|
Bugzilla – Full Text Bug Listing |
| Summary: | radeon/radeonhd [R3xx-R5xx] Screen distortion with KMS / Immediate Xserver crash with UMS (DRI related) | ||
|---|---|---|---|
| Product: | [openSUSE] openSUSE 11.4 | Reporter: | David Rankin <drankinatty> |
| Component: | X.Org | Assignee: | Stefan Dirsch <sndirsch> |
| Status: | RESOLVED FIXED | QA Contact: | E-mail List <xorg-maintainer-bugs> |
| Severity: | Critical | ||
| Priority: | P2 - High | CC: | andreas.bolsch, dmuhamedagic, forgotten_ho8rvtClXX, forgotten_U_TY3F7lDx, forgotten_Xl7iAZAFUm, jan, jes, keel_lambert, kkaempf, markus.walser, meissner, mr-kris, per, R.Vickers, ralf.schaefer, revealed, rnelson, robxu9, TLingk |
| Version: | Final | ||
| Target Milestone: | --- | ||
| Hardware: | x86 | ||
| OS: | Other | ||
| See Also: | https://bugs.freedesktop.org/show_bug.cgi?id=33929 | ||
| Whiteboard: | maint:released:11.4:40839 | ||
| Found By: | --- | Services Priority: | |
| Business Priority: | Blocker: | --- | |
| Marketing QA Status: | --- | IT Deployment: | --- |
| Deadline: | 2011-06-07 | ||
| Attachments: |
Screenshot of Bug
dmesg output Xorg.0.log nomodeset Xorg.0.log :( Xorg.0.log 50-device.conf 'Radeon' with nomodeset Screenshot of bug Segfault output + backtrace from startx Xorg.0.log with AIGLX error and no 3d support after start Xorg.0.log with "nomodeset" and no 3d support after start X server |
||
|
Description
David Rankin
2011-03-09 20:53:01 UTC
Created attachment 418405 [details]
dmesg output
This was on a brand new install - NO upgrade. Also, on my 11.3 box, (radeon 9600) I have had X lockup 3 times trying to file this bug report. Something is up. The 11.4 xorg install is: 16:01 alchemy:~> rpm -qa | grep xorg xorg-x11-libxkbfile-32bit-7.6-6.1.x86_64 xorg-x11-libXfixes-32bit-7.6_4.0.5-6.1.x86_64 xorg-x11-fonts-devel-7.6-7.1.x86_64 xorg-x11-libxkbfile-devel-7.6-6.1.x86_64 xorg-x11-libXprintUtil-devel-7.6-2.1.x86_64 xorg-x11-libICE-7.6-8.1.x86_64 xorg-x11-libfontenc-7.6-6.1.x86_64 xorg-x11-libSM-7.6-7.1.x86_64 xorg-x11-libXv-devel-7.6-6.1.x86_64 xorg-x11-libX11-32bit-7.6-10.2.x86_64 xorg-x11-libX11-7.6-10.2.x86_64 xorg-x11-libXfixes-7.6_4.0.5-6.1.x86_64 xorg-x11-xtrans-devel-7.6-6.1.x86_64 xorg-x11-libfontenc-devel-7.6-6.1.x86_64 xorg-x11-libxcb-devel-7.6_1.7-8.1.x86_64 xorg-x11-7.6-42.3.x86_64 xorg-x11-driver-video-nouveau-0.0.16_20110115_b795ca6-3.1.x86_64 xorg-x11-libXau-32bit-7.6_1.0.6-6.1.x86_64 xorg-x11-libXp-32bit-7.6-2.1.x86_64 xorg-x11-libxkbfile-7.6-6.1.x86_64 xorg-x11-libXmu-32bit-7.6-7.1.x86_64 xorg-x11-libXprintUtil-32bit-7.6-2.1.x86_64 xorg-x11-libXv-7.6-6.1.x86_64 xorg-x11-libXmu-7.6-7.1.x86_64 xorg-x11-libXpm-devel-7.6-6.1.x86_64 xorg-x11-libxcb-32bit-7.6_1.7-8.1.x86_64 xorg-x11-libXt-7.6_1.0.9-4.1.x86_64 xorg-x11-util-devel-7.6-8.1.x86_64 xorg-x11-libXprintUtil-7.6-2.1.x86_64 xorg-x11-libX11-devel-7.6-10.2.x86_64 xorg-x11-libX11-ccache-7.6-2.3.x86_64 xorg-x11-libXau-7.6_1.0.6-6.1.x86_64 xorg-x11-libxcb-7.6_1.7-8.1.x86_64 xorg-x11-libfontenc-32bit-7.6-6.1.x86_64 xorg-x11-libXpm-32bit-7.6-6.1.x86_64 xorg-x11-libXext-7.6_1.2.0-2.1.x86_64 xorg-x11-libXp-7.6-2.1.x86_64 xorg-x11-libs-32bit-7.6-16.1.x86_64 xorg-x11-libXt-devel-7.6_1.0.9-4.1.x86_64 xorg-x11-libXfixes-devel-7.6_4.0.5-6.1.x86_64 xorg-x11-server-7.6_1.9.3-15.18.4.x86_64 xorg-x11-driver-video-intel-legacy-2.9.1-8.1.x86_64 xorg-x11-libXrender-32bit-7.6_0.9.6-6.1.x86_64 xorg-x11-libXext-32bit-7.6_1.2.0-2.1.x86_64 xorg-x11-libICE-devel-7.6-8.1.x86_64 xorg-x11-libXp-devel-7.6-2.1.x86_64 xorg-x11-libICE-32bit-7.6-8.1.x86_64 xorg-x11-libSM-32bit-7.6-7.1.x86_64 xorg-x11-libXt-32bit-7.6_1.0.9-4.1.x86_64 xorg-x11-libXrender-7.6_0.9.6-6.1.x86_64 xorg-x11-proto-devel-7.6-17.1.x86_64 