[Acer Aspire V5-171] Backlight cannot be adjusted unless passing acpi_backlight=vendor

[Alan Jenkins]

The boot parameter is suggested as a fix for this model by
http://www.linlap.com/acer_aspire_v5-171

(so if it would help, I would suggest classing this bug as "confirmed").

ProblemType: Bug
DistroRelease: Ubuntu 13.04
Package: linux-image-3.8.0-19-generic 3.8.0-19.29 [modified: boot/vmlinuz-3.8.0-19-generic]
ProcVersionSignature: Ubuntu 3.8.0-19.29-generic 3.8.8
Uname: Linux 3.8.0-19-generic x86_64
ApportVersion: 2.9.2-0ubuntu8
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: alan 1919 F.... pulseaudio
Date: Sat Apr 27 19:59:23 2013
InstallationDate: Installed on 2013-04-27 (0 days ago)
InstallationMedia: Ubuntu 13.04 "Raring Ringtail" - Release amd64 (20130424)
MachineType: Acer V5-171
MarkForUpload: True
ProcFB: 0 inteldrmfb
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-3.8.0-19-generic.efi.signed root=UUID=ad03cbd3-39e0-4f7e-b0ab-e1ef3abaa88e ro acpi_backlight=vendor quiet splash vt.handoff=7
RelatedPackageVersions:
 linux-restricted-modules-3.8.0-19-generic N/A
 linux-backports-modules-3.8.0-19-generic N/A
 linux-firmware 1.106
SourcePackage: linux
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 09/14/2012
dmi.bios.vendor: Acer
dmi.bios.version: V2.04
dmi.board.asset.tag: Type2 - Board Asset Tag
dmi.board.name: Mimic
dmi.board.vendor: Acer
dmi.board.version: Type2 - Board Version
dmi.chassis.type: 10
dmi.chassis.vendor: Acer
dmi.chassis.version: V2.04
dmi.modalias: dmi:bvnAcer:bvrV2.04:bd09/14/2012:svnAcer:pnV5-171:pvrV2.04:rvnAcer:rnMimic:rvrType2-BoardVersion:cvnAcer:ct10:cvrV2.04:
dmi.product.name: V5-171
dmi.product.version: V2.04
dmi.sys.vendor: Acer

[peter.weber (peter.weber-linux-kernel-bugs)]

Created attachment 88171
config of kernel 3.7.0-rc7

Hello!

The backlight keys (Fn+F8, Fn+F9) stopped working with Kernel 3.7.0-rc7. It works with the current stable Kernel 3.6.8. I can still change the backlight brightness via "System Settings" > "Brightness & Lock" in GNOME. I can't see any message about not working keys in dmesg or the syslog.

I should note, that the brightness-indicator of GNOME shows if I press one of the keys and it moves the progress-bar a ittle bit. After I changedthe brigthness via "System Settings" it is possible to use the keys to change the brightness over a wider range, but not the complete range. Seems to be some side effect.

Maximal movement of of brightness, visuallay:
Default:
[ x x x x x x x - ]
After I changed the brightness to the middle, via "System Settings":
[ x x - - - - x x ]

Sorry for the weird report.
Thank you.

[peter.weber (peter.weber-linux-kernel-bugs)]

Created attachment 88181
dmesg output

[peter.weber (peter.weber-linux-kernel-bugs)]

Hardware is a ThinkPad t430s in UEFI-Mode, theCompatibility Support Module is switched off.

[rui.zhang (rui.zhang-linux-kernel-bugs)]

well, I guess the problem also exists in 3.3 kernel, right?

it seems that this patch introduces the "regression".
commit fba4e087361605d1eed63343bb08811f097c83ee
Author: Igor Murzov <email address hidden>
Date: Sat Oct 13 04:41:25 2012 +0400

    ACPI video: Ignore errors after _DOD evaluation.

    There are systems where video module known to work fine regardless
    of broken _DOD and ignoring returned value here doesn't cause
    any issues later. This should fix brightness controls on some laptops.

    Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=47861

    Signed-off-by: Igor Murzov <email address hidden>
    Reviewed-by: Sergey V <email address hidden>
    Signed-off-by: Zhang Rui <email address hidden>

But IMO, the ACPI backlight control on your laptop is always broken,
but this commit
commit ea9f8856bd6d4ed45885b06a338f7362cd6c60e5
Author: Igor Murzov <email address hidden>
Date: Fri Mar 30 21:32:08 2012 +0400

    ACPI video: Harden video bus adding.

    It is always better to check return values, so add some new checks and
    correct existing ones.

    v2: Be consistent and don't mix errors from -E* and AE_* namespaces.

    Signed-off-by: Igor Murzov <email address hidden>
    Signed-off-by: Len Brown <email address hidden>
happens to make your laptop stop using ACPI backlight control. :)

can you please attach the dmidecode output of your laptop?
so that I can backlist your laptop from ACPI Backlight control.

[peter.weber (peter.weber-linux-kernel-bugs)]

Thank you. I add the output of dmidecode in a few seconds.
You should take also a look at this (Daniel Vetter and someone with a Dell Laptop):
http://www.spinics.net/lists/kernel/msg1446647.html

[peter.weber (peter.weber-linux-kernel-bugs)]

Created attachment 88191
output of dmidecode on a thinkpad t430s, uefi only mode

[peter.weber (peter.weber-linux-kernel-bugs)]

Adding the parameter "acpi_backlight=vendor" as kernel option to the bootloader makes the Fn-Keys working again, like assumed by Daniel Vetter. But than the UI of GNOME is broken.

[peter.weber (peter.weber-linux-kernel-bugs)]

I can't confirm that the problem exists in Kernel 3.3 or not, because my first Kernel on that device was a Kernel of the 3.5 series. And with Kernel 3.5 and 3.6 everything worked.

[peter.weber (peter.weber-linux-kernel-bugs)]

I tried the change blacklist.c, but it doesn't work for me. See attachment. I also wonder about the last empty curly braces in the struct (not added by me)?

[peter.weber (peter.weber-linux-kernel-bugs)]

Created attachment 89941
not working modified backlist.c

[peter.weber (peter.weber-linux-kernel-bugs)]

Question:
$ ls -l
total 0
lrwxrwxrwx 1 root root 0 Dec 30 19:02 acpi_video0 -> ../../devices/pci0000:00/0000:00:02.0/backlight/acpi_video0
lrwxrwxrwx 1 root root 0 Dec 30 19:02 intel_backlight -> ../../devices/pci0000:00/0000:00:02.0/drm/card0/card0-LVDS-1/intel_backlight

If I understand your noted patches above, it is *okay* to see both directorie on kernel 3.6 and 3.7. The acpi_video0 just doesn't block intel_backlight on kernel 3.6?

[peter.weber (peter.weber-linux-kernel-bugs)]

Adding following parameters to boot command line seems to be a mostly working workaround:

# first switches of thinkpad_acpi, seconds switches of acpi_video
thinkpad-acpi.brightness_enable=0 acpi_backlight=vendor

* Fn+BrightnessDown, Fn+BrighnessUp works
* GNOME Power-Widget displays change of value through Fn-Keys
* GNOME > System Settings > Brighness works
* ls -l /sys/class/backlight only lists intel_backlight (not acpi or thinkpad)

Only automatic display-dimming through GNOME seems not to work.

[daniel.m.kao (daniel.m.kao-linux-kernel-bugs)]

I have done a git bisect and found the offending commit:
commit a57f7f9175b8ccbc9df83ac13860488913115de4
Author: Bob Moore <email address hidden>
Date: Fri Aug 17 10:55:02 2012 +0800

    ACPICA: Add Windows8/Server2012 string for _OSI method.

    This change adds a new _OSI string, "Windows 2012" for both Windows 8
    and Windows Server 2012.

    From Microsoft document "How to Identify the Windows Version in ACPI
    by Using _OSI", July 13, 2012.

