drmMap libGL error on "SuperSavage/IXC 64" and "Xorg 7.2"

Bug #88905 reported by Joachim Frieben
46
Affects Status Importance Assigned to Milestone
Linux
Fix Released
Medium
linux (Ubuntu)
Fix Released
Undecided
Unassigned
linux-source-2.6.20 (Ubuntu)
Won't Fix
Low
Unassigned
xserver-xorg-video-savage (Debian)
Fix Released
Unknown

Bug Description

Binary package hint: kernel-image-2.6.20-9-generic-di

Hardware rendering is broken for "SuperSavage/IXC 64" and "Xorg 7.2" in the current "feisty" tree. On my "IBM ThinkPad T23", "glxinfo" now reports:

...
libGL error: drmMap of framebuffer failed (Invalid argument)
libGL error: reverting to (slow) indirect rendering
display: :0 screen: 0
direct rendering: No
...

This used to work in previous releases of "Ubuntu". The responsible is an outdated "DRM" version in the current kernel source tree. Rebuilding "drm.ko" and "savage.ko" from the current "freedesktop.org" tree allows to recover working hardware rendering.

I have reported this bug upstream at: http://bugzilla.kernel.org/show_bug.cgi?id=7767

Revision history for this message
Kyle McMartin (kyle) wrote :

OK. I'll see what we can do about shipping an updated Savage DRI.

Changed in linux-source-2.6.20:
assignee: nobody → kyle
importance: Undecided → Medium
status: Unconfirmed → Confirmed
importance: Medium → Low
Revision history for this message
Ben Collins (ben-collins) wrote :

Confirmed bugs need to be assigned to ubuntu-kernel-team.

Changed in linux-source-2.6.20:
assignee: kyle → ubuntu-kernel-team
Revision history for this message
Joachim Frieben (jfrieben) wrote :

Still broken in kernel "2.6.20-12-generic". This probably won't work out for "feisty" final ...

Changed in xserver-xorg-video-savage:
status: Unknown → Unconfirmed
Changed in linux:
status: Unknown → Confirmed
Changed in xserver-xorg-video-savage:
status: Unconfirmed → Confirmed
Revision history for this message
catchit (catchit) wrote :

This issue is still present in the Ubuntu 7.04 release. Can you please provide instructions on how I can rebuild "drm.ko" and "savage.ko"?

Thanks.

Revision history for this message
Joachim Frieben (jfrieben) wrote :

The procedure is for instance described here:

    http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=393266;msg=37;att=0

Revision history for this message
catchit (catchit) wrote :

I've followed those instructions under Ubunto 7.04. However, when I make DRM_MODULES="savage" the follow error comes up.

make -C /lib/modules/2.6.20-15-generic/build SUBDIRS=`pwd` DRMSRCDIR=`pwd` modules
make[1]: Entering directory `/usr/src/linux-headers-2.6.20-15-generic'
  CC [M] /usr/src/drm/linux-core/drm_compat.o
/usr/src/drm/linux-core/drm_compat.c:190: error: static declaration of ‘vm_insert_pfn’ follows non-static declaration
include/linux/mm.h:1126: error: previous declaration of ‘vm_insert_pfn’ was here
make[2]: *** [/usr/src/drm/linux-core/drm_compat.o] Error 1
make[1]: *** [_module_/usr/src/drm/linux-core] Error 2
make[1]: Leaving directory `/usr/src/linux-headers-2.6.20-15-generic'
make: *** [modules] Error 2

Any idea what is wrong?

Revision history for this message
Eugenia Loli-Queru (eloli) wrote :

It seems that this bug you stumbled upon is a new ubuntu bug:
http://<email address hidden>/msg30392.html
It didn't exist with the month-old 2.6.20 kernel I tried.

Revision history for this message
catchit (catchit) wrote :

So I'm out of luck till a kernel update?

Revision history for this message
Eugenia Loli-Queru (eloli) wrote :

Unless you know how to fix the kernel, yeah, you will have to wait for either ubuntu or freedesktop to do the necessary changes to their code.

Revision history for this message
Joachim Frieben (jfrieben) wrote :

No, this is not a new "Ubuntu" bug. It also happens on "Fedora". I am sorry that I completely forgot about that issue that I encounter myself when using the current tree. I am using an older snapshot that compiles correctly. Try to checkout the "DRM" branch as of about 2 months ago. That one should work.

