* Wouter Verhelst wrote: > Hi, > > I have such a card, but it's sitting in a machine that runs Sarge (and > I'm not going to update it to etch before that's released). > > If it would help, I'd be willing to run some tests on that box. Just > tell me what you want me to do and what information you need back, and > I'll do so.
I've been asked to look at this bug as part of my T&S for the NM process. Wouter, could you try rebuilding glide with -fno-strict-aliasing? I have attached a diff that patches debian/rules to add that to CFLAGS. This is mainly just guessing but the same solution did fix a similar bug in mesa recently. If that doesn't work, perhaps you could try to get a backtrace of running one of the segfaulting binaries against a non-stripped glide library (i.e. rebuild with something like "DEB_BUILD_OPTIONS=nostrip debuild"). Thierry
signature.asc
Description: Digital signature