In article <Pine.NEB.4.58.0312311719230.15848@villain.mintex.demon.co.uk> you write:
>On Wed, 31 Dec 2003 chris@dokein.co.uk wrote:
>
>> > After a couple of solid days' hacking, I seem to have got the vidcaudio(4)
>> > driver to work, both on a VIDC20 (8-bit, mu-law) and an ARM7500FE (16-bit,
>> > linear).  I don't know whether it's ever worked before, but there's
>> > certainly precious little of the old driver left.  The new version is in
>> > NetBSD-current, and people should test it.  Risc PCs with 16-bit sound
>> > aren't supported, largely because I haven't got one, so I don't know how
>> > to reliably detect the 16-bit sound hardware.
>>
>> I believe the easiest method for detecting 16bit is to actually do what
>> RISC-OS does, which is rely on the user setting an entry in the CMOS
>> correctly.
>
>Do we currently have a utility in NetBSD that is able to read/write CMOS
>locations from userland?

No, unless you count dd(1).  I suspect perverting eeprom(8) might be the
right way to go if we want one.

-- 
Ben Harris                                                   <bjh21@NetBSD.org>
Portmaster, NetBSD/acorn26           <URL:http://www.NetBSD.org/Ports/acorn26/>