No, there is no existing way to save and restore V8's state at arbitrary
times. Sven has already explained why it's impossible to implement one.
Assuming that V8 (or a Chrome renderer process) is single-threaded doesn't
help, because it *isn't* single-threaded.
But even if what you're suggesting
https://code.google.com/p/v8/wiki/V8Profiler
On Tue, Dec 17, 2013 at 3:54 PM, Pawel Por porpa...@gmail.com wrote:
Hi
Can I profile javascript with v8 ?
I would like to log the time of execution of specific javascript functions
into file.
I want to have the following information: function
- dictionary backing
store is allocated, all elements are copied over.
Cheers
On Sunday, December 15, 2013 3:49:31 AM UTC-8, Jakob Kummerow wrote:
There, fixed that for ya: http://jsperf.com/preallocated-vs-dynamic-
arrays/3
Pre-allocated vs. dynamically grown is only part of the story
On Mon, Dec 16, 2013 at 5:37 PM, Pawel Por porpa...@gmail.com wrote:
Hi
I have a few questions about V8 virtual machine.
- is there some kind of intermediate byte code or everything is always
compiled into machine native code
No, there is no byte code.
The optimizing compiler does have an
There, fixed that for ya: http://jsperf.com/preallocated-vs-dynamic-arrays/3
Pre-allocated vs. dynamically grown is only part of the story. For
preallocated arrays, 100,000 just so happens to be the threshold where V8
decides to give you a slow (a.k.a. dictionary mode) array.
Also, growing
Thanks for the report, but there's nothing to be done here (or to be
concerned about). That test allocates a contiguous 2GB chunk of memory on
x64 (1GB on ia32). If it can't get that, it'll crash. Try running it in
isolation (just copy-paste the command from the line starting with
Command: ...)
On Sun, Dec 1, 2013 at 11:11 AM, B Lyon bradfl...@gmail.com wrote:
On an iMac, I'm trying a newbie build of v8 I just grabbed. It's treating
unused variables as an error, but I can't get these to be ignored.
Yeah, werror=no isn't implemented for iMac right now.
I created the XCode project
V8 is not an ahead-of-time compiler. Using V8 as a starting point for
developing an ahead-of-time compiled system is going to be a lot of work.
On Mon, Nov 25, 2013 at 11:29 AM, Axel Payan axelpa...@gmail.com wrote:
Hello,
I'm trying to save a compiled Javascript code and create an
[+v8-users]
V8 requires at least the ARMv6 + VFPv2 instruction sets. If I understand
correctly, the Cortex M3 only supports Thumb/Thumb2, so V8 won't run on it.
Generally speaking, running JavaScript is possible on any hardware -- for
example if you use an interpreter-based JS engine instead of
No, a drop in performance after a while does not at all sound like the
optimizer kicking in.
One thing I'd do is stop creating a new context on every iteration. If you
just repeat the script-Run() call, that would be enough, and avoid a
bunch of needless work.
Another thing I'd do is stop using
No.
On Fri, Nov 8, 2013 at 8:56 AM, Michael demiGod dem...@gmail.com wrote:
script:
1 document.write('foo');
2 var str = 'hello' + 'world';
3 ...
cpp:
v8::TryCatch try_catch;
v8::Handle result = script-Run();
result:
test.js:1: Uncaught ReferenceError: document is not defined
run-tests.py can run all the tests in V8's test/ directory. Look at its
--help output for additional options, such as verbose output (printing
every test that's run). You can specify which test suites to run, e.g.
run-tests.py mozilla will just run the mozilla tests.
On Sat, Nov 2, 2013 at 8:17
I think SVN doesn't let you checkout into existing directories. Try doing
it the other way round:
1.) checkout V8 (which automatically creates the v8 directory)
2.) cd v8
3.) checkout GYP
See also further dependencies listed at
https://code.google.com/p/v8/wiki/BuildingWithGYP.
On Fri, Oct 18,
The V8 project does not distribute binaries. You'll have to compile your
own.
On Thu, Oct 17, 2013 at 6:47 AM, shada laf...@gmail.com wrote:
where can i find v8 binary for visual c++?
--
--
v8-users mailing list
v8-users@googlegroups.com
http://groups.google.com/group/v8-users
---
You
I've never heard of anyone trying to cross-compile V8 for ARM on Windows. I
have no idea if it is possible. You're welcome to try it, and contribute
patches and/or documentation if/when you find out anything.
