which tests did you run?
I ran Classlib tests on DRLVM in "once" fork mode. it took about 20-25 minutes
on each x86 and x86_64 on Linux
2006/12/9, Geir Magnusson Jr. <[EMAIL PROTECTED]>:
I've been running tests on x86 and x86_64 linux, and things "feel" slow.
Problem is I don't have any times
it's not your commits. it's an intermittent problem we've seen before.
seems like either a bug in J9 or memory corruption in Harmony classlib
Thanks,
Mikhail
2006/12/9, Tim Ellison <[EMAIL PROTECTED]>:
Can anyone reproduce this?
All tests pass for me, and on the IBM build machine.
Strangely t
On 12/9/06, Tim Ellison <[EMAIL PROTECTED]> wrote:
Can anyone reproduce this?
All tests pass for me, and on the IBM build machine.
All tests pass on my machine (Windows XP SP2 + IBM VME) too. Thanks.
Strangely the line below:
> at java.io.ObjectStreamClass.computeSerialVersionUID(
ObjectSt
have we considered using
http://goog-perftools.sourceforge.net/
??
If I look at
http://wiki.wikked.net/wiki/Squid_memory_fragmentation_problem
it shows you some serious potential for improvement in memory management
(but I don't know enough about our GC to know how useful that is)
Thoughts?
Nadya : I'll leave it for you to fix :-)
Thanks
Tim
Geir Magnusson Jr. wrote:
> yes
>
> dev@harmony.apache.org
>
>
> Tim Ellison wrote:
>> [EMAIL PROTECTED] wrote:
>>> -
>>> +
>>
>>
>> huh? it *is* dev@harmony.apache.org, right?
>>
>> Tim
>
Can anyone reproduce this?
All tests pass for me, and on the IBM build machine.
Strangely the line below:
> at
> java.io.ObjectStreamClass.computeSerialVersionUID(ObjectStreamClass.java:54)
doesn't match what I see in that source file -- i.e. the
computeSerialVersionUID
method is at line 323. O
Alexei Fedotov wrote:
Gregory,
Thank you for explaining. As far as I unerstand writing to the thread
unsafe list is enough to cause segmentation fault on smp. You need to
update two fields, and this cannot be done atomically.
Writing to the list won't do any harm as long as the list is not use
Thanks!
On 12/8/06, Alexei Zakharov <[EMAIL PROTECTED]> wrote:
Done!
Regards,
2006/12/8, Alexei Fedotov <[EMAIL PROTECTED]>:
> Hello, Vladimir,
>
> There are two JIRA issues about this test, the most frequent problem
> even contain a patch made by Denis and tried by me.
> http://issues.apache.
Gregory,
Thank you for explaining. As far as I unerstand writing to the thread
unsafe list is enough to cause segmentation fault on smp. You need to
update two fields, and this cannot be done atomically.
According to the wiki page [1] I need to cleanup CodeChunkInfo
structures anyway. Access to
Gregory,
thank you for answers.
I remember there was a reason why I
moved the functions from compile_do_compilation, I think they were not
called in case of methods recompilation.
The situation is opposite (but probalby this is exactly you want to):
compile_me stub is called once during the fi
Alexei Fedotov wrote:
Gregory, All,
I started looking into the compile.cpp file and found new ti callbacks
in compile_jit_a_method jit stub. I've got several questions.
DebugUtilsTI *ti = VM_Global_State::loader_env->TI;
if(ti->isEnabled() && ti->is_single_step_enabled()
&& !method->is_nativ
Alexei Fedotov wrote:
Gregory, Ilya, TI gurus,
Could you please help me to understand why do we need a global list of
ia32 specific code chunks VM_Global_State::loader_env->dcList? Cannot
we use good old CodeChunkInfo structure?
The information contained in dcList is different from CodeChunks.
I updated the wiki with our test setup. I don't have Cruise control set to
send out any emails. Is that something that people would want?
Naveen
>> Is anyone else is running cruise control on ia32/RHEL4 or ia32/FC5? There
>> isn't much point of duplicating setups.
