bugGnash - The GNU Flash player - Bugs: bug #31868, Test run_swfdec_testsuite_m some...

 
 

bug #31868: Test run_swfdec_testsuite_m some times fail with a segfault

Submitted by:  Petter Reinholdtsen <pere>
Submitted on:  Mon 13 Dec 2010 10:28:33 PM UTC  
 
Category: testsuiteSeverity: 6 - Security
Release: masterStatus: Fixed
Privacy: PublicAssigned to: Benjamin Wolsey <bwy>
Open/Closed: Closed

Add a New Comment(Rich Markup)
   

You are not logged in

Please log in, so followups can be emailed to you.

 

(Jump to the original submission Jump to the original submission)

Sat 05 Mar 2011 12:10:53 PM UTC, comment #14:

Original reporter has no time to test, so we'll have to rely on my observations. The loop in comment #2 produced invalid accesses in approx 20% of cases for me before the patch at f74f541652c370888cac6889ea49e4821c686e1a and after the patch none in over one hundred iterations. If anyone else can confirm it would be welcome.

Benjamin Wolsey <bwy>
Project MemberIn charge of this item.
Sat 05 Mar 2011 08:35:11 AM UTC, comment #13:

The full diagnosis is as follows:

movie_definitions are kept alive by movie_root and the MovieLibrary. The latter is static.

SWFParsers are all owned by a movie_definition, running in a separate thread.

As long as movie_root is alive, its MovieLoader thread can continue to create new movie_definitions and place them in the library.

gprocessor called MovieFactory::clear(), which in turn clears the library, but movie_root was still alive and occasionally managed to add another movie_definition to the MovieLibrary after this was cleared. The movie_definition was then kept alive until static destruction occurred. Its parser thus continued to use resources from main (RunResources), as well as accessing static resources such as LogFile, which may be destroyed before or after MovieLibrary.

So the correct fix currently is indeed to make sure movie_root is destroyed before MovieFactory::clear() is called. In future the situation would be improved by reducing the number of statics further and cutting down on threading.

Benjamin Wolsey <bwy>
Project MemberIn charge of this item.
Thu 03 Mar 2011 01:35:46 PM UTC, comment #12:

My latest conjecture is that a movie_definition is parsed, created, and added to the MovieLibrary after MovieLibrary::clear() has been called, so that it's still there on exit.

Benjamin Wolsey <bwy>
Project MemberIn charge of this item.
Wed 02 Mar 2011 04:13:29 PM UTC, comment #11:

read_all_swf() continues to parse the SWF after movie_root and movie_definition is destroyed(). Perhaps someone forgot to tell it to stop.

Benjamin Wolsey <bwy>
Project MemberIn charge of this item.
Wed 02 Mar 2011 02:29:17 PM UTC, comment #10:

I'm not so sure about this now: it seems that the illegal read is of the TagLoadersTable, which is held by boost::shared_ptr and kept alive in gprocessor by the main() function.

I don't see any obvious place where intrusive_ptrs are missing: neither SWFMovieLoader nor SWFParser use intrusive_ptr, but they are both owned exclusively by movie_definitions that pass 'this' to their respective objects. Probably they should use intrusive_ptr for future refactoring, but it's not responsible for this bug.

THe only thing that does prove to fix the bug is what I suggested in comment #8, but I don't know why.

Benjamin Wolsey <bwy>
Project MemberIn charge of this item.
Wed 02 Mar 2011 11:20:32 AM UTC, comment #9:

movie_definition is hold by intrusive_ptr, if MovieLoader needs it, it should keep it alive, no ?

Sandro Santilli <strk>
Project Member
Wed 02 Mar 2011 11:08:32 AM UTC, comment #8:

movie_root contains a MovieLoader, which handles threaded movie loading. Its threads are cleared in the dtor, which is called from movie_root's dtor.

However, in gprocessor the movie_root is an auto variable that is destroyed on exit from play_movie, while movie_definition (and MovieLibrary, which may also be relevant) are cleared beforehand.

Probably this can be fixed by ensuring the movie_root is destroyed before the movie_definition.

Benjamin Wolsey <bwy>
Project MemberIn charge of this item.
Wed 02 Mar 2011 10:18:47 AM UTC, comment #7:

I can reproduce this.

Benjamin Wolsey <bwy>
Project MemberIn charge of this item.
Wed 02 Mar 2011 10:17:51 AM UTC, comment #6:

I can reproduce this.

