llvm.org GIT mirror llvm / master lib / Target / AMDGPU / AMDGPUPrintfRuntimeBinding.cpp
master

Tree @master (Download .tar.gz)

AMDGPUPrintfRuntimeBinding.cpp @masterraw · history · blame

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
//=== AMDGPUPrintfRuntimeBinding.cpp - OpenCL printf implementation -------===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
// \file
//
// The pass bind printfs to a kernel arg pointer that will be bound to a buffer
// later by the runtime.
//
// This pass traverses the functions in the module and converts
// each call to printf to a sequence of operations that
// store the following into the printf buffer:
// - format string (passed as a module's metadata unique ID)
// - bitwise copies of printf arguments
// The backend passes will need to store metadata in the kernel
//===----------------------------------------------------------------------===//

#include "AMDGPU.h"
#include "llvm/ADT/SmallString.h"
#include "llvm/ADT/StringExtras.h"
#include "llvm/ADT/Triple.h"
#include "llvm/Analysis/InstructionSimplify.h"
#include "llvm/Analysis/TargetLibraryInfo.h"
#include "llvm/CodeGen/Passes.h"
#include "llvm/IR/Constants.h"
#include "llvm/IR/DataLayout.h"
#include "llvm/IR/Dominators.h"
#include "llvm/IR/GlobalVariable.h"
#include "llvm/IR/IRBuilder.h"
#include "llvm/IR/Instructions.h"
#include "llvm/IR/Module.h"
#include "llvm/IR/Type.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/raw_ostream.h"
#include "llvm/Transforms/Utils/BasicBlockUtils.h"
using namespace llvm;

#define DEBUG_TYPE "printfToRuntime"
#define DWORD_ALIGN 4

namespace {
class LLVM_LIBRARY_VISIBILITY AMDGPUPrintfRuntimeBinding final
    : public ModulePass {

public:
  static char ID;

  explicit AMDGPUPrintfRuntimeBinding();

private:
  bool runOnModule(Module &M) override;
  void getConversionSpecifiers(SmallVectorImpl<char> &OpConvSpecifiers,
                               StringRef fmt, size_t num_ops) const;

  bool shouldPrintAsStr(char Specifier, Type *OpType) const;
  bool
  lowerPrintfForGpu(Module &M,
                    function_ref<const TargetLibraryInfo &(Function &)> GetTLI);

  void getAnalysisUsage(AnalysisUsage &AU) const override {
    AU.addRequired<TargetLibraryInfoWrapperPass>();
    AU.addRequired<DominatorTreeWrapperPass>();
  }

  Value *simplify(Instruction *I, const TargetLibraryInfo *TLI) {
    return SimplifyInstruction(I, {*TD, TLI, DT});
  }

  const DataLayout *TD;
  const DominatorTree *DT;
  SmallVector<CallInst *, 32> Printfs;
};
} // namespace

char AMDGPUPrintfRuntimeBinding::ID = 0;

INITIALIZE_PASS_BEGIN(AMDGPUPrintfRuntimeBinding,
                      "amdgpu-printf-runtime-binding", "AMDGPU Printf lowering",
                      false, false)
INITIALIZE_PASS_DEPENDENCY(TargetLibraryInfoWrapperPass)
INITIALIZE_PASS_DEPENDENCY(DominatorTreeWrapperPass)
INITIALIZE_PASS_END(AMDGPUPrintfRuntimeBinding, "amdgpu-printf-runtime-binding",
                    "AMDGPU Printf lowering", false, false)

char &llvm::AMDGPUPrintfRuntimeBindingID = AMDGPUPrintfRuntimeBinding::ID;

namespace llvm {
ModulePass *createAMDGPUPrintfRuntimeBinding() {
  return new AMDGPUPrintfRuntimeBinding();
}
} // namespace llvm

