bugGnash - The GNU Flash player - Bugs: bug #20076, Memory leak at end of...

 
 

bug #20076: Memory leak at end of "movie"

Submitter:  Sebastiá Matas <annonygmouse>
Submitted:  Mon 04 Jun 2007 10:42:43 PM UTC
   
 
Category:  core Severity:  5 - Blocker
Release:  None Status:  Fixed
Privacy:  Public Assigned to:  strk
Open/Closed:  Closed
* Mandatory Fields

Add a New Comment Rich Markup
   

Jump to the original submission

Sun 18 May 2008 09:42:52 AM UTC, comment #13: 

Thank you for good bug reports.

Sandro Santilli <strk>
Group Member
Sun 18 May 2008 09:41:55 AM UTC, comment #12: 


Sorry, my fault.

I was using a gnash executable from February!!!!

I've cleaned up everything, updated working copy, compiled,
installed and now it behaves correctly.

This bug can be closed.

Thanks a lot for your time and patiente. GREAT WORK!

Sebastiá Matas <annonygmouse>
Group Member
Sat 17 May 2008 06:35:52 AM UTC, comment #11: 

An update: I did unset GLIBCPP_FORCE_NEW and still don't get leaks. Tried with ffmpeg handler, and still don't get leaks.

What you can do:

1) make really sure you're using an updated version of head
2) check with valgrind --leak-check=full to see if there's any definitely lost record
3) if no hint on 2 use massif tool of valgrind

Sandro Santilli <strk>
Group Member
Sat 17 May 2008 06:32:07 AM UTC, comment #10: 

Here are my results, ps aux on the 'Game Over' screen:

USER       PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME COMMAND
strk      9358  9.1  1.2  75764 19168 pts/4    R+   08:26   0:05 /home/strk/src/gnash/build/head/allgui_agg_gst/gui/.libs/lt-gtk-gnash -v /home/strk/src/gnash/tests/swf5/curveball.swf
strk      9358 15.3  1.2  75764 19168 pts/4    R+   08:26   0:23 /home/strk/src/gnash/build/head/allgui_agg_gst/gui/.libs/lt-gtk-gnash -v /home/strk/src/gnash/tests/swf5/curveball.swf
strk      9358 17.0  1.2  75764 19168 pts/4    R+   08:26   0:45 /home/strk/src/gnash/build/head/allgui_agg_gst/gui/.libs/lt-gtk-gnash -v /home/strk/src/gnash/tests/swf5/curveball.swf

Note that I have GLIBCPP_FORCE_NEW=1 in environment, which
disables caches in the gnu STL.
Also, I'm using the gstreamer media handler.

I could reproduce before patching it, btw..

Sandro Santilli <strk>
Group Member
Fri 16 May 2008 08:21:13 PM UTC, comment #9: 


Sorry, bug still there.
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dv mai 16 22:16:03 CEST 2008
jo        8393  2.5  2.1 288008 21764 pts/0    S+   22:15   0:00 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo        8409  0.0  0.0    320    68 pts/5    R+   22:16   0:00 grep gnash
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dv mai 16 22:17:18 CEST 2008
jo        8393  5.5  2.5 370220 25604 pts/0    SLl+ 22:15   0:05 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo        8456  0.0  0.0   8600   800 pts/5    R+   22:17   0:00 grep gnash
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dv mai 16 22:17:51 CEST 2008
jo        8393  7.3  3.5 381080 36484 pts/0    RLl+ 22:15   0:09 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo        8467  0.0  0.0   8600   804 pts/5    R+   22:17   0:00 grep gnash
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dv mai 16 22:18:04 CEST 2008
jo        8393 12.0  7.1 417112 72516 pts/0    SLl+ 22:15   0:16 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo        8472  0.0  0.0   8600   808 pts/5    R+   22:18   0:00 grep gnash
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dv mai 16 22:18:14 CEST 2008
jo        8393 18.1  9.8 444356 99756 pts/0    RLl+ 22:15   0:26 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dv mai 16 22:18:35 CEST 2008
jo        8393 28.0 13.3 480268 135672 pts/0   RLl+ 22:15   0:47 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo        8511  0.0  0.0   8600   796 pts/5    R+   22:18   0:00 grep gnash
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dv mai 16 22:18:43 CEST 2008
jo        8393 31.3 14.4 491680 147084 pts/0   RLl+ 22:15   0:55 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo        8519  0.0  0.0   8600   800 pts/5    R+   22:18   0:00 grep gnash
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dv mai 16 22:18:50 CEST 2008
jo        8393 34.1 15.4 501420 156824 pts/0   RLl+ 22:15   1:02 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf


