Embedding v8 6.7.
I found the following test in test-modules.cc:
CHECK(module->Evaluate(env.local())
.ToLocalChecked()
->StrictEquals(v8_str("gaga")));
CHECK_EQ(Module::kEvaluated, module->GetStatus());
Embedding v8 6.7.
Would like to use the ICU library built with custom flags and run v8 unit
tests. What is the best way to do that?
Thanks in advance.
--
--
v8-users mailing list
v8-users@googlegroups.com
http://groups.google.com/group/v8-users
---
You received this message because you are
Is there a way to build static libraries of icu from v6.7 version of v8?
If so, how?
--
--
v8-users mailing list
v8-users@googlegroups.com
http://groups.google.com/group/v8-users
---
You received this message because you are subscribed to the Google Groups
"v8-users" group.
To unsubscribe
perties on the module
namespace object, so that each module has its own isolated variable
bindings. Is that possible?
On Wednesday, January 2, 2019 at 9:55:26 PM UTC-8, Jane Chen wrote:
>
> Hi Ben,
>
> I checked out Module API on v8 6.7 and it was working great, except one
> pr
Referring to test-modules.cc in v8 6.7, is the following part of the module
standard?
import {a, b} from 'export var a = 1; export let b = 2';
If so, how do I differentiate a module definition from a module name/path?
--
--
v8-users mailing list
v8-users@googlegroups.com
to do that?
Thanks a lot!
On Saturday, May 12, 2018 at 4:38:15 AM UTC-7, Ben Noordhuis wrote:
>
> On Sat, May 12, 2018 at 6:42 AM, Jane Chen > wrote:
> > Thanks Ben.
> >
> > I want to re-use my context for performance reasons, but some JavaScript
> > progra
alse)
}
With symbol_level=1, the size for libv8.so is only slightly higher than no
symbols: 24910336
Thanks again!
On Thursday, December 27, 2018 at 4:39:19 PM UTC-8, Jane Chen wrote:
>
> Embedding v8 6.7 using gcc 4.8.5 on RedHat 7.
>
> Building v8 optimized build with gn and ninj
That worked! Thanks!
The size difference is big. For example:
libv8.so, version 6.7, with debug info, 565773459, without debug info,
15134520.
On Thursday, December 27, 2018 at 4:39:19 PM UTC-8, Jane Chen wrote:
>
> Embedding v8 6.7 using gcc 4.8.5 on RedHat 7.
>
> Building
Attached patches to shell.cc.
On Friday, December 28, 2018 at 12:15:48 PM UTC-8, Jane Chen wrote:
>
> Embedding v8 6.7, open to upgrade if necessary.
>
> I modified v8 shell.cc to experiment with the new Module API and am trying
> to figure out how to handle error conditions
Embedding v8 6.7, open to upgrade if necessary.
I modified v8 shell.cc to experiment with the new Module API and am trying
to figure out how to handle error conditions while resolving a module.
If the resolve callback calls v8::Isolate::throwException, I get:
#
# Fatal error in
Embedding v8 6.7 using gcc 4.8.5 on RedHat 7.
Building v8 optimized build with gn and ninja with the following arguments:
gn gen out.gn/x64.release --args='is_debug=false is_official_build=true
is_component_build=true is_cfi=false is_clang=false use_custom_libcxx=false
use_sysroot=false
When I build v8 with gn and ninja on RedHat7, I'd like to specify the
location of gcc so that it doesn't use the default one. Is it something
specifiable through the gn configuration?
Thanks!
--
--
v8-users mailing list
v8-users@googlegroups.com
http://groups.google.com/group/v8-users
---
Thanks Ben!
Last time when we upgraded to 5.3, we used gcc 4.8 on Linux, VS 2015 on
Windows to compile v8 and our own product. Do we need to use a higher
version of compiler this time?