    Signed-off-by: Bob Moore <email address hidden>
    Signed-off-by: Feng Tang <email address hidden>
    Signed-off-by: Len Brown <email address hidden>

Reverting this fixes the issue for me on ThinkPad X230.

[rui.zhang (rui.zhang-linux-kernel-bugs)]

oh, please attach the acpidump output of your laptop.

[rui.zhang (rui.zhang-linux-kernel-bugs)]

BTW, does adding kernel parameter acpi_osi="!Windows 2012" help?

[daniel.m.kao (daniel.m.kao-linux-kernel-bugs)]

Created attachment 90681
acpidump output for ThinkPad X230

[daniel.m.kao (daniel.m.kao-linux-kernel-bugs)]

acpi_osi="!Windows 2012" works, I've been using that instead of the modified kernel.
Thanks.

[peter.weber (peter.weber-linux-kernel-bugs)]

Thats weird! I will test that also and tell you if it works.

Am I right, that this will tell the ACPI based functions of the kernel that the BIOS/UEFI doesn't support Windows 2012?

[peter.weber (peter.weber-linux-kernel-bugs)]

I can confirm that this works!

ls -l /sys/class/backlight/
total 0
lrwxrwxrwx 1 root root 0 Jan 8 22:20 acpi_video0 -> ../../devices/pci0000:00/0000:00:02.0/backlight/acpi_video0
lrwxrwxrwx 1 root root 0 Jan 8 22:20 intel_backlight -> ../../devices/pci0000:00/0000:00:02.0/drm/card0/card0-LVDS-1/intel_backlight

Booth seem to work, again, adjacent to another.
But the automatic display dimming via GNOME still seems not to work. I wonder why, but it is not really important.

[seth (seth-linux-kernel-bugs)]

I took a look at the ACPI tables from an x230. If the OS says it's Windows 2012 then _BCL gives an alternate table of brightness values to meet an arbitrary requirement of Win8 to have 101 brightness levels [1]. _BCM tries to find a match for the brightness level it's given in the smaller table it uses for non-Win8 OSes. If it doesn't find an exact match then it just returns without applying any brightness change, so of the 101 values that _BCL claims to support only a much smaller subset (16) actually affect a change in the brightness.

[1] http://msdn.microsoft.com/en-us/library/windows/hardware/jj128256.aspx

[seth (seth-linux-kernel-bugs)]

I'm not seeing any way to fix this other than doing some sort of quirking for the affected machines.

One option add a quirk to keep the kernel from claiming to be Windows 8. There are other places in the ACPI tables where behavior will be affected, but since folks have been running older kernels successfully on these machines I suspect everything should work okay.

The other option I can think of is to quirk acpi-video so that it will only use the brightness values that actually work. I don't see any way to do this without hard coding firmware implementation details into the driver though, which is ugly.

Any other ideas?

[daniel.m.kao (daniel.m.kao-linux-kernel-bugs)]

I see that _BCM and _BQC only work with the 16 step table (\_SB.PCI0.LPC.EC.BRTW).
When interacting with the backlight through sysfs, brightness node can be set to any value from _BCL, but actual_brightness reports its current value from _BQC. This causes a discrepancy between brightness and actual_brightness where brightness could be 75 but actual_brightness remains at 70.

Gnome uses RandR to change backlight, which reports actual_brightness when available. When Gnome tries to change brightness in reaction to media keys, it tries to do so in 5% increments based on actual_brightness. Now, since 75% is not supported in the BRTW table, stepping up from 70% would never succeed and so Gnome's brightness OSD gets stuck at 75% and cannot go any higher. Same thing happens when trying to step down from 20%.

I have created a test kernel by commenting out the get_brightness entry in acpi_backlight_ops. RandR seems to fallback to the brightness node and everything works fine in Gnome.

Brightness adjustment with media keys in virtual terminals still has no effect. But I have no idea how the brightness keys are handled in VT so I can't comment on that.

