4, !tbaa !7
+  %mul.45 = sub nsw i32 0, %157
+  store i32 %mul.45, i32 addrspace(1)* %arrayidx.45, align 4, !tbaa !7
+  %158 = load i32, i32* %arrayidx1.46, align 4, !tbaa !7
+  %mul.46 = sub nsw i32 0, %158
+  store i32 %mul.46, i32 addrspace(1)* %arrayidx.46, align 4, !tbaa !7
+  %159 = load i32, i32* %arrayidx1.47, align 4, !tbaa !7
+  %mul.47 = sub nsw i32 0, %159
+  store i32 %mul.47, i32 addrspace(1)* %arrayidx.47, align 4, !tbaa !7
+  %160 = load i32, i32* %arrayidx1.48, align 4, !tbaa !7
+  %mul.48 = sub nsw i32 0, %160
+  store i32 %mul.48, i32 addrspace(1)* %arrayidx.48, align 4, !tbaa !7
+  %161 = load i32, i32* %arrayidx1.49, align 4, !tbaa !7
+  %mul.49 = sub nsw i32 0, %161
+  store i32 %mul.49, i32 addrspace(1)* %arrayidx.49, align 4, !tbaa !7
+  %162 = load i32, i32* %arrayidx1.50, align 4, !tbaa !7
+  %mul.50 = sub nsw i32 0, %162
+  store i32 %mul.50, i32 addrspace(1)* %arrayidx.50, align 4, !tbaa !7
+  %163 = load i32, i32* %arrayidx1.51, align 4, !tbaa !7
+  %mul.51 = sub nsw i32 0, %163
+  store i32 %mul.51, i32 addrspace(1)* %arrayidx.51, align 4, !tbaa !7
+  %164 = load i32, i32* %arrayidx1.52, align 4, !tbaa !7
+  %mul.52 = sub nsw i32 0, %164
+  store i32 %mul.52, i32 addrspace(1)* %arrayidx.52, align 4, !tbaa !7
+  %165 = load i32, i32* %arrayidx1.53, align 4, !tbaa !7
+  %mul.53 = sub nsw i32 0, %165
+  store i32 %mul.53, i32 addrspace(1)* %arrayidx.53, align 4, !tbaa !7
+  %166 = load i32, i32* %arrayidx1.54, align 4, !tbaa !7
+  %mul.54 = sub nsw i32 0, %166
+  store i32 %mul.54, i32 addrspace(1)* %arrayidx.54, align 4, !tbaa !7
+  %167 = load i32, i32* %arrayidx1.55, align 4, !tbaa !7
+  %mul.55 = sub nsw i32 0, %167
+  store i32 %mul.55, i32 addrspace(1)* %arrayidx.55, align 4, !tbaa !7
+  %168 = load i32, i32* %arrayidx1.56, align 4, !tbaa !7
+  %mul.56 = sub nsw i32 0, %168
+  store i32 %mul.56, i32 addrspace(1)* %arrayidx.56, align 4, !tbaa !7
+  %169 = load i32, i32* %arrayidx1.57, align 4, !tbaa !7
+  %mul.57 = sub nsw i32 0, %169
+  store i32 %mul.57, i32 addrspace(1)* %arrayidx.57, align 4, !tbaa !7
+  %170 = load i32, i32* %arrayidx1.58, align 4, !tbaa !7
+  %mul.58 = sub nsw i32 0, %170
+  store i32 %mul.58, i32 addrspace(1)* %arrayidx.58, align 4, !tbaa !7
+  %171 = load i32, i32* %arrayidx1.59, align 4, !tbaa !7
+  %mul.59 = sub nsw i32 0, %171
+  store i32 %mul.59, i32 addrspace(1)* %arrayidx.59, align 4, !tbaa !7
+  %172 = load i32, i32* %arrayidx1.60, align 4, !tbaa !7
+  %mul.60 = sub nsw i32 0, %172
+  store i32 %mul.60, i32 addrspace(1)* %arrayidx.60, align 4, !tbaa !7
+  %173 = load i32, i32* %arrayidx1.61, align 4, !tbaa !7
+  %mul.61 = sub nsw i32 0, %173
+  store i32 %mul.61, i32 addrspace(1)* %arrayidx.61, align 4, !tbaa !7
+  %174 = load i32, i32* %arrayidx1.62, align 4, !tbaa !7
+  %mul.62 = sub nsw i32 0, %174
+  store i32 %mul.62, i32 addrspace(1)* %arrayidx.62, align 4, !tbaa !7
+  %175 = load i32, i32* %arrayidx1.63, align 4, !tbaa !7
+  %mul.63 = sub nsw i32 0, %175
