Kernel Bug Tracker – Bug 12401
2.6.28 regression: xbacklight broken on ThinkPad X61s
Last modified: 2013-04-09 06:23:26 UTC
Subject : 2.6.28 regression: xbacklight broken on ThinkPad X61s
Submitter : Tino Keitel <firstname.lastname@example.org>
Date : 2009-01-05 8:39
References : http://marc.info/?l=linux-kernel&m=123114479110314&w=4
This entry is being used for tracking a regression from 2.6.27. Please don't
close it until the problem is fixed in the mainline.
Notify-Also : Zhang Rui <email@example.com>
Notify-Also : Matthew Garrett <firstname.lastname@example.org>
Author: Thomas Renninger <email@example.com>
Date: Fri Aug 1 17:37:54 2008 +0200
ACPI: video: Ignore devices that aren't present in hardware
Signed-off-by: Thomas Renninger <firstname.lastname@example.org>
Acked-by: Zhang Rui <email@example.com>
Signed-off-by: Andi Kleen <firstname.lastname@example.org>
Signed-off-by: Len Brown <email@example.com>
First-Bad-Commit : 22c13f9d8179f4c9caecfcb60a95214562b9addc
On my T61 2.6.28 is the kernel which made the brightness keys work :) The OpRegion stuff fixed all for me, and xbacklight works fine too. So maybe there's some stuff missing for x61 but on other X3100 it's working.
This looks like a duplicate of #12361.
*** Bug 12361 has been marked as a duplicate of this bug. ***
Bug 12174 could also be a duplicate.
I remember Matthew and Rui saying that OpRegion does not always work due to IRQ problems with the Intel graphics driver? But I don't know how to find that out or how far it is solved. Hmm, Matthew does not have a bugzilla account...
Matthias: Can you switch brightness by echoing brightness levels into the remaining backlight device's sysfs files?
Tino, please attach the acpidump output.
you can get it by using the latest pmtools at http://www.lesswatts.org/projects/acpi/utilities.php
If this is still a problem, I strongly suspect that it's down to the 915 drm failing to get the IRQs. This has been a perennial problem - it's been fixed at various points.
However, the ACPI change means that that systems that expect opregion support *must* have the i915 drm loaded to get working brightness control.
The acpidump output is attached to bug #12361
If drm must be loaded, then this could be the real cause why xbacklight is not working, and why it worked the last time I tried 188.8.131.52. I disable drm most of the time, because it saves some power, and the X61s is a small and mobile laptop.
Oh, right. There's no way to make this work correctly without drm. Can you try with that built?
I guess this problem can also be fixed by the patch in http://bugzilla.kernel.org/show_bug.cgi?id=11259#c26
No. This bug relates to the fact that the DRM code is now required for working backlight control on some i915 machines.
first, I mean the patch in comment#25 bug#11259.
second, what we need is that i915 driver MUST be loaded on IGD OpRegion boxes to support backlight control, which the patch in comment#25 bug#11259 is not(it only fixes the oops when KMS is set).
well, I don't think we got a proposal for this bug in the IRC yesterday.
Matthew, any ideas?
If users with opregion-based hardware don't build the i915 DRM they don't get backlight control. I don't think there's any other way of dealing with this. Distros will all be building with DRM, so I don't think it's a real problem.
close this bug as the backlight works if DRM is loaded (comment #11).