On Monday, June 4, 2018 at 1:09:33 AM UTC-7, Ben Noordhuis wrote:
>
> On Mon, Jun 4, 2018 at 2:36 AM
I am making plans to upgrade v8 for our next major release. We are
currently running 5.3. It looks like that the current stable release is
6.7. Is this a good version to upgrade to? Is the module API ready to be
used? Any significant incompatibilities from version 5.3 that we should be
enclosing the
script with {}. Do you see any issue with doing it? Is there any better
way to do so?
On Friday, May 11, 2018 at 5:50:07 AM UTC-7, Ben Noordhuis wrote:
>
> On Thu, May 10, 2018 at 10:24 PM, Jane Chen <jxch...@gmail.com
> > wrote:
> > Embedding v8 v5.3.
>
Embedding v8 v5.3.
Is there any way to reset global object for a context so that the context
can be re-used?
How is Context::DetachGlobal() is meant to be used?
Thanks,
Jane
--
--
v8-users mailing list
v8-users@googlegroups.com
http://groups.google.com/group/v8-users
---
You received this
, Ben Noordhuis wrote:
>
> On Thu, Jul 13, 2017 at 12:40 AM, Jane Chen <jxch...@gmail.com
> > wrote:
> > Embedding v8 5.3.
> >
> > Seeing a crash if I have weak handles in many isolates:
> >
> > Thread [51] 27257 [core: 6] (Suspended : Signal : SI
Embedding v8 5.3.
Seeing a crash if I have weak handles in many isolates:
Thread [51] 27257 [core: 6] (Suspended : Signal : SIGILL:Illegal
instruction)
v8::base::OS::Abort() at platform-posix.cc:240 0x75f078a1
V8_Fatal() at logging.cc:116 0x75f02d88
I used to do it without specifying the file path for icudtl.dat. Now I
realize it's required. So passing in the file path fixed it.
Thanks a lot!
On Sunday, March 12, 2017 at 12:55:51 AM UTC-8, Ben Noordhuis wrote:
>
> On Sun, Mar 12, 2017 at 4:03 AM, Jane Chen <jxch...@gmail.com
Embedding v8 5.3.
This causes v8 to crash:
().toLocaleString();
2017-03-07 08:24:07.362 Critical: # Fatal error in , line 0
2017-03-07 08:24:07.362 Critical: # Failed to create ICU number format, are ICU
data files missing?2017-03-07 08:24:07.362 Critical: #2017-03-07 08:24:07.362
Why does v8::Object::Set() takes a context parameter now? Does it mean
that if an object is created in one context, it can only be used in that
context?
Thanks.
--
--
v8-users mailing list
v8-users@googlegroups.com
http://groups.google.com/group/v8-users
---
You received this message
I found the memory leak is caused by the following two flags: gdbjit_full
and harmony rest parameters. Removing them fixed the problem.
Thanks Ben for your help.
On Wednesday, December 14, 2016 at 11:36:06 PM UTC-8, Jane Chen wrote:
>
> I did. I didn't find anything related to isolate di
On Wed, Dec 14, 2016 at 2:47 AM, Jane Chen <jxch...@gmail.com
> > wrote:
> > For experiment, I just changed my implementation to always dispose and
> > recreate isolates after each JavaScript execution. I can easily
> reproduce a
> > memory leak and my server proce
of requests.
On Tuesday, December 13, 2016 at 10:35:40 AM UTC-8, Jane Chen wrote:
>
> Thanks Ben. I'm testing this on Linux. RH7 VM and Fedora 20. It's good
> to know that you haven't seen this yourself.
> I am seeing it when I dispose the isolate only when LowMemory
2929984 8.94 13224 696908 15415752 37.44 1416408
386944 0
I'll try with a different test to dispose and recreate isolates that don't
have issues with GC. I'll report back.
On Tuesday, December 13, 2016 at 4:34:39 AM UTC-8, Ben Noordhuis wrote:
>
> On Tue, Dec 13, 2016 at 3:15 AM, Jane Chen
environment, and I haven't been able to
reproduce with a debug build.