[daniel.m.kao (daniel.m.kao-linux-kernel-bugs)]

Sorry, made a mistake in the previous comment:
RandR backlight control doesn't work without a valid actual_brightness node so Gnome falls back to its own backlight helper which uses the brightness node.

[seth (seth-linux-kernel-bugs)]

(In reply to comment #21)
> I see that _BCM and _BQC only work with the 16 step table
> (\_SB.PCI0.LPC.EC.BRTW).
> When interacting with the backlight through sysfs, brightness node can be set
> to any value from _BCL, but actual_brightness reports its current value from
> _BQC. This causes a discrepancy between brightness and actual_brightness
> where
> brightness could be 75 but actual_brightness remains at 70.

Yes, since _BCM only discards the value and doesn't return an error, acpi_video thinks the brightness change was successful and thus updates the cached brightness value. actual_brightness shows that the firmware discarded the value.

> Gnome uses RandR to change backlight, which reports actual_brightness when
> available. When Gnome tries to change brightness in reaction to media keys,
> it
> tries to do so in 5% increments based on actual_brightness. Now, since 75% is
> not supported in the BRTW table, stepping up from 70% would never succeed and
> so Gnome's brightness OSD gets stuck at 75% and cannot go any higher. Same
> thing happens when trying to step down from 20%.
>
> I have created a test kernel by commenting out the get_brightness entry in
> acpi_backlight_ops. RandR seems to fallback to the brightness node and
> everything works fine in Gnome.

We shouldn't expose the backlight to userspace if it isn't going to work properly. So the possible solutions I can think of are quirking to use "!Windows 2012" by default, quirking to disable acpi_video backlight, or quirking somehow to make acpi_video only use the values that actually work.

> Brightness adjustment with media keys in virtual terminals still has no
> effect.
> But I have no idea how the brightness keys are handled in VT so I can't
> comment
> on that.

This depends to some degree on how the firmware behaves. Most of the time a keycode is emitted when you press the brightness keys which userspace needs to respond to. When you're in a VT there's probably no software responding to the brightness keycodes, thus no change in brightness.

[peter.weber (peter.weber-linux-kernel-bugs)]

Created attachment 91091
acpidump of thinkpad t430s with uefi 2.05 in uefi-only mode, without csm

acpidump in uefi-only mode, compatibility-support-module is off

[seth (seth-linux-kernel-bugs)]

Created attachment 91311
Disable Windows 8 compatibility for ThinkPad T430s and X230

Here's a patch (based on 3.8-rc3) which disables Windows 8 compatibility for the T430s and X230 in ACPI. Please give it a try.

[Brad Figg (brad-figg)]

This change was made by a bot.

[Joseph Salisbury (jsalisbury)]

Would it be possible for you to test the latest upstream kernel? Refer to https://wiki.ubuntu.com/KernelMainlineBuilds . Please test the latest v3.9 kernel[0].

If this bug is fixed in the mainline kernel, please add the following tag 'kernel-fixed-upstream'.

If the mainline kernel does not fix this bug, please add the tag: 'kernel-bug-exists-upstream'.

If you are unable to test the mainline kernel, for example it will not boot, please add the tag: 'kernel-unable-to-test-upstream'.
Once testing of the upstream kernel is complete, please mark this bug as "Confirmed".

Thanks in advance.

[0] http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.9-raring/

[Alan Jenkins (aj504)]

Thanks! Booted vmlinuz-3.9.0-030900-generic without the parameter, and the backlight still cannot be adjusted.

[penalvch (penalvch)]

Alan Jenkins, as per http://us.acer.com/ac/en/US/content/drivers an update is available for your BIOS (2.15). If you update to this, does it change anything?

If not, could you please both specify what happened, and provide the output of the following terminal command:
sudo dmidecode -s bios-version && sudo dmidecode -s bios-release-date

Thank you for your understanding.

[Alan Jenkins (aj504)]

Sorry I've been neglecting this (and my original report may have fluffed the detail).

