Tuesday, January 16, 2007

A non-beta Flash Player 9 for Linux release

We just released another version of the Flash Player 9 for Linux, the version number is 9.0.31.0. What is different this time is that we do no call it a beta anymore and that it is available on our main download site (Click on the Linux x86 link). For the impatient ones, the direct download of the tarball is here, and the .rpm file is here:

Alright, so far so good... That does not mean that this is a release which I consider feature complete. We do however consider this version to be vastly better than any previous Flash Player for Linux. Since customers were starting to ask for a final version we decided to feature freeze what we have. So two months ago we moved the Flash Player 9 for Linux code into a separate code branch for stabilization. QE (Quality Engineering) has been hammering on this branch for all this time. Every bug fix had to be approved by management. The build you can download now (9.0.31.0) represents that branch.

The internal unstable branch is currently open for all developers and for the first time we build the Windows, Mac and Linux versions on a daily basis from the same code branch. There are various large work items which still need to be tackled and we are in full swing getting this ready as soon as possible. These changes are disruptive for the stability of the build and will be for a while.

Here is a list of some of the Linux specific items we are working on right now, it is far from complete and each of the items might or might not make it into the next version:

  • XEmbed support (specifically GtkPlug). We have tons of bugs which can only be fixed this way.

  • WMODE support. This will be tricky since we need to coordinate this feature with Mozilla as the browsers on Linux have no support for this right now. Also note that XEmbed support is a prerequisite for this feature to work at all.

  • Support for Opera. Hopefully we will get it into a state where it runs stable for more than 5 minutes. Opera 9.10 solved a lot of issues, but not all.

  • Rewrite the braindead/non-standard http stack in the standalone player. Right now this bug can make the standalone player timeout and block for a long time if you need to open any external URL (local files are fine). I'll spare you the details, but this code is really embarassing in my mind.

  • Try to fix other functional bugs which require larger infrastructural changes.

  • Finish the flashsupport add-on library. Adding camera, microphone support etc. Move the project to sourceforge.net or similar site. BTW, it's my fault that this has not seen too many changes lately since I own this piece. :-)

  • Add full screen support.

  • 64bit work is ongoing. You can follow some of the daily progress in the Tamarin CVS tree. (Alright, not too much happened end of December. Forgive us, the holidays are always a little slow in a company like Adobe.)



So what changed between Flash Player 9.0.21.78 beta and 9.0.31.0? Here is a partial list of Linux specific bugs we fixed:


192642 Crash during "Browse multi"
191932 After plug-in player installed, firefox cannot launch (crash while launching)
191298 Firefox 1.5 crashes during tests execution (automated section)
191514 Non Functioning rtmps not functioning
192615 Crash hang navigating away from camera/mic
192747 LocalConnection not functioning between standalone players
192653 change standalone name from gflashplayer to flashplayer
192399 Crash - GrowableBuffer and/or Generated Code
191741 crash when "Allow" is clicked in camera settings UI popup when no camera driver is installed on SUSE 9
192006 Clicking on URL failed to launch websites
191749 mouseDown count advancing more than once per click - stand alone
187060 Rotating the image moves image out of the panel
177699 Bug in detection kit in combination with Linux player
191427 rtmpt not functioning
192036 Incorrectly Functioning File->Close is still selectable in the projector
191843 website crashes with a segmentation fault.
191267 Unable to copy/paste text from text fields using context menu.
191857 remove all printfs from release builds of the plugin and standalone


(Update: I started this entry last week to start listing the fixed bugs, but actually posted the news January 16th. Darn. So some you you saw January 11th as the post date... I am really sorry for the confusion...)

Labels:

47 Comments:

Anonymous Anonymous said...

I think you guys did a really good job. I can't even start to imagine the complications involved in developing a piece of software, like Flash Player, for three completely different platforms. Congratulations!

Tuesday, January 16, 2007 10:53:00 PM  
Anonymous Anonymous said...

It would be added value if you had signed the RPM package with an official GPG key.

Also, the package spec changelog has not been updated since the 9.0.21.55-4 release.

Please don't forget such details.

Wednesday, January 17, 2007 2:06:00 AM  
Anonymous joe said...