On Thursday, December 8, 2016 at 2:14:32 AM UTC-8, Ben Noordhuis wrote:
>
> On Thu, Dec 8, 2016 at 12:35 AM, Jane Chen <jxch...@gmail.com
> > wrote:
> > Just upgraded v8 from 4.6 to 5.3 and found a n
Just upgraded v8 from 4.6 to 5.3 and found a new behavior: garbage
collection can happen in a background thread after
ContextDisposedNotification and IdleNotificationDeadline return. My
idleNotificationDeadline runs in a loop with 1sec increments, and it
typically comes out after 1 or 2
Embedding v8 5.3. Seeing a segfault in our stress test:
Critical: #6 0x7fbf3f6c0173 in
v8::internal::GlobalHandles::Node::Release() () from lib/libv8.so
This happens when v8::Persistent::Reset is called.
Has anyone else seen this? Any clues on what might be wrong?
Thanks.
--
--
Embedding v8 4.6.88.
Got a segfault during stress test of my app:
2016-11-19 20:54:26.933 Critical: Segmentation fault in thread
0x7efbf9400700 2016-11-19 20:54:26.933 Critical: Thread 68 (Thread
0x7efbf9400700 (LWP 6602)): 2016-11-19 20:54:26.933 Critical: #0
0x00311ca0f279 in
Embedding v8 5.3.
I have a JavaScript function prof.eval which is implemented through a
callback function that creates a fresh v8 context, starts CPU profiling,
evaluates the inner script, stops profiling and returns profiling result.
The profiling result contains frames calling prof.eval in
I'm facing a similar issue on Mac. Any pointers to the mac build script
which includes any incompatible dependency?
Thanks!
On Saturday, November 12, 2016 at 9:21:32 PM UTC-8, Jane Chen wrote:
>
> Thanks Jochen. This did it for me:
>
> diff --git a/gypfiles/standalone.gyp
vember 12, 2016 at 2:36:12 PM UTC-8, Jochen Eisinger wrote:
>
> We're working on upgrading the sysroot, however, I'm not aware of any
> documentation on how to do that, sorry
>
> On Fri, Nov 11, 2016 at 3:16 PM Jane Chen <jxch...@gmail.com >
> wrote:
>
>> I now
:
>
> This is because the sysroot (in build/linux/) is outdated. I filed
> crbug.com/661506 to track this
>
> On Tue, Nov 1, 2016 at 4:41 PM Jane Chen <jxch...@gmail.com >
> wrote:
>
>> I'm investigating a crash in a debug build of our app embedding v8
>> 5.
Thanks Ben. You were exactly right. Fixing it now...
On Friday, November 11, 2016 at 12:35:56 AM UTC-8, Ben Noordhuis wrote:
>
> On Fri, Nov 11, 2016 at 5:40 AM, Jane Chen <jxch...@gmail.com
> > wrote:
> > Embedding v8 5.3. Running code to create a context and execute the
Embedding v8 5.3. Running code to create a context and execute the same
JavaScript code in a loop. After thousands of iterations, I get "Exception
thrown during bootstrapping". Here's the stack:
51 v8::internal::ReportBootstrappingException()
/space/projects/v8-git/v8/src/isolate.cc:960
gt;
> On Tue, Nov 1, 2016 at 4:41 PM Jane Chen <jxch...@gmail.com >
> wrote:
>
>> I'm investigating a crash in a debug build of our app embedding v8
>> 5.3-lkgr and notice libstdc++ from gcc4.6 is included and referenced from
>> some of the makefiles to build ICU and
deed also
> changed the implementation of stack traces a bit between 4.6 and 5.3. A
> (preferably reduced) repro would be great to find out what exactly is going
> on.
>
> Cheers,
>
> Yang
>
>
>
> On Tuesday, November 1, 2016 at 4:19:35 PM UTC+1, Jane Chen wro
I'm investigating a crash in a debug build of our app embedding v8 5.3-lkgr
and notice libstdc++ from gcc4.6 is included and referenced from some of
the makefiles to build ICU and v8, although v8 5.3-lkgr requires gcc 4.8 or
above to build on Linux. Am I missing something while building v8?