Log
jo@debianmbp:~/gnash$ gnash ../swftest/curveball-game.swf
8393] 22:15:48: Timer delay set to 50 milliseconds
8393] 22:15:48: DEBUG: gnash::media::sound_handler* gnash::media::create_sound_handler_gst() enter
8393] 22:15:49: Checking support for MIT-SHM...
8393] 22:15:49: NOTICE: MIT-SHM available (version 1.1)!
8393] 22:15:49: Your X server expects RGB24 pixmap data for standard mode.
8393] 22:15:50: Created top level window
8393] 22:15:51: X server pixel format is (R16:8, G8:8, B0:8, 32 bpp)
8393] 22:15:51: X server is using BGRA32 pixel format
8393] 22:15:51: framebuffer pixel format is BGRA32 (little-endian host)
8393] 22:15:51: DEBUG: Base url set to: file:///home/jo/swftest/curveball-game.swf
8393] 22:15:51: DEBUG: /home/jo/swftest/ appended to local sandboxes
8393] 22:15:51: DEBUG: /home/jo/swftest/curveball-game.swf appended to local sandboxes
8393] 22:15:51: SECURITY: Load of file /home/jo/swftest/curveball-game.swf granted (under local sandbox /home/jo/swftest/).
8393] 22:15:51: Movie file:///home/jo/swftest/curveball-game.swf (SWF5) added to library
8393] 22:15:51: GTK-AGG: Using shared memory image
8393] 22:15:51: initialized AGG buffer <0x2b5e7bb8e000>, 350000 bytes, 350x250, rowsize is 1400 bytes
8393] 22:15:51: UNIMPLEMENTED: MP3 delay seek 1661
8393] 22:15:52: UNIMPLEMENTED: MP3 delay seek 1661
8393] 22:15:52: UNIMPLEMENTED: MP3 delay seek 1661
8393] 22:15:52: UNIMPLEMENTED: MP3 delay seek 1661
8393] 22:15:52: UNIMPLEMENTED: MP3 delay seek 1661
8393] 22:15:52: UNIMPLEMENTED: Different stream/playback sound rate (5512/11025). This seems common in SWF files, so we'll warn only once.
8393] 22:15:52: UNIMPLEMENTED: Different stream/playback sample size (32/16). This seems common in SWF files, so we'll warn only once.
8393] 22:15:52: ERROR: No samples advertised for sound stream
8393] 22:15:52: ERROR: No samples advertised for sound stream
8393] 22:15:52: ERROR: No samples advertised for sound stream
8393] 22:15:52: DEBUG: Sprite _level0.bounds (depth -16383) won't check for user-defined LOAD event (is not dynamic, has a parent, no registered class and no clip events defined)
8393] 22:15:52: DEBUG: Global instance list grew to 2 entries
8393] 22:15:52: ERROR: No samples advertised for sound stream
8393] 22:15:52: ERROR: No samples advertised for sound stream
8393] 22:15:52: ERROR: No samples advertised for sound stream
8393] 22:15:52: ERROR: No samples advertised for sound stream
8393] 22:15:52: ERROR: No samples advertised for sound stream
8393] 22:15:52: ERROR: No samples advertised for sound stream
8393] 22:15:52: ERROR: No samples advertised for sound stream
8393] 22:15:52: GTK-AGG: Using shared memory image
8393] 22:15:52: initialized AGG buffer <0x2aaab17f1000>, 6680 bytes, 334x5, rowsize is 1336 bytes
8393] 22:15:52: GTK-AGG: Using shared memory image
8393] 22:15:52: initialized AGG buffer <0x2aaab17f3000>, 350000 bytes, 350x250, rowsize is 1400 bytes
8393] 22:16:10: DEBUG: Global instance list grew to 3 entries
8393] 22:16:12: DEBUG: Sprite _level0.bonus (depth -16362) won't check for user-defined LOAD event (is not dynamic, has a parent, no registered class and no clip events defined)
8393] 22:16:12: DEBUG: Sprite _level0.enemyLives (depth -16357) won't check for user-defined LOAD event (is not dynamic, has a parent, no registered class and no clip events defined)
8393] 22:16:12: DEBUG: Sprite _level0.playerLives (depth -16351) won't check for user-defined LOAD event (is not dynamic, has a parent, no registered class and no clip events defined)
8393] 22:16:12: DEBUG: Global instance list grew to 6 entries
8393] 22:16:12: DEBUG: Global instance list grew to 7 entries
8393] 22:16:12: DEBUG: Global instance list grew to 9 entries
8393] 22:16:23: DEBUG: Target of action_buffer (_level0.instance40 of type gnash::sprite_instance) unloaded by execution of opcode:
 PC:391 - EX: SetTarget ("")
Discarding 1 bytes of remaining opcodes:
 PC:395 - EX: <End>

