On Fri, 23 Jul 2010 01:12:04 +0200, Marc Lehmann wrote: > > > The only things I can suggest is playing around with the backend selection > > > (I can then hardcode defaults). > > Ok, thanks for your hints, we can try that. > Oh, the most obvious I forgot: on those platforms where pthreads are > incompatible with POSIX, it is usually possible to use the pthreads > backend - it will be much slower, but should work on any platform that > supports pthreads "reasonably" (not much is required). The reason is that > glibc contains special workarounds for pthreads.
I've now changed debian/rules in svn to use the pthread backend for the architectures where we see test failures. Could someone please take a look if this make sense? Cheers, gregor -- .''`. http://info.comodo.priv.at/ -- GPG key IDs: 0x8649AA06, 0x00F3CFE4 : :' : Debian GNU/Linux user, admin, & developer - http://www.debian.org/ `. `' Member of VIBE!AT & SPI, fellow of Free Software Foundation Europe `- NP: Nick Drake: Way To Blue
signature.asc
Description: Digital signature
