[v8-dev] Re: a tiny code review (sgjesse/fix-env-overrides)

gjesse/fix-env-overrides/bleeding_edge/samples/[EMAIL PROTECTED] > ) > @@ -31,6 +31,7 @@ Import('sample context') > def ConfigureObjectFiles(): > env = Environment() > env.Replace(**context.flags['sample']) > + context.ApplyEnvOverrides(env) > return env

[v8-dev] Re: a tiny code review (sgjesse/fix-env-overrides)

On Mon, Sep 8, 2008 at 5:26 PM, Christian Plesner Hansen < [EMAIL PROTECTED]> wrote: > Was ApplyEnvOverrides defined already? Yes - when it was introduced I forgot to use it in one place. > > > Lgtm. > > On Mon, Sep 8, 2008 at 5:20 PM, Søren Gjesse <[EMAIL PROTECTE

[v8-dev] Re: Add assembler.h the the Visual Studio project file.

Right - switched .cc and .h. The Visual Studio UI sorts the files independent of the order in the .vcproj file anyway though. /Søren On Fri, Sep 19, 2008 at 10:42 AM, Kasper Lund <[EMAIL PROTECTED]> wrote: > > Usually the .h files come after the .cc files in the list, but other > than that it LGT

[v8-dev] Re: DLL Access from other environments...

The V8 API is a C++ interface which on Windows is exported to the DLL using __declspec(dllexport). When an application is build the C++ interface is imported using __declspec(dllimport). This woks fine when Microsoft C++ is used for compiling both the DLL and the application using the DLL as the sa

[v8-dev] Re: Enable string pooling in the VS release build.

flag in the SConstruct build? > > On Sun, Sep 21, 2008 at 3:50 PM, <[EMAIL PROTECTED]> wrote: > > Reviewers: M-A, Søren Gjesse, > > > > Message: > > Apparently /GF doesn't come with /Ox (although it comes with /O1 and > > /02?). > > > > 12113

[v8-dev] Re: [v8 commit] r384 - branches/bleeding_edge/src

Sorry, I forgot change the description. I discovered that this need to be a signed comparison because we calculate what the SP will be after pushing the arguments, and that calculated value could end up wrapping around 0 becoming negative. So my change ended up with only adding test for the flag ch

[v8-dev] Re: Issue 94 in v8: Get rid of -fnostrict-aliasing

Using MSVC with /Ox (Full Optimization) all three generate the same code: pushOFFSET _s call_bar popecx /Søren On Mon, Sep 29, 2008 at 10:25 PM, Erik Corry <[EMAIL PROTECTED]> wrote: > I took a look at what code gcc-4.0 generates for these two ways of > doing things. I

[v8-dev] Re: Changed optimization from /Ox to the (more) commonly used /O2 (which includes...

Using /O1 causes about 20% reduction in the size of shell.exe but it is also about 10% slower on the V8 benchmark. Adding /Oi doesn't make much difference. Using /O2 and /Ox together with /GL (hole program optimization) gives the exact same size of shell.exe. /O2 it is. /Søren On Thu, Oct 2, 2008

[v8-dev] Re: Mark a test as flaky on ARM. Fails on hardware but passes in simulator.

Manually committed as r427. On Fri, Oct 3, 2008 at 2:08 PM, Søren Gjesse <[EMAIL PROTECTED]> wrote: > No, but I have decided to find out. > > > On Fri, Oct 3, 2008 at 1:56 PM, Kasper Lund <[EMAIL PROTECTED]> wrote: > >> LGTM. Do you know why this started happenin

[v8-dev] Re: Fixed unsafe code where a GC could occour after a Handle had been deferenced....

Sending the code review to you I should have guessed. Changed all Handle initializations to assignments in the function (previously 2 where assignments and 2 where not). On Mon, Oct 6, 2008 at 3:01 PM, Kasper Lund <[EMAIL PROTECTED]> wrote: > LGTM. I would use assignment for the initialization of

[v8-dev] Re: Fixed unsafe code where a GC could occour after a Handle had been deferenced....

Added comment explaining the problem with GetScriptWrapper calling GC. On Mon, Oct 6, 2008 at 3:02 PM, Kasper Lund <[EMAIL PROTECTED]> wrote: > ... and I would add a comment on why you can't just fold the code > together like it was originally... > > On Mon, Oct 6, 2008 at 3:01 PM, Kasper Lund <[

[v8-dev] Re: Marked a test parsing on ARM as parsing.

Manually committed as r446. On Mon, Oct 6, 2008 at 4:22 PM, Kasper Lund <[EMAIL PROTECTED]> wrote: > > LGTM. I think there was another problem win debug mode on ARMu: > > === debug debug-script-breakpoints === > Path: mjsunit/debug-script-breakpoints > Command: /auto/JavaScriptV8/flexo/linux64/bu

[v8-dev] Re: Posponed the setting of "external_caught_exception" to when leaving JavaScrip...