xorg-x11-xauth-7.6-42.3.x86_64 xorg-x11-libs-7.6-16.1.x86_64 xorg-x11-libSM-devel-7.6-7.1.x86_64 xorg-x11-driver-video-radeonhd-1.3.0_20100512_80ba041-2.1.x86_64 xorg-x11-libXpm-7.6-6.1.x86_64 xorg-x11-libXau-devel-7.6_1.0.6-6.1.x86_64 xorg-x11-libXdmcp-devel-7.6-7.1.x86_64 xorg-x11-libXrender-devel-7.6_0.9.6-6.1.x86_64 xorg-x11-libXext-devel-7.6_1.2.0-2.1.x86_64 xorg-x11-driver-video-7.6-52.4.x86_64 xorg-x11-libXdmcp-7.6-7.1.x86_64 xorg-x11-fonts-core-7.6-10.1.noarch xorg-x11-libXmu-devel-7.6-7.1.x86_64 xorg-x11-driver-input-7.6-29.1.x86_64 xorg-x11-devel-7.6-16.1.x86_64 xorg-x11-fonts-7.6-10.1.noarch xorg-x11-Xvnc-7.6_1.9.3-15.18.4.x86_64 Created attachment 418411 [details]
Xorg.0.log
Please try without KMS as mentioned in our release notes for openSUSE 11.4. Created attachment 418437 [details]
nomodeset Xorg.0.log :(
Stefan,
I tried nomodeset and it causes X to segfault. Here is the console output from a manual startx:
19:50 zephyr:~/bugs> cat startx-segfault.txt
xauth: file /root/.serverauth.3445 does not exist
xauth: (stdin):2: unknown command "fed36715b8176d4675e4edfa0c86c3f6"
X.Org X Server 1.9.3
Release Date: 2010-12-13
X Protocol Version 11, Revision 0
Build Operating System: openSUSE SUSE LINUX
Current Operating System: Linux alchemy 2.6.37.1-1.2-desktop #1 SMP PREEMPT 2011-02-21 10:34:10 +0100 x86_64
Kernel command line: root=/dev/disk/by-id/ata-WDC_WD6400BEVT-00A0RT0_WD-WX40A9996688-part6 resume=/dev/disk/by-id/ata-WDC_WD6400BEVT-00A0RT0_WD-WX40A9996688-part5 splash=silent quiet vga=0x317 nomodeset
Build Date: 22 February 2011 10:08:09PM
Current version of pixman: 0.20.0
Before reporting problems, check http://wiki.x.org
to make sure that you have the latest version.
Markers: (--) probed, (**) from config file, (==) default setting,
(++) from command line, (!!) notice, (II) informational,
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
(==) Log file: "/var/log/Xorg.0.log", Time: Wed Mar 9 19:41:23 2011
(==) Using config directory: "/etc/X11/xorg.conf.d"
(EE) Failed to load module "fglrx" (module does not exist, 0)
(EE) KMS is disabled. This is good for us, because RADEONHD conflicts with KMS.
(EE) RADEONHD(0): Unusupported PowerPlayInfo Revision
(EE) RADEONHD(0): Power Management: Cannot get known good chip configurations
None
Backtrace:
0: X (xorg_backtrace+0x28) [0x463678]
1: X (0x400000+0x6ae89) [0x46ae89]
2: /lib64/libc.so.6 (0x7f388d3f6000+0x32b30) [0x7f388d428b30]
Segmentation fault at address (nil)
Fatal server error:
Caught signal 11 (Segmentation fault). Server aborting
Hmm. What you also could try is without KMS, but "radeon" driver. --> /etc/X11/xorg.conf.d/50-devices.conf. I've tried with just the radeon driver and just about everything else I could think of :) # Driver "radeon" Driver "radeonhd" # Driver "fbdev" The only way X will start is with KSM and NO xorg.conf. 21:14 alchemy:~> lsmod | grep radeon radeon 984687 2 ttm 74373 1 radeon drm_kms_helper 36694 1 radeon drm 232019 4 radeon,ttm,drm_kms_helper i2c_algo_bit 6246 1 radeon I can unload it, but all I have is the 'radeon' driver, even though it is installed under: [21:22 alchemy:/home/david] # rpm -qa | grep radeon xorg-x11-driver-video-radeonhd-1.3.0_20100512_80ba041-2.1.x86_64 [21:23 alchemy:/home/david] # rpm -ql xorg-x11-driver-video-radeonhd /usr/bin/rhd_conntest /usr/lib64/xorg/modules/drivers /usr/lib64/xorg/modules/drivers/radeonhd_drv.la /usr/lib64/xorg/modules/drivers/radeonhd_drv.so /usr/share/doc/packages/xorg-x11-driver-video-radeonhd /usr/share/doc/packages/xorg-x11-driver-video-radeonhd/COPYING /usr/share/doc/packages/xorg-x11-driver-video-radeonhd/README /usr/share/doc/packages/xorg-x11-driver-video-radeonhd/README_conntest /usr/share/man/man4/radeonhd.4.gz /usr/share/sax /usr/share/sax/api /usr/share/sax/api/data /usr/share/sax/api/data/cdb /usr/share/sax/api/data/cdb/Cards.20.xorg-x11-driver-video-radeonhd /usr/share/sax/sysp /usr/share/sax/sysp/maps /usr/share/sax/sysp/maps/update /usr/share/sax/sysp/maps/update/Identity.map.20.xorg-x11-driver-video-radeonhd /var/lib/hardware /var/lib/hardware/ids /var/lib/hardware/ids/20.xorg-x11-driver-video-radeonhd Is there another just 'ati' driver (the old radeon driver) somewhere I can try? I've searched and all I see is the radeonhd: [21:24 alchemy:/home/david] # zypper se radeon Loading repository data... Reading installed packages... S | Name | <snip> i | xorg-x11-driver-video-radeonhd DUH - I understand your comment now -- I'll try and report back :) OK,
I changed xorg.conf.d/50-device.conf and uncommented:
Driver "radeon"
I then rebooted, cleared the grub boot line and added nomodeset. The results are the same. X segfaults. The console output from startx is:
22:05 zephyr:~/tmp> cat startx-console.txt
xauth: file /root/.serverauth.3332 does not exist
xauth: (stdin):2: unknown command "fed36715b8176d4675e4edfa0c86c3f6"
X.Org X Server 1.9.3
Release Date: 2010-12-13
X Protocol Version 11, Revision 0
Build Operating System: openSUSE SUSE LINUX
Current Operating System: Linux alchemy 2.6.37.1-1.2-desktop #1 SMP PREEMPT 2011-02-21 10:34:10 +0100 x86_64
Kernel command line: root=/dev/disk/by-id/ata-WDC_WD6400BEVT-00A0RT0_WD-WX40A9996688-part6 resume=/dev/disk/by-id/ata-WDC_WD6400BEVT-00A0RT0_WD-WX40A9996688-part5 splash=silent quiet nomodeset
Build Date: 22 February 2011 10:08:09PM
Current version of pixman: 0.20.0
Before reporting problems, check http://wiki.x.org
to make sure that you have the latest version.
Markers: (--) probed, (**) from config file, (==) default setting,
(++) from command line, (!!) notice, (II) informational,
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
(==) Log file: "/var/log/Xorg.0.log", Time: Wed Mar 9 21:57:20 2011
(==) Using config directory: "/etc/X11/xorg.conf.d"
(II) [KMS] drm report modesetting isn't supported.
XRANDR name: VGA-0
Connector: VGA
CRT1: INTERNAL_KLDSCP_DAC1
DDC reg: 0x7e50
XRANDR name: LVDS
Connector: LVDS
LCD1: INTERNAL_LVTM1
DDC reg: 0x7e40
Dac detection success
finished output detect: 0
finished output detect: 1
finished all detect
Dac detection success
Output LCD1 disable success
Blank CRTC 0 success
Disable CRTC 0 success
Blank CRTC 1 success
Disable CRTC 1 success
Output CRT1 disable success
Output LCD1 disable success
Blank CRTC 0 success
Disable CRTC 0 success
Blank CRTC 1 success
Disable CRTC 1 success
Output LCD1 disable success
Blank CRTC 0 success
Disable CRTC 0 success
Set CRTC 0 Source success
Mode 1440x900 - 1760 912 10
Picked PLL 0
best_freq: 96361
best_feedback_div: 323
best_frac_feedback_div: 0
best_ref_div: 4
best_post_div: 12
Set CRTC 0 PLL success
Set CRTC Timing success
Set CRTC 0 Overscan success
Not using RMX
scaler 0 setup success
Set CRTC 0 Source success
crtc 0 YUV disable setup success
Output digital setup success
Output LCD1 enable success
Enable CRTC 0 success
Unblank CRTC 0 success
Output CRT1 disable success
Blank CRTC 1 success
Disable CRTC 1 success
Backtrace:
0: X (xorg_backtrace+0x28) [0x463678]
1: X (0x400000+0x6ae89) [0x46ae89]
2: /lib64/libc.so.6 (0x7f1ec84ed000+0x32b30) [0x7f1ec851fb30]
Segmentation fault at address (nil)
Fatal server error:
Caught signal 11 (Segmentation fault). Server aborting
Please consult the The X.Org Foundation support
at http://wiki.x.org
for help.
Please also check the log file at "/var/log/Xorg.0.log" for additional information.
