lib/Support/MemoryBuffer.cpp

   1 //===--- MemoryBuffer.cpp - Memory Buffer implementation ------------------===//
   2 //
   3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
   4 // See https://llvm.org/LICENSE.txt for license information.
   5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
   6 //
   7 //===----------------------------------------------------------------------===//
   8 //
   9 //  This file implements the MemoryBuffer interface.
  10 //
  11 //===----------------------------------------------------------------------===//
  12
  13 #include "llvm/Support/MemoryBuffer.h"
  14 #include "llvm/ADT/SmallString.h"
  15 #include "llvm/Config/config.h"
  16 #include "llvm/Support/Errc.h"
  17 #include "llvm/Support/Errno.h"
  18 #include "llvm/Support/FileSystem.h"
  19 #include "llvm/Support/MathExtras.h"
  20 #include "llvm/Support/Path.h"
  21 #include "llvm/Support/Process.h"
  22 #include "llvm/Support/Program.h"
  23 #include "llvm/Support/SmallVectorMemoryBuffer.h"
  24 #include <cassert>
  25 #include <cerrno>
  26 #include <cstring>
  27 #include <new>
  28 #include <sys/types.h>
  29 #include <system_error>
  30 #if !defined(_MSC_VER) && !defined(__MINGW32__)
  31 #include <unistd.h>
  32 #else
  33 #include <io.h>
  34 #endif
  35 using namespace llvm;
  36
  37 //===----------------------------------------------------------------------===//
  38 // MemoryBuffer implementation itself.
  39 //===----------------------------------------------------------------------===//
  40
  41 MemoryBuffer::~MemoryBuffer() { }
  42
  43 /// init - Initialize this MemoryBuffer as a reference to externally allocated
  44 /// memory, memory that we know is already null terminated.
  45 void MemoryBuffer::init(const char *BufStart, const char *BufEnd,
  46                         bool RequiresNullTerminator) {
  47   assert((!RequiresNullTerminator || BufEnd[0] == 0) &&
  48          "Buffer is not null terminated!");
  49   BufferStart = BufStart;
  50   BufferEnd = BufEnd;
  51 }
  52
  53 //===----------------------------------------------------------------------===//
  54 // MemoryBufferMem implementation.
  55 //===----------------------------------------------------------------------===//
  56
  57 /// CopyStringRef - Copies contents of a StringRef into a block of memory and
  58 /// null-terminates it.
  59 static void CopyStringRef(char *Memory, StringRef Data) {
  60   if (!Data.empty())
  61     memcpy(Memory, Data.data(), Data.size());
  62   Memory[Data.size()] = 0; // Null terminate string.
  63 }
  64
  65 namespace {
  66 struct NamedBufferAlloc {
  67   const Twine &Name;
  68   NamedBufferAlloc(const Twine &Name) : Name(Name) {}
  69 };
  70 }
  71
  72 void *operator new(size_t N, const NamedBufferAlloc &Alloc) {
  73   SmallString<256> NameBuf;
  74   StringRef NameRef = Alloc.Name.toStringRef(NameBuf);
  75
  76   char *Mem = static_cast<char *>(operator new(N + NameRef.size() + 1));
  77   CopyStringRef(Mem + N, NameRef);
  78   return Mem;
  79 }
  80
  81 namespace {
  82 /// MemoryBufferMem - Named MemoryBuffer pointing to a block of memory.
  83 template<typename MB>
  84 class MemoryBufferMem : public MB {
  85 public:
  86   MemoryBufferMem(StringRef InputData, bool RequiresNullTerminator) {
  87     MemoryBuffer::init(InputData.begin(), InputData.end(),
  88                        RequiresNullTerminator);
  89   }
  90
  91   /// Disable sized deallocation for MemoryBufferMem, because it has
  92   /// tail-allocated data.
  93   void operator delete(void *p) { ::operator delete(p); }
  94
  95   StringRef getBufferIdentifier() const override {
  96     // The name is stored after the class itself.
