Re: Problems with TOD clock data on i386?

["John D. Baker" <jdbaker%mylinuxisp.com@localhost>]

On Mon, 16 Jul 2012, Brian Buhrow wrote:

>         Hello.  Probably a silly question, but is it safe to assume that the
> kernel doesn't update the TOD clock since it can't read it?  The only way I
> can think to truly test this is to set the time in the BIOS  to some time last
> week, reboot, let ntpdate set the time, and then drop into BIOS again and
> see if the clock is correct.
> I'm guessing the BIOS will still show you the time you set the clock to
> originally, plus the amount of time that passed between both BIOS sessions.
> If it shows the current time, and you didn't boot into another OS between
> BIOS sessions, then you're able to set the clock somehow, but can't read
> it, which would be even stranger but which might be useful information to
> have.

It it definitely strange.  I did this.  Set the TOD clock to about a
month ago, rebooted single-user then back to BIOS a couple of times to
make sure it stayed set that way, then rebooted multiuser so ntpd would
start up.

I used it for about an hour or so, checking the ntp peers to make sure
it had syncronized with and selected my time local time server, then
rebooted.

When I dropped into the BIOS setup, the TOD clock reflected current
UTC time.  So, it can set the time in the TOD clock, but seems unable
to initialize itself from it.  Is there debugging to see what it
thinks it read (and rejects as preposterous)?

--
|/"\ John D. Baker, KN5UKS               NetBSD     Darwin/MacOS X
|\ / jdbaker[snail]mylinuxisp[flyspeck]com    OpenBSD            FreeBSD
| X  No HTML/proprietary data in email.   BSD just sits there and works!
|/ \ GPGkeyID:  D703 4A7E 479F 63F8 D3F4  BD99 9572 8F23 E4AD 1645