Chris Tribo wrote:
>
> On Sep 12, 2006, at 10:18 AM, Garrett D'Amore wrote:
>
>> computer_help@rferl.org wrote:
>>> ...
>>> -I/usr/current/src/sys/lib/libkern/../../../common/lib/libc/quad
>>> -I/usr/current/src/sys/lib/libkern/../../../common/lib/libc/string
>>> -I/usr/current/src/sys/lib/libkern/../../../common/lib/libc/arch/arm/string
>>>
>>> -c /usr/current/src/sys/kern/kern_todr.c
>>> cc1: warnings being treated as errors
>>> /usr/current/src/sys/kern/kern_todr.c: In function 'todr_gettime':
>>> /usr/current/src/sys/kern/kern_todr.c:288: warning: implicit
>>> declaration of function 'todr_debug'
>>>
>>> *** Failed target:  kern_todr.o
>>> *** Failed command: echo '# ' "compile Astro/kern_todr.
>>>
>>> Pavel
>>
>> Thats really, really weird, since since todr_debug is either a function,
>> or is a macro.  I've test compiled this both ways for architectures
>> other than hpcarm.  I'll do an hpcarm compile and see if it is busted.
>>
>
> Something is definitely hinky, even on i386:
>
> WARNING: preposterous TOD clock time
> WARNING: using filesystem time
> WARNING: CHECK AND RESET THE DATE!
>
> The BIOS clock is fine, but the system time is off by more than 8 hours.
>

Hmm.... have you updated kern_todr in the past 12 hours?  There was a
bug where some rtcs were passing a bogus value for the weekday.  I
removed that check, and it is better now.

Failing that, please #define TODR_DEBUG (or stick it in your kernel
options) and update to the very latest code.  It will dump the exact
values being read from the system clock.

-- 
Garrett D'Amore, Principal Software Engineer
Tadpole Computer / Computing Technologies Division,
General Dynamics C4 Systems
http://www.tadpolecomputer.com/
Phone: 951 325-2134  Fax: 951 325-2191