[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Gnash-dev] --enable-jemalloc Problem

From: Udo Giacomozzi
Subject: [Gnash-dev] --enable-jemalloc Problem
Date: Wed, 7 May 2008 19:27:54 +0200

Rob, this is for you:   ;-)

I compiled Gnash on Debian 4 standard with these options:

./configure --with-extensions=fileio --disable-debugger
--enable-renderer=agg --enable-gui=fb --disable-cygnal
--with-pixelformat=BGRA32,BGR24 --enable-fps-debug --disable-klash
--enable-media=GST --disable-static --with-gcc-arch=i486
--enable-jemalloc CXX=ccache g++

fb-gnash aborts right at the start:


Starting program: /mnt/userpart/home/.gnash-2008-05-07/fb-gnash -v 
[Thread debugging using libthread_db enabled]
RcInitFile: couldn't open file: /usr/local/etc/gnashrc
RcInitFile: parsing /mnt/userpart/userhome/.gnashrc
[New Thread 0xb70e6920 (LWP 2828)]
2828:3071174944] 19:06:43 SECURITY: Checking security of URL 
2828:3071174944] 19:06:43 SECURITY: Load of file 
/mnt/userpart/home/.gnash-2008-01-23/test/gravity.swf granted (under local 
sandbox /mnt/userpart/home/.gnash-2008-01-23/test/)
2828:3071174944] 19:06:43 SECURITY: Extensions enabled, scanning plugin dir for 
2828:3071174944] 19:06:43 SECURITY: Loading module: fileio
2828:3071174944] 19:06:43 SECURITY: Initializing module: "fileio"
[New Thread 0xb6b7eb90 (LWP 2832)]
[Thread 0xb6b7eb90 (LWP 2832) exited]
Program received signal SIGABRT, Aborted.
[Switching to Thread 0xb70e6920 (LWP 2828)]
0xb7852ea6 in raise () from /lib/
(gdb) bt
#0  0xb7852ea6 in raise () from /lib/
#1  0xb78547b1 in abort () from /lib/
#2  0xb7ba6a95 in gnash::Property::setReachable (this=0x8190a60) at 
#3  0xb7babc1a in gnash::PropertyList::setReachable (this=0x81911c8) at 
#4  0xb7bc4512 in gnash::as_object::markAsObjectReachable (this=0x81911c0) at 
#5  0xb7bde778 in gnash::character::markCharacterReachable (this=0x81911c0) at 
#6  0xb7c1c2ce in gnash::sprite_instance::markReachableResources 
(this=0x81911c0) at sprite_instance.cpp:4697
#7  0xb7c1c1bd in gnash::sprite_instance::markReachableResources 
(this=0x8184180) at ../libbase/GC.h:126
#8  0xb7c065da in gnash::movie_root::markReachableResources (this=0x8122900) at 
#9  0xb7dc26d1 in gnash::VM::markReachableResources (this=0x816b060) at 
#10 0xb7c918d7 in gnash::GnashGcRoot::markReachableResources (this=0xb7f23388) 
at impl.cpp:865
#11 0xb7aea5c0 in gnash::GC::collect (this=0x81242e0) at GC.h:289
#12 0xb7c0a82b in gnash::movie_root::cleanupAndCollect (this=0x8122900) at 
#13 0xb7c0c34f in gnash::movie_root::setRootMovie (this=0x8122900, 
movie=0x8184180) at movie_root.cpp:200
#14 0x0806d307 in gnash::Gui::start (this=0x810b800) at gui.cpp:890
#15 0x0806d47a in gnash::Gui::advanceMovie (this=0x810b800) at gui.cpp:906
#16 0x0807cf60 in gnash::FBGui::run (this=0x810b800) at gui.h:238
#17 0x080766fc in gnash::Player::run (this=0xbfa4154c, argc=3, argv=0xbfa41674, 
address@hidden, address@hidden) at Player.cpp:401
#18 0x08061189 in main (argc=3, argv=0xbfa41674) at gnash.cpp:449

- doesn't happen with GTK Gnash
- fileio doesn't matter
- same effect on three different machines and with different movies
- using current HEAD, of course

I modified the code print the value in setReachable() (that's where
it aborts) and after that it just aborts with:

lt-fb-gnash: character.cpp:912: virtual void gnash::character::destroy(): 
Assertion `!_destroyed' failed.

...while which() always returned 1. So very likely there is memory


reply via email to

[Prev in Thread] Current Thread [Next in Thread]