| 1 | /* |
| 2 | * Copyright (C) 2009-2018 Apple Inc. All rights reserved. |
| 3 | * |
| 4 | * Redistribution and use in source and binary forms, with or without |
| 5 | * modification, are permitted provided that the following conditions |
| 6 | * are met: |
| 7 | * 1. Redistributions of source code must retain the above copyright |
| 8 | * notice, this list of conditions and the following disclaimer. |
| 9 | * 2. Redistributions in binary form must reproduce the above copyright |
| 10 | * notice, this list of conditions and the following disclaimer in the |
| 11 | * documentation and/or other materials provided with the distribution. |
| 12 | * |
| 13 | * THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY |
| 14 | * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
| 15 | * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR |
| 16 | * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL APPLE INC. OR |
| 17 | * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, |
| 18 | * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, |
| 19 | * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR |
| 20 | * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY |
| 21 | * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT |
| 22 | * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE |
| 23 | * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
| 24 | */ |
| 25 | |
| 26 | #pragma once |
| 27 | |
| 28 | #include <Platform.h> |
| 29 | |
| 30 | #if ENABLE(YARR_JIT) |
| 31 | |
| 32 | #include "MacroAssemblerCodeRef.h" |
| 33 | #include "MatchResult.h" |
| 34 | #include "Yarr.h" |
| 35 | #include "YarrPattern.h" |
| 36 | |
| 37 | #if CPU(X86) && !COMPILER(MSVC) |
| 38 | #define YARR_CALL __attribute__ ((regparm (3))) |
| 39 | #else |
| 40 | #define YARR_CALL |
| 41 | #endif |
| 42 | |
| 43 | #if ENABLE(YARR_JIT_ALL_PARENS_EXPRESSIONS) |
| 44 | constexpr size_t patternContextBufferSize = 8192; // Space caller allocates to save nested parenthesis context |
| 45 | #endif |
| 46 | |
| 47 | namespace JSC { |
| 48 | |
| 49 | class VM; |
| 50 | class ExecutablePool; |
| 51 | |
| 52 | namespace Yarr { |
| 53 | |
| 54 | enum class JITFailureReason : uint8_t { |
| 55 | DecodeSurrogatePair, |
| 56 | BackReference, |
| 57 | ForwardReference, |
| 58 | VariableCountedParenthesisWithNonZeroMinimum, |
| 59 | ParenthesizedSubpattern, |
| 60 | FixedCountParenthesizedSubpattern, |
| 61 | ExecutableMemoryAllocationFailure, |
| 62 | }; |
| 63 | |
| 64 | class YarrCodeBlock { |
| 65 | #if CPU(X86_64) || CPU(ARM64) |
| 66 | #if ENABLE(YARR_JIT_ALL_PARENS_EXPRESSIONS) |
| 67 | typedef MatchResult (*YarrJITCode8)(const LChar* input, unsigned start, unsigned length, int* output, void* freeParenContext, unsigned parenContextSize) YARR_CALL; |
| 68 | typedef MatchResult (*YarrJITCode16)(const UChar* input, unsigned start, unsigned length, int* output, void* freeParenContext, unsigned parenContextSize) YARR_CALL; |
| 69 | typedef MatchResult (*YarrJITCodeMatchOnly8)(const LChar* input, unsigned start, unsigned length, void*, void* freeParenContext, unsigned parenContextSize) YARR_CALL; |
| 70 | typedef MatchResult (*YarrJITCodeMatchOnly16)(const UChar* input, unsigned start, unsigned length, void*, void* freeParenContext, unsigned parenContextSize) YARR_CALL; |
| 71 | #else |
| 72 | typedef MatchResult (*YarrJITCode8)(const LChar* input, unsigned start, unsigned length, int* output) YARR_CALL; |
| 73 | typedef MatchResult (*YarrJITCode16)(const UChar* input, unsigned start, unsigned length, int* output) YARR_CALL; |
