[RS482] wine screen updates faulty in Jaunty

Compiz has always messed up Wine: https://bugs.edge.launchpad.net/ubuntu/+source/wine/+bug/237209

Thanks, Scott. They play together just fine in Intrepid, at least on my machine (same one as above).

I've tried turning off various effects, but haven't found a sub-set that restores Wine to normality in Jaunty. Now that Compiz seems to be "on" by default it would be good if Wine was working correctly.

Is anything happening? Wine and programs running under Wine in Jaunty continue to behave strangely. Some of the strange behavior (progressive failure of the Wine "desktop" window to update correctly, leaving blue patches in place of card images in Spider, for instance) does appear to be due to some interaction or failure of interaction with compiz, but even without compiz effects enabled Wine still doesn't feel as if it's working right under Jaunty. The problems with wineconfig and with browsing "C:" seem to have been fixed.

Strong contrast with Intrepid - Wine and programs in it work fine in Intrepid whether compiz is enabled or not.

Running (for instance) wine spider.exe in a terminal raises some alsa configuration errors, but no helpful terminal output.

Any chance of some workaround short of turning off compiz?

metacity --replace & wine foo.

That looks as if it switches window manager to metacity - thus turning off compiz.

If there is a general problem with Wine in jaunty, let's talk about it and help to get it fixed.

If there is a general problem with compiz in Jaunty, ditto.

If the problem is some esoteric interaction between them, ditto.

I'm sorry if my simple-minded approach is offending anyone - I thought the purpose of testing a new pre-release was to spot little niggles like this so the developers could fix them before the final release. I've drawn attention to this one - it's real, it's reproducible, at least on my machine, and it's minor - not really a show-stopper.

If anyone needs more information about what's going on they could:

a) ask me for logs, etc

b) try running both wine and compiz in Jaunty and see if they get the same effects as I've described.

Or just ignore it, in which case I will find Jaunty less satisfactory than Intrepid, where wine and compiz work together just fine.

Further information:

I get the same errors in screen re-drawing in Wine even with compiz disabled, i.e. using the "Normal" setting in Visual Effects tab of the Appearance Preferences. This suggests to me that the errors occur with any window manager that tries to use compositing, including metacity.

Once I reset the Visual Effects setting to "None", which I believe uses metacity without compositing, I get normal screen re-drawing in Wine.

I'm using amd64, kernel 2.6.28-7-generic, and the open-source ati/radeon drivers xserver-xorg-video-ati and xserver-xorg-video-radeon, on an old hp pavilion with the Radeon Xpress 200M 5955 chipset. In Jaunty the drivers are version 1:6.10.0-1ubuntu1. I'm attaching the glxinfo output for the situation when Wine is redrawing correctly.

Very disappointed to see no response! I have upgraded to the Wine 1.1.19 beta, but no change in this unacceptable behavior. Jaunty is up to date, now in beta; Wine is the latest available version, and the problem persists. It is NOT a problem with compiz, as it occurs with metacity/compositing as the window manager.

Hello Stephen,

You may not know it, but we switched to X.org server 1.6 in Jaunty, which may have, as a result, an increase of instability for the GPUs. This issue is, imho, nonely related to Wine, and you're likely to run into same trouble with any OpenGL game, because Compiz + openGL have always been messy, and because we use a new x.org server, and because ATi does take time to make good and stable drivers.

The only ways to make this significantly better would be to contribute in making great free drivers (since we can't contribute on proprietary ones), and find out a way to make compiz (and other composite wms, such as XFCE/KDE ones) behave more smoothly with OpenGL apps (obviously i don't know what's wrong with it, but remember compiz fusion is still under development).

I'm sorry to say it, but i don't think this is going to improve in a close future, and there is nothing the people at wine can do to change that.

Sorry I couldn't figure this one out sooner Stephen, but unfortunately I believe Steve is right that this is a binary driver bug.

You could both be right - should I file another bug against xserver-xorg 1.6 itself?

