Bug 35531 - Kernel 2.6.36 series, xf86-video-ati 6.14.0 kernel blows up
Kernel 2.6.36 series, xf86-video-ati 6.14.0 kernel blows up
Status: CLOSED FIXED
Product: DRI
Classification: Unclassified
Component: DRM/Radeon
unspecified
x86 (IA32) Linux (All)
: medium major
Assigned To: Default DRI bug account
https://bugs.gentoo.org/show_bug.cgi?...
:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2011-03-21 21:26 UTC by Christopher Head
Modified: 2011-03-22 22:42 UTC (History)
0 users

See Also:


Attachments
dmesg output from a successful attempt (22.77 KB, text/plain)
2011-03-21 21:26 UTC, Christopher Head
no flags Details
dmesg output from a failed attempt (30.48 KB, text/plain)
2011-03-21 21:26 UTC, Christopher Head
no flags Details
dmesg output with radeon.agpmode=-1 (24.37 KB, text/plain)
2011-03-21 21:26 UTC, Christopher Head
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description Christopher Head 2011-03-21 21:26:00 UTC
Created attachment 44703 [details]
dmesg output from a successful attempt

When booting up Xorg server 1.9.2 with xf86-video-ati-6.14.0 and kernel 2.6.36-gentoo-r5 (distro-patched kernel, but the same problem doesn't appear in 2.6.35-g-r12), as soon as X starts, my monitor goes into DPMS powersave and I get backtraces in dmesg.

A Gentoo developer suggested booting with kernel command-line parameter "radeon.agpmode=-1". This didn't help.

I've attached dmesg output from a successful startup with 2.6.35, and will attach dmesg output from failed startups with and without agpmode=-1 momentarily (looks like only one file can be attached at bug creation time).
Comment 1 Christopher Head 2011-03-21 21:26:26 UTC
Created attachment 44704 [details]
dmesg output from a failed attempt
Comment 2 Christopher Head 2011-03-21 21:26:41 UTC
Created attachment 44705 [details]
dmesg output with radeon.agpmode=-1
Comment 3 Dave Airlie 2011-03-21 21:42:27 UTC
that kernel doesn't have KMS enabled which is probably not what you want.
Comment 4 Christopher Head 2011-03-22 22:42:28 UTC
Oh, the two versions listed are those declared as ostensibly "stable" by my distro, but 2.6.37 fixes the bug.