| 1 | //===-- DataExtractor.cpp -------------------------------------------------===// |
| 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 | #include "llvm/Support/DataExtractor.h" |
| 10 | #include "llvm/ADT/StringExtras.h" |
| 11 | #include "llvm/Support/Errc.h" |
| 12 | #include "llvm/Support/ErrorHandling.h" |
| 13 | #include "llvm/Support/LEB128.h" |
| 14 | #include "llvm/Support/SwapByteOrder.h" |
| 15 | |
| 16 | using namespace llvm; |
| 17 | |
| 18 | bool DataExtractor::(uint64_t Offset, uint64_t Size, |
| 19 | Error *E) const { |
| 20 | if (isValidOffsetForDataOfSize(offset: Offset, length: Size)) |
| 21 | return true; |
| 22 | if (E) { |
| 23 | if (Offset <= Data.size()) |
| 24 | *E = createStringError( |
| 25 | EC: errc::illegal_byte_sequence, |
| 26 | Fmt: "unexpected end of data at offset 0x%zx while reading [0x%" PRIx64 |
| 27 | ", 0x%" PRIx64 ")" , |
| 28 | Vals: Data.size(), Vals: Offset, Vals: Offset + Size); |
| 29 | else |
| 30 | *E = createStringError(EC: errc::invalid_argument, |
| 31 | Fmt: "offset 0x%" PRIx64 |
| 32 | " is beyond the end of data at 0x%zx" , |
| 33 | Vals: Offset, Vals: Data.size()); |
| 34 | } |
| 35 | return false; |
| 36 | } |
| 37 | |
| 38 | static bool isError(Error *E) { return E && *E; } |
| 39 | |
| 40 | template <typename T> |
| 41 | T DataExtractor::(uint64_t *offset_ptr, Error *Err) const { |
| 42 | ErrorAsOutParameter ErrAsOut(Err); |
| 43 | T val = 0; |
| 44 | if (isError(E: Err)) |
| 45 | return val; |
| 46 | |
| 47 | uint64_t offset = *offset_ptr; |
| 48 | if (!prepareRead(Offset: offset, Size: sizeof(T), E: Err)) |
| 49 | return val; |
| 50 | std::memcpy(dest: &val, src: &Data.data()[offset], n: sizeof(val)); |
| 51 | if (sys::IsLittleEndianHost != IsLittleEndian) |
| 52 | sys::swapByteOrder(val); |
| 53 | |
| 54 | // Advance the offset |
| 55 | *offset_ptr += sizeof(val); |
| 56 | return val; |
| 57 | } |
| 58 | |
| 59 | template <typename T> |
| 60 | T *DataExtractor::(uint64_t *offset_ptr, T *dst, uint32_t count, |
| 61 | Error *Err) const { |
| 62 | ErrorAsOutParameter ErrAsOut(Err); |
| 63 | if (isError(E: Err)) |
| 64 | return nullptr; |
| 65 | |
| 66 | uint64_t offset = *offset_ptr; |
| 67 | |
| 68 | if (!prepareRead(Offset: offset, Size: sizeof(*dst) * count, E: Err)) |
| 69 | return nullptr; |
| 70 | for (T *value_ptr = dst, *end = dst + count; value_ptr != end; |
| 71 | ++value_ptr, offset += sizeof(*dst)) |
| 72 | *value_ptr = getU<T>(offset_ptr, Err); |
| 73 | // Advance the offset |
| 74 | *offset_ptr = offset; |
| 75 | // Return a non-NULL pointer to the converted data as an indicator of |
| 76 | // success |
| 77 | return dst; |
| 78 | } |
| 79 | |
| 80 | uint8_t DataExtractor::(uint64_t *offset_ptr, llvm::Error *Err) const { |
| 81 | return getU<uint8_t>(offset_ptr, Err); |
| 82 | } |
| 83 | |
| 84 | uint8_t *DataExtractor::(uint64_t *offset_ptr, uint8_t *dst, |
