:> >$ uptime; last -4 reboot shutdown
:> >12:32PM up 2 days, 14:46, 2 users, load averages: 0.00, 0.03, 0.00
:> >reboot ~ Sat Apr 3 11:53
:> >shutdown ~ Wed Mar 31 21:46
:> >reboot ~ Wed Mar 31 20:46
:> >shutdown ~ Sat Mar 27 18:59
:>
:> Hello YONETANI-san,
:>
:> I'm seeing a bug in DragonFlyBSD that was caused by the fix to your
:> uptime problem in April. Now, if you call settimeofday(2), such as
:> using 'ntpdate -b', the uptime is set to twice the real uptime.
:> boottime is definately set incorectly.
:>
:> I wonder, if you revert kern_clock.c:1.19 (i.e. set boottime =
:> basetime) do you still see your problem from April 2003? I can't
:> seem to reproduce it at all.
:
:I've confirmed that it doesn't. At lease uptime, ntpdate and ps are working
:correctly. I'm not 100% sure that this change affects other parts of the
:system that depend on this twisted boottime though.
Hmm.
I think the issue is settimeofday() not compensating the boottime
properly. I'll commit a fix for this today. I'd prefer not to
revert that rev but I'm investigating the issue now.
-Matt
Matthew Dillon
<dillon@backplane.com>