| Re: [chrony-dev] [GIT] chrony/chrony.git annotated tag, 1.27-pre1, created. 1.27-pre1 |
[ Thread Index |
Date Index
| More chrony.tuxfamily.org/chrony-dev Archives
]
On Mon, 12 Mar 2012, Miroslav Lichvar wrote:
On Sat, Mar 10, 2012 at 07:08:14PM -0800, Bill Unruh wrote:
I have just noticed that (in chrony 1.26) the offsets are now all reported
relative to the last second, rather than the nearest second. Thus positive
offsets are reported to ns, while negative offsets, since they start with
9.99999 only report offsets to the nearest 100ns (in refclock.log). Ie, there are no negative
offsets anymore. This seems a bit strange. While one could say that reporting
to the nearest 100ns is already too much accuracy, it would be aesthetically
more pleasing if negative and positive offsets were symmetrically handled.
That's only for raw PPS offset, right? The raw value is printed as
received from PPS refclock driver, which doesn't know the PPS rate and
can't easily convert 0.999 to -0.001.
Surely what is reported from the refclock is the time stamp, not the offset.
It is chrony which is stripping off the seconds to give an offset.
Ie, the raw offset should be the difference between the "true time" as set by
the refclock, and the system time. If the refclock does not know what second
it is (PPS) then that offset should be something like t-int(t+.5) not simply
t-int(t).
Note that this is the way it used to be AFAIK (chrony 1.24 with the shm
driver)-- you would get both pos and neg offsets in the raw time.
I do agree that the PPS rate is a more ticklish issue.
The cooked offset in the next column should be ok.
--
William G. Unruh | Canadian Institute for| Tel: +1(604)822-3273
Physics&Astronomy | Advanced Research | Fax: +1(604)822-5324
UBC, Vancouver,BC | Program in Cosmology | unruh@xxxxxxxxxxxxxx
Canada V6T 1Z1 | and Gravity | www.theory.physics.ubc.ca/
--
To unsubscribe email chrony-dev-request@xxxxxxxxxxxxxxxxxxxx with "unsubscribe" in the subject.
For help email chrony-dev-request@xxxxxxxxxxxxxxxxxxxx with "help" in the subject.
Trouble? Email listmaster@xxxxxxxxxxxxxxxxxxxx.