| Commit message (Collapse) | Author | Age | Files | Lines | ||
|---|---|---|---|---|---|---|
| ... | ||||||
| * | Stringer warning when KPPS not available. | Gary E. Miller | 2015-03-30 | 1 | -3/+10 | |
| | | ||||||
| * | Don't throw the baby out with the bathwater. | Gary E. Miller | 2015-03-30 | 1 | -225/+354 | |
| | | | | | 'reverting' 4336209683d0f5927e32a4edf9131a17e1618253 | |||||
| * | Remove dead code found by scan-build in ppsthread.c. | Eric S. Raymond | 2015-03-30 | 1 | -2/+0 | |
| | | | | | PPS observed live on GR601-W. | |||||
| * | Make TIOCMIWAIT PPS work again. | Eric S. Raymond | 2015-03-30 | 1 | -353/+226 | |
| | | | | | | | | | | Reverts most of 'Pull out RFC2783 edge finding into a new function: get_edge_rfc2783()', but keeps the removal of a lot of splint markup for cleaner alternatives. This version passes regressions, splints clean, and exhibits PPS on the GR601-W when running as either root or non-root. | |||||
| * | Fix a truncated PPS log message. | Gary E. Miller | 2015-03-29 | 1 | -1/+2 | |
| | | ||||||
| * | Do not try to use time_pps_getcap() if no pps_handle. | Gary E. Miller | 2015-03-29 | 1 | -1/+5 | |
| | | ||||||
| * | Beat the splint beast into submission. | Eric S. Raymond | 2015-03-29 | 1 | -12/+17 | |
| | | | | | | | Code splints clean, all regression tests pass, but 1PPS is not being detected and wasn't before this changeset. Debug messages suggest the trailing edge is being rejected. | |||||
| * | Last splint thing in ppsthread.c | Gary E. Miller | 2015-03-28 | 1 | -1/+1 | |
| | | ||||||
| * | ppsthread.c is splint clean, but one thing. YMMV. | Gary E. Miller | 2015-03-28 | 1 | -30/+49 | |
| | | ||||||
| * | work around incorrect definition of struct timespec in splint 3.12 | Eric S. Raymond | 2015-03-28 | 1 | -0/+6 | |
| | | ||||||
| * | Remove the last of no longer needed splint cruft from ppsthread.c | Gary E. Miller | 2015-03-25 | 1 | -34/+2 | |
| | | | | | scons splint runs fine, no code changes | |||||
| * | Remove unneeded splint overrides | Gary E. Miller | 2015-03-25 | 1 | -13/+0 | |
| | | | | | splint still happy | |||||
| * | Brain damaged pps-gpio sometimes never fills in PPS clear timestamp. | Gary E. Miller | 2015-03-25 | 1 | -1/+9 | |
| | | ||||||
| * | add yet a tad more PPS trace logging. | Gary E. Miller | 2015-03-25 | 1 | -0/+9 | |
| | | ||||||
| * | Remove unneeded -#ifndef S_SPLINT_S | Gary E. Miller | 2015-03-25 | 1 | -9/+8 | |
| | | | | | Cruft be gone! | |||||
| * | Splint fixes. | Gary E. Miller | 2015-03-25 | 1 | -10/+4 | |
| | | ||||||
| * | Fix another nit that bothers splint. | Gary E. Miller | 2015-03-25 | 1 | -1/+1 | |
| | | ||||||
| * | Make cppcheck happy, reduce scope of pther_err | Gary E. Miller | 2015-03-25 | 1 | -1/+2 | |
| | | ||||||
| * | gpsd no longer needs TIOCMIWAIT if it has RFC2783 PPS_CANWAIT | Gary E. Miller | 2015-03-25 | 1 | -10/+45 | |
| | | | | | Next step is to ensure gpsd can compile and run without TIOCMIWAIT | |||||
| * | Pull out RFC2783 edge finding into a new function: get_edge_rfc2783() | Gary E. Miller | 2015-03-25 | 1 | -194/+269 | |
| | | | | | Still depends on TIOCMIWAIT, but not for much longer. | |||||
| * | Clean up a binary and. | Gary E. Miller | 2015-03-24 | 1 | -1/+1 | |
| | | ||||||
| * | ppsthread.c broken on non-Linux | Hal Murray | 2015-03-24 | 1 | -1/+1 | |
| | | | | | Signed-off-by: Gary E. Miller <gem@rellim.com> | |||||
| * | Fix segfault in split off TIOCMIWAIT support. | Gary E. Miller | 2015-03-24 | 1 | -1/+1 | |
| | | ||||||
| * | Move TIOMCIWAIT to its own function. | Gary E. Miller | 2015-03-24 | 1 | -83/+126 | |
| | | | | | Soo it will be optional at run time. | |||||
| * | Remove a bit of temporary debug logging. | Gary E. Miller | 2015-03-24 | 1 | -3/+0 | |
| | | ||||||
| * | Disable the new PPS_CANWAIT code to allow shipping the ppsthread changes. | Gary E. Miller | 2015-03-23 | 1 | -3/+4 | |
| | | ||||||
| * | If we have PPS_CANWAIT, no need for TIOCMIWAIT | Gary E. Miller | 2015-03-23 | 1 | -92/+122 | |
| | | ||||||
| * | Comment typos | Gary E. Miller | 2015-03-21 | 1 | -1/+1 | |
| | | ||||||
