brightness dark and not working on acer 5715

Bug #401591 reported by alain57
30
This bug affects 5 people
Affects Status Importance Assigned to Milestone
Linux
Invalid
Medium
linux (Ubuntu)
Fix Released
Low
Andy Whitcroft

Bug Description

hi,
i currently have a brightness issue on my notebook. In fact i have this problem a little longer, but i was hoping that it will be corrected with the time... unfortunately it is not the case (may i be the only person trying karmic on this notebook ?!?)

so here is the problem :

when i boot on karmic, i can't increase or decrease the brightness, and by default it's really dark.
the only solution i found, is really a stupid one :(

switch from X to VT1
use the FN + brightness down key => i receive an error message "ACPI: Failed to switch the brightness", but it don't work
funnily the brightness up key don't work too, but don't give me any error message

i close the notebook
i open it and switch to X
if i'm lucky it work
if not i need to do all that stuff again (i don't need to do this a lot, maximum 3 or 4 times)

if first through it may be an KMS problem (because KMS disable brightness on my other notebook) so i already disabled it.
but even after disabling it, netheir my brightness keys, xbacklight, or the brightness applet worked :(

/sys/class/backlight is empty

Revision history for this message
alain57 (alain57) wrote :
Revision history for this message
alain57 (alain57) wrote :
Revision history for this message
alain57 (alain57) wrote :

even doing something like this don't work :(

lspci | grep VGA
00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 03)

second :
setpci -s 00:02.0 F4.B=66 (a value between 00 and FF, where FF is 100%)

Revision history for this message
alain57 (alain57) wrote :

this bug is really bad for me...
so if you guys needs further informations i can provide them, you just need to tell me what i need to do

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

Hi alain57,

Seems this might be related to the following upstream bug in which a patch has been applied. Could you attach your dmidecode if this issue remains with the latest 2.6.31-4 kernel? Thanks.

http://bugzilla.kernel.org/show_bug.cgi?id=13121

Might also be related to bug 333386 but that was reported against an Acer 6920G.

Changed in linux (Ubuntu):
status: New → Incomplete
Revision history for this message
alain57 (alain57) wrote :

here is my dmidecode
ps : i'm on 2.6.31-4 and my brightness is not working

Revision history for this message
alain57 (alain57) wrote :

well i read the first bugzilla topic you gave me, and i saw a patch for 5715Z (same notebook as i have)

but i checked the source, and .... well the patch for 5710, 5720, 5315 and 7720 are IN
but the patch for 5715Z is not in :(

can you please add following patch http://bugzilla.kernel.org/attachment.cgi?id=21801 to the next kernel release ?

i haven't test it, but on the topic, an other guy test it and said it worked

thanks in advance

Revision history for this message
alain57 (alain57) wrote :

i don't now if i did things right
i simply modify the video.c file
changed 5710Z into 5715Z
saved

cp Makefile Makefile.old

nano Makefile and put this

obj-m += video.o

all:
        make -C /lib/modules/$(shell uname -r)/build M=$(PWD) modules

clean:
        make -C /lib/modules/$(shell uname -r)/build M=$(PWD) clean

then i did a make

i boot on other kernel (to be sure i can write the new module)
i copy video.ko in /lib/modules/2.6.31-4-generic/kernel/drivers/acpi/

