From e3d2ba7c70b13a2165020e45abf02373a1e953f7 Mon Sep 17 00:00:00 2001 From: practicalswift Date: Mon, 21 Dec 2020 23:19:33 +0000 Subject: [PATCH] fuzz: Update FuzzedDataProvider.h from upstream (LLVM) Upstream revision: https://github.com/llvm/llvm-project/blob/6d0488f75bb2f37bcfe93fc8f59f6e78c9a0c939/compiler-rt/include/fuzzer/FuzzedDataProvider.h Changes: * [compiler-rt] FuzzedDataProvider: add ConsumeData and method. * [compiler-rt] Fix a typo in a comment in FuzzedDataProvider.h. * [compiler-rt] Add ConsumeRandomLengthString() version without arguments. * [compiler-rt] Refactor FuzzedDataProvider for better readability. * [compiler-rt] FuzzedDataProvider: make linter happy. * [compiler-rt] Mark FDP non-template methods inline to avoid ODR violations. --- src/test/fuzz/FuzzedDataProvider.h | 556 +++++++++++++++++------------ 1 file changed, 319 insertions(+), 237 deletions(-) diff --git a/src/test/fuzz/FuzzedDataProvider.h b/src/test/fuzz/FuzzedDataProvider.h index 3e069eba69..83bcd0134a 100644 --- a/src/test/fuzz/FuzzedDataProvider.h +++ b/src/test/fuzz/FuzzedDataProvider.h @@ -34,272 +34,354 @@ class FuzzedDataProvider { : data_ptr_(data), remaining_bytes_(size) {} ~FuzzedDataProvider() = default; - // Returns a std::vector containing |num_bytes| of input data. If fewer than - // |num_bytes| of data remain, returns a shorter std::vector containing all - // of the data that's left. Can be used with any byte sized type, such as - // char, unsigned char, uint8_t, etc. - template std::vector ConsumeBytes(size_t num_bytes) { - num_bytes = std::min(num_bytes, remaining_bytes_); - return ConsumeBytes(num_bytes, num_bytes); - } - - // Similar to |ConsumeBytes|, but also appends the terminator value at the end - // of the resulting vector. Useful, when a mutable null-terminated C-string is - // needed, for example. But that is a rare case. Better avoid it, if possible, - // and prefer using |ConsumeBytes| or |ConsumeBytesAsString| methods. + // See the implementation below (after the class definition) for more verbose + // comments for each of the methods. + + // Methods returning std::vector of bytes. These are the most popular choice + // when splitting fuzzing input into pieces, as every piece is put into a + // separate buffer (i.e. ASan would catch any under-/overflow) and the memory + // will be released automatically. + template std::vector ConsumeBytes(size_t num_bytes); template - std::vector ConsumeBytesWithTerminator(size_t num_bytes, - T terminator = 0) { - num_bytes = std::min(num_bytes, remaining_bytes_); - std::vector result = ConsumeBytes(num_bytes + 1, num_bytes); - result.back() = terminator; - return result; - } - - // Returns a std::string containing |num_bytes| of input data. Using this and - // |.c_str()| on the resulting string is the best way to get an immutable - // null-terminated C string. If fewer than |num_bytes| of data remain, returns - // a shorter std::string containing all of the data that's left. - std::string ConsumeBytesAsString(size_t num_bytes) { - static_assert(sizeof(std::string::value_type) == sizeof(uint8_t), - "ConsumeBytesAsString cannot convert the data to a string."); - - num_bytes = std::min(num_bytes, remaining_bytes_); - std::string result( - reinterpret_cast(data_ptr_), - num_bytes); - Advance(num_bytes); - return result; - } + std::vector ConsumeBytesWithTerminator(size_t num_bytes, T terminator = 0); + template std::vector ConsumeRemainingBytes(); - // Returns a number in the range [min, max] by consuming bytes from the - // input data. The value might not be uniformly distributed in the given - // range. If there's no input data left, always returns |min|. |min| must - // be less than or equal to |max|. - template T ConsumeIntegralInRange(T min, T max) { - static_assert(std::is_integral::value, "An integral type is required."); - static_assert(sizeof(T) <= sizeof(uint64_t), "Unsupported integral type."); + // Methods returning strings. Use only when you need a std::string or a null + // terminated C-string. Otherwise, prefer the methods returning std::vector. + std::string