AuroraMiddleware/v8
1
0
Fork 0
Code Issues 2 Pull Requests Releases Wiki Activity
fcaa2c2e57
v8/test/cctest/test-disasm-ia32.cc

864 lines
27 KiB
C++
Raw Normal View History

Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
// Copyright 2011 the V8 project authors. All rights reserved.
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are
// met:
//
// * Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
// * Redistributions in binary form must reproduce the above
// copyright notice, this list of conditions and the following
// disclaimer in the documentation and/or other materials provided
// with the distribution.
// * Neither the name of Google Inc. nor the names of its
// contributors may be used to endorse or promote products derived
// from this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#include <stdlib.h>
Use full include paths everywhere - this avoids using relative include paths which are forbidden by the style guide - makes the code more readable since it's clear which header is meant - allows for starting to use checkdeps BUG=none R=jkummerow@chromium.org, danno@chromium.org LOG=n Review URL: https://codereview.chromium.org/304153016 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21625 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-06-03 08:12:43 +00:00
#include "src/v8.h"
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
Simplify IC interfaces BUG= Review URL: https://codereview.chromium.org/1865873002 Cr-Commit-Position: refs/heads/master@{#35301}
2016-04-06 13:36:49 +00:00
#include "src/code-factory.h"
Debugger: move implementation to a separate folder. R=cbruni@chromium.org Review URL: https://codereview.chromium.org/1265923002 Cr-Commit-Position: refs/heads/master@{#29951}
2015-07-31 11:07:50 +00:00
#include "src/debug/debug.h"
Use full include paths everywhere - this avoids using relative include paths which are forbidden by the style guide - makes the code more readable since it's clear which header is meant - allows for starting to use checkdeps BUG=none R=jkummerow@chromium.org, danno@chromium.org LOG=n Review URL: https://codereview.chromium.org/304153016 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21625 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-06-03 08:12:43 +00:00
#include "src/disasm.h"
#include "src/disassembler.h"
Realize IWYU pattern for frames-inl.h header. R=bmeurer@chromium.org Review URL: https://codereview.chromium.org/1283183002 Cr-Commit-Position: refs/heads/master@{#30127}
2015-08-12 10:28:34 +00:00
#include "src/ia32/frames-ia32.h"
Use full include paths everywhere - this avoids using relative include paths which are forbidden by the style guide - makes the code more readable since it's clear which header is meant - allows for starting to use checkdeps BUG=none R=jkummerow@chromium.org, danno@chromium.org LOG=n Review URL: https://codereview.chromium.org/304153016 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21625 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-06-03 08:12:43 +00:00
#include "src/macro-assembler.h"
#include "test/cctest/cctest.h"
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
using namespace v8::internal;
#define __ assm.
static void DummyStaticFunction(Object* result) {
}
TEST(DisasmIa320) {
Unify the way cctest initalizes the VM for each test case. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/13483017 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14199 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-04-10 08:29:39 +00:00
CcTest::InitializeVM();
Print properly signed displacement in disassembler. R=titzer@chromium.org BUG= Review URL: https://codereview.chromium.org/178193028 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19667 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-03-05 09:28:26 +00:00
Isolate* isolate = CcTest::i_isolate();
Added a version of the v8::HandleScope constructor with an Isolate and use that consistently. I tried to limit the use of v8::Isolate::GetCurrent() and v8::internal::Isolate::Current() as much as possible, but sometimes this would have involved restructuring tests quite a bit, which is better left for a separate CL. BUG=v8:2487 Review URL: https://codereview.chromium.org/12716010 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@13953 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-03-15 12:06:53 +00:00
HandleScope scope(isolate);
[ia32] Add pextrb/pextrw, pinsrb, pshufb/pshuflw and AVX version Also add vpinsrw BUG= Review-Url: https://codereview.chromium.org/2931333002 Cr-Commit-Position: refs/heads/master@{#45883}
2017-06-13 02:06:17 +00:00
v8::internal::byte buffer[8192];
Added a version of the v8::HandleScope constructor with an Isolate and use that consistently. I tried to limit the use of v8::Isolate::GetCurrent() and v8::internal::Isolate::Current() as much as possible, but sometimes this would have involved restructuring tests quite a bit, which is better left for a separate CL. BUG=v8:2487 Review URL: https://codereview.chromium.org/12716010 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@13953 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-03-15 12:06:53 +00:00
Assembler assm(isolate, buffer, sizeof buffer);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
DummyStaticFunction(NULL); // just bloody use it (DELETE; debugging)
// Short immediate instructions
__ adc(eax, 12345678);
Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
__ add(eax, Immediate(12345678));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ or_(eax, 12345678);
Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
__ sub(eax, Immediate(12345678));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ xor_(eax, 12345678);
__ and_(eax, 12345678);
Deprecate FACTORY helper macro. This removes the FACTORY helper macro to avoid accidental TLS access when using the factory. Most internal code has access to the Isolate by now whereas tests which are not performance critical still heavily use TLS access through explicit Isolate::Current() calls. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/16337005 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14931 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-04 10:30:05 +00:00
Handle<FixedArray> foo = isolate->factory()->NewFixedArray(10, TENURED);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ cmp(eax, foo);
// ---- This one caused crash
__ mov(ebx, Operand(esp, ecx, times_2, 0)); // [esp+ecx*4]
// ---- All instructions that I can think of
Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
__ add(edx, ebx);
Rename RelocInfo::NONE to RelocInfo::NONE32. This CL only does renaming, nothing else. Followup to: https://chromiumcodereview.appspot.com/11695006/ There are now NONE and NONE64 RelocInfo types, but only ARM uses them both at the same time. They were added in: https://chromiumcodereview.appspot.com/11191029/ R= ulan@chromium.org Review URL: https://chromiumcodereview.appspot.com/11744020 Patch from JF Bastien <jfb@chromium.org>. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@13311 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-01-04 10:56:24 +00:00
__ add(edx, Operand(12, RelocInfo::NONE32));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ add(edx, Operand(ebx, 0));
__ add(edx, Operand(ebx, 16));
__ add(edx, Operand(ebx, 1999));
Print properly signed displacement in disassembler. R=titzer@chromium.org BUG= Review URL: https://codereview.chromium.org/178193028 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19667 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-03-05 09:28:26 +00:00
__ add(edx, Operand(ebx, -4));
__ add(edx, Operand(ebx, -1999));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ add(edx, Operand(esp, 0));
__ add(edx, Operand(esp, 16));
__ add(edx, Operand(esp, 1999));
Print properly signed displacement in disassembler. R=titzer@chromium.org BUG= Review URL: https://codereview.chromium.org/178193028 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19667 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-03-05 09:28:26 +00:00
__ add(edx, Operand(esp, -4));
__ add(edx, Operand(esp, -1999));
__ nop();
__ add(esi, Operand(ecx, times_4, 0));
__ add(esi, Operand(ecx, times_4, 24));
__ add(esi, Operand(ecx, times_4, -4));
__ add(esi, Operand(ecx, times_4, -1999));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ nop();
__ add(edi, Operand(ebp, ecx, times_4, 0));
__ add(edi, Operand(ebp, ecx, times_4, 12));
Print properly signed displacement in disassembler. R=titzer@chromium.org BUG= Review URL: https://codereview.chromium.org/178193028 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19667 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-03-05 09:28:26 +00:00
__ add(edi, Operand(ebp, ecx, times_4, -8));
__ add(edi, Operand(ebp, ecx, times_4, -3999));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ add(Operand(ebp, ecx, times_4, 12), Immediate(12));
__ nop();
Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
__ add(ebx, Immediate(12));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ nop();
[wasm] Int64Lowering of Int64Add on ia32 and arm. Int64Add is lowered to a new turbofan operator, Int32AddPair. The new operator takes 4 inputs an generates 2 outputs. The inputs are the low word of the left input, high word of the left input, the low word of the right input, and high word of the right input. The ouputs are the low and high word of the result of the addition. R=titzer@chromium.org, v8-arm-ports@googlegroups.com Review URL: https://codereview.chromium.org/1778493004 Cr-Commit-Position: refs/heads/master@{#34747}
