In message: <[EMAIL PROTECTED]> Daniel Eischen <[EMAIL PROTECTED]> writes: : On Thu, 19 Sep 2002, M. Warner Losh wrote: : : > In message: <[EMAIL PROTECTED]> : > Joe Marcus Clarke <[EMAIL PROTECTED]> writes: : > : On Thu, 2002-09-19 at 23:38, M. Warner Losh wrote: : > : > In message: <[EMAIL PROTECTED]> : > : > Joe Marcus Clarke <[EMAIL PROTECTED]> writes: : > : > : On Thu, 2002-09-19 at 23:22, M. Warner Losh wrote: : > : > : > In message: :<[EMAIL PROTECTED]> : > : > : > Julian Elischer <[EMAIL PROTECTED]> writes: : > : > : > : Is it statically linked? : > : > : > : > : > : > Yup. : > : > : : > : > : You'll need to rebuild against the new libc_r, then. : > : > : > : > Yuck. I'd rather nail jello to a tree than try to update a port that : > : > complicated. : > : : > : If it makes you feel better, I just ran a compile on my -CURRENT : > : machine, and it went fine. Both mozilla and mozilla-devel should build : > : out of the tree....with or without jello. : > : > Do you know if this breaks libc_r.so.4.0 too? I'm thinking of just : > grabbing the build from Mozilla.org... : : I think it does since it's a kernel ABI breakage in ucontext_t's : and sigreturn(). I think we either need to reformat the ucontext_t : in some way that is compatible with the old format, or make a : new sigreturn() using the current one for compatibility. It's : a shame 'cause it's only broken for i386, not the other archs.
I think that would be an excellent idea. I think that there are enough libc_r applications for 4.x that we need to retain binary compatibility at this level. Otherwise we're going to be raked over the coals. Warner To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