Benjamin Wolsey <bwy>
Project MemberIn charge of this item.
Sat 12 Feb 2011 09:07:46 PM UTC, comment #5:

Let's call it closed, altought pere had to run it in a tight loop to reproduce...

Sandro Santilli <strk>
Project Member
Sat 12 Feb 2011 08:53:38 PM UTC, comment #4:

That test doesn't fail anymore on any buildbot slaves.

Gabriele Giacone <gg0>
Project Member
Sat 18 Dec 2010 04:46:32 PM UTC, comment #3:

Short version:

==15567== Thread 3:
==15567== Invalid read of size 4
==15567== at 0x41108AB: gnash::SWF::TagLoadersTable::get(gnash::SWF::TagType, void (*&)(gnash::SWFStream&, gnash::SWF::TagType, gnash::movie_definition&, gnash::RunResources const&)) const (stl_tree.h:1002)
==15567== by 0x4280A94: gnash::SWFParser::read(int) (SWFParser.cpp:94)
==15567== by 0x4288D3D: gnash::SWFMovieDefinition::read_all_swf() (SWFMovieDefinition.cpp:487)
==15567== by 0x428944A: gnash::SWFMovieLoader::execute(gnash::SWFMovieLoader&, gnash::SWFMovieDefinition*) (SWFMovieDefinition.cpp:123)
==15567== by 0x428B111: boost::detail::thread_data<boost::_bi::bind_t<void, void ()(gnash::SWFMovieLoader&, gnash::SWFMovieDefinition), boost::_bi::list2<boost::reference_wrapper<gnash::SWFMovieLoader>, boost::_bi::value<gnash::SWFMovieDefinition*> > > >::run() (bind.hpp:313)
==15567== by 0x4D0D0D4: thread_proxy (in /usr/lib/libboost_thread.so.1.42.0)
==15567== by 0x4911954: start_thread (pthread_create.c:300)
==15567== by 0x4AE5E7D: clone (clone.S:130)
==15567== Address 0x1048 is not stack'd, malloc'd or (recently) free'd

Could this be a missing intrusive_ptr holding of movie_definition ?

Sandro Santilli <strk>
Project Member
Sat 18 Dec 2010 04:35:04 PM UTC, comment #2:

I was able to reproduce it, by running the test in a loop until it failed. Had to ask valgrind to report for childs too, to get useful information available. Here is the dump.