Output LCD1 disable success
Blank CRTC 0 success
Disable CRTC 0 success
Blank CRTC 1 success
Disable CRTC 1 success
Enable CRTC 0 success
Unblank CRTC 0 success
I'll attache the Xorg.0.log
Created attachment 418444 [details]
Xorg.0.log 50-device.conf 'Radeon' with nomodeset
Stefan - I'm going to switch drives and boot Arch and then I can give you the xorg version where this broke (I think). I also experienced this crash with 11.3 and had to downgrade xorg to get it working again, but I can't get to my mail right now (on the box that is down). It was posted to the suse list about 3 weeks ago or so. I'll report back after I get the info from the arch box. Here is the link to the 11.3 xorg versions that I had to downgrade to in order to recover a working X on 11.3: http://linux.derkeiler.com/Mailing-Lists/SuSE/2011-02/msg00239.html On Arch, the following versions caused the same problem: xf86-video-ati-6.14 xorg-server-1.9.4-1 Downgrading to xorg-server-1.9.3.901-1 alone didn't fix the issue, BUT xorg-server and downgrading to xf86-video-ati-6.13.2 made it work like a champ :) So the issue looks to be the radeon driver either alone, or in combination with the xorg-server. What do you think we can try with opensuse? For the ati driver, this one is the one that works: http://xorg.freedesktop.org/releases/individual/driver/xf86-video-ati-6.13.0 Do you have a .spec for it handy? I can try and build it (but then I have no idea how to integrate it with the new /etc/X11/xorg.conf.d/ udev setup. Is there a cheat-sheet somewhere that would explain it? David, for now please try this in /etc/X11/xorg.conf.d/50-device.conf Section "Module" Disable "dri" Disable "dri2" EndSection Section "Device" Identifier "Default Device" Driver "radeonhd" Option "AccelMethod" "xaa" EndSection That probably fixes the radeonhd related Xserver crash at the startup, switches from EXA to XAA as acceleration method and disabled 3D support completely. Hope this helps. Stefan - YOU DA MAN!! That worked! I'm a bit confused as to why EXA isn't working, but the display looks great! Got a neat 11.4 screenshot for you: http://www.3111skyline.com/dl/suse/bugs/114/X/working.jpg Is there anything else I should do to tweak the driver? Stefan, Is there something we can do to help performance? When the screensaver (Flux GL) comes on it is just (ca-chunk, ca-chunk, ...) drawing the screen frame by frame. It used to fly! What to try? (In reply to comment #17) > Stefan, > > Is there something we can do to help performance? When the screensaver (Flux > GL) comes on it is just (ca-chunk, ca-chunk, ...) drawing the screen frame by > frame. It used to fly! > > What to try? That's what I told you before. That way we've disabled 3D support completely. Same problem here. Hardware: HP Compaq NC6000 laptop with Radeon 9600 graphics (PCI ID: 1002:4e50) openSUSE version: 11.4 i586 Screen is garbled when booting without nomodeset option. When using nomodeset, X doesn't start - segfaults as in Comment #10. Using options mentioned in Comment #15 helps, but of course without 3D acceleration. Another thing I noticed: if I boot into WindowsXP first and then reboot into openSUSE 11.4 (without using nomodeset), everything works ok. Unfortunately, neither sleep nor hibernate work, so I have to do this every time I want to start openSUSE. Created attachment 418971 [details]
Screenshot of bug
(In reply to comment #20) > Created an attachment (id=418971) [details] > Screenshot of bug Heh, heh, heh... At least you can read your screen after the corruption occurs :) I'm sure the smart guys will get this solved in the near term. Glad I'm not the only one affected by it. *** Bug 679141 has been marked as a duplicate of this bug. *** Hello! Got a ASUS X1950 XTX; I am suffering the same problem. I have KDE crashes in addition to that i had to disable acceleration features. So KDE is quite unusable to me at the moment. I hope there will be a fix for this soon. I am waiting as i dont want to mess up existing setup. Many thanks! Greetings, R PS.: I am using KDE stable repos in addition to basic ones. Created attachment 419230 [details]
Segfault output + backtrace from startx
I was able to get a working configuration using the fbdev driver. Here's the corresponding xorg.conf.d/50-device.conf file. Section "Device" Identifier "RadeonX1400" # Driver "radeonhd" Driver "fbdev" ## Required magic for radeon/radeonhd drivers; output name ## (here: "DVI-0") can be figured out via 'xrandr -q' # Option "monitor-DVI-0" "Monitor[0]" EndSection I am going to try disabling 3D support and switching the acceleration method as Stephan suggested above. Hopefully it we perform better than the fbdev driver. *** Bug 680415 has been marked as a duplicate of this bug. *** I've just added this bug to the list of most annoying bugs for openSUSE 11.4. *** Bug 679105 has been marked as a duplicate of this bug. *** Workaround for R5xx users (Radeon X1xx series)
-----------------------------------------------
- disable KMS by adding 'nomodeset' to kernel boot options
(/boot/grub/menu.lst).
- add the following section to your /etc/X11/xorg.conf.d/50-device.conf
Section "Module"
Disable "dri"
Disable "dri2"
EndSection
- use this "Device" in your /etc/X11/xorg.conf./50-device.conf
Section "Device"
Identifier "Default Device"
Driver "radeonhd"
Option "AccelMethod" "xaa"
EndSection
Workaround for R3xx/R4xx users (Radeon 9600/X300/Xxxx)
------------------------------------------------------
- disabled KMS by adding 'nomodeset' to boot options.