2016-03-14 15:33:15 +00:00
__ adc(edx, Operand(ebx));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ adc(ecx, 12);
__ adc(ecx, 1000);
__ nop();
__ and_(edx, 3);
__ and_(edx, Operand(esp, 4));
__ cmp(edx, 3);
__ cmp(edx, Operand(esp, 4));
__ cmp(Operand(ebp, ecx, times_4, 0), Immediate(1000));
Deprecate FACTORY helper macro. This removes the FACTORY helper macro to avoid accidental TLS access when using the factory. Most internal code has access to the Isolate by now whereas tests which are not performance critical still heavily use TLS access through explicit Isolate::Current() calls. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/16337005 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14931 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-06-04 10:30:05 +00:00
Handle<FixedArray> foo2 = isolate->factory()->NewFixedArray(10, TENURED);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ cmp(ebx, foo2);
Implement inline string compare on ARM. Backport optimizations from x64 version to ia32. Review URL: http://codereview.chromium.org/546087 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3670 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-01-21 12:10:56 +00:00
__ cmpb(ebx, Operand(ebp, ecx, times_2, 0));
__ cmpb(Operand(ebp, ecx, times_2, 0), ebx);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ or_(edx, 3);
__ xor_(edx, 3);
__ nop();
Fix the CPU feature detection. Move all of the CPU detection logic to the CPU class, and make all other code use the CPU class for feature detection. This also fixes the ARM CPU feature detection logic, which was based on fragile string search in /proc/cpuinfo. Now we use ELF hwcaps if available, falling back to sane(!!) parsing of /proc/cpuinfo for CPU features. The ia32 and x64 code was also cleaned up to make it usable outside the assembler. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/23401002 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16315 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-26 09:37:39 +00:00
__ cpuid();
Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
__ movsx_b(edx, ecx);
__ movsx_w(edx, ecx);
__ movzx_b(edx, ecx);
__ movzx_w(edx, ecx);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ nop();
Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
__ imul(edx, ecx);
[wasm] Int64Lowering of I64Shl on ia32. I64Shl is lowered to a new turbofan operator, WasmWord64Shl. The new operator takes 3 inputs, the low-word input, the high-word input, and the shift, and produces 2 output, the low-word output and the high-word output. At the moment I implemented the lowering only for ia32, but I think the CL is already big enough. I will add the other platforms in separate CLs. R=titzer@chromium.org Review URL: https://codereview.chromium.org/1756863002 Cr-Commit-Position: refs/heads/master@{#34546}
2016-03-07 15:17:54 +00:00
__ shld(edx, ecx, 10);
__ shld_cl(edx, ecx);
[wasm] Int64Lowering of I64ShrU and I64ShrS on ia32. I implemented I64ShrU and I64ShrS the same as I64Shl in https://codereview.chromium.org/1756863002 R=titzer@chromium.org Review URL: https://codereview.chromium.org/1768233002 Cr-Commit-Position: refs/heads/master@{#34630}
2016-03-09 16:37:29 +00:00
__ shrd(edx, ecx, 10);
__ shrd_cl(edx, ecx);
Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
__ bts(edx, ecx);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ bts(Operand(ebx, ecx, times_4, 0), ecx);
__ nop();
__ pushad();
__ popad();
__ pushfd();
__ popfd();
__ push(Immediate(12));
__ push(Immediate(23456));
__ push(ecx);
__ push(esi);
__ push(Operand(ebp, JavaScriptFrameConstants::kFunctionOffset));
__ push(Operand(ebx, ecx, times_4, 0));
__ push(Operand(ebx, ecx, times_4, 0));
__ push(Operand(ebx, ecx, times_4, 10000));
__ pop(edx);
__ pop(eax);
__ pop(Operand(ebx, ecx, times_4, 0));
__ nop();
__ add(edx, Operand(esp, 16));
Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
__ add(edx, ecx);
__ mov_b(edx, ecx);
__ mov_b(ecx, 6);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ mov_b(Operand(ebx, ecx, times_4, 10000), 6);
__ mov_b(Operand(esp, 16), edx);
__ mov_w(edx, Operand(esp, 16));
__ mov_w(Operand(esp, 16), edx);
__ nop();
__ movsx_w(edx, Operand(esp, 12));
__ movsx_b(edx, Operand(esp, 12));
__ movzx_w(edx, Operand(esp, 12));
__ movzx_b(edx, Operand(esp, 12));
__ nop();
__ mov(edx, 1234567);
__ mov(edx, Operand(esp, 12));
__ mov(Operand(ebx, ecx, times_4, 10000), Immediate(12345));
__ mov(Operand(ebx, ecx, times_4, 10000), edx);
__ nop();
__ dec_b(edx);
Support dec_b instruction in ia32 and x64 disassembler. Review URL: http://codereview.chromium.org/3412028 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5515 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-09-23 19:01:56 +00:00
__ dec_b(Operand(eax, 10));
__ dec_b(Operand(ebx, ecx, times_4, 10000));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ dec(edx);
__ cdq();
__ nop();
__ idiv(edx);
Land the Fan (disabled) R=mstarzinger@chromium.org Review URL: https://codereview.chromium.org/426233002 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@22709 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-07-30 13:54:45 +00:00
__ idiv(Operand(edx, ecx, times_1, 1));
__ idiv(Operand(esp, 12));
__ div(edx);
__ div(Operand(edx, ecx, times_1, 1));
__ div(Operand(esp, 12));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ mul(edx);
__ neg(edx);
__ not_(edx);
__ test(Operand(ebx, ecx, times_4, 10000), Immediate(123456));
__ imul(edx, Operand(ebx, ecx, times_4, 10000));
__ imul(edx, ecx, 12);
Land the Fan (disabled) R=mstarzinger@chromium.org Review URL: https://codereview.chromium.org/426233002 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@22709 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-07-30 13:54:45 +00:00
__ imul(edx, Operand(edx, eax, times_2, 42), 8);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ imul(edx, ecx, 1000);
Land the Fan (disabled) R=mstarzinger@chromium.org Review URL: https://codereview.chromium.org/426233002 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@22709 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-07-30 13:54:45 +00:00
__ imul(edx, Operand(ebx, ecx, times_4, 1), 9000);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ inc(edx);
__ inc(Operand(ebx, ecx, times_4, 10000));
__ push(Operand(ebx, ecx, times_4, 10000));
__ pop(Operand(ebx, ecx, times_4, 10000));
__ call(Operand(ebx, ecx, times_4, 10000));
__ jmp(Operand(ebx, ecx, times_4, 10000));
__ lea(edx, Operand(ebx, ecx, times_4, 10000));
__ or_(edx, 12345);
__ or_(edx, Operand(ebx, ecx, times_4, 10000));
__ nop();
__ rcl(edx, 1);
__ rcl(edx, 7);
Add rcr to disassembler. Review URL: http://codereview.chromium.org/3108023 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5268 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-08-16 11:18:42 +00:00
__ rcr(edx, 1);
__ rcr(edx, 7);
[turbofan] support all shift operands on ia32 R=bmeurer@chromium.org BUG= Review URL: https://codereview.chromium.org/619663002 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24387 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-10-02 09:04:04 +00:00
__ ror(edx, 1);
__ ror(edx, 6);
__ ror_cl(edx);
__ ror(Operand(ebx, ecx, times_4, 10000), 1);
__ ror(Operand(ebx, ecx, times_4, 10000), 6);
__ ror_cl(Operand(ebx, ecx, times_4, 10000));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ sar(edx, 1);
__ sar(edx, 6);
Changes to Intel shift functions Change name of shifts picking the shift count from cl to sal_cl, shl_cl and shr_cl. Add special encoding of shift by one for shr which was missing it. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3314 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-11-17 08:35:43 +00:00
__ sar_cl(edx);
Land the Fan (disabled) R=mstarzinger@chromium.org Review URL: https://codereview.chromium.org/426233002 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@22709 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-07-30 13:54:45 +00:00
__ sar(Operand(ebx, ecx, times_4, 10000), 1);
__ sar(Operand(ebx, ecx, times_4, 10000), 6);
__ sar_cl(Operand(ebx, ecx, times_4, 10000));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ sbb(edx, Operand(ebx, ecx, times_4, 10000));
__ shl(edx, 1);
__ shl(edx, 6);
Changes to Intel shift functions Change name of shifts picking the shift count from cl to sal_cl, shl_cl and shr_cl. Add special encoding of shift by one for shr which was missing it. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3314 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-11-17 08:35:43 +00:00
__ shl_cl(edx);
Land the Fan (disabled) R=mstarzinger@chromium.org Review URL: https://codereview.chromium.org/426233002 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@22709 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-07-30 13:54:45 +00:00
__ shl(Operand(ebx, ecx, times_4, 10000), 1);
__ shl(Operand(ebx, ecx, times_4, 10000), 6);
__ shl_cl(Operand(ebx, ecx, times_4, 10000));
[wasm] Int64Lowering of I64ShrU and I64ShrS on ia32. I implemented I64ShrU and I64ShrS the same as I64Shl in https://codereview.chromium.org/1756863002 R=titzer@chromium.org Review URL: https://codereview.chromium.org/1768233002 Cr-Commit-Position: refs/heads/master@{#34630}
2016-03-09 16:37:29 +00:00