% while ionice -c3 nice valgrind --trace-children=yes utilities/gprocessor testsuite/swfdec/src/test/trace/movieclip-version-8.swf -r11 -f10 -v -d -1; do sleep 5; done
==15567== Memcheck, a memory error detector
==15567== Copyright (C) 2002-2010, and GNU GPL'd, by Julian Seward et al.
==15567== Using Valgrind-3.6.0.SVN-Debian and LibVEX; rerun with -h for copyright info
==15567== Command: utilities/gprocessor testsuite/swfdec/src/test/trace/movieclip-version-8.swf -r11 -f10 -v -d -1
==15567==
==15574==
==15574== HEAP SUMMARY:
==15574== in use at exit: 0 bytes in 0 blocks
==15574== total heap usage: 0 allocs, 0 frees, 0 bytes allocated
==15574==
==15574== All heap blocks were freed -- no leaks are possible
==15574==
==15574== For counts of detected and suppressed errors, rerun with: -v
==15574== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 16 from 7)
==15570==
==15570== HEAP SUMMARY:
==15570== in use at exit: 0 bytes in 0 blocks
==15570== total heap usage: 0 allocs, 0 frees, 0 bytes allocated
==15570==
==15570== All heap blocks were freed -- no leaks are possible
==15570==
==15570== For counts of detected and suppressed errors, rerun with: -v
==15570== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 16 from 7)
==15575==
==15575== HEAP SUMMARY:
==15575== in use at exit: 0 bytes in 0 blocks
==15575== total heap usage: 0 allocs, 0 frees, 0 bytes allocated
==15575==
==15575== All heap blocks were freed -- no leaks are possible
==15575==
==15575== For counts of detected and suppressed errors, rerun with: -v
==15575== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 16 from 7)
==15576==
==15576== HEAP SUMMARY:
==15576== in use at exit: 0 bytes in 0 blocks
==15576== total heap usage: 0 allocs, 0 frees, 0 bytes allocated
==15576==
==15576== All heap blocks were freed -- no leaks are possible
==15576==
==15576== For counts of detected and suppressed errors, rerun with: -v
==15576== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 16 from 7)
==15579== Memcheck, a memory error detector
==15579== Copyright (C) 2002-2010, and GNU GPL'd, by Julian Seward et al.
==15579== Using Valgrind-3.6.0.SVN-Debian and LibVEX; rerun with -h for copyright info
==15579== Command: /bin/sed -e 1s/^X// -e s%/[^/]*$%%
==15579==
==15578==
==15578== HEAP SUMMARY:
==15578== in use at exit: 0 bytes in 0 blocks
==15578== total heap usage: 0 allocs, 0 frees, 0 bytes allocated
==15578==
==15578== All heap blocks were freed -- no leaks are possible
==15578==
==15578== For counts of detected and suppressed errors, rerun with: -v
==15578== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 16 from 7)
==15579==
==15579== HEAP SUMMARY:
==15579== in use at exit: 11,947 bytes in 94 blocks
==15579== total heap usage: 137 allocs, 43 frees, 15,829 bytes allocated
==15579==
==15579== LEAK SUMMARY:
==15579== definitely lost: 2 bytes in 2 blocks
==15579== indirectly lost: 0 bytes in 0 blocks
==15579== possibly lost: 0 bytes in 0 blocks
==15579== still reachable: 11,945 bytes in 92 blocks
==15579== suppressed: 0 bytes in 0 blocks
==15579== Rerun with --leak-check=full to see details of leaked memory
==15579==
==15579== For counts of detected and suppressed errors, rerun with: -v
==15579== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 16 from 7)
==15577==
==15577== HEAP SUMMARY:
==15577== in use at exit: 0 bytes in 0 blocks
==15577== total heap usage: 0 allocs, 0 frees, 0 bytes allocated
==15577==
==15577== All heap blocks were freed -- no leaks are possible
==15577==
==15577== For counts of detected and suppressed errors, rerun with: -v
==15577== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 16 from 7)
==15582== Memcheck, a memory error detector
==15582== Copyright (C) 2002-2010, and GNU GPL'd, by Julian Seward et al.
==15582== Using Valgrind-3.6.0.SVN-Debian and LibVEX; rerun with -h for copyright info
==15582== Command: /bin/sed -n s/.*-\>\ //p
==15582==
==15581== Memcheck, a memory error detector
==15581== Copyright (C) 2002-2010, and GNU GPL'd, by Julian Seward et al.
==15581== Using Valgrind-3.6.0.SVN-Debian and LibVEX; rerun with -h for copyright info
==15581== Command: /bin/ls -ld utilities/gprocessor
==15581==
==15581==
==15581== HEAP SUMMARY:
==15581== in use at exit: 12,962 bytes in 7 blocks
==15581== total heap usage: 22 allocs, 15 frees, 17,005 bytes allocated
==15581==
==15581== LEAK SUMMARY:
==15581== definitely lost: 0 bytes in 0 blocks
==15581== indirectly lost: 0 bytes in 0 blocks
==15581== possibly lost: 0 bytes in 0 blocks
==15581== still reachable: 12,962 bytes in 7 blocks
==15581== suppressed: 0 bytes in 0 blocks
==15581== Rerun with --leak-check=full to see details of leaked memory
==15581==