- add the following section to your /etc/X11/xorg.conf.d/50-device.conf
Section "Module"
Disable "dri"
Disable "dri2"
EndSection
Just for the record (ref bnc#680415), _removing_ 'nomodeset' from the boot options fixed the issue for me. (In reply to comment #30) > Just for the record (ref bnc#680415), _removing_ 'nomodeset' from the boot > options fixed the issue for me. Then you are *very* lucky! Using Radon Xpress 200M: I tried nomodeset at first, but that just led to no GUI at all, then I went with "Workaround for R3xx/R4xx users (Radeon 9600/X300/Xxxx)" from comment 29 and that worked. Workaround for me (R300): enable desktop effects in KDE4. But system will be less responsible. I've got the same "pixelation" problem on RS690 (X1200). Using xorg.conf and radeon driver. I first tried "Option NoAccel true", it works but is very slow. Using the entries from comment 15 (but in xorg.conf instead) also works and feels a little bit faster. This might be the same bug as https://bugs.freedesktop.org/show_bug.cgi?id=33929 The 'Option "ColorTiling" "false"' mentioned there fixes my problem. There is also a patch, but I've not tried it. Thanks, Jan. I've added the patch to obs://X11:XOrg/xorg-x11-server package. Could you give it a try? The RPM changelog is: Fri Mar 25 03:20:58 UTC 2011 - sndirsch@novell.com - EXA-mixed-ModifyPixmapHeader-pitch-fixes.-bug-33929.patch * EXA/mixed: ModifyPixmapHeader pitch fixes. (bnc #678264, bfo #33929) Looks Good! I first commented out the workarounds to verify that I still had a distorted image. It's just to easy to earn a brown paperbag by forgetting something like that! Then after installing xorg-x11-server-7.6_1.9.3-138.1.x86_64.rpm everything looks normal again. Thanks, Stefan! Thanks Jan! Anybody else, who can confirm that the option and the patched xorg-x11-server package version fix the initial issue (screen distortion)? >Anybody else, who can confirm that the option and the patched
xorg-x11-server package version fix the initial issue (screen distortion)?
Yes, with patched xorg-x11-server works fine
Hallo, My card doesn´t work with KMS since there is KMS until now. So as i am using radeon;xorg.conf;nomodeset. I can only confirm that there is still a backtrace with a nil value using the patch from xorg repository. Greetings, R (In reply to comment #38) > Thanks Jan! Anybody else, who can confirm that the option and the patched > xorg-x11-server package version fix the initial issue (screen distortion)? I tried installing xorg-x11-server-7.6_1.9.3-138.1.i586.rpm, and the problem persists. The screen looks fine in kdm - login screen, but gets corrupted when KDE desktop loads. xorg-x11-server-7.6_1.9.3-138.1.i586.rpm fixes segfault on two machines with r690, nomodeset. No problems regarding dri or anything else any more. Thanks! I've never had problems with screen distortion. The problem with the segfaults, when dri/sri2 is activated persists with xorg-x11-server-7.6_1.9.3-138.1.i586.rpm for me. My configuration: r520, nomodeset Well, the patch is only considered to fix the distortion issue, not the segfault. Hello! i found that i got a somewhat different backtrace: (without kms enabled dri dri2) Backtrace: [ 258.493] 0: /usr/bin/Xorg (xorg_backtrace+0x37) [0x80e4a97] [ 258.493] 1: /usr/bin/Xorg (0x8048000+0x62f0a) [0x80aaf0a] [ 258.493] 2: (vdso) (__kernel_rt_sigreturn+0x0) [0xffffe40c] [ 258.493] Segmentation fault at address (nil) [ 258.493] Fatal server error: [ 258.493] Caught signal 11 (Segmentation fault). Server aborting [ 258.493] [ 258.493] When booting with KMS i get this in /var/log/messages: Mar 26 12:19:13 wild-thing kernel: [ 247.627332] radeon 0000:01:00.0: Could not find HDMI block for 0xf encoder Mar 26 12:20:56 wild-thing kernel: [ 58.892307] radeon 0000:01:00.0: Could not find HDMI block for 0x13 encoder Screen doesn´t get detected :( Screen: HPw2408h Chipset: "ATI Radeon X1950" (ChipID = 0x7240) Can you please give any advice on what to do? I have a bugreport about no picture with KMS since SuSE 11.3. But i upgraded to 11.4. And now i am getting this stacktrace. Should i open a different bugreport? Many thanks! Greetings, R (In reply to comment #45) > When booting with KMS i get this in /var/log/messages: > Mar 26 12:19:13 wild-thing kernel: [ 247.627332] radeon 0000:01:00.0: Could > not find HDMI block for 0xf encoder > Mar 26 12:20:56 wild-thing kernel: [ 58.892307] radeon 0000:01:00.0: Could > not find HDMI block for 0x13 encoder > > Screen doesn´t get detected :( > > Screen: HPw2408h > Chipset: "ATI Radeon X1950" (ChipID = 0x7240) > > Can you please give