Re: [gpsd-dev] Updated docs on NTP segment management

[Gary E. Miller]

Yo Hal!

On Wed, 25 Feb 2015 11:03:15 -0800
Hal Murray <address@hidden> wrote:

> address@hidden said:
> > Yeah, I've already found some fields that I'd like to add to the
> > protocol, but didn't do it to not break the compatibility. 
> 
> What did you have in mind?  I'd like to make sure we collect all the
> good ideas while this discussion is going on.

Well, the thing that had been byting me is the lack of a way to mark
an NTPSHM as 'in-use'.  If I am not really careful I get gpsd and
ptp4l both trying to step on shmid NTP0.

The root:rooot/600 and root:root/666 perms thing also recently bit me.
We gotta do ntpd:ntpd/660 or gpsd:ntpd:660.

Another possibility, still noodling on it, would be to put a real
POSIX semaphore inside the SHM, so the slaves do not need to keep
polling, plus that give you a memory barrier for free.

> Would it have made sense to use the dummy slots in the current SHM?
> (assume we could find a way to avoid collisions)

Yes, similar to the nSec addition.

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97701
        address@hidden  Tel:+1(541)382-8588