https://bugs.kde.org/show_bug.cgi?id=371579

            Bug ID: 371579
           Summary: Kaffeine 2.0.5 unresponsive on certain DVB-S channels
           Product: kaffeine
           Version: 2.0.1
          Platform: Mint (Ubuntu based)
                OS: other
            Status: UNCONFIRMED
          Severity: normal
          Priority: NOR
         Component: general
          Assignee: [email protected]
          Reporter: [email protected]

When watching certain DVB-S channels on ASTRA 19.2, kaffeine uses more than
100% CPU and does not respond any more.

For example, the issue is present on these transponders / channels:
11582 H, DVB-S2: "NDR FS HD"
12110 H, DVB-S: "MDR FS", "NDR FS", "rbb Berlin", "SWR Fernsehen" 

In this case, the output from kaffeine is, for example:

24-10-16 10:46:34.487 [Info    ] Using built-in dvb device manager
24-10-16 10:46:34.799 [Info    ] Found dvb device P14f188800070c12a: NXP
TDA10071
[00007fb980000c48] vdpau_avcodec generic error: decoder profile not supported:
2
[00007fb9ac00b158] vdpau_avcodec generic error: decoder profile not supported:
2
[00000000020caeb8] core input error: ES_OUT_SET_(GROUP_)PCR  is called too late
(pts_delay increased to 300 ms)
[00000000020caeb8] core input error: ES_OUT_RESET_PCR called
[00000000020caeb8] core input error: ES_OUT_SET_(GROUP_)PCR  is called too late
(pts_delay increased to 301 ms)
[00000000020caeb8] core input error: ES_OUT_RESET_PCR called
[00000000020caeb8] core input error: ES_OUT_SET_(GROUP_)PCR  is called too late
(pts_delay increased to 328 ms)
[00000000020caeb8] core input error: ES_OUT_RESET_PCR called
[00000000020caeb8] core input error: ES_OUT_SET_(GROUP_)PCR  is called too late
(pts_delay increased to 376 ms)
[00000000020caeb8] core input error: ES_OUT_RESET_PCR called
[00000000020caeb8] core input error: ES_OUT_SET_(GROUP_)PCR  is called too late
(pts_delay increased to 581 

The issue is never present on these transponders:
12188 H, DVB-S: "RTL Television"
12545 H, DVB-S: "ProSieben", "SAT.1"

It seems, the issue is caused by the big amount of EPG data on these channels,
because when commenting out the loop that checks for existing EPG entries in 
dvbepg.cpp - DvbEpgModel::addEntry(), the issue is not present.

//     for (Iterator it = entries.begin(); ConstIterator(it) !=
entries.constEnd(); it++) {
//         const DvbSharedEpgEntry &existingEntry = *it;
//
//         // Don't do anything if the event already exists
//         if (*existingEntry == entry)
//             return DvbSharedEpgEntry();
//
//         if (existingEntry->channel != entry.channel)
//             continue;
//
//         const QDateTime enEnd = existingEntry->begin.addSecs(QTime(0, 0,
0).secsTo(existingEntry->duration));
//
//         // A new event conflicts with an existing one
//         if (((entry.begin > existingEntry->begin) && (entry.begin < enEnd))
|| ((end > existingEntry->begin) && end < enEnd)) {
//             Debug("removed", existingEntry);
//             it = removeEntry(it);
//             break;
//         }
//         // New event data for the same event
//         if (((entry.begin == existingEntry->begin) && (end == enEnd))) {
//             if (existingEntry->details.isEmpty() &&
!entry.details.isEmpty()) {
//                 emit entryAboutToBeUpdated(existingEntry);
//                 const_cast<DvbEpgEntry
*>(existingEntry.constData())->details =
//                     entry.details;
//                 emit entryUpdated(existingEntry);
//                 Debug("updated", existingEntry);
//             }
//             return existingEntry;
//         }
//     }


Reproducible: Always




System Info:
Linux Mint 18 KDE
CPU: 4 x Intel Core 2 Quad CPU Q8200 @ 2.33GHz
Memory: 7.8 GiB RAM
Graphics Card: NVIDIA Corporation G92 [GeForce GT 230 OEM]

-- 
You are receiving this mail because:
You are watching all bug changes.

Reply via email to