reboot with kernel 2.6.31-4
and .... nothing changed :(

other stuff i did

cat /sys/devices/platform/acer-wmi/interface
WMID

grep -r . /proc/acpi/video
/proc/acpi/video/OVGA/DD05/EDID:<not supported>
/proc/acpi/video/OVGA/DD05/brightness:<not supported>
/proc/acpi/video/OVGA/DD05/state:state: 0x1d
/proc/acpi/video/OVGA/DD05/state:query: 0x00
/proc/acpi/video/OVGA/DD05/info:device_id: 0x0005
/proc/acpi/video/OVGA/DD05/info:type: UNKNOWN
/proc/acpi/video/OVGA/DD05/info:known by bios: no
/proc/acpi/video/OVGA/DD04/EDID:<not supported>
/proc/acpi/video/OVGA/DD04/brightness:<not supported>
/proc/acpi/video/OVGA/DD04/state:state: 0x1d
/proc/acpi/video/OVGA/DD04/state:query: 0x00
/proc/acpi/video/OVGA/DD04/info:device_id: 0x0004
/proc/acpi/video/OVGA/DD04/info:type: UNKNOWN
/proc/acpi/video/OVGA/DD04/info:known by bios: no
/proc/acpi/video/OVGA/DD03/EDID:<not supported>
/proc/acpi/video/OVGA/DD03/brightness:<not supported>
/proc/acpi/video/OVGA/DD03/state:state: 0x1d
/proc/acpi/video/OVGA/DD03/state:query: 0x00
/proc/acpi/video/OVGA/DD03/info:device_id: 0x0410
/proc/acpi/video/OVGA/DD03/info:type: UNKNOWN
/proc/acpi/video/OVGA/DD03/info:known by bios: no
/proc/acpi/video/OVGA/DD02/EDID:<not supported>
/proc/acpi/video/OVGA/DD02/brightness:<not supported>
/proc/acpi/video/OVGA/DD02/state:state: 0x1d
/proc/acpi/video/OVGA/DD02/state:query: 0x00
/proc/acpi/video/OVGA/DD02/info:device_id: 0x0240
/proc/acpi/video/OVGA/DD02/info:type: UNKNOWN
/proc/acpi/video/OVGA/DD02/info:known by bios: no
/proc/acpi/video/OVGA/DD01/EDID:<not supported>
/proc/acpi/video/OVGA/DD01/brightness:<not supported>
/proc/acpi/video/OVGA/DD01/state:state: 0x1d
/proc/acpi/video/OVGA/DD01/state:query: 0x00
/proc/acpi/video/OVGA/DD01/info:device_id: 0x0100
/proc/acpi/video/OVGA/DD01/info:type: UNKNOWN
/proc/acpi/video/OVGA/DD01/info:known by bios: no
/proc/acpi/video/OVGA/DOS:DOS setting: <0>
/proc/acpi/video/OVGA/POST:<not supported>
/proc/acpi/video/OVGA/POST_info:<not supported>
/proc/acpi/video/OVGA/ROM:<TODO>
/proc/acpi/video/OVGA/info:Switching heads: yes
/proc/acpi/video/OVGA/info:Video ROM: no
/proc/acpi/video/OVGA/info:Device to be POSTed on boot: no

and i case you need my acpidump, here it is

I'm not using KMS, because brigthness does not work on KMS with Intel (i now because on my other notebook brightness work only without KMS)

Revision history for this message
alain57 (alain57) wrote :

Hi, it seams Zhang Rui from Intel correct the bug
i opened a report at http://bugzilla.kernel.org/show_bug.cgi?id=13955

after applying the two patches :
http://bugzilla.kernel.org/attachment.cgi?id=21801
and
http://bugzilla.kernel.org/attachment.cgi?id=22672

i have now brightness working

i'll do some further test after work, and give feedback
if it this bug is resolved, can you guys please add the patches to the ubuntu kernel ?

Andy Whitcroft (apw)
tags: added: regression-potential
Revision history for this message
Andy Whitcroft (apw) wrote :

@alain57 -- you refer two patches associated with that upstream bug. The second one does look sensible in its own right. The first of those is simply a debug patch and not applicable to the general case, and therefore not applicable to our kernel. There does appear to be an updated patch in the bug, but that is predicated by a bunch of other patches:

According to upstream we would need these patches:
  http://bugzilla.kernel.org/attachment.cgi?id=21801
  http://patchwork.kernel.org/patch/40097/
  http://patchwork.kernel.org/patch/40098/
  http://patchwork.kernel.org/patch/40099/
  http://patchwork.kernel.org/patch/40100/
  http://patchwork.kernel.org/patch/40348/

Then this one:
  http://bugzilla.kernel.org/attachment.cgi?id=22675

I will put together a test kernel with this combination so we can help push this to mainline.

Changed in linux (Ubuntu):
assignee: nobody → Andy Whitcroft (apw)
status: Incomplete → In Progress
importance: Undecided → Low
Revision history for this message
alain57 (alain57) wrote :

yes it is exactly like you said
i did the same localy, but if you do a custom kernel i'll test it ^^

it would really be cool if brightness will work on any new ubuntu kernel

my next step is to help (as far as i can) to make brightness work on KMS

Changed in linux:
status: Unknown → Fix Released
Revision history for this message
Andy Whitcroft (apw) wrote :

Ok I have put together a test kernel with the stack of patches recommended by upstream. If you could test the kernels at the URL below and report back here that would be very helpful. The kernels are here:

    http://people.canonical.com/~apw/lp401591-karmic/

Changed in linux (Ubuntu):
status: In Progress → Incomplete
Revision history for this message
alain57 (alain57) wrote :

i installed the i386 version and it worked

but i was unable to install the headers
it tell me it depends on linux-headers-2.6.31-6 witch is not on my system.