+  store i32 %mul.63, i32 addrspace(1)* %arrayidx.63, align 4, !tbaa !7
+  %176 = load i32, i32* %arrayidx1.64, align 4, !tbaa !7
+  %mul.64 = sub nsw i32 0, %176
+  store i32 %mul.64, i32 addrspace(1)* %arrayidx.64, align 4, !tbaa !7
+  %177 = load i32, i32* %arrayidx1.65, align 4, !tbaa !7
+  %mul.65 = sub nsw i32 0, %177
+  store i32 %mul.65, i32 addrspace(1)* %arrayidx.65, align 4, !tbaa !7
+  %178 = load i32, i32* %arrayidx1.66, align 4, !tbaa !7
+  %mul.66 = sub nsw i32 0, %178
+  store i32 %mul.66, i32 addrspace(1)* %arrayidx.66, align 4, !tbaa !7
+  %179 = load i32, i32* %arrayidx1.67, align 4, !tbaa !7
+  %mul.67 = sub nsw i32 0, %179
+  store i32 %mul.67, i32 addrspace(1)* %arrayidx.67, align 4, !tbaa !7
+  %180 = load i32, i32* %arrayidx1.68, align 4, !tbaa !7
+  %mul.68 = sub nsw i32 0, %180
+  store i32 %mul.68, i32 addrspace(1)* %arrayidx.68, align 4, !tbaa !7
+  %181 = load i32, i32* %arrayidx1.69, align 4, !tbaa !7
+  %mul.69 = sub nsw i32 0, %181
+  store i32 %mul.69, i32 addrspace(1)* %arrayidx.69, align 4, !tbaa !7
+  %182 = load i32, i32* %arrayidx1.70, align 4, !tbaa !7
+  %mul.70 = sub nsw i32 0, %182
+  store i32 %mul.70, i32 addrspace(1)* %arrayidx.70, align 4, !tbaa !7
+  %183 = load i32, i32* %arrayidx1.71, align 4, !tbaa !7
+  %mul.71 = sub nsw i32 0, %183
+  store i32 %mul.71, i32 addrspace(1)* %arrayidx.71, align 4, !tbaa !7
+  %184 = load i32, i32* %arrayidx1.72, align 4, !tbaa !7
+  %mul.72 = sub nsw i32 0, %184
+  store i32 %mul.72, i32 addrspace(1)* %arrayidx.72, align 4, !tbaa !7
+  %185 = load i32, i32* %arrayidx1.73, align 4, !tbaa !7
+  %mul.73 = sub nsw i32 0, %185
+  store i32 %mul.73, i32 addrspace(1)* %arrayidx.73, align 4, !tbaa !7
+  %186 = load i32, i32* %arrayidx1.74, align 4, !tbaa !7
+  %mul.74 = sub nsw i32 0, %186
+  store i32 %mul.74, i32 addrspace(1)* %arrayidx.74, align 4, !tbaa !7
+  %187 = load i32, i32* %arrayidx1.75, align 4, !tbaa !7
+  %mul.75 = sub nsw i32 0, %187
+  store i32 %mul.75, i32 addrspace(1)* %arrayidx.75, align 4, !tbaa !7
+  %188 = load i32, i32* %arrayidx1.76, align 4, !tbaa !7
+  %mul.76 = sub nsw i32 0, %188
+  store i32 %mul.76, i32 addrspace(1)* %arrayidx.76, align 4, !tbaa !7
+  %189 = load i32, i32* %arrayidx1.77, align 4, !tbaa !7
+  %mul.77 = sub nsw i32 0, %189
+  store i32 %mul.77, i32 addrspace(1)* %arrayidx.77, align 4, !tbaa !7
+  %190 = load i32, i32* %arrayidx1.78, align 4, !tbaa !7
+  %mul.78 = sub nsw i32 0, %190
+  store i32 %mul.78, i32 addrspace(1)* %arrayidx.78, align 4, !tbaa !7
+  %191 = load i32, i32* %arrayidx1.79, align 4, !tbaa !7
+  %mul.79 = sub nsw i32 0, %191
+  store i32 %mul.79, i32 addrspace(1)* %arrayidx.79, align 4, !tbaa !7
+  %192 = load i32, i32* %arrayidx1.80, align 4, !tbaa !7
+  %mul.80 = sub nsw i32 0, %192
+  store i32 %mul.80, i32 addrspace(1)* %arrayidx.80, align 4, !tbaa !7
+  %193 = load i32, i32* %arrayidx1.81, align 4, !tbaa !7
+  %mul.81 = sub nsw i32 0, %193