8393] 22:16:30: DEBUG: Target of action_buffer (_level0.instance80 of type gnash::sprite_instance) unloaded by execution of opcode:
 PC:391 - EX: SetTarget ("")
Discarding 1 bytes of remaining opcodes:
 PC:395 - EX: <End>

8393] 22:17:28: DEBUG: Target of action_buffer (_level0.instance121 of type gnash::sprite_instance) unloaded by execution of opcode:
 PC:391 - EX: SetTarget ("")
Discarding 1 bytes of remaining opcodes:
 PC:395 - EX: <End>

8393] 22:17:37: DEBUG: Target of action_buffer (_level0.instance162 of type gnash::sprite_instance) unloaded by execution of opcode:
 PC:391 - EX: SetTarget ("")
Discarding 1 bytes of remaining opcodes:
 PC:395 - EX: <End>

8393] 22:17:42: DEBUG: Target of action_buffer (_level0.instance203 of type gnash::sprite_instance) unloaded by execution of opcode:
 PC:391 - EX: SetTarget ("")
Discarding 1 bytes of remaining opcodes:
 PC:395 - EX: <End>

8393] 22:17:48: DEBUG: Target of action_buffer (_level0.instance244 of type gnash::sprite_instance) unloaded by execution of opcode:
 PC:365 - EX: SetTarget ("")
Discarding 27 bytes of remaining opcodes:
 PC:369 - EX: ActionBranchAlways ( 21)
 PC:374 - EX: ActionPushData (dict_lookup[0])
 PC:379 - EX: ActionGetVariable
 PC:380 - EX: ActionSetTargetExpression
 PC:381 - EX: GotoLabel ("Serve")
 PC:390 - EX: Play
 PC:391 - EX: SetTarget ("")
 PC:395 - EX: <End>

8393] 22:17:48: get url: target=, url=file:///home/jo/swftest/checkscore.php?Score=350&Level=1 (checkscore.php?Score=350&Level=1), method=c0 (sendVars:0, loadTarget:1, loadVariable:1)
8393] 22:17:48: SECURITY: Load of file /home/jo/swftest/checkscore.php granted (under local sandbox /home/jo/swftest/).
8393] 22:17:48: getURL2 loadVariable
8393] 22:17:48: ERROR: get url: target  not found
8393] 22:18:55: DEBUG: static gboolean gnash::GtkGui::delete_event(GtkWidget*, GdkEvent*, void*) enter
Main loop ended, cleaning up
Any segfault past this message is likely due to improper threads cleanup.
jo@debianmbp:~/gnash$                                                          

Is there something I could do to help profiling?