On Mon, Oct 7, 2013 at 3:57 PM, Mihail Slavchev
mihail.slavc...@gmail.comwrote:
Hi
Are you sure you're doing your users (and yourself) a favor by shipping
trunk builds?
On Sat, Oct 5, 2013 at 2:07 PM, Anatol Pomozov anatol.pomo...@gmail.comwrote:
Hi
Thanks everyone, and especially Pawel. I was using gentoo package as a
great source of information.
Here the package
In addition to what has been said already:
1) The V8 build does not automatically pull in either ICU or GYP. You have
to explicitly run make dependencies to download them, which is merely a
convenience feature. So there's no need to turn that off -- simply run
make native or make ia32.release
You can see what version Chrome uses at omahaproxy.appspot.com.
Rule of thumb: the second newest V8 branch is usually the most stable. E.g.
right now that would be 3.20.
On Fri, Sep 27, 2013 at 10:03 AM, wingsky fly flywingsk...@gmail.comwrote:
can anyone tell me which version is stable? and
On Sun, Sep 1, 2013 at 8:25 PM, sihaowu shwu.c...@gmail.com wrote:
Let's say I run a webapp, and I call gc() in my javascript code.
Will the GC be triggered if the user runs my webapp on his side ?
(note that in the user's side, he does'n expose the gc flag)
No, that'll actually trigger a
As Toon said before, you cannot and you should not trigger GCs manually.
On Mon, Sep 2, 2013 at 2:09 PM, sihaowu shwu.c...@gmail.com wrote:
Thanks for your replies. I have another question.
If somehow I want it to do GC from the client side (let's say I just want
to call gc() function),
As always: when in doubt, measure it! Implement several approaches (in a
simplified version of your app if necessary) and see for yourself if any of
the options you're considering makes a difference.
Generally I would say that what applies to objects also applies to
prototypes (as they're objects
[+v8-users; chromium-discuss to BCC]
Hi,
On Mon, Aug 5, 2013 at 2:50 PM, Elias Weingärtner
elias.weingaert...@haufe-lexware.com wrote:
Folks,
we have recently run into troubles, because one of our Web-based software
products crashes the current stable of Chrome, more precicely V8.It is
Well, if you want to compile V8 with local modifications, you should
generally know how to use a debugger to find out what's wrong if something
doesn't work ;-)
In the case at hand, the current implementation of the snapshotting feature
doesn't support large object space, which is the memory
9日星期二UTC+8下午7时19分20秒，Jakob Kummerow写道：
On Tue, Jul 9, 2013 at 12:09 PM, QingHai Hu ustc@gmail.com wrote:
Hi,guys
I want to know the meaning of the v8 benchmark's score.
Higher is better.
Does it depend on the compile-time and the run-time ? Or only dependent
on run-time?
I'm
On Tue, Jul 9, 2013 at 12:09 PM, QingHai Hu ustc.sos...@gmail.com wrote:
Hi,guys
I want to know the meaning of the v8 benchmark's score.
Higher is better.
Does it depend on the compile-time and the run-time ? Or only dependent on
run-time?
I'm not sure what you mean.
C++ compilation is
For the record, the following should be enough:
export CXX=/usr/bin/arm-linux-gnueabihf-g++
export LINK=/usr/bin/arm-linux-gnueabihf-g++
make arm.release -j4 armfloatabi=hard armfpu=vfpv3
On Mon, Jul 8, 2013 at 1:52 PM, Benedikt Naessens th3sink...@gmail.comwrote:
Rodolphe,
I made a simple
See also https://code.google.com/p/v8/wiki/BuildingWithGYP#MinGW.
On Thu, Jun 27, 2013 at 2:34 PM, Mat Sutcliffe oktal3...@gmail.com wrote:
Sorry to necro a year-old thread, but this seemed like the best way to
announce that I've been able to build Node.js in MinGW+MSYS quite easily -
the
Using internal methods is, quite obviously, not supported. Don't do it.
Especially when you don't have a clear understanding of how V8 works
internally. You most certainly don't need to mess with internal methods to
get access to the built-in JSON parser, as that's exposed to JS land anyway.
If
On Thu, Jun 6, 2013 at 11:07 AM, Joran Dirk Greef jo...@ronomon.com wrote:
Hi Hitesh,
I have had pauses of up to a minute with the GC in V8.