+  store i32 %mul.81, i32 addrspace(1)* %arrayidx.81, align 4, !tbaa !7
+  %194 = load i32, i32* %arrayidx1.82, align 4, !tbaa !7
+  %mul.82 = sub nsw i32 0, %194
+  store i32 %mul.82, i32 addrspace(1)* %arrayidx.82, align 4, !tbaa !7
+  %195 = load i32, i32* %arrayidx1.83, align 4, !tbaa !7
+  %mul.83 = sub nsw i32 0, %195
+  store i32 %mul.83, i32 addrspace(1)* %arrayidx.83, align 4, !tbaa !7
+  %196 = load i32, i32* %arrayidx1.84, align 4, !tbaa !7
+  %mul.84 = sub nsw i32 0, %196
+  store i32 %mul.84, i32 addrspace(1)* %arrayidx.84, align 4, !tbaa !7
+  %197 = load i32, i32* %arrayidx1.85, align 4, !tbaa !7
+  %mul.85 = sub nsw i32 0, %197
+  store i32 %mul.85, i32 addrspace(1)* %arrayidx.85, align 4, !tbaa !7
+  %198 = load i32, i32* %arrayidx1.86, align 4, !tbaa !7
+  %mul.86 = sub nsw i32 0, %198
+  store i32 %mul.86, i32 addrspace(1)* %arrayidx.86, align 4, !tbaa !7
+  %199 = load i32, i32* %arrayidx1.87, align 4, !tbaa !7
+  %mul.87 = sub nsw i32 0, %199
+  store i32 %mul.87, i32 addrspace(1)* %arrayidx.87, align 4, !tbaa !7
+  %200 = load i32, i32* %arrayidx1.88, align 4, !tbaa !7
+  %mul.88 = sub nsw i32 0, %200
+  store i32 %mul.88, i32 addrspace(1)* %arrayidx.88, align 4, !tbaa !7
+  %201 = load i32, i32* %arrayidx1.89, align 4, !tbaa !7
+  %mul.89 = sub nsw i32 0, %201
+  store i32 %mul.89, i32 addrspace(1)* %arrayidx.89, align 4, !tbaa !7
+  %202 = load i32, i32* %arrayidx1.90, align 4, !tbaa !7
+  %mul.90 = sub nsw i32 0, %202
+  store i32 %mul.90, i32 addrspace(1)* %arrayidx.90, align 4, !tbaa !7
+  %203 = load i32, i32* %arrayidx1.91, align 4, !tbaa !7
+  %mul.91 = sub nsw i32 0, %203
+  store i32 %mul.91, i32 addrspace(1)* %arrayidx.91, align 4, !tbaa !7
+  %204 = load i32, i32* %arrayidx1.92, align 4, !tbaa !7
+  %mul.92 = sub nsw i32 0, %204
+  store i32 %mul.92, i32 addrspace(1)* %arrayidx.92, align 4, !tbaa !7
+  %205 = load i32, i32* %arrayidx1.93, align 4, !tbaa !7
+  %mul.93 = sub nsw i32 0, %205
+  store i32 %mul.93, i32 addrspace(1)* %arrayidx.93, align 4, !tbaa !7
+  %206 = load i32, i32* %arrayidx1.94, align 4, !tbaa !7
+  %mul.94 = sub nsw i32 0, %206
+  store i32 %mul.94, i32 addrspace(1)* %arrayidx.94, align 4, !tbaa !7
+  %207 = load i32, i32* %arrayidx1.95, align 4, !tbaa !7
+  %mul.95 = sub nsw i32 0, %207
+  store i32 %mul.95, i32 addrspace(1)* %arrayidx.95, align 4, !tbaa !7
+  %208 = load i32, i32* %arrayidx1.96, align 4, !tbaa !7
+  %mul.96 = sub nsw i32 0, %208
+  store i32 %mul.96, i32 addrspace(1)* %arrayidx.96, align 4, !tbaa !7
+  %209 = load i32, i32* %arrayidx1.97, align 4, !tbaa !7
+  %mul.97 = sub nsw i32 0, %209
+  store i32 %mul.97, i32 addrspace(1)* %arrayidx.97, align 4, !tbaa !7
+  %210 = load i32, i32* %arrayidx1.98, align 4, !tbaa !7
+  %mul.98 = sub nsw i32 0, %210
+  store i32 %mul.98, i32 addrspace(1)* %arrayidx.98, align 4, !tbaa !7
+  %211 = load i32, i32* %arrayidx1.99, align 4, !tbaa !7
+  %mul.99 = sub nsw i32 0, %211
+  store i32 %mul.99, i32 addrspace(1)* %arrayidx.99, align 4, !tbaa !7