Embedding v8 5.3. The embedding application allows JavaScript programs to
import other library modules similar to Node.js require. I noticed a
change of behavior after upgrading from v8 4.6.88 to 5.3. That is, if the
imported library module is in strict mode, the stack trace of a function
Embedding v8 5.3. In my embedding application, the user can make some
updates and the updates are committed as one transaction when the script
exits. When one statement throws an error which is caught by a try-catch
block, the scope of the try-catch block determines the aborted changes.
I figured out how to build with gyp now. Would still be nice to know how
to link libv8_libplatform.lib and libv8_libbase.lib with the gn approach.
Thanks.
On Sunday, October 2, 2016 at 3:12:02 PM UTC-7, Jane Chen wrote:
>
> I specified "is_component_build=true" on 5.3-lkgr. I
o generate
the .lib files out of the .obj files under v8_base etc. Any suggestions?
On Thursday, September 22, 2016 at 1:04:48 AM UTC-7, Jochen Eisinger wrote:
>
>
>
> On Thu, Sep 22, 2016 at 1:22 AM Jane Chen <jxch...@gmail.com >
> wrote:
>
>> Trying to bui
I also built v8 5.3-lkgr using gn, and I also needed to build
v8_libbase.lib and v8_libplatform.lib for embedding purpose. The problem
is I don't find samples.sln anywhere. How do I generate samples.sln? Or
more directly, how am I supposed to generate the .lib files from the .obj
files?
On
It does. --gdbjit_full gives me the complete stack. Thanks again!
On Wednesday, September 28, 2016 at 12:09:52 AM UTC-7, Ben Noordhuis wrote:
>
> On Sat, Sep 24, 2016 at 2:40 AM, Jane Chen <jxch...@gmail.com
> > wrote:
> > Thanks, Ben!
> >
> > I added tha
LazyCompile:~isPrime :1() 0x332193c6b884
I suppose that's as good as it gets, right? Any performance penalties to
leave it on always?
On Friday, September 23, 2016 at 11:42:08 AM UTC-7, Ben Noordhuis wrote:
>
> On Fri, Sep 23, 2016 at 1:10 AM, Jane Chen <jxch...@gmail.com
UTC-7, Ben Noordhuis wrote:
>
> On Thu, Sep 22, 2016 at 2:41 AM, Jane Chen <jxch...@gmail.com
> > wrote:
> > Embedding v8. Currently on 4.6.88, and in the process of upgrading to
> > 5.3.lkgr.
> >
> > When I do a pstack of my program embedding v8, some o
Embedding v8. Currently on 4.6.88, and in the process of upgrading to
5.3.lkgr.
When I do a pstack of my program embedding v8, some of the v8 frames show
up as ???. I wonder whether there's a build flag I could use to enable the
full stack trace of v8 frames.
Please advise.
Thanks.
--
Trying to build v8 5.3-lkgr with gn for embedding use.
When I build with GYP, I need to set "-Dv8_use_external_startup_data=0".
Is there an equivalent setting flag for gn?
Also if I build with gn and I need to link with libv8_libplatform.a,
libv8_libbase.a etc, is there a flag to generate
I'm embedding v8 and trying to upgrade to v8 5.3-lkgr. After I built v8
using GYP with GYPFLAGS="-Dv8_use_external_startup_data=0
-Dcomponent=shared_library", I got the following files under
out/x64.release/obj.target/src/:
js2c.stamp libv8_nosnapshot.a
libv8_base.a
Hi Ben,
What's a good low value to set for gc_interval for debugging purpose? 1? 10?
Thanks!
On Friday, September 2, 2016 at 3:57:20 PM UTC-7, Ben Noordhuis wrote:
>
> On Fri, Sep 2, 2016 at 1:17 AM, Jane Chen <jxch...@gmail.com >
> wrote:
> > Embedding v8 4.6.88.