| 85 | uint32_t count) const { |
| 86 | return getUs<uint8_t>(offset_ptr, dst, count, Err: nullptr); |
| 87 | } |
| 88 | |
| 89 | uint8_t *DataExtractor::(Cursor &C, uint8_t *Dst, uint32_t Count) const { |
| 90 | return getUs<uint8_t>(offset_ptr: &C.Offset, dst: Dst, count: Count, Err: &C.Err); |
| 91 | } |
| 92 | |
| 93 | uint16_t DataExtractor::(uint64_t *offset_ptr, llvm::Error *Err) const { |
| 94 | return getU<uint16_t>(offset_ptr, Err); |
| 95 | } |
| 96 | |
| 97 | uint16_t *DataExtractor::(uint64_t *offset_ptr, uint16_t *dst, |
| 98 | uint32_t count) const { |
| 99 | return getUs<uint16_t>(offset_ptr, dst, count, Err: nullptr); |
| 100 | } |
| 101 | |
| 102 | uint32_t DataExtractor::(uint64_t *OffsetPtr, Error *Err) const { |
| 103 | uint24_t = getU<uint24_t>(offset_ptr: OffsetPtr, Err); |
| 104 | // The 3 bytes are in the correct byte order for the host. |
| 105 | return ExtractedVal.getAsUint32(IsLittleEndian: sys::IsLittleEndianHost); |
| 106 | } |
| 107 | |
| 108 | uint32_t DataExtractor::(uint64_t *offset_ptr, llvm::Error *Err) const { |
| 109 | return getU<uint32_t>(offset_ptr, Err); |
| 110 | } |
| 111 | |
| 112 | uint32_t *DataExtractor::(uint64_t *offset_ptr, uint32_t *dst, |
| 113 | uint32_t count) const { |
| 114 | return getUs<uint32_t>(offset_ptr, dst, count, Err: nullptr); |
| 115 | } |
| 116 | |
| 117 | uint64_t DataExtractor::(uint64_t *offset_ptr, llvm::Error *Err) const { |
| 118 | return getU<uint64_t>(offset_ptr, Err); |
| 119 | } |
| 120 | |
| 121 | uint64_t *DataExtractor::(uint64_t *offset_ptr, uint64_t *dst, |
| 122 | uint32_t count) const { |
| 123 | return getUs<uint64_t>(offset_ptr, dst, count, Err: nullptr); |
| 124 | } |
| 125 | |
| 126 | uint64_t DataExtractor::(uint64_t *offset_ptr, uint32_t byte_size, |
| 127 | llvm::Error *Err) const { |
| 128 | switch (byte_size) { |
| 129 | case 1: |
| 130 | return getU8(offset_ptr, Err); |
| 131 | case 2: |
| 132 | return getU16(offset_ptr, Err); |
| 133 | case 4: |
| 134 | return getU32(offset_ptr, Err); |
| 135 | case 8: |
| 136 | return getU64(offset_ptr, Err); |
| 137 | } |
| 138 | llvm_unreachable("getUnsigned unhandled case!" ); |
| 139 | } |
| 140 | |
| 141 | int64_t |
| 142 | DataExtractor::(uint64_t *offset_ptr, uint32_t byte_size) const { |
| 143 | switch (byte_size) { |
| 144 | case 1: |
| 145 | return (int8_t)getU8(offset_ptr); |
| 146 | case 2: |
| 147 | return (int16_t)getU16(offset_ptr); |
| 148 | case 4: |
| 149 | return (int32_t)getU32(offset_ptr); |
| 150 | case 8: |
| 151 | return (int64_t)getU64(offset_ptr); |
| 152 | } |
| 153 | llvm_unreachable("getSigned unhandled case!" ); |
| 154 | } |
| 155 | |
| 156 | StringRef DataExtractor::(uint64_t *OffsetPtr, Error *Err) const { |
| 157 | ErrorAsOutParameter ErrAsOut(Err); |
| 158 | if (isError(E: Err)) |
| 159 | return StringRef(); |
