Code: Select all

0x6ffc1410 in __kernel_vsyscall ()
(gdb) bt
#0  0x6ffc1410 in __kernel_vsyscall ()
#1  0x6e4bac1e in ?? () from /lib/libpthread.so.0
#2  0x0a46e23c in ?? ()
#3  0x6e4bfff4 in ?? () from /lib/libpthread.so.0
#4  0x6e4b72c0 in ?? () from /lib/libpthread.so.0
#5  0x6e4b7564 in ?? () from /lib/libpthread.so.0
#6  0x00000011 in ?? ()
#7  0x6ea86d26 in pthread_mutex_lock () from /lib/libc.so.6
#8  0x6f23c487 in QRecursiveMutexPrivate::lock () from /usr/qt/3/lib/libqt-mt.so.3
#9  0x6f23bfd2 in QMutex::lock () from /usr/qt/3/lib/libqt-mt.so.3
#10 0x0809455e in QMutexLocker (this=0x77b53b3c, m=0x9c08700) at /usr/qt/3/include/qmutex.h:96
#11 0x6fe95b78 in bt::PacketWriter::queuePacket (this=0x9c086d8, p=0xa1ee590) at packetwriter.cpp:74
#12 0x6fe9678e in bt::PacketWriter::sendCancel (this=0x9c086d8, r=@0x1675dd90) at packetwriter.cpp:137
#13 0x6fe8d734 in bt::PeerDownloader::checkTimeouts (this=0xaed27c0) at peerdownloader.cpp:226
#14 0x6fe78033 in bt::Downloader::update (this=0x8e3bd58) at downloader.cpp:149
#15 0x6fe83ebf in bt::TorrentControl::update (this=0xaa1e628) at torrentcontrol.cpp:187
#16 0x08082c9e in KTorrentCore::update (this=0x82dbb28) at ktorrentcore.cpp:624
#17 0x080854f9 in KTorrentCore::qt_invoke (this=0x82dbb28, _id=5, _o=0x77b53dfc) at ktorrentcore.moc:243
#18 0x6eefd213 in QObject::activate_signal () from /usr/qt/3/lib/libqt-mt.so.3
#19 0x6eefdfac in QObject::activate_signal () from /usr/qt/3/lib/libqt-mt.so.3
#20 0x6f2fc1be in QTimer::timeout () from /usr/qt/3/lib/libqt-mt.so.3
#21 0x6ef2bb15 in QTimer::event () from /usr/qt/3/lib/libqt-mt.so.3
#22 0x6ee88943 in QApplication::internalNotify () from /usr/qt/3/lib/libqt-mt.so.3
#23 0x6ee8a8bb in QApplication::notify () from /usr/qt/3/lib/libqt-mt.so.3
#24 0x6f5ed3ca in KApplication::notify () from /usr/kde/3.5/lib/libkdecore.so.4
#25 0x77b54400 in ?? ()
#26 0x082dbb5c in ?? ()
#27 0x77b54158 in ?? ()
#28 0x6eb9c439 in operator new () from /usr/lib/gcc/i686-pc-linux-gnu/4.1.1/libstdc++.so.6
Backtrace stopped: previous frame inner to this frame (corrupt stack?)


Looks like a nice deadlock bug I hope it isnt hard to fix. Locks up the app completely, so I'm guessing its getting stuck between the GUI update and one of the transfer threads (or both, that'd suck).

Can you get a stacktrace of the other threads ?

This is probably a rather complex situation with a couple of mutexes. Threading can be a tricky thing.

Sorry, I already exited.. I'll let you know when It happens again. Will probably happen again later today.

I'm getting nothing from "info threads". literally nothing. so I don't know the thread IDs..

Also, ps/top/htop seem to think ktorrent was using up 50% cpu+ when it was frozen. Not sure whats going on. I'd figure a deadlock would make the threads just wait on the mutex, and yield to the os.

Tomasu wrote:I'm getting nothing from "info threads". literally nothing. so I don't know the thread IDs..

Also, ps/top/htop seem to think ktorrent was using up 50% cpu+ when it was frozen. Not sure whats going on. I'd figure a deadlock would make the threads just wait on the mutex, and yield to the os.

There are 3 threads, so one will probably still be running, probably the download thread.

Ok, thats odd that its still using 50%-80% cpu when just the download thread is running, since I disabled my download limits yesterday or so. it shouldn't be doing what it's doing no?

And I don't get why gdb wasn't working.