AMDGPUPrintfRuntimeBinding::AMDGPUPrintfRuntimeBinding()
    : ModulePass(ID), TD(nullptr), DT(nullptr) {
  initializeAMDGPUPrintfRuntimeBindingPass(*PassRegistry::getPassRegistry());
}

void AMDGPUPrintfRuntimeBinding::getConversionSpecifiers(
    SmallVectorImpl<char> &OpConvSpecifiers, StringRef Fmt,
    size_t NumOps) const {
  // not all format characters are collected.
  // At this time the format characters of interest
  // are %p and %s, which use to know if we
  // are either storing a literal string or a
  // pointer to the printf buffer.
  static const char ConvSpecifiers[] = "cdieEfgGaosuxXp";
  size_t CurFmtSpecifierIdx = 0;
  size_t PrevFmtSpecifierIdx = 0;

  while ((CurFmtSpecifierIdx = Fmt.find_first_of(
              ConvSpecifiers, CurFmtSpecifierIdx)) != StringRef::npos) {
    bool ArgDump = false;
    StringRef CurFmt = Fmt.substr(PrevFmtSpecifierIdx,
                                  CurFmtSpecifierIdx - PrevFmtSpecifierIdx);
    size_t pTag = CurFmt.find_last_of("%");
    if (pTag != StringRef::npos) {
      ArgDump = true;
      while (pTag && CurFmt[--pTag] == '%') {
        ArgDump = !ArgDump;
      }
    }

    if (ArgDump)
      OpConvSpecifiers.push_back(Fmt[CurFmtSpecifierIdx]);

    PrevFmtSpecifierIdx = ++CurFmtSpecifierIdx;
  }
}

bool AMDGPUPrintfRuntimeBinding::shouldPrintAsStr(char Specifier,
                                                  Type *OpType) const {
  if (Specifier != 's')
    return false;
  const PointerType *PT = dyn_cast<PointerType>(OpType);
  if (!PT || PT->getAddressSpace() != AMDGPUAS::CONSTANT_ADDRESS)
    return false;
  Type *ElemType = PT->getContainedType(0);
  if (ElemType->getTypeID() != Type::IntegerTyID)
    return false;
  IntegerType *ElemIType = cast<IntegerType>(ElemType);
  return ElemIType->getBitWidth() == 8;
}

bool AMDGPUPrintfRuntimeBinding::lowerPrintfForGpu(
    Module &M, function_ref<const TargetLibraryInfo &(Function &)> GetTLI) {
  LLVMContext &Ctx = M.getContext();
  IRBuilder<> Builder(Ctx);
  Type *I32Ty = Type::getInt32Ty(Ctx);
  unsigned UniqID = 0;
  // NB: This is important for this string size to be divizable by 4
  const char NonLiteralStr[4] = "???";

  for (auto CI : Printfs) {
    unsigned NumOps = CI->getNumArgOperands();

    SmallString<16> OpConvSpecifiers;
    Value *Op = CI->getArgOperand(0);

    if (auto LI = dyn_cast<LoadInst>(Op)) {
      Op = LI->getPointerOperand();
      for (auto Use : Op->users()) {
        if (auto SI = dyn_cast<StoreInst>(Use)) {
          Op = SI->getValueOperand();
          break;
        }
      }
    }

    if (auto I = dyn_cast<Instruction>(Op)) {
      Value *Op_simplified = simplify(I, &GetTLI(*I->getFunction()));
      if (Op_simplified)
        Op = Op_simplified;
    }

    ConstantExpr *ConstExpr = dyn_cast<ConstantExpr>(Op);

    if (ConstExpr) {
      GlobalVariable *GVar = dyn_cast<GlobalVariable>(ConstExpr->getOperand(0));