>
Embedding v8 4.6.88.
Running a memory-consuming script. Sometimes I get a fatal error callback
due to "process out of memory" which I now handle. Other times v8 just
crashes in CollectGarbage:
2016-08-25 17:25:49.649 Critical: #5
2016-08-25 17:25:49.649 Critical: #6 0x7f71f69d67fd in
I have. The PrintF calls happen before the user callback is called.
With the embedded scenario, it would be desirable to have some API to
suppress or redirect stdout output.
On Tuesday, August 23, 2016 at 9:27:37 PM UTC-7, Ben Noordhuis wrote:
>
> On Wed, Aug 24, 2016 at 2:17 AM, Jan
Embedding v8 4.6.88.
When a JavaScript program executes and runs out of memory, I get some
stdout output from the subject function logging the last few GC and stack
frames.
Is there anyway to either suppress or redirect the output?
Thanks in advance.
--
--
v8-users mailing list
to reproduce.
Are we supposed to always turn on profiling before compiling the script to
be profiled?
On Monday, August 22, 2016 at 2:29:55 PM UTC-7, Jane Chen wrote:
>
> Ben,
>
> Thanks for the tip. I'll look into that option.
>
> In general we are open to upgrade to a later rel
, Ben Noordhuis wrote:
>
> On Sat, Aug 20, 2016 at 2:23 AM, Jane Chen <jxch...@gmail.com
> > wrote:
> > Embedding v8 4.6.88.
> >
> > Seeing a hang on Windows during cpu profiling. The hang happens in any
> > random place calling v8, such as Isol
Embedding v8 4.6.88.
Seeing a hang on Windows during cpu profiling. The hang happens in any
random place calling v8, such as Isolate::New, Context::New etc. Running
optimized build, my stack doesn't show v8 frames.
Not sure whether it's relevant, I captured this stack once during a hang
> I suspect that profiling wasn't enabled at the time the function was
> compiled. In that case, the information required for mapping code positions
> back to line numbers isn't available.
>
> Jane Chen <jxch...@gmail.com > schrieb am Do., 2. Juni 2016,
> 20:40:
>
&
umbers isn't available.
>
> Jane Chen <jxch...@gmail.com > schrieb am Do., 2. Juni 2016,
> 20:40:
>
>> Testing profiling against v8 4.6.88.
>>
>> I have functions that are exposed through accessor callbacks. In the
>> CpuProfileNode, the line and column nu
Testing profiling against v8 4.6.88.
I have functions that are exposed through accessor callbacks. In the
CpuProfileNode, the line and column numbers are 0 for these functions. Is
that expected? Is there anything I can do to get the real line and column
numbers? Is this something fixed in
attempts to LoadLibraryA() that DLL.
>
> On Sat, Feb 27, 2016 at 8:30 PM Jane Chen <jxch...@gmail.com >
> wrote:
>
>> Embedding v8 4.6.88.
>>
>> On windows, I get a crash when I run the following:
>>
>> new Intl.Collator().compare('a','c')
>>
>&
Embedding v8 4.6.88.
On windows, I get a crash when I run the following:
new Intl.Collator().compare('a','c')
#
# Fatal error in , line 0
# Failed to create ICU collator, are ICU data files missing?
#
In my server startup code, I have:
v8::V8::InitializeICU();
I didn't specify an ICU data
I'm getting
warning: ignoring return value of function declared with warn_unused_result
attribute [-Wunused-result]
obj->Set(isolate->GetCurrentContext(),
^~~~ ~
Presumably I should get the return result of Object::Set. But what shall I
do if it
ke you suggest can just call "set" on themselves and they'll
> be transformed to data properties.
>
> Does that work?
