... #13549 0x00007ffff7bd0675 in play_PGC (vm=vm@entry=0x7fffb4056e10) at src/vm/play.c:83 #13550 0x00007ffff7bd07e0 in play_PGC_post (vm=vm@entry=0x7fffb4056e10) at src/vm/play.c:153 #13551 0x00007ffff7bd0542 in play_PG (vm=vm@entry=0x7fffb4056e10) at src/vm/play.c:168 #13552 0x00007ffff7bd0675 in play_PGC (vm=vm@entry=0x7fffb4056e10) at src/vm/play.c:83 #13553 0x00007ffff7bd07e0 in play_PGC_post (vm=vm@entry=0x7fffb4056e10) at src/vm/play.c:153 #13554 0x00007ffff7bd0542 in play_PG (vm=vm@entrye=0x7fffb4056e10) at src/vm/play.c:168 #13555 0x00007ffff7bd0675 in play_PGC (vm=vm@entry=0x7fffb4056e10) at src/vm/play.c:83 #13556 0x00007ffff7bd07e0 in play_PGC_post (vm=vm@entry=0x7fffb4056e10) at src/vm/play.c:153 #13557 0x00007ffff7bd0542 in play_PG (vm=vm@entry=0x7fffb4056e10) at src/vm/play.c:168 #13558 0x00007ffff7bd0675 in play_PGC (vm=vm@entry=0x7fffb4056e10) at src/vm/play.c:83 ...
looks like it's definitely infinitely recursing to me. I seem to recall some ancient patch that mentions ref counts something, wonder if that's related... http://lists.mplayerhq.hu/pipermail/dvdnav-discuss/2012-December/001816.html re-proposed once here http://comments.gmane.org/gmane.comp.video.dvdnav.general/1658 [BTW does that patch have merit, even if unrelated, I wonder?] On Fri, Aug 15, 2014 at 10:23 AM, Michael Russo <[email protected]> wrote: > That did help and I was now able to compile libdvdnav with TRACE defined. > I still ran vlc through gdb in order to get a backtrace. There was a LOT > of tracing going on, and now the SIGSEGV occurred inside of printf() > because it was trying to print out the invalid memory address. Hopefully > there's something here you can use to track down what's going on. It's > likely some weird form of copy protection. > > > On Thu, Aug 14, 2014 at 10:50 PM, gnosygnu <[email protected]> wrote: > > > On Thu, Aug 14, 2014 at 9:01 AM, Michael Russo <[email protected]> > > wrote: > > > > > I tried to define TRACE in vm.h (uncommenting the line) but I get a > > > message about an undefined constant FP_DOMAIN. The compilation > > > succeeds without TRACE enabled. I also can't find where this constant > > > is defined inside of libdvdnav. Perhaps this is a simple oversight > > > related to all the cleanup work that took place in 5.0.0? > > > > > > > > Actually, it was probably related to an earlier commit. See: > > > > > http://git.videolan.org/?p=libdvdnav.git;a=commitdiff;h=331781f98a8ea6842beee38539f6470c3e8a819c > > > > I updated the constants just now: > > > > > http://git.videolan.org/?p=libdvdnav.git;a=commitdiff;h=da8a809a069c954872391acf7d6c97af1cdbc3a1 > > > > I also cleaned up some compiler warnings in my earlier checkin that I > > didn't notice before. Sorry about that. > > > > > http://git.videolan.org/?p=libdvdnav.git;a=commitdiff;h=03cb72be09295686cf1af0af5d8661ab8d35b537 > > > > Hope this helps. > > _______________________________________________ > > DVDnav-discuss mailing list > > [email protected] > > https://lists.mplayerhq.hu/mailman/listinfo/dvdnav-discuss > > > > _______________________________________________ > DVDnav-discuss mailing list > [email protected] > https://lists.mplayerhq.hu/mailman/listinfo/dvdnav-discuss > _______________________________________________ DVDnav-discuss mailing list [email protected] https://lists.mplayerhq.hu/mailman/listinfo/dvdnav-discuss
