Den 2010-06-09 20:30 skrev Ralf Wildenhues:
* Peter Rosin wrote on Wed, Jun 09, 2010 at 09:00:06AM CEST:
Den 2010-06-09 07:14 skrev Ralf Wildenhues:
I would like to see more of the story here. If this is still
reproducible for you, post the output of
make check-local TESTSUITEFLAGS='-v -d -
* Peter Rosin wrote on Wed, Jun 09, 2010 at 09:00:06AM CEST:
> Den 2010-06-09 07:14 skrev Ralf Wildenhues:
> >I would like to see more of the story here. If this is still
> >reproducible for you, post the output of
> > make check-local TESTSUITEFLAGS='-v -d -x 77'
>
> It is very reproducible, e
Hello Jay,
* Jay K wrote on Wed, Jun 09, 2010 at 06:17:49PM CEST:
> bash-4.1$ /home/jayk/src/gcc-4.5.0/config.guess
> alphaev67-dec-osf5.1
>
> ash-4.1$ uname
> OSF1 V5.1 732 alpha
>
> An old odd out of support host, not even 5.1a, but that's your point, eh? :)
>
> Not sure there's any rpath s
On 06/08/2010 11:55 PM, Ralf Wildenhues wrote:
This skips a bit more than is absolutely needed (only the demo-exec run
after demo-nopic and demo-make would need to be skipped), but I don't
see a big problem with that. We can fix that when we rename tests to
have unique names (e.g., to allow par
Hallo Ralf,
On 9 Jun 2010, at 03:48, Ralf Wildenhues wrote:
> We have workarounds in most Java tests but not in this one, e.g.,
> convenience.at or infer-tag.at. So I guess that calls for factorization
> of the issue. The convenience test needs executability, the rest only
> linkability. These