Revision history for this message
catchit (catchit) wrote :

Sounds a lil too advanced for me... I maybe just have to wait this one out.

Revision history for this message
Joachim Frieben (jfrieben) wrote :

Simply unpack the attachment, go to "linux-core" and type 'make DRM_MODULES="savage"' to build "drm.ko" and "savage.ko" or simply type "make" to build all the kernel modules. Make sure to load the new modules "drm.ko" and "savage.ko" before the "X" server starts or simply replace the original ones by the new ones as described elsewhere. Good luck!

Revision history for this message
Joachim Frieben (jfrieben) wrote :
Revision history for this message
catchit (catchit) wrote :

Ok, did as you suggested. Everything went find. However, when I type "glxinfo | grep direct" direct rendering = no

Any further suggestions?

Revision history for this message
Giorgio Vazzana (mywing) wrote :

Thank you Joachim, that worked for me (I have an IBM Thinkpad T23, 2647-4MG).
I've run the following commands:

tar xjvf drm.tar.bz2
cd drm/linux-core/
make DRM_MODULES="savage"
sudo mv /lib/modules/2.6.20-15-generic/kernel/drivers/char/drm/savage.ko /lib/modules/2.6.20-15-generic/kernel/drivers/char/drm/savage.ko.old
sudo mv /lib/modules/2.6.20-15-generic/kernel/drivers/char/drm/drm.ko /lib/modules/2.6.20-15-generic/kernel/drivers/char/drm/drm.ko.old
sudo cp savage.ko /lib/modules/2.6.20-15-generic/kernel/drivers/char/drm/savage.ko
sudo cp drm.ko /lib/modules/2.6.20-15-generic/kernel/drivers/char/drm/drm.ko
sudo depmod -a

After rebooting, "glxinfo | grep direct" reports "direct rendering: Yes".

@catchit: I hope this helps.

Revision history for this message
catchit (catchit) wrote :

Giorgio Vazzana.

I've got the same laptop and taken the exact same steps. The only thing that differs is that I ran the commands as Root. Direct rendering = no.

Revision history for this message
Joachim Frieben (jfrieben) wrote :

@catchit

What model is your notebook exactly? Is direct rendering supported by the "X" server as reported in "Xorg.0.log"? Did you ever manage to get direct rendering work for previous releases? Do you still obtain the error message:

...
 libGL error: drmMap of framebuffer failed (Invalid argument)
 libGL error: reverting to (slow) indirect rendering
 display: :0 screen: 0
 direct rendering: No
 ... ?

I have an "IBM ThinkPad T23" like Giorgio. For older models T20-T22, success is less certain. Also make sure that the video memory is sufficient for direct rendering. Try "16bpp" and a resolution like "1024x768" to start with.

Revision history for this message
catchit (catchit) wrote :

I have the same notebook, an IBM T23.

In the Xorg.0.log I saw:
(II) SAVAGE(0): Direct rendering enabled
(==) RandR enabled

When I do a "glxinfo" I don't see the libGL errors... but still get:
name of display: :0.0
display: :0 screen: 0
direct rendering: No

I've had direct rendering enabled in the last version of Ubuntu, no configuration necessary.

I'm stumped....

Revision history for this message
Eugenia Loli-Queru (eloli) wrote :

Catchit, you shouldn't even be doing all the things you do to fix it. This is a major bug that Ubuntu has shipped with, and they should have been the ones that fixed it within a week that this bug was confirmed. But they didn't. So, don't bother and just wait for a fix.

However, if you want to continue trying, something is wrong with your xorg setup, maybe you haven't dri/gl loaded on your xorg.conf. There is no other explanation if you have taken all the right steps to recompile the drm driver. I have a T23 too and it works. If it still doesn't for you, check your xorg to make sure you enable all the extensions needed.

Revision history for this message
catchit (catchit) wrote :

Well it's just weird how I've done exactly as the others, yet mine doesn't work. glx and dir are in my xorg.conf.

Not sure what else to look for so... maybe you're right, just wait.

Revision history for this message
Tom Shaw (firephoto) wrote :

I've posted this xorg.conf elsewhere but maybe that bug is closed now.

