🚀 go-pugleaf

ajudge@maths.tcd.ie (Alan Judge) writes:

>However, the primary problem turned out to be the MIPS kernel.  All
>MIPS workstations have time of day chips, and it seems that the kernel
>intermittently checks this clock and adjusts the CPU clock accordingly
>(both by using an internal adjtime(), and by changing the basic speed
>of the CPU clock causing the "Adjust CLOCK Hz" messages in syslog).
>Obviously, this has bad interactions with the correct operation of
>NTP, which is doing something similar based on NTP clocks outside.
>Each time the kernel does this, NTP has a fit.

>It seems from the code that this has always been done, even on R3000
>machines and in RISCos 4.51, so NTP has always been a bit funny

>There is no MIPS-provided way to turn this off (like Sun's
>dosynctodr).  However, MIPS do provide source code for all their
>drivers with RISCos, so it is a fairly simple patch on most MIPS boxes

I've run xntp3 on CDC's EPIX 1.4.3 on R3000 and R6000 boxes. EPIX 1.4.3
is based on RISCos 4.52. Under EPIX 1.4.3 its just the same, we always
got the funny "Adjust CLOCK Hz" messages as well as "Previous time adjustment
didn't complete messages". I've now disabled the verify_time 
code in /usr/reconfig/io/todc_r{3030,6000}.c, and gee xntp runs more stable
now. No more didn't complete messages, and very good synchronization
to our stratum 1.

Before I had offset's in the 5-15 msec range, now i get offsets in the
0.1-2 msec range!!!!! (On a CD4320, aka RC3230.)

That's great !

>I have put in a request for an easier way to disable the clock syncing.
I will put in a similar request to CDC.

Sincerely,
Klaus
--
Klaus Steinberger               Beschleunigerlabor der TU und LMU Muenchen
Phone: (+49 89)3209 4287        Hochschulgelaende
FAX:   (+49 89)3209 4280        D-8046 Garching, Germany
Internet: Klaus.Steinberger@Physik.Uni-Muenchen.DE