I don't play many games - mostly the standard gnome set, which don't show any problems for me. The only one I use Wine for is spider solitaire. From what you say, this might be a problem with the composite extension in the x-x-v-ati driver, which should show up with other OpenGL apps. glxgears shows a normal output screen, but a low frame rate, with or without compiz. Compiz without Wine/spider works fine, smooth rotating cube, fire effect, all that jazz.

Any suggestions about what else I can check - combinations of video settings/effects/apps that might show problems?

Hello,

You might want to try OpenArena / ET:QW / Nexuiz in order to see if you have a big FPS difference between Intrepid and Jaunty, this may help to confirm the problem.

That would not work for me - I don't want my machine to play violent shooting games. I want it to run regular everyday apps without showing video artefacts and crashing. If OpenGL is only useful for fast-action games it doesn't bother me whether it works or not; but please would developers not try to use it for anything else in that case. If it is, on the other hand, a general purpose video layer that everyone uses, then I believe it is important that it does that well with the latest and greatest xserver-xorg, the cutting-edge ubuntu, and the OS drivers.

Please attach your Xorg.0.log, and the output of 'lspci -vvnn'. If you've made any customizations to your xorg.conf, attach that as well.

One point of confusion we need cleared up: This bug is filed against the open source -ati driver, but some comments are referring to the proprietary driver (-fglrx). If you could, please clarify which driver you are using.

Also, if you have had -fglrx installed previously but are now using -ati, be aware that the latter can cause some incompatibilities with the former if not properly uninstalled. See the guide at http://wiki.ubuntu.com/X/Troubleshooting if this could be the situation; it includes instructions for how to rectify it.

Hello Bryce,

I'm the one who referred to proprietary drivers but I was probably wrong, i really can't recognise ATi prop/free drivers' names, I apologize for that.

Cordially, SD.

On Sat, Apr 11, 2009 at 09:42:49PM -0000, Steve Dodier wrote:
> Hello Bryce,
>
> I'm the one who referred to proprietary drivers but I was probably
> wrong, i really can't recognise ATi prop/free drivers' names, I
> apologize for that.

No prob, once you attach your Xorg.0.log that should tell. Also, do
refer to the troubleshooting page to make sure you don't have leftover
fglrx bits installed - they can sometimes cause problems not unlike
what's described here.

Bryce

I am using the opensource drivers - have been exclusively since early testing Intrepid, AFAIK. The current Jaunty install was upgraded from Intrepid last week, after beta release, and updated daily since.

Here is lspci -vvnn, and Xorg.0.log as an attachment:

sc@sc-laptop:~$ lspci -vvnn
00:00.0 Host bridge [0600]: ATI Technologies Inc RS480 Host Bridge [1002:5950]
 Subsystem: Hewlett-Packard Company Device [103c:3085]
 Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
 Status: Cap- 66MHz+ UDF- FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort+ >SERR- <PERR- INTx-
 Latency: 64

00:01.0 PCI bridge [0604]: ATI Technologies Inc RS480 PCI Bridge [1002:5a3f]
 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
 Status: Cap+ 66MHz+ UDF- FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
 Latency: 64
 Bus: primary=00, secondary=01, subordinate=01, sec-latency=64
 I/O behind bridge: 00009000-00009fff
 Memory behind bridge: b0100000-b01fffff
 Prefetchable memory behind bridge: 00000000c0000000-00000000cfffffff
 Secondary status: 66MHz+ FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- <SERR- <PERR-
 BridgeCtl: Parity- SERR- NoISA+ VGA+ MAbort- >Reset- FastB2B-
  PriDiscTmr- SecDiscTmr- DiscTmrStat- DiscTmrSERREn-
 Capabilities: <access denied>
 Kernel modules: shpchp

00:04.0 PCI bridge [0604]: ATI Technologies Inc RS480 PCI Bridge [1002:5a36]
 Control: I/O- Mem- BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
 Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
 Latency: 0, Cache Line Size: 32 bytes
 Bus: primary=00, secondary=02, subordinate=02, sec-latency=0
 Secondary status: 66MHz- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- <SERR- <PERR-
 BridgeCtl: Parity- SERR- NoISA+ VGA- MAbort- >Reset- FastB2B-
  PriDiscTmr- SecDiscTmr- DiscTmrStat- DiscTmrSERREn-
 Capabilities: <access denied>
 Kernel driver in use: pcieport-driver
 Kernel modules: shpchp