__ shrd_cl(Operand(ebx, ecx, times_4, 10000), edx);
Add missing case for shr in IA-32 disassembler Review URL: http://codereview.chromium.org/397024 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3316 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-11-17 10:24:27 +00:00
__ shr(edx, 1);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ shr(edx, 7);
Changes to Intel shift functions Change name of shifts picking the shift count from cl to sal_cl, shl_cl and shr_cl. Add special encoding of shift by one for shr which was missing it. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3314 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-11-17 08:35:43 +00:00
__ shr_cl(edx);
Land the Fan (disabled) R=mstarzinger@chromium.org Review URL: https://codereview.chromium.org/426233002 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@22709 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-07-30 13:54:45 +00:00
__ shr(Operand(ebx, ecx, times_4, 10000), 1);
__ shr(Operand(ebx, ecx, times_4, 10000), 6);
__ shr_cl(Operand(ebx, ecx, times_4, 10000));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
// Immediates
__ adc(edx, 12345);
Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
__ add(ebx, Immediate(12));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ add(Operand(edx, ecx, times_4, 10000), Immediate(12));
__ and_(ebx, 12345);
__ cmp(ebx, 12345);
Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
__ cmp(ebx, Immediate(12));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ cmp(Operand(edx, ecx, times_4, 10000), Immediate(12));
Extends testb and cmpb/cmpw instruction support in the ia32 assembler. This is in preparation for a CL that does the equivalent of http://crrev.com/1780193003 for ia32. BUG= Review URL: https://codereview.chromium.org/1815213002 Cr-Commit-Position: refs/heads/master@{#34925}
2016-03-21 10:08:43 +00:00
__ cmpb(eax, Immediate(100));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ or_(ebx, 12345);
Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
__ sub(ebx, Immediate(12));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ sub(Operand(edx, ecx, times_4, 10000), Immediate(12));
__ xor_(ebx, 12345);
__ imul(edx, ecx, 12);
__ imul(edx, ecx, 1000);
Faster filling newly allocated arrays with the holes from the Array construction stub. Review URL: http://codereview.chromium.org/661245 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3995 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-03-02 12:42:15 +00:00
__ cld();
Add missing instructions to the IA-32 disasembler Added newly added instructions to test-disasem-ia32.cc and implemented the missi ng ones in the disasembler. Added some asserts to 8-bit instructions which only work with eax, ebx, ecx and edx (al, bl, cl and dl). Removed the loope instruction. Review URL: http://codereview.chromium.org/548002 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3577 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-01-11 15:19:53 +00:00
__ rep_movs();
Faster filling newly allocated arrays with the holes from the Array construction stub. Review URL: http://codereview.chromium.org/661245 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3995 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-03-02 12:42:15 +00:00
__ rep_stos();
Use direct loop when filling small arrays. r3995 (http://code.google.com/p/v8/source/detail?r=3995) introduce performance regression for the case when arrat size is small (think new Array(4)). It turns out that in those cases rep stos is slower than plain loop (apprently due to ecx increment, but I didn't check this hypothesis.) The next thing to try could be direct jump into right place of long sequence of stos'es. Review URL: http://codereview.chromium.org/2082006 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4664 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-05-17 14:14:34 +00:00
__ stos();
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ sub(edx, Operand(ebx, ecx, times_4, 10000));
Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
__ sub(edx, ebx);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ test(edx, Immediate(12345));
__ test(edx, Operand(ebx, ecx, times_8, 10000));
Add test_b(Operand, imm8) to ia32 disassembler. Review URL: http://codereview.chromium.org/2765001 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4822 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-06-08 12:44:24 +00:00
__ test(Operand(esi, edi, times_1, -20000000), Immediate(300000000));
__ test_b(edx, Operand(ecx, ebx, times_2, 1000));
Extends testb and cmpb/cmpw instruction support in the ia32 assembler. This is in preparation for a CL that does the equivalent of http://crrev.com/1780193003 for ia32. BUG= Review URL: https://codereview.chromium.org/1815213002 Cr-Commit-Position: refs/heads/master@{#34925}
2016-03-21 10:08:43 +00:00
__ test_b(Operand(eax, -20), Immediate(0x9A));
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ nop();
__ xor_(edx, 12345);
__ xor_(edx, Operand(ebx, ecx, times_8, 10000));
__ bts(Operand(ebx, ecx, times_8, 10000), edx);
__ hlt();
__ int3();
__ ret(0);
__ ret(8);
// Calls
Label L1, L2;
__ bind(&L1);
__ nop();
__ call(&L1);
__ call(&L2);
__ nop();
__ bind(&L2);
__ call(Operand(ebx, ecx, times_4, 10000));
__ nop();
Reland "[builtins] Remove Builtins::Name() accessors" This is a reland of 2f79e035602640b9e2b63e3cbdf07c565df7be04 Original change's description: > [builtins] Remove Builtins::Name() accessors > > Instead of auto-generating the Name() convenience accessor, use a macro to > avoid wasting code space. > > BUILTIN_CODE(isolate, Name) > > expands to > > isolate->builtins()->builtin_handle(Builtins::kName); > > This reduces the size of libv8.so by 134,752 bytes on a x64 release build. > > Bug: v8:6624 > Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng > Change-Id: Idff7ee5c45e344e73412c0f47e92553c7c7ff75f > Reviewed-on: https://chromium-review.googlesource.com/593607 > Reviewed-by: Andreas Haas <ahaas@chromium.org> > Reviewed-by: Benedikt Meurer <bmeurer@chromium.org> > Commit-Queue: Jakob Gruber <jgruber@chromium.org> > Cr-Commit-Position: refs/heads/master@{#47010} TBR=bmeurer@chromium.org,ahaas@chromium.org Bug: v8:6624 Change-Id: I4733731e56dc8873ee06c2b36cac1918c0a658b2 Reviewed-on: https://chromium-review.googlesource.com/594087 Commit-Queue: Jakob Gruber <jgruber@chromium.org> Reviewed-by: Andreas Haas <ahaas@chromium.org> Reviewed-by: Jakob Gruber <jgruber@chromium.org> Cr-Commit-Position: refs/heads/master@{#47037}
2017-08-01 08:15:46 +00:00
Handle<Code> ic = BUILTIN_CODE(isolate, LoadIC);
Refactor the enum RelocMode changing the naming scheme from lower case to upper case. Moved it into the RelocInfo class together with the associated is_xxx functions. Renamed is_xxx to IsXxx in the process. Removed the exit_js_frame mode as it was no longer used. Patch Set 2 renames RELOC_MODE_COUNT to NUMBER_OF_MODES and fixes a couple of lint errors. Review URL: http://codereview.chromium.org/3186 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@354 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-09-22 13:57:03 +00:00
__ call(ic, RelocInfo::CODE_TARGET);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ nop();
Refactor the enum RelocMode changing the naming scheme from lower case to upper case. Moved it into the RelocInfo class together with the associated is_xxx functions. Renamed is_xxx to IsXxx in the process. Removed the exit_js_frame mode as it was no longer used. Patch Set 2 renames RELOC_MODE_COUNT to NUMBER_OF_MODES and fixes a couple of lint errors. Review URL: http://codereview.chromium.org/3186 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@354 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-09-22 13:57:03 +00:00
__ call(FUNCTION_ADDR(DummyStaticFunction), RelocInfo::RUNTIME_ENTRY);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ nop();
__ jmp(&L1);
__ jmp(Operand(ebx, ecx, times_4, 10000));
Refactor the enum RelocMode changing the naming scheme from lower case to upper case. Moved it into the RelocInfo class together with the associated is_xxx functions. Renamed is_xxx to IsXxx in the process. Removed the exit_js_frame mode as it was no longer used. Patch Set 2 renames RELOC_MODE_COUNT to NUMBER_OF_MODES and fixes a couple of lint errors. Review URL: http://codereview.chromium.org/3186 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@354 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-09-22 13:57:03 +00:00
__ jmp(ic, RelocInfo::CODE_TARGET);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ nop();
Label Ljcc;
__ nop();
// long jumps
__ j(overflow, &Ljcc);
__ j(no_overflow, &Ljcc);
__ j(below, &Ljcc);
__ j(above_equal, &Ljcc);
__ j(equal, &Ljcc);
__ j(not_equal, &Ljcc);
__ j(below_equal, &Ljcc);
__ j(above, &Ljcc);
__ j(sign, &Ljcc);
__ j(not_sign, &Ljcc);
__ j(parity_even, &Ljcc);
__ j(parity_odd, &Ljcc);
__ j(less, &Ljcc);
__ j(greater_equal, &Ljcc);
__ j(less_equal, &Ljcc);
__ j(greater, &Ljcc);
__ nop();
__ bind(&Ljcc);
// short jumps
__ j(overflow, &Ljcc);
__ j(no_overflow, &Ljcc);
__ j(below, &Ljcc);
__ j(above_equal, &Ljcc);
__ j(equal, &Ljcc);
__ j(not_equal, &Ljcc);
__ j(below_equal, &Ljcc);
__ j(above, &Ljcc);
__ j(sign, &Ljcc);
__ j(not_sign, &Ljcc);
__ j(parity_even, &Ljcc);
__ j(parity_odd, &Ljcc);
__ j(less, &Ljcc);
__ j(greater_equal, &Ljcc);
__ j(less_equal, &Ljcc);
__ j(greater, &Ljcc);
// 0xD9 instructions