      StringRef Str("unknown");
      if (GVar && GVar->hasInitializer()) {
        auto Init = GVar->getInitializer();
        if (auto CA = dyn_cast<ConstantDataArray>(Init)) {
          if (CA->isString())
            Str = CA->getAsCString();
        } else if (isa<ConstantAggregateZero>(Init)) {
          Str = "";
        }
        //
        // we need this call to ascertain
        // that we are printing a string
        // or a pointer. It takes out the
        // specifiers and fills up the first
        // arg
        getConversionSpecifiers(OpConvSpecifiers, Str, NumOps - 1);
      }
      // Add metadata for the string
      std::string AStreamHolder;
      raw_string_ostream Sizes(AStreamHolder);
      int Sum = DWORD_ALIGN;
      Sizes << CI->getNumArgOperands() - 1;
      Sizes << ':';
      for (unsigned ArgCount = 1; ArgCount < CI->getNumArgOperands() &&
                                  ArgCount <= OpConvSpecifiers.size();
           ArgCount++) {
        Value *Arg = CI->getArgOperand(ArgCount);
        Type *ArgType = Arg->getType();
        unsigned ArgSize = TD->getTypeAllocSizeInBits(ArgType);
        ArgSize = ArgSize / 8;
        //
        // ArgSize by design should be a multiple of DWORD_ALIGN,
        // expand the arguments that do not follow this rule.
        //
        if (ArgSize % DWORD_ALIGN != 0) {
          llvm::Type *ResType = llvm::Type::getInt32Ty(Ctx);
          VectorType *LLVMVecType = llvm::dyn_cast<llvm::VectorType>(ArgType);
          int NumElem = LLVMVecType ? LLVMVecType->getNumElements() : 1;
          if (LLVMVecType && NumElem > 1)
            ResType = llvm::VectorType::get(ResType, NumElem);
          Builder.SetInsertPoint(CI);
          Builder.SetCurrentDebugLocation(CI->getDebugLoc());
          if (OpConvSpecifiers[ArgCount - 1] == 'x' ||
              OpConvSpecifiers[ArgCount - 1] == 'X' ||
              OpConvSpecifiers[ArgCount - 1] == 'u' ||
              OpConvSpecifiers[ArgCount - 1] == 'o')
            Arg = Builder.CreateZExt(Arg, ResType);
          else
            Arg = Builder.CreateSExt(Arg, ResType);
          ArgType = Arg->getType();
          ArgSize = TD->getTypeAllocSizeInBits(ArgType);
          ArgSize = ArgSize / 8;
          CI->setOperand(ArgCount, Arg);
        }
        if (OpConvSpecifiers[ArgCount - 1] == 'f') {
          ConstantFP *FpCons = dyn_cast<ConstantFP>(Arg);
          if (FpCons)
            ArgSize = 4;
          else {
            FPExtInst *FpExt = dyn_cast<FPExtInst>(Arg);
            if (FpExt && FpExt->getType()->isDoubleTy() &&
                FpExt->getOperand(0)->getType()->isFloatTy())
              ArgSize = 4;
          }
        }
        if (shouldPrintAsStr(OpConvSpecifiers[ArgCount - 1], ArgType)) {
          if (ConstantExpr *ConstExpr = dyn_cast<ConstantExpr>(Arg)) {
            GlobalVariable *GV =
                dyn_cast<GlobalVariable>(ConstExpr->getOperand(0));
            if (GV && GV->hasInitializer()) {
              Constant *Init = GV->getInitializer();
              ConstantDataArray *CA = dyn_cast<ConstantDataArray>(Init);
              if (Init->isZeroValue() || CA->isString()) {
                size_t SizeStr = Init->isZeroValue()
                                     ? 1
                                     : (strlen(CA->getAsCString().data()) + 1);
                size_t Rem = SizeStr % DWORD_ALIGN;
                size_t NSizeStr = 0;
                LLVM_DEBUG(dbgs() << "Printf string original size = " << SizeStr
                                  << '\n');
                if (Rem) {
                  NSizeStr = SizeStr + (DWORD_ALIGN - Rem);
                } else {
                  NSizeStr = SizeStr;
                }
                ArgSize = NSizeStr;
              }
            } else {
              ArgSize = sizeof(NonLiteralStr);
            }
          } else {
            ArgSize = sizeof(NonLiteralStr);
          }
        }
        LLVM_DEBUG(dbgs() << "Printf ArgSize (in buffer) = " << ArgSize
                          << " for type: " << *ArgType << '\n');
        Sizes << ArgSize << ':';
        Sum += ArgSize;
      }
      LLVM_DEBUG(dbgs() << "Printf format string in source = " << Str.str()
                        << '\n');
      for (size_t I = 0; I < Str.size(); ++I) {
        // Rest of the C escape sequences (e.g. \') are handled correctly
        // by the MDParser
        switch (Str[I]) {
        case '\a':
          Sizes << "\\a";
          break;
        case '\b':
          Sizes << "\\b";
          break;
        case '\f':
          Sizes << "\\f";
          break;
        case '\n':
          Sizes << "\\n";
          break;
        case '\r':
          Sizes << "\\r";
          break;
        case '\v':
          Sizes << "\\v";
          break;
        case ':':
          // ':' cannot be scanned by Flex, as it is defined as a delimiter
          // Replace it with it's octal representation \72
          Sizes << "\\72";
          break;
        default:
          Sizes << Str[I];
          break;
        }
      }