I will clean up this in a new changelist. It will probably change the catcher_ from a TryCatch pointer to a boolean flag. On Thu, Oct 23, 2008 at 2:45 PM, <[EMAIL PROTECTED]> wrote: > Reviewers: Christian Plesner Hansen, Kasper Lund, > > Description: > Posponed the setting of "external_caught_exc

[v8-dev] Re: a tiny code review ([EMAIL PROTECTED]/runtime_calls)

LGTM - nice to get rid of that. On Fri, Oct 24, 2008 at 11:32 AM, <[EMAIL PROTECTED]> wrote: > Kasper & Søren, > > I'd like you to do a code review. To review this change, run > > gvn review --project https://v8.googlecode.com/svn > [EMAIL PROTECTED]/[EMAIL PROTECTED] > > Alternatively, to revi

[v8-dev] Re: By default disable the general protection fault message box when...

New snapshot uploaded. On Tue, Oct 28, 2008 at 11:18 PM, <[EMAIL PROTECTED]> wrote: > > http://codereview.chromium.org/8676/diff/1/3 > File src/platform-win32.cc (right): > > http://codereview.chromium.org/8676/diff/1/3#newcode786 > Line 786: abort(); > On 2008/10/28 22:15:57, sgjesse wrote: > >>

[v8-dev] Re: Skip the tests which currently crash on the ARM simulator.

I just changed SKIP to CRASH in the status file as the outcome of a crashed test was already being set to CRASH in the test frameweok. On Wed, Oct 29, 2008 at 2:27 PM, Kasper Lund <[EMAIL PROTECTED]> wrote: > > LGTM. Would it make sense to have a CRASH marker that would run the > test and expect i

[v8-dev] Re: Changed the handling of Win32 function SetErrorMode to be more correct. The f...

Changed oring ot merging in comment. On Wed, Nov 19, 2008 at 11:00 AM, <[EMAIL PROTECTED]> wrote: > Reviewers: Christian Plesner Hansen, > > Description: > Changed the handling of Win32 function SetErrorMode to be more correct. > The flag to prevent error dialogs is now merged with existing flags

[v8-dev] Re: Change Windows daylight saving time calculations to...

We already use _USE_32BIT_TIME_T in both SCons and Visual Studio builds. /Søren On Mon, Dec 1, 2008 at 3:19 PM, Mads Ager <[EMAIL PROTECTED]> wrote: > > Thanks Dean. Bill put in explicit checks for that and we make sure to > convert to a reasonable date in the caller. > > -- Mads > > On Mon, Dec

[v8-dev] Re: Fix trailing whitespace.

LGTM, thanks On Mon, Dec 8, 2008 at 12:23 PM, <[EMAIL PROTECTED]> wrote: > Reviewers: Søren Gjesse, > > Description: > Fix trailing whitespace. > > TBR=sgjesse > > Please review this at http://codereview.chromium.org/13616 > > SVN Base: http://v8.googl

[v8-dev] Re: [v8 commit] r983 - branches/bleeding_edge/tools/v8.xcodeproj

Thanks for fixing this Ivan. /Søren On Mon, Dec 15, 2008 at 10:56 PM, wrote: > > Author: ipo...@chromium.org > Date: Mon Dec 15 12:40:12 2008 > New Revision: 983 > > Modified: >branches/bleeding_edge/tools/v8.xcodeproj/project.pbxproj > > Log: > Add missing files (debug-.cc) to the Xcode pr

[v8-dev] Re: Refactored the recording of source position in the generated code. The code g...

ine 1993: ASSERT(pos != RelocInfo::kNoPosition); > ARM also has a peephole optimizer so it should mirror what we do for > IA32. We have just not investigated the code quality as much at this > point. > > > 2008/12/17 08:20:26, Søren Gjesse wrote: > >> On 2008/12/17 07:45:3

[v8-dev] Re: Added command line debugger to D8 shell....

There are two identical issues in Rietveld. I assume that it is http://codereview.chromium.org/14509 which was created as the last that should be reviewed. /Søren On Wed, Dec 17, 2008 at 4:34 PM, wrote: > > Reviewers: Mads Ager, > > Description: > Added command line debugger to D8 shell. > > b

[v8-dev] Re: Merge bleeding-edge change 985, RecordStatementPosition reform, into...

LGTM On Fri, Dec 19, 2008 at 1:58 PM, wrote: > Reviewers: Søren Gjesse, Kevin Millikin, > > Message: > This change merged fine, except for a restructured for loop case in > VisitLoopStatement. That should be fine too. Change number is 985, not > 984. > > Descriptio

[v8-dev] Re: Changed the debugger API to allow only one debug event listener to be registe...