==15581== For counts of detected and suppressed errors, rerun with: -v
==15581== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 24 from 7)
==15582==
==15582== HEAP SUMMARY:
==15582== in use at exit: 8,433 bytes in 38 blocks
==15582== total heap usage: 69 allocs, 31 frees, 11,061 bytes allocated
==15582==
==15582== LEAK SUMMARY:
==15582== definitely lost: 1 bytes in 1 blocks
==15582== indirectly lost: 0 bytes in 0 blocks
==15582== possibly lost: 0 bytes in 0 blocks
==15582== still reachable: 8,432 bytes in 37 blocks
==15582== suppressed: 0 bytes in 0 blocks
==15582== Rerun with --leak-check=full to see details of leaked memory
==15582==
==15582== For counts of detected and suppressed errors, rerun with: -v
==15582== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 16 from 7)
==15580==
==15580== HEAP SUMMARY:
==15580== in use at exit: 0 bytes in 0 blocks
==15580== total heap usage: 0 allocs, 0 frees, 0 bytes allocated
==15580==
==15580== All heap blocks were freed -- no leaks are possible
==15580==
==15580== For counts of detected and suppressed errors, rerun with: -v
==15580== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 16 from 7)
==15590==
==15590== HEAP SUMMARY:
==15590== in use at exit: 0 bytes in 0 blocks
==15590== total heap usage: 0 allocs, 0 frees, 0 bytes allocated
==15590==
==15590== All heap blocks were freed -- no leaks are possible
==15590==
==15590== For counts of detected and suppressed errors, rerun with: -v
==15590== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 16 from 7)
==15593== Memcheck, a memory error detector
==15593== Copyright (C) 2002-2010, and GNU GPL'd, by Julian Seward et al.
==15593== Using Valgrind-3.6.0.SVN-Debian and LibVEX; rerun with -h for copyright info
==15593== Command: /bin/sed 1q
==15593==
==15593==
==15593== HEAP SUMMARY:
==15593== in use at exit: 5,274 bytes in 8 blocks
==15593== total heap usage: 18 allocs, 10 frees, 6,562 bytes allocated
==15593==
==15593== LEAK SUMMARY:
==15593== definitely lost: 0 bytes in 0 blocks
==15593== indirectly lost: 0 bytes in 0 blocks
==15593== possibly lost: 0 bytes in 0 blocks
==15593== still reachable: 5,274 bytes in 8 blocks
==15593== suppressed: 0 bytes in 0 blocks
==15593== Rerun with --leak-check=full to see details of leaked memory
==15593==
==15593== For counts of detected and suppressed errors, rerun with: -v
==15593== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 16 from 7)
==15591==
==15591== HEAP SUMMARY:
==15591== in use at exit: 0 bytes in 0 blocks
==15591== total heap usage: 0 allocs, 0 frees, 0 bytes allocated
==15591==
==15591== All heap blocks were freed -- no leaks are possible
==15591==
==15591== For counts of detected and suppressed errors, rerun with: -v
==15591== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 16 from 7)
==15567== Memcheck, a memory error detector
==15567== Copyright (C) 2002-2010, and GNU GPL'd, by Julian Seward et al.
==15567== Using Valgrind-3.6.0.SVN-Debian and LibVEX; rerun with -h for copyright info
==15567== Command: /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/utilities/.libs/lt-gprocessor testsuite/swfdec/src/test/trace/movieclip-version-8.swf -r11 -f10 -v -d -1
==15567==
15567:1] 16:21:13 SECURITY: Checking security of URL 'file:///var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-8.swf'
15567:1] 16:21:14 SECURITY: Load of file /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-8.swf granted (under local sandbox /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/)
==15567== Conditional jump or move depends on uninitialised value(s)
==15567== at 0x4BD4A2E: inflateReset2 (in /usr/lib/libz.so.1.2.3.4)
==15567== by 0x4BD4B0C: inflateInit2_ (in /usr/lib/libz.so.1.2.3.4)
==15567== by 0x4BD4B82: inflateInit_ (in /usr/lib/libz.so.1.2.3.4)
==15567== by 0x46EAF4C: gnash::zlib_adapter::InflaterIOChannel::InflaterIOChannel(std::auto_ptr<gnash::IOChannel>) (zlib_adapter.cpp:304)
==15567== by 0x46EB14B: gnash::zlib_adapter::make_inflater(std::auto_ptr<gnash::IOChannel>) (zlib_adapter.cpp:315)
==15567== by 0x42874D4: gnash::SWFMovieDefinition::readHeader(std::auto_ptr<gnash::IOChannel>, std::string const&) (SWFMovieDefinition.cpp:311)
==15567== by 0x4161466: gnash::MovieFactory::makeMovie(std::auto_ptr<gnash::IOChannel>, std::string const&, gnash::RunResources const&, bool) (MovieFactory.cpp:247)
==15567== by 0x4161CD7: gnash::MovieFactory::makeMovie(gnash::URL const&, gnash::RunResources const&, char const, bool, std::string const) (MovieFactory.cpp:321)
==15567== by 0x8051248: play_movie(std::string const&, gnash::RunResources const&) (processor.cpp:421)
==15567== by 0x8052A55: main (processor.cpp:371)
==15567==
15567:1] 16:21:16 SECURITY: Extensions disabled