I fixed it with a fairly oversized hammer:
1. Open node/deps/v8/src/heap.cc
2. Comment out everything inside the CollectAllAvailableGarbage
On Thu, Jun 6, 2013 at 2:20 PM, Hitesh Gupta hiteshgupt...@gmail.comwrote:
Hi Joran,
Thanks for sharing what worked for you.
We do not want to disable the GC entirely. But perhaps we'd like to
reduce kMaxNumberOfAttempts constant in Heap::CollectAllAvailableGarbage
from 7 to 1 or 2.
That
you.
在 2013年5月23日星期四UTC+8下午4时51分21秒，Jakob Kummerow写道：
Here's what the builtin forEach function does: https://code.google.com/**
p/v8/source/browse/branches/**bleeding_edge/src/array.js#**1078https://code.google.com/p/v8/source/browse/branches/bleeding_edge/src/array.js#1078
Notably, for every
Sure, given how make works, you can build any individual output file. The
default all meta target is just convenience.
On Thu, May 23, 2013 at 11:47 PM, mschwartz myk...@gmail.com wrote:
Is there a way to build v8 without building the cctest/ tree?
When I do
make x64.release
I see it
Here's what the builtin forEach function does:
https://code.google.com/p/v8/source/browse/branches/bleeding_edge/src/array.js#1078
Notably, for every index it checks if the array actually contains that
index. This check is mandated by the ECMAScript specification, see
Looks like you've forgotten to link in libv8_base.a.
On Mon, May 20, 2013 at 7:49 AM, Crossle Song xming.s...@gmail.com wrote:
$ gcc -Iinclude hello_world.cpp -o hello_world
out/x64.release/obj.target/tools/gyp/libv8_snapshot.a -lpthread
/tmp/ccmNdbH0.o: In function `main':
On Sun, May 5, 2013 at 5:30 PM, Shan Thomas shantin...@gmail.com wrote:
Javascript is a good
That's debatable.
simple
No, definitely not.
language, and v8 is a fast enough engine
Nope, if it were fast enough, we'd stop working on it, but in fact we're
working very hard to make it
One more clarification: to avoid allocation of HeapNumbers (and therefore
GC) in existing V8 versions, it's not necessary to re-write the app using
Float64Arrays -- just using regular JavaScript Arrays is enough (e.g.:
this.position = [x, y, z];). As long as there are only numbers in the
array, V8
On Tue, Apr 23, 2013 at 1:51 AM, johnjbar...@chromium.org wrote:
Hi. I am trying to modify the debug-data in the blink binding of V8. I'm
stuck because I can't find the cause for a crash. The call stack enters GC
and crashes, but the starting point varies with how I run the test (eg one
test
On Tue, Apr 23, 2013 at 6:06 PM, johnjbar...@chromium.org wrote:
On Tuesday, April 23, 2013 1:52:13 AM UTC-7, Jakob Kummerow wrote:
...
Is there any strategy for this kind of problem?
1) Build and run in Debug mode. It contains a bunch of checks that help
catch at least some types
On Sunday, April 21, 2013 10:26:53 AM UTC-7, Jakob Kummerow wrote:
On Sat, Apr 20, 2013 at 10:13 AM, Toon Verwaest verw...@chromium.orgwrote:
Hi Andrei,
V8 has an optimization that turns arrays into unboxed double arrays. If
you build a debug build of V8 you could do %DebugPrint(arr1
On Sat, Apr 20, 2013 at 10:13 AM, Toon Verwaest verwa...@chromium.orgwrote:
Hi Andrei,
V8 has an optimization that turns arrays into unboxed double arrays. If
you build a debug build of V8 you could do %DebugPrint(arr1) for example,
and you'll see that the elements kind is set to
There's no direct relationship between AddSimulate() and
FinishExitWithDeoptimization(). Both are helper methods to add certain
HInstructions to the Hydrogen instruction graph.
Simulates are safe points or snapshots of the compiler's environment,
which reflects the execution state of the program,
possible deoptimizations explicitly on the Hydrogen
level. Look for uses of DeoptimizeIf() in
src/$ARCH/lithium-codegen-$ARCH.cc for more.
Sincerely,
Amy
On Sun, Apr 14, 2013 at 1:22 PM, Jakob Kummerow jkumme...@chromium.orgwrote:
There's no direct relationship between AddSimulate
Makes sense. There's not just a maximum heap size, there are also various
per-object size limits.