It works fine here with the default install and this xorg.conf except it takes a cold start sometimes for the acceleration to be working, otherwise it paints to the screen really slow. Also I don't think the savage chips support 24 bit color on linux (or in general perhaps) so you need to be using 16 bit as the default.

Revision history for this message
catchit (catchit) wrote :

Tom,

I just tried this xorg.conf, still no luck.

Thanks though.

Revision history for this message
Tom Shaw (firephoto) wrote :

It does work here. I did remember that I enabled acpi in grub but I can't recall if this affected X or not. Again, this may not work if you don't power off the laptop since rebooting doesn't reinitialize the bios with these laptops. Powering off then starting up again is 100% working every time with this setup, rebooting isn't always working (properly).

Revision history for this message
Tormod Volden (tormodvolden) wrote :

catchit, if you don't see the libGL error messages any longer, you probably have another problem than the one in this bug report. Please file a support request, and attach your xorg.conf and /var/log/Xorg.0.log. Well, last time I checked the support tracker didn't handle attachments, so you might have to attach them here, off-topic or not.

Revision history for this message
catchit (catchit) wrote :

Tormod,

My xorg.conf is the default from Ubuntu 7.04. See attached. (Can only do one attachment at a time, so I've left out the log)

Revision history for this message
compartist (compartist) wrote :

Thanks Tom and Eugenia! I also have a T23 and with all your help, I was able to get my laptop working.

Catchit, I've also been looking at your Xorg.conf file because I know a couple of others who are having the same problem, have tried everything and still no luck. I'll let you know if I come up with anything...
BTW, can you attach your Xorg.0.log? It would help a lot...

Revision history for this message
catchit (catchit) wrote :

Compartist,

Here is my Xorg.0.log

Revision history for this message
MiDWaN (midwan) wrote :

This might help:

I did manage to get Direct Rendering: Yes in "glxinfo" by following the compiling instructions found here:
http://home.comcast.net/~jyavner/T23/#savage-bug

and then following the troubleshooting instructions, found here: http://dri.freedesktop.org/wiki/DriTroubleshooting

What really helped was turning the environment variable export LIBGL_DEBUG to "verbose", with: "export LIBGL_DEBUG=verbose" (for a bash shell), as described in the above link.

It seems that the "savage_dri.so" file was not in the expected path after "depmod -a" was executed, so it couldn't be loaded.

After moving the file to "/usr/X11R6/lib/modules/drm/" it worked magically.

I now have Direct Rendering, and GL output is much smoother (it reports 500FPS after 5 secs in glxgears).

I hope this helps!

Revision history for this message
Joachim Frieben (jfrieben) wrote :

>It seems that the "savage_dri.so" file was not in the
>expected path after "depmod -a" was executed, so it
>couldn't be loaded.

The "depmod" command has nothing to do with the "Xorg" driver module "savage_dri.so". It only ensures that new kernel modules "drm.ko" and "savage.ko" do get used instead of the old ones after moving them to the modules directory.

Revision history for this message
MiDWaN (midwan) wrote : Re: [Bug 88905] Re: No hardware rendering for "SuperSavage/IXC 64" and "Xorg 7.2"

Oops, sorry about that then. Still learning... :)

Joachim Frieben wrote:
>> It seems that the "savage_dri.so" file was not in the
>> expected path after "depmod -a" was executed, so it
>> couldn't be loaded.
>
> The "depmod" command has nothing to do with the "Xorg" driver module
> "savage_dri.so". It only ensures that new kernel modules "drm.ko" and
> "savage.ko" do get used instead of the old ones after moving them to the
> modules directory.
>

Revision history for this message
Nick (morrownr) wrote : Re: No hardware rendering for "SuperSavage/IXC 64" and "Xorg 7.2"

The subject video chipset is the one used in the entire Thinkpad T2x series. This was a very popular line and the systems are powerful enough to still be in widespread use. A quick search of the forums shows this bug is causing a lot of questions. I've reverted my wife's T23 back to Windows 2000 because I don't have much hope of this bug being fixed very soon given the LOW priority that it is assigned.

Request the priority of this bug be elevated.

Revision history for this message
Tormod Volden (tormodvolden) wrote :

It would be good if all those with the same issue as the original reporter would try out those DRM snapshot modules and say if they work or not.

