On Tue, Jul 29, 2003 at 08:10:21PM -0400, Sean Davis wrote:
> To followup...
> 
> I just updated my sources, and rebuilt a kernel with DDB (DDB was the one
> and only thing I added to the conf, nothing else changed) and now pcictl
> pci1 no longer deadlocks the machine. I tried it as root, as a normal user,
> and on serial console. When it happened the first time, I was ssh'd in as a
> normal user. So either the bug went away in some recent commit to sys/ that
> affected this alpha, or it took a stranger set of circumstances than I
> thought and pcictl pci1 list just happened to be the last straw, as it were.
> Anyway, unless anyone feels this warrants further looking into, the PR can
> be closed.
> 
> On Tue, Jul 29, 2003 at 11:11:00PM -0000, gnats-admin@netbsd.org wrote:
> > Thank you very much for your problem report.
> > It has the internal identification `port-alpha/22301'.
> > The individual assigned to look at your
> > report is: port-alpha-maintainer. 
> > 
> > >Category:       port-alpha
> > >Responsible:    port-alpha-maintainer
> > >Synopsis:       running pcictl pci1 list causes deadlock on an AlphaStation 600 5/266
> > >Arrival-Date:   Tue Jul 29 23:11:00 UTC 2003
> > 

Following up to the followup... with a kernel without DDB, it happens again.
Current sources. I added pseudo-device wsmux, but I don't see why that would
have anything to do with a problem that occured before it was in the kernel.
This time, it hung on pcictl pci0 list, not pcictl pci1 list. This leads me
to believe there is a bug lurking either in the pcictl code, or in the pci
code... I dunno...

-- 
/~\ The ASCII
\ / Ribbon Campaign                   Sean Davis
 X  Against HTML                       aka dive
/ \ Email!