bugGnash - The GNU Flash player - Bugs: bug #23808, assertion failed while loading...

 
 

bug #23808: assertion failed while loading extensions

Submitted by:  Sandro Santilli <strk>
Submitted on:  Thu 10 Jul 2008 07:35:30 AM UTC  
 
Category: coreSeverity: 5 - Blocker
Release: NoneStatus: Invalid
Privacy: PublicAssigned to: Sandro Santilli <strk>
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)

Thu 04 Sep 2008 09:53:51 AM UTC, comment #9:

Well, I filed it, I'll close it...

Sandro Santilli <strk>
Project MemberIn charge of this item.
Wed 03 Sep 2008 01:13:08 PM UTC, comment #8:

Yep, it seems that was the problem.
Maybe a version checking would help when loading plugins.

Sandro Santilli <strk>
Project MemberIn charge of this item.
Wed 03 Sep 2008 01:10:50 PM UTC, comment #7:

setting GNASH_PLUGINS to point to just-built plugins seems
to fix the issue, maybe I had a bogus format around..
will take a closer look later.

Sandro Santilli <strk>
Project MemberIn charge of this item.
Wed 03 Sep 2008 01:07:39 PM UTC, comment #6:

Actually, even loading an empty SWF triggers this.
In my case I enable fileio,gtk,dejagnu.
The segfault comes from dejagnu (the first found)

Sandro Santilli <strk>
Project MemberIn charge of this item.
Wed 03 Sep 2008 12:48:49 PM UTC, comment #5:

I just found out the configure switch to enable extensions
changed. It takes --enable-extensions=... now (not --with-extensions).
I rebuilt with extensions enabled and can confirm the segfault.
Try utilities/gprocessor -v extensions/fileio/test.swf

Sandro Santilli <strk>
Project MemberIn charge of this item.
Tue 26 Aug 2008 01:48:29 PM UTC, comment #4:

Well, still can't reproduce with fileio enabled.

Benjamin Wolsey <bwy>
Project Member
Wed 30 Jul 2008 08:06:28 PM UTC, comment #3:

It was fileio. Could be an error on my side though, since 'distcheck' doesn't fail there I guess it was (distcheck enables all extensions, on mx1 I just dropped the mysql one)

Sandro Santilli <strk>
Project MemberIn charge of this item.
Tue 29 Jul 2008 10:32:41 PM UTC, comment #2:

I can't reproduce the assertion failure. Which extensions are enabled?

I do get valgrind errors on exit when running with extensions enabled. I'm cleaning up the SharedLib and extension code at the moment, so I'll have a look at it.

Benjamin Wolsey <bwy>
Project Member
Thu 10 Jul 2008 07:46:22 AM UTC, comment #1:

Dunno if this is related, but:

==31936== Invalid read of size 4
==31936== at 0x40150C8: (within /lib/ld-2.7.so)
==31936== by 0x400D5D5: (within /lib/ld-2.7.so)
==31936== by 0x4010F5D: (within /lib/ld-2.7.so)
==31936== by 0x58D4C18: (within /lib/tls/i686/cmov/libdl-2.7.so)
==31936== by 0x400D5D5: (within /lib/ld-2.7.so)
==31936== by 0x58D52BB: (within /lib/tls/i686/cmov/libdl-2.7.so)
==31936== by 0x58D4B50: dlopen (in /lib/tls/i686/cmov/libdl-2.7.so)
==31936== by 0x4D665EB: sys_dl_open (ltdl.c:1113)
==31936== by 0x4D66D65: tryall_dlopen (ltdl.c:2446)
==31936== by 0x4D680CD: try_dlopen (ltdl.c:2534)
==31936== by 0x4D6873A: lt_dlopenext (ltdl.c:3501)
==31936== by 0x4560D95: gnash::SharedLib::openLib(char const*) (sharedlib.cpp:170)
==31936== Address 0x5e48880 is 8 bytes inside a block of size 10 alloc'd
==31936== at 0x4022AB8: malloc (vg_replace_malloc.c:207)
==31936== by 0x4D6650C: lt_emalloc (ltdl.c:1022)
==31936== by 0x4D67AA1: trim (ltdl.c:3024)
==31936== by 0x4D68175: try_dlopen (ltdl.c:3269)
==31936== by 0x4D6873A: lt_dlopenext (ltdl.c:3501)
==31936== by 0x4560D95: gnash::SharedLib::openLib(char const*) (sharedlib.cpp:170)
==31936== by 0x4560E93: gnash::SharedLib::openLib() (sharedlib.cpp:138)
==31936== by 0x45385D0: gnash::Extension::initModule(char const*, gnash::as_object&) (extension.cpp:145)
==31936== by 0x4538788: gnash::Extension::scanAndLoad(gnash::as_object&) (extension.cpp:126)
==31936== by 0x4281D8A: gnash::Global::loadExtensions() (Global.cpp:699)
==31936== by 0x4283973: gnash::Global::Global(gnash::VM&, gnash::ClassHierarchy*) (Global.cpp:255)
==31936== by 0x43B7BDC: gnash::VM::init(gnash::movie_definition&, gnash::VirtualClock&) (VM.cpp:65)

Sandro Santilli <strk>
Project MemberIn charge of this item.
Thu 10 Jul 2008 07:35:30 AM UTC, original submission:

I've no idea why, but the VM::get() assertion (_singleton.get())
fails when extensions are enabled.
The backtrace shows that VM::get() is called after VM::init()
so can't figure why _singleton.get() fails as VM::init() was
carefully written to make sure the singleton was created before
registering the global object to it (with the global object constructor doing extention scan).

Sandro Santilli <strk>
Project MemberIn charge of this item.

 

(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 strk (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 7 latest changes.

    Date Changed By Updated Field Previous Value => Replaced By
    Thu 04 Sep 2008 09:53:51 AM UTCstrkStatusReady For Test=>Invalid
      Open/ClosedOpen=>Closed
    Wed 03 Sep 2008 01:13:08 PM UTCstrkStatusConfirmed=>Ready For Test
    Wed 03 Sep 2008 01:07:39 PM UTCstrkAssigned tobwy=>strk
    Wed 03 Sep 2008 12:48:49 PM UTCstrkStatusReady For Test=>Confirmed
    Wed 30 Jul 2008 08:06:28 PM UTCstrkStatusNone=>Ready For Test
    Tue 29 Jul 2008 10:32:41 PM UTCbwyAssigned toNone=>bwy

    Back to the top


    Powered by Savane 3.1-cleanup1