Re: Problems with clock on 1.1-CURRENT

["Freddie Cash" <fcash-ml@xxxxxxxxxx>]

> :Re-syncing the clock is not the problem.  :)  Having a CPU that is
> :running at hyperspeed and causing heat problems is.  When the system
> :clock starts running faster, the whole system starts running faster.
> :I've watched the P166 run through a buildworld in very little time
> :(around 30 minutes wall time).  The resulting binaries don't work,
> :though.  Not to mention the system becomes unusable as things like
> :screensaver timeouts are in the sub-minute (wall time) range.

> :Once the clock is out-of-sync like that, the only solution is to
> :reboot the system.  Otherwise, heat buildup will most likely kill it.

>     Well, I'm pretty sure it isn't running at hyperspeed.  A pentium
>     166 doesn't really have any low power / clock enhancement modes
>     that DragonFly could twiddle anyway.

Yeah, it's not really, but it certainly seems like it.  :) Watching a
day flash pass in less than an hour is interesting.  Couldn't figure
out why the system felt so sluggish until I noticed there were about 7
instances of the daily periodic scripts.  :)

>     But the clock issue is one I'd like to solve.  There's something
>     going on somewhere but I haven't yet been able to reproduce it
>     the problem.

>     When you are experiencing the clock issue are you doing anything
> that
>     might trigger BIOS calls, such as video bios calls or ACPI events
>     (sleep mode, that sort of thing) ?  Or does it appear to solely be
>     triggered by a heavy load?

On the P166, it occurs whenever I run a buildworld.  And I've had it
happen a few times during a cvsup of the ports tree using a Linksys
16-bit PCCard NIC using the ed(4) driver.

On the Celeron, it occurs about every third time I start XFree86.  I
can start top in one virtual terminal and watch as XFree86 loads, and
just as KDE starts initialising devices, the clock goes crazy.  It
also happened once during the initial buildworld (the make output just
flew by onscreen, a hell of a lot faster than it should have).

The P166 if a Fujitsu Lifebook 765DX with a Trident CyberBlade 2M
video card, a 2.1 GB UDMA 33 harddrive, and 80 MB RAM.

The Celeron is a Seanix Sea-Note with an S3 SavagePro PN133 video card
configure to use 32 MB of RAM as video RAM, a 20 GB UDMA100 harddrive,
and 256 MB RAM.

The laptops are at work right now.  I can make dmesg and vmstat output
available tomorrow if need be.

-- 
Freddie Cash, CCNT CCLP        Helpdesk / Network Support Tech.
School District 73             (250) 377-HELP [377-4357]
fcash-ml@xxxxxxxxxx