ConsumeBytesAsString(size_t num_bytes); + std::string ConsumeRandomLengthString(size_t max_length); + std::string ConsumeRandomLengthString(); + std::string ConsumeRemainingBytesAsString(); - if (min > max) - abort(); + // Methods returning integer values. + template T ConsumeIntegral(); + template T ConsumeIntegralInRange(T min, T max); - // Use the biggest type possible to hold the range and the result. - uint64_t range = static_cast(max) - min; - uint64_t result = 0; - size_t offset = 0; - - while (offset < sizeof(T) * CHAR_BIT && (range >> offset) > 0 && - remaining_bytes_ != 0) { - // Pull bytes off the end of the seed data. Experimentally, this seems to - // allow the fuzzer to more easily explore the input space. This makes - // sense, since it works by modifying inputs that caused new code to run, - // and this data is often used to encode length of data read by - // |ConsumeBytes|. Separating out read lengths makes it easier modify the - // contents of the data that is actually read. - --remaining_bytes_; - result = (result << CHAR_BIT) | data_ptr_[remaining_bytes_]; - offset += CHAR_BIT; - } + // Methods returning floating point values. + template T ConsumeFloatingPoint(); + template T ConsumeFloatingPointInRange(T min, T max); - // Avoid division by 0, in case |range + 1| results in overflow. - if (range != std::numeric_limits::max()) - result = result % (range + 1); + // 0 <= return value <= 1. + template T ConsumeProbability(); - return static_cast(min + result); - } + bool ConsumeBool(); - // Returns a std::string of length from 0 to |max_length|. When it runs out of - // input data, returns what remains of the input. Designed to be more stable - // with respect to a fuzzer inserting characters than just picking a random - // length and then consuming that many bytes with |ConsumeBytes|. - std::string ConsumeRandomLengthString(size_t max_length) { - // Reads bytes from the start of |data_ptr_|. Maps "\\" to "\", and maps "\" - // followed by anything else to the end of the string. As a result of this - // logic, a fuzzer can insert characters into the string, and the string - // will be lengthened to include those new characters, resulting in a more - // stable fuzzer than picking the length of a string independently from - // picking its contents. - std::string result; - - // Reserve the anticipated capaticity to prevent several reallocations. - result.reserve(std::min(max_length, remaining_bytes_)); - for (size_t i = 0; i < max_length && remaining_bytes_ != 0; ++i) { - char next = ConvertUnsignedToSigned(data_ptr_[0]); - Advance(1); - if (next == '\\' && remaining_bytes_ != 0) { - next = ConvertUnsignedToSigned(data_ptr_[0]); - Advance(1); - if (next != '\\') - break; - } - result += next; - } + // Returns a value chosen from the given enum. + template T ConsumeEnum(); - result.shrink_to_fit(); - return result; - } + // Returns a value from the given array. + template T PickValueInArray(const T (&array)[size]); + template T PickValueInArray(std::initializer_list list); - // Returns a std::vector containing all remaining bytes of the input data. - template std::vector ConsumeRemainingBytes() { - return ConsumeBytes(remaining_bytes_); - } + // Writes data to the given destination and returns number of bytes written. + size_t ConsumeData(void *destination, size_t num_bytes); - // Returns a std::string containing all remaining bytes of the input data. - // Prefer using |ConsumeRemainingBytes| unless you actually need a std::string - // object. - std::string ConsumeRemainingBytesAsString() { - return ConsumeBytesAsString(remaining_bytes_); - } + // Reports the remaining bytes available for fuzzed input. + size_t remaining_bytes() { return remaining_bytes_; } - // Returns a number in the range [Type's min, Type's max]. The value might - // not be uniformly distributed in the given range. If there's no input data - // left, always returns |min|. - template T ConsumeIntegral() { - return ConsumeIntegralInRange(std::numeric_limits::min(), - std::numeric_limits::max()); - } + private: + FuzzedDataProvider(const FuzzedDataProvider &) = delete; + FuzzedDataProvider &operator=(const FuzzedDataProvider &) = delete; - // Reads one byte and returns a bool, or false when no data remains. - bool ConsumeBool() { return 1 & ConsumeIntegral(); } + void CopyAndAdvance(void *destination, size_t num_bytes); - // Returns a copy of the value selected from the given fixed-size |array|. - template - T PickValueInArray(const T (&array)[size]) { - static_assert(size > 0, "The array must be non empty."); - return array[ConsumeIntegralInRange(0, size - 1)]; - } + void Advance(size_t num_bytes); template - T PickValueInArray(std::initializer_list list) { - // TODO(Dor1s): switch to static_assert once C++14 is allowed. - if (!list.size()) - abort(); - - return *(list.begin() + ConsumeIntegralInRange(0, list.size() - 1)); - } - - // Returns an enum value. The enum must start at 0 and be contiguous. It must - // also contain |kMaxValue| aliased to its largest (inclusive) value. Such as: - // enum class Foo { SomeValue, OtherValue, kMaxValue = OtherValue }; - template T ConsumeEnum() { - static_assert(std::is_enum::value, "|T| must be an enum type."); - return static_cast(ConsumeIntegralInRange( - 0, static_cast(T::kMaxValue))); - } + std::vector ConsumeBytes(size_t size, size_t num_bytes); - // Returns a floating point number in the range [0.0, 1.0]. If there's no - // input data left, always returns 0. - template T ConsumeProbability() { - static_assert(std::is_floating_point::value, - "A floating point type is required."); + template TS ConvertUnsignedToSigned(TU value); - // Use different integral types for different floating point types in order - // to provide better density of the resulting values. - using IntegralType = - typename std::conditional<(sizeof(T) <= sizeof(uint32_t)), uint32_t, - uint64_t>::type; + const uint8_t *data_ptr_; + size_t remaining_bytes_; +}; - T result = static_cast(ConsumeIntegral()); - result /= static_cast(std::numeric_limits::max()); - return result; +// Returns a std::vector containing |num_bytes| of input data. If fewer than +// |num_bytes| of data remain, returns a shorter std::vector containing all +// of the data that's left. Can be used with any byte sized type, such as +// char, unsigned char, uint8_t, etc. +template +std::vector FuzzedDataProvider::ConsumeBytes(size_t num_bytes) { + num_bytes = std::min(num_bytes, remaining_bytes_); + return ConsumeBytes(num_bytes, num_bytes); +} + +// Similar to |ConsumeBytes|, but also appends the terminator value at the end +// of the resulting vector. Useful, when a mutable null-terminated C-string is +// needed, for example. But that is a rare case. Better avoid it, if possible, +// and prefer using |ConsumeBytes| or |ConsumeBytesAsString| methods. +template +std::vector FuzzedDataProvider::ConsumeBytesWithTerminator(size_t num_bytes, + T terminator) { + num_bytes = std::min(num_bytes, remaining_bytes_); + std::vector result = ConsumeBytes(num_bytes + 1, num_bytes); + result.back() = terminator; + return result; +} + +// Returns a std::vector containing all remaining bytes of the input data. +template +std::vector FuzzedDataProvider::ConsumeRemainingBytes() { + return ConsumeBytes(remaining_bytes_); +} + +// Returns a std::string containing |num_bytes| of input data. Using this and +// |.c_str()| on the resulting string is the best way to get an immutable +// null-terminated C string. If fewer than |num_bytes| of data remain, returns +// a shorter std::string containing all of the data that's left. +inline std::string FuzzedDataProvider::ConsumeBytesAsString(size_t num_bytes) { + static_assert(sizeof(std::string::value_type) == sizeof(uint8_t), + "ConsumeBytesAsString cannot convert the data to a string."); + + num_bytes = std::min(num_bytes, remaining_bytes_); + std::string result( + reinterpret_cast(data_ptr_), num_bytes); + Advance(num_bytes); + return result; +} + +// Returns a std::string of length from 0 to |max_length|. When it runs out of +// input data, returns what remains of the input. Designed to be more stable +// with respect to a fuzzer inserting characters than just picking a random +// length and then