banshee eats all my CPU with SMP systems

[Ariel Vardi]

After a while, banshee suddenly uses all my CPU resources. It still plays the current song until the end but doesn't start the next one.

I'm not sure what else, I could tell you about it.
I'm running a dapper up-to-date, and I think this bug appeared pretty recently (maybe 2 weeks or sthg...)

[Sebastian Dröge (slomo)]

Do you still have this problem?
Could you try attaching gdb to the CPU-eating banshee process?
gdb -p `pidof banshee`

and then enter
"thread apply all bt" and paste the output here?

Also does this happen with all types of files?

[Ariel Vardi (ariel-vardi)]

I still have the problem on dapper up-to-date. It just froze again on an mp3. I think it also crashed before on an ogg.

I did what you told me with gdb and here is the output:

(gdb) thread apply all bt

Thread 1 (Thread -1212421824 (LWP 11610)):
#0 0xffffe410 in __kernel_vsyscall ()
#1 0xb7c7e8c4 in poll () from /lib/tls/i686/cmov/libc.so.6
#2 0xb4f0132c in snd_pcm_direct_check_interleave ()
   from /usr/lib/libasound.so.2
#3 0xb4f0178e in snd_pcm_direct_server_create () from /usr/lib/libasound.so.2
#4 0xb4f034c4 in snd_pcm_dmix_open () from /usr/lib/libasound.so.2
#5 0xb4f03b87 in _snd_pcm_dmix_open () from /usr/lib/libasound.so.2
#6 0xb4ecce69 in snd_pcm_free () from /usr/lib/libasound.so.2
#7 0xb4ecd4e9 in snd_pcm_free () from /usr/lib/libasound.so.2
#8 0xb4ecd714 in snd_pcm_open_slave () from /usr/lib/libasound.so.2
#9 0xb4f09a72 in _snd_pcm_softvol_open () from /usr/lib/libasound.so.2
#10 0xb4ecce69 in snd_pcm_free () from /usr/lib/libasound.so.2
#11 0xb4ecd737 in snd_pcm_open_slave () from /usr/lib/libasound.so.2
#12 0xb4eeb165 in _snd_pcm_plug_open () from /usr/lib/libasound.so.2
#13 0xb4ecce69 in snd_pcm_free () from /usr/lib/libasound.so.2
#14 0xb4ecd737 in snd_pcm_open_slave () from /usr/lib/libasound.so.2
#15 0xb4f076a6 in _snd_pcm_asym_open () from /usr/lib/libasound.so.2
#16 0xb4ecce69 in snd_pcm_free () from /usr/lib/libasound.so.2
#17 0xb4ecd4e9 in snd_pcm_free () from /usr/lib/libasound.so.2
#18 0xb4b1c425 in ?? () from /usr/lib/gstreamer-0.10/libgstalsa.so
#19 0x00000000 in ?? ()
---Type <return> to continue, or q <return> to quit---
#0 0xffffe410 in __kernel_vsyscall ()
(gdb)

Hope it helps

[Sebastian Dröge (slomo)]

are you sure that this is the complete output? there should be more than one thread otherwise something is really broken :(

[Ariel Vardi (ariel-vardi)]

That's all the output after the command 'thread apply all bt' yeah

On Sun, 2006-04-16 at 00:03 +0000, Sebastian Dröge wrote:
> are you sure that this is the complete output? there should be more than one thread otherwise something is really broken :(

[Wade Menard (wade-ezri)]

I have been seeing this as well. This is with both the latest banshee deb and the latest CVS HEAD. I am still trying to debug it myself. Here is the bt you wanted.

Thread 10 (Thread -1210516560 (LWP 16115)):
#0 0xffffe410 in __kernel_vsyscall ()
#1 0xb7ce6d56 in __nanosleep_nocancel ()
   from /lib/tls/i686/cmov/libpthread.so.0
#2 0xb7ec4956 in mono_disasm_code_one () from /usr/lib/libmono.so.0
#3 0xb7ce1341 in start_thread () from /lib/tls/i686/cmov/libpthread.so.0
#4 0xb7c4b4ee in clone () from /lib/tls/i686/cmov/libc.so.6