__ nop();
Implemented missing instructions in ia32 and x64 disassembler. ia32: fld(i), fldpi x64: fld(i), fldpi, cvtsd2si, cvttsd2si Review URL: http://codereview.chromium.org/3471011 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5525 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-09-24 12:55:17 +00:00
__ fld(1);
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ fld1();
__ fldz();
Implemented missing instructions in ia32 and x64 disassembler. ia32: fld(i), fldpi x64: fld(i), fldpi, cvtsd2si, cvttsd2si Review URL: http://codereview.chromium.org/3471011 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5525 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2010-09-24 12:55:17 +00:00
__ fldpi();
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ fabs();
__ fchs();
__ fprem();
__ fprem1();
__ fincstp();
__ ftst();
__ fxch(3);
__ fld_s(Operand(ebx, ecx, times_4, 10000));
__ fstp_s(Operand(ebx, ecx, times_4, 10000));
__ ffree(3);
__ fld_d(Operand(ebx, ecx, times_4, 10000));
__ fstp_d(Operand(ebx, ecx, times_4, 10000));
__ nop();
__ fild_s(Operand(ebx, ecx, times_4, 10000));
__ fistp_s(Operand(ebx, ecx, times_4, 10000));
__ fild_d(Operand(ebx, ecx, times_4, 10000));
__ fistp_d(Operand(ebx, ecx, times_4, 10000));
__ fnstsw_ax();
__ nop();
__ fadd(3);
__ fsub(3);
__ fmul(3);
__ fdiv(3);
__ faddp(3);
__ fsubp(3);
__ fmulp(3);
__ fdivp(3);
__ fcompp();
__ fwait();
[x64] add disasm for two fp instructions BUG= R=yangguo@chromium.org Review URL: https://codereview.chromium.org/146583002 Patch from Weiliang Lin <weiliang.lin@intel.com>. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19022 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-02-03 08:13:12 +00:00
__ frndint();
__ fninit();
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ nop();
Introduce addps/subps/mulps/divps for IA32/X64 BUG= R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/60093005 Patch from Weiliang Lin <weiliang.lin@intel.com>. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17842 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-11-18 15:24:41 +00:00
// SSE instruction
{
Require SSE2 support for the ia32 port. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/275433004 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-05-09 13:01:50 +00:00
// Move operation
__ movaps(xmm0, xmm1);
[Turbofan] Add ia32 support for 4 and 16 byte moves and swaps. - Adds move/swap handling for 4 and 16 bytes to ia32. - Register allocator now only requests 4 bytes for floats on ia32 and arm. - We probably need similar support in mips. LOG=N BUG=v8:4124 Review-Url: https://codereview.chromium.org/2027043002 Cr-Commit-Position: refs/heads/master@{#37714}
2016-07-13 13:53:04 +00:00
__ movups(xmm0, xmm1);
__ movups(xmm0, Operand(edx, 4));
__ movups(Operand(edx, 4), xmm0);
Require SSE2 support for the ia32 port. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/275433004 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-05-09 13:01:50 +00:00
__ shufps(xmm0, xmm0, 0x0);
[turbofan] IA: TruncateFloat64ToFloat32 supports mem operand BUG= R=dcarney@chromium.org Review URL: https://codereview.chromium.org/639283003 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24541 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-10-13 07:12:57 +00:00
__ cvtsd2ss(xmm0, xmm1);
__ cvtsd2ss(xmm0, Operand(ebx, ecx, times_4, 10000));
Require SSE2 support for the ia32 port. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/275433004 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-05-09 13:01:50 +00:00
// logic operation
__ andps(xmm0, xmm1);
__ andps(xmm0, Operand(ebx, ecx, times_4, 10000));
__ orps(xmm0, xmm1);
__ orps(xmm0, Operand(ebx, ecx, times_4, 10000));
__ xorps(xmm0, xmm1);
__ xorps(xmm0, Operand(ebx, ecx, times_4, 10000));
// Arithmetic operation
[ia32] Introduce FMA3 instructions on scalar data elements. port 83a635e0d70f144300ea83be0d7effc1eb1bf6ef BUG= Review URL: https://codereview.chromium.org/773783002 Cr-Commit-Position: refs/heads/master@{#25619}
2014-12-02 15:30:03 +00:00
__ addss(xmm1, xmm0);
__ addss(xmm1, Operand(ebx, ecx, times_4, 10000));
__ mulss(xmm1, xmm0);
__ mulss(xmm1, Operand(ebx, ecx, times_4, 10000));
__ subss(xmm1, xmm0);
__ subss(xmm1, Operand(ebx, ecx, times_4, 10000));
__ divss(xmm1, xmm0);
__ divss(xmm1, Operand(ebx, ecx, times_4, 10000));
[turbofan] Add backend support for float32 operations. This adds the basics necessary to support float32 operations in TurboFan. The actual functionality required to detect safe float32 operations will be added based on this later. Therefore this does not affect production code except for some cleanup/refactoring. In detail, this patchset contains the following features: - Add support for float32 operations to arm, arm64, ia32 and x64 backends. - Add float32 machine operators. - Add support for float32 constants to simplified lowering. - Handle float32 representation for phis in simplified lowering. In addition, contains the following (related) cleanups: - Fix/unify naming of backend instructions. - Use AVX comparisons when available. - Extend ArchOpcodeField to 9 bits (required for arm64). - Refactor some code duplication in instruction selectors. BUG=v8:3589 LOG=n R=dcarney@chromium.org Review URL: https://codereview.chromium.org/1044793002 Cr-Commit-Position: refs/heads/master@{#27509}
2015-03-30 07:33:46 +00:00
__ maxss(xmm1, xmm0);
__ maxss(xmm1, Operand(ebx, ecx, times_4, 10000));
__ minss(xmm1, xmm0);
__ minss(xmm1, Operand(ebx, ecx, times_4, 10000));
__ sqrtss(xmm1, xmm0);
__ sqrtss(xmm1, Operand(ebx, ecx, times_4, 10000));
Require SSE2 support for the ia32 port. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/275433004 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-05-09 13:01:50 +00:00
__ addps(xmm1, xmm0);
__ addps(xmm1, Operand(ebx, ecx, times_4, 10000));
__ subps(xmm1, xmm0);
__ subps(xmm1, Operand(ebx, ecx, times_4, 10000));
__ mulps(xmm1, xmm0);
__ mulps(xmm1, Operand(ebx, ecx, times_4, 10000));
__ divps(xmm1, xmm0);
__ divps(xmm1, Operand(ebx, ecx, times_4, 10000));
[ia32] Add minps/maxps and AVX v_ps/v_pd for add/sub/mul/div/min/max Also use vinstr for vps/vpd/vss/vsd BUG= Review-Url: https://codereview.chromium.org/2747103002 Cr-Commit-Position: refs/heads/master@{#43809}
2017-03-15 02:01:05 +00:00
__ minps(xmm1, xmm0);
__ minps(xmm1, Operand(ebx, ecx, times_4, 10000));
__ maxps(xmm1, xmm0);
__ maxps(xmm1, Operand(ebx, ecx, times_4, 10000));
[ia32] Add rcpps, rsqrtps, cvtdq2ps, cvttps2dq instr(xmm, xmm/mem) vinstr(xmm, xmm/mem) BUG= Review-Url: https://codereview.chromium.org/2870253003 Cr-Commit-Position: refs/heads/master@{#45293}
2017-05-15 09:00:42 +00:00
__ rcpps(xmm1, xmm0);
__ rcpps(xmm1, Operand(ebx, ecx, times_4, 10000));
__ rsqrtps(xmm1, xmm0);
__ rsqrtps(xmm1, Operand(ebx, ecx, times_4, 10000));
[ia32] Introduce FMA3 instructions on scalar data elements. port 83a635e0d70f144300ea83be0d7effc1eb1bf6ef BUG= Review URL: https://codereview.chromium.org/773783002 Cr-Commit-Position: refs/heads/master@{#25619}
2014-12-02 15:30:03 +00:00
[ia32] Add cmpps/vcmpps for eq/lt/le/neq BUG= Review-Url: https://codereview.chromium.org/2847683002 Cr-Commit-Position: refs/heads/master@{#44955}
2017-04-28 01:56:26 +00:00
__ cmpeqps(xmm5, xmm1);
__ cmpeqps(xmm5, Operand(ebx, ecx, times_4, 10000));
__ cmpltps(xmm5, xmm1);
__ cmpltps(xmm5, Operand(ebx, ecx, times_4, 10000));
__ cmpleps(xmm5, xmm1);
__ cmpleps(xmm5, Operand(ebx, ecx, times_4, 10000));
__ cmpneqps(xmm5, xmm1);
__ cmpneqps(xmm5, Operand(ebx, ecx, times_4, 10000));
[ia32] Introduce FMA3 instructions on scalar data elements. port 83a635e0d70f144300ea83be0d7effc1eb1bf6ef BUG= Review URL: https://codereview.chromium.org/773783002 Cr-Commit-Position: refs/heads/master@{#25619}
2014-12-02 15:30:03 +00:00
__ ucomiss(xmm0, xmm1);
__ ucomiss(xmm0, Operand(ebx, ecx, times_4, 10000));
Introduce addps/subps/mulps/divps for IA32/X64 BUG= R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/60093005 Patch from Weiliang Lin <weiliang.lin@intel.com>. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17842 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-11-18 15:24:41 +00:00
}
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
{
Require SSE2 support for the ia32 port. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/275433004 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-05-09 13:01:50 +00:00
__ cvttss2si(edx, Operand(ebx, ecx, times_4, 10000));
__ cvtsi2sd(xmm1, Operand(ebx, ecx, times_4, 10000));
[turbofan]IA: ChangeFloat32ToFloat64 supports mem operand BUG= R=titzer@chromium.org Review URL: https://codereview.chromium.org/641153003 Patch from Jing Bao <jing.bao@intel.com>. git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24542 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-10-13 07:42:35 +00:00
__ cvtss2sd(xmm1, Operand(ebx, ecx, times_4, 10000));
__ cvtss2sd(xmm1, xmm0);
[ia32] Add rcpps, rsqrtps, cvtdq2ps, cvttps2dq instr(xmm, xmm/mem) vinstr(xmm, xmm/mem) BUG= Review-Url: https://codereview.chromium.org/2870253003 Cr-Commit-Position: refs/heads/master@{#45293}
2017-05-15 09:00:42 +00:00
__ cvtdq2ps(xmm1, xmm0);
__ cvtdq2ps(xmm1, Operand(ebx, ecx, times_4, 10000));
__ cvttps2dq(xmm1, xmm0);
__ cvttps2dq(xmm1, Operand(ebx, ecx, times_4, 10000));
Require SSE2 support for the ia32 port. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/275433004 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-05-09 13:01:50 +00:00
__ movsd(xmm1, Operand(ebx, ecx, times_4, 10000));
__ movsd(Operand(ebx, ecx, times_4, 10000), xmm1);
// 128 bit move instructions.
