After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 697236 - v4l2src: SEGFAULT when changing state of uvch264src pipeline
v4l2src: SEGFAULT when changing state of uvch264src pipeline
Status: RESOLVED DUPLICATE of bug 695981
Product: GStreamer
Classification: Platform
Component: gst-plugins-good
git master
Other Linux
: Normal normal
: git master
Assigned To: GStreamer Maintainers
GStreamer Maintainers
Depends on:
Blocks:
 
 
Reported: 2013-04-04 02:22 UTC by Tristan Matthews
Modified: 2013-04-17 19:38 UTC
See Also:
GNOME target: ---
GNOME version: ---

Attachments
Check if pool is NULL before calling gst_v4l2_buffer_pool_process (487 bytes, patch)
2013-04-04 02:22 UTC, Tristan Matthews 		none Details | Review
log with GST_DEBUG=*basesrc*:8,*v4l2*:8 up until SEGFAULT (243.51 KB, text/x-log)
2013-04-05 03:27 UTC, Tristan Matthews 		  Details
output of thread apply all bt (15.83 KB, text/plain)
2013-04-05 03:28 UTC, Tristan Matthews 		  Details
log with GST_DEBUG=*basesrc*:8,*v4l2*:8 up until Abort (1.05 MB, text/plain)
2013-04-17 06:52 UTC, Tristan Matthews 		  Details

Description Tristan Matthews 2013-04-04 02:22:00 UTC 
Created attachment 240561 [details] [review]
Check if pool is NULL before calling gst_v4l2_buffer_pool_process

To reproduce, run the uvch264 test program, change the state to playing after hitting capture. 

This bug does not happen every time.

I've attached a patch which fixes the immediate problem of the NULL pointer access, however I'm not sure if it's the right approach.

Here's the backtrace.

+ Trace 231723

  • #0 gst_v4l2_buffer_pool_process
    at gstv4l2bufferpool.c line 1073
  • #1 gst_v4l2src_fill
    at gstv4l2src.c line 771
  • #2 gst_push_src_fill
    at gstpushsrc.c line 169
  • #3 gst_base_src_default_create
    at gstbasesrc.c line 1434
  • #4 gst_push_src_create
    at gstpushsrc.c line 134
  • #5 gst_push_src_create
    at gstpushsrc.c line 122
  • #6 gst_base_src_get_range
    at gstbasesrc.c line 2357
  • #7 gst_base_src_loop
    at gstbasesrc.c line 2618
  • #8 gst_task_func
    at gsttask.c line 316
  • #9 default_func
    at gsttaskpool.c line 70
  • #10 g_thread_pool_thread_proxy
    at /build/buildd/glib2.0-2.34.1/./glib/gthreadpool.c line 309
  • #11 g_thread_proxy
    at /build/buildd/glib2.0-2.34.1/./glib/gthread.c line 797
  • #12 start_thread
    at pthread_create.c line 308
  • #13 clone
    at ../sysdeps/unix/sysv/linux/i386/clone.S line 130 






Comment 1


Tim-Philipp Müller





          2013-04-04 09:09:07 UTC
        



I think there's a similar bug somewhere just with v4l2src, not sure if it got fixed or not.






Comment 2


Wim Taymans





          2013-04-04 12:04:56 UTC
        



I don't understand in what condition the pool can be NULL. The only reason I can think of is when negotiation fails but then it should never reach the fill method.

Do you have a debug log? something like:

 GST_DEBUG=*basesrc*:8,*v4l2*:8

should be enough






Comment 3


Tim-Philipp Müller





          2013-04-04 12:14:37 UTC
        



 Bug #695981 looks the same, and has a script to reproduce it.






Comment 4


Tristan Matthews





          2013-04-05 03:27:33 UTC
        



Created attachment 240688 [details]
log with GST_DEBUG=*basesrc*:8,*v4l2*:8 up until SEGFAULT






Comment 5


Tristan Matthews





          2013-04-05 03:28:44 UTC
        



Created attachment 240689 [details]
output of thread apply all bt






Comment 6


Tristan Matthews





          2013-04-05 03:29:17 UTC
        



Ok I've added the logs plus the backtrace from GDB for all the threads, let me know if you need more info.






Comment 7


Olivier Crête





          2013-04-15 22:26:28 UTC
        



I assume you have a Logitech C920 ?