Revision history for this message
Andy Whitcroft (apw) wrote :

@alain57 -- there are three headers packages at the URL above, one each for i386 and amd64, plus a separate _all package which provides that dependancy. Thanks for testing, that is great.

Changed in linux (Ubuntu):
status: Incomplete → In Progress
Revision history for this message
alain57 (alain57) wrote :

ooops sorry i didn't saw it.
well even without header it worked
i'm installing the header right now
if its no more working, i'll tell you

Revision history for this message
alain57 (alain57) wrote :

is the patch going to be included ?
i'm asking, because i had 2 kernel updates and each time i loose the brightness control because it doesn't seems to be included

Revision history for this message
MaximAtmashkin (atmashkinmi) wrote :

Same problem on Karmic (and other distros with kernels 2.6.30-2.6.31) on Acer Aspire 5530. On kernel 2.6.29 brightness is controllable, but also there are some problems : during boot it decreases to the lowest level (10); when i try to increase it, sometimes it's resisting to increasing :); when i'm decreasing it, it decreases by 30 (not 10).

Changed in linux:
status: Fix Released → Confirmed
Changed in linux:
status: Confirmed → Fix Released
Revision history for this message
alain57 (alain57) wrote :

brightness is now almost working on kernel 2.6.31-9.29

i can change the brightness level, but it's bugging
for exemple if i press brightness up for a couple of seconds, the brightness is blinking (it goes up and down)
the same happen if i change the brightness level with the brightness applet

so i don't think the fix was commited in 2.6.31-9.29, but a part of it is in there because i can change the level (badly)

Revision history for this message
alain57 (alain57) wrote :

still not working like it should on 2.6.31-10.30 ....

Revision history for this message
alain57 (alain57) wrote :

well regression ...
now on 2.6.31-10.32 its does not work at all
on all previous versions it was allmost working, now nothing ....

Changed in linux:
status: Fix Released → Confirmed
Revision history for this message
Andy Whitcroft (apw) wrote :

Ok, some acpi brightess fixes hit Ubuntu-2.6.31-11.38, if you could test the latest kernel 2.6.32-12.39 or later and report back here. Thanks!

Changed in linux (Ubuntu):
status: In Progress → Incomplete
Revision history for this message
nicky.7 (nickkkk7) wrote : Re: [Bug 401591] Re: brightness dark and not working on acer 5715

it's workin now.. thanks

2009/10/7 Andy Whitcroft <email address hidden>

> Ok, some acpi brightess fixes hit Ubuntu-2.6.31-11.38, if you could test
> the latest kernel 2.6.32-12.39 or later and report back here. Thanks!
>
> ** Changed in: linux (Ubuntu)
> Status: In Progress => Incomplete
>
> --
> brightness dark and not working on acer 5715
> https://bugs.launchpad.net/bugs/401591
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

Marking this Fix Released per the previous comment. Thanks!

Changed in linux (Ubuntu):
status: Incomplete → Fix Released
Revision history for this message
htruong (htruong) wrote :

I have a Lenovo U350 and the problem is *not* fixed as of kernel 2.6.32

$ uname -a
Linux lenovo-u350 2.6.32-020632rc5-generic #020632rc5 SMP Fri Oct 16 09:06:17 UTC 2009 x86_64 GNU/Linux

$ dmesg |tail
[ 240.118833] ACPI: Failed to switch the brightness
[ 240.614684] ACPI: Failed to switch the brightness

Revision history for this message
Jess Sightler (jesse-sightler) wrote :

I can confirm that the bug is still present with a Lenovo U350.

Revision history for this message
nicky.7 (nickkkk7) wrote :

It is resolved for acer 5715z. You can open another bug. Or change the name of this one maybe?
Bye, Nicola

Revision history for this message
Jess Sightler (jesse-sightler) wrote :

Yeah, someone marked this as a duplicate of bug #414658. However, it is now seems this assessment is false. I'm asking for the other one to be reopened.

Changed in linux:
status: Confirmed → Invalid
Changed in linux:
importance: Unknown → Medium
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.