consuming that many bytes with |ConsumeBytes|. +inline std::string +FuzzedDataProvider::ConsumeRandomLengthString(size_t max_length) { + // Reads bytes from the start of |data_ptr_|. Maps "\\" to "\", and maps "\" + // followed by anything else to the end of the string. As a result of this + // logic, a fuzzer can insert characters into the string, and the string + // will be lengthened to include those new characters, resulting in a more + // stable fuzzer than picking the length of a string independently from + // picking its contents. + std::string result; + + // Reserve the anticipated capaticity to prevent several reallocations. + result.reserve(std::min(max_length, remaining_bytes_)); + for (size_t i = 0; i < max_length && remaining_bytes_ != 0; ++i) { + char next = ConvertUnsignedToSigned(data_ptr_[0]); + Advance(1); + if (next == '\\' && remaining_bytes_ != 0) { + next = ConvertUnsignedToSigned(data_ptr_[0]); + Advance(1); + if (next != '\\') + break; + } + result += next; } - // Returns a floating point value in the range [Type's lowest, Type's max] by - // consuming bytes from the input data. If there's no input data left, always - // returns approximately 0. - template T ConsumeFloatingPoint() { - return ConsumeFloatingPointInRange(std::numeric_limits::lowest(), - std::numeric_limits::max()); + result.shrink_to_fit(); + return result; +} + +// Returns a std::string of length from 0 to |remaining_bytes_|. +inline std::string FuzzedDataProvider::ConsumeRandomLengthString() { + return ConsumeRandomLengthString(remaining_bytes_); +} + +// Returns a std::string containing all remaining bytes of the input data. +// Prefer using |ConsumeRemainingBytes| unless you actually need a std::string +// object. +inline std::string FuzzedDataProvider::ConsumeRemainingBytesAsString() { + return ConsumeBytesAsString(remaining_bytes_); +} + +// Returns a number in the range [Type's min, Type's max]. The value might +// not be uniformly distributed in the given range. If there's no input data +// left, always returns |min|. +template T FuzzedDataProvider::ConsumeIntegral() { + return ConsumeIntegralInRange(std::numeric_limits::min(), + std::numeric_limits::max()); +} + +// Returns a number in the range [min, max] by consuming bytes from the +// input data. The value might not be uniformly distributed in the given +// range. If there's no input data left, always returns |min|. |min| must +// be less than or equal to |max|. +template +T FuzzedDataProvider::ConsumeIntegralInRange(T min, T max) { + static_assert(std::is_integral::value, "An integral type is required."); + static_assert(sizeof(T) <= sizeof(uint64_t), "Unsupported integral type."); + + if (min > max) + abort(); + + // Use the biggest type possible to hold the range and the result. + uint64_t range = static_cast(max) - min; + uint64_t result = 0; + size_t offset = 0; + + while (offset < sizeof(T) * CHAR_BIT && (range >> offset) > 0 && + remaining_bytes_ != 0) { + // Pull bytes off the end of the seed data. Experimentally, this seems to + // allow the fuzzer to more easily explore the input space. This makes + // sense, since it works by modifying inputs that caused new code to run, + // and this data is often used to encode length of data read by + // |ConsumeBytes|. Separating out read lengths makes it easier modify the + // contents of the data that is actually read. + --remaining_bytes_; + result = (result << CHAR_BIT) | data_ptr_[remaining_bytes_]; + offset += CHAR_BIT; } - // Returns a floating point value in the given range by consuming bytes from - // the input data. If there's no input data left, returns |min|. Note that - // |min| must be less than or equal to |max|. - template T ConsumeFloatingPointInRange(T min, T max) { - if (min > max) - abort(); - - T range = .0; - T result = min; - constexpr T zero(.0); - if (max > zero && min < zero && max > min + std::numeric_limits::max()) { - // The diff |max - min| would overflow the given floating point type. Use - // the half of the diff as the range and consume a bool to decide whether - // the result is in the first of the second part