+  %212 = load i32, i32* %arrayidx1.100, align 4, !tbaa !7
+  %mul.100 = sub nsw i32 0, %212
+  store i32 %mul.100, i32 addrspace(1)* %arrayidx.100, align 4, !tbaa !7
+  %213 = load i32, i32* %arrayidx1.101, align 4, !tbaa !7
+  %mul.101 = sub nsw i32 0, %213
+  store i32 %mul.101, i32 addrspace(1)* %arrayidx.101, align 4, !tbaa !7
+  %214 = load i32, i32* %arrayidx1.102, align 4, !tbaa !7
+  %mul.102 = sub nsw i32 0, %214
+  store i32 %mul.102, i32 addrspace(1)* %arrayidx.102, align 4, !tbaa !7
+  %215 = load i32, i32* %arrayidx1.103, align 4, !tbaa !7
+  %mul.103 = sub nsw i32 0, %215
+  store i32 %mul.103, i32 addrspace(1)* %arrayidx.103, align 4, !tbaa !7
+  %216 = load i32, i32* %arrayidx1.104, align 4, !tbaa !7
+  %mul.104 = sub nsw i32 0, %216
+  store i32 %mul.104, i32 addrspace(1)* %arrayidx.104, align 4, !tbaa !7
+  %217 = load i32, i32* %arrayidx1.105, align 4, !tbaa !7
+  %mul.105 = sub nsw i32 0, %217
+  store i32 %mul.105, i32 addrspace(1)* %arrayidx.105, align 4, !tbaa !7
+  %218 = load i32, i32* %arrayidx1.106, align 4, !tbaa !7
+  %mul.106 = sub nsw i32 0, %218
+  store i32 %mul.106, i32 addrspace(1)* %arrayidx.106, align 4, !tbaa !7
+  %219 = load i32, i32* %arrayidx1.107, align 4, !tbaa !7
+  %mul.107 = sub nsw i32 0, %219
+  store i32 %mul.107, i32 addrspace(1)* %arrayidx.107, align 4, !tbaa !7
+  %220 = load i32, i32* %arrayidx1.108, align 4, !tbaa !7
+  %mul.108 = sub nsw i32 0, %220
+  store i32 %mul.108, i32 addrspace(1)* %arrayidx.108, align 4, !tbaa !7
+  %221 = load i32, i32* %arrayidx1.109, align 4, !tbaa !7
+  %mul.109 = sub nsw i32 0, %221
+  store i32 %mul.109, i32 addrspace(1)* %arrayidx.109, align 4, !tbaa !7
+  %222 = load i32, i32* %arrayidx1.110, align 4, !tbaa !7
+  %mul.110 = sub nsw i32 0, %222
+  store i32 %mul.110, i32 addrspace(1)* %arrayidx.110, align 4, !tbaa !7
+  call void @llvm.lifetime.end(i64 444, i8* %0) #2
+  ret void
+}
+
+; Function Attrs: argmemonly nounwind
+declare void @llvm.lifetime.start(i64, i8* nocapture) #1
+
+; Function Attrs: argmemonly nounwind
+declare void @llvm.lifetime.end(i64, i8* nocapture) #1
+
+attributes #0 = { nounwind "amdgpu-num-vgpr"="16" "amdgpu-num-active-waves-per-eu"="1,10" "disable-tail-calls"="false" "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "target-cpu"="fiji" "target-features"="+16-bit-insts,+s-memrealtime" "unsafe-fp-math"="false" "use-soft-float"="false" }
+attributes #1 = { argmemonly nounwind }
+attributes #2 = { nounwind }
+
+!opencl.kernels = !{!0}
+!llvm.ident = !{!6}
+
+!0 = !{void (i32 addrspace(1)*)* @attr_amdgpu_num_vgpr_16, !1, !2, !3, !4, !5}
+!1 = !{!"kernel_arg_addr_space", i32 1}
+!2 = !{!"kernel_arg_access_qual", !"none"}
+!3 = !{!"kernel_arg_type", !"int*"}
+!4 = !{!"kernel_arg_base_type", !"int*"}
+!5 = !{!"kernel_arg_type_qual", !""}
+!6 = !{!"clang version 3.9.0 (trunk 273003)"}
+!7 = !{!8, !8, i64 0}
+!8 = !{!"int", !9, i64 0}
+!9 = !{!"omnipotent char", !10, i64 0}
+!10 = !{!"Simple C/C++ TBAA"}