> Toon
>
> On Fri, Oct 30, 2015 at 10:36 PM Jane Chen <jxch...@gmail.com
> > wrote:
>
>> If it was only a performance concer
nate the thread depends on the threading
> library you use. Note that you'll also have to free up all C++ memory
> you've allocated, because exiting the Isolate won't invoke weak callbacks
> etc..
>
> On Tue, Jan 26, 2016 at 3:13 AM Jane Chen <jxch...@gmail.com >
> wrot
Thu, Jan 21, 2016 at 11:59 PM Jane Chen <jxch...@gmail.com
> > wrote:
>
>> Yes, I do get fatal errors like these from v8, but my fatal error
>> callback is never invoked. I register my callback after I create the
>> isolate like this:
>>
>> isolat
Now that we are on this, can I ask more questions about GC:
Is calling LowMemoryNotification() expensive? Should I only do it when I
absolutely need more memory to continue, or should I do it proactively and
in small batches?
On Friday, January 22, 2016 at 11:41:08 PM UTC-8, Jane Chen wrote
On Friday, January 22, 2016 at 12:54:25 AM UTC-8, Jochen Eisinger wrote:
>
> would you mind filing a bug for this? Ideally, with a self-contained repro
> case!
>
> thanks
> -jochen
>
> On Fri, Jan 22, 2016 at 12:23 AM Jane Chen <jxch...@gmail.com
> > wrote:
>
>
> 11: 0x5d4554
> 12: 0x5d42e8
> 13: 0x5a98a5
> 14: 0x685bea
> 15: 0x6836df
> 16: 0x6a5f82
> 17: 0x59017f
> 18: 0x91a56a
> 19: 0x919c7d
> 20: 0x5bf3e806458
> Illegal instruction (core dumped)
>
>
> On Tue, Jan 19, 2016 at 1:05 AM Jane Chen <jxch...@gmail.com
Embedding v8 4.6.88.
Suppose my native function sequence() returns a JavaScript iterable
wrapping a long sequence of internal objects; function gc() invokes
v8::Isolate::LowMemoryNotification(). To reduce memory consumption of the
iterable, gc() needs to be called periodically to release the
ght be better than
> crashing, although I'm not sure whether you won't run into other problems...
>
>
>
> On Thu, Jan 14, 2016 at 6:49 AM Jane Chen <jxch...@gmail.com >
> wrote:
>
>> Unfortunately, the following link is no longer valid:
>>
>> https://code.goog
We embed v8 version 4.6.88.
Looking at flag-definitions.h, I know some ES6 features are tracked by a
flag and can be turned on/off. How do I get a list of features that are
always turned on and not tracked by a flag, such as Iterator, Generator and
Symbols, for example? Is there a
Once the .sln and .vcxproj files are generated through python, how is one
supposed to re-generate them with different gyp flags? When I run:
python build/gyp_v8 -Dtarget_arch=x64 -Dv8_use_external_startup_data=0
for example, I get "updating project files...", but then only to find out
v8 on Windows has the default _ITERATOR_DEBUG_LEVEL value, which conflicts
with another Windows library I'm embedding. So I'd like to set the
_ITERATOR_DEBUG_LEVEL to 0 while building v8. I haven't quite figured out
how to do it either by editing v8.gyp or the generated VS solution files.
I added it to toolchain.gypi and that seems to do it.
On Monday, December 28, 2015 at 2:45:46 PM UTC-8, Jane Chen wrote:
>
> v8 on Windows has the default _ITERATOR_DEBUG_LEVEL value, which conflicts
> with another Windows library I'm embedding. So I'd like to set the
> _ITERATOR
Ben,
Manually adding WinMM.lib to my project did it for me. Thanks a lot.
Jane
On Tuesday, December 22, 2015 at 3:09:48 AM UTC-8, Ben Noordhuis wrote:
>
> On Tue, Dec 22, 2015 at 6:20 AM, Jane Chen <jxch...@gmail.com
> > wrote:
> > I was able to build v8 4.6.88 successful
It's intermittent for me somehow in cygwin. Even after I set
DEPOT_TOOLS_WIN_TOOLCHAIN=0. But what worked for me is to run "fetch v8"
in Windows PowerShell. Still a mystery...