Thread 9 (Thread -1221842000 (LWP 16116)):
#0 0xb7d1bc5b in g_hash_table_remove () from /usr/lib/libglib-2.0.so.0
#1 0xb6bd1677 in gdk_event_free () from /usr/lib/libgdk-x11-2.0.so.0
#2 0xb6a5094a in g_value_dup_boxed () from /usr/lib/libgobject-2.0.so.0
#3 0xb6a72b28 in g_value_unset () from /usr/lib/libgobject-2.0.so.0
#4 0xb6a73918 in g_value_array_free () from /usr/lib/libgobject-2.0.so.0
#5 0xb452456c in ?? ()
#6 0x089beaa0 in ?? ()
#7 0x002c2288 in ?? ()
#8 0xb7df7dd3 in mini_method_compile () from /usr/lib/libmono.so.0
#9 0xb452451a in ?? ()
#10 0x089beaa0 in ?? ()
#11 0x0850e008 in ?? ()
#12 0xb72c225c in ?? ()
#13 0xb4524428 in ?? ()
#14 0x002208d0 in ?? ()
#15 0x00000000 in ?? ()

Thread 8 (Thread -1234646096 (LWP 16119)):
#0 0xffffe410 in __kernel_vsyscall ()
#1 0xb7ce6d56 in __nanosleep_nocancel ()
   from /lib/tls/i686/cmov/libpthread.so.0
#2 0xb7ed6af6 in SleepEx () from /usr/lib/libmono.so.0
#3 0xb7e8a384 in ves_icall_System_Threading_Thread_Sleep_internal ()
   from /usr/lib/libmono.so.0
#4 0xb67354d5 in ?? ()
#5 0x0000000a in ?? ()
#6 0x0822e1e8 in ?? ()
#7 0x0822daa8 in ?? ()
#8 0xb5300490 in ?? ()
#9 0x0002d300 in ?? ()
#10 0x0000000a in ?? ()
#11 0x00000000 in ?? ()

Thread 7 (Thread -1286853712 (LWP 16123)):
#0 0xffffe410 in __kernel_vsyscall ()
#1 0xb7c418c4 in poll () from /lib/tls/i686/cmov/libc.so.6
#2 0xb7d29788 in g_main_context_check () from /usr/lib/libglib-2.0.so.0
#3 0xb7d29c58 in g_main_loop_run () from /usr/lib/libglib-2.0.so.0
#4 0xb679c70e in link_thread_io_context () from /usr/lib/libORBit-2.so.0
#5 0xb7d42582 in g_static_private_free () from /usr/lib/libglib-2.0.so.0
#6 0xb7ce1341 in start_thread () from /lib/tls/i686/cmov/libpthread.so.0
#7 0xb7c4b4ee in clone () from /lib/tls/i686/cmov/libc.so.6

Thread 6 (Thread -1236415568 (LWP 16158)):
#0 0xffffe410 in __kernel_vsyscall ()
#1 0xb7ce3c76 in pthread_cond_wait@@GLIBC_2.3.2 ()
   from /lib/tls/i686/cmov/libpthread.so.0
#2 0xb511e061 in gst_task_get_type () from /usr/lib/libgstreamer-0.10.so.0
#3 0xb7d44428 in g_thread_pool_free () from /usr/lib/libglib-2.0.so.0
#4 0xb7d42582 in g_static_private_free () from /usr/lib/libglib-2.0.so.0
#5 0xb7ce1341 in start_thread () from /lib/tls/i686/cmov/libpthread.so.0
#6 0xb7c4b4ee in clone () from /lib/tls/i686/cmov/libc.so.6

Thread 5 (Thread -1300792400 (LWP 16159)):
#0 0xffffe410 in __kernel_vsyscall ()
#1 0xb7ce3c76 in pthread_cond_wait@@GLIBC_2.3.2 ()
   from /lib/tls/i686/cmov/libpthread.so.0
#2 0xb511e061 in gst_task_get_type () from /usr/lib/libgstreamer-0.10.so.0
#3 0xb7d4442...

[Wade Menard (wade-ezri)]

I am working on this upstream with Aaron Bockover, it turns out this is SMP related.

[Ariel Vardi (ariel-vardi)]

I confirm that this happens on my laptop and it's an Intel Dual Core

[Sebastian Dröge (slomo)]

thanks... I saw your talk on IRC so let's set this to confirmed :)

[Wade Menard (wade-ezri)]

This appears to be fixed by the patch in Bug #43727

[Sebastian Dröge (slomo)]

this is fixed in gtk-sharp2 2.8.2-0ubuntu5 which will be uploaded in a few seconds :)