Re: Possible binary problem in current X release
On Sam, 2002-12-21 at 14:04, Raffaele Sandrini wrote:
> Michel Dänzer schrieb:
>
> >On Fre, 2002-12-20 at 23:47, Raffaele Sandrini wrote:
> >
> >>Michel Dänzer schrieb:
> >>
> >>>On Fre, 2002-12-20 at 08:40, Raffaele Sandrini wrote:
> >>>
> >>>>I think i found a binary error in the current X4.2 Debain distri:
> >>>>
> >>>>I get a symbol missmatch when i use the "s3" driver. Before Debian released
> >>>>X4.2 debs i used the precompiled native XFree86 4.2 version. I had n
> >>>>
> >>>>problems with that one (also with the s3 driver).
> >>>
> >>>The server seems to work though? Unresolved symbols usually aren't a
> >>>problem.
> >>
> >>No, the server does NOT run. It terminates with this error.
> >
> >What error? It doesn't fail due to the unresolved symbol, in fact the
> >log doesn't show any error that would cause the server to fail. It looks
> >like the server works perfectly, if it terminates immediately that's
> >probably a problem with the client(s).
> >
> Hmm this makes that whole thing a lot stranger. I dirscribe my situation
> here: I'm talking about an Debian Installation wich is distributed to
> many and different workstations. On every workstation ecxept the one
> wich uses the "s3" module, the whole thing works perfect. And if i take
> a native installation,
Native installation of what? Binary tarballs from xfree86.org? Do you
only replace the server side with those?
> then also the "s3" module works perfect. That means to me that there must be
> something wrong with the debian packs.
If replacing xserver-xfree86 only (or even only the _drv.o) fixes the
problem, that would point there, yes.
> What does then this message "Failed to set up write-combining range
> (0xff000000,0x100000)" mean?
Failure setting up write-combining isn't a fatal error but just means
performance won't be optimal.
--
Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer
XFree86 and DRI project member / CS student, Free Software enthusiast
Reply to: