| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Move the memory barrier instructions into the header file | Beat Bolli | 2011-03-27 | 3 | -12/+13 |
| | | | | | | | This allows adding other compilers and/or architectures later as needed. Signed-off-by: Eric S. Raymond <esr@thyrsus.com> | ||||
| * | gpxlogger improvements | Beat Bolli | 2011-03-27 | 1 | -15/+17 |
| | | | | | | | | - update the usage string with the new options - introduce and use the NITEMS() macro Signed-off-by: Eric S. Raymond <esr@thyrsus.com> | ||||
| * | Re-identify most *_IS flags with corresponding *_SET flags. | Eric S. Raymond | 2011-03-27 | 30 | -318/+292 |
| | | | | | | | | | | | | Required so clients looking at flag masks in data exported via the shared-memory interface will see the right thing. These were separated originally in order to avoid pushing the reqyuired width of the gpsdata.set flag mask over 32 bits. It became 64 bits in the Version 5 API change, so that constraint went away. All regression tests pass. | ||||
| * | Clear up some confusion about *_SET vs.. *_IS defines. | Eric S. Raymond | 2011-03-26 | 3 | -3/+3 |
| | | | | | | | | | | *_SET masks are library-side and for the set member of the fisx structure. *_IS masks are daemon-side and used only in the status mask returned by the packet analyzer methods, and code in the core library that uses same. Never the twain shall meet. All regression tests pass. | ||||
| * | Add note on GPSTether. | Eric S. Raymond | 2011-03-26 | 1 | -1/+5 |
| | | |||||
| * | Startlingly, memcpy() is *less* of a problem for the seqlock than I thought. | Eric S. Raymond | 2011-03-26 | 3 | -14/+17 |
| | | |||||
| * | A Java binding is in early development. | Eric S. Raymond | 2011-03-26 | 1 | -0/+4 |
| | | |||||
| * | Remove obsolete caveat. | Eric S. Raymond | 2011-03-26 | 1 | -6/+1 |
| | | |||||
| * | Plant warnings about odd memcpy() implementations. | Eric S. Raymond | 2011-03-26 | 2 | -0/+15 |
| | | |||||
| * | Fix a glitch in the SiRF driver exposed by the SHM export. | Eric S. Raymond | 2011-03-26 | 3 | -38/+45 |
| | | | | | | When we set TINE_IS, we actually have to report time. Required one regression test rebuild. | ||||
| * | Experimental export via shared memory is almost working. | Eric S. Raymond | 2011-03-26 | 4 | -32/+109 |
| | | | | | | | Fix fields look correct but time is garbled. All regression tests pass. Code splints clean. | ||||
| * | First cut at read side of shared-memory exporter. | Eric S. Raymond | 2011-03-26 | 3 | -2/+78 |
| | | | | | Untested, probably buggy. All regression tests pass. | ||||
| * | Casts can easily screw you up. | Eric S. Raymond | 2011-03-25 | 1 | -5/+23 |
| | | | | | All regression tests pass, *with* shm export configured on. | ||||
| * | First cut at shared-memory export. Daemon-side only. | Eric S. Raymond | 2011-03-25 | 7 | -14/+148 |
| | | | | | | Not configured in by default yet because it produces a strange regression failure. | ||||
| * | Enable gpsd to be built withot socket and subscription support. | Eric S. Raymond | 2011-03-25 | 5 | -147/+202 |
| | | | | | | | | | | Built this way it camn only export results via DBUS - or a shared-memory interface not yet implemented. The DBUS option and enabling define are renamed so we have a consistent suffix of export/EXPORT. No logic changes, just some #ifdefs and rearrangements. All regression tests pass. Code splints clean. | ||||
| * | Use srcdir vs abs_builddir (more) appropriately. | Greg Troxel | 2011-03-24 | 1 | -4/+16 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | VPATH builds have been broken probably since python was introduced; the basic issue is that python's build infrastructure is not cross friendly or VPATH friendly. This commit makes minor changes to allow VPATH builds to work, and thus 'make distcheck' now gets to the point of running test cases. (VPATH builds refer to having the build directory not be the source directory, by running configure in another directory, enabling multiple different builds from the same source tree and read-only source trees.) NB: conventional builds have srcdir==abs_builddir, and therefore these changes have nearly zero effect in that case. Specific changes are: specify that the PYTHON_DISTUTILS_{LIBDIR,SCRIPTDIR} directories are in the build directory, not the source directory. make the source directory writable with chmod. This is gross, but setup.py wants to create a build subdirectory in the sources and place a temporary directory there. when symlinking the python shlibs into gps/, do so in a directory gps in the build directory, rather than the source directory. (Note that this is the same directory when running configure in the source directory.) "make check" results in all regression tests passing (with conventional src==build). | ||||
| * | Remove a fossil from the Python binding. | Eric S. Raymond | 2011-03-24 | 1 | -1/+0 |
| | | |||||
| * | Documentation improvement. | Eric S. Raymond | 2011-03-24 | 2 | -2/+8 |