00:13.0 USB Controller [0c03]: ATI Technologies Inc IXP SB400 USB Host Controller [1002:4374] (prog-if 10)
 Subsystem: ATI Technologies Inc IXP SB400 USB Host Controller [1002:4374]
 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
 Status: Cap+ 66MHz+ UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
 Latency: 64, Cache Line Size: 64 bytes
 Interrupt: pin A routed to IRQ 19
 Region 0: Memory at b0000000 (32-bit, non-prefetchable) [size=4K]
 Capabilities: <access denied>
 Kernel driver in use: ohci_hcd

00:13.1 USB Controller [0c03]: ATI Technologies Inc IXP SB400 USB Host Controller [1002:4375] (prog-if 10)
 Subsystem: ATI Technologies Inc IXP SB400 USB Host Controller [1002:4375]
 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
 Status: Cap+ 66MHz+ UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
 Latency: 64, Cache Line Size: 64 bytes
 Interrupt: pin A routed to IRQ 19
 Region 0: Memory at b0001...

I checked, but could only find fglrx-modaliases; I've removed that to be quite sure. The dpkg -l '*fglrx*' call gave a list including the xorg-driver-fglrx and fglrx-driver, but they don't appear to be installed (no version #,flagged pn and un respectively. apt-get remove fglrx* didn't find them.

Marking as invalid in compiz:
- see #237209 in wine for the desktop effects / wine issue
- follow this bug in xserver-xorg-video-ati for the probable driver issue

Thanks Julien; there doesn't seem to have been any progress on #237209, does there? I have confirmed that the problem is not related to compiz itself; it happens with metacity set to use compositing. So there is a problem with the Composite extension, that shows itself in Wine for some reason on my system. Interestingly it doesn't show up with Photoshop 7 in wine, as far as I can see, only with the spider solitaire game!

I believe I have found the problem. The "empty" xorg.conf results in my case in the video driver (radeon) trying to use exa acceleration, not uxa, which I believe is the new default. Adding the Option "AccelMethod" "XAA" line to the device section removes the problem. I haven't tried explicitly requesting "UXA" instead - maybe that would work.

Nope - setting Option "AccelMethod" "UXA" results in the driver defaulting to EXA. It seems I shall have to continue to specify XAA acceleration......

Further checking: comparing Intrepid and Jaunty on same machine -

both using radeon OS drivers, most recent versions in each case;

Intrepid - specifying either XAA or EXA in Device section of xorg.conf (or leaving no option specified) gives normal behavior in the specific situation I described - Compiz or metacity+compositing as window manager, spider.exe running in Wine.

Jaunty - specifiying XAA gives normal behavior as above. Specifying EXA or leaving no option specified gives the faulty screen-redraws originally described.

Please let me know if any further information is required/helpful; it is disconcerting to send the info asked for and see the bug still marked Incomplete.

"UXA" is only applicable for the -intel driver. -ati only supports "XAA" and "EXA".

Yes, "EXA" is now the default for -ati starting with Jaunty.

Pardon for your discomfort; I simply have been completely focused on intel problems the last couple months. With over 1800 X bugs it is hard to give immediate attention to every bug. I need to learn how Santa Claus does it...

Thanks for getting back to this one. I have added the following two Option lines in xorg.conf, and they allow me to use EXA:

Section "Device"
 Identifier "Configured Video Device"
 Option "MigrationHeuristic" "smart"
 Option "EXANoComposite" "Yes"
EndSection

This works in Jaunty and also in Karmic - Wine screen updates complete correctly even when compiz is running.

Hope that helps

[This is an automatic notification.]

A new version of the -ati driver is now available in Karmic.

This is a significant update to -ati which brings in kernel mode-setting
(currently disabled) and scores of fixes for DRI2, EXA, etc.