If it can help anything, I compiled the modules for the current 2.6.20-16-generic kernel version. The attached archive file has two kernel modules that should be copied to /lib/modules/2.6.20-16-generic/kernel/drivers/char/drm/ (and nowhere else). Make a backup of the old modules before you copy. Reboot and test.

Revision history for this message
Carbonfish (ironfish) wrote :

I too think that there are far too many ThinkPads still in use to allow this bug to have such a low priority. While I realize that the devs only have so much time on their hands and a Herculean task to do, but having to find and install a workaround for this myself is pretty frustrating. I'm an anthropologist, not a coder. I need my bullet-proof old ThinkPad to run optimally and shouldn't have to chase this bug myself. Please address this.

Thank you.

Revision history for this message
Eugenia Loli-Queru (eloli) wrote :

Well said Carbonfish. The Thinkpads are used a lot and the S3 3D *and* 2D driver has major bugs (especially the 2D driver, I already filed 3-4 bugs for that already). And it's not lack of documentation, it's just bugs. And it makes the usage of a laptop with Savage unbearable. I promised my old T23 to my brother as a gift, and my idea was to give it away with Ubuntu. But seeing all these bugs, I have now asked my brother to save his old copy of XP so he can put that in there when he gets the laptop (he lives in Greece, I am in USA).

Revision history for this message
Carbonfish (ironfish) wrote :

Would anyone care to be an angel and walk me through the workaround for this bug? I just installed Feisty with the 2.6.20-16-generic kernel this afternoon and of course, thought that my problems would be solved. Guess not.

If someone could give me an algorithm for getting my T23 to render directly I would be in your debt. But please be gentle, I'm a middle aged man and my heart isn't what it used to be.

Thank you in advance for any assistance you might offer.

KC

Revision history for this message
Joachim Frieben (jfrieben) wrote :

@ Carbonfish:

As I have pointed out several comments above, the procedure is for instance described here:

    http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=393266;msg=37;att=0

The current "git" tree is fully functional. So, just go ahead!

Revision history for this message
Michael Kanis (mkanis) wrote :

Following the instructions Joachim Frieben kindly linked in this post [1] and again in the last one, I had absolutely no problems installing the driver from the snapshot he posted here [2]. It took me less than 5 minutes including the reboot and works like a charme.
However, to make the procedure even a bit faster, I wrote the commands from [1] together in a very simple (no error handling!) script, which I attach here. It belongs in the drm folder, from [2] and must be executed with sudo. The kernel-headers package matching your installed kernel must be installed. Please use at your own risk.

[1] https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/88905/comments/5
[2] https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/88905/comments/13

Revision history for this message
Carbonfish (ironfish) wrote :

Thank you both Joachim and Michael. I do no have time to apply this patch today, but I will follow your recommendations to the letter and I am sure that things will go well. I greatly appreciate you both for taking the time to follow up on this and give me this guidance.

Thanks again,

KC

Revision history for this message
catchit (catchit) wrote :

Tormod Volden I just tried your built for 2.6.20-16-generic with no success. Direct rendering = no

Revision history for this message
Tom Shaw (firephoto) wrote :

For those not getting direct rendering, or other issues releated to this chipset, you need to understand that you can't do 3D with 24 bit color, there's not enough video ram for this. You have to change the default depth to 16.

To repeat something else, the default install of feisty is working fine with a proper xorg.conf (posted above and elsewhere) unless you do a warm reboot without powering the system off so that is something that might be fixed by newer DRM modules.

So the most critical things I see here is an improper xorg configuration by default for this particular savage chip and the need for the newer DRM modules to fix (maybe) some other issues.

Revision history for this message
Joachim Frieben (jfrieben) wrote :

@ Tom Shaw :
The hint concerning the video memory was already given in comment 17. The important thing to retain is that in this case, already "Xorg.0.log" reports that direct rendering is disabled due to insufficient video memory! In "catchit's" case, however, this seems to be ok, whereas "glxinfo/glxgears" [you want to set "LIBGL_DEBUG" to "verbose" before this step ] reports "libGL error: drmMap of framebuffer failed ..".
In case you ["catchit"] do not obtain this message, you have a different problem, and then, this is not the place to post your observations.
Again: restarting the "X" server is not sufficient after you have replaced the kernel modules following the procedure outlined above! You do need to reboot the system [or shutdown the "X" server, unload the old modules, load the new modules and restart it].