any advice on what to do? I have a bugreport about no > picture with KMS since SuSE 11.3. But i upgraded to 11.4. And now i am getting > this stacktrace. Should i open a different bugreport? Many thanks! Yes, please open a seperate bugreport (kernel component, since it's a KMS issue). (In reply to comment #46) > (In reply to comment #45) > > When booting with KMS i get this in /var/log/messages: > > Mar 26 12:19:13 wild-thing kernel: [ 247.627332] radeon 0000:01:00.0: Could > > not find HDMI block for 0xf encoder > > Mar 26 12:20:56 wild-thing kernel: [ 58.892307] radeon 0000:01:00.0: Could > > not find HDMI block for 0x13 encoder > > > > Screen doesn´t get detected :( > > > > Screen: HPw2408h > > Chipset: "ATI Radeon X1950" (ChipID = 0x7240) > > > > Can you please give any advice on what to do? I have a bugreport about no > > picture with KMS since SuSE 11.3. But i upgraded to 11.4. And now i am getting > > this stacktrace. Should i open a different bugreport? Many thanks! > > Yes, please open a seperate bugreport (kernel component, since it's a KMS > issue). I already opened this issue as a separate bug report. It was merged in here by you. https://bugzilla.novell.com/show_bug.cgi?id=679141 Should it be repopened and changed to kernel component? (In reply to comment #47) > I already opened this issue as a separate bug report. It was merged in here by > you. > > https://bugzilla.novell.com/show_bug.cgi?id=679141 > > Should it be repopened and changed to kernel component? No, that one is about the segfault with radeonhd driver (KMS disabled). Please open a *new* bugreport against kernel component. Since we've seen mixed results here (and it doesn't address only the screen distortion), I've submitted the patch only for openSUSE:Factory (SR #65325). (In reply to comment #49) > Since we've seen mixed results here (and it doesn't address only the screen > distortion), ... ... and it does address only the screen distortion ... I am sorry for the confusion. This is the link to my bugreport about KMS Kernel component since OpenSuSE 11.3 (): http://bugzilla.novell.com/show_bug.cgi?id=669937 The Stacktrace is Kernelcomponent, too? Stefan, all, for the record, here are the specifics of the card that started this mess:
01:05.0 VGA compatible controller: ATI Technologies Inc RS690M [Radeon X1200 Series] (prog-if 00 [VGA controller])
Subsystem: Toshiba America Info Systems Device ff00
Flags: bus master, fast devsel, latency 64, IRQ 18
Memory at f0000000 (64-bit, prefetchable) [size=128M]
Memory at f8100000 (64-bit, non-prefetchable) [size=64K]
I/O ports at 9000 [size=256]
Memory at f8000000 (32-bit, non-prefetchable) [size=1M]
Expansion ROM at <unassigned> [disabled]
Capabilities: [50] Power Management version 2
Capabilities: [80] MSI: Enable- Count=1/1 Maskable- 64bit+
Kernel driver in use: radeon
The dmesg information is:
[ 2.680472] udev[126]: starting version 166
[ 2.839085] [drm] Initialized drm 1.1.0 20060810
[ 2.882958] [drm] radeon defaulting to kernel modesetting.
[ 2.882961] [drm] radeon kernel modesetting enabled.
[ 2.883142] checking generic (f0000000 1000000) vs hw (f0000000 8000000)
[ 2.883148] fb: conflicting fb hw usage radeondrmfb vs VESA VGA - removing generic driver
[ 2.883178] Console: switching to colour dummy device 80x25
[ 2.883481] radeon 0000:01:05.0: power state changed by ACPI to D0
[ 2.883490] radeon 0000:01:05.0: power state changed by ACPI to D0
[ 2.883499] radeon 0000:01:05.0: PCI INT A -> GSI 18 (level, low) -> IRQ 18
[ 2.886581] [drm] initializing kernel modesetting (RS690 0x1002:0x791F).
[ 2.886709] [drm] register mmio base: 0xF8100000
[ 2.886712] [drm] register mmio size: 65536
[ 2.886836] ATOM BIOS: ATI
[ 2.886867] radeon 0000:01:05.0: VRAM: 128M 0x00000000D8000000 - 0x00000000DFFFFFFF (128M used)
[ 2.886871] radeon 0000:01:05.0: GTT: 512M 0x00000000A0000000 - 0x00000000BFFFFFFF
[ 2.886903] [drm] radeon: irq initialized.
[ 2.887131] [drm] Detected VRAM RAM=128M, BAR=128M
[ 2.887136] [drm] RAM width 128bits DDR
[ 2.887242] [TTM] Zone kernel: Available graphics memory: 1962886 kiB.
[ 2.887245] [TTM] Initializing pool allocator.
[ 2.887268] [drm] radeon: 128M of VRAM memory ready
[ 2.887270] [drm] radeon: 512M of GTT memory ready.
[ 2.887300] [drm] GART: num cpu pages 131072, num gpu pages 131072
[ 2.891190] [drm] radeon: 1 quad pipes, 1 z pipes initialized.
[ 2.899728] radeon 0000:01:05.0: WB enabled
[ 2.899816] [drm] Loading RS690/RS740 Microcode
[ 2.902946] [drm] radeon: ring at 0x00000000A0001000
[ 2.902967] [drm] ring test succeeded in 1 usecs
[ 2.903177] [drm] radeon: ib pool ready.
