%PDF- <> %âãÏÓ endobj 2 0 obj <> endobj 3 0 obj <>/ExtGState<>/ProcSet[/PDF/Text/ImageB/ImageC/ImageI] >>/Annots[ 28 0 R 29 0 R] /MediaBox[ 0 0 595.5 842.25] /Contents 4 0 R/Group<>/Tabs/S>> endobj ºaâÚÎΞ-ÌE1ÍØÄ÷{òò2ÿ ÛÖ^ÔÀá TÎ{¦?§®¥kuµùÕ5sLOšuY>endobj 2 0 obj<>endobj 2 0 obj<>endobj 2 0 obj<>endobj 2 0 obj<> endobj 2 0 obj<>endobj 2 0 obj<>es 3 0 R>> endobj 2 0 obj<> ox[ 0.000000 0.000000 609.600000 935.600000]/Fi endobj 3 0 obj<> endobj 7 1 obj<>/ProcSet[/PDF/Text/ImageB/ImageC/ImageI]>>/Subtype/Form>> stream
// Copyright 2012 the V8 project authors. All rights reserved. // Use of this source code is governed by a BSD-style license that can be // found in the LICENSE file. // CPU specific code for x64 independent of OS goes here. #if defined(__GNUC__) && !defined(__MINGW64__) #include "src/third_party/valgrind/valgrind.h" #endif #if V8_TARGET_ARCH_X64 #include "src/codegen/cpu-features.h" namespace v8 { namespace internal { void CpuFeatures::FlushICache(void* start, size_t size) { // No need to flush the instruction cache on Intel. On Intel instruction // cache flushing is only necessary when multiple cores running the same // code simultaneously. V8 (and JavaScript) is single threaded and when code // is patched on an intel CPU the core performing the patching will have its // own instruction cache updated automatically. // If flushing of the instruction cache becomes necessary Windows has the // API function FlushInstructionCache. // By default, valgrind only checks the stack for writes that might need to // invalidate already cached translated code. This leads to random // instability when code patches or moves are sometimes unnoticed. One // solution is to run valgrind with --smc-check=all, but this comes at a big // performance cost. We can notify valgrind to invalidate its cache. #ifdef VALGRIND_DISCARD_TRANSLATIONS unsigned res = VALGRIND_DISCARD_TRANSLATIONS(start, size); USE(res); #endif } } // namespace internal } // namespace v8 #endif // V8_TARGET_ARCH_X64