15567:2] 16:21:17 UNIMPLEMENTED: FileAttributes tag in the SWF requests that network access is not granted to this movie (or application?) when loaded from the filesystem. Anyway Gnash won't care; use white/black listing in your .gnashrc instead
15567:1] 16:21:18 TRACE: _level0
15567:1] 16:21:18 TRACE: LNX
15567:1] 16:21:18 TRACE: LNX
15567:3] 16:21:19 SECURITY: Checking security of URL 'file:///var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-7.swf'
15567:3] 16:21:19 SECURITY: Load of file /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-7.swf granted (under local sandbox /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/)
15567:1] 16:21:19 TRACE: _level8
15567:1] 16:21:19 TRACE: LNX
15567:1] 16:21:19 TRACE: LNX
15567:3] 16:21:19 SECURITY: Checking security of URL 'file:///var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-6.swf'
15567:3] 16:21:19 SECURITY: Load of file /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-6.swf granted (under local sandbox /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/)
15567:4] 16:21:19 UNIMPLEMENTED: FileAttributes tag in the SWF requests that network access is not granted to this movie (or application?) when loaded from the filesystem. Anyway Gnash won't care; use white/black listing in your .gnashrc instead
15567:1] 16:21:19 TRACE: _level0.m8
15567:1] 16:21:19 TRACE: undefined
15567:1] 16:21:19 TRACE: undefined
15567:1] 16:21:19 TRACE: undefined
==15567== Thread 3:
==15567== Invalid read of size 4
==15567== at 0x41108AB: gnash::SWF::TagLoadersTable::get(gnash::SWF::TagType, void (*&)(gnash::SWFStream&, gnash::SWF::TagType, gnash::movie_definition&, gnash::RunResources const&)) const (stl_tree.h:1002)
==15567== by 0x4280A94: gnash::SWFParser::read(int) (SWFParser.cpp:94)
==15567== by 0x4288D3D: gnash::SWFMovieDefinition::read_all_swf() (SWFMovieDefinition.cpp:487)
==15567== by 0x428944A: gnash::SWFMovieLoader::execute(gnash::SWFMovieLoader&, gnash::SWFMovieDefinition*) (SWFMovieDefinition.cpp:123)
==15567== by 0x428B111: boost::detail::thread_data<boost::_bi::bind_t<void, void ()(gnash::SWFMovieLoader&, gnash::SWFMovieDefinition), boost::_bi::list2<boost::reference_wrapper<gnash::SWFMovieLoader>, boost::_bi::value<gnash::SWFMovieDefinition*> > > >::run() (bind.hpp:313)
==15567== by 0x4D0D0D4: thread_proxy (in /usr/lib/libboost_thread.so.1.42.0)
==15567== by 0x4911954: start_thread (pthread_create.c:300)
==15567== by 0x4AE5E7D: clone (clone.S:130)
==15567== Address 0x1048 is not stack'd, malloc'd or (recently) free'd
==15567==
==15567==
==15567== Process terminating with default action of signal 11 (SIGSEGV)
==15567== Access not within mapped region at address 0x1048
==15567== at 0x41108AB: gnash::SWF::TagLoadersTable::get(gnash::SWF::TagType, void (*&)(gnash::SWFStream&, gnash::SWF::TagType, gnash::movie_definition&, gnash::RunResources const&)) const (stl_tree.h:1002)
==15567== by 0x4280A94: gnash::SWFParser::read(int) (SWFParser.cpp:94)
==15567== by 0x4288D3D: gnash::SWFMovieDefinition::read_all_swf() (SWFMovieDefinition.cpp:487)
==15567== by 0x428944A: gnash::SWFMovieLoader::execute(gnash::SWFMovieLoader&, gnash::SWFMovieDefinition*) (SWFMovieDefinition.cpp:123)
==15567== by 0x428B111: boost::detail::thread_data<boost::_bi::bind_t<void, void ()(gnash::SWFMovieLoader&, gnash::SWFMovieDefinition), boost::_bi::list2<boost::reference_wrapper<gnash::SWFMovieLoader>, boost::_bi::value<gnash::SWFMovieDefinition*> > > >::run() (bind.hpp:313)
==15567== by 0x4D0D0D4: thread_proxy (in /usr/lib/libboost_thread.so.1.42.0)
==15567== by 0x4911954: start_thread (pthread_create.c:300)
==15567== by 0x4AE5E7D: clone (clone.S:130)
==15567== If you believe this happened as a result of a stack
==15567== overflow in your program's main thread (unlikely but
==15567== possible), you can try to increase the size of the
==15567== main thread stack using the --main-stacksize= flag.
==15567== The main thread stack size used in this run was 8388608.
==15567==
==15567== HEAP SUMMARY:
==15567== in use at exit: 198,147 bytes in 382 blocks
==15567== total heap usage: 6,676 allocs, 6,294 frees, 509,253 bytes allocated
==15567==
==15567== LEAK SUMMARY:
==15567== definitely lost: 8 bytes in 1 blocks
==15567== indirectly lost: 0 bytes in 0 blocks
==15567== possibly lost: 5,267 bytes in 228 blocks
==15567== still reachable: 192,872 bytes in 153 blocks
==15567== suppressed: 0 bytes in 0 blocks
==15567== Rerun with --leak-check=full to see details of leaked memory
==15567==
==15567== For counts of detected and suppressed errors, rerun with: -v
==15567== Use --track-origins=yes to see where uninitialised values come from
==15567== ERROR SUMMARY: 4 errors from 2 contexts (suppressed: 184 from 7)
Killed
%

