Re: Call for bge(4) testers

On Thu, Nov 19, 2009 at 07:32:27PM -0300, Gonzalo Nemmi wrote:
> On Thu, Nov 19, 2009 at 5:37 PM, Pyun YongHyeon <pyunyh_at_gmail.com> wrote:
> 
> > On Thu, Nov 19, 2009 at 01:27:03AM -0300, Gonzalo Nemmi wrote:
> > > On Wed, Nov 18, 2009 at 8:32 PM, Pyun YongHyeon <pyunyh_at_gmail.com>
> > wrote:
> > >
> > > > On Wed, Nov 18, 2009 at 01:40:24AM -0300, Gonzalo Nemmi wrote:
> > > >
> > > > [...]
> > > >
> > > > > I just tried ...
> > > > > echo 'hw.bge.allow_asf="1"' >> /boot/loader.conf
> > > > > reboot
> > > > > load if_bge
> > > > > acpiconf -s3
> > > > > same results :(
> > > > >
> > > >
> > > > Ok, here is new try.
> > > >
> > >
> > > Let's get on to it then !
> > >
> > > Well now, the situation has improved .. here's what I got:
> > > http://pastebin.com/f2d152f91
> > >
> > > lines 2 to 8 == kldload if_bge
> > > line 9 == acpiconf -s3
> > > lines 10 to 18 == resume (notice there are only 2 messages now: lines 12
> > and
> > > 13!)
> > > lines 19 to 36 == ifconfig bge0
> > > lines 37 to 42 == me mounting the pendrive to get the messages from
> > > /var/log/messages =P
> > > lines 43 to 46 == kldunload if_bge0
> > >
> > > I think you narrowed it down quite a lot this time !
> > >
> >
> > Hmm, not actually. I guess it just removed unnecessary operation
> > for PHY but it still failed to get functional state. :-(
> >
> > > I have to warn you though, that this time, my kernel was compiled using a
> > > ... certainly modified /etc/make.conf file ... just in case you need to
> > know
> > > how it looks like, you'll fin it in here:
> > > http://pastebin.com/f42e356d2
> > >
> > > If you think that make.conf config may interfere with your pourposes or
> > > tests, just let me know and I'll use a default one.
> >
> > I think that's ok.
> >
> > > The good thing about that make.conf is that it saves me quite a time on
> > > every recompile ;)
> > >
> > > > im at your service .. tell me what to do and I'll do it :)
> > > > >
> > > >
> > > > Thanks a lot for your patience and continuous support to fixing
> > > > bugs.
> > > >
> > >
> > > Thank _YOU_ for keeping the good work up and for trying to solve a really
> > > nasty bug that makes every bge(4) user (think of it
> > > _as_every_dell_notebook_owner_) unable to get his laptop to resume .. or
> > > even use FreeBSD altoghether just because of this.
> > >
> > > Dear Pyung, rest assured that as long as you remain commited to fix this,
> > or
> > > any other bug that I can help you with, you can count on me to do
> > everything
> > > that may be within the reach of my hand. As long as you remain commited,
> > > I'll be there, commited just as well :)
> > >
> >
> > Thanks a lot! This really helps me a lot to fix the bug.
> > Here is 4-th try. I added a couple of debug messages to see what
> > register contents it have after resume.
> >
> > Thanks in advance.
> >
> 
> Pyung, I have problems applying your patch. 4 out 13 hunks fail.
> Take a look in here.
> http://pastebin.com/f7eab233f
> 
> I tried to recompile anyways but buildkernel failed at bge ...
> 
> Can you tell me what to do to apply it cleanly and recompile?
> 

Hmm, I guess it was caused by the difference between HEAD and 8.0.
I've uploaded complete bge(4) files to the following URL.
http://people.freebsd.org/~yongari/bge/resume/if_bge.c
http://people.freebsd.org/~yongari/bge/resume/if_bgereg.h

Save your original bge(4) files in sys/dev/bge directory and
download above files and then rebuild.

Hope this helps.

> Best Regards
> Gonzalo Nemmi