Red Hat Bugzilla – Bug 813949
radeon KMS is broken on big-endian
Last modified: 2014-02-05 06:56:50 EST
Description of problem:
When loading the radeon driver on my Apple iBook (32bit ppc), it tries to use xaa accelleration. This fails, since the xaa module is removed from the xorg X server.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. Start X with no other config than what's provided by the package
The radeon automatically tries to load xaa:
(==) RADEON(0): Using XAA acceleration architecture
X then dies with
(EE) Failed to load module "xaa" (module does not exist, 0)
Radeon driver choosing an existing acceleration mode
Workaround: Tell radeon to use the exa acceleration mode instead. Something like this in /etc/X11/xorg.conf.d/radeon_fix.conf :
Option "AccelMethod" "EXA"
Output from lspci -vv
0000:00:10.0 VGA compatible controller: ATI Technologies Inc Radeon Mobility M7 LW [Radeon Mobility 7500] (prog-if 00 [VGA controller])
Subsystem: ATI Technologies Inc Radeon Mobility M7 LW [Radeon Mobility 7500]
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: 255 (2000ns min), Cache Line Size: 32 bytes
Interrupt: pin A routed to IRQ 48
Region 0: Memory at 98000000 (32-bit, prefetchable) [size=128M]
Region 1: I/O ports at 0400 [size=256]
Region 2: Memory at 90000000 (32-bit, non-prefetchable) [size=64K]
Expansion ROM at 90020000 [disabled] [size=128K]
Capabilities:  AGP version 2.0
Status: RQ=48 Iso- ArqSz=0 Cal=0 SBA+ ITACoh- GART64- HTrans- 64bit- FW- AGP3- Rate=x1,x2
Command: RQ=8 ArqSz=0 Cal=0 SBA+ AGP+ GART64- 64bit- FW- Rate=x2
Capabilities:  Power Management version 2
Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
Kernel driver in use: radeonfb
Kernel driver in use: radeonfb
Booooo. You should be using KMS. Why aren't you using KMS? That line should say "radeon" not "radeonfb".
Booting with KMS gives color distrotions, and the machine freezes after a few minutes. Running with the boot option "nomodeset", it works fine (with the workaround described above, that is).
That KMS does not work properly is perhaps worth another bug, though.
I'd still say this is an xorg radeon bug.
In the sense that you're ending up on a path that shouldn't be possible, sure. In 6.14.3-12 and later the non-KMS path isn't even built in the radeon X driver, and you're just going to fall back to fbdev.
Updating description to reflect the real issue.
But this is not a xorg-x11-drv-ati related issue. The last working kernel is 3.1.5-1 and every latter one leads to the issue described in comment #2 .
Latest working means that it works with this
Option "AccelMethod" "EXA"
May or may not be relevant. Running with the setup above, I get artifacts and picture distortion on anything that's not basic Xlib (xterm works). See attached screenshot.
And btw, using fbdev, I get a color palette of 8 colors or something like that, rendering the screen quite unusable.
So, what's next? Just wait for the kernel to catch up?
Created attachment 580797 [details]
Photograph of screen with distorted graphics
Created attachment 583967 [details]
Without KMS using last workign kernel
This is the X.org's log with the last working kernel - 3.1.5
Created attachment 583968 [details]
With KMS (and with bugs and distortions)
And this is the Xorg's log wuth the latest (broken) kernel. It produces garbled colors right after the start, wrong display size (800x600 instead of 1920x1200) and the X locks after a few seconds (normally 30-60 seconds after start)
After a bit of investigation I think that this bug is in fact a superposition of a several different bugs.
For example with the kernels with version 3.1.5+ X.org wrongly believes that I've got a S-video output connected. Can we fix somehow?
I'm updating Xorg logs with 3.1.5 and with 3.5.0 latest kernels. See diff -u between them for the details.
Created attachment 587002 [details]
Without KMS using last working kernel (3.1.5)
Created attachment 587003 [details]
With KMS (bugs, S-video, and distortions) using kernel 3.5.0
Created attachment 587005 [details]
diff -u between Xorg logs with nokms (works) and kms (doesn't work)
Just a sidenote: Font rendering through gtk(?) or pango(?) trigs this bug. Other widget sets are not affected. This means that apps/desktop environments using Qt (Razor-qt, KDE), or even Firefox (why?) are not affected.
My workaround for now is just using Razor-qt.
On fedora 18, with version 7.0 of the driver, UMS is disabled, and with KMS, it's impossible to get a stable display at all. Color distortions and crashes as soon as I start using the display for anything but running xdm.
Recompiling xorg-x11-drv-ati-6.14.4-6.20120602git930760942.fc17, reenabling UMS (removing patch #14), and the display is stable again.
Comment 3 suggests that this should not be possible. Well, I can assure that this works, and it's not the fbdev driver.
Which side of KMS is trigging the bug? The kernel or the xorg driver?
With this, I can run X and Qt apps fine. GTK apps still have distortions, though. So at least, I have a working display on my iBook as long as I can recompile version 6.x of the driver against a current version of Fedora.
See also https://bugs.freedesktop.org/show_bug.cgi?id=55745
FYI, with a small patch, I was able to recompile xorg-x11-drv-ati-6.14.4-6.20120602git930760942 against f19 xorg, and it works with the workarounds and limitations mentioned in Comment 14.
This message is a reminder that Fedora 18 is nearing its end of life.
Approximately 4 (four) weeks from now Fedora will stop maintaining
and issuing updates for Fedora 18. It is Fedora's policy to close all
bug reports from releases that are no longer maintained. At that time
this bug will be closed as WONTFIX if it remains open with a Fedora
'version' of '18'.
Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, simply change the 'version'
to a later Fedora version prior to Fedora 18's end of life.
Thank you for reporting this issue and we are sorry that we may not be
able to fix it before Fedora 18 is end of life. If you would still like
to see this bug fixed and are able to reproduce it against a later version
of Fedora, you are encouraged change the 'version' to a later Fedora
version prior to Fedora 18's end of life.
Although we aim to fix as many bugs as possible during every release's
lifetime, sometimes those efforts are overtaken by events. Often a
more recent Fedora release includes newer upstream software that fixes
bugs or makes them obsolete.
Fedora 18 changed to end-of-life (EOL) status on 2014-01-14. Fedora 18 is
no longer maintained, which means that it will not receive any further
security or bug fix updates. As a result we are closing this bug.
If you can reproduce this bug against a currently maintained version of
Fedora please feel free to reopen this bug against that version. If you
are unable to reopen this bug, please file a new report against the
current release. If you experience problems, please add a comment to this
Thank you for reporting this bug and we are sorry it could not be fixed.