Re: [Bug-wget] Memory leak in idn_encode; Valgrind suppression file

[Tim Rühsen]

Hi Hubert,

Am Mittwoch, 8. April 2015, 18:59:59 schrieb Hubert Tarasiuk:
> I forgot to add Makefile.am to commit message. Please find the updated
> patch attached,
> 
> W dniu 08.04.2015 o 18:55, Hubert Tarasiuk pisze:
> > W dniu 06.04.2015 o 22:26, Tim Rühsen pisze:
> >> Valgrind suppressions are a bit compiler/architecture/distribution
> >> dependent. Maybe you could you add a comment into the suppression file
> >> with these infos. As a quick reference and explanation.
> > 
> > I have added a reference URL to Redhat's bugzilla, where the problem is
> > described.
> > https://bugzilla.redhat.com/show_bug.cgi?id=678518
> > 
> >> Sorry, one point I missed: Please put the suppression file into
> >> EXTRA_DIST
> >> variable in tests/Makefile.am. Else it won't go into the tarball (make
> >> dist).> 
> > Done.
> > 
> > I also refactored a little my modification in the Perl script. (Put the
> > suppression file path into a variable, and use Cwd to get the correct
> > path, instead of making relative path from the test being run.)

Could you use $srcdir instead Cwd ? I didn't test it, but it could fail when 
using DISTCHECK_CONFIGURE_FLAGS="VALGRIND_TESTS=1" make distcheck. It least we 
should use the same path mechanism as with e.g. 'certs' in 
Test-proxied-https-auth.px.

> > 
> > And I also found a simpler way to suppress that error. Valgrind has an
> > option --partial-loads-ok which would suppress it (more details can be
> > found in the URL above).
> > 
> > I am not sure which workaround is better for Wget (suppression file vs.
> > --partial-loads-ok=yes). What do you think?
> > (The --partial-loads-ok could suppress actual mistakes in future Wget's
> > code, while the former just works for the particular function in
> > libidn.so.)> 

You name it. IMO, that is why we should stick with a suppression file. Finer 
granularity, more control.

Regards, Tim

signature.asc
Description: This is a digitally signed message part.