>
>Take a look here to see
Thanks! I think I have this automated to upload results every night. I'll
check tomorrow to see if everything is working.
BTW, I uploaded duplicate results to harmonytest.org while I was testing
things. Sorry.
Naveen
>On 12/7/06, Naveen Neelakantam <[EMAIL PROTECTED]> wrote:
>
>> Thanks Vla
Done!
Regards,
2006/12/8, Alexei Fedotov <[EMAIL PROTECTED]>:
Hello, Vladimir,
There are two JIRA issues about this test, the most frequent problem
even contain a patch made by Denis and tried by me.
http://issues.apache.org/jira/browse/HARMONY-2353
http://issues.apache.org/jira/browse/HARMONY
Tim Ellison wrote:
Alexey Petrenko wrote:
2006/12/8, Tim Ellison <[EMAIL PROTECTED]>:
Alexey Petrenko wrote:
The only difference is that my version is switching debug option for
javac as well. And I think that I'll commit this modification.
Is it worth it? The javac compiler doesn't generate
Stefano Mazzocchi wrote:
Stepan Mishura wrote:
On 12/8/06, Geir Magnusson Jr. wrote:
We've talked about this before because we knew we'd run into the
problem. I'll see if I can remember what we concluded...
geir
See[1] - we talked about two approaches: repackaging and 'dedicated class
l
I've been running tests on x86 and x86_64 linux, and things "feel" slow.
Problem is I don't have any times to compare to - can we add that to
the data recorded in harmonytest?
Does anyone else feel that things are slow?
geir
I spoke a little too soon.
On ubuntu 6, I am seeing stress.Threads fail... the much dreaded OOM.
"340 threads creation resulted in java.lang.OutOfMemoryError: Failed to
create new thread
PASSED, 340 threads created and started"
Now, this is a tad confusing :)
It's the old thread creation pr
Stepan Mishura wrote:
> On 12/8/06, Geir Magnusson Jr. wrote:
>>
>>
>>
>> Stefano Mazzocchi wrote:
>> > Robin Garner (JIRA) wrote:
>> >> [
>> http://issues.apache.org/jira/browse/HARMONY-2130?page=comments#action_12456374]
>>
>> >>
>> >> Robin Garner commented on HARMONY-2130:
>> >> ---
I like to keep 'em guessing :)
A website is a website. It can have the docs - should have the docs.
And maybe in the end there's nothing to really distinguish... but I
always worry about the website tail wagging the documentation dog...
geir
Morozova, Nadezhda wrote:
I'd would prefer if
yes
dev@harmony.apache.org
Tim Ellison wrote:
[EMAIL PROTECTED] wrote:
-
+
huh? it *is* dev@harmony.apache.org, right?
Tim
[EMAIL PROTECTED] wrote:
> -
> +
huh? it *is* dev@harmony.apache.org, right?
Tim
Nice work - I'm testing it now and will commit. If someone can test
today on x86_64, that would be great, or I will do it tomorrow
geir
Elena Semukhina wrote:
After a few days of runs I can conclude that almost 40 tests are valid and
should be removed from the exclude lists. I've prepared a
>I'd would prefer if we distinguish between "website" and
"documentation"
>as I think they are actually different.
I've had an impression that we've actually been pulling these together,
so your idea is somewhat revolutionary to me :)
Cheers,
Nadya
>-Original Message-
>From: Geir Magnu
Thomas Hawtin wrote:
Ivanov, Alexey A wrote:
From: Nathan Beyer [mailto:[EMAIL PROTECTED]
Do you know of any specific examples of code that's required to be
thread
safe?
Yes, I do. For example: javax.swing.text.AbstractDocument.remove,
.insertString, .createPosition;
javax.swing.text.JText
Morozova, Nadezhda wrote:
As a result, some docs are distributed with code and copied to site,
some are stored at the site only, and some are duplicated and require
syncing.
I can do the cleanup to find files that exist in two variants and try
the merge, but I need to know where we're mergin
have fun :)
Tony Wu wrote:
I have used winK before, and I'm an amateur at movie making ;-)
I would like to take this job if no one objects.