Revision history for this message
Joachim Frieben (jfrieben) wrote :

@ Tom Shaw :
I am running an "IBM ThinkPad T23" connected to an external monitor at "1280x1024@24bpp" with hardware rendering up and running which is clearly possible for this model sporting 16 MB of video memory.

Revision history for this message
Tormod Volden (tormodvolden) wrote :

Tom, this bug is about one of the "(maybe) some other issues": See the error message in the original post, and the fact that the new DRM modules fixed it for the reporter, which has a SuperSavage/IXC 64.

It is true that Option BusType PCI avoids the bugs that exist in the AGP handling of many Savage cards (and other cards as well, ATI comes to my mind). But the other options in your xorg.conf above are not needed and might be wrong for other users.

It is also true that you need enough memory available to run DRI. Otherwise DRI will just be disabled and this is reported in Xorg.0.log.

Revision history for this message
catchit (catchit) wrote :

@ Joachim Frieben

Why isn't this thread for me? It's regarding: No hardware rendering for "SuperSavage/IXC 64" and "Xorg 7.2"

I've replaced the modules and rebooted/cold boot. You're telling me this is incorrect?

Revision history for this message
Carbonfish (ironfish) wrote :

Hello,

I must apologize for being thick, but I was preparing to open a terminal and follow Joachim's instructions from post # 36, but then I lost my nerve as I didn't understand the warning that the "kernel headers must be installed."

Again, I am sorry that I don't completely understand, but I am trying to pick up the command line as fast as I can and I am still pretty low on the learning curve.

If it is not too much trouble, could someone please elucidate? I have never expanded a tarball before and so am not sure what I am doing.

As embarrassing as this confession was, I would still like to have direct rendering work on my machine, so your indulgence would be greatly appreciated.

KC

Revision history for this message
Tormod Volden (tormodvolden) wrote :

For those who really need the new kernel modules, but feel uncomfortable with the command line, I put together what I think is a "fool-proof" double-click-able installer script. It has a bunch of error checking so no harm should be done if trying. Maybe a bit of a overkill for this issue here, but I wrote it as a possible basis for other things as well.

1. Download the attached package
2. Double-click on the package to open the archive
3. Click "Extract" to unpack the archive
4. Double-click the newly created "drm-easy-installer" directory (probably on your desktop)
5. Double click on the "easy-module-installer"

Just press Enter for the yes/no questions and enter your password when asked for it. Enjoy!

Note: Downloading the current git today does not work on the current Feisty kernel. Therefore use the version that is shipped with the package, which is the same as in Joachim's post. If you download a "bad" version, just delete it again from the directory, otherwise it will be proposed as the newest candidate for building, instead of the old, working one.

Of course, this is one of these third-party installer script your mother warned you about...

Revision history for this message
Bayla (bayla) wrote :

I started out with Midwan's advice on his webblog and got stuck after this:* Install Debian packages "linux-source-2.6.20", "gcc", "libc6-dev", "linux-headers-2.6.20-15-generic", and "linux-kernel-headers".
* cd /usr/src/
* sudo tar jxvf linux-source-2.6.20.tar.bz2
* cd linux-source-2.6.20
* sudo cp /boot/config-2.6.20-15-generic .config
* sudo make oldconfig && sudo make prepare
* cd /lib/modules/2.6.20-15-generic
* sudo ln -s /usr/src/linux-source-2.6.20 build
Download this package to ~/drm.tar.bz2

I don't think I have built the kernel???

This is what happened after I ran the easy installer:

Found DRM source directory drm-20070225
Would you like to download the newest 'git' version, instead
of using these sources [y/N]?
Will use existing source drm-20070225
Building from drm-20070225

Makefile:153: *** Cannot find a kernel config file. Gestopt.

Building modules failed!
Leaving build tree in /tmp/tmp.ZNRAH12640
Press Enter to close this window

I'm pretty much a newbie on Linux, really am trying my very best, but please keep it simple to me since my head starts spinning quite easily since I am not very familiar with a lot of terms.

Did I do something very wrong? How to continue now?

I appreciate your help very much!