No Chromium patch will be required as Chromium uses the JSON message based API which was not changed by this changelist. On Tue, Feb 3, 2009 at 8:47 AM, wrote: > I will prepare a Chromium patch. > > > http://codereview.chromium.org/19753/diff/1/28 > File include/v8.h (right): > > http://coderevi

[v8-dev] Re: Added the simplest call stack sampling and call profile in tick processor output.

403/diff/31/32 > File src/log.cc (right): > > http://codereview.chromium.org/21403/diff/31/32#newcode866 > Line 866: fprintf(logfile_, ",0x%x", (unsigned int)(sample->stack[i])); > On 2009/02/17 14:14:50, Søren Gjesse wrote: > >> Please use reinterpret_cast(..

[v8-dev] Re: A refinement of cl 28068.

You are probably right. Lets keep an eye on the crash reports when this code makes it into a DEV release of Chrome. /Søren On Tue, Feb 24, 2009 at 3:56 PM, wrote: > I'm not sure that it's worth dealing with external resources in test > because StringMatch only branches on ASCII vs. UC16 strings

[v8-dev] Re: exposing JSON serialization?

Safari does not have it, but It is in WebKit Bugzilla https://bugs.webkit.org/show_bug.cgi?id=20031, but it looks as if it is not being worked on. /Søren On Wed, Feb 25, 2009 at 9:11 AM, Erik Corry wrote: > > > On Thu, Feb 19, 2009 at 2:07 AM, James Robinson wrote: > >> ECMAScript 3.1 supports

[v8-dev] Re: Adding unit tests for profiler's stack tracer.

I think there needs to be two kind of tests 1. Tests that we get expected results when the correct FP is sampled 2. Tests that don't crash when wild FP's are sampled For 1. the test with the code patch will provide a correct FP, and using a runtime function with a JavaScriptFrameIterator will

[v8-dev] Re: Adding unit tests for profiler's stack tracer.

shared, CLEAR_EXCEPTION, 0); Regards, Søren On Wed, Feb 25, 2009 at 4:40 PM, wrote: > > http://codereview.chromium.org/28112/diff/1/3 > File src/log.h (right): > > http://codereview.chromium.org/28112/diff/1/3#newcode276 > Line 276: class StackTracer { > On 2009/02/25 12:36:13, Søren Gj

[v8-dev] Re: Introduction and issues to work on?

Hi Matt Welcome to the project Matt it is great to have people working on V8. With regard to getting started then two the issues you have located could be good starting points. I suggest starting with issue 213 which is basically adding binary search. Parts of the code involved (locating line ends

[v8-dev] Re: Building v8 on cygwin with gcc

I guess that you are missing including stdarg.h. If you look at platform.h you will see that for MinGW stdarg.h is included directly in platform.h. If you want to use gcc on Windows I suggest that you try out MinGW as this should work now. Support was added with http://codereview.chromium.org/20177

[v8-dev] Re: Minor changes to allow V8 to build on Ubuntu Jaunty.

LGTM On Mon, Mar 9, 2009 at 9:30 AM, wrote: > Reviewers: Søren Gjesse, > > Description: > Minor changes to allow V8 to build on Ubuntu Jaunty. > > Please review this at http://codereview.chromium.org/40289 > > SVN Base: http://v8.googlecode.com/svn/branches/bleeding_

[v8-dev] Re: cpplint does not like multiple semicolons on the same line. Do not...

LGTM On Mon, Mar 9, 2009 at 9:53 AM, wrote: > Reviewers: Søren Gjesse, > > Description: > cpplint does not like multiple semicolons on the same line. Do not > lint that line. > > tbr=sgje...@chromium.org > > Please review this at http://codereview.chromium.org/39

[v8-dev] Re: Fix test for building on 64-bit Linux.

LGTM On Mon, Mar 9, 2009 at 8:47 PM, wrote: > Reviewers: Søren Gjesse, > > Description: > Fix test for building on 64-bit Linux. > > TBR=sgjesse > > Please review this at http://codereview.chromium.org/41007 > > SVN Base: http://v8.googlecode.com/svn/branches/ble

[v8-dev] Re: Disable a test that fails on some Linux versions

It was late and I was lazy... On Thu, Mar 19, 2009 at 10:47 PM, Erik Corry wrote: > Why not just mark it as failing in the status file? > > 2009/3/19 : > > Reviewers: Erik Corry, > > > > Description: > > Disable a test that fails on some Linux versions. > > > > tbr=erik.co...@gmail.com > > > >

[v8-dev] Re: Reluctantly turn off the debug flags in optimized builds....

