Hi Alexander, Alexander Zimmermann <[EMAIL PROTECTED]> wrote: > There are no segmentation faults anymore, but lots of > signal SIGPIPE, Broken pipe > at random places.
SIGPIPE is a normal signal which occurs all the time. If you want to run GTKG within gdb you have type this command before running it: ha SIGPIPE noprint nostop pass > I switched to the gtk2 version and this seems to work. > (But debugging this version failed due to an internal gdb error). It's quite normal that gdb is a little broken on non-x86 platforms. You can ignore those internal errors by answering both questions with n(o). I'm not sure whether you want to continue running GTKG after this, but for getting a backtrace etc. it's not a problem, usually. > If someone's interesseted I can do some additional debugging of the > gtk1 version, but I'm not sure if this is necessary or desired. Sure, the gtk1version is not deprecated and GTKG is only occasionally tested on Solaris. BTW, someone else reported nearly the same bug as you. Instead of SIGSEGV he wrote that it crashed by SIGBUS. You're probably using a different CPU or even gdb reported a wrong signal(?), so the actual signal used to propagate the exception might be different. Maybe your bug is already fixed in CVS. Christian
pgp00000.pgp
Description: PGP signature