Comment 8


Olivier Crête





          2013-04-16 00:00:34 UTC
        



I can't reproduce the bug, but I'm pretty sure Tim is right that this is the same as #695981, can you please test the two patches attached there ?

*** This bug has been marked as a duplicate of bug 695981 ***






Comment 9


Tristan Matthews





          2013-04-16 14:24:16 UTC
        



(In reply to comment #7)
> I assume you have a Logitech C920 ?

yup






Comment 10


Tristan Matthews





          2013-04-16 14:24:41 UTC
        



(In reply to comment #8)
> I can't reproduce the bug, but I'm pretty sure Tim is right that this is the
> same as #695981, can you please test the two patches attached there ?
> 
> *** This bug has been marked as a duplicate of bug 695981 ***

Will do, I should be able to test that tonight.






Comment 11


Tristan Matthews





          2013-04-17 03:28:21 UTC
        



Ok I've tested the patches from Bug #695981 and I'm no longer getting the segfault, but I do get a SIGABORT, here are the logs:

0:00:29.317479955 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool5> error requesting 4 buffers: Device or resource busy
0:00:29.318569774 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool5> error requesting 4 buffers: Device or resource busy
0:00:29.318629140 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool5> error requesting 4 buffers: Device or resource busy
0:00:29.318681243 18999  0x82da430 ERROR             bufferpool gstbufferpool.c:485:gst_buffer_pool_set_active:<v4l2bufferpool5> pool was not configured
0:00:29.318717351 18999  0x82da430 ERROR                basesrc gstbasesrc.c:2871:gst_base_src_set_allocation:<v4l2src2> failed to activate bufferpool.
0:00:30.623800030 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool6> error requesting 4 buffers: Device or resource busy
0:00:30.624452498 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool6> error requesting 4 buffers: Device or resource busy
0:00:30.624511724 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool6> error requesting 4 buffers: Device or resource busy
0:00:30.624550068 18999  0x82da430 ERROR             bufferpool gstbufferpool.c:485:gst_buffer_pool_set_active:<v4l2bufferpool6> pool was not configured
0:00:30.624580798 18999  0x82da430 ERROR                basesrc gstbasesrc.c:2871:gst_base_src_set_allocation:<v4l2src3> failed to activate bufferpool.
0:00:32.574675228 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool7> error requesting 4 buffers: Device or resource busy
0:00:32.575328533 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool7> error requesting 4 buffers: Device or resource busy
0:00:32.575386921 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool7> error requesting 4 buffers: Device or resource busy
0:00:32.575420725 18999  0x82da430 ERROR             bufferpool gstbufferpool.c:485:gst_buffer_pool_set_active:<v4l2bufferpool7> pool was not configured
0:00:32.575449500 18999  0x82da430 ERROR                basesrc gstbasesrc.c:2871:gst_base_src_set_allocation:<v4l2src4> failed to activate bufferpool.
0:00:33.779579837 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool8> error requesting 4 buffers: Device or resource busy
0:00:33.780233840 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool8> error requesting 4 buffers: Device or resource busy
0:00:33.780291739 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool8> error requesting 4 buffers: Device or resource busy
0:00:33.780330641 18999  0x82da430 ERROR             bufferpool gstbufferpool.c:485:gst_buffer_pool_set_active:<v4l2bufferpool8> pool was not configured
0:00:33.780353759 18999  0x82da430 ERROR                basesrc gstbasesrc.c:2871:gst_base_src_set_allocation:<v4l2src5> failed to activate bufferpool.
0:00:34.706867921 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool9> error requesting 4 buffers: Device or resource busy
0:00:34.707544344 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool9> error requesting 4 buffers: Device or resource busy
0:00:34.707602732 18999  0x82da430 ERROR                   v4l2 gstv4l2bufferpool.c:414:gst_v4l2_buffer_pool_set_config:<v4l2bufferpool9> error requesting 4 buffers: Device or resource busy
0:00:34.707636186 18999  0x82da430 ERROR             bufferpool gstbufferpool.c:485:gst_buffer_pool_set_active:<v4l2bufferpool9> pool was not configured
0:00:34.707660142 18999  0x82da430 ERROR                basesrc gstbasesrc.c:2871:gst_base_src_set_allocation:<v4l2src6> failed to activate bufferpool.
**
ERROR:gstv4l2bufferpool.c:922:gst_v4l2_buffer_pool_release_buffer: code should not be reached