Mike, why do you want to turn off profiling in release builds. With this change it is no longer possible to use the V8 profiler for e.g. test_shell, right? How much does the v8::internal::Logger::is_enabled() check cost when the not active. Maybe we should split regular logging calls and profiling

[v8-dev] Re: Fixed test memory leaks

When you commit the CL is not updated with the revision of the actual commit. Are you not using gcl to commit? /Søren On Fri, Mar 27, 2009 at 01:24, wrote: > > http://codereview.chromium.org/53089/diff/1/2 > File include/v8.h (right): > > http://codereview.chromium.org/53089/diff/1/2#newcode2042

[v8-dev] Re: Fixed the step in handling for function.apply

Step in for function.call does not do anything (neither the right thing nor crash) as the function.call code does not have any IC's. I was planning to have the function.call fix as a separate CL. On Tue, Apr 7, 2009 at 11:45, Kasper Lund wrote: > > How does it behave for Function.prototype.call i

[v8-dev] Re: If an external string enters the symbol table, make sure to set the...

Thank you for taking care of this. I will find a better solution for finding scripts. As a start making the debugger keep weak references to all scripts should work. This will also make the retreival of scripts much faster, and provide the debugger with the ability to give notifications when script

[v8-dev] Re: Issue 290 in v8: undefined clock_gettime symbol in r1603 lib

Please provide additional information to which symbol from librt is required as the current V8 source does no longer use clock_gettime. On Thu, Apr 23, 2009 at 05:25, wrote: > > > Comment #8 on issue 290 by pkoppstein: undefined clock_gettime symbol in > r1603 lib > http://code.google.com/p/v8/i

[v8-dev] Re: Prepare to push version 1.2.0 to trunk. Now worker on version 1.2.1.

LGTM On Thu, Apr 23, 2009 at 09:29, wrote: > Reviewers: Søren Gjesse, > > Description: > Prepare to push version 1.2.0 to trunk. Now worker on version 1.2.1. > > Please review this at http://codereview.chromium.org/92066 > > SVN Base: http://v8.googlecode.com/sv

[v8-dev] Re: Move backend specific files to separate directories.

What about removing the -ia32 and -arm suffixes would that be to dangerous? You are missing the file tools/visual_studio/v8_base_arm.vcproj. /Søren On Thu, Apr 23, 2009 at 10:41, wrote: > > This sounds reasonable. There are a few issues as discussed: > > 1. Be 100% sure that these files are mo

[v8-dev] Re: Added ProfileView object for performing sorting, searching and filtering operations on a profile.

gt; On 2009/04/24 10:55:20, Søren Gjesse wrote: > >> I woud prefer to have this function defined as an inner function in >> devtools.profiler.ViewBuilder.prototype.buildView instead of being >> > defined > >> anonomous in the arguments list. >> > > I see two

[v8-dev] Re: Changed the debugger message API to receive an object instead of a JSON string

I don't know why Rietweld garbeled my nicely formatted description... Changed the debugger message API to receive an object instead of a JSON string. The object delivered to the debug message handler contains additional information on the current break handling the messages. Clients which require

[v8-dev] Re: Added better version information...

I have looked more into the SONAME issue, and from what I can find SONAME have to match for a library to be compatible. An executable build against some shared object will have the SONAME of that shared object embedded as a needed library. Adding the full version to the SONAME should then make an e

[v8-dev] Re: Make Object::GetIdentityHash() never return 0....

Please update comment in include/v8.h to sya that GetIdentityHash() will never return 0. /Søren On Wed, Apr 29, 2009 at 03:50, wrote: > > Reviewers: iposva, > > Description: > Make Object::GetIdentityHash() never return 0. > > This is convenient when using identity hashes in data structures that

[v8-dev] Re: Modify assert so it doesn't crash when 'location' is NULL....

I cannot see how this changes anything. When a V8 ASSERT is hit OS::Abort() is called, so I can't see how you can continue after the first ASSERT anyway. Also 'return *location_' will dereference the NULL pointer causing a crash. Regards, Søren On Thu, May 14, 2009 at 21:34, wrote: > > Reviewers

[v8-dev] Re: Use JSON.stringify to serialize debugger messages

thin these files through the builtins object. > RegExp; > Date; > > > > http://codereview.chromium.org/113399/diff/34/49 > File src/debug-delay.js (right): > > http://codereview.chromium.org/113399/diff/34/49#newcode1808 > Line 1808: * @return {Object} JSON formatted obj

[v8-dev] Re: Disable compilation when debugger is active

compilation -> compilation cache! On Wed, May 20, 2009 at 10:22, wrote: > Reviewers: Kasper Lund, > > Description: > Disable compilation when debugger is active. > > BUG=343 > > Please review this at http://codereview.chromium.org/113625 > > SVN Base: http://v8.googlecode.com/svn/branches/bleedi

[v8-dev] Re: [v8 commit] r2009 - Introduce Logger::LogCompiledFunctions that logs current map of compiled code.