On Tuesday, December 22, 2015 at 5:43:22 AM UTC-8, Maxime Marousez wrote:
>
> set DEPOT_TOOLS_WIN_TOOLCHAIN=0 before
I was able to build v8 4.6.88 successfully with Visual Studio 2013 with the
following flags:
-Dtarget_arch=x64 -Dv8_use_external_startup_data=0
I got the following libraries:
$ ls -l build/Release/lib
total 353468
-rwxr-xr-x 1 jchen Domain Users 3140002 Dec 21 17:20 gmock.lib
-rwxr-xr-x 1
This seems only a problem with debug build. Optimized build works fine.
On Friday, December 11, 2015 at 2:05:20 PM UTC-8, Jane Chen wrote:
>
> 3.24 crashes too. Same JavaScript query, but different stack:
>
> C stack trace ===
>
> 1: V8_Fatal
I am seeing a SIGILL crash too with my embedding app during profiling.
Here's the stack:
Thread [134] (Suspended: Signal 'SIGILL' received. Description: Illegal
instruction.)
40 v8::base::OS::Abort()
/space/projects/v8-git/v8/src/base/platform/platform-posix.cc:229
0x764bcd11
internal::FunctionCallbackArguments::Call(void
(*)(v8::FunctionCallbackInfo const&))
On Friday, December 11, 2015 at 1:46:16 PM UTC-8, Jane Chen wrote:
>
> I am seeing a SIGILL crash too with my embedding app during profiling.
> Here's the stack:
>
> Thread [134] (Suspende
Thanks a lot! It sure helps.
I found CpuProfileNode.GetNodeId(). Maybe that's the pre-computed id which
is the same as what you show here?
On Wednesday, December 9, 2015 at 2:09:49 PM UTC-8, Ben Noordhuis wrote:
>
> On Wed, Dec 9, 2015 at 8:55 PM, Jane Chen <jxch...@gmail.com &
t 2:32 AM, Jane Chen <jxch...@gmail.com >
> wrote:
> > I'm trying to serialize the CpuProfile object to a JSON serialization
> > compatible to Chrome's .cpuprofile file. The part I'm not clear about
> is
> > the samples array in the .cpuprofile file:
> >
Is my understanding correct that turning on v8 debugging is global for all
isolates? Is there a way to turn on v8 debug for just one isolate? What
is the performance implication to turn on v8 debug?
Thanks in advance.
--
--
v8-users mailing list
v8-users@googlegroups.com
Ben,
In v8 4.6.88, I can no longer find src/debug-debugger.js, nor
src/{liveedit,mirror}-debugger.js. Does it mean that the function-based
API is also deprecated and removed?
Jane
On Saturday, March 14, 2015 at 3:53:55 AM UTC-7, Ben Noordhuis wrote:
>
> On Sat, Mar 14, 2015 at 8:30 AM, Deguo
to the embedder so it is easier to do "spot
> consulting'.
>
> What are you planning to do?
>
> On Monday, November 9, 2015 at 11:58:14 PM UTC+1, Jane Chen wrote:
>>
>> Just went to this link:
>>
>> https://chromium.googlesource.com/v8/v8/+/master/docs/de
Just went to this link:
https://chromium.googlesource.com/v8/v8/+/master/docs/debugger_protocol.md
and found that "*The message based API is no longer maintained.*"
Does that mean that users (or embedders) should no longer use the message
based API since it is no longer supported?
On
With the change of behavior in Iterator discussed here:
https://groups.google.com/forum/#!topic/v8-users/8LkrofVr0aA
If I have an iterator (natively defined) which supports for ... of loop,
what is the equivalent of calling next to advance it?
I hope the following js example demonstrates what
4.6.88. Just tried it in v8 shell. myIterable[Symbol.iterator].next is
undefined.