Sebastiá Matas <annonygmouse>
Group Member
Fri 16 May 2008 10:21:44 AM UTC, comment #8: 

Should be fixed now, please test.

Sandro Santilli <strk>
Group Member
Fri 16 May 2008 08:34:36 AM UTC, comment #7: 

So, this is not a proper leak, in that memory isn't lost.
massif reports the peak being in instantiation of textfields
during goto_frame actions.

I suspect the problem being unloaded textfield being
marked as reachable anyway from the target of their
registered variable.

Sandro Santilli <strk>
Group Member
Thu 15 May 2008 06:29:48 PM UTC, comment #6: 

Yes, confirmed, even with -r1 (no sound).

Sandro Santilli <strk>
Group Member
Thu 15 May 2008 04:32:41 PM UTC, comment #5: 


Sorry strk, the bug is still there.
It seems the "leak while moving pointer" is solved, but when
the game over text appears the big leak is still there... (and
eats all the CPU it can get).

Here's the test:
Just run, waiting to press "start game" button.
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dj mai 15 18:20:09 CEST 2008
jo       13448  3.2  2.1 288444 22260 pts/1    S+   18:20   0:00 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo       13455  0.0  0.0   8604   816 pts/6    S+   18:20   0:00 grep gnash

I loose one ball
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dj mai 15 18:20:37 CEST 2008
jo       13448  3.2  2.4 353500 25152 pts/1    SLl+ 18:20   0:01 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo       13479  0.0  0.0   8604   804 pts/6    R+   18:20   0:00 grep gnash

I loose one ball
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dj mai 15 18:20:48 CEST 2008
jo       13448  4.5  2.5 353896 25472 pts/1    SLl+ 18:20   0:02 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo       13484  0.0  0.0   8604   808 pts/6    R+   18:20   0:00 grep gnash

I loose one ball
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dj mai 15 18:21:00 CEST 2008
jo       13448  5.5  2.5 354176 25652 pts/1    SLl+ 18:20   0:03 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo       13491  0.0  0.0   8604   804 pts/6    R+   18:21   0:00 grep gnash

I loose one ball
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dj mai 15 18:21:10 CEST 2008
jo       13448  6.1  2.5 354340 25816 pts/1    SLl+ 18:20   0:04 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo       13496  0.0  0.0   8604   804 pts/6    R+   18:21   0:00 grep gnash

I loose one ball
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dj mai 15 18:21:22 CEST 2008
jo       13448  6.9  3.0 359604 31100 pts/1    SLl+ 18:20   0:05 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo       13501  0.0  0.0   8604   800 pts/6    R+   18:21   0:00 grep gnash

After game over and letting the game over on screen (BTW, if I remember right, with prior versions I could
click and start again... but now I click and nothing happens...)
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dj mai 15 18:21:47 CEST 2008
jo       13448 22.7  9.9 428992 100484 pts/1   RLl+ 18:20   0:24 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo       13509  0.0  0.0   8604   812 pts/6    S+   18:21   0:00 grep gnash
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dj mai 15 18:22:12 CEST 2008
jo       13448 37.6 14.1 472360 143852 pts/1   RLl+ 18:20   0:49 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo       13522  0.0  0.0   8604   796 pts/6    R+   18:22   0:00 grep gnash
jo@debianmbp:/usr/local/jboss-3.2.8-caib5/server/default/deploycaib$ date ; ps aux|grep gnash
dj mai 15 18:22:25 CEST 2008
jo       13448 43.3 15.8 489564 161060 pts/1   RLl+ 18:20   1:02 /usr/local/bin/gtk-gnash ../swftest/curveball-game.swf
jo       13532  0.0  0.0   8604   796 pts/6    R+   18:22   0:00 grep gnash