Mikhail, It looks as if strtoll is not available on Windows. For now strtol should work. /Søren On Wed, May 20, 2009 at 11:04, wrote: > > Author: mikhail.naga...@gmail.com > Date: Wed May 20 02:04:13 2009 > New Revision: 2009 > > Modified: >branches/bleeding_edge/src/compiler.cc >branc

[v8-dev] Re: Execute accessor for the debugged property if it's a native function

LGTM On Mon, May 25, 2009 at 16:55, wrote: > On 2009/05/25 14:13:43, Søren Gjesse wrote: > >> LGTM >> > > How about adding a test where the accessor throws an exception? >> > > On top of this we also needs to handle interceptors. They are used in >>

[v8-dev] Re: How can I dump the machine code that v8 generates when it compiles JS file

The flag is --print-code which will print all code generated, and the flag in only available for a debug build. When your first functions are compiled you will see compilation of some internal stuff, and as a lot of the builtins as lazily compiled you will see auxiliary code generation first time s

[v8-dev] Re: Can V8 be separated from chrome alone?

V8 is a separate open source project located at http://code.google.com/p/v8. Take a look at the shell sample in samples/shell.cc which can the evaluate JavaScript code and pront the result. Build the sample it using: $ scons sample=shell To build libv8.a just run: $ scons For a debug build add

[v8-dev] Re: MakeFunction in compiler.cc

If you are using XCode, take a look at the following thread on chromium-dev: http://groups.google.com/group/chromium-dev/browse_thread/thread/20bc9002733aebe5/bb90c80f9bc86911?lnk=gst&q=Where+is+the+integration+point+betwee+chromium+and+V8#bb90c80f9bc86911 . Regards, Søren On Sun, May 31, 2009 at

[v8-dev] Re: [chromium-reviews] Turn on DevTools Profiler....

It is really cool to now have the V8 profiler exposed in Chromium by default. Regards, Søren On Mon, Jun 8, 2009 at 16:48, wrote: > > Reviewers: pfeldman, > > Description: > Turn on DevTools Profiler. > > Works on Windows and Linux. On Mac V8's sampling doesn't work with > Chromium due to an unk

[v8-dev] Re: Change locationFromPosition() and locationFromLine() to use a binary search t...

le; > just let me know to go another round. > > > > On 2009/06/08 08:35:34, Søren Gjesse wrote: > >> LGTM >> > > http://codereview.chromium.org/118371/diff/1/3 >> File src/messages.js (right): >> > > http://codereview.chromium.org/118371/di

[v8-dev] Re: Fix lint error

Yep On Wed, Jun 17, 2009 at 16:14, Kasper Lund wrote: > LGTM. Does it lint? > > On Wed, Jun 17, 2009 at 4:13 PM, wrote: > > Reviewers: Kasper Lund, > > > > Description: > > Fix lint error. > > > > tbr=kasp...@chromium.org > > > > > > Please review this at http://codereview.chromium.org/126276 >

[v8-dev] Re: Include basic scopes information in 'backtrace' response

On Mon, Jun 22, 2009 at 10:45, wrote: > > http://codereview.chromium.org/131107/diff/1/2 > File src/mirror-delay.js (right): > > http://codereview.chromium.org/131107/diff/1/2#newcode2241 > Line 2241: ref: scope.scopeObject().handle() > On 2009/06/22 06:42:28, Søren G

[v8-dev] Re: Regarding Creating Object and maitaining javascript variablr scope using V8 apis

I am not sure what you mean with *suppose I have pushed a scope named "session" through v8 api*, but I suggest you take a look at the API tests (test/cctest/test-api-cc) where most ways of using the API is tested. Also I suggest that you check for exceptions after script->Run() by either checking f

[v8-dev] Re: Regarding Creating Object and maitaining javascript variablr scope using V8 apis

scope that i currently am using session.sess1 > or application.app1.Also if application is created inside the session > then we can access app1 using session.application.app1. > > On Aug 7, 11:53 am, Søren Gjesse wrote: > > I am not sure what you mean with *suppose I have pushed

[v8-dev] Re: Support "break"/suspend command

Hi Peter See answers inlined below: On Sun, Aug 30, 2009 at 21:21, wrote: > Hi Soren > > Thank you for explanation. Let me withdraw my CL now. > > On the other hand I'm interested in developing support of V8 debugger > because of chromedevtools project I'm working on. > > As I understand our ma

