gpsd-commit-watch
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[gpsd-commit-watch] [SCM] GPSD branch, master, updated. release-3.16-489


From: Gary E. Miller
Subject: [gpsd-commit-watch] [SCM] GPSD branch, master, updated. release-3.16-489-gc0c9ed8
Date: Fri, 23 Sep 2016 00:16:15 +0000 (UTC)

This is an automated email from the git hooks/post-receive script. It was
generated because a ref change was pushed to the repository containing
the project "GPSD".

The branch, master has been updated
       via  c0c9ed85b0d3a199b51b4a9f4eeed7bd24712225 (commit)
       via  89bc05db05ec3f8f073852a46caa5b37ed87e2e1 (commit)
       via  21eb174671474947688c5afee8e92a637b2fb149 (commit)
      from  a3ff750ce4d73dbdc7e7f017ef81f3f1df7347a4 (commit)

Those revisions listed above that are new to this repository have
not appeared on any other notification email; so we list those
revisions in full, below.

- Log -----------------------------------------------------------------
commit c0c9ed85b0d3a199b51b4a9f4eeed7bd24712225
Author: Fred Wright <address@hidden>
Date:   Fri Sep 16 18:18:28 2016 -0700

    Fixes inverted offset signs in gpsmon's PPS lines.
    
    In gpsmon, the PPS offsets reported in the display pane and in the
    scrolling packet data has opposite signs.  Since the display version
    is consistent with the value reported by the ppswatch utility
    (from the pps-tools package), this is assumed to be the correct
    version, and the "text" version is corrected to match.
    
    TESTED:
    Ran gpsmon on both OSX and Linux, verifying that the two forms
    of PPS offset reported by gpsmon and the ppswatch version are all
    consistent.
    
    Signed-off-by: Gary E. Miller <address@hidden>

commit 89bc05db05ec3f8f073852a46caa5b37ed87e2e1
Author: Fred Wright <address@hidden>
Date:   Fri Sep 16 18:18:27 2016 -0700

    Fixes Ill-formed TOFF/PPS packet error in gpsmon.
    
    The code  for reporting the  combined host address and  device address
    modifies the  same 'path' field that  it uses to  form the combination
    This was resulting in a growing accumulation of device names until the
    string became too long.  Although the way this code works could use
    some improvement, a cheap fix is to check for the presence of the
    device name in the path, and skip it when creating the new one.
    
    TESTED:
    On both OSX and Linux, ran "scons build-all check" and tested gpsmon
    both with and without the -a option, verifying that it no longer
    fails and that the reported "device" is the expected combined form.
    
    Signed-off-by: Gary E. Miller <address@hidden>

commit 21eb174671474947688c5afee8e92a637b2fb149
Author: Fred Wright <address@hidden>
Date:   Fri Sep 16 18:18:26 2016 -0700

    Fixes crash when quitting gpsmon in -a mode.
    
    There were two bugs in the alternate input code used in -a mode:
    
    1) It was using strlen() instead of sizeof() to determine the
    available space in the input buffer.  This is uninitialized data,
    making the result nondeterministic, though it would return zero
    in the case observed.
    
    2) It was failing to handle the NULL return from fgets() correctly,
    resulting in a call to do_command() with a garbage argument.  This
    caused a segfault.
    
    Also, casting the strlen/sizeof value to int was superfluous.
    
    TESTED:
    Typing a "q" in -a mode now outputs a prompt (an incorrect one,
    but that's another story) and can then quits without crashing.
    
    Signed-off-by: Gary E. Miller <address@hidden>

-----------------------------------------------------------------------

Summary of changes:
 drivers.c |    4 +++-
 gpsmon.c  |    7 ++++---
 2 files changed, 7 insertions(+), 4 deletions(-)


hooks/post-receive
-- 
GPSD



reply via email to

[Prev in Thread] Current Thread [Next in Thread]