of the diff. - range = (max / 2.0) - (min / 2.0); - if (ConsumeBool()) { - result += range; - } - } else { - range = max - min; + // Avoid division by 0, in case |range + 1| results in overflow. + if (range != std::numeric_limits::max()) + result = result % (range + 1); + + return static_cast(min + result); +} + +// Returns a floating point value in the range [Type's lowest, Type's max] by +// consuming bytes from the input data. If there's no input data left, always +// returns approximately 0. +template T FuzzedDataProvider::ConsumeFloatingPoint() { + return ConsumeFloatingPointInRange(std::numeric_limits::lowest(), + std::numeric_limits::max()); +} + +// Returns a floating point value in the given range by consuming bytes from +// the input data. If there's no input data left, returns |min|. Note that +// |min| must be less than or equal to |max|. +template +T FuzzedDataProvider::ConsumeFloatingPointInRange(T min, T max) { + if (min > max) + abort(); + + T range = .0; + T result = min; + constexpr T zero(.0); + if (max > zero && min < zero && max > min + std::numeric_limits::max()) { + // The diff |max - min| would overflow the given floating point type. Use + // the half of the diff as the range and consume a bool to decide whether + // the result is in the first of the second part of the diff. + range = (max / 2.0) - (min / 2.0); + if (ConsumeBool()) { + result += range; } - - return result + range * ConsumeProbability(); + } else { + range = max - min; } - // Reports the remaining bytes available for fuzzed input. - size_t remaining_bytes() { return remaining_bytes_; } - - private: - FuzzedDataProvider(const FuzzedDataProvider &) = delete; - FuzzedDataProvider &operator=(const FuzzedDataProvider &) = delete; - - void Advance(size_t num_bytes) { - if (num_bytes > remaining_bytes_) + return result + range * ConsumeProbability(); +} + +// Returns a floating point number in the range [0.0, 1.0]. If there's no +// input data left, always returns 0. +template T FuzzedDataProvider::ConsumeProbability() { + static_assert(std::is_floating_point::value, + "A floating point type is required."); + + // Use different integral types for different floating point types in order + // to provide better density of the resulting values. + using IntegralType = + typename std::conditional<(sizeof(T) <= sizeof(uint32_t)), uint32_t, + uint64_t>::type; + + T result = static_cast(ConsumeIntegral()); + result /= static_cast(std::numeric_limits::max()); + return result; +} + +// Reads one byte and returns a bool, or false when no data remains. +inline bool FuzzedDataProvider::ConsumeBool() { + return 1 & ConsumeIntegral(); +} + +// Returns an enum value. The enum must start at 0 and be contiguous. It must +// also contain |kMaxValue| aliased to its largest (inclusive) value. Such as: +// enum class Foo { SomeValue, OtherValue, kMaxValue = OtherValue }; +template T FuzzedDataProvider::ConsumeEnum() { + static_assert(std::is_enum::value, "|T| must be an enum type."); + return static_cast( + ConsumeIntegralInRange(0, static_cast(T::kMaxValue))); +} + +// Returns a copy of the value selected from the given fixed-size |array|. +template +T FuzzedDataProvider::PickValueInArray(const T (&array)[size]) { + static_assert(size > 0, "The array must be non empty."); + return array[ConsumeIntegralInRange(0, size - 1)]; +} + +template +T FuzzedDataProvider::PickValueInArray(std::initializer_list list) { + // TODO(Dor1s): switch to static_assert once C++14 is allowed. + if (!list.size()) + abort(); + + return *(list.begin() + ConsumeIntegralInRange(0, list.size() - 1)); +} + +// Writes |num_bytes| of input data to the given destination pointer. If there +// is not enough data left, writes all remaining bytes. Return value is the +// number of bytes written. +// In general, it's better to avoid using this function, but it may be useful +// in cases when it's necessary to fill a certain buffer or object with +// fuzzing data. +inline size_t FuzzedDataProvider::ConsumeData(void *destination, + size_t num_bytes) { + num_bytes = std::min(num_bytes, remaining_bytes_); + CopyAndAdvance(destination, num_bytes); + return