[v8-dev] Re: Let embedding application process pending debug requests

Peter, Adding an additional parameter to the EnableAgent is probably where this should go. However the callback should be a function like for the rest of the V8 debugger API (e.g. EventCallback). Also I don't see the need for the function ProcessDebugRequests(), The embedder should just ensure that

[v8-dev] Re: linker error in latest trunk 1.3.9

Unfortunately the configuration for building the Windows DLL was lost when adding x64 build for Windows to the SCons file. This is beeing fixed. See http://codereview.chromium.org/197036 for the patch. Regards, Søren On Mon, Sep 7, 2009 at 03:34, Alexander M Beedie wrote: > > First time this ha

[v8-dev] Re: Fixed SCons build of Windows DLL

Done - V 1.3.9.1, r2829. On Mon, Sep 7, 2009 at 09:05, wrote: > LGTM > > Will you merge it to trunk? > > > http://codereview.chromium.org/197036 > --~--~-~--~~~---~--~~ v8-dev mailing list v8-dev@googlegroups.com http://groups.google.com/group/v8-dev -~--~

[v8-dev] Re: Fix lint error

We use the Google cpplint tool which is written Python and cehcks the code against the Google C++ Style Guide. Both these resources are now open sources and available at http://code.google.com/p/google-styleguide/. The cpplint tool can be found here http://code.google.com/p/google-styleguide/source

[v8-dev] Re: Set V8::Script::Compile() scope?

It is correct that V8::Script::Compile() is being executed in the global scope. You can of cause use eval in JavaScript to evaluate code in the context of another function but this requires that you place the eval in the function source where you want to use it. There is no support in the API for c

[v8-dev] Re: [v8] r2943 committed - Tag version 1.2.14.19.

I think it was my fault as I created the same tag as well just after Mads did. I will clean it up. /Søren On Tue, Sep 22, 2009 at 09:43, Kasper Lund wrote: > > This does not look right to me. There's a 1.2 directory inside > tags/1.2.14.19 for some reason? > > On Tue, Sep 22, 2009 at 9:39 AM,

[v8-dev] Re: Generate inline code for contextual loads on ARM.... (issue5140002)

Will do. On Thu, Nov 18, 2010 at 15:54, wrote: > Søren, could you have a look at this? > > > http://codereview.chromium.org/5140002/ > -- v8-dev mailing list v8-dev@googlegroups.com http://groups.google.com/group/v8-dev

[v8-dev] Re: Optimize instanceof further... (issue5990005)

Now it does (lint that is). /Søren On Wed, Dec 22, 2010 at 14:38, wrote: > Does it lint? ;) > > I'll perform a real review ASAP. > > > http://codereview.chromium.org/5990005/ > -- v8-dev mailing list v8-dev@googlegroups.com http://groups.google.com/group/v8-dev

Re: [v8-dev] current work on lithium stubs

Alexandre, Thank you for looking int completing the crankshaft ARM port. In r6155 the following went in DoJSArrayLength DoLoadNamedField DoCallConstantFunction DoBoundsCheck DoCheckPrototypeMaps We will keep from working on the rest you mentioned. Regards, Søren On Tue, Jan 4, 2011

[v8-dev] Re: Misc debugger enhancements and bug fixes.... (issue5980006)

I had to revert this again as it caused test failures, see http://build.chromium.org/p/client.v8/builders/V8%20Linux/builds/343/steps/Check/logs/stdio. It was reverted in r6182. Regards, Søren On Wed, Jan 5, 2011 at 14:57, wrote: > LGTM > > We are not linting the line lengths of the JS files, s

[v8-dev] Re: Misc debugger enhancements and bug fixes.... (issue5980006)

r6182 -> r6183 On Wed, Jan 5, 2011 at 15:06, Søren Gjesse wrote: > I had to revert this again as it caused test failures, see > http://build.chromium.org/p/client.v8/builders/V8%20Linux/builds/343/steps/Check/logs/stdio. > It was reverted in r6182. > > Regards, > Søren >

Re: [v8-dev] Implementing Math.floor and Math.sqrt for crankshaft. (issue6250002)

Please do that. I think that if we only have double_scratchX(), and use .high() or .low() for single precision registers the overlapping will be quite explicit. Regards, Søren On Thu, Jan 13, 2011 at 14:38, wrote: > On 2011/01/13 12:33:06, Søren Gjesse wrote: > >> On 2011/01/13 12

[v8-dev] Re: Add function for setting breakpoint by scriptId and position to debugger script (issue6301001)

OK, will this be tested in some of the DevTools tests? Otherwise this might get broken without anybody noticing. /Søren On Thu, Jan 13, 2011 at 18:32, wrote: > On 2011/01/13 15:46:14, Søren Gjesse wrote: > >> LGTM, but the function is not used anywhere. >> > > It w

