144ebf8b0331cb3a994a8df16fb63fdcdbc37d927Nick Lewycky//===- StreamableMemoryObject.h - Streamable data interface -----*- C++ -*-===//
22ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff//
32ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff//                     The LLVM Compiler Infrastructure
42ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff//
52ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff// This file is distributed under the University of Illinois Open Source
62ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff// License. See LICENSE.TXT for details.
72ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff//
82ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff//===----------------------------------------------------------------------===//
92ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
102ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
11674be02d525d4e24bc6943ed9274958c580bcfbcJakub Staszak#ifndef LLVM_SUPPORT_STREAMABLEMEMORYOBJECT_H
12674be02d525d4e24bc6943ed9274958c580bcfbcJakub Staszak#define LLVM_SUPPORT_STREAMABLEMEMORYOBJECT_H
132ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
14f630e49efc7bf3f1716b6daab3c2cc11a908754aCraig Topper#include "llvm/Support/Compiler.h"
152ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff#include "llvm/Support/DataStream.h"
16cd81d94322a39503e4a3e87b6ee03d4fcb3465fbStephen Hines#include "llvm/Support/ErrorHandling.h"
17255f89faee13dc491cb64fbeae3c763e7e2ea4e6Chandler Carruth#include "llvm/Support/MemoryObject.h"
1836b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines#include <cassert>
1936b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines#include <memory>
202ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff#include <vector>
212ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
222ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuffnamespace llvm {
232ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
242ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// StreamableMemoryObject - Interface to data which might be streamed.
252ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// Streamability has 2 important implications/restrictions. First, the data
262ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// might not yet exist in memory when the request is made. This just means
272ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// that readByte/readBytes might have to block or do some work to get it.
282ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// More significantly, the exact size of the object might not be known until
292ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// it has all been fetched. This means that to return the right result,
302ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// getExtent must also wait for all the data to arrive; therefore it should
312ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// not be called on objects which are actually streamed (this would defeat
322ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// the purpose of streaming). Instead, isValidAddress and isObjectEnd can be
332ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// used to test addresses without knowing the exact size of the stream.
342ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// Finally, getPointer can be used instead of readBytes to avoid extra copying.
352ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuffclass StreamableMemoryObject : public MemoryObject {
362ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff public:
372ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// Destructor      - Override as necessary.
382ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  virtual ~StreamableMemoryObject();
392ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
402ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// getBase         - Returns the lowest valid address in the region.
412ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///
422ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @result         - The lowest valid address.
4336b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines  uint64_t getBase() const override = 0;
442ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
452ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// getExtent       - Returns the size of the region in bytes.  (The region is
462ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                   contiguous, so the highest valid address of the region
472ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                   is getBase() + getExtent() - 1).
482ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                   May block until all bytes in the stream have been read
492ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///
502ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @result         - The size of the region.
5136b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines  uint64_t getExtent() const override = 0;
522ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
532ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// readByte        - Tries to read a single byte from the region.
542ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                   May block until (address - base) bytes have been read
552ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @param address  - The address of the byte, in the same space as getBase().
562ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @param ptr      - A pointer to a byte to be filled in.  Must be non-NULL.
572ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @result         - 0 if successful; -1 if not.  Failure may be due to a
582ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                   bounds violation or an implementation-specific error.
5936b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines  int readByte(uint64_t address, uint8_t *ptr) const override = 0;
602ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
612ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// readBytes       - Tries to read a contiguous range of bytes from the
622ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                   region, up to the end of the region.
632ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                   May block until (address - base + size) bytes have
642ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                   been read. Additionally, StreamableMemoryObjects will
652ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                   not do partial reads - if size bytes cannot be read,
662ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                   readBytes will fail.
672ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///
682ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @param address  - The address of the first byte, in the same space as
692ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                   getBase().
7049a6a8d8f2994249c81b7914b07015714748a55cBenjamin Kramer  /// @param size     - The number of bytes to copy.
712ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @param buf      - A pointer to a buffer to be filled in.  Must be non-NULL
722ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                   and large enough to hold size bytes.
732ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @result         - 0 if successful; -1 if not.  Failure may be due to a
742ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                   bounds violation or an implementation-specific error.
7536b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines  int readBytes(uint64_t address, uint64_t size,
7636b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines                uint8_t *buf) const override = 0;
772ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
782ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// getPointer  - Ensures that the requested data is in memory, and returns
792ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///               A pointer to it. More efficient than using readBytes if the
802ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///               data is already in memory.
812ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///               May block until (address - base + size) bytes have been read
822ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @param address - address of the byte, in the same space as getBase()
832ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @param size    - amount of data that must be available on return
842ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @result        - valid pointer to the requested data
85adef06a71458ded0716935a61b3d43d164d4df12Derek Schuff  virtual const uint8_t *getPointer(uint64_t address, uint64_t size) const = 0;
862ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
872ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// isValidAddress - Returns true if the address is within the object
882ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                  (i.e. between base and base + extent - 1 inclusive)
892ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                  May block until (address - base) bytes have been read
902ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @param address - address of the byte, in the same space as getBase()
912ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @result        - true if the address may be read with readByte()
92adef06a71458ded0716935a61b3d43d164d4df12Derek Schuff  virtual bool isValidAddress(uint64_t address) const = 0;
932ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
942ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// isObjectEnd    - Returns true if the address is one past the end of the
952ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                  object (i.e. if it is equal to base + extent)
962ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  ///                  May block until (address - base) bytes have been read
972ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @param address - address of the byte, in the same space as getBase()
982ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// @result        - true if the address is equal to base + extent
99adef06a71458ded0716935a61b3d43d164d4df12Derek Schuff  virtual bool isObjectEnd(uint64_t address) const = 0;
1002ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff};
1012ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
1022ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// StreamingMemoryObject - interface to data which is actually streamed from
1032ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// a DataStreamer. In addition to inherited members, it has the
1042ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// dropLeadingBytes and setKnownObjectSize methods which are not applicable
1052ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff/// to non-streamed objects.