Program received signal SIGABRT, Aborted.
0xb7fdd424 in __kernel_vsyscall ()






Comment 12


Tristan Matthews





          2013-04-17 03:29:20 UTC
        



And the backtrace:





+
Trace
    231804






Thread 1
          (Thread 0xb6e4a840 (LWP 18999))


    

  • 
#0
__kernel_vsyscall

    • 

  • 
#1
__GI_raise

    
                at ../nptl/sysdeps/unix/sysv/linux/raise.c
                  line
                  64

    

    • 

  • 
#2
__GI_abort

    
                at abort.c
                  line
                  91

    

    • 

  • 
#3
g_assertion_message

    
                at /build/buildd/glib2.0-2.34.1/./glib/gtestutils.c
                  line
                  1877

    

    • 

  • 
#4
gst_v4l2_buffer_pool_release_buffer

    
                at gstv4l2bufferpool.c
                  line
                  922

    

    • 

  • 
#5
gst_buffer_pool_release_buffer

    
                at gstbufferpool.c
                  line
                  1155

    

    • 

  • 
#6
_gst_buffer_dispose

    
                at gstbuffer.c
                  line
                  540

    

    • 

  • 
#7
gst_mini_object_unref

    
                at gstminiobject.c
                  line
                  448

    

    • 

  • 
#8
gst_buffer_unref

    
                at ../../../gst/gstbuffer.h
                  line
                  354

    

    • 

  • 
#9
gst_base_sink_set_last_buffer_unlocked

    
                at gstbasesink.c
                  line
                  954

    

    • 

  • 
#10
gst_base_sink_set_last_buffer

    
                at gstbasesink.c
                  line
                  966

    

    • 

  • 
#11
gst_base_sink_set_last_buffer

    
                at gstbasesink.c
                  line
                  960

    

    • 

  • 
#12
gst_base_sink_change_state

    
                at gstbasesink.c
                  line
                  4957

    

    • 

  • 
#13
gst_xvimagesink_change_state

    
                at xvimagesink.c
                  line
                  856

    

    • 

  • 
#14
gst_element_change_state

    
                at gstelement.c
                  line
                  2605

    

    • 

  • 
#15
gst_element_set_state_func

    
                at gstelement.c
                  line
                  2561

    

    • 

  • 
#16
gst_element_set_state

    
                at gstelement.c
                  line
                  2462

    

    • 

  • 
#17
gst_bin_element_set_state

    
                at gstbin.c
                  line
                  2297

    

    • 

  • 
#18
gst_bin_change_state_func

    
                at gstbin.c
                  line
                  2599

    

    • 

  • 
#19
gst_pipeline_change_state

    
                at gstpipeline.c
                  line
                  471

    

    • 

  • 
#20
gst_element_change_state

    
                at gstelement.c
                  line
                  2605

    

    • 

  • 
#21
gst_element_continue_state

    
                at gstelement.c
                  line
                  2315

    

    • 

  • 
#22
gst_element_change_state

    
                at gstelement.c
                  line
                  2649

    

    • 

  • 
#23
gst_element_set_state_func

    
                at gstelement.c
                  line
                  2561

    

    • 

  • 
#24
gst_element_set_state

    
                at gstelement.c
                  line
                  2462

    

    • 

  • 
#25
on_button_ready_clicked

    
                at test-uvch264.c
                  line
                  267

    

    • 

  • 
#26
g_cclosure_marshal_VOID__VOIDv

    
                at /build/buildd/glib2.0-2.34.1/./gobject/gmarshal.c
                  line
                  115

    

    • 

  • 
#27
_g_closure_invoke_va

    
                at /build/buildd/glib2.0-2.34.1/./gobject/gclosure.c
                  line
                  840

    

    • 

  • 
#28
g_signal_emit_valist

    
                at /build/buildd/glib2.0-2.34.1/./gobject/gsignal.c
                  line
                  3211

    

    • 

  • 
#29
g_signal_emit

    
                at /build/buildd/glib2.0-2.34.1/./gobject/gsignal.c
                  line
                  3356

    

    • 

  • 
#30
gtk_button_clicked

    
                from
                /usr/lib/i386-linux-gnu/libgtk-x11-2.0.so.0

    

    • 

  • 
#31
??

    
                from
                /usr/lib/i386-linux-gnu/libgtk-x11-2.0.so.0

    

