[RC410] [TOSHIBA Satellite L35] suspend/resume failure with DRI (Xpress 200M 5a62)

[gene]

Binary package hint: xserver-xorg-video-radeon

Here am I again. Upgraded to jaunty with a little bit of hope to be able to use the open source xserver-xorg-video-radeon instead of odious fglrx (damn ATI!) No luck. Got worse compared to hardy, couldn't login with the radeon 5 times in a row.
Will be trying some options before going back to fglrx. Wish me luck.
PS I have trouble understanding the policy though. My bug # 305301 is still new and undecided.

ProblemType: KernelOops
Annotation: This occured during a previous suspend and prevented it from resuming properly.
Architecture: i386
DistroRelease: Ubuntu 9.04
ExecutablePath: /usr/share/apport/apportcheckresume
Failure: suspend/resume
InterpreterPath: /usr/bin/python2.6
Lsusb:
 Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
 Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
 Bus 001 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
MachineType: TOSHIBA Satellite L35
Package: linux-image-2.6.28-8-generic 2.6.28-8.28
ProcAttrCurrent: unconfined
ProcCmdLine: root=UUID=8ed12067-caa1-4af3-b3b4-c97a2fe5cc3d ro quiet splash
ProcCmdline: /usr/bin/python /usr/share/apport/apportcheckresume
ProcEnviron: PATH=(custom, no user)
ProcVersionSignature: Ubuntu 2.6.28-8.28-generic
SourcePackage: linux
StressLog: Error: [Errno 2] No such file or directory: '/var/lib/pm-utils/stress.log'
Tags: resume suspend
Title: [TOSHIBA Satellite L35] suspend/resume failure
UserGroups:

[gene (eugenios)]

Liked this novelty in jaunty: semi-automated crash report!
Well done!

[gene (eugenios)]

I am not sure what is the definition of a "duplicate bug". There is no question that this bug is related to #305301.
If anyone could elucidate me on that, I would appreciate. However, from my experience here at launchpad, nobody talks to you...

[gene (eugenios)]

Should correct myself: in my first post I meant not "hardy heron", but "intrepid ibex". In hardy I had no 3dc functionalities, but had no problems like this one

[gene (eugenios)]

Turns out, when specifying "Option GARTSize 128" in the xorg.conf file, the Xserver turns off dri which solves the described problem (from the X log):
[ 0.436460] (EE) RADEON(0): [pci] Out of memory (-12)
[ 0.436498] (EE) RADEON(0): [pci] PCI failed to initialize. Disabling the DRI
I can perfectly suspend/resume and login without any lock-ups.

It goes without saying though, that no 3d/compiz can be used. So it is not a fix, but an option.

It is strange, since according to the specs of RC410 (Radeon X200M), it can go up to 128MB VRAM.

I attach my xorg.conf file. It should be pointed out, that when trying different from "128" values for GARTSize (64,32) gives the described problem , as well as without Option "AGPMode" "2" or "1"

[Tormod Volden (tormodvolden)]

Please try with a default, or no xorg.conf, and attach the new Xorg.0.log.

[gene (eugenios)]

Thanks Tormod,
Yes I did that. I had 4 system crashes after I removed my xorg.conf . I am attaching my logs (I did it many times). Do you need any other logs? I also attach my kernel log below. Thanks

[Tormod Volden (tormodvolden)]

To turn off DRI, use Option "DRI" "off".
Can you please try other values for AGPMode, like 2 and 4?

Things _should_ work well with the default xorg.conf, otherwise file bugs. Sometimes "cruft" in xorg.conf has negative impact.

[gene (eugenios)]

Thank you Tormod!
I have tried all of that already! In my current xorg.conf file I do have
Option "DRI" "off"
I tried different values for AGPMode, like 1,2,4,8 - it gave me the same:
- frequent/complete system lockups when starting X server (with gdm)
- NOWAY to resume from suspend, if I can get it started

However, I will try different values for AGPMode (without DRI off)

[Tormod Volden (tormodvolden)]

Are you saying you get the crashes even with DRI off?

In that case, AGPMode won't help. Looking closer, I don't think you have an AGP card either :)

[gene (eugenios)]

I get no crashes with DRI turned off. I can resume from suspend (seemingly ) without any problem (however, as I speak, apparently I have no vt running on my machine due to one of suspenses, I feel though that this is yet another problem with the marvelous ATI card, not related to this one)

[gene (eugenios)]

Well, when I change AGPMode value it does not make any difference (dri on or off), so your statement about my card does make sense to me.

[jetdog (slicksterdave)]

Hey all... just a tip:

* The fglrx driver works amazing currently. Make sure you use the version that still includes support for the xpress 200m - I have been using it for two years and it works wounders. (Also, this will likely be your last chance to experience aiglx with opengl 3.0 - since >= 2.6.29 is no longer supported on R200-R500 series ati cards via their proprietary driver - meaning this october you will be stuck with the radeon driver once again).

The proprietary driver had no suspend issues, with catalyst 9.3 (last support ever for < 2.6.29 (i.e. THIS version of ubuntu)) - and had amazing 3d support. Remember to use the aticonfig tool on a fresh Xorg.conf if you were to try and get this driver working.

(Information for x11-drivers/xf86-video-ati, version 6.12.2 - on xorg-x11-7.2):

I'm actually having problems with suspend on the x200m card too (only with radeon driver) - Although, on my gentoo system with the above versions, the system doesn't hang, but instead I get a friendly dialog from gnome indicating that the suspend failed. This kind of bug tends to lie within xorg's domain, which is difficult support to get here on launchpad, on a hardware-specific basis, without going right to xorg.

http://www.x.org/wiki/radeon

The above link contains some documentation and support-related links for the radeon driver. Currently, on their website, it says the the status of an R400-series's suspend ability is 'unknown'. A quick google indicates that all versions of the X200, including intel, amd, and mobile variants, are R400-series cards. If I get any further, I'll be sure to let you know.

Uploaded, is my Xorg.conf for the radeon driver. Again this is =NOT= an ubuntu version, but I would suggest that if you're trying to get somewhere with crash-related issues, that you try and tweak based on the information I have (or have not) put in there.

Please be aware that the radeon driver is actually pretty good at guessing most defaults (as evident by how much I have left out)

Also, good news: apparently, there are some improvements coming for this driver, with the arrival of 2.6.31, according to phronix.