This is Windows 8 hardware. Manually poking the intel_backlight driver in sysfs (echo 100 | sudo tee /sys/class/intel_backlight/brightness) works fine. So for now I'm assuming the problem is what's referred to here:

http://permalink.gmane.org/gmane.linux.kernel.commits.head/396675

(I first saw this referred to when it was reverted in 3.11-rc3. Linus' release announcement says "But never fear, we have top people looking at it").

That should be nice & easy to test compared to the _other_ bug I'm tracking on this laptop. So I'll see if I can get there & feedback to upstream.

[Alan Jenkins (aj504)]

That is, I'm hoping the brightness keys will work (e.g. under KDE) when using the commit "ACPI / video / i915: No ACPI backlight if firmware expects Windows 8". With the non-working acpi_backlight removed, KDE would use the working intel_backlight instead.

[Alan Jenkins (aj504)]

Sorry, I don't think the backlight problem on the V5-171 is a regression. I've never had it work as expected.

I've now tried the acpi_osi="!Windows 2012" test, to see if this machine could be added to the temporary blacklist. I.e. whether my bug was a regression caused when Linux added the "Windows 2012" OSI string.[1] Unfortunately that boot option didn't help.

[1] https://bugzilla.kernel.org/show_bug.cgi?id=51231#c137

The result from testing v3.11-rc2 is that it does fix the problem. And as expected, the revert in v3.11-rc3 breaks it again.

To answer the question about firmware upgrades, I'm not going to try that unless it helps upstream. It's not worth reflashing "at my own risk" just to avoid passing acpi_backlight=vendor.

[peter.weber (peter.weber-linux-kernel-bugs)]

Created attachment 111591
dmidecode video.use_native_backlight -> not set

kernel 3.12.0-rc5 with patch_set_v5

[peter.weber (peter.weber-linux-kernel-bugs)]

Created attachment 111601
kernel config, kernel 3.12.0-rc5 with patch_set_v5

[peter.weber (peter.weber-linux-kernel-bugs)]

Thanks for your work on this Aaron! I've upgraded to kernel 3.12.0-rc5 and applied the patches. When I set video.use_native_backlight=1:

* everything works fine!
* /sys/class/backlight contains only intel_backlight
* there seem to be 21 brightness-levels available with gnome-power-manager
* the lowest level turns the backlight completely off (this doesn't happend with acpi_osi="!Windows 2012"), is this intentional?
* as already expected it is not longer possible to tune the brightness-level with the Fn+Keys, while working on a virtual-terminal
* thinkpad_acpi seem unhappy about the brightness-interface (see dmesg)

> thinkpad_acpi: http://ibm-acpi.sf.net/
> thinkpad_acpi: Unsupported brightness interface, please contact
> <email address hidden>

Looks good.

[peter.weber (peter.weber-linux-kernel-bugs)]

Created attachment 111611
dmesg with complain from thinkpad_acpi about brightness interface

unsupported brightness interface

[TomWij (tomwij-linux-kernel-bugs)]

Peter: Can you change this bug from NEEDINFO to UNCONFIRMED or CONFIRMED? I don't know for sure; but, due to the high traffic, the assigned person might miss your responses if the bug doesn't appear back in his list. It might be the possibility that only the assigned person can change, in which case never mind...

[peter.weber (peter.weber-linux-kernel-bugs)]

Ohhhh. I thought the active developers change that if appropriate. I changed the status to VERIFIED, because there is no CONFIRMED available.

[TomWij (tomwij-linux-kernel-bugs)]

Eh, VERIFIED is usually a status to mark it resolved and verified by QA; you'll not want to pick that one, see https://bugzilla.kernel.org/page.cgi?id=fields.html#status for details on the various statuses.

[peter.weber (peter.weber-linux-kernel-bugs)]

The naming is pretty weird, isn't it? I don't have the canconfirm-permission. I hesitate to set this bug to "RESOLVED" but the description fits. Maybe Aaron would set this to IN_PROGRESS.

[aaron.lu (aaron.lu-linux-kernel-bugs)]

Thanks Peter for the test. If you show me your dmi info, I can add your system into a list so that the mentioned param will be automatically set for you.