[ 2.903261] [drm] ib test succeeded in 0 usecs
[ 2.903264] [drm] Enabling audio support
[ 2.903574] [drm] Radeon Display Connectors
[ 2.903577] [drm] Connector 0:
[ 2.903578] [drm] VGA
[ 2.903581] [drm] DDC: 0x7e50 0x7e40 0x7e54 0x7e44 0x7e58 0x7e48 0x7e5c 0x7e4c
[ 2.903583] [drm] Encoders:
[ 2.903585] [drm] CRT1: INTERNAL_KLDSCP_DAC1
[ 2.903587] [drm] Connector 1:
[ 2.903589] [drm] LVDS
[ 2.903591] [drm] DDC: 0x7e40 0x7e50 0x7e44 0x7e54 0x7e48 0x7e58 0x7e4c 0x7e5c
[ 2.903593] [drm] Encoders:
[ 2.903595] [drm] LCD1: INTERNAL_LVTM1
[ 2.903597] [drm] Connector 2:
[ 2.903598] [drm] S-video
[ 2.903600] [drm] Encoders:
[ 2.903601] [drm] TV1: INTERNAL_KLDSCP_DAC1
[ 3.459254] [drm] fb mappable at 0xF0040000
[ 3.459257] [drm] vram apper at 0xF0000000
[ 3.459259] [drm] size 5300224
[ 3.459261] [drm] fb depth is 24
[ 3.459263] [drm] pitch is 5888
[ 3.913104] Console: switching to colour frame buffer device 176x52
[ 3.982881] fb0: radeondrmfb frame buffer device
[ 3.982884] drm: registered panic notifier
[ 3.982892] [drm] Initialized radeon 2.7.0 20080528 for 0000:01:05.0 on minor 0
<snip>
[ 7.687953] input: Video Bus as /devices/LNXSYSTM:00/device:00/PNP0A08:00/device:25/LNXVIDEO:01/input/input7
[ 7.688052] ACPI: Video Device [VGA1] (multi-head: yes rom: no post: no)
[ 7.733548] MCE: In-kernel MCE decoding enabled.
I'll go try the patched Xorg and report back. Thanks for your help Stefan!
Stefan, I installed the xorg packages from X11/Xorg and removed the modification to 50-devices.conf. I am running without any mods to disable 3D and it is working! -- sort of :) ksplash still rips and distorts (primarily the lower 1/2 of the ksplash display), but the normal X display looks good! FluxGL screensaver actually moves now -- although performance is still a bit slow. However it is MUCH, MUCH better than disabling 3D. I'm marking the "This provides the needed information" button. If you still need more, Just let me know. The current Xorg.0.log and dmesg information are available at the following links: [39k] http://www.3111skyline.com/dl/suse/bugs/114/X/patched/Xorg.0.log [67k] http://www.3111skyline.com/dl/suse/bugs/114/X/patched/dmesg.txt Stefan, I have also placed a current list of the xorg packages I have installed here: http://www.3111skyline.com/dl/suse/bugs/114/X/patched/xorg-packages.txt e.g. xorg-x11-libICE-7.6-23.1.x86_64 xorg-x11-fonts-7.6-69.1.noarch xorg-x11-libXfixes-7.6_4.0.5-16.1.x86_64 xorg-x11-server-7.6_1.9.3-138.2.x86_64 <snip> (In reply to comment #29) > Workaround for R5xx users (Radeon X1xx series) > ----------------------------------------------- > > - disable KMS by adding 'nomodeset' to kernel boot options > (/boot/grub/menu.lst). > > - add the following section to your /etc/X11/xorg.conf.d/50-device.conf > > Section "Module" > Disable "dri" > Disable "dri2" > EndSection > > - use this "Device" in your /etc/X11/xorg.conf./50-device.conf > > Section "Device" > Identifier "Default Device" > > Driver "radeonhd" > Option "AccelMethod" "xaa" > EndSection > > > Workaround for R3xx/R4xx users (Radeon 9600/X300/Xxxx) > ------------------------------------------------------ > > - disabled KMS by adding 'nomodeset' to boot options. > > - add the following section to your /etc/X11/xorg.conf.d/50-device.conf > > Section "Module" > Disable "dri" > Disable "dri2" > EndSection Hardware: 02:00.0 VGA compatible controller: ATI Technologies Inc RV530LE [Radeon X1600/X1650 PRO] Worked for me with Driver "radeon". If Driver "radeonhd" selected then card detect as RV535 Radeon X1300, X1650 which has a different output configuration. *** Bug 685018 has been marked as a duplicate of this bug. *** After upgrade from 11.3, I, too, tripped on this bug. Display of login screen was fine, but the Gnome desktop was distorted. Driver is: xorg-x11-driver-video-radeonhd-1.3.0_20100512_80ba041-2.1.x86_64 Found an interesting workaround. If you use the "Configure Display Settings" to request an "Upside Down" rotation, then return the rotation to "normal", the distortion goes away - and stays away during subsequent logins (coming back after a power cycle). Perhaps this indicates that something needs reinitialization. As an aside, I came up with this idea while thinking of resetting an Etch-A-Sketch. Ran in the same bug after installing 11.4, KDE was fine here, until I opened an xterm. The configuration from comment#29 had no success. With 'nomodeset' and 'radeonhd' my X-Server fails to start at all. I can combine 'nomodeset' with 'radeon' or 'fbdev' -- this works and survives opening an xterm. 