Description of problem: I upgraded the kernel and kernel-devel packages from testing repository ( kernel-2.6.35.10-68.fc14). I see only "broken" screen of X after I rebooted to this new kernel and the system log is full of "[drm:radeon_cs_ioctl] *ERROR* Invalid command stream !" (see dmesg.out). Version-Release number of selected component (if applicable): kernel-2.6.35.10-68.fc14 How reproducible: boot to runlevel 5 Steps to Reproduce: 1. each boot to runlevel 5 2. 3. Actual results: see attached gdm.jpg, gnome-session.jpg and gnome-panel.jpg photos Expected results: Additional info: see attached dmesg.out, lspci.out and Xorg.0.log files
Created attachment 469569 [details] gdm.jpg
Created attachment 469570 [details] gnome-panel.jpg
Created attachment 469571 [details] gnome-session.jpg
Created attachment 469572 [details] dmesg.out
Created attachment 469573 [details] lspci.out
Created attachment 469574 [details] Xorg.0.log
valid for kernel-2.6.35.10-69 too
I will try revert this commit "881fe6c1d06bf49f4ab7aef212cdaf66bd059614" later this day. I expect that it will help for my HW. http://git.kernel.org/?p=linux/kernel/git/longterm/linux-2.6.35.y.git;a=commitdiff;h=b8e9a4a45f8427837f4dba89bda4d4e3f3a5c726;hp=cde76274952b114223f4b36123ffa3450c5bbf8a
I see the same problem with a Radeon HD 4870.
Dave, is stable 2.6.35 missing something? Should we back out that patch for now? --Kyle
commit f30df2fad0c901e74ac9a52a488a54c69a373a41 Author: Dave Airlie <airlied> Date: Thu Oct 21 13:55:40 2010 +1000 drm/radeon/r600: fix tiling issues in CS checker. The CS checker had some incorrect alignment requirements for 2D surfaces, this made rendering to mipmap levels that were 2D broken. Also the CB height was being worked out from the BO size, this doesn't work at all when rendering mipmap levels, instead we work out what height userspace wanted from slice max and use that to check it fits inside the BO, however the DDX send the wrong slice max for an unaligned buffer so we have to workaround for that even though its a userspace bug. Reviewed-by: Alex Deucher <alexdeucher> Signed-off-by: Dave Airlie <airlied> looks like it might help?
That might help, but note that there was a subsequent commit fixing typos in that commit: commit d33ef52d9db8a36900dd53f2e32db9a521ace259 drm/radeon/kms: fix typo in r600 cs checker Looks like a typo in: drm/radeon/r600: fix tiling issues in CS checker. (f30df2fad0c901e74ac9a52a488a54c69a373a41) Signed-off-by: Alex Deucher <alexdeucher> Signed-off-by: Dave Airlie <airlied>
And a subsequent revert of that...
Ah yes, good point.
http://koji.fedoraproject.org/koji/taskinfo?taskID=2677662 Could you try this scratch build when it completes?
Kernel build from comment 15 is OK :)
Great, thanks. I'll talk to airlied and try to get something fixed in the next build.
I'll chime in and say that that the kernel-2.6.35.10-69.bz664206.fc14 build WORKSFORME as well.
I can see this error: Dec 21 06:59:47 nbone kernel: [ 313.658517] radeon 0000:01:00.0: r600_cs_track_validate_cb:203 cb pitch (64) invalid Dec 21 06:59:47 nbone kernel: [ 313.658530] radeon 0000:01:00.0: r600_packet3_check:1108 invalid cmd stream 423 Dec 21 06:59:47 nbone kernel: [ 313.658537] [drm:radeon_cs_ioctl] *ERROR* Invalid command stream ! in both kernel-2.6.35.10-68.fc14.x86_64 kernel-2.6.35.10-69.fc14.x86_64
kernel-2.6.35.10-72.fc14.x86_64 works fine
Yes, kernel kernel-2.6.35.10-72.fc14.x86_64 include new patch drm-radeon-r600-cs-checker-fixes.patch which fixed this issue.
kernel-2.6.35.10-72.fc14 has been submitted as an update for Fedora 14. https://admin.fedoraproject.org/updates/kernel-2.6.35.10-72.fc14
kernel-2.6.35.10-72.fc14 has been pushed to the Fedora 14 testing repository. If problems still persist, please make note of it in this bug report. If you want to test the update, you can install it with su -c 'yum --enablerepo=updates-testing update kernel'. You can provide feedback for this update here: https://admin.fedoraproject.org/updates/kernel-2.6.35.10-72.fc14
kernel-2.6.35.10-72.fc14 has been pushed to the Fedora 14 stable repository. If problems still persist, please make note of it in this bug report.