test/CodeGen/AMDGPU/indirect-private-64.ll

define void @private_access_v2i64_alloca(<2 x i64> addrspace(1)* noalias %out, <2 x i64> addrspace(1)* noalias %in, i32 %b) #1 {
   store <2 x i64> %val, <2 x i64>* %ptr, align 16
   call void @llvm.amdgcn.s.barrier()
   %result = load <2 x i64>, <2 x i64>* %ptr, align 16
   store <2 x i64> %result, <2 x i64> addrspace(1)* %out, align 16
   ret void
 }
 
 attributes #0 = { convergent nounwind }
-attributes #1 = { nounwind "amdgpu-max-waves-per-eu"="2" "amdgpu-max-work-group-size"="64" }
+attributes #1 = { nounwind "amdgpu-num-active-waves-per-eu"="2,2" "amdgpu-flat-work-group-size"="64,64" }

test/CodeGen/AMDGPU/large-work-group-promote-alloca.ll

entry:
   store i32 %2, i32 addrspace(1)* %out, align 4
   %arrayidx12 = getelementptr inbounds [5 x i32], [5 x i32]* %stack, i32 0, i32 1
   %3 = load i32, i32* %arrayidx12
   %arrayidx13 = getelementptr inbounds i32, i32 addrspace(1)* %out, i32 1
   store i32 %3, i32 addrspace(1)* %arrayidx13
   ret void
 }
 