'fbdev' without nomodeset is also sufficient. Juergen, comment #35/36 is interesting here. Not comment #29. *** Bug 691244 has been marked as a duplicate of this bug. *** *** Bug 691234 has been marked as a duplicate of this bug. *** The issue has been identified as a Mesa bug fixed by git f0fa040 dri_util: fail driCreateNewScreen if InitScreen is NULL Mesa 7.10.2 has *many* additional fixes, and is currently prepared in X11:XOrg. Please test when it's done compiling. At the moment it should work with 11.4 as well, without need to install all other X packages from that directory. For testing the workaround described in comment #29 needs to be reverted. And please make sure that RPM changelog contains the following entry: Fri May 6 16:46:21 UTC 2011 - mhopf@novell.com - Update to 7.10.2 * Many crash and rendering fixes * Fixes bug 678264 (Xserver crash with radeon/radeonhd UMS) Stefan, Matthias: I updated with the packages from 5/6 and things seem to be working well. I have reverted all prior config changes so I'm running with the default setup in /etc/X11/xorg.conf.d. Performance is much improved. (we may have found the bug from a year ago that killed radeonhd for the RS690M) The only thing I see is 'momentary' screen distortion (like a garbled frame or two) when kde ksplash starts, and then when you 'test screen saver' from the kde control panel. The emphasis is on 'momentary'. After the modes are switched (or whatever happens in those two instances), the display is fine. I'm marking the 'provides neeeded information' box. If you want/need more information from others, please just uncheck it and I won't re-check it. Thanks for the great work! Let me drop another note on performance -- it is the best I've seen out of the radeonhd driver yet. I use the Flux(GL) screensaver to give me a relative idea on how the openGL code is working with the driver and hardware. When this bug hit, Flux(GL) was so slow it was litterally 'clunking' across the screen. It was terrible. With this release, the screensave is smooth and fluid without much noteable delay compared to the 8-9 fglrx driver that was available for 11.0. That says a LOT about how much performance has improved with these fixes. Again -- great job! I would like to update Mesa (7.10 --> 7.10.2; only a bugfix relase) and add the EXA fix to xorg-x11-server for openSUSE 11.4 to finally address that issue. It's one of the most annoying bugs on 11.4 we have. "bugfix only" means "low risk" for breaking stuff? then I would be fine. +1 Yes, .n (n >= 1) releases of Mesa contain only bugfixes. Low risk, yes. yes, we can do +1 The SWAMPID for this issue is 40808. This issue was rated as low. Please submit fixed packages until 2011-06-07. Also create a patchinfo file using this link: https://swamp.suse.de/webswamp/wf/40808 I tried that new Mesa 7.10.2 and the Segfault are gone, and Xorg start just fine, but no 3d support with errors: (EE) AIGLX error: Calling driver entry point failed (EE) AIGLX: reverting to software rendering Yours is most likely a different issue, given the result in comment 65. Please attach the full Xorg.0.log. Created attachment 428924 [details]
Xorg.0.log with AIGLX error and no 3d support after start
[ 63.403] (II) [KMS] drm report modesetting isn't supported. Please read comment 63 again - you have to remove the workarounds, and you still seem to have nomodeset active in the kernel command line. xorg-x11-server: SR #70051 Mesa: SR #70056 Patchinfo: https://swamp.suse.de/webswamp/swamp/template/DisplayWorkflow.vm/workflowid/40839 Created attachment 429151 [details]
Xorg.0.log with "nomodeset" and no 3d support after start X server
Matthias Hopf, please, check this log too - all workarounds are gone here. (In reply to comment #77) > Created an attachment (id=429151) [details] > Xorg.0.log with "nomodeset" and no 3d support after start X server [ 5504.926] (II) [KMS] drm report modesetting isn't supported. Again. *Remove* "nomodeset" from the kernel boot options. I can't remove "nomodeset", because with KMS and "suspend to disk" doesn't work at all. Do I need open new bug for 3d issue ? This is a complete different issue. Update released for: Mesa, Mesa-debuginfo, Mesa-debugsource, Mesa-devel, Mesa-nouveau3d, Mesa-nouveau3d-debuginfo, xorg-x11-Xvnc, xorg-x11-Xvnc-debuginfo, xorg-x11-server, xorg-x11-server-debuginfo, xorg-x11-server-debugsource, xorg-x11-server-extra, xorg-x11-server-extra-debuginfo, xorg-x11-server-sdk Products: openSUSE 11.4 (debug, i586, x86_64) I have a problem which I'm not sure if its related to any of the above, I'm using OpenSUSE 11-4 64 and Gnome 3 with an ATI 4830 card. When updating the Mesa 4546.noarch patch using Software Update in Gnome Shell, I get the following error message: patch:Mesa-4546.noarch conflicts with Mesa.x86_64 < 7.10.2-7.31 provided by Mesa-7.10-1.3.x86_64 I can apply the update using YaST but I have a problem after I reboot the system and log in to my account. When I click on the 'Activities' label, or some of the icons on the top panel, the screen flickers once but then everything appears to be normal. However,when I try and click again, a new screen appears stating there is an unrecoverable error and asks me to log out and try again. This does not happen when I click on the Date/Calender label on the middle of the panel, or the User/Account label on the far right of the panel-they work fine. I've tried to find an answer to the problem on the forums but with no luck. I'm new to OpenSUSE so there could be a simple fix that I'm not aware of. Is this a known problem or can anyone help me find a solution? Thanks Last build with removal of radeonhd package now causes Flux GL screensaver to freeze (no movement). It will still unlock, but there is no screensaving motion. Please keep this bugreport closed. The main issue (als described in the title of the bug) has been fixed. If really necessary open a new one ... Gotcha Stefan,
Wasn't sure whether it fit here or if a new one was needed, but I knew it was a last build issue w.r.t. this bug so I gave it a shot here. Latest build seems to have fixed it anyway. Thanks.
|