| 74 | typedef MatchResult (*YarrJITCodeMatchOnly8)(const LChar* input, unsigned start, unsigned length) YARR_CALL; |
| 75 | typedef MatchResult (*YarrJITCodeMatchOnly16)(const UChar* input, unsigned start, unsigned length) YARR_CALL; |
| 76 | #endif |
| 77 | #else |
| 78 | typedef EncodedMatchResult (*YarrJITCode8)(const LChar* input, unsigned start, unsigned length, int* output) YARR_CALL; |
| 79 | typedef EncodedMatchResult (*YarrJITCode16)(const UChar* input, unsigned start, unsigned length, int* output) YARR_CALL; |
| 80 | typedef EncodedMatchResult (*YarrJITCodeMatchOnly8)(const LChar* input, unsigned start, unsigned length) YARR_CALL; |
| 81 | typedef EncodedMatchResult (*YarrJITCodeMatchOnly16)(const UChar* input, unsigned start, unsigned length) YARR_CALL; |
| 82 | #endif |
| 83 | |
| 84 | public: |
| 85 | YarrCodeBlock() = default; |
| 86 | ~YarrCodeBlock() { clear(); } |
| 87 | |
| 88 | static void replaceCodeRef(MacroAssemblerCodeRef &target, const MacroAssemblerCodeRef &source); |
| 89 | |
| 90 | void setFallBackWithFailureReason(JITFailureReason failureReason) { m_failureReason = failureReason; } |
| 91 | std::optional<JITFailureReason> failureReason() { return m_failureReason; } |
| 92 | |
| 93 | bool has8BitCode() { return m_ref8.size(); } |
| 94 | bool has16BitCode() { return m_ref16.size(); } |
| 95 | void set8BitCode(MacroAssemblerCodeRef ref) { replaceCodeRef(target&: m_ref8, source: ref); } |
| 96 | void set16BitCode(MacroAssemblerCodeRef ref) { replaceCodeRef(target&: m_ref16, source: ref); } |
| 97 | |
| 98 | bool has8BitCodeMatchOnly() { return m_matchOnly8.size(); } |
| 99 | bool has16BitCodeMatchOnly() { return m_matchOnly16.size(); } |
| 100 | void set8BitCodeMatchOnly(MacroAssemblerCodeRef matchOnly) |
| 101 | { |
| 102 | replaceCodeRef(target&: m_matchOnly8, source: matchOnly); |
| 103 | } |
| 104 | void set16BitCodeMatchOnly(MacroAssemblerCodeRef matchOnly) |
| 105 | { |
| 106 | replaceCodeRef(target&: m_matchOnly16, source: matchOnly); |
| 107 | } |
| 108 | |
| 109 | #if ENABLE(YARR_JIT_ALL_PARENS_EXPRESSIONS) |
| 110 | bool usesPatternContextBuffer() { return m_usesPatternContextBuffer; } |
| 111 | void setUsesPatternContextBuffer() { m_usesPatternContextBuffer = true; } |
| 112 | |
| 113 | MatchResult execute(const LChar* input, unsigned start, unsigned length, int* output, void* freeParenContext, unsigned parenContextSize) |
| 114 | { |
| 115 | ASSERT(has8BitCode()); |
| 116 | return MatchResult(reinterpret_cast<YarrJITCode8>(m_ref8.code().executableAddress())(input, start, length, output, freeParenContext, parenContextSize)); |
| 117 | } |
| 118 | |
| 119 | MatchResult execute(const UChar* input, unsigned start, unsigned length, int* output, void* freeParenContext, unsigned parenContextSize) |
| 120 | { |
| 121 | ASSERT(has16BitCode()); |
| 122 | return MatchResult(reinterpret_cast<YarrJITCode16>(m_ref16.code().executableAddress())(input, start, length, output, freeParenContext, parenContextSize)); |
| 123 | } |
| 124 | |
| 125 | MatchResult execute(const LChar* input, unsigned start, unsigned length, void* freeParenContext, unsigned parenContextSize) |
| 126 | { |
| 127 | ASSERT(has8BitCodeMatchOnly()); |
| 128 | return MatchResult(reinterpret_cast<YarrJITCodeMatchOnly8>(m_matchOnly8.code().executableAddress())(input, start, length, 0, freeParenContext, parenContextSize)); |
| 129 | } |
| 130 | |
| 131 | MatchResult execute(const UChar* input, unsigned start, unsigned length, void* freeParenContext, unsigned parenContextSize) |