__ movdqa(xmm0, Operand(ebx, ecx, times_4, 10000));
__ movdqa(Operand(ebx, ecx, times_4, 10000), xmm0);
__ movdqu(xmm0, Operand(ebx, ecx, times_4, 10000));
__ movdqu(Operand(ebx, ecx, times_4, 10000), xmm0);
[ia32][wasm] Support AVX instructions for I32x4Splat/ReplaceLane/ExtractLane Add vpinsrd, vpextrd, vmovd. Supplement pshufd and add vpshufd. Add Movd and Pshufd macro. Add AVX support to Pextrd. Fix disassembler to recognize "pinsrd/pextrd xmm,[mem],imm8". BUG= Review-Url: https://codereview.chromium.org/2916093002 Cr-Commit-Position: refs/heads/master@{#45833}
2017-06-12 01:57:05 +00:00
__ movd(xmm0, edi);
__ movd(xmm0, Operand(ebx, ecx, times_4, 10000));
__ movd(eax, xmm1);
__ movd(Operand(ebx, ecx, times_4, 10000), xmm1);
Require SSE2 support for the ia32 port. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/275433004 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-05-09 13:01:50 +00:00
__ addsd(xmm1, xmm0);
IA: Double arithmetic binops support memory operand BUG= R=dcarney@chromium.org Review URL: https://codereview.chromium.org/662813002 Patch from Weiliang Lin <weiliang.lin@intel.com>. Cr-Commit-Position: refs/heads/master@{#25052} git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@25052 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-11-03 05:56:55 +00:00
__ addsd(xmm1, Operand(ebx, ecx, times_4, 10000));
Require SSE2 support for the ia32 port. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/275433004 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-05-09 13:01:50 +00:00
__ mulsd(xmm1, xmm0);
IA: Double arithmetic binops support memory operand BUG= R=dcarney@chromium.org Review URL: https://codereview.chromium.org/662813002 Patch from Weiliang Lin <weiliang.lin@intel.com>. Cr-Commit-Position: refs/heads/master@{#25052} git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@25052 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-11-03 05:56:55 +00:00
__ mulsd(xmm1, Operand(ebx, ecx, times_4, 10000));
Require SSE2 support for the ia32 port. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/275433004 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-05-09 13:01:50 +00:00
__ subsd(xmm1, xmm0);
[turbofan] IA: Float64ToUint32 supports mem operand BUG= R=titzer@chromium.org Review URL: https://codereview.chromium.org/582713002 Patch from Weiliang Lin <weiliang.lin@intel.com>. git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24092 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-09-19 14:25:13 +00:00
__ subsd(xmm1, Operand(ebx, ecx, times_4, 10000));
Require SSE2 support for the ia32 port. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/275433004 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-05-09 13:01:50 +00:00
__ divsd(xmm1, xmm0);
IA: Double arithmetic binops support memory operand BUG= R=dcarney@chromium.org Review URL: https://codereview.chromium.org/662813002 Patch from Weiliang Lin <weiliang.lin@intel.com>. Cr-Commit-Position: refs/heads/master@{#25052} git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@25052 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-11-03 05:56:55 +00:00
__ divsd(xmm1, Operand(ebx, ecx, times_4, 10000));
[turbofan] Introduce optional Float64Min and Float64Max machine operators. Basically recognize certain x < y ? x : y constructs and turn that into Float64Min/Float64Max operations, if the target machine supports that. On x86 we lower to (v)minsd/(v)maxsd. R=dcarney@chromium.org Review URL: https://codereview.chromium.org/998283002 Cr-Commit-Position: refs/heads/master@{#27160}
2015-03-12 14:07:28 +00:00
__ minsd(xmm1, xmm0);
__ minsd(xmm1, Operand(ebx, ecx, times_4, 10000));
__ maxsd(xmm1, xmm0);
__ maxsd(xmm1, Operand(ebx, ecx, times_4, 10000));
Require SSE2 support for the ia32 port. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/275433004 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-05-09 13:01:50 +00:00
__ ucomisd(xmm0, xmm1);
__ cmpltsd(xmm0, xmm1);
__ andpd(xmm0, xmm1);
[ia32] Supplement shift instructions for SIMD psll/rl/raw(xmm, imm) psrad(xmm, imm) vpsll/rl/raw(xmm, xmm, imm) vpsll/rl/rad(xmm, xmm, imm) BUG= Review-Url: https://codereview.chromium.org/2747783004 Cr-Commit-Position: refs/heads/master@{#43762}
2017-03-14 08:01:20 +00:00
__ psllw(xmm0, 17);
__ pslld(xmm0, 17);
__ psrlw(xmm0, 17);
__ psrld(xmm0, 17);
__ psraw(xmm0, 17);
__ psrad(xmm0, 17);
Require SSE2 support for the ia32 port. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/275433004 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-05-09 13:01:50 +00:00
__ psllq(xmm0, 17);
__ psllq(xmm0, xmm1);
__ psrlq(xmm0, 17);
__ psrlq(xmm0, xmm1);
Add several SIMD opcodes to IA32 CreateInt32x4, Int32x4ExtractLane, Int32x4ReplaceLane Int32x4Add, Int32x4Sub Also add paddd, psubd, vpaddd, vpsubd, pinsrw to ia32-assembler BUG= Review-Url: https://codereview.chromium.org/2695613004 Cr-Original-Commit-Position: refs/heads/master@{#43483} Committed: https://chromium.googlesource.com/v8/v8/+/4deb9ffdecf121c69a3db7eae6698eae23a80a15 Review-Url: https://codereview.chromium.org/2695613004 Cr-Commit-Position: refs/heads/master@{#43708}
2017-03-10 02:40:06 +00:00
[ia32] Add pextrb/pextrw, pinsrb, pshufb/pshuflw and AVX version Also add vpinsrw BUG= Review-Url: https://codereview.chromium.org/2931333002 Cr-Commit-Position: refs/heads/master@{#45883}
2017-06-13 02:06:17 +00:00
__ pshuflw(xmm5, xmm1, 5);
__ pshuflw(xmm5, Operand(edx, 4), 5);
[ia32][wasm] Support AVX instructions for I32x4Splat/ReplaceLane/ExtractLane Add vpinsrd, vpextrd, vmovd. Supplement pshufd and add vpshufd. Add Movd and Pshufd macro. Add AVX support to Pextrd. Fix disassembler to recognize "pinsrd/pextrd xmm,[mem],imm8". BUG= Review-Url: https://codereview.chromium.org/2916093002 Cr-Commit-Position: refs/heads/master@{#45833}
2017-06-12 01:57:05 +00:00
__ pshufd(xmm5, xmm1, 5);
__ pshufd(xmm5, Operand(edx, 4), 5);
Add several SIMD opcodes to IA32 CreateInt32x4, Int32x4ExtractLane, Int32x4ReplaceLane Int32x4Add, Int32x4Sub Also add paddd, psubd, vpaddd, vpsubd, pinsrw to ia32-assembler BUG= Review-Url: https://codereview.chromium.org/2695613004 Cr-Original-Commit-Position: refs/heads/master@{#43483} Committed: https://chromium.googlesource.com/v8/v8/+/4deb9ffdecf121c69a3db7eae6698eae23a80a15 Review-Url: https://codereview.chromium.org/2695613004 Cr-Commit-Position: refs/heads/master@{#43708}
2017-03-10 02:40:06 +00:00
__ pinsrw(xmm5, edx, 5);
__ pinsrw(xmm5, Operand(edx, 4), 5);
#define EMIT_SSE2_INSTR(instruction, notUsed1, notUsed2, notUsed3) \
__ instruction(xmm5, xmm1); \
__ instruction(xmm5, Operand(edx, 4));
SSE2_INSTRUCTION_LIST(EMIT_SSE2_INSTR)
#undef EMIT_SSE2_INSTR
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
}
Add support for comisd and cmov in IA-32 disassembler. Review URL: http://codereview.chromium.org/266014 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3040 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-10-08 13:28:46 +00:00
// cmov.
{
Require CMOV support for the ia32 port. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/275253004 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21279 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-05-13 08:16:26 +00:00
__ cmov(overflow, eax, Operand(eax, 0));
__ cmov(no_overflow, eax, Operand(eax, 1));
__ cmov(below, eax, Operand(eax, 2));
__ cmov(above_equal, eax, Operand(eax, 3));
__ cmov(equal, eax, Operand(ebx, 0));
__ cmov(not_equal, eax, Operand(ebx, 1));
__ cmov(below_equal, eax, Operand(ebx, 2));
__ cmov(above, eax, Operand(ebx, 3));
__ cmov(sign, eax, Operand(ecx, 0));
__ cmov(not_sign, eax, Operand(ecx, 1));
__ cmov(parity_even, eax, Operand(ecx, 2));
__ cmov(parity_odd, eax, Operand(ecx, 3));
__ cmov(less, eax, Operand(edx, 0));
__ cmov(greater_equal, eax, Operand(edx, 1));
__ cmov(less_equal, eax, Operand(edx, 2));
__ cmov(greater, eax, Operand(edx, 3));
Add support for comisd and cmov in IA-32 disassembler. Review URL: http://codereview.chromium.org/266014 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3040 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2009-10-08 13:28:46 +00:00
}
[ia32] Add pextrb/pextrw, pinsrb, pshufb/pshuflw and AVX version Also add vpinsrw BUG= Review-Url: https://codereview.chromium.org/2931333002 Cr-Commit-Position: refs/heads/master@{#45883}
2017-06-13 02:06:17 +00:00
{
if (CpuFeatures::IsSupported(SSSE3)) {
CpuFeatureScope scope(&assm, SSSE3);
__ pshufb(xmm5, xmm1);
__ pshufb(xmm5, Operand(edx, 4));
}
}
Change the code for materializing double constants on ia32. Instead of using the stack, use a temporary integer register and avoid memory access. Review URL: http://codereview.chromium.org/6452002 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6696 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-02-09 12:39:15 +00:00