Thnx in advance :-))) Bayla

Revision history for this message
Carbonfish (ironfish) wrote :

Hello Tormod,

Thank you very much for creating the "fool proof" installer script. I ran it a few minutes ago. I am sorry to say that I must be a something more problematic than a plain old fool, as I did not have success. The script ran fine, but failed to build the modules. I have attached a screenshot of the terminal window at the error message.

Thank you again for your efforts in assisting those of us that lack your skills (although I must say that I am glad that I do not share Bayla's "head-spinning" problem).

I shall endeavor to persevere.

K. D. Cline

Revision history for this message
Carbonfish (ironfish) wrote :

Greetings all!

Success!! Thank you very much Tormod, and everyone else who helped to get this little machine closer to correct operation. I really need this old devil for work and you have been a first rate team to have in my corner. THANK YOU AGAIN!

What happened the first time was that I answered "yes" when the installer asked me if I wanted to download the latest git modules. I should have used the git package that Tormod supplied. When I ran the script again, I used the supplied git information and everything went very smoothly.

The attached screenshot should say it all.

One more thank you,

K. D. Cline

Revision history for this message
Joachim Frieben (jfrieben) wrote :

@ catchit:

This bug is not for you **unless** you observe the already well-known "libGL error: drmMap of framebuffer failed .." when executing "glxgears/glxinfo", preferably after setting "LIBGL_DEBUG" to "verbose". Since you do not seem to encounter this message, you have benn bitten by a different bug. That's why this bug is likely to be not your bug ..
All other users who have reported the "DRM" map problem have been able to settle the issue by the fix described above.

Revision history for this message
Tormod Volden (tormodvolden) wrote :

Bayla, your system got a little messed up when you followed that _Debian_ weblog. You can fix it with:
cd /lib/modules/2.6.20-15-generic
sudo rm build
sudo ln -s /usr/src/linux-headers-2.6.20-15-generic build

Or by installing the new kernel version -16, which you should do anyway.

Revision history for this message
catchit (catchit) wrote :

@ Joachim Frieben

Well I'm at a loss then. This was a fresh install of Ubuntu 7.04.

Revision history for this message
Tormod Volden (tormodvolden) wrote :

catchit, when your Xorg server says DRI is enabled, but the clients not, it's usually due to installation of 3rd-party drivers (from ATI Inc in this case). See https://help.ubuntu.com/community/RadeonDriver#head-229f59879c2a2b6c6635d1e189706d97f836b879 for how to fix it.

Revision history for this message
catchit (catchit) wrote :

@ Tormod Volden

Ok, so I did the "sudo modprobe -r fglrx" and rebooted... nothing has changed.

glxinfo | grep vendor returns the following with no mention of ATI:

server glx vendor stringe: SGI
client glx vendor string: SGI
OpenGL vendor string: S3 Graphics Inc.

Revision history for this message
Bayla (bayla) wrote :

@Tormod-

I'm sorry but I do not have the knowledge to do such a thing...Is there anywhere where I can look this up or get the command lines on how to do this safely? Or a place /person besides the local forum's where I can ask q's like these?
(I have no experience with IRC)

Thnx in advance
Bayla

Revision history for this message
Tormod Volden (tormodvolden) wrote :

Bayla, you enter those three lines in a Terminal window, that's it. Or to upgrade your kernel, use Update Manager.

Catchit and Bayla, you can ask your questions at https://answers.launchpad.net/ubuntu.

Revision history for this message
Nick (morrownr) wrote :

Report:

The script provided by Tormod Volden in comment 46 worked here on my T23.

FPS increased from ~50 to ~500.

I used the default answers for all questions in the script.

Hopefully this problem is resolved "out of the box" in Gutsy.

Revision history for this message
shaggy (slimshaggy) wrote :

There is this bug, the problem has been found and a solution, too.

Please make your users happy and release a fix for feisty.

Revision history for this message
Joachim Frieben (jfrieben) wrote :

Issue fixed for me by kernel "2.6.22-rc3-git6".

Changed in linux:
status: Confirmed → Rejected
Revision history for this message
QuentinHartman (qhartman) wrote :

@Tormod

Your "automagic" git script worked for me using the snapshot you included. Using the most fresh git checkout failed to build, however. In any case, thanks for the idiot proofing!

