[cctalk] Re: Two items for RSTS/E

2023-11-21 Thread Paul Koning via cctalk



> On Nov 20, 2023, at 10:05 PM, Tim Sneddon via cctalk  
> wrote:
> 
> On Tue, 21 Nov 2023 at 06:13, Paul Koning via cctalk 
> wrote:
> 
>> I just pushed two additions to https://github.com/pkoning2/decstuff :
>> 
>> In "patches" a new patch for the DEUNA driver.  This fixes a problem seen
>> when doing user (as opposed to DECnet) I/O, as well as two errors that show
>> up when using units beyond the first.
>> 
>> Directory "ntp" is new.  This is a simple NTP protocol client for RSTS,
>> which will synchronize the system clock with an NTP server on the LAN.  It
>> includes handling of timezone rules, so the right thing will happen at
>> daylight savings time (summer time) boundaries.  The clock is maintained to
>> the full RSTS resolution -- typically 1/50th or 1/60th second, but can be
>> as low as 10 ms if the KW-11/P clock is used.
>> 
> 
> This is awesome! Thanks, Paul.

Glad you like it.  Comments welcome.

When I first posted this I still had trouble with the QNA driver.  Those are 
now fixed (patch file patches/xhdvr.cmd) so it now works on both UNA and QNA 
devices.

I've tested this in SIMH.  I would expect it to work on real hardware too, but 
I don't have any to try, so reports on actual machines would be particularly 
interesting.

paul



[cctalk] Re: Two items for RSTS/E

2023-11-20 Thread Tim Sneddon via cctalk
On Tue, 21 Nov 2023 at 06:13, Paul Koning via cctalk 
wrote:

> I just pushed two additions to https://github.com/pkoning2/decstuff :
>
> In "patches" a new patch for the DEUNA driver.  This fixes a problem seen
> when doing user (as opposed to DECnet) I/O, as well as two errors that show
> up when using units beyond the first.
>
> Directory "ntp" is new.  This is a simple NTP protocol client for RSTS,
> which will synchronize the system clock with an NTP server on the LAN.  It
> includes handling of timezone rules, so the right thing will happen at
> daylight savings time (summer time) boundaries.  The clock is maintained to
> the full RSTS resolution -- typically 1/50th or 1/60th second, but can be
> as low as 10 ms if the KW-11/P clock is used.
>

This is awesome! Thanks, Paul.

Regards, Tim.