    • 

  • 
#32
g_cclosure_marshal_VOID__VOID

    
                at /build/buildd/glib2.0-2.34.1/./gobject/gmarshal.c
                  line
                  85

    

    • 

  • 
#33
g_type_class_meta_marshal

    
                at /build/buildd/glib2.0-2.34.1/./gobject/gclosure.c
                  line
                  970

    

    • 

  • 
#34
g_closure_invoke

    • 

  • 
#35
signal_emit_unlocked_R

    
                at /build/buildd/glib2.0-2.34.1/./gobject/gsignal.c
                  line
                  3481

    

    • 

  • 
#36
g_signal_emit_valist

    
                at /build/buildd/glib2.0-2.34.1/./gobject/gsignal.c
                  line
                  3300

    

    • 

  • 
#37
g_signal_emit

    
                at /build/buildd/glib2.0-2.34.1/./gobject/gsignal.c
                  line
                  3356

    

    • 

  • 
#38
gtk_button_released

    
                from
                /usr/lib/i386-linux-gnu/libgtk-x11-2.0.so.0

    

    • 

  • 
#39
??

    
                from
                /usr/lib/i386-linux-gnu/libgtk-x11-2.0.so.0

    

    • 

  • 
#40
??

    
                from
                /usr/lib/i386-linux-gnu/libgtk-x11-2.0.so.0

    

    • 

  • 
#41
g_type_class_meta_marshal

    
                at /build/buildd/glib2.0-2.34.1/./gobject/gclosure.c
                  line
                  970

    

    • 

  • 
#42
g_closure_invoke

    
                at /build/buildd/glib2.0-2.34.1/./gobject/gclosure.c
                  line
                  777

    

    • 

  • 
#43
signal_emit_unlocked_R

    • 

  • 
#44
g_signal_emit_valist

    
                at /build/buildd/glib2.0-2.34.1/./gobject/gsignal.c
                  line
                  3310

    

    • 

  • 
#45
g_signal_emit

    
                at /build/buildd/glib2.0-2.34.1/./gobject/gsignal.c
                  line
                  3356

    

    • 

  • 
#46
??

    
                from
                /usr/lib/i386-linux-gnu/libgtk-x11-2.0.so.0

    

    • 

  • 
#47
gtk_propagate_event

    
                from
                /usr/lib/i386-linux-gnu/libgtk-x11-2.0.so.0

    

    • 

  • 
#48
gtk_main_do_event

    
                from
                /usr/lib/i386-linux-gnu/libgtk-x11-2.0.so.0

    

    • 

  • 
#49
??

    
                from
                /usr/lib/i386-linux-gnu/libgdk-x11-2.0.so.0

    

    • 

  • 
#50
g_main_dispatch

    
                at /build/buildd/glib2.0-2.34.1/./glib/gmain.c
                  line
                  2715

    

    • 

  • 
#51
g_main_context_dispatch

    
                at /build/buildd/glib2.0-2.34.1/./glib/gmain.c
                  line
                  3219

    

    • 

  • 
#52
g_main_context_iterate

    
                at /build/buildd/glib2.0-2.34.1/./glib/gmain.c
                  line
                  3290

    

    • 

  • 
#53
g_main_loop_run

    
                at /build/buildd/glib2.0-2.34.1/./glib/gmain.c
                  line
                  3484

    

    • 

  • 
#54
gtk_main

    
                from
                /usr/lib/i386-linux-gnu/libgtk-x11-2.0.so.0

    

    • 

  • 
#55
main

    
                at test-uvch264.c
                  line
                  657

    

    • 













Comment 13


Olivier Crête





          2013-04-17 05:37:24 UTC
        



Can you attach more of the log? with GST_DEBUG=*v4l2*:8,basesrc:8 ?






Comment 14


Tristan Matthews





          2013-04-17 06:52:20 UTC
        



Created attachment 241713 [details]
log with GST_DEBUG=*basesrc*:8,*v4l2*:8 up until Abort






Comment 15


Olivier Crête





          2013-04-17 19:19:15 UTC
        



I'm not sure what you're application is.. but with the current v4l2src, you have to shut down the entire pipeline while stopping v4l2src. Work to fix that has been done on bug #682770






Comment 16


Tristan Matthews





          2013-04-17 19:38:23 UTC
        



All this testing was with:

http://cgit.freedesktop.org/gstreamer/gst-plugins-bad/tree/tests/examples/uvch264/test-uvch264.c