Thank you for getting this done. Works without crashing on fedora core 6, firefox-2.0.0.1-x86_64, and nspluginwrapper. Two sites that still don't work are foxnews.com and abcnews.com.

Wednesday, January 17, 2007 2:08:00 AM  
Anonymous Anonymous said...

Not going to thank you for something that should have been done years ago.

And now what, should the BSD people have to beg now?

Wednesday, January 17, 2007 2:09:00 AM  
Anonymous Anonymous said...

I really hope you guys focus to fixing the Opera related problems.

Wednesday, January 17, 2007 2:21:00 AM  
Anonymous Anonymous said...

Wonderful to hear that you are working on 64 bit support.

Wednesday, January 17, 2007 3:26:00 AM  
Anonymous Anonymous said...

An AMD64 version of the Linux Flash player - YES!

Wednesday, January 17, 2007 5:53:00 AM  
Anonymous Anonymous said...

Uooooh!

Good News!
Continue this way.... maybe a free software version in the future? ;-)

Wednesday, January 17, 2007 6:03:00 AM  
Anonymous Anonymous said...

Happy to see this progress! I love it! Great work all you guys/gals working on the Linux-player!

Wednesday, January 17, 2007 7:59:00 AM  
Anonymous Zajec said...

One more annoying bug in Opera is that I can not close ad on http://zajec.net/bug/flash_xss test-case page.

Thursday, January 18, 2007 4:09:00 AM  
Anonymous Anonymous said...

jeah a 64 Bit version :D

Thursday, January 18, 2007 8:14:00 AM  
Anonymous Anonymous said...

Hi!

There is a small bug in libflashsupport in
oss_thread:

FPI_SoundOutput_FillBuffer(ptr,buffer,4096);
len = 4096;
while ( len ) {
written = write(instance->oss_fd, buffer, len);

The last line should be:
write(instance->oss_fd, buffer+4096-len, len);

It is also helpful to increase buffersize ten times, e.g. 40000 instead of 4096.
So the buffer has more reserve.
These modifications result in uninterrupted sound when cpu load is high.

regards,

Peter peter dot heckert at arcor.de

Thursday, January 18, 2007 2:43:00 PM  
Anonymous Anonymous said...

thanks for that update to flashsupport.c anonymous, I was going to ask for an update to the buggy factors you mentioned. At least we can all make that change before waiting for kaourantin to do it.

Kaourantin: please put this code on sourceforge to ease your maintainer responsibilities and everyone benefits!

Thursday, January 25, 2007 1:44:00 AM  
Anonymous Anonymous said...

Still waiting for an Opera update.

Friday, January 26, 2007 3:50:00 AM  
Anonymous Anonymous said...

Mhm, whatabout the arm port?

More Ideas for flashsupport:

* fonts
* codecs (atleast mp3 since there exists dedicated hw resources for it)

In general, using openMAX or gstreamer would make people really happy.

Friday, January 26, 2007 5:41:00 AM  
Anonymous Anonymous said...

why is there still no workable flash player
for freebsd ? really, why doesn't someone
answer this simple question ?
linux flashplayer 9 will likely never work
well on freebsd, with its alsa audio
and huge list of deps.
so is this comment section just for
backslapping and this question is
just gonna be ignored like all
the petitions and queries or what ?

-disgruntled-second-class-web-browsing-type

-

Saturday, January 27, 2007 9:52:00 AM  
Anonymous Dj Joyrider said...

On my Ubuntu installation Opera 9.1 doesn't even detect the new flashplayer-plugin if I throw it into it's own plugin-folder...is this to protect me from using an unstable flashplayer?

Saturday, January 27, 2007 10:02:00 AM  
Anonymous subgeniusd said...

Just checked the Opera/Linux forums and looks like staying with flash 7 a good idea for the time being.

But thanx for the hard work anyway.

Wednesday, January 31, 2007 12:50:00 PM  
Anonymous ven said...

Just superb!

Im impressed over Adobes effort to reach as many as possible, and this flashplayer is what gives Adobe the edge in the time to come when it comes to RIA development. Sure, its not 100% yet, but it just gets closer.

