> > On Sat, 6 Dec 1997, Chris wrote:
> >
> > > After upgrading my squake to 1.0 last night, I was still sad that it still
> > > coredumped on my mmx. Today, after upgrading to svgalib 1.2.11, the
> > > problem has disapeared. I can load squake as many times as I like. Just
> > > thought i'd say this...
> >
> > after upgrading to 1.2.11, to "support" my ViRGE card, everything works
> > great... except
> >
> > zgv will lock up my system solid... but not all the time, often it will
> > cause DMA errors on my first IDE disk, and i can Ctrl-C out of it..
> > sometimes it will work great, and (often after 4+ days) uptime, it will
> > lock solid, nothing in syslog, sysrq doesnt seem to work, and i have to
> > cold reboot. the chances that it completly locks seems to be amplified if
> > i am using my soundcard at the same time.
>
> If it's any help, loading ViRGE into memory-mapped IO mode seems to setup
> a situation where the IDE (if it's in DMA mode) locks on triton (and
> maybe others).
>
> If anyone can figure this one out, the GGI folks would seriously
> appreciate it!
> AFAIK this is still a problem... at least there's some kind of conflict
> between ViRGE and Triton-IDE.....
Would it help to disable the Triton support in the kernel. It may be a
hackish (maybe not the right word...) solution, but should work unless
someone desperately needs Triton support.
>
> I don't think this is particularily ontopic for linux-kernel though...
Probably is... the kernel gurus would be able to help with the problem.
Maybe the person/group of people who worked on the Triton support in the
kernel could help you straighten things out.
> except that it probably hits a few people out there - and someone may have
> a solution (please?)
>
> G'day, eh? :)
> - Teunis
>
>
----------------------------------------------------------------------
Phil Brutsche
"Be of stout heart, Number One. We've handled the Borg. We can
certainly handle Admiral Jellico." - Jean-Luc Picard
----------------------------------------------------------------------