| | | |||||
| * | Document the unpack() method in C and Python APIs. | Eric S. Raymond | 2011-03-24 | 4 | -3/+15 |
| | | | | | | | | | We're not bumping the minor API number because the C library has had this all along. Alas, it can't be made to fit the RAII model of the C++ binding. All regression tests pass. | ||||
| * | Minor buld fixes, including shipping the doc directory in the tarball. | Eric S. Raymond | 2011-03-24 | 1 | -1/+4 |
| | | |||||
| * | DEVICE::activated becomes an ISO8601 timestamp. | Eric S. Raymond | 2011-03-24 | 5 | -7/+24 |
| | | |||||
| * | Eliminate an ugly magic number. | Eric S. Raymond | 2011-03-24 | 1 | -1/+1 |
| | | |||||
| * | Documentation update. All regression tests pass. | Eric S. Raymond | 2011-03-24 | 2 | -2/+7 |
| | | |||||
| * | Pull the last NTPSHM code dependency out of the core library. | Eric S. Raymond | 2011-03-24 | 2 | -38/+38 |
| | | | | | | | The npshm stuff is now coupled to everything else in the daemon only through the dispatcher layer. This means that auxiliary tools like gpsmon and gpsctl will no longer try to talk to ntpd at device-initialization time. | ||||
| * | Try to move NTP link activation and deactivation up to the disparcher layer. | Eric S. Raymond | 2011-03-24 | 4 | -21/+20 |
| | | |||||
| * | Remove unused SiRF constants: TIME_SEEN_GPS_1, etc. | Gary E. Miller | 2011-03-23 | 2 | -15/+0 |
| | | | | | No need to set flags that are never used. | ||||
| * | Fix SiRF PPSTIME_IS | Gary E. Miller | 2011-03-23 | 1 | -0/+12 |
| | | |||||
| * | PPSTIME_IS and CLEAR_IS may be in the first packet. | Gary E. Miller | 2011-03-23 | 1 | -1/+1 |
| | | |||||
| * | Set ntpd_ship_to per-cycle based on the driver PPSTIME_IS flag. | Eric S. Raymond | 2011-03-24 | 1 | -25/+28 |
| | | |||||
| * | Merge branch 'master' of ssh://git.berlios.de/gitroot/gpsd | Eric S. Raymond | 2011-03-24 | 1 | -5/+5 |
| |\ | |||||
| | * | SiRF sets PPSTIME_IS when 3 or more sats in view. | Gary E. Miller | 2011-03-23 | 1 | -5/+5 |
| | | | |||||
| * | | Merge branch 'master' of ssh://git.berlios.de/gitroot/gpsd | Eric S. Raymond | 2011-03-23 | 1 | -21/+24 |
| |\ \ | |/ | |||||
| | * | Add fixcnt constraint to packet time as well. | Gary E. Miller | 2011-03-23 | 1 | -24/+24 |
| | | | |||||
| | * | 3 sats needed for good time, not 4. | Gary E. Miller | 2011-03-23 | 1 | -1/+1 |
| | | | |||||
| | * | Revert over conditioning ship_to_ntpd | Gary E. Miller | 2011-03-23 | 1 | -2/+5 |
| | | | |||||
| * | | Reindent code in order to minimize some near-future diffs. | Eric S. Raymond | 2011-03-23 | 1 | -27/+27 |
| |/ | |||||
| * | Refactoring step. Hide the internals of NTPD communication better. | Eric S. Raymond | 2011-03-23 | 3 | -16/+19 |
| | | |||||
| * | Add some explanatory comments about where root privs are needed. | Eric S. Raymond | 2011-03-23 | 1 | -3/+15 |
| | | | | | | Add conditioned-out code as a demonstration of how to spin the thread inexpensively until a device is open. | ||||
| * | Note a bug fix. | Eric S. Raymond | 2011-03-23 | 1 | -0/+2 |
| | | |||||
| * | New regression test for Typhoon 3100. | Eric S. Raymond | 2011-03-23 | 2 | -0/+1098 |
| | | | | | | This revealed a library bug that could cause the timestamp on every second fix to get clobberec due to GSV not having a timestamp. | ||||
| * | Fix bonehead error in the code for backward compatibility with Unix timestamps. | Eric S. Raymond | 2011-03-23 | 1 | -1/+1 |
| | | |||||
| * | Enable client debugging to actually work in gpxlogger. | Eric S. Raymond | 2011-03-23 | 1 | -0/+1 |
| | | |||||
| * | The ship_to_ntpd flag shouldn't latch. | Eric S. Raymond | 2011-03-23 | 1 | -0/+1 |
| | | |||||
| * | Add missing include file needed for open(). | Gary E. Miller | 2011-03-23 | 1 | -0/+4 |
| | | | | | A side effect of the recent PPS restructuring. | ||||
| * | Move as much of the PPS support as possible to ntpshm.c. | Eric S. Raymond | 2011-03-23 | 3 | -552/+554 |
| | | | | | All regression tests pass. | ||||
| * | Attempt to use the new PPSTIME_IS mask bit to control updating ntpd. | Eric S. Raymond | 2011-03-23 | 2 | -20/+7 |
| | | | | | | | | This commit may break ntpd support and should be live-tested by someone who can - I don't have an RS232 GPS with PPS to try it on. All regression tests pass. | ||||
| * | Make all drivers set the PPSTIME_IS mask where appropriate. It's not yet used. | Eric S. Raymond | 2011-03-23 | 14 | -32/+56 |
| | | | | | All regression tests pass. | ||||
| * | Use truly 64-bit constants for the 64-bit gps_mask_t flags. | Eric S. Raymond | 2011-03-23 | 3 | -63/+63 |
| | | | | | All regression tests pass. Code splints clean. | ||||
| * | Revert "Remove a failed experiment." | Eric S. Raymond | 2011-03-23 | 1 | -1/+2 |
| | | |||||
| * | Remove a failed experiment. | Eric S. Raymond | 2011-03-23 | 1 | -2/+1 |
| | | |||||
 |
index : delta/gpsd.git | |
| git.savannah.gnu.org: git/gpsd.git |
| summaryrefslogtreecommitdiff |