Merge Chromium + Blink git repositories
[chromium-blink-merge.git] / third_party / WebKit / Source / wtf / text / ASCIIFastPath.h
blob392966e507e4e3e05f8f873bf804a03061ad5da4
1 /*
2 * Copyright (C) 2011, 2012 Apple Inc. All rights reserved.
3 * Copyright (C) 2011 Nokia Corporation and/or its subsidiary(-ies).
5 * This library is free software; you can redistribute it and/or
6 * modify it under the terms of the GNU Library General Public
7 * License as published by the Free Software Foundation; either
8 * version 2 of the License, or (at your option) any later version.
10 * This library is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13 * Library General Public License for more details.
15 * You should have received a copy of the GNU Library General Public License
16 * along with this library; see the file COPYING.LIB. If not, write to
17 * the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
18 * Boston, MA 02110-1301, USA.
22 #ifndef ASCIIFastPath_h
23 #define ASCIIFastPath_h
25 #include "wtf/Alignment.h"
26 #include "wtf/CPU.h"
27 #include "wtf/StdLibExtras.h"
28 #include "wtf/text/Unicode.h"
29 #include <stdint.h>
31 #if OS(MACOSX) && (CPU(X86) || CPU(X86_64))
32 #include <emmintrin.h>
33 #endif
35 namespace WTF {
37 // Assuming that a pointer is the size of a "machine word", then
38 // uintptr_t is an integer type that is also a machine word.
39 typedef uintptr_t MachineWord;
40 const uintptr_t machineWordAlignmentMask = sizeof(MachineWord) - 1;
42 inline bool isAlignedToMachineWord(const void* pointer)
44 return !(reinterpret_cast<uintptr_t>(pointer) & machineWordAlignmentMask);
47 template<typename T> inline T* alignToMachineWord(T* pointer)
49 return reinterpret_cast<T*>(reinterpret_cast<uintptr_t>(pointer) & ~machineWordAlignmentMask);
52 template<size_t size, typename CharacterType> struct NonASCIIMask;
53 template<> struct NonASCIIMask<4, UChar> {
54 static inline uint32_t value() { return 0xFF80FF80U; }
56 template<> struct NonASCIIMask<4, LChar> {
57 static inline uint32_t value() { return 0x80808080U; }
59 template<> struct NonASCIIMask<8, UChar> {
60 static inline uint64_t value() { return 0xFF80FF80FF80FF80ULL; }
62 template<> struct NonASCIIMask<8, LChar> {
63 static inline uint64_t value() { return 0x8080808080808080ULL; }
67 template<typename CharacterType>
68 inline bool isAllASCII(MachineWord word)
70 return !(word & NonASCIIMask<sizeof(MachineWord), CharacterType>::value());
73 // Note: This function assume the input is likely all ASCII, and
74 // does not leave early if it is not the case.
75 template<typename CharacterType>
76 inline bool charactersAreAllASCII(const CharacterType* characters, size_t length)
78 MachineWord allCharBits = 0;
79 const CharacterType* end = characters + length;
81 // Prologue: align the input.
82 while (!isAlignedToMachineWord(characters) && characters != end) {
83 allCharBits |= *characters;
84 ++characters;
87 // Compare the values of CPU word size.
88 const CharacterType* wordEnd = alignToMachineWord(end);
89 const size_t loopIncrement = sizeof(MachineWord) / sizeof(CharacterType);
90 while (characters < wordEnd) {
91 allCharBits |= *(reinterpret_cast_ptr<const MachineWord*>(characters));
92 characters += loopIncrement;
95 // Process the remaining bytes.
96 while (characters != end) {
97 allCharBits |= *characters;
98 ++characters;
101 MachineWord nonASCIIBitMask = NonASCIIMask<sizeof(MachineWord), CharacterType>::value();
102 return !(allCharBits & nonASCIIBitMask);
105 inline void copyLCharsFromUCharSource(LChar* destination, const UChar* source, size_t length)
107 #if OS(MACOSX) && (CPU(X86) || CPU(X86_64))
108 const uintptr_t memoryAccessSize = 16; // Memory accesses on 16 byte (128 bit) alignment
109 const uintptr_t memoryAccessMask = memoryAccessSize - 1;
111 size_t i = 0;
112 for (;i < length && !isAlignedTo<memoryAccessMask>(&source[i]); ++i) {
113 ASSERT(!(source[i] & 0xff00));
114 destination[i] = static_cast<LChar>(source[i]);
117 const uintptr_t sourceLoadSize = 32; // Process 32 bytes (16 UChars) each iteration
118 const size_t ucharsPerLoop = sourceLoadSize / sizeof(UChar);
119 if (length > ucharsPerLoop) {
120 const size_t endLength = length - ucharsPerLoop + 1;
121 for (; i < endLength; i += ucharsPerLoop) {
122 #if ENABLE(ASSERT)
123 for (unsigned checkIndex = 0; checkIndex < ucharsPerLoop; ++checkIndex)
124 ASSERT(!(source[i+checkIndex] & 0xff00));
125 #endif
126 __m128i first8UChars = _mm_load_si128(reinterpret_cast<const __m128i*>(&source[i]));
127 __m128i second8UChars = _mm_load_si128(reinterpret_cast<const __m128i*>(&source[i+8]));
128 __m128i packedChars = _mm_packus_epi16(first8UChars, second8UChars);
129 _mm_storeu_si128(reinterpret_cast<__m128i*>(&destination[i]), packedChars);
133 for (; i < length; ++i) {
134 ASSERT(!(source[i] & 0xff00));
135 destination[i] = static_cast<LChar>(source[i]);
137 #elif COMPILER(GCC) && CPU(ARM_NEON) && !(CPU(BIG_ENDIAN) || CPU(MIDDLE_ENDIAN)) && defined(NDEBUG)
138 const LChar* const end = destination + length;
139 const uintptr_t memoryAccessSize = 8;
141 if (length >= (2 * memoryAccessSize) - 1) {
142 // Prefix: align dst on 64 bits.
143 const uintptr_t memoryAccessMask = memoryAccessSize - 1;
144 while (!isAlignedTo<memoryAccessMask>(destination))
145 *destination++ = static_cast<LChar>(*source++);
147 // Vector interleaved unpack, we only store the lower 8 bits.
148 const uintptr_t lengthLeft = end - destination;
149 const LChar* const simdEnd = end - (lengthLeft % memoryAccessSize);
150 do {
151 asm("vld2.8 { d0-d1 }, [%[SOURCE]] !\n\t"
152 "vst1.8 { d0 }, [%[DESTINATION],:64] !\n\t"
153 : [SOURCE]"+r" (source), [DESTINATION]"+r" (destination)
155 : "memory", "d0", "d1");
156 } while (destination != simdEnd);
159 while (destination != end)
160 *destination++ = static_cast<LChar>(*source++);
161 #else
162 for (size_t i = 0; i < length; ++i) {
163 ASSERT(!(source[i] & 0xff00));
164 destination[i] = static_cast<LChar>(source[i]);
166 #endif
169 } // namespace WTF
171 #endif // ASCIIFastPath_h