      // Insert the printf_alloc call
      Builder.SetInsertPoint(CI);
      Builder.SetCurrentDebugLocation(CI->getDebugLoc());

      AttributeList Attr = AttributeList::get(Ctx, AttributeList::FunctionIndex,
                                              Attribute::NoUnwind);

      Type *SizetTy = Type::getInt32Ty(Ctx);

      Type *Tys_alloc[1] = {SizetTy};
      Type *I8Ptr = PointerType::get(Type::getInt8Ty(Ctx), 1);
      FunctionType *FTy_alloc = FunctionType::get(I8Ptr, Tys_alloc, false);
      FunctionCallee PrintfAllocFn =
          M.getOrInsertFunction(StringRef("__printf_alloc"), FTy_alloc, Attr);

      LLVM_DEBUG(dbgs() << "Printf metadata = " << Sizes.str() << '\n');
      std::string fmtstr = itostr(++UniqID) + ":" + Sizes.str().c_str();
      MDString *fmtStrArray = MDString::get(Ctx, fmtstr);

      // Instead of creating global variables, the
      // printf format strings are extracted
      // and passed as metadata. This avoids
      // polluting llvm's symbol tables in this module.
      // Metadata is going to be extracted
      // by the backend passes and inserted
      // into the OpenCL binary as appropriate.
      StringRef amd("llvm.printf.fmts");
      NamedMDNode *metaD = M.getOrInsertNamedMetadata(amd);
      MDNode *myMD = MDNode::get(Ctx, fmtStrArray);
      metaD->addOperand(myMD);
      Value *sumC = ConstantInt::get(SizetTy, Sum, false);
      SmallVector<Value *, 1> alloc_args;
      alloc_args.push_back(sumC);
      CallInst *pcall =
          CallInst::Create(PrintfAllocFn, alloc_args, "printf_alloc_fn", CI);

      //
      // Insert code to split basicblock with a
      // piece of hammock code.
      // basicblock splits after buffer overflow check
      //
      ConstantPointerNull *zeroIntPtr =
          ConstantPointerNull::get(PointerType::get(Type::getInt8Ty(Ctx), 1));
      ICmpInst *cmp =
          dyn_cast<ICmpInst>(Builder.CreateICmpNE(pcall, zeroIntPtr, ""));
      if (!CI->use_empty()) {
        Value *result =
            Builder.CreateSExt(Builder.CreateNot(cmp), I32Ty, "printf_res");
        CI->replaceAllUsesWith(result);
      }
      SplitBlock(CI->getParent(), cmp);
      Instruction *Brnch =
          SplitBlockAndInsertIfThen(cmp, cmp->getNextNode(), false);

      Builder.SetInsertPoint(Brnch);

      // store unique printf id in the buffer
      //
      SmallVector<Value *, 1> ZeroIdxList;
      ConstantInt *zeroInt =
          ConstantInt::get(Ctx, APInt(32, StringRef("0"), 10));
      ZeroIdxList.push_back(zeroInt);

      GetElementPtrInst *BufferIdx =
          dyn_cast<GetElementPtrInst>(GetElementPtrInst::Create(
              nullptr, pcall, ZeroIdxList, "PrintBuffID", Brnch));

      Type *idPointer = PointerType::get(I32Ty, AMDGPUAS::GLOBAL_ADDRESS);
      Value *id_gep_cast =
          new BitCastInst(BufferIdx, idPointer, "PrintBuffIdCast", Brnch);

      StoreInst *stbuff =
          new StoreInst(ConstantInt::get(I32Ty, UniqID), id_gep_cast);
      stbuff->insertBefore(Brnch); // to Remove unused variable warning

      SmallVector<Value *, 2> FourthIdxList;
      ConstantInt *fourInt =
          ConstantInt::get(Ctx, APInt(32, StringRef("4"), 10));

      FourthIdxList.push_back(fourInt); // 1st 4 bytes hold the printf_id
      // the following GEP is the buffer pointer
      BufferIdx = cast<GetElementPtrInst>(GetElementPtrInst::Create(
          nullptr, pcall, FourthIdxList, "PrintBuffGep", Brnch));