| * | Move istty() and PPS caps tests to where they are needed. | Gary E. Miller | 2015-03-20 | 1 | -14/+50 | |
| | | ||||||
| * | init_kernel_pps() has no need to know if device is a tty. | Gary E. Miller | 2015-03-20 | 1 | -10/+8 | |
| | | | | | Also log if we have PPS_CANWAIT | |||||
| * | Doc & interface cleanup of ppsthread code. PPS observed live. | Eric S. Raymond | 2015-03-20 | 1 | -5/+8 | |
| | | ||||||
| * | Add code to handle less than perfect RFC2783 drivers. | Gary E. Miller | 2015-03-20 | 1 | -38/+59 | |
| | | | | | | The Linux pps-gpio driver only track the clear edge. Fair warning your PPS may be offset by the pulse width. | |||||
| * | Remove redundant define. | Gary E. Miller | 2015-03-20 | 1 | -1/+0 | |
| | | ||||||
| * | PPS_TSFMT_TSPEC is the default, make it obvious. | Gary E. Miller | 2015-03-20 | 1 | -1/+3 | |
| | | ||||||
| * | Tell time_pps_setparam() what version API to use. | Gary E. Miller | 2015-03-19 | 1 | -0/+2 | |
| | | ||||||
| * | Improve logging of time_pps_setparams() and time_pps_getcap() | Gary E. Miller | 2015-03-19 | 1 | -5/+5 | |
| | | | | | Direct open of /dev/pps0 still fails. Old functionality still good. | |||||
| * | Add root device name to PPS debug output. | Gary E. Miller | 2015-03-19 | 1 | -44/+89 | |
| | | | | | | | | | With more than one PPS thread going at a time hard to know which is which without help. Also remove restriction that devicename is a tty since /dev/ppsX is not a tty. | |||||
| * | Drop an unneeded root check. | Gary E. Miller | 2015-03-19 | 1 | -6/+3 | |
| | | ||||||
| * | It only makes sense to look for 1PPS on DCE signals. Add an explanation. | Eric S. Raymond | 2015-03-18 | 1 | -2/+19 | |
| | | | | | All regressiion tests pass. | |||||
| * | In ppsthread.c, add TIOCM_RTS to the PPS mask. | Eric S. Raymond | 2015-03-17 | 1 | -1/+1 | |
| | | | | | | | | Motivation: I happen to have been looking at RS-232 level-shifter boards recently. The second most common set of supported data lines (after TX/RX alone and before the full set CD/RX/TX/DTR/DSR/RTS/CTS/RI) is TX/RX/CTS/RTS. | |||||
| * | splint/cppcheck/coverity prerelease cleanup. | Eric S. Raymond | 2015-03-13 | 1 | -2/+2 | |
| | | ||||||
| * | In ppsthread.c, ensure errno thread-safety. All regression tests pass. | Eric S. Raymond | 2015-03-11 | 1 | -0/+1 | |
| | | ||||||
| * | Simplify the thread-monitor interface. | Eric S. Raymond | 2015-03-11 | 1 | -15/+12 | |
| | | | | | PPS observed live on GR601W. | |||||
| * | Remove a redundancy. | Eric S. Raymond | 2015-03-11 | 1 | -3/+3 | |
| | | ||||||
| * | Take a whack at addressing pre-existing /dev/ppsN devices. | Eric S. Raymond | 2015-03-11 | 1 | -56/+64 | |
| | | | | | | | PPS observed live on GR-601W, but code not tested on anything like the Raspberry Pi that would invoke the new case. | |||||
| * | splint cleanup of new code. | Eric S. Raymond | 2015-03-08 | 1 | -2/+3 | |
| | | ||||||
| * | ppsthread.c is now fully decoupled from the libgpsd_core.c structures. | Eric S. Raymond | 2015-03-08 | 1 | -1/+51 | |
| | | | | | | | | | | | | | | | | | | The files ntpshm.h ntpshmread.c ntpshmwrite.c ppsthread.[ch] timespec_str.[ch] now form a standalone group which can be used for NTP communication and PPS thread monitoring. Separation is not yet perfect, as ppsthread.c requires gpsd_config.h for HAVE_SYS_TIMEPPS_H and npshmread.c/ntpshmwrite.c both require compiler.h. All regression tests pass. PPS is live on a GR-601W. | |||||
| * | Break timespec_str out of gpsutils.c. | Eric S. Raymond | 2015-03-08 | 1 | -0/+1 | |
| | | | | | | | It's not used on the client side, and it's needed separately for ntplib. All regression tesrs pass. PPS is live. | |||||
| * | Eliminate timestamp-T use from PPS thread code. | Eric S. Raymond | 2015-03-08 | 1 | -12/+13 | |
| | | | | | PPS observed live on GR-601W. All regression yests pass. | |||||
| * | Clean up minor ppsthread.c dependencies. No code changes. | Eric S. Raymond | 2015-03-08 | 1 | -2/+4 | |
| | | ||||||
 |
index : delta/gpsd.git | |
| git.savannah.gnu.org: git/gpsd.git |
| summaryrefslogtreecommitdiff |
path: root/ppsthread.c