  97     return StringRef(reinterpret_cast<const char *>(this + 1));
  98   }
  99
 100   MemoryBuffer::BufferKind getBufferKind() const override {
 101     return MemoryBuffer::MemoryBuffer_Malloc;
 102   }
 103 };
 104 }
 105
 106 template <typename MB>
 107 static ErrorOr<std::unique_ptr<MB>>
 108 getFileAux(const Twine &Filename, int64_t FileSize, uint64_t MapSize,
 109            uint64_t Offset, bool RequiresNullTerminator, bool IsVolatile);
 110
 111 std::unique_ptr<MemoryBuffer>
 112 MemoryBuffer::getMemBuffer(StringRef InputData, StringRef BufferName,
 113                            bool RequiresNullTerminator) {
 114   auto *Ret = new (NamedBufferAlloc(BufferName))
 115       MemoryBufferMem<MemoryBuffer>(InputData, RequiresNullTerminator);
 116   return std::unique_ptr<MemoryBuffer>(Ret);
 117 }
 118
 119 std::unique_ptr<MemoryBuffer>
 120 MemoryBuffer::getMemBuffer(MemoryBufferRef Ref, bool RequiresNullTerminator) {
 121   return std::unique_ptr<MemoryBuffer>(getMemBuffer(
 122       Ref.getBuffer(), Ref.getBufferIdentifier(), RequiresNullTerminator));
 123 }
 124
 125 static ErrorOr<std::unique_ptr<WritableMemoryBuffer>>
 126 getMemBufferCopyImpl(StringRef InputData, const Twine &BufferName) {
 127   auto Buf = WritableMemoryBuffer::getNewUninitMemBuffer(InputData.size(), BufferName);
 128   if (!Buf)
 129     return make_error_code(errc::not_enough_memory);
 130   memcpy(Buf->getBufferStart(), InputData.data(), InputData.size());
 131   return std::move(Buf);
 132 }
 133
 134 std::unique_ptr<MemoryBuffer>
 135 MemoryBuffer::getMemBufferCopy(StringRef InputData, const Twine &BufferName) {
 136   auto Buf = getMemBufferCopyImpl(InputData, BufferName);
 137   if (Buf)
 138     return std::move(*Buf);
 139   return nullptr;
 140 }
 141
 142 ErrorOr<std::unique_ptr<MemoryBuffer>>
 143 MemoryBuffer::getFileOrSTDIN(const Twine &Filename, int64_t FileSize,
 144                              bool RequiresNullTerminator) {
 145   SmallString<256> NameBuf;
 146   StringRef NameRef = Filename.toStringRef(NameBuf);
 147
 148   if (NameRef == "-")
 149     return getSTDIN();
 150   return getFile(Filename, FileSize, RequiresNullTerminator);
 151 }
 152
 153 ErrorOr<std::unique_ptr<MemoryBuffer>>
 154 MemoryBuffer::getFileSlice(const Twine &FilePath, uint64_t MapSize,
 155                            uint64_t Offset, bool IsVolatile) {
 156   return getFileAux<MemoryBuffer>(FilePath, -1, MapSize, Offset, false,
 157                                   IsVolatile);
 158 }
 159
 160 //===----------------------------------------------------------------------===//
 161 // MemoryBuffer::getFile implementation.
 162 //===----------------------------------------------------------------------===//
 163
 164 namespace {
 165 /// Memory maps a file descriptor using sys::fs::mapped_file_region.
 166 ///
 167 /// This handles converting the offset into a legal offset on the platform.
 168 template<typename MB>
 169 class MemoryBufferMMapFile : public MB {
 170   sys::fs::mapped_file_region MFR;
 171
 172   static uint64_t getLegalMapOffset(uint64_t Offset) {
 173     return Offset & ~(sys::fs::mapped_file_region::alignment() - 1);
 174   }
 175
 176   static uint64_t getLegalMapSize(uint64_t Len, uint64_t Offset) {
 177     return Len + (Offset - getLegalMapOffset(Offset));
 178   }
 179
 180   const char *getStart(uint64_t Len, uint64_t Offset) {
 181     return MFR.const_data() + (Offset - getLegalMapOffset(Offset));
 182   }
 183
 184 public:
 185   MemoryBufferMMapFile(bool RequiresNullTerminator, sys::fs::file_t FD, uint64_t Len,
 186                        uint64_t Offset, std::error_code &EC)
 187       : MFR(FD, MB::Mapmode, getLegalMapSize(Len, Offset),
 188             getLegalMapOffset(Offset), EC) {
 189     if (!EC) {
 190       const char *Start = getStart(Len, Offset);
 191       MemoryBuffer::init(Start, Start + Len, RequiresNullTerminator);
 192     }
 193   }
 194
 195   /// Disable sized deallocation for MemoryBufferMMapFile, because it has
 196   /// tail-allocated data.
