>I admit I only briefly looked at the latest code, but afaik
>especially with the 'demand timer patch' Linux uses the TOD
>to tell time, and I don't recall seeing the drift corrections
>applied. This would mean an ntpd cannot adjust time of Linux
>running on z/VM, but one might be able to change that. I can
>even imagine hwclock to issue SCK instruction to update the
>virtual machine's private offset of the VM TOD clock (to
>live through a reboot, but not logoff).

You have looked very briefly ;-) Linux uses the TOD clock to set
the internal xtime at ipl. Afterwards Linux uses the clock
comparator to get an interrupt every 1/100 second. In the interrupt
handler this "tick" is added to xtime by calling the common code
function do_timer(). This is where ntp time adjustment is taking
place as well. It is completly indepedent from the architecture.
The timer patch optimizes the code a bit so that it can account
for more than a single tick in one go. This is the basic idea
behind the timer patch: stop the HZ timer until something relevant
happens and then catch up with the ticks/xtime.
In short the ntpd time adjustment should work just fine on s/390.

blue skies,
   Martin

Linux/390 Design & Development, IBM Deutschland Entwicklung GmbH
Sch�naicherstr. 220, D-71032 B�blingen, Telefon: 49 - (0)7031 - 16-2247
E-Mail: [EMAIL PROTECTED]

Reply via email to