Bug 6951 - dkms built fglrx disabled 3rd monitor on triple head 6870 system, works w/std kernel/driver
Summary: dkms built fglrx disabled 3rd monitor on triple head 6870 system, works w/std...
Status: RESOLVED WONTFIX
Alias: None
Product: Mageia
Classification: Unclassified
Component: RPM Packages (show other bugs)
Version: 2
Hardware: x86_64 Linux
Priority: Normal normal
Target Milestone: ---
Assignee: Mageia Bug Squad
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2012-08-04 18:14 CEST by Greg McGee
Modified: 2012-08-05 13:39 CEST (History)
0 users

See Also:
Source RPM: dkms-fglrx-8.961-1.mga2.nonfree
CVE:
Status comment:


Attachments

Description Greg McGee 2012-08-04 18:14:07 CEST
Description of problem:
dkms built fglrx on recompiled custom kernel (only turned off tickless and set for low latency desktop) disabled 3rd monitor on triple head 6870/2G and AMD 785-based quad core system, works perfectly w/std kernel/driver

(just boots at incredibly slow rate for unknown reasons with std kernel, but I digress, did a fresh install from release, mistake, was running from beta2 DVD install, worked better)

Only posting a bug as it's a head scratcher.

Version-Release number of selected component (if applicable):
dkms-fglrx-8.961-1.mga2.nonfree
kernel-3.3.6-2.mga2รข


How reproducible: Boot custom kernel with dkms built fglrx, tickless set off and low latency desktop set.

I'm going to rebuild with std options and see if it's the LL desktop or tickless options causing the issue for some bizarre reason, mainly posting bug to document the issue.
Comment 1 Greg McGee 2012-08-05 13:39:40 CEST
Duplicated after building more variations of the std kernel than I'd like to admit, disabling ONLY "tickless kernel" as an option (all else being std, same xorg.conf and all other kernel options) also disables the third monitor functionalityfor some unknown reason. 

As it really hurts nothing to leave "tickless enabled (as it must be explicitly turned on in the kernel command line ala "nohz=1" or such...) ...the moral of the story is "don't do that"

The low latency etc made no difference.

This likely only matters to the three people running a triple-headed AMD//ATI card configs with fglrx.

Closed and documented for posterity?

Status: NEW => RESOLVED
Resolution: (none) => WONTFIX


Note You need to log in before you can comment on or make changes to this bug.