[Simh] VAX RTC implemented ?

[A.Bucher at alcatel.de]

>as far as I know, there is a battery buffered RTC in the VAXes, which
keeps
>running even if the VAX is powered off.
>
>Is this RTC somehow implemented in SIMH ?
>
>Why I ask:
>Using SIMH on a Laptop, I tried to hibernate it - works without problems
>(ok, except for telnet connections and some network related things) - but:
>the time within the emulator stays as it was before the laptop was sent to
>sleep.  Besides, very convenient to have a "VAX to go" in the pocket :-)
>
>Okay, you may say that there may be some NTP or similar client for VMS,
but
>what's for older or non-networked OSes ?
>
>Therefore, is it possible to have the simulated RTC updated from the hosts
>clock ?! Or did I just miss something ?
>
>Thanks you for the assistance !
>_______________________________________________

I run multinet on my emulation and with it I can do a    $ multinet
rdate 192.168.0.4
and it will sync the time to the node specified.  I don't know if you
can do something
simuliar with the tcp/ip package you have installed but might be worth a
try.
Maybe you could create a job to run fairly often to resync the clock
this way if
you want it to be automatic or if ntp is supported you could try that -
but I suspect
ntp wouldn't sync if the time is off by a large amount - you would need
to rdate
or ntpupdate the time first to get it close before ntp would sync up.
>_______________________________________________


Hi,
Rick, thanks for the hints.
And, of course, as Bob said, VAX/VMS does not support any hibernation
concept, I know :-)

But, in my opinion, hibernation of the host system should not affect the
contents of the emulator !
(see explanation below)

I am running the DEC TCPIP 5.0 package, and I think there is
no time server client - but I did not check thoroughly, I've to admit.
Neither did I check
the shareware/freeware available for time sync.

But I thought of something different:
As far as I remember from the 11/750 HW docs, the machine has two clocks:
a) the RTC, a battery backed-up clock, which runs permanently
b) an interval timer (or something similar), which is used for the VMS
time.
right ?
I am not sure, but I think all VAXes use the same.

Now, if you set the time (with VMS), or enter time/date at bootup, both
"clocks" are updated (I mean SET TIME=xxx).

At boot of VMS, the OS checks if the RTC No. a) (Battery backed up one) is
valid (if not, it asks for date/time and sets it correct) and loads this
time into the timer No. b), which is used by the OS.
If you do a SET TIME w/o parameters, the same effect occurs.

If this is true, the emulator could update this RTC No. a) from the host OS
in intervals and would therefore always deliver a correct time, whether it
was hibernated or not.
To correct the time within VMS, it would be sufficient to issue SET TIME
commands in frequent intervals. For other OS, similar things would be
possible.

Additionally, of course, the time sync between hosts should be done for
running systems, but this is another story.

I am only addressing my suspection that the emulated clock registers of
simh are only updated at startup of the emulation.

Regards,
Andreas