-attributes #0 = { nounwind "amdgpu-max-work-group-size"="63" }
-attributes #1 = { nounwind "amdgpu-max-waves-per-eu"="3" "amdgpu-max-work-group-size"="256" }
-attributes #2 = { nounwind "amdgpu-max-waves-per-eu"="1" "amdgpu-max-work-group-size"="1600" }
-attributes #3 = { nounwind "amdgpu-max-waves-per-eu"="0" }
-attributes #4 = { nounwind "amdgpu-max-waves-per-eu"="-1" }
+attributes #0 = { nounwind "amdgpu-flat-work-group-size"="63,63" }
+attributes #1 = { nounwind "amdgpu-num-active-waves-per-eu"="1,3" "amdgpu-flat-work-group-size"="256,256" }
+attributes #2 = { nounwind "amdgpu-num-active-waves-per-eu"="1,1" "amdgpu-flat-work-group-size"="1600,1600" }
+attributes #3 = { nounwind "amdgpu-num-active-waves-per-eu"="1,10" }
+attributes #4 = { nounwind "amdgpu-num-active-waves-per-eu"="1,10" }

test/CodeGen/AMDGPU/large-work-group-registers.ll

 ; RUN: llc -march=amdgcn -mcpu=tonga -post-RA-scheduler=0 < %s | FileCheck %s
 
-; CHECK: NumVgprs: 64
+; CHECK: NumVgprs: 32
 define void @main([9 x <16 x i8>] addrspace(2)* byval, [17 x <16 x i8>] addrspace(2)* byval, [17 x <8 x i32>] addrspace(2)* byval, [16 x <8 x i32>] addrspace(2)* byval, [16 x <4 x i32>] addrspace(2)* byval, <3 x i32> inreg, <3 x i32> inreg, <3 x i32>) #0 {
 main_body:
   %8 = getelementptr [16 x <4 x i32>], [16 x <4 x i32>] addrspace(2)* %4, i64 0, i64 8
   %9 = load <4 x i32>, <4 x i32> addrspace(2)* %8, align 16, !tbaa !0
   %10 = extractelement <3 x i32> %7, i32 0
   %11 = extractelement <3 x i32> %7, i32 1
   %12 = mul i32 %10, %11
   %bc = bitcast <3 x i32> %7 to <3 x float>
main_body:
   call void @llvm.amdgcn.buffer.store.format.v4f32(<4 x float> %27, <4 x i32> %9, i32 0, i32 %19, i1 false, i1 false)
   ret void
 }
 
 declare void @llvm.amdgcn.s.barrier() #1
 
 declare void @llvm.amdgcn.buffer.store.format.v4f32(<4 x float>, <4 x i32>, i32, i32, i1, i1) #2
 
-attributes #0 = { "amdgpu-max-work-group-size"="1024" }
+attributes #0 = { "amdgpu-flat-work-group-size"="1024,1024" }
 attributes #1 = { convergent nounwind }
 attributes #2 = { nounwind }
 
 !0 = !{!1, !1, i64 0, i32 1}
 !1 = !{!"const", null}

test/CodeGen/AMDGPU/private-memory-r600.ll

define void @ptrtoint(i32 addrspace(1)* %out, i32 %a, i32 %b) #0 {
   %tmp4 = getelementptr inbounds i32, i32* %tmp3, i32 %b
   %tmp5 = load i32, i32* %tmp4
   store i32 %tmp5, i32 addrspace(1)* %out
   ret void
 }
 
