Hello everybody,
I'm trying to track (offline) the drift of my gps clock relative to my
hw clock (the clock that I would have obtained without GPS).
My final goal is to convert a series of gps timestamps to the
equivalent hw timestamps (with microsec precision).
My GPS clock is built from the shmPPS driver & a garmin 18 LVC GPS.
I'm using linux 2.6.25 with ntp 4.2.4.
1)My first idea was to recreate (off line) the hw timestamps by
-inverting the NTP kernel clock discipline algorithm
-feeding it with the loopstats statistics and my gps time series
but the kernel discipline does much more that applying the loopstats
input data
(it runs its own PLL in the kernel)
so it doesn't seem practically feasible
2)I try to disable the kernel disciple (disable kernel) to run the PLL
in user space, and capture the basic adjtime systems calls. But in
this case when I strace the adjtimex system calls,
most parameters are empty & it returns an error:
15:03:44.881187 adjtimex({modes=ADJ_OFFSET|0x8000, offset=0,
freq=5528944, maxerror=731664, esterror=16, status=STA_UNSYNC,
constant=4, precision=1, tolerance=33554432, time={1232373824,
881222}, tick=10000, ppsfreq=0, jitter=0, shift=0, stabil=0, jitcnt=0,
calcnt=0, errcnt=0, stbcnt=0}) = 5 (TIME_ERROR)
so it doesn't seem to work. (furthemore it doesn't not seem to be a
good solution when high accuracy is needed)
3)Is there a simple way to do that? It seems very simple but I'm not
from the domain.
Has a hw clock driver clock already been implemented to do such an
operation?
If so, I imagine the peerstats it would generate could help me.
Regards
Ryad
_______________________________________________
questions mailing list
[email protected]
https://lists.ntp.org/mailman/listinfo/questions