Internal backing stores for dictionaries have a maximum size, effectively
limiting arrays with dictionary mode elements to ~22.3 million elements. We
may (or may not) be able to increase that limit,
https://code.google.com/p/v8/source/browse/branches/bleeding_edge/Makefile.android#51
On Tue, Apr 9, 2013 at 10:00 PM, Joe Millenbach jmillenb...@gmail.comwrote:
I've managed to get android builds to work, but this is just by setting up
the environment variable to point at my NDK folder. In
Sure, theoretically the optimization could be extended to floor of
multiplication. That's probably an uncommon case, though, so it may not be
worth it.
On Sun, Apr 7, 2013 at 11:43 PM, Ben Noordhuis i...@bnoordhuis.nl wrote:
On Sun, Apr 7, 2013 at 11:02 PM, Jakob Kummerow jkumme
Sounds like you haven't checked out GYP into v8\build\gyp\.
On Fri, Apr 5, 2013 at 2:38 PM, Kapil Rakh kapilr...@gmail.com wrote:
C:\V8third_party\python_26\**python.exe build\gyp
C:\V8\third_party\python_26\**python.exe: can't find '__main__.py' in
'build\\gyp'
I have followed the
On Fri, Apr 5, 2013 at 12:52 AM, SiBest simonjb...@googlemail.com wrote:
I've seen that there are platform-specific classes available in the V8
code base. These are in files platform-linux.cc, platform-macos.cc, etc.
and are used in some tests, for example see
:49 AM UTC+1, Jakob Kummerow wrote:
On Fri, Apr 5, 2013 at 12:52 AM, SiBest simon...@googlemail.com wrote:
I've seen that there are platform-specific classes available in the V8
code base. These are in files platform-linux.cc, platform-macos.cc, etc.
and are used in some tests, for example see
Earlier you said that several scripts are running [in] parallel. That's
precisely what's not supported. Sven was pointing out that when several
threads need to access an isolate, they can do so, but only *one at a
time*. No two threads can be allowed to run in the same isolate at the same
time, or
PM, Kapil Rakh kapilr...@gmail.com wrote:
I did checkout gyp using this command. I did a flle search for __main__.py
but couldn't find it.
svn co -q http://gyp.googlecode.com/svn/trunk build/gyp
On Mon, Apr 8, 2013 at 5:23 AM, Jakob Kummerow jkumme...@chromium.orgwrote:
Sounds like you
This is expected. V8 has a special optimization for Math.floor of a
division.
The only way of fixing this would be to remove that optimization, in
which case the /2 case would become about as slow as the *.5 case.
That's not what you want, is it? ;-)
Background:
All of this is due to JavaScript,
might to
explain in more detail?
Thank you.
Jakob Kummerow於 2013年4月1日星期一UTC+8上午3時59分10秒寫道：
On Sun, Mar 31, 2013 at 4:16 PM, Sam Lin ckjbo...@gmail.com wrote:
Hi,
1. Numbers:
About the tagging technique, Objects and small integer are both to be
set as 32 bits.
Further, they use
No.
How is Mark-and-Sweep a contradiction to stop-the-world, generational,
accurate?
On Sun, Mar 31, 2013 at 9:23 PM, Sam Lin ckjboy2...@gmail.com wrote:
Hi,
I noticed that while google io 2009, V8 took Mark-and-Sweep as its GC
algorithm.
However, when I went through the official site of
On Sun, Mar 31, 2013 at 4:16 PM, Sam Lin ckjboy2...@gmail.com wrote:
Hi,
1. Numbers:
About the tagging technique, Objects and small integer are both to be
set as 32 bits.
Further, they use the bottom bit as the tagged value. if the bit is
set, then this means that 31 bits are object
Look for the bindings/v8 directory in the WebKit sources.
On Sat, Mar 30, 2013 at 5:37 PM, Sam Lin ckjboy2...@gmail.com wrote:
Hi,
I have been searching around the way that how google replaces original
javascript engine with V8.
Does google provide any document or any information about
For many types of inline caches (loads/stores/calls, in particular), the
handler stub is chosen/generated depending on the hidden class of the
receiver object.
On Fri, Mar 29, 2013 at 10:38 AM, Sam Lin ckjboy2...@gmail.com wrote:
Hi,
I think I am clear with how hidden classes and inline
On Thu, Mar 28, 2013 at 2:58 PM, Sam Lin ckjboy2...@gmail.com wrote:
Hi,
about the type of garbage collection,
I have two questions,
1. Full non-compacting collection
I have no idea about the free memory gets added to free lists.
