On Sun, 27 Nov 2011 00:06:44 +0100 Andreas Volz <li...@brachttal.net> said:
> Am Sat, 26 Nov 2011 20:38:35 +0900 schrieb Carsten Haitzler (The > Rasterman): > > Hm, would you vote for an EFL library bug or for a bug in my > application? > > I have a multi-threading application. But I'm using pthreads and sure > not to call ecore lock/unlock directly, but the pthread locking > functions. aree you calling any ecore, edje, elementary etc.calls from a thread? take a good audit of them and see. > So in which component should I start searching/tracing this bug? > > regards > Andreas > > > On Sat, 26 Nov 2011 11:00:55 +0100 Andreas Volz <li...@brachttal.net> > > said: > > > > oh yes you're right - its just the ecore thread correctness detection > > code. somethng is not doing lock/unlock correctly on enter/exit of > > ecore funcs... we just dont know what. > > > > > Am Sat, 26 Nov 2011 13:21:11 +0900 schrieb Carsten Haitzler (The > > > Rasterman): > > > > > > I don't think that I've activated threadsafety in ecore. If I run > > > configure again in ecore I get: > > > > > > Ecore........................: always > > > Thread support.............: POSIX > > > Thread safety..............: no > > > > > > So I don't know about using this. And the assert() is outside the > > > #ifdef HAVE_THREAD_SAFETY. > > > > > > So any other hints? > > > > > > regards > > > Andreas > > > > > > > well this is a bug in the threadsafety locks in ecore - they > > > > really haven't been tested much or used, so best is not to enable > > > > them. they are disabled by default. > > > > > > > > > Hello, > > > > > > > > > > in rare cases I get this assert in my application: > > > > > > > > > > lt-oiui: ecore_private.h:248: _ecore_unlock: Zusicherung > > > > > »_ecore_main_lock_count == 0« nicht erfüllt. lt-oiui: > > > > > ecore_private.h:241: _ecore_lock: Zusicherung > > > > > »_ecore_main_lock_count == 1« nicht erfüllt. > > > > > > > > > > This is gdb backtrace: > > > > > Program terminated with signal 6, Aborted. > > > > > #0 0x00dd5422 in __kernel_vsyscall () > > > > > (gdb) bt > > > > > #0 0x00dd5422 in __kernel_vsyscall () > > > > > #1 0x0101e651 in *__GI_raise (sig=6) > > > > > #at ../nptl/sysdeps/unix/sysv/linux/raise.c:64 2 0x01021a82 in > > > > > *__GI_abort #() at abort.c:92 3 0x01017718 in > > > > > *__GI___assert_fail ( assertion=0x360a0c > > > > > "_ecore_main_lock_count == 0", file=0x3609e0 "ecore_private.h", > > > > > line=248, function=0x360de7 "_ecore_unlock") at assert.c:81 > > > > > #4 0x00353604 in _ecore_unlock () at ecore_private.h:248 > > > > > #5 0x00354b63 in _ecore_main_select > > > > > (timeout=0.00068952832589275204) at ecore_main.c:1393 > > > > > #6 0x00355707 in _ecore_main_loop_iterate_internal > > > > > (once_only=0) at #ecore_main.c:1810 7 0x0035402a in > > > > > ecore_main_loop_begin () at #ecore_main.c:887 8 0x002ad4d8 in > > > > > elm_run () at elm_main.c:828 #9 0x001e1eb3 in > > > > > Elmxx::Application::run () at Application.cpp:26 #10 0x08061642 > > > > > in Main (this=0xbf918cf0, argc=1, argv=0xbf918e94) at > > > > > #Main.cpp:106 11 0x08061bad in main (argc=1, argv=0xbf918e94) > > > > > at Main.cpp:206 > > > > > > > > > > It happens while pressing on an elm_button inside a elm_layout > > > > > loaded from edj external. > > > > > > > > > > I tried to reproduce this by hitting dozen times on the button > > > > > - no chance. But "sometimes" it happens. > > > > > > > > > > Any ideas where to start searching this bug? > > > > > > > > > > regards > > > > > Andreas > > > > > > > > > > -- > > > > > Technical Blog <http://andreasvolz.wordpress.com/> > > > > > > > > > > ------------------------------------------------------------------------------ > > > > > All the data continuously generated in your IT infrastructure > > > > > contains a definitive record of customers, application > > > > > performance, security threats, fraudulent activity, and more. > > > > > Splunk takes this data and makes sense of it. IT sense. And > > > > > common sense. http://p.sf.net/sfu/splunk-novd2d > > > > > _______________________________________________ > > > > > enlightenment-devel mailing list > > > > > enlightenment-devel@lists.sourceforge.net > > > > > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > > > > > > > > > > > > -- > > > > ------------- Codito, ergo sum - "I code, therefore I am" > > > > -------------- The Rasterman (Carsten Haitzler) > > > > ras...@rasterman.com > > > > > > > > > > > > > -- > > > Technical Blog <http://andreasvolz.wordpress.com/> > > > > > > ------------------------------------------------------------------------------ > > > All the data continuously generated in your IT infrastructure > > > contains a definitive record of customers, application performance, > > > security threats, fraudulent activity, and more. Splunk takes this > > > data and makes sense of it. IT sense. And common sense. > > > http://p.sf.net/sfu/splunk-novd2d > > > _______________________________________________ > > > enlightenment-devel mailing list > > > enlightenment-devel@lists.sourceforge.net > > > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > > > > > > -- > > ------------- Codito, ergo sum - "I code, therefore I am" > > -------------- The Rasterman (Carsten Haitzler) > > ras...@rasterman.com > > > > > > ------------------------------------------------------------------------------ > > All the data continuously generated in your IT infrastructure > > contains a definitive record of customers, application performance, > > security threats, fraudulent activity, and more. Splunk takes this > > data and makes sense of it. IT sense. And common sense. > > http://p.sf.net/sfu/splunk-novd2d > > _______________________________________________ > > enlightenment-devel mailing list > > enlightenment-devel@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > > > -- > Technical Blog <http://andreasvolz.wordpress.com/> > > ------------------------------------------------------------------------------ > All the data continuously generated in your IT infrastructure > contains a definitive record of customers, application performance, > security threats, fraudulent activity, and more. Splunk takes this > data and makes sense of it. IT sense. And common sense. > http://p.sf.net/sfu/splunk-novd2d > _______________________________________________ > enlightenment-devel mailing list > enlightenment-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel -- ------------- Codito, ergo sum - "I code, therefore I am" -------------- The Rasterman (Carsten Haitzler) ras...@rasterman.com ------------------------------------------------------------------------------ All the data continuously generated in your IT infrastructure contains a definitive record of customers, application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-novd2d _______________________________________________ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