 197   void operator delete(void *p) { ::operator delete(p); }
 198
 199   StringRef getBufferIdentifier() const override {
 200     // The name is stored after the class itself.
 201     return StringRef(reinterpret_cast<const char *>(this + 1));
 202   }
 203
 204   MemoryBuffer::BufferKind getBufferKind() const override {
 205     return MemoryBuffer::MemoryBuffer_MMap;
 206   }
 207 };
 208 }
 209
 210 static ErrorOr<std::unique_ptr<WritableMemoryBuffer>>
 211 getMemoryBufferForStream(sys::fs::file_t FD, const Twine &BufferName) {
 212   const ssize_t ChunkSize = 4096*4;
 213   SmallString<ChunkSize> Buffer;
 214   // Read into Buffer until we hit EOF.
 215   for (;;) {
 216     Buffer.reserve(Buffer.size() + ChunkSize);
 217     Expected<size_t> ReadBytes = sys::fs::readNativeFile(
 218         FD, makeMutableArrayRef(Buffer.end(), ChunkSize));
 219     if (!ReadBytes)
 220       return errorToErrorCode(ReadBytes.takeError());
 221     if (*ReadBytes == 0)
 222       break;
 223     Buffer.set_size(Buffer.size() + *ReadBytes);
 224   }
 225
 226   return getMemBufferCopyImpl(Buffer, BufferName);
 227 }
 228
 229
 230 ErrorOr<std::unique_ptr<MemoryBuffer>>
 231 MemoryBuffer::getFile(const Twine &Filename, int64_t FileSize,
 232                       bool RequiresNullTerminator, bool IsVolatile) {
 233   return getFileAux<MemoryBuffer>(Filename, FileSize, FileSize, 0,
 234                                   RequiresNullTerminator, IsVolatile);
 235 }
 236
 237 template <typename MB>
 238 static ErrorOr<std::unique_ptr<MB>>
 239 getOpenFileImpl(sys::fs::file_t FD, const Twine &Filename, uint64_t FileSize,
 240                 uint64_t MapSize, int64_t Offset, bool RequiresNullTerminator,
 241                 bool IsVolatile);
 242
 243 template <typename MB>
 244 static ErrorOr<std::unique_ptr<MB>>
 245 getFileAux(const Twine &Filename, int64_t FileSize, uint64_t MapSize,
 246            uint64_t Offset, bool RequiresNullTerminator, bool IsVolatile) {
 247   Expected<sys::fs::file_t> FDOrErr =
 248       sys::fs::openNativeFileForRead(Filename, sys::fs::OF_None);
 249   if (!FDOrErr)
 250     return errorToErrorCode(FDOrErr.takeError());
 251   sys::fs::file_t FD = *FDOrErr;
 252   auto Ret = getOpenFileImpl<MB>(FD, Filename, FileSize, MapSize, Offset,
 253                                  RequiresNullTerminator, IsVolatile);
 254   sys::fs::closeFile(FD);
 255   return Ret;
 256 }
 257
 258 ErrorOr<std::unique_ptr<WritableMemoryBuffer>>
 259 WritableMemoryBuffer::getFile(const Twine &Filename, int64_t FileSize,
 260                               bool IsVolatile) {
 261   return getFileAux<WritableMemoryBuffer>(Filename, FileSize, FileSize, 0,
 262                                           /*RequiresNullTerminator*/ false,
 263                                           IsVolatile);
 264 }
 265
 266 ErrorOr<std::unique_ptr<WritableMemoryBuffer>>
 267 WritableMemoryBuffer::getFileSlice(const Twine &Filename, uint64_t MapSize,
 268                                    uint64_t Offset, bool IsVolatile) {
 269   return getFileAux<WritableMemoryBuffer>(Filename, -1, MapSize, Offset, false,
 270                                           IsVolatile);
 271 }
 272
 273 std::unique_ptr<WritableMemoryBuffer>
 274 WritableMemoryBuffer::getNewUninitMemBuffer(size_t Size, const Twine &BufferName) {
 275   using MemBuffer = MemoryBufferMem<WritableMemoryBuffer>;
 276   // Allocate space for the MemoryBuffer, the data and the name. It is important
 277   // that MemoryBuffer and data are aligned so PointerIntPair works with them.
