On Sat, Mar 10, 2007 at 03:21:43PM +0100, Bernd Ernesti wrote:
> On Sat, Mar 10, 2007 at 11:11:27AM +0100, Manuel Bouyer wrote:
> > On Sat, Mar 10, 2007 at 09:03:44AM +0100, Bernd Ernesti wrote:
> > > On Fri, Mar 09, 2007 at 09:16:00PM +0100, Pavel Cahyna wrote:
> > > > I do not understand it:
> > > > 
> > > > On Fri, Mar 09, 2007 at 06:59:55AM +0000, Jeff Rizzo wrote:
> > > > > For the benefit of modern hardware, enable ACPI by default in INSTALL and
> > > > > GENERIC kernels. If ACPI is an issue on your hardware, 'boot -c' and
> > > > > 'disable acpi' should be a workaround. ACPI-enabled kernels works fine
> > > > > on pre-acpi hardware.
> > > 
> > > Again, I already raised my concerns, because this will not work on all systems.
> > > An usb keyboard does not work with 'boot -c'.
> > 
> > It worked for me on a Dell 2950 (this one doesn't boot with acpi). As
> > long as the hardware provides legacy keyboard emulation, I can't see
> > why boot -c would not work. The USB controllers have not been touched
> > at this point.
> 
> It does not work for me if the pckbc (maybe it is pckbd) driver is in the kernel, 
> and this is IMHO a known problem.
> Same for DDB which doesn't work either in this case.

For ddb it's expected, as it's typically called after autoconf is completed.
userconf is called before autoconf has started. I didn't try but I suspect
boot -d would also work on the dell.

So OK, there are some system where boot -c won't work. Should we add NOACPI
kernels for amd64 ?

-- 
Manuel Bouyer <bouyer@antioche.eu.org>
     NetBSD: 26 ans d'experience feront toujours la difference
--