Petter Reinholdtsen <pere>
Project Member
Wed 15 Dec 2010 07:53:01 AM UTC, comment #1:

Could you try using:

utilities/gprocessor -r11 -f10 -vv testsuite/swfdec/src/test/trace/movieclip-version-6.swf

Would like to know if it's on exit or not.
Also, a valgrind report would be helpful.

Sandro Santilli <strk>
Project Member
Mon 13 Dec 2010 10:28:33 PM UTC, original submission:

The test suite running on my squeeze buildbot slave some times fail like this:

FAIL: run_swfdec_testsuite_m: movieclip-version-6.swf:40e09c3d4c0ed3f7763d30c05725ef1a crashed gnash (stdout in movieclip-version-6.swf.out, stderr in movieclip-version-6.swf.err)

The failing build log is available via
http://gnashdev.org:8010/builders/squeeze-linux-x86/builds/59/steps/make_anal/logs/stdio . Note that the in previous build the test suite had no failures.

This is the content of the .out file:

23676:1] 21:37:23 SECURITY: Checking security of URL 'file:///var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-6.swf'
23676:1] 21:37:23 SECURITY: Load of file /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-6.swf granted (under local sandbox /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/)
23676:1] 21:37:23 SECURITY: Extensions disabled
23676:2] 21:37:23 UNIMPLEMENTED: FileAttributes tag in the SWF requests that network access is not granted to this movie (or application?) when loaded from the filesystem. Anyway Gnash won't care; use white/black listing in your .gnashrc instead
23676:1] 21:37:23 TRACE: _level0
23676:1] 21:37:23 TRACE: WIN
23676:1] 21:37:23 TRACE: WIN
23676:3] 21:37:23 SECURITY: Checking security of URL 'file:///var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-8.swf'
23676:3] 21:37:23 SECURITY: Load of file /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-8.swf granted (under local sandbox /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/)
23676:3] 21:37:23 SECURITY: Checking security of URL 'file:///var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-7.swf'
23676:3] 21:37:23 SECURITY: Load of file /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-7.swf granted (under local sandbox /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/)
23676:4] 21:37:23 UNIMPLEMENTED: FileAttributes tag in the SWF requests that network access is not granted to this movie (or application?) when loaded from the filesystem. Anyway Gnash won't care; use white/black listing in your .gnashrc instead
23676:3] 21:37:23 SECURITY: Checking security of URL 'file:///var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-6.swf'
23676:3] 21:37:23 SECURITY: Load of file /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-6.swf granted (under local sandbox /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/)
23676:5] 21:37:23 UNIMPLEMENTED: FileAttributes tag in the SWF requests that network access is not granted to this movie (or application?) when loaded from the filesystem. Anyway Gnash won't care; use white/black listing in your .gnashrc instead
23676:3] 21:37:23 SECURITY: Checking security of URL 'file:///var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-5.swf'
23676:3] 21:37:23 SECURITY: Load of file /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-5.swf granted (under local sandbox /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/)
23676:6] 21:37:23 UNIMPLEMENTED: FileAttributes tag in the SWF requests that network access is not granted to this movie (or application?) when loaded from the filesystem. Anyway Gnash won't care; use white/black listing in your .gnashrc instead
23676:7] 21:37:23 UNIMPLEMENTED: FileAttributes tag in the SWF requests that network access is not granted to this movie (or application?) when loaded from the filesystem. Anyway Gnash won't care; use white/black listing in your .gnashrc instead
23676:1] 21:37:23 TRACE: _level8
23676:1] 21:37:23 TRACE: WIN
23676:1] 21:37:23 TRACE: WIN
23676:1] 21:37:23 TRACE: _level0.m8
23676:1] 21:37:23 TRACE: undefined
23676:1] 21:37:23 TRACE: undefined
23676:1] 21:37:23 TRACE: undefined
23676:1] 21:37:23 TRACE: _level7
23676:1] 21:37:23 TRACE: WIN
23676:1] 21:37:23 TRACE: WIN
23676:1] 21:37:23 TRACE: _level0.m7
23676:1] 21:37:23 TRACE: undefined
23676:1] 21:37:23 TRACE: undefined
23676:1] 21:37:23 TRACE: undefined
23676:1] 21:37:23 TRACE: _level6
23676:1] 21:37:23 TRACE: WIN
23676:1] 21:37:23 TRACE: WIN
23676:1] 21:37:23 TRACE: _level0.m6
23676:1] 21:37:23 TRACE: undefined
23676:1] 21:37:23 TRACE: undefined
23676:1] 21:37:23 TRACE: undefined
23676:1] 21:37:23 TRACE: _level5
23676:1] 21:37:23 TRACE: WIN
23676:1] 21:37:23 TRACE: WIN
23676:1] 21:37:23 TRACE: _level0.m5
23676:1] 21:37:23 TRACE: undefined
23676:1] 21:37:23 TRACE: undefined
23676:1] 21:37:23 TRACE: undefined
23676:3] 21:37:23 SECURITY: Checking security of URL 'file:///var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-8.swf'
23676:3] 21:37:23 SECURITY: Load of file /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/movieclip-version-8.swf granted (under local sandbox /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/swfdec/src/test/trace/)

