Aaron Green 6708186c91 [crt][fuzzer] Fix up various numeric conversions
Attempting to build a standalone libFuzzer in Fuchsia's default toolchain for the purpose of cross-compiling the unit tests  revealed a number of not-quite-proper type conversions. Fuchsia's toolchain include `-std=c++17` and `-Werror`, among others, leading to many errors like `-Wshorten-64-to-32`, `-Wimplicit-float-conversion`, etc.

Most of these have been addressed by simply making the conversion explicit with a `static_cast`. These typically fell into one of two categories: 1) conversions between types where high precision isn't critical, e.g. the "energy" calculations for `InputInfo`, and 2) conversions where the values will never reach the bits being truncated, e.g. `DftTimeInSeconds` is not going to exceed 136 years.

The major exception to this is the number of features: there are several places that treat features as `size_t`, and others as `uint32_t`. This change makes the decision to cap the features at 32 bits. The maximum value of a feature as produced by `TracePC::CollectFeatures` is roughly:
  (NumPCsInPCTables + ValueBitMap::kMapSizeInBits + ExtraCountersBegin() - ExtraCountersEnd() + log2(SIZE_MAX)) * 8

It's conceivable for extremely large targets and/or extra counters that this limit could be reached. This shouldn't break fuzzing, but it will cause certain features to collide and lower the fuzzers overall precision. To address this, this change adds a warning to TracePC::PrintModuleInfo about excessive feature size if it is detected, and recommends refactoring the fuzzer into several smaller ones.

Reviewed By: morehouse

Differential Revision: https://reviews.llvm.org/D97992
2021-03-11 16:01:28 -08:00

