[mythtv-users] Segfault on starting playback

Thu Aug 30 18:30:38 UTC 2012

On Thu, Aug 30, 2012 at 7:27 PM, Stephen Robertson
<stephengrobertson at gmail.com> wrote:
> On Thu, Aug 30, 2012 at 5:41 PM, Jim Stichnoth <stichnot at gmail.com> wrote:
>> On Sun, Aug 26, 2012 at 1:12 PM, Stephen Robertson
>> <stephengrobertson at gmail.com> wrote:
>>> On Sunday, August 26, 2012, Douglas Mackay wrote:
>>>>
>>>> On 26 August 2012 10:53, Douglas Mackay <douglas.mackay at gmail.com> wrote:
>>>> > On 26 August 2012 10:01, Douglas Mackay <douglas.mackay at gmail.com>
>>>> > wrote:
>>>> >> Not sure if this helps but I'm also experiencing a similar issue.
>>>> >>
>>>> >> I've got two crash reports on Launchpad (one from a remote frontend
>>>> >> and the other from the backend)
>>>> >>
>>>> >> https://bugs.launchpad.net/mythbuntu/+bug/1041788
>>>> >> https://bugs.launchpad.net/mythbuntu/+bug/1041797
>>>> >
>>>> > If I manually run mythpreviewgen on the recording then is crashes
>>>> > about 50% of the time. This recording was from Freeview (UK).
>>>>
>>>> I've done further testing and from what I can see it just affects BBC
>>>> channels (SD, HD on both UK Freeview and UK Freesat). These channels
>>>> have digital teletext services.
>>>>
>>>> Sample (6MB) - http://www.filedropper.com/100120120826105500
>>>> ______________________________________________
>>>
>>>
>>> I hadn't noticed this as the vast majority of my recordings are BBC but this
>>> seems to be the case for me too.
>>
>> This is a request for help from the people who are experiencing these
>> crashes.  I am trying to investigate this problem (there are at least
>> 3 tickets that seem to be reporting this same thing), but it is
>> challenging because I can't reproduce the crash on my system.
>> However, there are indications from valgrind that uninitialized memory
>> may be involved.  Specifically, if I use valgrind in this way:
>>
>> valgrind --leak-check=full --error-limit=no --show-reachable=yes
>> /usr/mythtrunk/bin/mythavtest -v general,playback,libav --loglevel
>> debug myth://Videos@192.168.0.205:6543/broken.mpg
>>
>> Part of the valgrind output looks like this:
>>
>> ...
>> 2012-08-29 13:17:33.668634 D  mpegts_add_stream: stream #11, has id
>> 0x1c25 and codec DSMCC_B, type Data at 0x134ef380
>> ==15679== Conditional jump or move depends on uninitialised value(s)
>> ==15679==    at 0x411497E: mpegts_push_section (mpegts-mythtv.c:674)
>> ==15679==    by 0x4114B6E: write_section_data (mpegts-mythtv.c:401)
>> ==15679==    by 0x41173CD: handle_packet (mpegts-mythtv.c:2527)
>> ==15679==    by 0x7021BC63: ???
>> ==15679==
>> ==15679== Conditional jump or move depends on uninitialised value(s)
>> ==15679==    at 0x4114985: mpegts_push_section (mpegts-mythtv.c:684)
>> ==15679==    by 0x4114B6E: write_section_data (mpegts-mythtv.c:401)
>> ==15679==    by 0x41173CD: handle_packet (mpegts-mythtv.c:2527)
>> ==15679==    by 0x7021BC63: ???
>> ==15679==
>> ==15679== Conditional jump or move depends on uninitialised value(s)
>> ==15679==    at 0x411497E: mpegts_push_section (mpegts-mythtv.c:674)
>> ==15679==    by 0x4114B6E: write_section_data (mpegts-mythtv.c:401)
>> ==15679==    by 0x41173CD: handle_packet (mpegts-mythtv.c:2527)
>> ==15679==    by 0x8792: ???
>> ==15679==
>> ==15679== Conditional jump or move depends on uninitialised value(s)
>> ==15679==    at 0x4114985: mpegts_push_section (mpegts-mythtv.c:684)
>> ==15679==    by 0x4114B6E: write_section_data (mpegts-mythtv.c:401)
>> ==15679==    by 0x41173CD: handle_packet (mpegts-mythtv.c:2527)
>> ==15679==    by 0x8792: ???
>> ==15679==
>> ==15679== Conditional jump or move depends on uninitialised value(s)
>> ==15679==    at 0x411497E: mpegts_push_section (mpegts-mythtv.c:674)
>> ==15679==    by 0x4114B6E: write_section_data (mpegts-mythtv.c:401)
>> ==15679==    by 0x41173CD: handle_packet (mpegts-mythtv.c:2527)
>> ==15679==    by 0x6FDC685: ??? (in /usr/lib/libQtGui.so.4.6.2)
>> ==15679==
>> ==15679== Conditional jump or move depends on uninitialised value(s)
>> ==15679==    at 0x4114985: mpegts_push_section (mpegts-mythtv.c:684)
>> ==15679==    by 0x4114B6E: write_section_data (mpegts-mythtv.c:401)
>> ==15679==    by 0x41173CD: handle_packet (mpegts-mythtv.c:2527)
>> ==15679==    by 0x6FDC685: ??? (in /usr/lib/libQtGui.so.4.6.2)
>> ==15679==
>> ==15679== Conditional jump or move depends on uninitialised value(s)
>> ==15679==    at 0x411497E: mpegts_push_section (mpegts-mythtv.c:674)
>> ==15679==    by 0x4114B6E: write_section_data (mpegts-mythtv.c:401)
>> ==15679==    by 0x41173CD: handle_packet (mpegts-mythtv.c:2527)
>> ==15679==    by 0x1F03EB59: ???
>> ==15679==
>> ==15679== Conditional jump or move depends on uninitialised value(s)
>> ==15679==    at 0x4114985: mpegts_push_section (mpegts-mythtv.c:684)
>> ==15679==    by 0x4114B6E: write_section_data (mpegts-mythtv.c:401)
>> ==15679==    by 0x41173CD: handle_packet (mpegts-mythtv.c:2527)
>> ==15679==    by 0x1F03EB59: ???
>> ==15679==
>> 2012-08-29 13:17:35.174329 I  AFD: Stream #0, has id 0x12d codec id
>> MPEG2VIDEO, type Video, bitrate 15000000 at 0x12c8d720
>> ...
>>
>> This valgrind output related to mpegts-mythtv.c lines 674 and 684
>> seems to correlate to the stack traces in the crash reports, and
>> appears just after the last bit of log output before the crashes.
>>
>> So I'd like to ask anyone who can reproduce this crash, if they can
>> run valgrind in the same way, and see where the crash happens with
>> respect to the valgrind output.
>>
>> Jim
>> _______________________________________________
>
> Not at all familiar with valgrind so don't know what to pull out so
> the whole thing (>8MB) is at
> https://www.dropbox.com/s/lf9m1vm94b5ndlu/valgrind.txt?m
>
> Anything else I can help with just let me know.
>
> Stephen

myyth log file that goes along with the valgrind output

https://www.dropbox.com/s/0jalsu58gd6xq99/log.txt

Stephen