num_bytes; +} + +// Private methods. +inline void FuzzedDataProvider::CopyAndAdvance(void *destination, + size_t num_bytes) { + std::memcpy(destination, data_ptr_, num_bytes); + Advance(num_bytes); +} + +inline void FuzzedDataProvider::Advance(size_t num_bytes) { + if (num_bytes > remaining_bytes_) + abort(); + + data_ptr_ += num_bytes; + remaining_bytes_ -= num_bytes; +} + +template +std::vector FuzzedDataProvider::ConsumeBytes(size_t size, size_t num_bytes) { + static_assert(sizeof(T) == sizeof(uint8_t), "Incompatible data type."); + + // The point of using the size-based constructor below is to increase the + // odds of having a vector object with capacity being equal to the length. + // That part is always implementation specific, but at least both libc++ and + // libstdc++ allocate the requested number of bytes in that constructor, + // which seems to be a natural choice for other implementations as well. + // To increase the odds even more, we also call |shrink_to_fit| below. + std::vector result(size); + if (size == 0) { + if (num_bytes != 0) abort(); - - data_ptr_ += num_bytes; - remaining_bytes_ -= num_bytes; - } - - template - std::vector ConsumeBytes(size_t size, size_t num_bytes_to_consume) { - static_assert(sizeof(T) == sizeof(uint8_t), "Incompatible data type."); - - // The point of using the size-based constructor below is to increase the - // odds of having a vector object with capacity being equal to the length. - // That part is always implementation specific, but at least both libc++ and - // libstdc++ allocate the requested number of bytes in that constructor, - // which seems to be a natural choice for other implementations as well. - // To increase the odds even more, we also call |shrink_to_fit| below. - std::vector result(size); - if (size == 0) { - if (num_bytes_to_consume != 0) - abort(); - return result; - } - - std::memcpy(result.data(), data_ptr_, num_bytes_to_consume); - Advance(num_bytes_to_consume); - - // Even though |shrink_to_fit| is also implementation specific, we expect it - // to provide an additional assurance in case vector's constructor allocated - // a buffer which is larger than the actual amount of data we put inside it. - result.shrink_to_fit(); return result; } - template TS ConvertUnsignedToSigned(TU value) { - static_assert(sizeof(TS) == sizeof(TU), "Incompatible data types."); - static_assert(!std::numeric_limits::is_signed, - "Source type must be unsigned."); - - // TODO(Dor1s): change to `if constexpr` once C++17 becomes mainstream. - if (std::numeric_limits::is_modulo) - return static_cast(value); - - // Avoid using implementation-defined unsigned to signer conversions. - // To learn more, see https://stackoverflow.com/questions/13150449. - if (value <= std::numeric_limits::max()) { - return static_cast(value); - } else { - constexpr auto TS_min = std::numeric_limits::min(); - return TS_min + static_cast(value - TS_min); - } + CopyAndAdvance(result.data(), num_bytes); + + // Even though |shrink_to_fit| is also implementation specific, we expect it + // to provide an additional assurance in case vector's constructor allocated + // a buffer which is larger than the actual amount of data we put inside it. + result.shrink_to_fit(); + return result; +} + +template +TS FuzzedDataProvider::ConvertUnsignedToSigned(TU value) { + static_assert(sizeof(TS) == sizeof(TU), "Incompatible data types."); + static_assert(!std::numeric_limits::is_signed, + "Source type must be unsigned."); + + // TODO(Dor1s): change to `if constexpr` once C++17 becomes mainstream. + if (std::numeric_limits::is_modulo) + return static_cast(value); + + // Avoid using implementation-defined unsigned to signed conversions. + // To learn more, see https://stackoverflow.com/questions/13150449. + if (value <= std::numeric_limits::max()) { + return static_cast(value); + } else { + constexpr auto TS_min = std::numeric_limits::min(); + return TS_min + static_cast(value - TS_min); } - - const uint8_t *data_ptr_; - size_t remaining_bytes_; -}; +} #endif // LLVM_FUZZER_FUZZED_DATA_PROVIDER_H_