harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alexei Fedotov" <alexei.fedo...@gmail.com>
Subject Re: [general][testing] need a piece of advise: how we can handle VM crashes
Date Wed, 06 Dec 2006 19:05:00 GMT
Alexander,

It is great that you started this work. Please, excuse my dumb
questions, I just want to understand the subject and its correlations
to other activities better

1. Does you patch capture an output from crashed VM? This would be
quite useful, because VM prints a native stack trace when crashes.

2. I support Alexei's question about functional differences between
Vladimir's approach and your one. As for now I more or less understand
why Vladimir's solution is good (simple and catches crashes with zero
return code), and understand why it is bad (it doesn't capture VM's
output).

3. Can your solution be extended to modes other than perTest (for
example, to test reliability on simple cases)?

4. You wrote,
> Note that H000 regression test is failed because of DRLVM crash.

Have I missed something? What is this H000?

5. Is there anything Harmony specific in your task? Could it be useful
for Ant project?

6. Is it difficult to add handling of test decorators in a following way?

  <junitperf fork="false">
      <decorator class="junit.extensions.RepeatedTest">
            <ctorarg test/>
            <ctorarg intvalue="${junit.extensions.repeatedtest.repetions}"/>
      </decorator>
      <test name="org.apache.ecs.InternationalCharTest" />
  </junitperf>

(I just oticed that posted this example privately).

To my mind idea is quite promising, and I even have started to
interfere with your field. You could think of a decorator which
preallocates a big amount of memory and runs given tests in the rest
of the heap space, or a decorator which runs GC in parallel, or even a
decorator which runs all tests on a finalization thread.

I put two coffees for the fact that the last decorator to class
library unit tests would produce many interesting DRLVM artifacts and
JIRA issues.

Alexander, thanks again for starting this job! All, thank you for
reading up to this point. :-)

On 12/6/06, Alexei Zakharov <alexei.zakharov@gmail.com> wrote:
> Vladimir,
>
> Well, initially I was going to commit your patch. But now we have two
> patches so I am trying to understand what is the best thing to do
> here. In same situations in the past (when we had two separate
> contributions of math for example) we have chosen one implementation
> as primary and then  transferred cool features from another
> implementation as a set of patches. Can we do something like it in
> this case? Alexander was telling something about DRLVM regression
> tests. Is your solution applicable there?
>
> Thanks,
>
> 2006/12/6, Vladimir Ivanov <ivavladimir@gmail.com>:
> > On 12/6/06, Alexander Kleymenov <kleymenov@gmail.com> wrote:
> > >
> > > Hi Alexey,
> > >
> > > To answer your question it is needed to understand what is the problem
> > > you are questioning about? If the problem is as
> > >
> > > On 11/29/06, Alexei Fedotov <alexei.fedotov@gmail.com> wrote:
> > > > detecting crashes during a test run
> > >
> > > then you are right. But if you meant
> > >
> > > On 12/1/06, Vladimir Ivanov <ivavladimir@gmail.com> wrote:
> > > > generate 'crashed' result for test
> > >
> > > you are wrong.
> >
> >
> >
> > No, my patch also generates some results for crashed tests :)
> >
> >
> >
> > These patches are implement the similar functionality but with different
> > approach.
> >
> > Due to unexpected nature of crashes both of them will be useful. In the case
> > when even no report file was created the second one is better. But if vm
> > crashed after creation zero-length file the first one is better.
> >
> >
> > Thanks, Vladimir
> >
> >
> > > Thanks,
> > > Alexander
> > >
> > > On 12/5/06, Alexei Zakharov <alexei.zakharov@gmail.com> wrote:
> > > > Hi Alexander,
> > > >
> > > > But what is the difference between junit extension your attached to
> > > > HARMONY-2416 and Vladimir's work from HARMONY-2349? It seems it solves
> > > > the same problem. Am I wrong?
> > > >
> > > > Thanks,
> > > >
> > > > 2006/12/5, Alexander Kleymenov <kleymenov@gmail.com>:
> > > > > Hello,
> > > > >
> > > > > As a part of DRLVM regression testing infrastructure I've developped
> > > > > JVM crash handler. Please, look at it in HARMONY-2416 report [1].
More
> > > > > information can be found in corresponding mail thread [2].
> > > > >
> > > > > Thank you,
> > > > > Alexander
> > > > >
> > > > > [1] http://issues.apache.org/jira/browse/HARMONY-2416
> > > > > [2]
> > > http://mail-archives.apache.org/mod_mbox/harmony-dev/200612.mbox/%3ce09a11790612040450g7412a7b3l648ca690d5deb3f8@mail.gmail.com%3e
>
>
>
>
> --
> Alexei Zakharov,
> Intel Enterprise Solutions Software Division
>


-- 
Thank you,
Alexei

Mime
View raw message