{
Require SSE2 support for the ia32 port. R=svenpanne@chromium.org Review URL: https://codereview.chromium.org/275433004 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-05-09 13:01:50 +00:00
if (CpuFeatures::IsSupported(SSE4_1)) {
Cleaned up CpuFeature scope handling. First of all, it has nothing to do with Isolates, it is related to the assembler at hand. Furthermore, the saving/restoring is platform-independent. Cleaned up some platform-specific stuff on the way. Note that there are some things which still need some cleanup, like e.g. using EnumSet instead of uint64_t, making Probe() more uniform across platforms etc., but the CL is already big enough. BUG=v8:2487 Review URL: https://codereview.chromium.org/12391055 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@13823 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-03-05 10:48:16 +00:00
CpuFeatureScope scope(&assm, SSE4_1);
[ia32] Add pextrb/pextrw, pinsrb, pshufb/pshuflw and AVX version Also add vpinsrw BUG= Review-Url: https://codereview.chromium.org/2931333002 Cr-Commit-Position: refs/heads/master@{#45883}
2017-06-13 02:06:17 +00:00
__ pextrb(eax, xmm0, 1);
__ pextrb(Operand(edx, 4), xmm0, 1);
__ pextrw(eax, xmm0, 1);
__ pextrw(Operand(edx, 4), xmm0, 1);
Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
__ pextrd(eax, xmm0, 1);
[ia32][wasm] Support AVX instructions for I32x4Splat/ReplaceLane/ExtractLane Add vpinsrd, vpextrd, vmovd. Supplement pshufd and add vpshufd. Add Movd and Pshufd macro. Add AVX support to Pextrd. Fix disassembler to recognize "pinsrd/pextrd xmm,[mem],imm8". BUG= Review-Url: https://codereview.chromium.org/2916093002 Cr-Commit-Position: refs/heads/master@{#45833}
2017-06-12 01:57:05 +00:00
__ pextrd(Operand(edx, 4), xmm0, 1);
[ia32] Add pextrb/pextrw, pinsrb, pshufb/pshuflw and AVX version Also add vpinsrw BUG= Review-Url: https://codereview.chromium.org/2931333002 Cr-Commit-Position: refs/heads/master@{#45883}
2017-06-13 02:06:17 +00:00
__ pinsrb(xmm1, eax, 0);
__ pinsrb(xmm1, Operand(edx, 4), 0);
Clean up the x86 assembler API. The API is inconsistent about when a register must be coerced to an operand and when it can be used as a register. Simplify usage by never requiring it to be wrapped. R=fschneider@chromium.org BUG= TEST= Review URL: http://codereview.chromium.org/8086021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9507 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-10-03 11:44:39 +00:00
__ pinsrd(xmm1, eax, 0);
[ia32][wasm] Support AVX instructions for I32x4Splat/ReplaceLane/ExtractLane Add vpinsrd, vpextrd, vmovd. Supplement pshufd and add vpshufd. Add Movd and Pshufd macro. Add AVX support to Pextrd. Fix disassembler to recognize "pinsrd/pextrd xmm,[mem],imm8". BUG= Review-Url: https://codereview.chromium.org/2916093002 Cr-Commit-Position: refs/heads/master@{#45833}
2017-06-12 01:57:05 +00:00
__ pinsrd(xmm1, Operand(edx, 4), 0);
Fixed bug in extractps instruction on ia32 and x64 This is a fixed version of https://codereview.chromium.org/27097002/ which was originally written by weiliang.lin@intel.com. R=bmeurer@chromium.org Review URL: https://codereview.chromium.org/27301003 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17217 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-10-15 12:51:58 +00:00
__ extractps(eax, xmm1, 0);
[ia32] Add some SSE2, SSE4 instructions and AVX version for SIMD Add instructions in following format: instr(xmm, xmm/mem) vinstr(xmm, xmm, xmm/mem) BUG= Review-Url: https://codereview.chromium.org/2744643004 Cr-Commit-Position: refs/heads/master@{#43760}
2017-03-14 02:00:13 +00:00
#define EMIT_SSE4_INSTR(instruction, notUsed1, notUsed2, notUsed3, notUsed4) \
__ instruction(xmm5, xmm1); \
__ instruction(xmm5, Operand(edx, 4));
SSE4_INSTRUCTION_LIST(EMIT_SSE4_INSTR)
#undef EMIT_SSE4_INSTR
Change the code for materializing double constants on ia32. Instead of using the stack, use a temporary integer register and avoid memory access. Review URL: http://codereview.chromium.org/6452002 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6696 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-02-09 12:39:15 +00:00
}
}
[ia32] Introduce vex prefix version of float64 arithmetic binop port 50c4d8826b838596cda892d76ec84eb7a0eab9c1 BUG= Review URL: https://codereview.chromium.org/770183002 Cr-Commit-Position: refs/heads/master@{#25595}
2014-12-02 08:09:39 +00:00
// AVX instruction
{
if (CpuFeatures::IsSupported(AVX)) {
CpuFeatureScope scope(&assm, AVX);
__ vaddsd(xmm0, xmm1, xmm2);
__ vaddsd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vmulsd(xmm0, xmm1, xmm2);
__ vmulsd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vsubsd(xmm0, xmm1, xmm2);
__ vsubsd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vdivsd(xmm0, xmm1, xmm2);
__ vdivsd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
[turbofan] Introduce optional Float64Min and Float64Max machine operators. Basically recognize certain x < y ? x : y constructs and turn that into Float64Min/Float64Max operations, if the target machine supports that. On x86 we lower to (v)minsd/(v)maxsd. R=dcarney@chromium.org Review URL: https://codereview.chromium.org/998283002 Cr-Commit-Position: refs/heads/master@{#27160}
2015-03-12 14:07:28 +00:00
__ vminsd(xmm0, xmm1, xmm2);
__ vminsd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vmaxsd(xmm0, xmm1, xmm2);
__ vmaxsd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
[turbofan] Add backend support for float32 operations. This adds the basics necessary to support float32 operations in TurboFan. The actual functionality required to detect safe float32 operations will be added based on this later. Therefore this does not affect production code except for some cleanup/refactoring. In detail, this patchset contains the following features: - Add support for float32 operations to arm, arm64, ia32 and x64 backends. - Add float32 machine operators. - Add support for float32 constants to simplified lowering. - Handle float32 representation for phis in simplified lowering. In addition, contains the following (related) cleanups: - Fix/unify naming of backend instructions. - Use AVX comparisons when available. - Extend ArchOpcodeField to 9 bits (required for arm64). - Refactor some code duplication in instruction selectors. BUG=v8:3589 LOG=n R=dcarney@chromium.org Review URL: https://codereview.chromium.org/1044793002 Cr-Commit-Position: refs/heads/master@{#27509}
2015-03-30 07:33:46 +00:00
__ vaddss(xmm0, xmm1, xmm2);
__ vaddss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vmulss(xmm0, xmm1, xmm2);
__ vmulss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vsubss(xmm0, xmm1, xmm2);
__ vsubss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vdivss(xmm0, xmm1, xmm2);
__ vdivss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vminss(xmm0, xmm1, xmm2);
__ vminss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vmaxss(xmm0, xmm1, xmm2);
__ vmaxss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
[x86] Introduce vandps/vandpd/vxorps/vxorpd. R=bmeurer@chromium.org Review URL: https://codereview.chromium.org/1072343002 Cr-Commit-Position: refs/heads/master@{#27768}
2015-04-11 01:02:22 +00:00
__ vandps(xmm0, xmm1, xmm2);
__ vandps(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vxorps(xmm0, xmm1, xmm2);
__ vxorps(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
[ia32] Add minps/maxps and AVX v_ps/v_pd for add/sub/mul/div/min/max Also use vinstr for vps/vpd/vss/vsd BUG= Review-Url: https://codereview.chromium.org/2747103002 Cr-Commit-Position: refs/heads/master@{#43809}
2017-03-15 02:01:05 +00:00
__ vaddps(xmm0, xmm1, xmm2);
__ vaddps(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vmulps(xmm0, xmm1, xmm2);
__ vmulps(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vsubps(xmm0, xmm1, xmm2);
__ vsubps(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vminps(xmm0, xmm1, xmm2);
__ vminps(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vdivps(xmm0, xmm1, xmm2);
__ vdivps(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vmaxps(xmm0, xmm1, xmm2);
__ vmaxps(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
[ia32] Add rcpps, rsqrtps, cvtdq2ps, cvttps2dq instr(xmm, xmm/mem) vinstr(xmm, xmm/mem) BUG= Review-Url: https://codereview.chromium.org/2870253003 Cr-Commit-Position: refs/heads/master@{#45293}
2017-05-15 09:00:42 +00:00
__ vrcpps(xmm1, xmm0);
__ vrcpps(xmm1, Operand(ebx, ecx, times_4, 10000));
__ vrsqrtps(xmm1, xmm0);
__ vrsqrtps(xmm1, Operand(ebx, ecx, times_4, 10000));
[x86] Introduce vandps/vandpd/vxorps/vxorpd. R=bmeurer@chromium.org Review URL: https://codereview.chromium.org/1072343002 Cr-Commit-Position: refs/heads/master@{#27768}
