[mythtv-commits] Ticket #13519: seg fault while channel scanning hdhomerun connect quatro

MythTV noreply at mythtv.org
Fri Nov 29 10:16:30 UTC 2019


#13519: seg fault while channel scanning hdhomerun connect quatro
-----------------------------------+-------------------------------
 Reporter:  Mike Bibbings          |          Owner:  Klaas de Waal
     Type:  Bug Report - Crash     |         Status:  assigned
 Priority:  minor                  |      Milestone:  31.0
Component:  MythTV - Mythtv-setup  |        Version:  Master Head
 Severity:  medium                 |     Resolution:
 Keywords:                         |  Ticket locked:  0
-----------------------------------+-------------------------------

Comment (by Klaas de Waal):

 Mike,

 I've just added a patch that can be applied to today's master. This should
 fix the segfault that you have with scanning and it should also fix the
 crash in the backend as reported in ticket #9027. The HDHR debug output is
 also still present. Can you please test this with the United Kingdom
 frequency list?

 As it happens the mutex to lock the HDHR tuner instance was already
 present and it was already used in the TunerGet and TunerSet functions,
 but not in UpdateValues.


 About why it happens on the UK scans

 The UK scans have offsets defined; if there is no lock on the exact
 frequency there is a try on +160kHz and on -160kHz. Log output shows that
 for the offset tuning the signal monitor is not stopped. For a non-offset
 scan, such as with the Netherlands channel list,the signal monitor is
 started and stopped for each frequency.

 About the function value of GetTunerStatus

 You are correct about the function value of
 hdhomerun_device_get_tuner_status.  However, this value is ignored. The
 status values are returned in a struct hdhomerun_tuner_status_t and this
 is, as I can see it, done correct.


 About ticket #9027.

 Gary, thanks for digging this up. I was not aware of this one. The gdb log
 shows that it is indeed a similar problem; here the call to
 HDHRStreamHandler::UpdateValues from the HDHR signal monitor clashes with
 the HDHRStreamHandler::run. This problem is still present in today's code
 but should now also be fixed.

-- 
Ticket URL: <https://code.mythtv.org/trac/ticket/13519#comment:17>
MythTV <http://www.mythtv.org>
MythTV Media Center


More information about the mythtv-commits mailing list