Keep up the good spirit and work!

Thursday, February 01, 2007 4:10:00 PM  
Anonymous ash said...

I have installed the latest flash player (9.0.31.0) on both opera 9.10 and firefox 2.0.0.1 folders on linux ubuntu edgy, but Opera still fails to detect the plugin, even though it's there and now my firefox cant open any site; it crashes as soon as it tries to open something, eg: about:plugins. And i thought this problem was fixed. (???)

Tuesday, February 06, 2007 4:36:00 AM  
Blogger Rei said...

How about AMD64? If you can't get it together to put together flash for other systems and architectures open up the source or divulge the necessary information to a group that can make a descent imitation (Gnash?). I don't understand why a company bent on delivering content like this would turn down an opportunity to allow more people to view said content. I'm sure there are plenty of people out there who would write you a port or imitation for other systems for free (Gnash of course comes immediately to mind).

Friday, February 23, 2007 7:30:00 AM  
Blogger Endoplasmic said...

I was digging through the libflashsupport and noticed support for openSSL.

I'm currently developing software in Flash that uses XML sockets and I want to encrypt the data that is sent from the client to the socket using SSL.

Is there some direction on how to implement this in the Flash development environment or and I out in the cold until the Flash 9 dev environment is out?

Thursday, March 22, 2007 12:37:00 PM  
Anonymous Wolfram Gottfried said...

I am trying to get this working, but even the release version is giving me grief on Flash video playback - it plays a short segment (between 5 and 15 seconds) properly with sound, and then the sound skips a little and the browser crashes with a segmentation fault. I compiled and installed the flashsupport library and that made a little improvments (10-30 seconds before crashes now) but still not there. I had an older version of ALSA on the system (1.0.5a) and thought that was the problem so I updated it to 1.0.12 - still no change. I've seen the plugin work properly on other systems, so I'm sure its just a minor inconsistancy between my system and the plugin, but I can't figure out WHAT the problem is to correct it.

Friday, March 23, 2007 4:16:00 PM  
Anonymous Anonymous said...

Hi, I really appreciate all your hard work for a Linux version of Flash. Since this blog post was made a while back, I was wondering if any progress had been made on the x86_64 version of Flash? Do you foresee it possibly being released in the next year? Thanks.

Thursday, July 19, 2007 11:09:00 AM  
Blogger saritha said...

can any one help me out what is the difference between standard alone player and plugin,Gnash can be used as a standalone player or as a plugin,so which version of gnash is better to run as standard alone player .where to download just help me out

Thursday, July 19, 2007 11:55:00 PM  
Anonymous <a href="http://drugscenterhere.com">ShopMan</a> said...

I like articles like this. Thanks!

Saturday, August 25, 2007 10:32:00 PM  
Anonymous <a href="http://drugscenterhere.com">ShopMan</a> said...

I like articles like this. Thanks!

Saturday, August 25, 2007 10:32:00 PM  
Anonymous <a href="http://m1.aol.com/phentermine4">Phentermine</a> said...

Great Article! Thank You!

Tuesday, August 28, 2007 3:08:00 PM  
Anonymous <a href="http://m1.aol.com/phentermine4">Buy Phentermine</a> said...

Thanks to author! I like articles like this, very interesting.

Wednesday, August 29, 2007 2:37:00 AM  
Anonymous <a href="http://buy-viagra2007.blogspot.com">Buy Viagra</a> said...

very good!

Sunday, September 02, 2007 2:10:00 AM  
Anonymous <a href="http://free-metro-pcs-ringtones.blogspot.com">Free Ringtones</a> said...

nice blog!

Sunday, September 02, 2007 11:25:00 AM  
Anonymous <a href="http://buy-viagra2007.blogspot.com">buy viagra</a> said...

nice blog!Nice information

Monday, September 03, 2007 2:42:00 PM  
Anonymous <a href="http://payday-loans-online--ooz.blogspot.com">Payday loans Online</a> said...

very good site! Best!

Monday, September 03, 2007 9:17:00 PM  
Anonymous <a href="http://buy-levitra--ooz.blogspot.com">Levitra</a> said...