      Type *Int32Ty = Type::getInt32Ty(Ctx);
      Type *Int64Ty = Type::getInt64Ty(Ctx);
      for (unsigned ArgCount = 1; ArgCount < CI->getNumArgOperands() &&
                                  ArgCount <= OpConvSpecifiers.size();
           ArgCount++) {
        Value *Arg = CI->getArgOperand(ArgCount);
        Type *ArgType = Arg->getType();
        SmallVector<Value *, 32> WhatToStore;
        if (ArgType->isFPOrFPVectorTy() &&
            (ArgType->getTypeID() != Type::VectorTyID)) {
          Type *IType = (ArgType->isFloatTy()) ? Int32Ty : Int64Ty;
          if (OpConvSpecifiers[ArgCount - 1] == 'f') {
            ConstantFP *fpCons = dyn_cast<ConstantFP>(Arg);
            if (fpCons) {
              APFloat Val(fpCons->getValueAPF());
              bool Lost = false;
              Val.convert(APFloat::IEEEsingle(), APFloat::rmNearestTiesToEven,
                          &Lost);
              Arg = ConstantFP::get(Ctx, Val);
              IType = Int32Ty;
            } else {
              FPExtInst *FpExt = dyn_cast<FPExtInst>(Arg);
              if (FpExt && FpExt->getType()->isDoubleTy() &&
                  FpExt->getOperand(0)->getType()->isFloatTy()) {
                Arg = FpExt->getOperand(0);
                IType = Int32Ty;
              }
            }
          }
          Arg = new BitCastInst(Arg, IType, "PrintArgFP", Brnch);
          WhatToStore.push_back(Arg);
        } else if (ArgType->getTypeID() == Type::PointerTyID) {
          if (shouldPrintAsStr(OpConvSpecifiers[ArgCount - 1], ArgType)) {
            const char *S = NonLiteralStr;
            if (ConstantExpr *ConstExpr = dyn_cast<ConstantExpr>(Arg)) {
              GlobalVariable *GV =
                  dyn_cast<GlobalVariable>(ConstExpr->getOperand(0));
              if (GV && GV->hasInitializer()) {
                Constant *Init = GV->getInitializer();
                ConstantDataArray *CA = dyn_cast<ConstantDataArray>(Init);
                if (Init->isZeroValue() || CA->isString()) {
                  S = Init->isZeroValue() ? "" : CA->getAsCString().data();
                }
              }
            }
            size_t SizeStr = strlen(S) + 1;
            size_t Rem = SizeStr % DWORD_ALIGN;
            size_t NSizeStr = 0;
            if (Rem) {
              NSizeStr = SizeStr + (DWORD_ALIGN - Rem);
            } else {
              NSizeStr = SizeStr;
            }
            if (S[0]) {
              char *MyNewStr = new char[NSizeStr]();
              strcpy(MyNewStr, S);
              int NumInts = NSizeStr / 4;
              int CharC = 0;
              while (NumInts) {
                int ANum = *(int *)(MyNewStr + CharC);
                CharC += 4;
                NumInts--;
                Value *ANumV = ConstantInt::get(Int32Ty, ANum, false);
                WhatToStore.push_back(ANumV);
              }
              delete[] MyNewStr;
            } else {
              // Empty string, give a hint to RT it is no NULL
              Value *ANumV = ConstantInt::get(Int32Ty, 0xFFFFFF00, false);
              WhatToStore.push_back(ANumV);
            }
          } else {
            uint64_t Size = TD->getTypeAllocSizeInBits(ArgType);
            assert((Size == 32 || Size == 64) && "unsupported size");
            Type *DstType = (Size == 32) ? Int32Ty : Int64Ty;
            Arg = new PtrToIntInst(Arg, DstType, "PrintArgPtr", Brnch);
            WhatToStore.push_back(Arg);
          }
        } else if (ArgType->getTypeID() == Type::VectorTyID) {
          Type *IType = NULL;
          uint32_t EleCount = cast<VectorType>(ArgType)->getNumElements();
          uint32_t EleSize = ArgType->getScalarSizeInBits();
          uint32_t TotalSize = EleCount * EleSize;
          if (EleCount == 3) {
            IntegerType *Int32Ty = Type::getInt32Ty(ArgType->getContext());
            Constant *Indices[4] = {
                ConstantInt::get(Int32Ty, 0), ConstantInt::get(Int32Ty, 1),
                ConstantInt::get(Int32Ty, 2), ConstantInt::get(Int32Ty, 2)};
            Constant *Mask = ConstantVector::get(Indices);
            ShuffleVectorInst *Shuffle = new ShuffleVectorInst(Arg, Arg, Mask);
            Shuffle->insertBefore(Brnch);
            Arg = Shuffle;
            ArgType = Arg->getType();
            TotalSize += EleSize;
          }
          switch (EleSize) {
          default:
            EleCount = TotalSize / 64;
            IType = dyn_cast<Type>(Type::getInt64Ty(ArgType->getContext()));
            break;
          case 8:
            if (EleCount >= 8) {
              EleCount = TotalSize / 64;
              IType = dyn_cast<Type>(Type::getInt64Ty(ArgType->getContext()));
            } else if (EleCount >= 3) {
              EleCount = 1;
              IType = dyn_cast<Type>(Type::getInt32Ty(ArgType->getContext()));
            } else {
              EleCount = 1;
              IType = dyn_cast<Type>(Type::getInt16Ty(ArgType->getContext()));
            }
            break;
          case 16:
            if (EleCount >= 3) {
              EleCount = TotalSize / 64;
              IType = dyn_cast<Type>(Type::getInt64Ty(ArgType->getContext()));
            } else {
              EleCount = 1;
              IType = dyn_cast<Type>(Type::getInt32Ty(ArgType->getContext()));
            }
            break;
          }
          if (EleCount > 1) {
            IType = dyn_cast<Type>(VectorType::get(IType, EleCount));
          }
          Arg = new BitCastInst(Arg, IType, "PrintArgVect", Brnch);
          WhatToStore.push_back(Arg);
        } else {
          WhatToStore.push_back(Arg);
        }
        for (unsigned I = 0, E = WhatToStore.size(); I != E; ++I) {
          Value *TheBtCast = WhatToStore[I];
          unsigned ArgSize =
              TD->getTypeAllocSizeInBits(TheBtCast->getType()) / 8;
          SmallVector<Value *, 1> BuffOffset;
          BuffOffset.push_back(ConstantInt::get(I32Ty, ArgSize));