225 lines
5.7 KiB
C++

//===- FuzzerSHA1.h - Private copy of the SHA1 implementation ---*- C++ -* ===//
//
// 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
//
//===----------------------------------------------------------------------===//
// This code is taken from public domain
// (http://oauth.googlecode.com/svn/code/c/liboauth/src/sha1.c)
// and modified by adding anonymous namespace, adding an interface
// function fuzzer::ComputeSHA1() and removing unnecessary code.
//
// lib/Fuzzer can not use SHA1 implementation from openssl because
// openssl may not be available and because we may be fuzzing openssl itself.
// For the same reason we do not want to depend on SHA1 from LLVM tree.
//===----------------------------------------------------------------------===//
#include "FuzzerSHA1.h"
#include "FuzzerDefs.h"
#include "FuzzerPlatform.h"
/* This code is public-domain - it is based on libcrypt
* placed in the public domain by Wei Dai and other contributors.
*/
#include <iomanip>
#include <sstream>
#include <stdint.h>
#include <string.h>
namespace { // Added for LibFuzzer
#ifdef __BIG_ENDIAN__
# define SHA_BIG_ENDIAN
// Windows is always little endian and MSVC doesn't have <endian.h>
#elif defined __LITTLE_ENDIAN__ || LIBFUZZER_WINDOWS
/* override */
#elif defined __BYTE_ORDER
# if __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
# define SHA_BIG_ENDIAN
# endif
#else // ! defined __LITTLE_ENDIAN__
# include <endian.h> // machine/endian.h
# if __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
# define SHA_BIG_ENDIAN
# endif
#endif
/* header */
#define HASH_LENGTH 20
#define BLOCK_LENGTH 64
typedef struct sha1nfo {
uint32_t buffer[BLOCK_LENGTH/4];
uint32_t state[HASH_LENGTH/4];
uint32_t byteCount;
uint8_t bufferOffset;
uint8_t keyBuffer[BLOCK_LENGTH];
uint8_t innerHash[HASH_LENGTH];
} sha1nfo;
/* public API - prototypes - TODO: doxygen*/
/**
*/
void sha1_init(sha1nfo *s);
/**
*/
void sha1_writebyte(sha1nfo *s, uint8_t data);
/**
*/
void sha1_write(sha1nfo *s, const char *data, size_t len);
/**
*/
uint8_t* sha1_result(sha1nfo *s);
/* code */
#define SHA1_K0 0x5a827999
#define SHA1_K20 0x6ed9eba1
#define SHA1_K40 0x8f1bbcdc
#define SHA1_K60 0xca62c1d6
void sha1_init(sha1nfo *s) {
s->state[0] = 0x67452301;
s->state[1] = 0xefcdab89;
s->state[2] = 0x98badcfe;
s->state[3] = 0x10325476;
s->state[4] = 0xc3d2e1f0;
s->byteCount = 0;
s->bufferOffset = 0;
}
uint32_t sha1_rol32(uint32_t number, uint8_t bits) {
return ((number << bits) | (number >> (32-bits)));
}
void sha1_hashBlock(sha1nfo *s) {
uint8_t i;
uint32_t a,b,c,d,e,t;
a=s->state[0];
b=s->state[1];
c=s->state[2];
d=s->state[3];
e=s->state[4];
for (i=0; i<80; i++) {
if (i>=16) {
t = s->buffer[(i+13)&15] ^ s->buffer[(i+8)&15] ^ s->buffer[(i+2)&15] ^ s->buffer[i&15];
s->buffer[i&15] = sha1_rol32(t,1);
}
if (i<20) {
t = (d ^ (b & (c ^ d))) + SHA1_K0;
} else if (i<40) {
t = (b ^ c ^ d) + SHA1_K20;
} else if (i<60) {
t = ((b & c) | (d & (b | c))) + SHA1_K40;
} else {
t = (b ^ c ^ d) + SHA1_K60;
}
t+=sha1_rol32(a,5) + e + s->buffer[i&15];
e=d;
d=c;
c=sha1_rol32(b,30);
b=a;
a=t;
}
s->state[0] += a;
s->state[1] += b;
s->state[2] += c;
s->state[3] += d;
s->state[4] += e;
}
// Adds the least significant byte of |data|.
void sha1_addUncounted(sha1nfo *s, uint32_t data) {
uint8_t *const b = (uint8_t *)s->buffer;
#ifdef SHA_BIG_ENDIAN
b[s->bufferOffset] = static_cast<uint8_t>(data);
#else
b[s->bufferOffset ^ 3] = static_cast<uint8_t>(data);
#endif
s->bufferOffset++;
if (s->bufferOffset == BLOCK_LENGTH) {
sha1_hashBlock(s);
s->bufferOffset = 0;
}
}
void sha1_writebyte(sha1nfo *s, uint8_t data) {
++s->byteCount;
sha1_addUncounted(s, data);
}
void sha1_write(sha1nfo *s, const char *data, size_t len) {
for (;len--;) sha1_writebyte(s, (uint8_t) *data++);
}
void sha1_pad(sha1nfo *s) {
// Implement SHA-1 padding (fips180-2 §5.1.1)
// Pad with 0x80 followed by 0x00 until the end of the block
sha1_addUncounted(s, 0x80);
while (s->bufferOffset != 56) sha1_addUncounted(s, 0x00);
// Append length in the last 8 bytes
sha1_addUncounted(s, 0); // We're only using 32 bit lengths
sha1_addUncounted(s, 0); // But SHA-1 supports 64 bit lengths
sha1_addUncounted(s, 0); // So zero pad the top bits
sha1_addUncounted(s, s->byteCount >> 29); // Shifting to multiply by 8
sha1_addUncounted(s, s->byteCount >> 21); // as SHA-1 supports bitstreams as well as
sha1_addUncounted(s, s->byteCount >> 13); // byte.
sha1_addUncounted(s, s->byteCount >> 5);
sha1_addUncounted(s, s->byteCount << 3);
}
uint8_t* sha1_result(sha1nfo *s) {
// Pad to complete the last block
sha1_pad(s);
#ifndef SHA_BIG_ENDIAN
// Swap byte order back
int i;
for (i=0; i<5; i++) {
s->state[i]=
(((s->state[i])<<24)& 0xff000000)
| (((s->state[i])<<8) & 0x00ff0000)
| (((s->state[i])>>8) & 0x0000ff00)
| (((s->state[i])>>24)& 0x000000ff);
}
#endif
// Return pointer to hash (20 characters)
return (uint8_t*) s->state;
}
} // namespace; Added for LibFuzzer
namespace fuzzer {
// The rest is added for LibFuzzer
void ComputeSHA1(const uint8_t *Data, size_t Len, uint8_t *Out) {
sha1nfo s;
sha1_init(&s);
sha1_write(&s, (const char*)Data, Len);
memcpy(Out, sha1_result(&s), HASH_LENGTH);
}
std::string Sha1ToString(const uint8_t Sha1[kSHA1NumBytes]) {
std::stringstream SS;
for (int i = 0; i < kSHA1NumBytes; i++)
SS << std::hex << std::setfill('0') << std::setw(2) << (unsigned)Sha1[i];
return SS.str();
}
std::string Hash(const Unit &U) {
uint8_t Hash[kSHA1NumBytes];
ComputeSHA1(U.data(), U.size(), Hash);
return Sha1ToString(Hash);
}
}