[v8-dev] Re: Add function for setting breakpoint by scriptId and position to debugger script (issue6301001)

d by layout tests. > > BR, > Pavel Podivilov > > > > On Thu, Jan 13, 2011 at 11:40 PM, Søren Gjesse wrote: > >> OK, will this be tested in some of the DevTools tests? Otherwise this >> might get broken without anybody noticing. >> >> /Søren >>

[v8-dev] Re: Fix Windows compilation (issue6056006)

No, that should not have worked, but it seems that Visual Studio have some strncpy_s overloads behind its back. I changed it to use the OS::StrNCpy in http://code.google.com/p/v8/source/detail?r=6382. Thanks for letting us know. Regards, Søren On Mon, Jan 17, 2011 at 21:04, wrote: > Does this a

[v8-dev] Re: ARM: Implement DoHasCachedArrayIndex and DoHasCachedArrayIndexAndBranch. (issue6519009)

14, 2011 at 13:01, wrote: > Reviewers: Søren Gjesse, > > Description: > ARM: Implement DoHasCachedArrayIndex and DoHasCachedArrayIndexAndBranch. > > Please review this at http://codereview.chromium.org/6519009/ > > SVN Base: https://v8.googlecode.com/svn/branches/

[v8-dev] Re: Enable mksnapshot on ARM when cross-compiling using gyp

Committed: http://code.google.com/p/v8/source/detail?r=3235 /Søren On Thu, Nov 5, 2009 at 20:28, Antoine Labour wrote: > > > On Thu, Nov 5, 2009 at 4:14 AM, wrote: > >> LGTM >> >> Thank you for the explanation on the gyp cross-compile details. > > > Cool. > > Since I don't have commit access,

[v8-dev] Re: Skip debugger frames when looking for calling context

review.chromium.org/398002/diff/4002/4003 >> > File test/cctest/test-debug.cc (right): >> > >> > http://codereview.chromium.org/398002/diff/4002/4003#newcode5674 >> > Line 5674: // Create a function for testing beak. >> >

[v8-dev] Re: initial v8 openbsd support (issue431047)

As far as I can see this is also the way we have organized the platform specific parts of V8. We added the platform_posix.cc after having duplicated quite a lot in _linux, _mac and _freebsd. This changelist follows this pattern. On Thu, Nov 26, 2009 at 12:10, wrote: > On 2009/11/25 22:24:20, Ev

[v8-dev] Re: initial v8 openbsd support (issue431047)

Thank you for the patch. Before we can add this code to the repository you need to sign the Google Individual Contributor License Agreement, see http://code.google.com/p/v8/wiki/Contributing for more details. Regards, Søren On Thu, Nov 26, 20

Re: [v8-dev] Re: initial v8 openbsd support (issue431047)

using scons? $ scons sample=shell $ scons d8 As "Google Inc." is in the AUTHORS file you don't need to sign anything. Regards, Søren On Tue, Dec 1, 2009 at 23:08, Evan Martin wrote: > 2009/12/1 Peter Valchev : > > 2009/11/26 Søren Gjesse > >> > >> Thank

Re: [v8-dev] Re: Create the framework for adding a thumb2 backend for ARM. So far, it... (issue464070)

We should consider reflecting the SCons changes in the v8.gyp file as well. /Søren On Wed, Dec 9, 2009 at 10:55, wrote: > Reviewers: Erik Corry, andreip_google.com, > > Description: > Create the framework for adding a thumb2 backend for ARM. So far, it > is exactly the same as the ARM one. The

[v8-dev] Re: Implement "wait for connection" feature (issue489005)

Yes, v8::Debug::DebugBreak() will set the flag. /Søren On Thu, Dec 10, 2009 at 23:04, wrote: > > Should the debug break flag be raised to >> > ensure that a break will happen as soon as any code is executed? >> I didn't think about this. I though about manually putting "debugger;" >> > statemen

[v8-dev] Re: Extend the maximum size map space... (issue504026)

Mads have tried to run with the patch and it solved that particular problem. I was not able to reproduce the problem myself. /Søren On Wed, Dec 16, 2009 at 16:18, wrote: > LGTM > > just curious: do you know if it fixes the Chromium issue w/ dashboard? > > > http://codereview.chromium.org/504026

Re: [v8-dev] Re: Force mark sweep if size of map space is to big to allow forward pointers enc... (issue507025)

And Anton please go ahead with an implementation /Søren On Fri, Dec 18, 2009 at 08:41, Kevin Millikin wrote: > On Thu, Dec 17, 2009 at 11:18 PM, wrote: > >> >> Anyway, I'd love to implement this map compaction. >> >> > Hi Anton. What you described would work as well, but it's three passes > (