 ; OPT: !0 = !{i32 0, i32 2048}
 
-attributes #0 = { nounwind "amdgpu-max-waves-per-eu"="2" }
+attributes #0 = { nounwind "amdgpu-num-active-waves-per-eu"="1,2" }

test/CodeGen/AMDGPU/promote-alloca-mem-intrinsics.ll

 define void @promote_with_objectsize(i32 addrspace(1)* %out) #0 {
   %alloca = alloca [17 x i32], align 4
   %alloca.bc = bitcast [17 x i32]* %alloca to i8*
   %size = call i32 @llvm.objectsize.i32.p0i8(i8* %alloca.bc, i1 false)
   store i32 %size, i32 addrspace(1)* %out
   ret void
 }
 
-attributes #0 = { nounwind "amdgpu-max-work-group-size"="64" "amdgpu-max-waves-per-eu"="3" }
+attributes #0 = { nounwind "amdgpu-flat-work-group-size"="64,64" "amdgpu-num-active-waves-per-eu"="3,3" }
 attributes #1 = { nounwind readnone }

test/CodeGen/AMDGPU/promote-alloca-no-opts.ll

entry:
   store i32 0, i32* %gep0
   store i32 1, i32* %gep1
   %gep2 = getelementptr inbounds [2 x [2 x i32]], [2 x [2 x i32]]* %alloca, i32 0, i32 0, i32 %index
   %load = load i32, i32* %gep2
   store i32 %load, i32 addrspace(1)* %out
   ret void
 }
 
-attributes #0 = { nounwind "amdgpu-max-work-group-size"="64" }
-attributes #1 = { nounwind optnone noinline "amdgpu-max-work-group-size"="64" }
+attributes #0 = { nounwind "amdgpu-flat-work-group-size"="64,64" }
+attributes #1 = { nounwind optnone noinline "amdgpu-flat-work-group-size"="64,64" }

test/CodeGen/AMDGPU/promote-alloca-padding-size-estimate.ll

entry:
   store volatile i32 0, i32 addrspace(3)* %gep.lds3, align 4
 
   %gep.lds4 = getelementptr inbounds [63 x <4 x i32>], [63 x <4 x i32>] addrspace(3)* @lds4, i32 0, i32 %idx
   store volatile <4 x i32> zeroinitializer, <4 x i32> addrspace(3)* %gep.lds4, align 16
 
   ret void
 }
 
-attributes #0 = { nounwind "amdgpu-max-work-group-size"="64" }
+attributes #0 = { nounwind "amdgpu-flat-work-group-size"="64,64" "amdgpu-num-active-waves-per-eu"="1,7" }

test/CodeGen/AMDGPU/promote-alloca-to-lds-icmp.ll

define void @lds_promoted_alloca_icmp_unknown_ptr(i32 addrspace(1)* %out, i32 %a, i32 %b) #0 {
   %cmp = icmp eq i32* %ptr0, %ptr1
   %zext = zext i1 %cmp to i32
   store volatile i32 %zext, i32 addrspace(1)* %out
   ret void
 }
 
 declare i32* @get_unknown_pointer() #0
 
-attributes #0 = { nounwind "amdgpu-max-waves-per-eu"="1" }
+attributes #0 = { nounwind "amdgpu-num-active-waves-per-eu"="1,1" }

test/CodeGen/AMDGPU/promote-alloca-to-lds-phi.ll

for.body:                                         ; preds = %for.body, %for.body.preheader
   %incdec.ptr = getelementptr inbounds i32, i32* %p.08, i32 1
   %inc = add nuw nsw i32 %i.09, 1
   %cmp = icmp eq i32* %incdec.ptr, %call
   br i1 %cmp, label %for.cond.cleanup.loopexit, label %for.body
 }
 
 declare i32* @get_unknown_pointer() #0
 
-attributes #0 = { nounwind "amdgpu-max-waves-per-eu"="1" }
+attributes #0 = { nounwind "amdgpu-num-active-waves-per-eu"="1,1" }

test/CodeGen/AMDGPU/promote-alloca-to-lds-select.ll

bb:
   %tmp2 = icmp eq i32 %arg1, 0
   %tmp3 = select i1 %tmp2, double* null, double* %tmp
   store double 1.000000e+00, double* %tmp3, align 8
   %tmp4 = load double, double* %tmp, align 8
   store double %tmp4, double addrspace(1)* %arg
   ret void
 }
 