2015-04-11 01:02:22 +00:00
[ia32] Add cmpps/vcmpps for eq/lt/le/neq BUG= Review-Url: https://codereview.chromium.org/2847683002 Cr-Commit-Position: refs/heads/master@{#44955}
2017-04-28 01:56:26 +00:00
__ vcmpeqps(xmm5, xmm4, xmm1);
__ vcmpeqps(xmm5, xmm4, Operand(ebx, ecx, times_4, 10000));
__ vcmpltps(xmm5, xmm4, xmm1);
__ vcmpltps(xmm5, xmm4, Operand(ebx, ecx, times_4, 10000));
__ vcmpleps(xmm5, xmm4, xmm1);
__ vcmpleps(xmm5, xmm4, Operand(ebx, ecx, times_4, 10000));
__ vcmpneqps(xmm5, xmm4, xmm1);
__ vcmpneqps(xmm5, xmm4, Operand(ebx, ecx, times_4, 10000));
[x86] Introduce vandps/vandpd/vxorps/vxorpd. R=bmeurer@chromium.org Review URL: https://codereview.chromium.org/1072343002 Cr-Commit-Position: refs/heads/master@{#27768}
2015-04-11 01:02:22 +00:00
__ vandpd(xmm0, xmm1, xmm2);
__ vandpd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vxorpd(xmm0, xmm1, xmm2);
__ vxorpd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
[ia32] Add minps/maxps and AVX v_ps/v_pd for add/sub/mul/div/min/max Also use vinstr for vps/vpd/vss/vsd BUG= Review-Url: https://codereview.chromium.org/2747103002 Cr-Commit-Position: refs/heads/master@{#43809}
2017-03-15 02:01:05 +00:00
__ vaddpd(xmm0, xmm1, xmm2);
__ vaddpd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vmulpd(xmm0, xmm1, xmm2);
__ vmulpd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vsubpd(xmm0, xmm1, xmm2);
__ vsubpd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vminpd(xmm0, xmm1, xmm2);
__ vminpd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vdivpd(xmm0, xmm1, xmm2);
__ vdivpd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vmaxpd(xmm0, xmm1, xmm2);
__ vmaxpd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
Add several SIMD opcodes to IA32 CreateInt32x4, Int32x4ExtractLane, Int32x4ReplaceLane Int32x4Add, Int32x4Sub Also add paddd, psubd, vpaddd, vpsubd, pinsrw to ia32-assembler BUG= Review-Url: https://codereview.chromium.org/2695613004 Cr-Original-Commit-Position: refs/heads/master@{#43483} Committed: https://chromium.googlesource.com/v8/v8/+/4deb9ffdecf121c69a3db7eae6698eae23a80a15 Review-Url: https://codereview.chromium.org/2695613004 Cr-Commit-Position: refs/heads/master@{#43708}
2017-03-10 02:40:06 +00:00
[ia32] Supplement shift instructions for SIMD psll/rl/raw(xmm, imm) psrad(xmm, imm) vpsll/rl/raw(xmm, xmm, imm) vpsll/rl/rad(xmm, xmm, imm) BUG= Review-Url: https://codereview.chromium.org/2747783004 Cr-Commit-Position: refs/heads/master@{#43762}
2017-03-14 08:01:20 +00:00
__ vpsllw(xmm0, xmm7, 21);
__ vpslld(xmm0, xmm7, 21);
__ vpsrlw(xmm0, xmm7, 21);
__ vpsrld(xmm0, xmm7, 21);
__ vpsraw(xmm0, xmm7, 21);
__ vpsrad(xmm0, xmm7, 21);
[ia32] Add rcpps, rsqrtps, cvtdq2ps, cvttps2dq instr(xmm, xmm/mem) vinstr(xmm, xmm/mem) BUG= Review-Url: https://codereview.chromium.org/2870253003 Cr-Commit-Position: refs/heads/master@{#45293}
2017-05-15 09:00:42 +00:00
[ia32] Add pextrb/pextrw, pinsrb, pshufb/pshuflw and AVX version Also add vpinsrw BUG= Review-Url: https://codereview.chromium.org/2931333002 Cr-Commit-Position: refs/heads/master@{#45883}
2017-06-13 02:06:17 +00:00
__ vpshufb(xmm5, xmm0, xmm1);
__ vpshufb(xmm5, xmm0, Operand(edx, 4));
__ vpshuflw(xmm5, xmm1, 5);
__ vpshuflw(xmm5, Operand(edx, 4), 5);
[ia32][wasm] Support AVX instructions for I32x4Splat/ReplaceLane/ExtractLane Add vpinsrd, vpextrd, vmovd. Supplement pshufd and add vpshufd. Add Movd and Pshufd macro. Add AVX support to Pextrd. Fix disassembler to recognize "pinsrd/pextrd xmm,[mem],imm8". BUG= Review-Url: https://codereview.chromium.org/2916093002 Cr-Commit-Position: refs/heads/master@{#45833}
2017-06-12 01:57:05 +00:00
__ vpshufd(xmm5, xmm1, 5);
__ vpshufd(xmm5, Operand(edx, 4), 5);
[ia32] Add pextrb/pextrw, pinsrb, pshufb/pshuflw and AVX version Also add vpinsrw BUG= Review-Url: https://codereview.chromium.org/2931333002 Cr-Commit-Position: refs/heads/master@{#45883}
2017-06-13 02:06:17 +00:00
__ vpextrb(eax, xmm0, 1);
__ vpextrb(Operand(edx, 4), xmm0, 1);
__ vpextrw(eax, xmm0, 1);
__ vpextrw(Operand(edx, 4), xmm0, 1);
[ia32][wasm] Support AVX instructions for I32x4Splat/ReplaceLane/ExtractLane Add vpinsrd, vpextrd, vmovd. Supplement pshufd and add vpshufd. Add Movd and Pshufd macro. Add AVX support to Pextrd. Fix disassembler to recognize "pinsrd/pextrd xmm,[mem],imm8". BUG= Review-Url: https://codereview.chromium.org/2916093002 Cr-Commit-Position: refs/heads/master@{#45833}
2017-06-12 01:57:05 +00:00
__ vpextrd(eax, xmm0, 1);
__ vpextrd(Operand(edx, 4), xmm0, 1);
[ia32] Add pextrb/pextrw, pinsrb, pshufb/pshuflw and AVX version Also add vpinsrw BUG= Review-Url: https://codereview.chromium.org/2931333002 Cr-Commit-Position: refs/heads/master@{#45883}
2017-06-13 02:06:17 +00:00
__ vpinsrb(xmm0, xmm1, eax, 0);
__ vpinsrb(xmm0, xmm1, Operand(edx, 4), 0);
__ vpinsrw(xmm0, xmm1, eax, 0);
__ vpinsrw(xmm0, xmm1, Operand(edx, 4), 0);
[ia32][wasm] Support AVX instructions for I32x4Splat/ReplaceLane/ExtractLane Add vpinsrd, vpextrd, vmovd. Supplement pshufd and add vpshufd. Add Movd and Pshufd macro. Add AVX support to Pextrd. Fix disassembler to recognize "pinsrd/pextrd xmm,[mem],imm8". BUG= Review-Url: https://codereview.chromium.org/2916093002 Cr-Commit-Position: refs/heads/master@{#45833}
2017-06-12 01:57:05 +00:00
__ vpinsrd(xmm0, xmm1, eax, 0);
__ vpinsrd(xmm0, xmm1, Operand(edx, 4), 0);
[ia32] Add rcpps, rsqrtps, cvtdq2ps, cvttps2dq instr(xmm, xmm/mem) vinstr(xmm, xmm/mem) BUG= Review-Url: https://codereview.chromium.org/2870253003 Cr-Commit-Position: refs/heads/master@{#45293}
2017-05-15 09:00:42 +00:00
__ vcvtdq2ps(xmm1, xmm0);
__ vcvtdq2ps(xmm1, Operand(ebx, ecx, times_4, 10000));
__ vcvttps2dq(xmm1, xmm0);
__ vcvttps2dq(xmm1, Operand(ebx, ecx, times_4, 10000));
[ia32][wasm] Support AVX instructions for I32x4Splat/ReplaceLane/ExtractLane Add vpinsrd, vpextrd, vmovd. Supplement pshufd and add vpshufd. Add Movd and Pshufd macro. Add AVX support to Pextrd. Fix disassembler to recognize "pinsrd/pextrd xmm,[mem],imm8". BUG= Review-Url: https://codereview.chromium.org/2916093002 Cr-Commit-Position: refs/heads/master@{#45833}
2017-06-12 01:57:05 +00:00
__ vmovd(xmm0, edi);
__ vmovd(xmm0, Operand(ebx, ecx, times_4, 10000));
__ vmovd(eax, xmm1);
__ vmovd(Operand(ebx, ecx, times_4, 10000), xmm1);
Add several SIMD opcodes to IA32 CreateInt32x4, Int32x4ExtractLane, Int32x4ReplaceLane Int32x4Add, Int32x4Sub Also add paddd, psubd, vpaddd, vpsubd, pinsrw to ia32-assembler BUG= Review-Url: https://codereview.chromium.org/2695613004 Cr-Original-Commit-Position: refs/heads/master@{#43483} Committed: https://chromium.googlesource.com/v8/v8/+/4deb9ffdecf121c69a3db7eae6698eae23a80a15 Review-Url: https://codereview.chromium.org/2695613004 Cr-Commit-Position: refs/heads/master@{#43708}
2017-03-10 02:40:06 +00:00
#define EMIT_SSE2_AVXINSTR(instruction, notUsed1, notUsed2, notUsed3) \
__ v##instruction(xmm7, xmm5, xmm1); \
__ v##instruction(xmm7, xmm5, Operand(edx, 4));
SSE2_INSTRUCTION_LIST(EMIT_SSE2_AVXINSTR)
#undef EMIT_SSE2_AVXINSTR
[ia32] Add some SSE2, SSE4 instructions and AVX version for SIMD Add instructions in following format: instr(xmm, xmm/mem) vinstr(xmm, xmm, xmm/mem) BUG= Review-Url: https://codereview.chromium.org/2744643004 Cr-Commit-Position: refs/heads/master@{#43760}
2017-03-14 02:00:13 +00:00
#define EMIT_SSE4_AVXINSTR(instruction, notUsed1, notUsed2, notUsed3, \
notUsed4) \
__ v##instruction(xmm7, xmm5, xmm1); \
__ v##instruction(xmm7, xmm5, Operand(edx, 4));
SSE4_INSTRUCTION_LIST(EMIT_SSE4_AVXINSTR)
#undef EMIT_SSE4_AVXINSTR
[ia32] Introduce vex prefix version of float64 arithmetic binop port 50c4d8826b838596cda892d76ec84eb7a0eab9c1 BUG= Review URL: https://codereview.chromium.org/770183002 Cr-Commit-Position: refs/heads/master@{#25595}
2014-12-02 08:09:39 +00:00
}
}
[ia32] Introduce FMA3 instructions on scalar data elements. port 83a635e0d70f144300ea83be0d7effc1eb1bf6ef BUG= Review URL: https://codereview.chromium.org/773783002 Cr-Commit-Position: refs/heads/master@{#25619}
2014-12-02 15:30:03 +00:00
// FMA3 instruction
{
if (CpuFeatures::IsSupported(FMA3)) {
CpuFeatureScope scope(&assm, FMA3);
