Subject: Re: we driver still broken.
To: Feico Dillema <dillema@huygens.org>
From: Alfred Perlstein <bright@hotjobs.com>
List: netbsd-help
Date: 11/24/1998 12:10:54
Ok, i'll be checking the list archives more often, sorry for the noise.

Compiling if_we.c sans -Ox workes fine for me.

I think i'll install gcc 2.7.2.2 in /usr/local/ to do my kernel compiling
from now on.

Thanks for the help,
Alfred Perlstein - Programmer, HotJobs Inc. - www.hotjobs.com
-- There are operating systems, and then there's FreeBSD.
-- http://www.freebsd.org/                        3.0-current

On Tue, 24 Nov 1998, Feico Dillema wrote:

> Your message dated: Tue, 24 Nov 1998 10:26:35 EST
> >On Tue, 24 Nov 1998, Feico Dillema wrote:
> 
> >I'm booting right now to check tht i don't have any overlapping drivers, i
> >should have done this a while back, however i did compile my kernel and
> >thought i had taken care of all conflicts.
> 
> Well, in theory it should work that my CDROM and we0 share an interrupt
> (edge-triggered E+) (at least that's what I've been told and what I 
> understand from the code). Tomorrow I may have access to this machine
> again, and I can try a -current kernel with an ugly hack that prevents
> this interrupt sharing. That might show whether the we0 driver is
> to blame and broken...
> 
> >Last night i compiled a debug kernel and tried to trace through the code,
> >however, i'm a bit inept with using the kernel debugger and couldn't
> >figure out how to get a source listing of where i am in the kernel.  nor
> >how to jump around stack frames to examine local variables in functions :(
> >
> >Need more practice with this thing. :)
> Me too. It's still a black art to me, kernel debugging that is.
> 
> >I wasn't aware of this issue, I think i'm going to subscribe to -current
> >and review the mailing list archives.
> It's a good idea to subscribe to -current, at least it has thought me
> a lot and healed some wounds afflicted by using bleeding-edge technology 
> ;-}. I've sent you  the mail about that issue, so you don't have to 
> start digging in the archives... 
> 
> >btw, any word on the ultrasparc port?
> No idea, I'm just a user. Although I'd be interested to remove Slowlaris
> from some Ultrasparcs in our lab here, and raise some interesting
> political issues here ;-}.
> 
> Feico.
>