[Bug 608256] New: Radeon driver does not support special desktop effects on ATI 9200PRO
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256 http://bugzilla.novell.com/show_bug.cgi?id=608256#c0 Summary: Radeon driver does not support special desktop effects on ATI 9200PRO Classification: openSUSE Product: openSUSE 11.3 Version: Milestone 7 Platform: i686 OS/Version: openSUSE 11.3 Status: NEW Severity: Normal Priority: P5 - None Component: X.Org AssignedTo: bnc-team-xorg-bugs@forge.provo.novell.com ReportedBy: lee_matheson@hotmail.com QAContact: xorg-maintainer-bugs@forge.provo.novell.com Found By: --- Blocker: --- User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.3) Gecko/20100401 SUSE/3.6.3-1.2 Firefox/3.6.3 Special desktop effects do not work with the Radeon driver when booting from the 32-bit openSUSE-11.3 M7 (build 0625) live CD to an ATI 9200PRO hardware.
From lspci -nnk : 01:00.0 VGA compatible controller [0300]: ATI Technologies Inc RV280 [Radeon 9200 PRO] [1002:5960] (rev 01) Subsystem: Club-3D BV Device [196d:1052]
I note this in the /var/log/Xorg.0.log file: [ 56.393] (--) RADEON(0): Chipset: "ATI Radeon 9250 5960 (AGP)" (ChipID = 0x5960) .. [ 56.393] (II) RADEON(0): AGP card detected ... [ 59.032] (EE) RADEON(0): [dri] RADEONDRIGetVersion failed because of a version mismatch. [dri] This chipset requires a kernel module version of 1.8.0, [dri] but the kernel reports a version of 2.3.0.[dri] If using legacy modesetting, upgrade your kernel. [dri] If using kernel modesetting, make sure your module is [dri] loaded prior to starting X, and that this driver was built [dri] with support for KMS. [dri] Disabling DRI. and if the PC is rebooted and the "nomodeset" boot code inserted, the radeon driver is still loaded, and one gets an all black background after X boots, with the task bar present, and special desktops sort of enabled (for example the cube rotation works, although there are no icons on black background of screen). I can confirm it is the radeon driver the "nomodeset" boot code (as confirmed by the /var/log/Xorg.0.log file). As soon as I applied a desktop background (to remove the black and show a different colour and display the icons) the special desktop effects were disabled with a dialog box noting compositing was too slow. But the proper desktop comes back with special desktop effects disabled. For the "nomodeset case" I note this in the /var/log/Xorg.0.log file: [ 55.744] (--) RADEON(0): Chipset: "ATI Radeon 9250 5960 (AGP)" (ChipID = 0x5960) .. [ 55.744] (II) RADEON(0): AGP card detected .... [ 56.946] (II) RADEON(0): [dri] Found DRI library version 1.3.0 and kernel module version 1.33.0] I will copy complete Xorg.0.log files in separate posts. I also have /var/log/message and dmesg output for each of these two boot cases (one with and one without) that I will attach ONLY upon request if believed relevant. Some background: Special desktop effects did work with the Radeon driver on openSUSE-11.1 and then briefly on openSUSE-11.2 with the Xorg : X11 repository Mesa and xorg-x11-driver-video rpm (that is until the Xorg : X11 repository got too cutting edge). Reproducible: Always Steps to Reproduce: 1.boot PC in one case without "nomodeset" and in the other case with "nomodeset" 2. 3. Actual Results: No special desktop effects with "nomodeset". With "nomodeset" there are special desktop effects but desktop has black background with no icons. Expected Results: Expect proper 3D special desktop effects -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c1
--- Comment #1 from Lee Matheson
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c2
--- Comment #2 from Lee Matheson
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c3
Stefan Dirsch
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c4
--- Comment #4 from Lee Matheson
Well, Radeon 9200 is becoming old and apparently nobody is testing the support for it any longer upstream. It's unlikely we'll see a fix for that ...
I confess to being a bit more hopeful :) I tested today's Fedora-13 release on this same PC, using their 32-bit KDE liveCD. Special desktop effects work ok on it, using the Radeon driver. Cube rotation, etc ... Running top, cpu load was low ( less than 10% during rotations). I have copied here the Fedora Xorg.0.log file. What else can I provide, so that we can see what they are doing different? I note Fedora13 have mesa-dri-driver-7.8.1-6, mesa-libGL-7.8.1-6, mesa-libGLU-7.8.1-6 rpms and xorg-x11-drivers-7.3.8.1-6 rpms. Their packaging is different from openSUSE, so I don't know what that means. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c5
--- Comment #5 from Stefan Dirsch
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c6
--- Comment #6 from Lee Matheson
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c7
--- Comment #7 from Lee Matheson
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c8
--- Comment #8 from Lee Matheson
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c9
Jan Bouwhuis
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c10
--- Comment #10 from Stefan Dirsch
I can confirm this problem with the Radeon X1270 (RS690) using the Gnome Live CD 64 bit and a Radeon 7000 (R100) using the Gnome Live CD 64 bit. Both live CD's are 11.3 M7. Work-a-round is adding 'radeon.modeset=0' as startup parameter.