Why V8 does this and what's this for?
As you
Yang, Dave has already uploaded the patch at
https://codereview.chromium.org/13071005/. It would be great if you had
time to take a look, I'm not sure when I'll get around to it myself.
On Wed, Mar 27, 2013 at 11:10 AM, Yang Guo yang...@chromium.org wrote:
Hi Dave,
Not sure if this really
make arm.release hardfp=on
On Fri, Mar 22, 2013 at 7:49 AM, nagarjuna atluri nagajun...@gmail.comwrote:
Hi,
In Common.gypi i ahve noticed the Below menationed flag
*'v8_use_arm_eabi_hardfloat%':*
*
*
Can any one please let me know the changes to enable this for gyp Build..
Regards,
You mean, some milliseconds are more accurate than others? As in, they are *
precisely* 1.0ms long each, rather than just *roughly* 1ms?
Now, assuming there is a high-accuracy kind of milliseconds, do you really
care whether a so-called benchmark finishes in 2ms or 2.0ms? When there's
easily a
, Jakob Kummerow wrote:
You mean, some milliseconds are more accurate than others? As in, they
are *precisely* 1.0ms long each, rather than just *roughly* 1ms?
Now, assuming there is a high-accuracy kind of milliseconds, do you
really care whether a so-called benchmark finishes in 2ms or 2.0ms
On Fri, Mar 15, 2013 at 6:08 AM, William Wang william.w...@gmail.comwrote:
My mean is if in d8 shell,I type:
a=1+1; when complete by return button the 2 is printed.
I only want at code:* print(a) *to output 2.
any d8 option disable this default behavior?
No.
You can easily edit d8's
On Tue, Mar 12, 2013 at 11:56 PM, Benjamin Kalman kal...@chromium.orgwrote:
I'm on the Chrome Extensions team, and we've run into a problem where
extensions override Array.prototype.forEach in a way that breaks our
internal JS.
A workaround we've done is to write our own forEach method, but
On Wed, Mar 13, 2013 at 4:44 PM, Benjamin Kalman kal...@chromium.orgwrote:
Not really feasible, monkey patching is an idiom that JS developers would
get very upset about being taken away from them. Plus lots of libraries do
it. We had a hard enough problem (and in fact failed) forbidding eval.
I've updated the wiki page to say a few times per week instead of
weekly.
On Fri, Mar 8, 2013 at 8:22 AM, Sven Panne svenpa...@chromium.org wrote:
The description in the Wiki is not really accurate: We copy bleeding_edge
to trunk on a per-need basis, and things have been mildly tested by our
All trunk revisions compile fine, if you follow the official build
instructions.
You have discovered the fact that building with Scons used to work in the
past, but is not supported anymore.
Please be aware that the 3.9 branch is about one year old, is a lot slower
than more recent versions,
On Wed, Mar 6, 2013 at 3:36 AM, William Wang william.w...@gmail.com wrote:
I want get v8.dll but link fail.only get libv8.a.
Environment:
Window 7 64bit
v8 source code:
svn checkout http://v8.googlecode.com/svn/trunk/ v8
today( 2013.3.6) refresh again version is 13833
mingw:
There should be no need to use your own toolchain, as the Android NDK ships
with one. Since you're working on OSX, why did you download the NDK for
Linux? Please try again with the toolchain that comes with the Android NDK
for Mac (just unset your CXX/LINK variables, as it will be used
This is expected. Support for the arguments object in the d8 shell
requires stuff that's not available in shared library builds. Please use make
native to build a statically linked d8 for running the tick processor.
Note that profiling involves two steps:
1.) Generate the log file by running your
On Mon, Feb 25, 2013 at 1:43 PM, Егор Яковлев goga@gmail.com wrote:
But its really strangely, that standard example not worked in a public
revision. :)
All V8 revisions are public. Sometimes they're broken, and sometime they're
newer than the documentation. It happens with projects that
That doesn't change the fact that benchmarks do not provide reliable
information when they don't run long enough.
On Thu, Feb 21, 2013 at 12:11 AM, Dave Pacheco d...@cs.brown.edu wrote:
On Fri, Feb 1, 2013 at 12:57 AM, Jakob Kummerow jkumme...@chromium.orgwrote:
Date.now() or new Date
I could reproduce this a few minutes ago, but now it seems to be fixed.
