summaryrefslogtreecommitdiff
path: root/serial.c
Commit message (Collapse)AuthorAgeFilesLines
* Sepatate character from retry counters.Eric S. Raymond2005-07-261-3/+3
|
* Slightly better unstrumentation of file descriptor operations.Eric S. Raymond2005-07-021-0/+1
|
* Cosmetic fix.Eric S. Raymond2005-07-011-1/+2
|
* Pull the DGPSIP code into its own module.Eric S. Raymond2005-06-291-0/+2
| | | | | | | | The DGPSIP socket is now part of the session context; it's guarded by the same select as all other socket reads. The latest RTCM104 report from it is relayed to each attached GPS when appropriate. (The old code only passed a correction to the first GPS to be polled after the report came in from the server.)
* Set VMIN in GPS reads.Eric S. Raymond2005-06-291-0/+6
|
* Remove more unneeded annotations.Eric S. Raymond2005-06-111-2/+2
|
* OpenBSD port change: CBAUDEX removal.Eric S. Raymond2005-06-101-1/+1
|
* 104 split errors. Some iffy changes to the TSIP driver.Eric S. Raymond2005-06-021-2/+2
|
* 879 splint warnings.Eric S. Raymond2005-05-271-3/+3
|
* Down to 1545 splint warnings.Eric S. Raymond2005-05-261-1/+1
|
* Down to 1543 splint warnings.Eric S. Raymond2005-05-261-9/+13
|
* More splint cleanup. Down to 1689 warnings.Eric S. Raymond2005-05-261-12/+13
|
* Make a start at cleaning up splint warnings.Eric S. Raymond2005-05-261-1/+1
|
* Re-apply the fix for Gary.Eric S. Raymond2005-05-261-0/+2
|
* Handle packet acceptance more cleanly.Eric S. Raymond2005-05-241-2/+0
|
* This may fix Gary Miller's bug.Eric S. Raymond2005-05-241-0/+2
|
* Clean up some device pointer references that must now be conditional Eric S. Raymond2005-05-241-2/+1
| | | | on non-NULL.
* Preparation for the TSIP driver.Eric S. Raymond2005-05-241-10/+21
|
* Tweak the code so the B command still works.Eric S. Raymond2005-05-241-0/+1
|
* Correct error in hunt logic.Eric S. Raymond2005-05-241-3/+4
|
* Fix a serious bug in multi-device operation.Eric S. Raymond2005-05-241-19/+36
| | | | | | | | | | | | Rob Janssen quite correctly reported as a bug the fact that when gpsd went hunting on a new device it fell out of the main loop and ignored pending input from devices already connected. This is because packet_sniff was written to loop until it found a packet or timed out. The fix: opening a device no longer does a packet sniff. Instead, gpsd_poll() doubles as the packet sniffer when the driver-type pointer is null, and the driver-type pointer for the device is only set when it finds a whole packet.
* Various driver fixes and cleanups from Rob Janssen.Eric S. Raymond2005-05-221-6/+2
|
* Set speed early as it's used for the delays in packet_sniff().Eric S. Raymond2005-05-111-3/+3
|
* Speed up synchronization on a port that hassn't been tweaked recently.Eric S. Raymond2005-05-061-13/+8
| | | | | | | Get the speed we try first from looking at the device's current speed, rather than the last speed stored for it by gpsd. This means gpsd will never have to try more than one speed when the device and GPSD speeds are matched, even if hotplug vicissitudes have renamed a device.
* Extend speed range, bug reported by Ulrich Ruess.Eric S. Raymond2005-04-051-1/+3
|
* Device list editing through a separate control socket.Eric S. Raymond2005-04-031-2/+2
|
* Better progress message.Eric S. Raymond2005-04-021-1/+1
|
* Change gps_session_t to gps_device_t.Eric S. Raymond2005-03-281-3/+3
|
* Garmin: more certain init of the serial portGary E. Miller2005-03-241-1/+3
|
* Big structure reorganization.Eric S. Raymond2005-03-221-24/+24
| | | | Timestamp fields are gone, replaced by validity mask bits.
* Fix some bugs in feature conditionalization.Eric S. Raymond2005-03-181-1/+1
|
* Put the initialization further back...Eric S. Raymond2005-03-161-1/+1
|
* Make sure the out-of-band packet-type value gets set.Eric S. Raymond2005-03-161-0/+1
|
* Clean up device initialization.Eric S. Raymond2005-03-161-4/+2
|
* N command is now fully implemented.Eric S. Raymond2005-03-151-1/+1
|
* We don't actually need O_SYNC.Eric S. Raymond2005-03-121-4/+1
|
* Three small fixes.Derrick Brashear2005-03-121-1/+4
| | | | | | serial.c needs sys/aio.h on macos (O_SYNC) configure.ac needs to test for externed timezone and daylight nmea_parse.c needs to not use daylight and timezone when they don't exist
* Mac OS port fixes. Describe WAAS/EGNOS in the GPS Hacking document.Eric S. Raymond2005-03-121-1/+1
| | | | Fix bonehead error in PGRME interpretation.
* Get a heads-up when the basic serial-device open fails.Eric S. Raymond2005-03-081-1/+3
|
* Fix some errors in code conditioning, remove an unused variable.Eric S. Raymond2005-03-051-1/+1
|
* Probe functions are now a per-driver capability. Eric S. Raymond2005-03-051-7/+8
| | | | Yo, Gary...test this, please! :-)
* Oops. Put in the actual driver switch.Eric S. Raymond2005-03-041-0/+1
|
* We have zero configuration. The -T option is dead.Eric S. Raymond2005-03-041-1/+1
|
* Aha! The Garmin probe works now.Eric S. Raymond2005-03-041-1/+1
|
* Packet engine now copes gracefully when reads span packet boundaries.Eric S. Raymond2005-02-281-6/+9
|
* Skip autobauding if Garmin type is selected.Gary E. Miller2005-02-281-0/+9
|
* Turns out reliable sync isn't good enough for some SiRF chips. you reallyEric S. Raymond2005-02-261-17/+3
| | | | need the timeout.
* Cache both speed and stopbits.Eric S. Raymond2005-02-261-0/+1
|
* Baud- and stopbit-hunting works well enough now that we no longer need to keepEric S. Raymond2005-02-261-11/+3
| | | | communications parameters in the driver.
* Explicit support for the FV18 is gone, instead we ship its init string at everyEric S. Raymond2005-02-261-1/+1
| | | | NMEA device just in case.
View Lorry Controller Status