On 12/8/06, Morozova, Nadezhda <[EMAIL PROTECTED]> wrote:
I'd be glad to help - but I haven't worked with Wink. Any volunteers to
lend a hand?
Cheers,
Na
Ivan Popov wrote:
I've tried to build jdktools with the latest changes in its build
system. On Linux build works well, but on Windows it fails mostly
because of lack of makefiles for JPDA module.
Yep, was still working on that.
I've created missed makefiles for JPDA module, prepared patche
Cool!
I did not work on application test for so long time that I have not
tried harmonytest.org ever since...
Thanks for improvement! I can't help trying it immediately.
On 11/13/06, Richard Liang <[EMAIL PROTECTED]> wrote:
Great Job. ;-)
Anton Luht wrote:
> Hello,
>
> Yesterday I've deployed
I have used winK before, and I'm an amateur at movie making ;-)
I would like to take this job if no one objects.
On 12/8/06, Morozova, Nadezhda <[EMAIL PROTECTED]> wrote:
I'd be glad to help - but I haven't worked with Wink. Any volunteers to
lend a hand?
Cheers,
Nadya
>-Original Message-
Nathan Beyer wrote:
On 12/7/06, Stefano Mazzocchi <[EMAIL PROTECTED]> wrote:
Vladimir Strigun (JIRA) wrote:
> [
http://issues.apache.org/jira/browse/HARMONY-2130?page=comments#action_12456467
]
>
> Vladimir Strigun commented on HARMONY-2130:
> ---
This would probably be well served with a password for each user :)
Vladimir Ivanov wrote:
On 12/7/06, Naveen Neelakantam <[EMAIL PROTECTED]> wrote:
Thanks Vladimir!
Is anyone else is running cruise control on ia32/RHEL4 or ia32/FC5?
There
isn't much point of duplicating setups.
I have i
I've tried to build jdktools with the latest changes in its build
system. On Linux build works well, but on Windows it fails mostly
because of lack of makefiles for JPDA module.
I've created missed makefiles for JPDA module, prepared patches and
added them to HARMONY-2180. With these patches jdkt
On 12/8/06, Zakharov, Vasily M <[EMAIL PROTECTED]> wrote:
Ok, Andrew, looks like you're right.
I'm withdrawing my objections. :)
:-)
Vasily
-Original Message-
From: Andrew Zhang [mailto:[EMAIL PROTECTED]
Sent: Friday, December 08, 2006 4:29 AM
To: dev@harmony.apache.org
Subject:
There is enough info in the manifests to do this classloader trickery at
runtime -- by explicitly listing what we export/import from a module we
can hide purely implementation-specific packages.
The remaining pieces are to (1) define a manifest that exports only the
JSE-defined packages (to remove
Naveen Neelakantam wrote:
> Is anyone else is running cruise control on ia32/RHEL4 or ia32/FC5? There
> isn't much point of duplicating setups.
Take a look here to see what we have going so far
http://wiki.apache.org/harmony/Automated_Testing
Regards,
Tim
I'd be glad to help - but I haven't worked with Wink. Any volunteers to
lend a hand?
Cheers,
Nadya
>-Original Message-
>From: Tim Ellison [mailto:[EMAIL PROTECTED]
>Sent: Friday, December 08, 2006 6:19 PM
>To: dev@harmony.apache.org
>Subject: Re: [doc] Authors of dev_eclipse movie need
Morozova, Nadezhda wrote:
> I'd vote for 2 - because this seems simpler and unambiguous and fewer
> copies of files actually exist. However, I'd go with what gurus say.
I agree, keep a single copy in the site tree and remove the use of
externals.
Regards,
Tim
Alexey Petrenko wrote:
> 2006/12/8, Tim Ellison <[EMAIL PROTECTED]>:
>> Alexey Petrenko wrote:
>> > The only difference is that my version is switching debug option for
>> > javac as well. And I think that I'll commit this modification.
