[TriLUG] Re: NVidia drivers cause Segmentation Faults in RH9 and Fedora Core beta

Roberto J. Dohnert webwarrior at gnu-darwin.org
Thu Oct 2 14:49:19 EDT 2003


Hey guys here is an update:

I reinstalled Red Hat and Fedora Core and started using the updated
drivers that NVidia distributes.  It didnt work properly, still same thing
was happening with no change, segmentation faults everywhere and it was a
nightmare returned.  I took the machine to CompUSA and had a diagnostic
done, everything from their tests appeared it worked fine, by the way, if
you ever have to take it to Comp make sure you have a Windows Dual boot,
Funny thing was when i booted into Windows at the store, little lines
started to appear on the screen, colors and resolution started switching. 
So I went ahead and bought a new NVidia card, I wont use ATI and I wont
use any of those el cheapos everyone else sells. Everything works fine now
under Red Hat Linux 9, Open Office runs, XMMS runs fine, everything starts
and boots like it should except for Jahshaka binary on the web site but I
did a compile from source and it started working.  I also switched out the
ram since they had some really cheap and the RAM I had was old.  I
stressed tested the whole thing and it seems okay.  But Fedora Core still
acts funny, it will boot and now I get GDM and it will boot into the OS
but it takes exactly 52 minutes to boot from the NVidia splash screen, the
techs at NVidia when i called back told me it had something to do with the
modes and that as soon as the specifics of Red Hat 10 aka Fedora Core came
out that they will make the enhancements, I told them that Fedora Core was
already available as a download for Beta, they said they will not write
any new drivers based on a beta build but that they wait for the final
release of the OS.

> On Wed, Oct 01, 2003 at 05:25:35PM -0700, Chuck Wolber wrote:
>> Strange, I've been running RH9 under my GF4 4600 Ti for months with
>> the NVIDIA 4363 drivers without a single problem.
>
> My experience with 1.0-4[34]* for Linux-x86 has shown that a few
> multithreaded apps misbehave when compiled and run on some combination
> of RH9-nonRH9 systems. Of course not all apps fall into this enumeration
> which probably explains why not everyone has been "hit"...
>
> --
> Daniel T. Chen          crimsun at fungus.sh.nu
> GPG key:   www.sh.nu/~crimsun/pubkey.gpg.asc


-- 
To view my Linux help page and howto's goto

http://www.geocities.com/kane121975/

I have information on x86 Linux and PowerPC Linux





More information about the TriLUG mailing list