[i945gm] Adobe flashplayer crashes on GPU enabled SWF files

[OverSoft]

Binary package hint: xserver-xorg-video-intel

Using the newest flash player (standalone and Firefox plugin) (10.0.32.18) with a GPU enabled SWF file on Karmic beta with the Intel 2.9.0-b1 driver crashes flashplayer (segmentation fault, core dumped).
This wasn't a problem in Intrepid and Jaunty.

Steps to test this:
1) Download standalone player from http://www.adobe.com/support/flashplayer/downloads.html
2) Extract the standalone/release player from the archive and put it somewhere useful (like /usr/bin)
3) Create a file named: /etc/adobe/mms.cfg and put this line in it: "OverrideGPUValidation=true"
4) Download a GPU enabled SWF file (for example: http://oversoft.no-ip.org/gpuswftest/ (be sure to download the FLV file as well))
5) Start the player from the terminal: flashplayer flvplayer.swf

It is supposed to go fullscreen and then display a really anoying FLV file (flashes colors, i used it to test vsync).

The player will try to go to fullscreen for a couple of seconds and crash.
Try it again after that and it will insta-crash.
Use a SWF which uses direct-rendering or software-rendering doesn't crash the player (but also, no acceleration obviously).

I don't know if it's related to UXA or to KMS (seems highly impropable) or just to a new bug in the Intel driver.
(Step 3 is needed to set the flash player to hardware acceleration, as the player has a weird way of discovering GPU acceleration capabilities and won't enable acceleration on Intel cards)

[lspci]
00:00.0 Host bridge [0600]: Intel Corporation Mobile 945GM/PM/GMS, 943/940GML and 945GT Express Memory Controller Hub [8086:27a0] (rev 03)
     Subsystem: Intel Corporation Mobile 945GM/PM/GMS, 943/940GML and 945GT Express Memory Controller Hub [8086:27a0]
00:02.0 VGA compatible controller [0300]: Intel Corporation Mobile 945GM/GMS, 943/940GML Express Integrated Graphics Controller [8086:27a2] (rev 03)
     Subsystem: Intel Corporation Mobile 945GM/GMS, 943/940GML Express Integrated Graphics Controller [8086:27a2]

[OverSoft (ubuntu-oversoft)]

(It also did this in the Karmic alpha versions)

[Bryce Harrington (bryce)]

Hi ubuntu-oversoft,

Please attach the output of `lspci -vvnn` and `dmesg`, and attach your /var/log/Xorg.0.log (and maybe Xorg.0.log.old) file from after reproducing this issue. If you're using a custom /etc/X11/xorg.conf please attach that as well.

[This is an automated message. Apologies if it has reached you inappropriately; please just reply to this message indicating so.]

[OverSoft (ubuntu-oversoft)]

Custom xorg.conf

Created like this:
- service gdm stop
- Xorg -configure
- add monitor specs
(the rest is default)

[OverSoft (ubuntu-oversoft)]

Terminal output:

pop@BB-player:~/Downloads/flv$ flashplayer flvplayer.swf
Segmentation fault (core dumped)

[OverSoft (ubuntu-oversoft)]

It's been 2 days, can this bug at least be confirmed?

[Geir Ove Myhr (gomyhr)]

Could you get a full backtrace? That should hopefully tell us if this is a bug in the flash player itself or in the intel driver.

https://wiki.ubuntu.com/Backtrace

[OverSoft (ubuntu-oversoft)]

Seems that it's crashing at libGL.so.1

[OverSoft (ubuntu-oversoft)]

I've just tested a couple of Nvidia ION boards and they don't have this problem.
This is going to be a biggie when people are going to use either flashplayer standalone or flashplayer plugin under Ubuntu 9.10.
For example: Vimeo crashes the browser (disable hardware acceleration and it runs properly)

[OverSoft (ubuntu-oversoft)]

Except for summary and description changes, is there anyone who would actually take a look at this?
It has already been almost 5 months and it still isn't fixed!

[bugbot (bugbot)]

This bug report was filed against an old version of Ubuntu.
Can you confirm whether this is still an issue in natty?

Please also ensure this bug has tags for each Ubuntu release
that the bug is confirmed as affecting.

[bugbot (bugbot)]

We're closing this bug since it is has been some time with no response from the original reporter. However, if the issue still exists please feel free to reopen with the requested information. Also, if you could, please test against the latest development version of Ubuntu, since this confirms the bug is one we may be able to pass upstream for help.