          Type *ArgPointer = PointerType::get(TheBtCast->getType(), 1);
          Value *CastedGEP =
              new BitCastInst(BufferIdx, ArgPointer, "PrintBuffPtrCast", Brnch);
          StoreInst *StBuff = new StoreInst(TheBtCast, CastedGEP, Brnch);
          LLVM_DEBUG(dbgs() << "inserting store to printf buffer:\n"
                            << *StBuff << '\n');
          (void)StBuff;
          if (I + 1 == E && ArgCount + 1 == CI->getNumArgOperands())
            break;
          BufferIdx = dyn_cast<GetElementPtrInst>(GetElementPtrInst::Create(
              nullptr, BufferIdx, BuffOffset, "PrintBuffNextPtr", Brnch));
          LLVM_DEBUG(dbgs() << "inserting gep to the printf buffer:\n"
                            << *BufferIdx << '\n');
        }
      }
    }
  }

  // erase the printf calls
  for (auto CI : Printfs)
    CI->eraseFromParent();

  Printfs.clear();
  return true;
}

bool AMDGPUPrintfRuntimeBinding::runOnModule(Module &M) {
  Triple TT(M.getTargetTriple());
  if (TT.getArch() == Triple::r600)
    return false;

  auto PrintfFunction = M.getFunction("printf");
  if (!PrintfFunction)
    return false;

  for (auto &U : PrintfFunction->uses()) {
    if (auto *CI = dyn_cast<CallInst>(U.getUser())) {
      if (CI->isCallee(&U))
        Printfs.push_back(CI);
    }
  }

  if (Printfs.empty())
    return false;

  TD = &M.getDataLayout();
  auto DTWP = getAnalysisIfAvailable<DominatorTreeWrapperPass>();
  DT = DTWP ? &DTWP->getDomTree() : nullptr;
  auto GetTLI = [this](Function &F) -> TargetLibraryInfo & {
    return this->getAnalysis<TargetLibraryInfoWrapperPass>().getTLI(F);
  };

  return lowerPrintfForGpu(M, GetTLI);
}