__ vfmadd132sd(xmm0, xmm1, xmm2);
__ vfmadd132sd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfmadd213sd(xmm0, xmm1, xmm2);
__ vfmadd213sd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfmadd231sd(xmm0, xmm1, xmm2);
__ vfmadd231sd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfmsub132sd(xmm0, xmm1, xmm2);
__ vfmsub132sd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfmsub213sd(xmm0, xmm1, xmm2);
__ vfmsub213sd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfmsub231sd(xmm0, xmm1, xmm2);
__ vfmsub231sd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfnmadd132sd(xmm0, xmm1, xmm2);
__ vfnmadd132sd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfnmadd213sd(xmm0, xmm1, xmm2);
__ vfnmadd213sd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfnmadd231sd(xmm0, xmm1, xmm2);
__ vfnmadd231sd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfnmsub132sd(xmm0, xmm1, xmm2);
__ vfnmsub132sd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfnmsub213sd(xmm0, xmm1, xmm2);
__ vfnmsub213sd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfnmsub231sd(xmm0, xmm1, xmm2);
__ vfnmsub231sd(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfmadd132ss(xmm0, xmm1, xmm2);
__ vfmadd132ss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfmadd213ss(xmm0, xmm1, xmm2);
__ vfmadd213ss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfmadd231ss(xmm0, xmm1, xmm2);
__ vfmadd231ss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfmsub132ss(xmm0, xmm1, xmm2);
__ vfmsub132ss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfmsub213ss(xmm0, xmm1, xmm2);
__ vfmsub213ss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfmsub231ss(xmm0, xmm1, xmm2);
__ vfmsub231ss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfnmadd132ss(xmm0, xmm1, xmm2);
__ vfnmadd132ss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfnmadd213ss(xmm0, xmm1, xmm2);
__ vfnmadd213ss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfnmadd231ss(xmm0, xmm1, xmm2);
__ vfnmadd231ss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfnmsub132ss(xmm0, xmm1, xmm2);
__ vfnmsub132ss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfnmsub213ss(xmm0, xmm1, xmm2);
__ vfnmsub213ss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
__ vfnmsub231ss(xmm0, xmm1, xmm2);
__ vfnmsub231ss(xmm0, xmm1, Operand(ebx, ecx, times_4, 10000));
}
}
[ia32] Introduce BMI instructions. BUG=v8:4015 LOG=n Review URL: https://codereview.chromium.org/1069683002 Cr-Commit-Position: refs/heads/master@{#27683}
2015-04-09 06:50:52 +00:00
// BMI1 instructions
{
if (CpuFeatures::IsSupported(BMI1)) {
CpuFeatureScope scope(&assm, BMI1);
__ andn(eax, ebx, ecx);
__ andn(eax, ebx, Operand(ebx, ecx, times_4, 10000));
__ bextr(eax, ebx, ecx);
__ bextr(eax, Operand(ebx, ecx, times_4, 10000), ebx);
__ blsi(eax, ebx);
__ blsi(eax, Operand(ebx, ecx, times_4, 10000));
__ blsmsk(eax, ebx);
__ blsmsk(eax, Operand(ebx, ecx, times_4, 10000));
__ blsr(eax, ebx);
__ blsr(eax, Operand(ebx, ecx, times_4, 10000));
__ tzcnt(eax, ebx);
__ tzcnt(eax, Operand(ebx, ecx, times_4, 10000));
}
}
// LZCNT instructions
{
if (CpuFeatures::IsSupported(LZCNT)) {
CpuFeatureScope scope(&assm, LZCNT);
__ lzcnt(eax, ebx);
__ lzcnt(eax, Operand(ebx, ecx, times_4, 10000));
}
}
// POPCNT instructions
{
if (CpuFeatures::IsSupported(POPCNT)) {
CpuFeatureScope scope(&assm, POPCNT);
__ popcnt(eax, ebx);
__ popcnt(eax, Operand(ebx, ecx, times_4, 10000));
}
}
// BMI2 instructions
{
if (CpuFeatures::IsSupported(BMI2)) {
CpuFeatureScope scope(&assm, BMI2);
__ bzhi(eax, ebx, ecx);
__ bzhi(eax, Operand(ebx, ecx, times_4, 10000), ebx);
__ mulx(eax, ebx, ecx);
__ mulx(eax, ebx, Operand(ebx, ecx, times_4, 10000));
__ pdep(eax, ebx, ecx);
__ pdep(eax, ebx, Operand(ebx, ecx, times_4, 10000));
__ pext(eax, ebx, ecx);
__ pext(eax, ebx, Operand(ebx, ecx, times_4, 10000));
__ sarx(eax, ebx, ecx);
__ sarx(eax, Operand(ebx, ecx, times_4, 10000), ebx);
__ shlx(eax, ebx, ecx);
__ shlx(eax, Operand(ebx, ecx, times_4, 10000), ebx);
__ shrx(eax, ebx, ecx);
__ shrx(eax, Operand(ebx, ecx, times_4, 10000), ebx);
__ rorx(eax, ebx, 31);
__ rorx(eax, Operand(ebx, ecx, times_4, 10000), 31);
}
}
Land the Fan (disabled) R=mstarzinger@chromium.org Review URL: https://codereview.chromium.org/426233002 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@22709 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-07-30 13:54:45 +00:00
// xchg.
{
Add cmpxchg and lock instructions to x64 and ia32 {dis,}assemblers Review-Url: https://codereview.chromium.org/1986113004 Cr-Commit-Position: refs/heads/master@{#36341}
2016-05-19 00:55:08 +00:00
__ xchg_b(eax, Operand(eax, 8));
__ xchg_w(eax, Operand(ebx, 8));
Land the Fan (disabled) R=mstarzinger@chromium.org Review URL: https://codereview.chromium.org/426233002 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@22709 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-07-30 13:54:45 +00:00
__ xchg(eax, eax);
__ xchg(eax, ebx);
__ xchg(ebx, ebx);
__ xchg(ebx, Operand(esp, 12));
}
Add cmpxchg and lock instructions to x64 and ia32 {dis,}assemblers Review-Url: https://codereview.chromium.org/1986113004 Cr-Commit-Position: refs/heads/master@{#36341}
2016-05-19 00:55:08 +00:00
// cmpxchg.
{
__ cmpxchg_b(Operand(esp, 12), eax);
__ cmpxchg_w(Operand(ebx, ecx, times_4, 10000), eax);
__ cmpxchg(Operand(ebx, ecx, times_4, 10000), eax);
}
// lock prefix.
{
__ lock();
__ cmpxchg(Operand(esp, 12), ebx);
__ lock();
__ xchg_w(eax, Operand(ecx, 8));
}
Support multi-byte nop instructions as recommended by the optimization guides from AMD and Intel. Review URL: http://codereview.chromium.org/8776033 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10134 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-12-02 10:01:50 +00:00
// Nop instructions
for (int i = 0; i < 16; i++) {
__ Nop(i);
}
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
__ ret(0);
CodeDesc desc;
[compiler] Delay allocation of code-embedded heap numbers. Instead of allocating and embedding certain heap numbers into the code during code assembly, emit dummies but record the allocation requests. Later then, in Assembler::GetCode, allocate the heap numbers and patch the code by replacing the dummies with the actual objects. The RelocInfos for the embedded objects are already recorded correctly when emitting the dummies. R=jarin@chromium.org BUG=v8:6048 Review-Url: https://codereview.chromium.org/2900683002 Cr-Commit-Position: refs/heads/master@{#45635}
2017-05-31 14:00:11 +00:00
assm.GetCode(isolate, &desc);
Handlify code allocation. R=mstarzinger@chromium.org Review URL: https://codereview.chromium.org/235153003 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20795 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-04-16 11:38:56 +00:00
Handle<Code> code = isolate->factory()->NewCode(
desc, Code::ComputeFlags(Code::STUB), Handle<Code>());
Fix unused variable warnings. TBR=jarin@chromium.org Review URL: https://codereview.chromium.org/238543008 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20797 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-04-16 11:50:24 +00:00
USE(code);
Changed #define symbols in test to OBJECT_PRINT from PRINT_OBJECT because this is actually what is being used. Patch by Mark Lam from Hewlett-Packard Development Company, LP Review URL: http://codereview.chromium.org/6037015 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@6128 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2011-01-03 09:32:42 +00:00
#ifdef OBJECT_PRINT
More OStreamsUse OStreams more often. This is a mostly mechanical CL (more than 90% Emacs macros and query-replace-regexp) moving FILE*/StringStream*-based APIs to OStream-based APIs. There are a few places where this had to stop, otherwise the CL would be even bigger, but this can easily and incrementally cleaned up later. R=bmeurer@chromium.org Review URL: https://codereview.chromium.org/363323003 git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@22232 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-07-07 09:57:29 +00:00
OFStream os(stdout);
code->Print(os);
Handlify code allocation. R=mstarzinger@chromium.org Review URL: https://codereview.chromium.org/235153003 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20795 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-04-16 11:38:56 +00:00
byte* begin = code->instruction_start();
byte* end = begin + code->instruction_size();
Included mjsunit JavaScript test suite and C++ unit tests. In the shell sample don't print the result of executing a script, only evaluating expressions. Fixed issue when building samples on Windows using a shared V8 library. Added visibility option on Linux build which makes the generated library 18% smaller. Changed build system to accept multiple build modes in one build and generate seperate objects, libraries and executables for each mode. Removed deferred negation optimization (a * -b => -(a * b)) since this visibly changes operand conversion order. Improved parsing performance by introducing stack guard in preparsing. Without a stack guard preparsing always bails out with stack overflow. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-08-22 13:33:59 +00:00
disasm::Disassembler::Disassemble(stdout, begin, end);
#endif
}
#undef __
Reference in New Issue Copy Permalink