And this is the content of the .err file:

Warning: missing value for variable "Renderer" in rcfile /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/gnashrc, line 51
Warning: missing value for variable "MediaHandler" in rcfile /var/lib/buildbot/gnash/build/squeeze-linux-x86_32/build/testsuite/gnashrc, line 54
Segmentation fault

Petter Reinholdtsen <pere>
Project Member

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach File(s):
   
   
Comment:
   

No files currently attached

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -unavailable- added by bwy (Posted a comment)
  • -unavailable- added by gg0 (Posted a comment)
  • -unavailable- added by strk (Posted a comment)
  • -unavailable- added by pere (Submitted the item)
  •  

    Do you think this task is very important?
    If so, you can click here to add your encouragement to it.
    This task has 0 encouragements so far.

    Only logged-in users can vote.

     

    Please enter the title of George Orwell's famous dystopian book (it's a date):

     

     

    Follow 13 latest changes.

    Date Changed By Updated Field Previous Value => Replaced By
    Sat 05 Mar 2011 12:36:27 PM UTCbwyAssigned toNone=>bwy
    Sat 05 Mar 2011 12:10:53 PM UTCbwyStatusReady For Test=>Fixed
      Open/ClosedOpen=>Closed
    Sat 05 Mar 2011 08:35:17 AM UTCbwySummaryTes run_swfdec_testsuite_m some times fail with a segfault=>Test run_swfdec_testsuite_m some times fail with a segfault
    Sat 05 Mar 2011 08:35:11 AM UTCbwyStatusConfirmed=>Ready For Test
    Wed 02 Mar 2011 10:18:47 AM UTCbwyOpen/ClosedClosed=>Open
    Wed 02 Mar 2011 10:17:51 AM UTCbwyStatusReady For Test=>Confirmed
      Open/ClosedOpen=>Closed
    Sat 12 Feb 2011 09:13:16 PM UTCstrkStatusFixed=>Ready For Test
      Open/ClosedClosed=>Open
    Sat 12 Feb 2011 09:07:46 PM UTCstrkStatusNone=>Fixed
      Open/ClosedOpen=>Closed
    Wed 15 Dec 2010 07:53:01 AM UTCstrkSeverity3 - Normal=>6 - Security

    Back to the top


    Powered by Savane 3.1-cleanup1