cvs commit: src/sys/dev/kbdmux kbdmux.c

[Maksim Yevmenkin]

On 3/1/06, Scott Long <scottl at samsco.org> wrote:
> Maksim Yevmenkin wrote:
> > On 3/1/06, Scott Long <scottl at samsco.org> wrote:
> >
> >>Eric Anderson wrote:
> >>
> >>>Ceri Davies wrote:
> >>>
> >>>
> >>>>On Wed, Mar 01, 2006 at 12:39:36AM -0700, Scott Long wrote:
> >>>>
> >>>>
> >>>>
> >>>>>Coleman Kane wrote:
> >>>>>
> >>>>>
> >>>>>
> >>>>>>On 2/28/06, *Dag-Erling Sm?rgrav* <des at des.no <mailto:des at des.no>>
> >>>>>>wrote:
> >>>>>>
> >>>>>>   "Maksim Yevmenkin" <maksim.yevmenkin at gmail.com
> >>>>>>   <mailto:maksim.yevmenkin at gmail.com>> writes:
> >>>>>>    > we should make a decision which scancodes keyboard drivers
> >>>>>>should be
> >>>>>>    > returning and change kbdmux(4), syscons(4) etc. to support it.
> >>>>>>    >
> >>>>>>    > until then, i'm afraid, sun rs232 and powerpc adb keyboard
> >>>>>>drivers
> >>>>>>    > will have to translate their own scancodes into at scancodes
> >>>>>>(just
> >>>>>>    > like ukbd(4) does).
> >>>>>>
> >>>>>>   Considering that ukbd(4) is the only kbd driver that runs on all
> >>>>>>   platforms, it might make sense to have all other drivers translate
> >>>>>>   their scancodes to ukbd(4) scancodes and use that in kbdmux(4) and
> >>>>>>   syscons(4).
> >>>>>>
> >>>>>>
> >>>>>>I agree,  especially since atkbdc's are rapidly being replaced by
> >>>>>>USB on the platforms they were made for.
> >>>>>>
> >>>>>
> >>>>>Long term, yes, but what is the risk/benefit of doing this for 6.1?
> >>>>>
> >>>>
> >>>>
> >>>>If it doesn't work, we look like bloody idiots.  6.2 isn't that far
> >>>>away, I'd think that anybody could manage waiting for it a little more
> >>>>(especially if it goes into RELENG_6 just after the 6.1 branch).
> >>>>
> >>>>The above not to be taken as a comment on the change, on which I am not
> >>>>informed enough to hold an opinion.
> >>>>
> >>>
> >>>
> >>>Can we have it all set, built in to GENERIC, but the device hint set to
> >>>disabled?  Lots of systems, Dell PowerEdge and similar, come with a
> >>>remote access card, which uses a USB tie in.  The only way I've found to
> >>>make the console keyboard work after initial install, is to boot single
> >>>user, hack /etc/devd.conf, and reboot, which is really ugly.  Many
> >>>FreeBSD users get stuck with this, so it would be very nice to have a
> >>>quick way to enable the kbdmux at loader time for 6.1, but if it goes in
> >>>right after, I'm not going to cry. :)
> >>>
> >>>Thanks again Maksim for all the work on this!
> >>
> >>I'd like to turn this on for the next BETA build, get feedback, then
> >>decide whether or not to leave it enabled for the final 6.1 release.
> >>Maksim, is that OK with you?
> >
> >
> > yes, it should be fine. i still need to add/test hints support in
> > kbdmux(4). i plan to finish it today. i will also need to MFC all my
> > changes to RELENG_6. i'm not sure how much testing we want before MFC.
> > should i get a blanket approval from re@?
>
> Instead of rushing this in, let's build BETA3 now without it and then
> spend the next week getting it working at tested and polished.

fine with me. more testing is always good. btw, kbdmux(4) hints
support was committed to HEAD. it was actually pretty trivial to add.
i briefly tested it with module and it worked for me.

thanks,
max