diff options
| author | Gary E. Miller <gem@rellim.com> | 2013-12-01 12:48:05 -0800 |
|---|---|---|
| committer | Gary E. Miller <gem@rellim.com> | 2013-12-01 12:48:05 -0800 |
| commit | 9182eda0e08865deb05861a4fe1aa42bd5893f94 (patch) | |
| tree | e46d1200ab545366d95aac6ea73b8736a91d0556 /ppsthread.c | |
| parent | 46c537a61148f436ffba1f2e9e7f6408964edbc6 (diff) | |
| download | gpsd-9182eda0e08865deb05861a4fe1aa42bd5893f94.tar.gz | |
Document a broken assumption about serial and PPS time.
The GR-601W can send the serial time fix 10 mSec before the PPS
status line change. My guess is some sort of race condition. The
consequence is the PPS is matched to the wrong secnod. Leading to
an off by one result.
Diffstat (limited to 'ppsthread.c')
| -rw-r--r-- | ppsthread.c | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/ppsthread.c b/ppsthread.c index 580651f6..bbadc37f 100644 --- a/ppsthread.c +++ b/ppsthread.c @@ -574,6 +574,9 @@ static /*@null@*/ void *gpsd_ppsmonitor(void *arg) * serial stream *after* emitting PPS for the top of second. * Thus, when we see PPS our available report is from the * previous cycle and we must increment. + * + * FIXME! The GR-601W at 38,400 or faster can send the + * serial fix before PPS by about 10 mSec! */ /*@+relaxtypes@*/ |