Here's the log:
jo@debianmbp:~/gnash$ gnash ../swftest/curveball-game.swf
13448] 18:20:02: Timer delay set to 50 milliseconds
13448] 18:20:02: DEBUG: gnash::media::sound_handler* gnash::media::create_sound_handler_gst() enter
13448] 18:20:02: Checking support for MIT-SHM...
13448] 18:20:02: NOTICE: MIT-SHM available (version 1.1)!
13448] 18:20:02: Your X server expects RGB24 pixmap data for standard mode.
13448] 18:20:02: Created top level window
13448] 18:20:02: X server pixel format is (R16:8, G8:8, B0:8, 32 bpp)
13448] 18:20:02: X server is using BGRA32 pixel format
13448] 18:20:02: framebuffer pixel format is BGRA32 (little-endian host)
13448] 18:20:02: DEBUG: Base url set to: file:///home/jo/swftest/curveball-game.swf
13448] 18:20:02: DEBUG: /home/jo/swftest/ appended to local sandboxes
13448] 18:20:02: DEBUG: /home/jo/swftest/curveball-game.swf appended to local sandboxes
13448] 18:20:02: SECURITY: Load of file /home/jo/swftest/curveball-game.swf granted (under local sandbox /home/jo/swftest/).
13448] 18:20:02: Movie file:///home/jo/swftest/curveball-game.swf (SWF5) added to library
13448] 18:20:02: GTK-AGG: Using shared memory image
13448] 18:20:02: initialized AGG buffer <0x2b5b900ca000>, 350000 bytes, 350x250, rowsize is 1400 bytes
13448] 18:20:02: UNIMPLEMENTED: MP3 delay seek 1661
13448] 18:20:02: UNIMPLEMENTED: MP3 delay seek 1661
13448] 18:20:02: UNIMPLEMENTED: MP3 delay seek 1661
13448] 18:20:02: UNIMPLEMENTED: MP3 delay seek 1661
13448] 18:20:02: UNIMPLEMENTED: MP3 delay seek 1661
13448] 18:20:02: UNIMPLEMENTED: Different stream/playback sound rate (5512/11025). This seems common in SWF files, so we'll warn only once.
13448] 18:20:02: UNIMPLEMENTED: Different stream/playback sample size (32/16). This seems common in SWF files, so we'll warn only once.
13448] 18:20:02: ERROR: No samples advertised for sound stream
13448] 18:20:02: ERROR: No samples advertised for sound stream
13448] 18:20:02: ERROR: No samples advertised for sound stream
13448] 18:20:02: DEBUG: Sprite _level0.bounds (depth -16383) won't check for user-defined LOAD event (is not dynamic, has a parent, no registered class and no clip events defined)
13448] 18:20:02: DEBUG: Global instance list grew to 2 entries
13448] 18:20:02: ERROR: No samples advertised for sound stream
13448] 18:20:02: ERROR: No samples advertised for sound stream
13448] 18:20:02: ERROR: No samples advertised for sound stream
13448] 18:20:02: ERROR: No samples advertised for sound stream
13448] 18:20:02: GTK-AGG: Using shared memory image
13448] 18:20:02: initialized AGG buffer <0x2aaab17f1000>, 6680 bytes, 334x5, rowsize is 1336 bytes
13448] 18:20:02: ERROR: No samples advertised for sound stream
13448] 18:20:02: ERROR: No samples advertised for sound stream
13448] 18:20:02: GTK-AGG: Using shared memory image
13448] 18:20:02: initialized AGG buffer <0x2aaab17f3000>, 350000 bytes, 350x250, rowsize is 1400 bytes
13448] 18:20:02: ERROR: No samples advertised for sound stream
13448] 18:20:28: DEBUG: Global instance list grew to 3 entries
13448] 18:20:30: DEBUG: Sprite _level0.bonus (depth -16362) won't check for user-defined LOAD event (is not dynamic, has a parent, no registered class and no clip events defined)
13448] 18:20:30: DEBUG: Sprite _level0.enemyLives (depth -16357) won't check for user-defined LOAD event (is not dynamic, has a parent, no registered class and no clip events defined)
13448] 18:20:30: DEBUG: Sprite _level0.playerLives (depth -16351) won't check for user-defined LOAD event (is not dynamic, has a parent, no registered class and no clip events defined)
13448] 18:20:30: DEBUG: Global instance list grew to 6 entries
13448] 18:20:30: DEBUG: Global instance list grew to 7 entries
13448] 18:20:30: DEBUG: Global instance list grew to 9 entries
13448] 18:20:35: DEBUG: Target of action_buffer (_level0.instance40 of type gnash::sprite_instance) unloaded by execution of opcode:
 PC:391 - EX: SetTarget ("")
Discarding 1 bytes of remaining opcodes:
 PC:395 - EX: <End>