But that are completely different GPUs than the R2xx based GPU on Radeon 9200. BTW, with radeon.modeset=0 radeonhd will be used on GPUs, which are supported by that driver (RS690 in your case).
The radeonhd driver will appear for the Radeon X1270 ( RS690 ) but crash. By explictly assigning the radeon driver (remove the #) and restarting X the Radeon driver takes over and works correctly with 3D.
An alternative is to uninstall the xorg-x11-driver-video-radeonhd package.
On the R100 system adding the radeon.modeset=0 gives a working 3D Why is still the radeonhd driver used as default fallback?
That's not correct. radeonhd is still our default. radeon is the fallback, if the GPU is not supported by radeonhd driver or if KMS is active (and therefore radeonhd driver bailed out). KMS is not supported by radeonhd driver at the moment. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c11
Lee Matheson
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c12
--- Comment #12 from Lee Matheson
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c13
--- Comment #13 from Lee Matheson
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c14
--- Comment #14 from Lee Matheson
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c15
--- Comment #15 from Lee Matheson
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c16
Stefan Dirsch
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c17
Lee Matheson
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c18
Stefan Dirsch
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.novell.com/show_bug.cgi?id=608256
http://bugzilla.novell.com/show_bug.cgi?id=608256#c19
--- Comment #19 from Lee Matheson
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
https://bugzilla.novell.com/show_bug.cgi?id=608256
https://bugzilla.novell.com/show_bug.cgi?id=608256#c20
--- Comment #20 from Lee Matheson
Unfortunately we have no ressources to work on radeon driver bugs. Hence the issue needs to be fixed upstream. You can report the issue via the ATI mailing list
http://lists.freedesktop.org/mailman/listinfo/xorg-driver-ati
or X.Org's bugzilla
https://bugs.freedesktop.org/ (product: xorg, component: Driver/Radeon)
I managed to obtain 3D support (ie special desktop effects, cube rotation, etc
.. ). I have been running for 1 hour thus far. It appears a bit slow, but
that is subjective. I note there is a bug report on the 9200pro running slow
with 3D on bugs.freedesktop.org but I do not know if that is the same.
There is alot of good information on those links you provided (thankyou very
much for those), albeit searching takes time. Based on what I read on various
posts and bugs on those links, I discovered that specifying 'nomodeset' as a
boot argument, together with the following edit to the
/etc/X11/xorg.conf.d/50-device.conf file, restores/fixes the broken 3D support
(with KMS, I think - see below) :
Section "Device"
Identifier "Default Device"
Driver "radeon"
## Required magic for radeon/radeonhd drivers; output name
## (here: "DVI-0") can be figured out via 'xrandr -q'
#Option "monitor-DVI-0" "Default Monitor"
#oldcpu added following 2 lines
Option "BusType" "PCI"
Option "AGPSize" "64"
EndSection
As noted, I successfully tested the above on a 32-bit openSUSE-11.3.
While I note this bug report is closed, I will attach a copy of the
/var/log/Xorg.0.log file 'for prosperity' in case anyone is curious (in a
subsequent post). While I obtained this solution from lists.freedesktop.org
and bugs.freedesktop.org, I still have not determined if there is an
appropriate bug raised in bugs.freedesktop.org as the threads with this
information were massively confusing to myself.
I also note that 'nomodeset' works but that specifying yast > System >
/etc/sysconfig Editor > System > Kernel > NO_KMS_IN_INITRD and change it to
"yes" DOES NOT work. This value must be No.
Note the radeon 9200 pro hardware I am using is (from "lspci -vv") :
01:00.0 VGA compatible controller: ATI Technologies Inc RV280 [Radeon 9200 PRO]
(rev 01) (prog-if 00 [VGA controller])
Subsystem: Club-3D BV Device 1052
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR- FastB2B- DisINTx-
Status: Cap+ 66MHz+ UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
https://bugzilla.novell.com/show_bug.cgi?id=608256
https://bugzilla.novell.com/show_bug.cgi?id=608256#c21
--- Comment #21 from Lee Matheson
participants (1)
-
bugzilla_noreply@novell.com