 278   // TODO: Is 16-byte alignment enough?  We copy small object files with large
 279   // alignment expectations into this buffer.
 280   SmallString<256> NameBuf;
 281   StringRef NameRef = BufferName.toStringRef(NameBuf);
 282   size_t AlignedStringLen = alignTo(sizeof(MemBuffer) + NameRef.size() + 1, 16);
 283   size_t RealLen = AlignedStringLen + Size + 1;
 284   char *Mem = static_cast<char*>(operator new(RealLen, std::nothrow));
 285   if (!Mem)
 286     return nullptr;
 287
 288   // The name is stored after the class itself.
 289   CopyStringRef(Mem + sizeof(MemBuffer), NameRef);
 290
 291   // The buffer begins after the name and must be aligned.
 292   char *Buf = Mem + AlignedStringLen;
 293   Buf[Size] = 0; // Null terminate buffer.
 294
 295   auto *Ret = new (Mem) MemBuffer(StringRef(Buf, Size), true);
 296   return std::unique_ptr<WritableMemoryBuffer>(Ret);
 297 }
 298
 299 std::unique_ptr<WritableMemoryBuffer>
 300 WritableMemoryBuffer::getNewMemBuffer(size_t Size, const Twine &BufferName) {
 301   auto SB = WritableMemoryBuffer::getNewUninitMemBuffer(Size, BufferName);
 302   if (!SB)
 303     return nullptr;
 304   memset(SB->getBufferStart(), 0, Size);
 305   return SB;
 306 }
 307
 308 static bool shouldUseMmap(sys::fs::file_t FD,
 309                           size_t FileSize,
 310                           size_t MapSize,
 311                           off_t Offset,
 312                           bool RequiresNullTerminator,
 313                           int PageSize,
 314                           bool IsVolatile) {
 315   // mmap may leave the buffer without null terminator if the file size changed
 316   // by the time the last page is mapped in, so avoid it if the file size is
 317   // likely to change.
 318   if (IsVolatile)
 319     return false;
 320
 321   // We don't use mmap for small files because this can severely fragment our
 322   // address space.
 323   if (MapSize < 4 * 4096 || MapSize < (unsigned)PageSize)
 324     return false;
 325
 326   if (!RequiresNullTerminator)
 327     return true;
 328
 329   // If we don't know the file size, use fstat to find out.  fstat on an open
 330   // file descriptor is cheaper than stat on a random path.
 331   // FIXME: this chunk of code is duplicated, but it avoids a fstat when
 332   // RequiresNullTerminator = false and MapSize != -1.
 333   if (FileSize == size_t(-1)) {
 334     sys::fs::file_status Status;
 335     if (sys::fs::status(FD, Status))
 336       return false;
 337     FileSize = Status.getSize();
 338   }
 339
 340   // If we need a null terminator and the end of the map is inside the file,
 341   // we cannot use mmap.
 342   size_t End = Offset + MapSize;
 343   assert(End <= FileSize);
 344   if (End != FileSize)
 345     return false;
 346
 347   // Don't try to map files that are exactly a multiple of the system page size
 348   // if we need a null terminator.
 349   if ((FileSize & (PageSize -1)) == 0)
 350     return false;
 351
 352 #if defined(__CYGWIN__)
 353   // Don't try to map files that are exactly a multiple of the physical page size
 354   // if we need a null terminator.