13448] 18:20:47: DEBUG: Target of action_buffer (_level0.instance80 of type gnash::sprite_instance) unloaded by execution of opcode:
 PC:391 - EX: SetTarget ("")
Discarding 1 bytes of remaining opcodes:
 PC:395 - EX: <End>

13448] 18:20:56: DEBUG: Target of action_buffer (_level0.instance120 of type gnash::sprite_instance) unloaded by execution of opcode:
 PC:391 - EX: SetTarget ("")
Discarding 1 bytes of remaining opcodes:
 PC:395 - EX: <End>

13448] 18:21:08: DEBUG: Target of action_buffer (_level0.instance160 of type gnash::sprite_instance) unloaded by execution of opcode:
 PC:391 - EX: SetTarget ("")
Discarding 1 bytes of remaining opcodes:
 PC:395 - EX: <End>

13448] 18:21:20: DEBUG: Target of action_buffer (_level0.instance200 of type gnash::sprite_instance) unloaded by execution of opcode:
 PC:365 - EX: SetTarget ("")
Discarding 27 bytes of remaining opcodes:
 PC:369 - EX: ActionBranchAlways ( 21)
 PC:374 - EX: ActionPushData (dict_lookup[0])
 PC:379 - EX: ActionGetVariable
 PC:380 - EX: ActionSetTargetExpression
 PC:381 - EX: GotoLabel ("Serve")
 PC:390 - EX: Play
 PC:391 - EX: SetTarget ("")
 PC:395 - EX: <End>

13448] 18:21:20: get url: target=, url=file:///home/jo/swftest/checkscore.php?Score=0&Level=1 (checkscore.php?Score=0&Level=1), method=c0 (sendVars:0, loadTarget:1, loadVariable:1)
13448] 18:21:20: SECURITY: Load of file /home/jo/swftest/checkscore.php granted (under local sandbox /home/jo/swftest/).
13448] 18:21:20: getURL2 loadVariable
13448] 18:21:20: ERROR: get url: target  not found
13448] 18:22:32: DEBUG: static gboolean gnash::GtkGui::delete_event(GtkWidget*, GdkEvent*, void*) enter
Main loop ended, cleaning up
Any segfault past this message is likely due to improper threads cleanup.

Sebastiá Matas <annonygmouse>
Group Member
Thu 15 May 2008 11:15:06 AM UTC, comment #4: 

seems to be fixed in head, can you confirm sebastia' ?

Sandro Santilli <strk>
Group Member
Tue 03 Jul 2007 01:21:33 AM UTC, comment #3: 

Tried again with current head, with garbage collection enabled by default.
Right now, with no user intervention, there's no leak in the initial screen.

Anyway, while moving the mouse the memory keeps increasing !

Sandro Santilli <strk>
Group Member
Sat 16 Jun 2007 10:57:39 AM UTC, comment #2: 

I verified leaks to exist even while being still in the initial frame. Just let it run. It's a small leak, but still there.

Sandro Santilli <strk>
Group Member
Mon 04 Jun 2007 10:44:11 PM UTC, comment #1: 

Sorry, forgot to submit my specs:

Debian unstable (updated nearly daily)
amd64 (Intel core2duo)

Sebastià

Sebastiá Matas <annonygmouse>
Group Member
Mon 04 Jun 2007 10:42:43 PM UTC, original submission:  


Hi, I was trying the game curveball-game.swf [1]
and it works GREAT! but... at the "game over" part, if
you let it open and don't start a new game (using "restart movie"),
it keeps the "game over" frame visible and starts eating all
the memory it can.