[tytso (tytso-linux-kernel-bugs)]

I can confirm that I also need video.use_native_backlight=1 plus this patch (applied to 3.12-rc5) in order to make things work on my T430s laptop:

Handle 0x000F, DMI type 1, 27 bytes
System Information
        Manufacturer: LENOVO
        Product Name: 2352CTO
        Version: ThinkPad T430s
        Wake-up Type: Power Switch
        SKU Number: LENOVO_MT_2352
        Family: ThinkPad T430s

[aaron.lu (aaron.lu-linux-kernel-bugs)]

Created attachment 111821
Add Thinkpad T430s into the list

For Thinkpad T430s, apply on top of the existing v5 patchset.

[peter.weber (peter.weber-linux-kernel-bugs)]

(In reply to Aaron Lu from comment #152)
> Thanks Peter for the test. If you show me your dmi info, I can add your
> system into a list so that the mentioned param will be automatically set for
> you.

Theodore posted it in meanwhile :-)
You can find the complete output in comment 143. I will add the patch on top of v5 within the next days, but doesn't expect any surprise.

[peter.weber (peter.weber-linux-kernel-bugs)]

Created attachment 112141
fixed typo, ThinkPad is written in camelCase, Thinkpad -> ThinkPad

[peter.weber (peter.weber-linux-kernel-bugs)]

Created attachment 112151
fixed typo, Thinkpad -> ThinkPad

ThinkPad brand name is written in CamelCaseas, as reported by "dmidecode -t 1".

[peter.weber (peter.weber-linux-kernel-bugs)]

Great! I'm myself not able to type CamelCase without a typo. To be clear, is necessary to change it to "ThinkPad" to make it working!

By the way, let me be pedantic :-)
0003-ACPI-video-Do-not-register-backlight-if-win8-and-nat.patch
Line 54:
- * For Windows 8 systems: if set ture and the GPU driver has
+ * For Windows 8 systems: if set true and the GPU driver has

Thank you! Good night!

[thirdwiggin (thirdwiggin-linux-kernel-bugs)]

I tried the v5 patchset on 3.12.0-rc5 on my Lenovo Yoga 13, and video.use_native_backlight=1 also fixes the issue on this machine. I'm glad to help with any future testing needed to get this patch merged. Thanks for fixing this!

[aaron.lu (aaron.lu-linux-kernel-bugs)]

*** Bug 63811 has been marked as a duplicate of this bug. ***

[aaron.lu (aaron.lu-linux-kernel-bugs)]

*** Bug 67031 has been marked as a duplicate of this bug. ***

[aaron.lu (aaron.lu-linux-kernel-bugs)]

*** Bug 68751 has been marked as a duplicate of this bug. ***

[aaron.lu (aaron.lu-linux-kernel-bugs)]

*** Bug 62941 has been marked as a duplicate of this bug. ***

[aaron.lu (aaron.lu-linux-kernel-bugs)]

commit 1811fcb029fa3eca5bd1d25eab386b0c4b80fb93
Author: Aaron Lu <email address hidden>
Date: Tue Feb 18 13:54:20 2014 +0800

    ACPI / video: Add systems that should favour native backlight interface

has entered Rafael's linux-next tree.

[fuffi.il.fuffo (fuffi.il.fuffo-linux-kernel-bugs)]

Hi,
Thinkpad T430 here (without the s), I'm using linux 3.13.4 and systemd 208 from arch linux booting in bios mode with the last bios update (2.62).

The bug seems the same as for T430s and I can "fix" it with acpi_osi='!Windows 2012' or using video.use_native_backlight=1 as kernel command line in grub.

It works kinda well with video.use_native_backlight=1 except that I have now 4437 level of brightness in /sys/class/backlight/intel_backlight and as reported by Weber, systemd complains about this:

[ 9.226063] thinkpad_acpi: Unsupported brightness interface, please contact <email address hidden>

System Information
        Manufacturer: LENOVO
        Product Name: 2349KDG
        Version: ThinkPad T430
        (...)
        Wake-up Type: Power Switch
        SKU Number: LENOVO_MT_2349
        Family: ThinkPad T430

It is because of so many levels of backlight that I have in KDE this behavior switching brightness with the keys:
from 0 to 100: 0-10-20-30-40-50-60-70-80-90-100
from 100 to 0: 100-89-79-69...etc.