:-) ochen\' zaebatyj blog!

Tuesday, September 04, 2007 2:41:00 AM  
Anonymous <a href="http://adobe-photo-shop-cs2.blogspot.com">Gringo Andre</a> said...

Write something else. Thanks! Best Blog...

Thursday, September 06, 2007 12:19:00 AM  
Anonymous <a href="http://buy-soma--ooz.blogspot.com">Buy Soma</a> said...

soglasen s vami ochen\' zaebatyj blog!

Thursday, September 06, 2007 3:08:00 AM  
Anonymous <a href="http://courses.cvcc.vccs.edu/ENG112_GROSS/_Chat_Room/000008e8.htm">Anonymous</a> said...

Great work,webmaster,nice design!

Sunday, September 09, 2007 6:15:00 AM  
Anonymous <a href="http://courses.cvcc.vccs.edu/ENG112_GROSS/_Chat_Room/000008fd.htm">Anonimous</a> said...

Well done. Keep up the great work. Best regards!

Sunday, September 09, 2007 8:47:00 AM  
Anonymous <a href="http://search.cnn.com/search?query=site:cialis-online-2007.blogspot.com?cialis_online.html">Anonymous</a> said...

Keep up the great work. It very impressive. Enjoyed the visit!

Sunday, September 09, 2007 1:46:00 PM  
Anonymous <a href="http://buy4soma.eamped.com">Anonimous</a> said...

I like it a lot! Nice site, I will bookmark!

Monday, September 10, 2007 7:17:00 AM  
Anonymous Speed Dating said...

Happy to see this progress! I love it! Great work all you guys/gals working on the Linux-player!

Wednesday, October 03, 2007 2:09:00 AM  
Blogger Federico, A. Moretti (aka fedalmor) said...

After some monthes... the same "problem": there are no ports on amd64... OK, there's nspluginwrapper. But it's a little bit frustrating! :(

BTW... I prefer to use the "closed" one instead of Gnash (for now). It's a pity that transparency issues are not even been solved. We'll wait for it! :P

Wednesday, October 03, 2007 9:12:00 PM  
Anonymous Terramir said...

Cheers for the write up. Keep up the great work :)

Tuesday, October 09, 2007 3:37:00 AM  
Anonymous Tricor said...

Thank you for this blog, I spent my time effectivelly browsing it!

Monday, October 22, 2007 7:39:00 PM  
Anonymous muanis said...

I've just downloaded version 9.0.0.48 and it's not working in my amd64 ubuntu.

ERROR: Your architecture, \'x86_64\', is not supported by the
Adobe Flash Player installer.

I did some error or the support was discontinued?

Wednesday, October 24, 2007 9:46:00 AM  
Blogger  said...

I want know if the moviestar include the H.264 encoder function? If we can use moviestar to talk to each other with the H.264 vide stream by use FMS?

Tuesday, November 20, 2007 12:31:00 AM  
Blogger Alexander said...

Hello,


I'm using libflashsupport (http://labs.adobe.com/wiki/index.php/Flash_Player:Additional_Interface_Support_for_Linux) in order to support Flash Player for Linux version 9.0.124.0 sound for proprietary Linux embedded system. I'm following ALSA/OSS/PulseAudio reference implementation and I noticed that there is a deadlock/crash.



Scenario:
Flash Player calls FPX_SoundOutput_Close() while "data pump" thread (like alsa_thread/oss_thread) is blocked inside FPX_SoundOutput_FillBuffer(). In such case FPX_SoundOutput_Close() can not pthread_join "data pump" thread which causes browser to deadlock. In case "data pump" thread is killed by FPX_SoundOutput_Close() using pthread_cancel() Flash Player deadlocks or crashes, since "data pump" thread is holding mutex by calling FPX_SoundOutput_FillBuffer().



Similar issue was also reported by PulseAudio: http://permalink.gmane.org/gmane.comp.audio.pulseaudio.bugs/42



I'd appreciate any help with resolving this issue. Let me know if any additional info is required.



Thank you for your help.

Thursday, May 08, 2008 12:11:00 PM  

Post a Comment

<< Home