[v8-dev] Re: V8 debugger and contexts

I have nothing to add to the accurate answers Yury provided, but Yury can you provide some details on the security related problems you are working on solving? /Søren On Wed, Jan 6, 2010 at 10:37, Yury Semikhatsky wrote: > Resending this question to the public mailing list. It may be of interes

Re: [v8-dev] Re: V8 debugger and contexts

Peter, Please note that there is a V8 context for each tab, and for each frame. Whether it is possible to access data in one context from another controlled through the V8 security. Through the API this can be controlled both through tokens and security callbacks. Chrome implements the web model u

[v8-dev] Re: Implementing ProcessDebuggerRequests method

Hi Peter, It will good to get rid of having to execute some JavaScript in order to process debugger commands. It might not be necessary to actually execute any JavaScript in order to process the debugger requests. We can also enter the debugger when there is a compile error (e.g syntax error) in

[v8-dev] Re: Debugger expressions evaluation when V8 is idle

Hi Peter What you are suggesting is that when there is no active JavaScript execution and the embedder calls v8::Debug::ProcessDebugMessages() the embedder can enter a context before the call, and the commands will be processed in this context. I think that sounds fine. However I am not sure wheth

Re: [v8-dev] Evaluate with breaks protocol issue

I agree that adding evaluation with breaks enabled is something that should work well. Instead of introducing another command how about any evaluate with breaks enabled does an implicit continue. This continue will run the evaluation expression, and the compiled code to the evaluation expression wi

[v8-dev] Re: Debugger expressions evaluation when V8 is idle

be useful, when id's became too large (with many digits) and thus > inconvenient. > Yes, and we might have to add some cap in the internal cache as well and flush (some of) it if it uses to much memory. > > Peter > > > >> >> /Søren >> >> On Mon, J

Re: [v8-dev] Visual Studio build files in V8's tree

Well, I use them when building and debugging V8 on Windows, which is not as frequent as it used to be. I think there should be a simple way to work work with V8 in Visual Studio, and making a set of project files that will work is not a simple task, so I vote for keeping them. There is also a vers

Re: [v8-dev] Visual Studio build files in V8's tree

As of r3722 the Visual Studio project files should be usable again. Regards, Søren On Tue, Jan 26, 2010 at 21:31, Søren Gjesse wrote: > Well, I use them when building and debugging V8 on Windows, which is not as > frequent as it used to be. > > I think there should be a simple way

[v8-dev] Re: Port direct call from JavaScript to native RegExp to x86... (issue548179)

Typo, x86 -> x64. On Thu, Jan 28, 2010 at 11:33, wrote: > Reviewers: Lasse Reichstein, > > Description: > Port direct call from JavaScript to native RegExp to x86 > > Code tested on both Linux and Windows. > > Added a bit more abstraction to calling a C function from generated code. > > Minor tw

[v8-dev] Re: Rewrite CsvParser.parseLine to make it simpler and gain some performance improvement. (issue548198)

#x27;)); > On 2010/01/29 13:21:34, Søren Gjesse wrote: > >> Is this covered by any tests? >> > > You mean double quotes? Lines 53--63 in test/mjsunit/tools/csvparser.js > > > http://codereview.chromium.org/548198 > -- v8-dev mailing list v8-dev@googlegroups.com http://groups.google.com/group/v8-dev

[v8-dev] Re: Show user script source line that caused exception intead of a line in a nati... (issue555170)

bably it's not that bad to show builtin >> > functions > >> in the stack trace since in the case of stack trace user is still able to >> > track > >> the source of error (unlike error message handler where the stack may not >> be >> available). >

Re: [v8-dev] ARMv7 ubfx support (issue569015)

Thank you for the patch to the V8 ARM code. However, you will need to sign the Google Individual Contributor License Agreement before we can use it. You will find the CLA on http://code.google.com/legal/individual-cla-v1.0.html, where you can sign it electronically. Regards, Søren On Tue, Feb 2,

[v8-dev] Re: That is a patch for v8 debug api.... (issue565007)

I think looking at performance is the right thing to do. Please try to use the GetMirror API in the code in V8Proxy.cpp, before adding devtools specific functions to the V8 debugger API. Remember that you can compile the JavaScript function taking the FunctionMirror object once. If using the mirror

[v8-dev] Re: That is a patch for v8 debug api.... (issue565007)

HECK_EQ(3, result->Int32Value()); I suggest you return a JavaScript object which contains all the information needed that is something like function functionInfo(mirror) { return { script: script().name(), line: mirror.sourceLocation().line } } Look in src/mirror-delay.js to see wha

  1   2   >