Please try again.
On Thu, Feb 14, 2013 at 1:07 PM, Taneli Leppä ro...@sektori.com wrote:
Hello guys,
I seem to be having some problems checking out or exporting
code from the SVN server:
$ svn co
On Sat, Feb 9, 2013 at 2:32 AM, Leo Lunatic lle...@gmail.com wrote:
Given:
a=0x8000
b=0x8
c=0x8
we find
a !== b
but
a === c
and this is LOVELY!
a !== ~~a
I realized this is probably and artifact of the optimizations we know and
love,
It's an artifact
On my box (Core i7, Ubuntu 12.04), running the example in the d8 shell,
ia32 and x64 have almost identical performance: 23ms vs. 24ms. Maybe
node.js adds some overhead to the x64 version?
The generated machine code is pretty much the same, too.
What you've created here is mostly a memory/GC
Well, doing something 100,000 times takes time. Why do you expect it to be
faster? What does your JavaScript function do? How long does it take to
execute it a single time? Have you profiled your app? Have you narrowed
down whether the majority of the time is spent in JS or C++? And in which
On Tue, Jan 22, 2013 at 6:37 AM, nagarjuna atluri nagajun...@gmail.comwrote:
Hi ,
Which all versions of Javascript does v8 support?
See http://v8.googlecode.com:
V8 implements ECMAScript as specified in
ECMA-262http://www.ecma-international.org/publications/standards/Ecma-262.htm,
5th
The SCons build is unmaintained and no longer being tested. It is entirely
possible that it is broken. You should really migrate away from it, as we
will delete the SCons-related files Really Soon Now™.
In this particular case, code-stubs-hydrogen.cc is probably missing. If
you're lucky, adding
/android_arm.release/mksnapshot: ELF 32-bit LSB executable, Intel 80386,
version 1 (GNU/Linux), dynamically linked (uses shared libs), for GNU/Linux
2.6.16, not stripped
-L
On Fri, Jan 11, 2013 at 1:16 AM, Jakob Kummerow jkumme...@chromium.orgwrote:
On Fri, Jan 11, 2013 at 1:59 AM, Louis
On Thursday, January 10, 2013, Jakob Kummerow wrote:
Yes, building with snapshot=on is the default.
Disregard the SConstruct file. It is no longer used and about to be
deleted. The SCons based build has been dead for a while now.
On Thu, Jan 10, 2013 at 9:11 PM, utkonos rsimmo...@gmail.com wrote
On Fri, Jan 11, 2013 at 10:34 AM, Neha nehatripathi28...@gmail.com wrote:
*Solution for compiling v8 on Linux*:
gcc version used : 4.1.2. Also tried with gcc 4.2.3.
The code did not compile with the existing version of python(2.4)
Download python 2.7.3. Configure and Build the same.
use
On Wed, Jan 9, 2013 at 10:37 PM, belgiamir...@gmail.com wrote:
I am attempting to change the chromium in the following manner:
I would like to inject some code to the end of each basic block of the AST
while V8 compiles the javascript code,
You're not saying why you want to do this, but
On Thu, Jan 10, 2013 at 6:12 PM, Matt Horsnell
matthew.horsn...@gmail.comwrote:
With the old scons build system you used to be able to build statically,
what changed?
Nothing has changed. You can link all of V8 into a single binary (and the
make targets do that by default). You cannot, and
Yes, building with snapshot=on is the default.
Disregard the SConstruct file. It is no longer used and about to be
deleted. The SCons based build has been dead for a while now.
On Thu, Jan 10, 2013 at 9:11 PM, utkonos rsimmo...@gmail.com wrote:
It seems from the instructions that snapshot=on
On Wed, Jan 9, 2013 at 2:50 AM, Joe Millenbach jmillenb...@gmail.comwrote:
I'm trying to build current versions of the v8 shell to use while running
the v8 benchmark, Kraken, and SunSpider. But we want the build to be
statically linked so it will run without any specific libraries or versions
Please realize that V8 is purely a JavaScript engine. It doesn't render
anything. It doesn't know what HTML is.
Arguably, being indexable by search engines (which essentially means being
navigatable by their crawlers) is a feature request for the server-side
application framework.
On Wed, Jan
On Wed, Jan 9, 2013 at 4:50 PM, daslicht anso...@gmail.com wrote:
Arguably, being indexable by search engines (which essentially means
being navigatable by their crawlers) is a feature request for the
server-side application framework.