| 132 | { |
| 133 | ASSERT(has16BitCodeMatchOnly()); |
| 134 | return MatchResult(reinterpret_cast<YarrJITCodeMatchOnly16>(m_matchOnly16.code().executableAddress())(input, start, length, 0, freeParenContext, parenContextSize)); |
| 135 | } |
| 136 | #else |
| 137 | MatchResult execute(const LChar* input, unsigned start, unsigned length, int* output) |
| 138 | { |
| 139 | ASSERT(has8BitCode()); |
| 140 | return MatchResult(reinterpret_cast<YarrJITCode8>(m_ref8.code().executableAddress())(input, start, length, output)); |
| 141 | } |
| 142 | |
| 143 | MatchResult execute(const UChar* input, unsigned start, unsigned length, int* output) |
| 144 | { |
| 145 | ASSERT(has16BitCode()); |
| 146 | return MatchResult(reinterpret_cast<YarrJITCode16>(m_ref16.code().executableAddress())(input, start, length, output)); |
| 147 | } |
| 148 | |
| 149 | MatchResult execute(const LChar* input, unsigned start, unsigned length) |
| 150 | { |
| 151 | ASSERT(has8BitCodeMatchOnly()); |
| 152 | return MatchResult(reinterpret_cast<YarrJITCodeMatchOnly8>(m_matchOnly8.code().executableAddress())(input, start, length)); |
| 153 | } |
| 154 | |
| 155 | MatchResult execute(const UChar* input, unsigned start, unsigned length) |
| 156 | { |
| 157 | ASSERT(has16BitCodeMatchOnly()); |
| 158 | return MatchResult(reinterpret_cast<YarrJITCodeMatchOnly16>(m_matchOnly16.code().executableAddress())(input, start, length)); |
| 159 | } |
| 160 | #endif |
| 161 | |
| 162 | #if ENABLE(REGEXP_TRACING) |
| 163 | void *get8BitMatchOnlyAddr() |
| 164 | { |
| 165 | if (!has8BitCodeMatchOnly()) |
| 166 | return 0; |
| 167 | |
| 168 | return m_matchOnly8.code().executableAddress(); |
| 169 | } |
| 170 | |
| 171 | void *get16BitMatchOnlyAddr() |
| 172 | { |
| 173 | if (!has16BitCodeMatchOnly()) |
| 174 | return 0; |
| 175 | |
| 176 | return m_matchOnly16.code().executableAddress(); |
| 177 | } |
| 178 | |
| 179 | void *get8BitMatchAddr() |
| 180 | { |
| 181 | if (!has8BitCode()) |
| 182 | return 0; |
| 183 | |
| 184 | return m_ref8.code().executableAddress(); |
| 185 | } |
| 186 | |
| 187 | void *get16BitMatchAddr() |
| 188 | { |
| 189 | if (!has16BitCode()) |
| 190 | return 0; |
| 191 | |
| 192 | return m_ref16.code().executableAddress(); |
| 193 | } |
| 194 | #endif |
| 195 | |
| 196 | size_t size() const |
| 197 | { |
| 198 | return m_ref8.size() + m_ref16.size() + m_matchOnly8.size() + m_matchOnly16.size(); |
| 199 | } |
| 200 | |
| 201 | void clear() |
| 202 | { |
| 203 | replaceCodeRef(target&: m_ref8, source: MacroAssemblerCodeRef()); |
| 204 | replaceCodeRef(target&: m_ref16, source: MacroAssemblerCodeRef()); |
| 205 | replaceCodeRef(target&: m_matchOnly8, source: MacroAssemblerCodeRef()); |
| 206 | replaceCodeRef(target&: m_matchOnly16, source: MacroAssemblerCodeRef()); |
| 207 | m_failureReason = std::nullopt; |
| 208 | } |
| 209 | |
| 210 | private: |
| 211 | MacroAssemblerCodeRef m_ref8; |
| 212 | MacroAssemblerCodeRef m_ref16; |
| 213 | MacroAssemblerCodeRef m_matchOnly8; |
| 214 | MacroAssemblerCodeRef m_matchOnly16; |
| 215 | #if ENABLE(YARR_JIT_ALL_PARENS_EXPRESSIONS) |
| 216 | bool m_usesPatternContextBuffer; |
| 217 | #endif |
| 218 | std::optional<JITFailureReason> m_failureReason; |
| 219 | }; |
| 220 | |
| 221 | enum YarrJITCompileMode { |
| 222 | MatchOnly, |
| 223 | IncludeSubpatterns |
| 224 | }; |
| 225 | void jitCompile(YarrPattern&, YarrCharSize, VM*, YarrCodeBlock& jitObject, YarrJITCompileMode = IncludeSubpatterns); |
| 226 | |
| 227 | } } // namespace JSC::Yarr |
| 228 | |
| 229 | #endif |
| 230 | |