>> Is it worth it? The javac compiler doesn't generate differe
Ivan Popov wrote:
> Yes, it's possible, but requires additional efforts. I'm not sure it
> makes sense to spend these efforts, because that JDWP agent is not
> positioned to work with other JVMs and may use some VM dependent
> features.
FWIW we tried it with the IBM VME and it seemed to work quite
Oliver Deakin wrote:
> I also notice that the compiler settings are changed to produce 1.4
> classes, so this video is fairly out of date now. I seem to remember
> that Tim was the original creator.
I created it ages ago, using Wink -- but have no emotional attachment to
it, so feel free to edit
Vladimir Ivanov wrote:
> The link http://harmony.apache.org/subcomponents/buildtest/index.html is a
> little bit outdated but it keeps the main idea:
It would be good if you could fix the out of date bits, or at least
raise a JIRA to remind us to do it. Out of date doc is very frustrating.
Thank
2006/12/8, Tim Ellison <[EMAIL PROTECTED]>:
Alexey Petrenko wrote:
> The only difference is that my version is switching debug option for
> javac as well. And I think that I'll commit this modification.
Is it worth it? The javac compiler doesn't generate different code with
the -g option, it simp
Alexey Petrenko wrote:
> The only difference is that my version is switching debug option for
> javac as well. And I think that I'll commit this modification.
Is it worth it? The javac compiler doesn't generate different code with
the -g option, it simply adds the debug attributes to the .class fi
Evgueni Brevnov wrote:
> Tim,
>
> On 11/21/06, Tim Ellison <[EMAIL PROTECTED]> wrote:
>>
>> However, the resource management API
>> we define should be generic, so we can register a callback to such
>> third-party libraries if they have a clean-up function we need to call.
>
> I agree, we need t
Ok, Andrew, looks like you're right.
I'm withdrawing my objections. :)
Vasily
-Original Message-
From: Andrew Zhang [mailto:[EMAIL PROTECTED]
Sent: Friday, December 08, 2006 4:29 AM
To: dev@harmony.apache.org
Subject: Re: [classlib][luni] java.net.ServerSocketTest fails - need
help
After a few days of runs I can conclude that almost 40 tests are valid and
should be removed from the exclude lists. I've prepared a patch to smoke
tests: https://issues.apache.org/jira/browse/HARMONY-2543
After it is applied we'll have 3 stable issues:
1) gc.Mark craches on Windows
2) - 3) stre
+1 for site-only
2006/12/7, Morozova, Nadezhda <[EMAIL PROTECTED]>:
Folks,
I am sorry to re-iterate over the same topic, but I haven't seen any
solution and do not know what to do. In fact, the question seems to be
large enough for both classlib and drlvm docs. Opinions welcome.
There seems to
Hello, Vladimir,
There are two JIRA issues about this test, the most frequent problem
even contain a patch made by Denis and tried by me.
http://issues.apache.org/jira/browse/HARMONY-2353
http://issues.apache.org/jira/browse/HARMONY-2387
Dear committers, could you please commit HARMONY-2353?
Th
Tim,
The patch has been applied to the Pack200 code, and I've raised a
separate bug for the separation of the pack200 stuff:
http://issues.apache.org/jira/browse/HARMONY-2479
I still think a separate module makes sense, because even if the
build.xml compiles against 1.4, anyone who checks out t
Hello everybody,
seems, the test HttpURLConnectionTest is unstable now. It failed time to
time on the DRLVM and IBMVM.
Could somebody from net-guru take care about this intermittent failure?
Thanks, Vladimir
BUILD FAILED:
Unit Tests: (19301)
error: testGetOutputStream
org.apache.harmony.tes
On 12/7/06, Rana Dasgupta <[EMAIL PROTECTED]> wrote:
On 12/7/06, Mikhail Fursov <[EMAIL PROTECTED]> wrote:
>
>
>
> >I mean that if the whole test is in the one method the optimized method
> version will never be >executed unless we have on stack replacement
support.
>
>
Yes, I understand now th
53 matches
Mail list logo