| 160 | |
| 161 | uint64_t Start = *OffsetPtr; |
| 162 | StringRef::size_type Pos = Data.find(C: '\0', From: Start); |
| 163 | if (Pos != StringRef::npos) { |
| 164 | *OffsetPtr = Pos + 1; |
| 165 | return StringRef(Data.data() + Start, Pos - Start); |
| 166 | } |
| 167 | if (Err) |
| 168 | *Err = createStringError(EC: errc::illegal_byte_sequence, |
| 169 | Fmt: "no null terminated string at offset 0x%" PRIx64, |
| 170 | Vals: Start); |
| 171 | return StringRef(); |
| 172 | } |
| 173 | |
| 174 | StringRef DataExtractor::(uint64_t *OffsetPtr, |
| 175 | uint64_t Length, |
| 176 | StringRef TrimChars) const { |
| 177 | StringRef Bytes(getBytes(OffsetPtr, Length)); |
| 178 | return Bytes.trim(Chars: TrimChars); |
| 179 | } |
| 180 | |
| 181 | StringRef DataExtractor::(uint64_t *OffsetPtr, uint64_t Length, |
| 182 | Error *Err) const { |
| 183 | ErrorAsOutParameter ErrAsOut(Err); |
| 184 | if (isError(E: Err)) |
| 185 | return StringRef(); |
| 186 | |
| 187 | if (!prepareRead(Offset: *OffsetPtr, Size: Length, E: Err)) |
| 188 | return StringRef(); |
| 189 | |
| 190 | StringRef Result = Data.substr(Start: *OffsetPtr, N: Length); |
| 191 | *OffsetPtr += Length; |
| 192 | return Result; |
| 193 | } |
| 194 | |
| 195 | template <typename T> |
| 196 | static T getLEB128(StringRef Data, uint64_t *OffsetPtr, Error *Err, |
| 197 | T (&Decoder)(const uint8_t *p, unsigned *n, |
| 198 | const uint8_t *end, const char **error)) { |
| 199 | ArrayRef<uint8_t> Bytes = arrayRefFromStringRef(Input: Data); |
| 200 | assert(*OffsetPtr <= Bytes.size()); |
| 201 | ErrorAsOutParameter ErrAsOut(Err); |
| 202 | if (isError(E: Err)) |
| 203 | return T(); |
| 204 | |
| 205 | const char *error = nullptr; |
| 206 | unsigned bytes_read; |
| 207 | T result = |
| 208 | Decoder(Bytes.data() + *OffsetPtr, &bytes_read, Bytes.end(), &error); |
| 209 | if (error) { |
| 210 | if (Err) |
| 211 | *Err = createStringError(EC: errc::illegal_byte_sequence, |
| 212 | Fmt: "unable to decode LEB128 at offset 0x%8.8" PRIx64 |
| 213 | ": %s" , |
| 214 | Vals: *OffsetPtr, Vals: error); |
| 215 | return T(); |
| 216 | } |
| 217 | *OffsetPtr += bytes_read; |
| 218 | return result; |
| 219 | } |
| 220 | |
| 221 | uint64_t DataExtractor::(uint64_t *offset_ptr, Error *Err) const { |
| 222 | return getLEB128(Data, OffsetPtr: offset_ptr, Err, Decoder&: decodeULEB128); |
| 223 | } |
| 224 | |
| 225 | int64_t DataExtractor::(uint64_t *offset_ptr, Error *Err) const { |
| 226 | return getLEB128(Data, OffsetPtr: offset_ptr, Err, Decoder&: decodeSLEB128); |
| 227 | } |
| 228 | |
| 229 | void DataExtractor::(Cursor &C, uint64_t Length) const { |
| 230 | ErrorAsOutParameter ErrAsOut(C.Err); |
| 231 | if (isError(E: &C.Err)) |
| 232 | return; |
| 233 | |
| 234 | if (prepareRead(Offset: C.Offset, Size: Length, E: &C.Err)) |
| 235 | C.Offset += Length; |
| 236 | } |
| 237 | |