Thats the thing with 'modern' JS based webapps which
On Tue, Jan 8, 2013 at 11:51 AM, Neha Mishra nehatripathi28...@gmail.comwrote:
Hi,
I am getting the same error on Windows. I have the v8-master code. Have
downloaded the cygwin and python_26.
I have kept python_26 and cygwin at v8-master\tools\third_party.
Well, observe the difference
On Sun, Dec 23, 2012 at 12:09 PM, jasonleste...@gmail.com wrote:
What's the perfrmance difference between creating objects and arrays?
Common sense makes me believe that arrays would be quicker because there's
less to them internally.
Why would there be less to them internally, whatever that
See my response on the other thread. As far as I can see the documentation
is correct.
On Sun, Dec 16, 2012 at 4:29 PM, jasonleste...@gmail.com wrote:
I'm having the same exact problem. Everything was done by the book, and
everything was double checked. Is the guide in need of being updated?
Have you tried using your favorite debugger to get a backtrace? That should
help you figure out what's going on.
On Wed, Dec 12, 2012 at 5:41 PM, Paul Harris harris...@gmail.com wrote:
Hello again,
I'm seeing a crash that seems a bit strange.
I adjusted d8.cc's Load method so that it
- How did you build V8?
- Which version of V8 did you build?
- I think your -L path may be wrong, at least on Linux libv8.so is in
out/x64.release/lib.target/, not sure about Mac but the situation is
probably similar.
- I also think this may be a question for the jav8 project rather than
upstream
, December 11, 2012 3:10:57 AM UTC-6, Jakob Kummerow wrote:
Looks like the Sunspider runner expects the shell to expose a run()
method. I have no idea what that method is supposed to do, and considering
how useless Sunspider is as a benchmark, I don't think
investigating/implementing it is a good
On Sun, Dec 2, 2012 at 3:20 PM, Erez erezro...@gmail.com wrote:
Hi.
Thanks for the answer.
How can I support the code in the site?
- rename the hidden input method
- use document.getElementById()
Thanks,
Erez
On Sunday, December 2, 2012 4:00:04 PM UTC+2, Jakob Kummerow wrote
JavaScript objects can't have a method and a property of the same name.
Rename one of the two.
On Sun, Dec 2, 2012 at 10:20 AM, Erez erezro...@gmail.com wrote:
Hi.
I'm using v8 to process web pages.
I have the following problem with a code on a certain site.
The site has the following
On Sat, Dec 1, 2012 at 6:25 PM, Maksim Lukyanchenko
maksim.lukyanche...@gmail.com wrote:
I'm trying to compile V8 on Win7 for VS2010:
svn co http://gyp.googlecode.com/svn/trunk build/gyp
svn co http://src.chromium.org/svn/trunk/tools/third_party/python_26
third_party/python_26
svn co
This is the sixth or so nonsensical mail that you've sent to this list in
the past few days. Would you please stop that? If you'd like to
unsubscribe, you can do that at
https://groups.google.com/forum/?fromgroups#!forum/v8-users.
On Wed, Nov 14, 2012 at 10:33 AM, joko suwito joko...@gmail.com
Hard to say without knowing more about what you did and what you observed.
E.g. if your browser was an x64 build and your command-line V8 was an ia32
build, that would explain all sorts of differences. Telling us *which*
tests exactly were faster in the browser would also add quite a bit of
Who's optimizing for a specific version of V8? Avoiding inherently slow
things (such as for-in loops over slow objects, or needless polymorphism,
etc) in your JS code makes a lot of sense regardless of the VM you're
running on.
Sure, some deopts may no longer happen in an upcoming version, but
On Fri, Oct 5, 2012 at 3:29 PM, Paweł Hajdan, Jr.
phajdan...@chromium.orgwrote:
On Tue, Oct 2, 2012 at 4:32 PM, Jakob Kummerow jkumme...@chromium.orgwrote:
V8 does not support x32 at this time. Sorry.
OK - do you plan to add that support? Just asking e.g. so I can take the
right action
Please follow the provided instructions:
Compile hello_world.cpp, linking to the static libraries created in the
build process. For example, on 64bit Linux using the GNU compiler:
g++ -Iinclude hello_world.cc -o hello_world
out/x64.release/obj.target/tools/gyp/libv8_{base,snapshot}.a -lpthread
401 - 500 of 588 matches
Mail list logo