 355   // FIXME: We should reorganize again getPageSize() on Win32.
 356   if ((FileSize & (4096 - 1)) == 0)
 357     return false;
 358 #endif
 359
 360   return true;
 361 }
 362
 363 static ErrorOr<std::unique_ptr<WriteThroughMemoryBuffer>>
 364 getReadWriteFile(const Twine &Filename, uint64_t FileSize, uint64_t MapSize,
 365                  uint64_t Offset) {
 366   Expected<sys::fs::file_t> FDOrErr = sys::fs::openNativeFileForReadWrite(
 367       Filename, sys::fs::CD_OpenExisting, sys::fs::OF_None);
 368   if (!FDOrErr)
 369     return errorToErrorCode(FDOrErr.takeError());
 370   sys::fs::file_t FD = *FDOrErr;
 371
 372   // Default is to map the full file.
 373   if (MapSize == uint64_t(-1)) {
 374     // If we don't know the file size, use fstat to find out.  fstat on an open
 375     // file descriptor is cheaper than stat on a random path.
 376     if (FileSize == uint64_t(-1)) {
 377       sys::fs::file_status Status;
 378       std::error_code EC = sys::fs::status(FD, Status);
 379       if (EC)
 380         return EC;
 381
 382       // If this not a file or a block device (e.g. it's a named pipe
 383       // or character device), we can't mmap it, so error out.
 384       sys::fs::file_type Type = Status.type();
 385       if (Type != sys::fs::file_type::regular_file &&
 386           Type != sys::fs::file_type::block_file)
 387         return make_error_code(errc::invalid_argument);
 388
 389       FileSize = Status.getSize();
 390     }
 391     MapSize = FileSize;
 392   }
 393
 394   std::error_code EC;
 395   std::unique_ptr<WriteThroughMemoryBuffer> Result(
 396       new (NamedBufferAlloc(Filename))
 397           MemoryBufferMMapFile<WriteThroughMemoryBuffer>(false, FD, MapSize,
 398                                                          Offset, EC));
 399   if (EC)
 400     return EC;
 401   return std::move(Result);
 402 }
 403
 404 ErrorOr<std::unique_ptr<WriteThroughMemoryBuffer>>
 405 WriteThroughMemoryBuffer::getFile(const Twine &Filename, int64_t FileSize) {
 406   return getReadWriteFile(Filename, FileSize, FileSize, 0);
 407 }
 408
 409 /// Map a subrange of the specified file as a WritableMemoryBuffer.
 410 ErrorOr<std::unique_ptr<WriteThroughMemoryBuffer>>
 411 WriteThroughMemoryBuffer::getFileSlice(const Twine &Filename, uint64_t MapSize,
 412                                        uint64_t Offset) {
 413   return getReadWriteFile(Filename, -1, MapSize, Offset);
 414 }
 415
 416 template <typename MB>
 417 static ErrorOr<std::unique_ptr<MB>>
 418 getOpenFileImpl(sys::fs::file_t FD, const Twine &Filename, uint64_t FileSize,
 419                 uint64_t MapSize, int64_t Offset, bool RequiresNullTerminator,
 420                 bool IsVolatile) {
 421   static int PageSize = sys::Process::getPageSizeEstimate();
 422
 423   // Default is to map the full file.
 424   if (MapSize == uint64_t(-1)) {
 425     // If we don't know the file size, use fstat to find out.  fstat on an open
 426     // file descriptor is cheaper than stat on a random path.
 427     if (FileSize == uint64_t(-1)) {
 428       sys::fs::file_status Status;
 429       std::error_code EC = sys::fs::status(FD, Status);
 430       if (EC)
 431         return EC;
 432
 433       // If this not a file or a block device (e.g. it's a named pipe
 434       // or character device), we can't trust the size. Create the memory
 435       // buffer by copying off the stream.