-attributes #0 = { norecurse nounwind "amdgpu-max-waves-per-eu"="1" }
-attributes #1 = { norecurse nounwind }
+attributes #0 = { norecurse nounwind "amdgpu-num-active-waves-per-eu"="1,1" }
+attributes #1 = { norecurse nounwind }
 

test/CodeGen/AMDGPU/vgpr-spill-emergency-stack-slot.ll

 ; RUN: llc -march=amdgcn -mcpu=tahiti -mattr=+vgpr-spilling -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=SI %s
 ; RUN: llc -march=amdgcn -mcpu=fiji -mattr=+vgpr-spilling -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=VI %s
 
 ; This ends up using all 255 registers and requires register
 ; scavenging which will fail to find an unsued register.
 
 ; Check the ScratchSize to avoid regressions from spilling
 ; intermediate register class copies.
 
 ; FIXME: The same register is initialized to 0 for every spill.
 ; FIXME: The unused arguments are removed
 
 ; GCN-LABEL: {{^}}main:
 
-; GCN-DAG: s_mov_b32 s13, s12
-; GCN-DAG: s_mov_b32 s16, SCRATCH_RSRC_DWORD0
-; GCN-DAG: s_mov_b32 s17, SCRATCH_RSRC_DWORD1
-; GCN-DAG: s_mov_b32 s18, -1
-; SI-DAG: s_mov_b32 s19, 0xe8f000
-; VI-DAG: s_mov_b32 s19, 0xe80000
-
-; s13 is offset system SGPR
-; GCN: buffer_store_dword {{v[0-9]+}}, off, s[16:19], s13 offset:{{[0-9]+}} ; 16-byte Folded Spill
-; GCN: buffer_load_dword v{{[0-9]+}}, off, s[16:19], s13 offset:{{[0-9]+}} ; 16-byte Folded Reload
+; GCN-DAG: s_mov_b32 [[SCRATCH_OFFSET_SGPR:s[0-9]+]], s12
+; GCN-DAG: s_mov_b32 s12, SCRATCH_RSRC_DWORD0
+; GCN-DAG: s_mov_b32 s13, SCRATCH_RSRC_DWORD1
+; GCN-DAG: s_mov_b32 s14, -1
+; SI-DAG: s_mov_b32 s15, 0xe8f000
+; VI-DAG: s_mov_b32 s15, 0xe80000
+
+; [[SCRATCH_OFFSET_SGPR]] is offset system SGPR
+; GCN: buffer_store_dword {{v[0-9]+}}, off, s[12:15], [[SCRATCH_OFFSET_SGPR]] offset:{{[0-9]+}} ; 16-byte Folded Spill
+; GCN: buffer_load_dword v{{[0-9]+}}, off, s[12:15], [[SCRATCH_OFFSET_SGPR]] offset:{{[0-9]+}} ; 16-byte Folded Reload
 
 ; GCN: NumVgprs: 256
 ; GCN: ScratchSize: 1024
 
 define amdgpu_vs void @main([9 x <16 x i8>] addrspace(2)* byval %arg, [17 x <16 x i8>] addrspace(2)* byval %arg1, [17 x <4 x i32>] addrspace(2)* byval %arg2, [34 x <8 x i32>] addrspace(2)* byval %arg3, [16 x <16 x i8>] addrspace(2)* byval %arg4, i32 inreg %arg5, i32 inreg %arg6, i32 %arg7, i32 %arg8, i32 %arg9, i32 %arg10) {
 bb:
   %tmp = getelementptr [17 x <16 x i8>], [17 x <16 x i8>] addrspace(2)* %arg1, i64 0, i64 0
   %tmp11 = load <16 x i8>, <16 x i8> addrspace(2)* %tmp, align 16, !tbaa !0