On Tuesday, November 3, 2015 at 3:40:07 PM UTC-8, Ben Noordhuis wrote:
>
> On Wed, Nov 4, 2015 at 12:17 AM, Jane Chen <jxch...@gmail.com
> > wrote:
> > With the change of behavior in It
t; http://tc39.github.io/ecma262/#sec-getiterator
>
> On Tue, Nov 3, 2015 at 5:45 PM, Jane Chen <jxch...@gmail.com
> > wrote:
>
>> 4.6.88. Just tried it in v8 shell. myIterable[Symbol.iterator].next is
>> undefined.
>>
>> On Tuesday, November 3, 2015 at
> accessor. But I assume you're doing that?
>
> On Friday, October 30, 2015 at 3:03:27 PM UTC-7, Jane Chen wrote:
>>
>> Is there anyway to use something like CreateDataProperty to override the
>> accessor so that the actual function is associated with the property,
&
ntiation
> would make perfect sense.
>
> On Saturday, October 31, 2015 at 11:03:10 AM UTC-7, Jane Chen wrote:
>>
>> Yes, the first time the property is accessed, it gets the correct native
>> function. In my accessor before I return, I have:
>>
>
stuff like dynamically loading the functions or maybe compiling JavaScript
> callbacks with your function or whatever, in which case lazy instantiation
> would make perfect sense.
>
> On Saturday, October 31, 2015 at 11:03:10 AM UTC-7, Jane Chen wrote:
>>
>> Yes, the first time th
rty produced behavior similar to what you're seeing -- my
> accessor was entered repeatedly.
>
> Is ForceSet just silently failing for you? Is it returning false?
>
> On Saturday, October 31, 2015 at 4:11:22 PM UTC-7, Jane Chen wrote:
>>
>> ForceSet worked in 3.24.
he point is that for a single function, it can only infer a single
> name.
>
> I think this is working as intended.
>
> On Thu, Oct 29, 2015 at 11:57 PM, Jane Chen <jxch...@gmail.com
> > wrote:
>
>> A test case to demonstrate the issue can be found at:
>>
>>
>
> On Tue, Oct 6, 2015, 21:57 Jakob Kummerow <jkum...@chromium.org
> > wrote:
>
>> Have you checked bit.ly/v8-api-changes ?
>>
>> On Tue, Oct 6, 2015 at 8:44 PM, Jane Chen <jxch...@gmail.com
>> > wrote:
>>
>>> In v8 4.7.0, Object.ForceSet
Testing profiling against v8 4.6.88.
I have functions that are exposed through accessor callbacks. If a native
function is set for multiple accessor properties, the last property shows
up as the function name in the CpuProfileNode, regardless what name you set
to the function returned by the
A test case to demonstrate the issue can be found at:
https://code.google.com/p/v8/issues/detail?id=4527
On Thursday, October 29, 2015 at 3:46:21 PM UTC-7, Jane Chen wrote:
>
> Testing profiling against v8 4.6.88.
>
> I have functions that are exposed through accessor callbacks.
Embedding v8 3.24. Running stress tests with 30 concurrent threads serving
JavaScript requests on one node. Got a segmentation fault in the 23rd hour
of the run at this line:
v8.h 5546:
template
Local Local::New(Isolate* isolate, const PersistentBase& that) {
return New(isolate,
> best
> -jochen
>
> On Tue, Oct 13, 2015 at 6:41 AM Jane Chen <jxch...@gmail.com >
> wrote:
>
>> Looking at v8 4.7.0, both shell.cc and d8.cc calls:
>>
>> while (v8::platform::PumpMessageLoop(platform, isolate)) continue;
>>
>> after
Looking at v8 4.7.0, both shell.cc and d8.cc calls:
while (v8::platform::PumpMessageLoop(platform, isolate)) continue;
after running a script. What's the purpose of this?
As always, thanks in advance.
--
--
v8-users mailing list
v8-users@googlegroups.com
1 - 100 of 193 matches
Mail list logo