 436       sys::fs::file_type Type = Status.type();
 437       if (Type != sys::fs::file_type::regular_file &&
 438           Type != sys::fs::file_type::block_file)
 439         return getMemoryBufferForStream(FD, Filename);
 440
 441       FileSize = Status.getSize();
 442     }
 443     MapSize = FileSize;
 444   }
 445
 446   if (shouldUseMmap(FD, FileSize, MapSize, Offset, RequiresNullTerminator,
 447                     PageSize, IsVolatile)) {
 448     std::error_code EC;
 449     std::unique_ptr<MB> Result(
 450         new (NamedBufferAlloc(Filename)) MemoryBufferMMapFile<MB>(
 451             RequiresNullTerminator, FD, MapSize, Offset, EC));
 452     if (!EC)
 453       return std::move(Result);
 454   }
 455
 456   auto Buf = WritableMemoryBuffer::getNewUninitMemBuffer(MapSize, Filename);
 457   if (!Buf) {
 458     // Failed to create a buffer. The only way it can fail is if
 459     // new(std::nothrow) returns 0.
 460     return make_error_code(errc::not_enough_memory);
 461   }
 462
 463   // Read until EOF, zero-initialize the rest.
 464   MutableArrayRef<char> ToRead = Buf->getBuffer();
 465   while (!ToRead.empty()) {
 466     Expected<size_t> ReadBytes =
 467         sys::fs::readNativeFileSlice(FD, ToRead, Offset);
 468     if (!ReadBytes)
 469       return errorToErrorCode(ReadBytes.takeError());
 470     if (*ReadBytes == 0) {
 471       std::memset(ToRead.data(), 0, ToRead.size());
 472       break;
 473     }
 474     ToRead = ToRead.drop_front(*ReadBytes);
 475     Offset += *ReadBytes;
 476   }
 477
 478   return std::move(Buf);
 479 }
 480
 481 ErrorOr<std::unique_ptr<MemoryBuffer>>
 482 MemoryBuffer::getOpenFile(sys::fs::file_t FD, const Twine &Filename, uint64_t FileSize,
 483                           bool RequiresNullTerminator, bool IsVolatile) {
 484   return getOpenFileImpl<MemoryBuffer>(FD, Filename, FileSize, FileSize, 0,
 485                          RequiresNullTerminator, IsVolatile);
 486 }
 487
 488 ErrorOr<std::unique_ptr<MemoryBuffer>>
 489 MemoryBuffer::getOpenFileSlice(sys::fs::file_t FD, const Twine &Filename, uint64_t MapSize,
 490                                int64_t Offset, bool IsVolatile) {
 491   assert(MapSize != uint64_t(-1));
 492   return getOpenFileImpl<MemoryBuffer>(FD, Filename, -1, MapSize, Offset, false,
 493                                        IsVolatile);
 494 }
 495
 496 ErrorOr<std::unique_ptr<MemoryBuffer>> MemoryBuffer::getSTDIN() {
 497   // Read in all of the data from stdin, we cannot mmap stdin.
 498   //
 499   // FIXME: That isn't necessarily true, we should try to mmap stdin and
 500   // fallback if it fails.
 501   sys::ChangeStdinToBinary();
 502
 503   return getMemoryBufferForStream(sys::fs::getStdinHandle(), "<stdin>");
 504 }
 505
 506 ErrorOr<std::unique_ptr<MemoryBuffer>>
 507 MemoryBuffer::getFileAsStream(const Twine &Filename) {
 508   Expected<sys::fs::file_t> FDOrErr =
 509       sys::fs::openNativeFileForRead(Filename, sys::fs::OF_None);
 510   if (!FDOrErr)
 511     return errorToErrorCode(FDOrErr.takeError());
 512   sys::fs::file_t FD = *FDOrErr;
 513   ErrorOr<std::unique_ptr<MemoryBuffer>> Ret =
 514       getMemoryBufferForStream(FD, Filename);
 515   sys::fs::closeFile(FD);
 516   return Ret;
 517 }
 518
 519 MemoryBufferRef MemoryBuffer::getMemBufferRef() const {
 520   StringRef Data = getBuffer();
 521   StringRef Identifier = getBufferIdentifier();
 522   return MemoryBufferRef(Data, Identifier);
 523 }
 524
 525 SmallVectorMemoryBuffer::~SmallVectorMemoryBuffer() {}