I've created snapshots and uploaded them to the site as usual.
But I've not created snapshots for Linux with libstdc++.so.6 - the
build failed to compile classlib - ant can not find ecj_3.2.jar. I
checked that the jar exists in ANT_HOME/lib but I still see error
message. If I add the jar to classpath then the build fails with:
compile-java:
[echo] Compiling ACCESSIBILITY classes
copy-resources:
[copy] Copying 1 file to
/export/users/smishura/snapshot/working_classlib/build/classes
svn-info:
build-jar:
BUILD FAILED
/export/users/smishura/snapshot/working_classlib/build.xml:113: The
following error occurred while executing this line:
/export/users/smishura/snapshot/working_classlib/make/build-java.xml:193:
The following error occurred while executing this line:
/export/users/smishura/snapshot/working_classlib/make/properties.xml:420:
The following error occurred while executing this line:
/export/users/smishura/snapshot/working_classlib/modules/accessibility/build.xml:108:
No supported regular expression matcher found
Any idea what is wrong?
Thanks,
Stepan.
On 4/28/07, Alexey Petrenko <[EMAIL PROTECTED]> wrote:
Let's try to build snapshots right now and test them.
SY, Alexey
2007/4/28, Stepan Mishura <[EMAIL PROTECTED]>:
> On 4/28/07, Mikhail Loenko wrote:
> > no, it's all platforms except lin32 ;)
> >
> > it looks like a stable failure. Not necessarily a blocker though
> >
>
> Yes, it is stable failure. Do we consider it is as a blocker that must
> be fixed? Or we can live with it and build snapshots right now?
>
> Thanks,
> Stepan.
>
> > Thanks,
> > Mikhail
> >
> > 2007/4/27, Alexey Varlamov <[EMAIL PROTECTED]>:
> > > If this is about [EMAIL PROTECTED], I recall Vladimir complained about
this
> > > failure some time ago. So it's "known" intermittent failure (though
> > > sometimes it is reproduced quite stably in batch runs). I don't think
> > > this is real blocker.
> > >
> > > 2007/4/27, Mikhail Loenko <[EMAIL PROTECTED]>:
> > > > test_freeMemory started to fail
> > > >
> > > > 2007/4/27, Mikhail Loenko <[EMAIL PROTECTED]>:
> > > > > AFAIK the tests are running still - they run 6 hours
> > > > >
> > > > > 2007/4/27, Yang Paulex <[EMAIL PROTECTED]>:
> > > > > > 2007/4/27, Tim Ellison <[EMAIL PROTECTED]>:
> > > > > > >
> > > > > > > It seems like we are very close, so let's freeze the code here --
no
> > > > > > > more commits please (including even simple low risk bug fixes)
without
> > > > > > > agreement from two committers on the dev list that it should go
in [1].
> > > > > > > This is true until we can declare the build good enough for a
> > > > > > > milestone, then it is back to business as usual.
> > > > > > >
> > > > > > > There is an outstanding bug that looks like a MUST fix for this
milestone
> > > > > > > :
> > > > > > >
> > > > > > > HARMONY-3783 [classlib][luni] java cannot find class in a
directory
> > > > > > > started with dot
> > > > > > >
> > > > > > > When that is fixed we will have a candidate for everyone to try.
Please
> > > > > > > raise anything you think is critical/blocker for consideration.
> > > > > >
> > > > > >
> > > > > > Seems HARMONY-3783 has OK now, but there is a CC failure for
> > > > > > DRLVM/win.x86_64, any ideas what's the problem?
> > > > > >
> > > > > > [1] I just made up that policy, let me know if you disagree with it.
> > > > > > >
> > > > > > > Regards,
> > > > > > > Tim
> > > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > --
> > > > > > Paulex Yang
> > > > > > China Software Development laboratory
> > > > > > IBM
> > > > > >
> > > > >
> > > >
> > >
> >
>
>
> --
> Stepan Mishura
> Intel Enterprise Solutions Software Division
>
--
Stepan Mishura
Intel Enterprise Solutions Software Division