1062ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuffclass StreamingMemoryObject : public StreamableMemoryObject {
1072ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuffpublic:
1082ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  StreamingMemoryObject(DataStreamer *streamer);
10936b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines  uint64_t getBase() const override { return 0; }
11036b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines  uint64_t getExtent() const override;
11136b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines  int readByte(uint64_t address, uint8_t *ptr) const override;
11236b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines  int readBytes(uint64_t address, uint64_t size,
11336b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines                uint8_t *buf) const override;
11436b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines  const uint8_t *getPointer(uint64_t address, uint64_t size) const override {
1152ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff    // This could be fixed by ensuring the bytes are fetched and making a copy,
1162ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff    // requiring that the bitcode size be known, or otherwise ensuring that
1172ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff    // the memory doesn't go away/get reallocated, but it's
1182ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff    // not currently necessary. Users that need the pointer don't stream.
119cd81d94322a39503e4a3e87b6ee03d4fcb3465fbStephen Hines    llvm_unreachable("getPointer in streaming memory objects not allowed");
120dce4a407a24b04eebc6a376f8e62b41aaa7b071fStephen Hines    return nullptr;
1212ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  }
12236b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines  bool isValidAddress(uint64_t address) const override;
12336b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines  bool isObjectEnd(uint64_t address) const override;
1242ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
1252ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// Drop s bytes from the front of the stream, pushing the positions of the
1262ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// remaining bytes down by s. This is used to skip past the bitcode header,
1272ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// since we don't know a priori if it's present, and we can't put bytes
1282ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// back into the stream once we've read them.
1292ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  bool dropLeadingBytes(size_t s);
1302ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
1312ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// If the data object size is known in advance, many of the operations can
1322ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// be made more efficient, so this method should be called before reading
1332ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  /// starts (although it can be called anytime).
1342ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  void setKnownObjectSize(size_t size);
1352ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
1362ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuffprivate:
1372ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  const static uint32_t kChunkSize = 4096 * 4;
138adef06a71458ded0716935a61b3d43d164d4df12Derek Schuff  mutable std::vector<unsigned char> Bytes;
13936b56886974eae4f9c5ebc96befd3e7bfe5de338Stephen Hines  std::unique_ptr<DataStreamer> Streamer;
140adef06a71458ded0716935a61b3d43d164d4df12Derek Schuff  mutable size_t BytesRead;   // Bytes read from stream
1412ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  size_t BytesSkipped;// Bytes skipped at start of stream (e.g. wrapper/header)
142adef06a71458ded0716935a61b3d43d164d4df12Derek Schuff  mutable size_t ObjectSize; // 0 if unknown, set if wrapper seen or EOF reached
143adef06a71458ded0716935a61b3d43d164d4df12Derek Schuff  mutable bool EOFReached;
1442ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
1452ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  // Fetch enough bytes such that Pos can be read or EOF is reached
1462ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  // (i.e. BytesRead > Pos). Return true if Pos can be read.
1472ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  // Unlike most of the functions in BitcodeReader, returns true on success.
1482ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  // Most of the requests will be small, but we fetch at kChunkSize bytes
1492ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  // at a time to avoid making too many potentially expensive GetBytes calls
150adef06a71458ded0716935a61b3d43d164d4df12Derek Schuff  bool fetchToPos(size_t Pos) const {
1512ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff    if (EOFReached) return Pos < ObjectSize;
1522ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff    while (Pos >= BytesRead) {
153cd4f803ac1b366084bdf691d5da5563704e36dfeDerek Schuff      Bytes.resize(BytesRead + BytesSkipped + kChunkSize);
1542ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff      size_t bytes = Streamer->GetBytes(&Bytes[BytesRead + BytesSkipped],
1552ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff                                        kChunkSize);
1562ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff      BytesRead += bytes;
1572ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff      if (bytes < kChunkSize) {
158cd81d94322a39503e4a3e87b6ee03d4fcb3465fbStephen Hines        assert((!ObjectSize || BytesRead >= Pos) &&
159cd81d94322a39503e4a3e87b6ee03d4fcb3465fbStephen Hines               "Unexpected short read fetching bitcode");
1602ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff        if (BytesRead <= Pos) { // reached EOF/ran out of bytes
1612ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff          ObjectSize = BytesRead;
1622ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff          EOFReached = true;
1632ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff          return false;
1642ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff        }
1652ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff      }
1662ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff    }
1672ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff    return true;
1682ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff  }
1692ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
170f630e49efc7bf3f1716b6daab3c2cc11a908754aCraig Topper  StreamingMemoryObject(const StreamingMemoryObject&) LLVM_DELETED_FUNCTION;
171f630e49efc7bf3f1716b6daab3c2cc11a908754aCraig Topper  void operator=(const StreamingMemoryObject&) LLVM_DELETED_FUNCTION;
1722ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff};
1732ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
1742ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek SchuffStreamableMemoryObject *getNonStreamedMemoryObject(
1752ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff    const unsigned char *Start, const unsigned char *End);
1762ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff
1772ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff}
1782ea93875b2f2900b9d244dfd7649c9ed02a34cd7Derek Schuff#endif  // STREAMABLEMEMORYOBJECT_H_
179