Here's the prove:

Running, before starting to play

jo@debian-mbp:~$ date ; ps aux|grep gnash
dt jun  5 00:41:26 CEST 2007
jo       25472  3.7  2.3 245888 23732 pts/4    S+   00:41   0:00 gnash ../swftest/curveball-game.swf
jo       25510  0.0  0.0   5068   752 pts/1    R+   00:41   0:00 grep gnash


jo@debian-mbp:~$ date ; ps aux|grep gnash
dt jun  5 00:41:59 CEST 2007
jo       25472  0.8  2.3 245888 23732 pts/4    S+   00:41   0:00 gnash ../swftest/curveball-game.swf
jo       25621  0.0  0.0   5072   752 pts/1    R+   00:41   0:00 grep gnash


No leak so far.

Playing a little bit...
jo@debian-mbp:~$ date ; ps aux|grep gnash
dt jun  5 00:42:35 CEST 2007
jo       25472  7.1  2.5 254796 26108 pts/4    RLl+ 00:41   0:05 gnash ../swftest/curveball-game.swf
jo       25741  0.0  0.0   5068   756 pts/1    R+   00:42   0:00 grep gnash

still more playing

jo@debian-mbp:~$ date ; ps aux|grep gnash
dt jun  5 00:43:40 CEST 2007
jo       25472 18.1  2.8 255192 28888 pts/4    SLl+ 00:41   0:26 gnash ../swftest/curveball-game.swf
jo       25960  0.0  0.0   5072   756 pts/1    R+   00:43   0:00 grep gnash

just after "game over", here starts eating memory.
jo@debian-mbp:~$ date ; ps aux|grep gnash
dt jun  5 00:44:01 CEST 2007
jo       25472 19.5  4.3 266804 43852 pts/4    SLl+ 00:41   0:32 gnash ../swftest/curveball-game.swf
jo       26030  0.0  0.0   5072   760 pts/1    S+   00:44   0:00 grep gnash


jo@debian-mbp:~$ date ; ps aux|grep gnash
dt jun  5 00:44:27 CEST 2007
jo       25472 20.9 18.3 409028 186068 pts/4   SLl+ 00:41   0:39 gnash ../swftest/curveball-game.swf
jo       26119  0.0  0.0   5068   760 pts/1    S+   00:44   0:00 grep gnash

jo@debian-mbp:~$ date ; ps aux|grep gnash
dt jun  5 00:45:12 CEST 2007
jo       25472 22.0 43.4 662956 440004 pts/4   SLl+ 00:41   0:52 gnash ../swftest/curveball-game.swf
jo       26274  0.0  0.0   5068   756 pts/1    S+   00:45   0:00 grep gnash

This is on a CVS compiled version from this afternoon and using "agg" (opengl is too
slow for my machine as I don't have opengl acceleration, dammm ati).

Hope it helps.

[1] http://www.fetchfido.co.uk/games/curve_ball/curve_ball.htm

Kind regards
Sebastià

Sebastiá Matas <annonygmouse>
Group Member

 

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

Attach Files:
   
   
Comment:
   

No files currently attached

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -email is unavailable- added by strk (Posted a comment)
  • -email is unavailable- added by annonygmouse (Submitted the item)
  •  

    There are 0 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.

    Only logged-in users can vote.

     

    Follow 12 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2008-07-09 strk Attached File#16063 Removed
    2008-07-09 None Attached File- Added get_apache_80x15_1.png, #16063
    2008-05-18 strk StatusReady For Test Fixed
        Open/ClosedOpen Closed
    2008-05-16 strk StatusIn Progress Ready For Test
    2008-05-16 strk CategoryNone core
        Severity3 - Normal 5 - Blocker
        StatusConfirmed In Progress
        Assigned toNone strk
    2008-05-15 strk StatusReady For Test Confirmed
    2008-05-15 strk StatusConfirmed Ready For Test
    2007-06-16 strk StatusNone Confirmed

    Back to the top

    Powered by Savane 3.13-4b48.
    Corresponding source code