I've posted the new version of this driver to the following PPA,
would you mind testing it and seeing if it resolves the bug you
reported?

  https://edge.launchpad.net/~bryceharrington/+archive/ppa/+sourcepub/709908/+listing-archive-extra

If you're not running this release of Ubuntu, you can try booting the Karmic
LiveCD and loading the PPA onto it, and then log out/in to restart X.
ISOs are available at http://cdimages.ubuntu.com/releases/

After testing Karmic, report back here whether it's still an issue or not,
and if it is please post a fresh Xorg.0.log and 'dmesg' output.

Note there could be new bugs... please file these as new reports using
the command 'ubuntu-bug xorg'.

Thanks, Bryce.

I have installed the amd64 versions from your ppa in a fresh Karmic install, and there are no apparent problems. That's good!

However, that particular Karmic install was working fine with the repo versions of the drivers, even without the extra lines I had had to add to xorg.conf in May to get Jaunty working correctly. In fact, as a fresh install, it has no xorg.conf at all, so it was using EXA by default without any options.

Will the ppa drivers work in Jaunty? I could try installing them and checking to see if the extra option lines are still needed.....

I have now tried the new drivers with an older Karmic, upgraded from jaunty, that still has xorg.conf and uses it to set the options I mentioned above (EXANoComposite=yes, MigrationHeuristic=smart).

The new versions work whether the options are included or not; the difference is apparent in Xorg.0.log, but does not seem to alter behavior. (Before the new versions, disabling those options gave the behavior I was complaining about).

That means the workaround (including options in xorg.conf) is no longer necessary.

Thanks!

Hi Stephen, thanks for the swift feedback! To answer some of your questions...

I don't know if the ppa drivers would work in Jaunty. Maybe once the driver's been deployed for karmic I'll look into porting it if it doesn't have messy depends. I think the majority of the changed code will have no effect on Jaunty (except to potentially introduce new bugs) but might be some relevant fixes.

The no-xorg.conf situation is the new default for Karmic. The code paths are slightly different whether you're using a blank xorg.conf vs. none at all, but that probably doesn't matter here.

I checked the possibility of installing in Jaunty - the new driver packages require xserver-xorg-core >= 2:1.6.2, but Jaunty only goes as far as 2:1.6.0. So that's out for now.

I'm curious as to why and when the Karmic upgrades made it possible for my system to work without EXANoComposite, which clearly happened some time before Alpha4, when I did the fresh install (without xorg.conf).

Anyway, it works for me now.

This bug was fixed in the package xserver-xorg-video-ati - 1:6.12.99+git20090825.fc74e119-0ubuntu1

---------------
xserver-xorg-video-ati (1:6.12.99+git20090825.fc74e119-0ubuntu1) karmic; urgency=low

  * Checkout from git 20090825 (master branch) up to commit
    fc74e1194c980d978667e02c60a29a761a694bde
    + Adds DRI2 / redirected direct rendering
      (LP: #96991)
    + Fix freeze on opengl games (bzflag, alienarena) (when KMS on)
      (LP: #348450)
    + Fix intermittent short screen blanking behavior
      (LP: #310864)
    + Fix faulty wine screen updates with EXA Compositing
      (LP: #314205)
    + Fix screen corruption issues when switching windows
      (LP: #406731)
    + Fix SIGSEGV crash in drmCommandNone()
      (LP: #352567)
    + Fix smeared/tearing of display on reboot
      (LP: #367741

  [Tormod Volden]
  * 199_add_git_version_to_log.diff: Log git commit id in RadeonPreInit()
  * sed -i s/DRI2BufferPtr/DRI2Buffer2Ptr/ src/radeon_dri2.c
      so it compiles against xserver 1.6.2.+

  [Bryce Harrington]
  * Restore patch system and quilt to build-depends
  * Drop patches already present upstream
    + 107_check_unsupported_composit_ops.patch
    + 108_quirk_agpmode_m6_ali.patch: AGPMode quirk.
    + 109_quirk_agpmode_m7_intel.patch: AGPMode quirk.
    + 110_quirk_agpmode_r420_sis.patch: AGPMode quirk.
    + 111_use_xaa_for_lowmem_or_nodri.patch

 -- Bryce Harrington <email address hidden> Sat, 29 Aug 2009 12:16:53 -0700