Revision history for this message
aquo (sbauch) wrote :

Is there any solution that doesn't require compiling own modules? I have feisty and would like to have a solution for my T23. Is there a kernel-patch for 2.6.20-15/16?

The bug is still open.

Revision history for this message
Tormod Volden (tormodvolden) wrote :

AFAIK, it has not been tracked down exactly what the bug is and whether a simple fix can be applied, just that the newer drm works. The changes that probably make the difference are discussed in https://bugs.freedesktop.org/show_bug.cgi?id=5714.

Revision history for this message
Joachim Frieben (jfrieben) wrote :

@aquo: I recommend you to install the current development kernel 2.6.22-7 from the Gutsy tree in addition to 2.6.20-15/16. For that one the drm problem has been clearly fixed upstream.

Changed in linux:
status: Invalid → Fix Released
Revision history for this message
bambi (svuillemenot) wrote :

As midwan; I did manage to get Direct Rendering: Yes in "glxinfo" by following the compiling instructions found here:
http://home.comcast.net/~jyavner/T23/#savage-bug.
Thanks to all for advices; and good luck to you catchit.

Revision history for this message
Javier Rovegno Campos (jrovegno) wrote :

Thanks Tormod Volden , the "fool-proof" double-click-able installer script works fine
on 2.6.20-16-generic kernel.
System: T23

Revision history for this message
Tormod Volden (tormodvolden) wrote :

The script works for the old drm version which is included in the attachment in comment 46. For newer drm versions and general usage, please get the updated easy-drm-modules-installer.v3 from https://wiki.ubuntu.com/XorgOnTheEdge

Revision history for this message
shaggy (slimshaggy) wrote :

In gutsy the bug does not exist, since it has been fixed upstream.

Changed in xserver-xorg-video-savage:
status: Confirmed → Fix Released
Revision history for this message
Launchpad Janitor (janitor) wrote : This bug is now reported against the 'linux' package

Beginning with the Hardy Heron 8.04 development cycle, all open Ubuntu kernel bugs need to be reported against the "linux" kernel package. We are automatically migrating this bug to the new "linux" package. However, development has already began for the upcoming Intrepid Ibex 8.10 release. It would be helpful if you could test the upcoming release and verify if this is still an issue - http://www.ubuntu.com/testing . If the issue still exists, please update this report by changing the Status of the "linux" task from "Incomplete" to "New". We appreciate your patience and understanding as we make this transition. Thanks!

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

The Ubuntu Kernel Team is planning to move to the 2.6.27 kernel for the upcoming Intrepid Ibex 8.10 release. As a result, the kernel team would appreciate it if you could please test this newer 2.6.27 Ubuntu kernel. There are one of two ways you should be able to test:

1) If you are comfortable installing packages on your own, the linux-image-2.6.27-* package is currently available for you to install and test.

--or--

2) The upcoming Alpha5 for Intrepid Ibex 8.10 will contain this newer 2.6.27 Ubuntu kernel. Alpha5 is set to be released Thursday Sept 4. Please watch http://www.ubuntu.com/testing for Alpha5 to be announced. You should then be able to test via a LiveCD.

Please let us know immediately if this newer 2.6.27 kernel resolves the bug reported here or if the issue remains. More importantly, please open a new bug report for each new bug/regression introduced by the 2.6.27 kernel and tag the bug report with 'linux-2.6.27'. Also, please specifically note if the issue does or does not appear in the 2.6.26 kernel. Thanks again, we really appreicate your help and feedback.

Revision history for this message
Joachim Frieben (jfrieben) wrote :

Issue was already solved in Gutsy. A current live CD works nicely, too. Package info:
- kernel 2.6.27-2-generic
- X.Org 1.4.99.906 (1.5.0 RC 6)
I suggest to close this bug as being solved.

Changed in linux:
status: Incomplete → Fix Released
Revision history for this message
Launchpad Janitor (janitor) wrote : Kernel team bugs

Per a decision made by the Ubuntu Kernel Team, bugs will longer be assigned to the ubuntu-kernel-team in Launchpad as part of the bug triage process. The ubuntu-kernel-team is being unassigned from this bug report. Refer to https://wiki.ubuntu.com/KernelTeamBugPolicies for more information. Thanks.

Changed in linux:
importance: Unknown → Medium
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.