[aaron.lu (aaron.lu-linux-kernel-bugs)]

(In reply to edm from comment #165)
> Hi,
> Thinkpad T430 here (without the s), I'm using linux 3.13.4 and systemd 208
> from arch linux booting in bios mode with the last bios update (2.62).
>
> The bug seems the same as for T430s and I can "fix" it with
> acpi_osi='!Windows 2012' or using video.use_native_backlight=1 as kernel
> command line in grub.
>
> It works kinda well with video.use_native_backlight=1 except that I have now
> 4437 level of brightness in /sys/class/backlight/intel_backlight and as
> reported by Weber, systemd complains about this:
>
> [ 9.226063] thinkpad_acpi: Unsupported brightness interface, please
> contact <email address hidden>

This is harmless, I'll send a patch to fix this.

>
>
> System Information
> Manufacturer: LENOVO
> Product Name: 2349KDG
> Version: ThinkPad T430
> (...)
> Wake-up Type: Power Switch
> SKU Number: LENOVO_MT_2349
> Family: ThinkPad T430
>
> It is because of so many levels of backlight that I have in KDE this
> behavior switching brightness with the keys:
> from 0 to 100: 0-10-20-30-40-50-60-70-80-90-100
> from 100 to 0: 100-89-79-69...etc.

I'll prepare a new patch for your laptop later(together with some other model once that model is confirmed to work well with video.use_native_backlight=1).

[aaron.lu (aaron.lu-linux-kernel-bugs)]

commit 0e9f81d3b7cd0649a3bc437391b6a0650f98f844
Author: Aaron Lu <email address hidden>
Date: Tue Feb 18 13:54:20 2014 +0800

    ACPI / video: Add systems that should favour native backlight interface

entered Linus' tree as v3.14 material.

[aaron.lu (aaron.lu-linux-kernel-bugs)]

BTW, edm, please also attach your acpidump here, thanks.

[fuffi.il.fuffo (fuffi.il.fuffo-linux-kernel-bugs)]

Created attachment 127991
acpidump T430

Output of
# acpidump > acpidump-T430.txt

[fuffi.il.fuffo (fuffi.il.fuffo-linux-kernel-bugs)]

Just a stupid question: does it make any difference doing "acpidump" with or without acpi_osi="!Windows 2012" as boot option?

[aaron.lu (aaron.lu-linux-kernel-bugs)]

(In reply to edm from comment #170)
> Just a stupid question: does it make any difference doing "acpidump" with or
> without acpi_osi="!Windows 2012" as boot option?

No difference.
I'll add your system into the table shortly, sorry for the long delay.

[lyude (lyude-linux-kernel-bugs)]

Created attachment 131691
acpi dump for ThinkPad Helix (UEFI)

[lyude (lyude-linux-kernel-bugs)]

Created attachment 131701
dmidecode dump for ThinkPad Helix

That should be all the information you need on the Helix regarding the patch I submitted ( https://patchwork.kernel.org/patch/3938911/ ), let me know if you need anything else

[Jonatã Bolzan Loss (jbopen)]

Bug still present in Ubuntu 14.04

[penalvch (penalvch)]

jonata, thank you for your comment. So your hardware and problem may be tracked, could you please file a new report with Ubuntu by executing the following in a terminal while booted into a Ubuntu repository kernel (not a mainline one) via:
ubuntu-bug linux

For more on this, please read the official Ubuntu documentation:
Ubuntu Bug Control and Ubuntu Bug Squad: https://wiki.ubuntu.com/Bugs/BestPractices#X.2BAC8-Reporting.Focus_on_One_Issue
Ubuntu Kernel Team: https://wiki.ubuntu.com/KernelTeam/KernelTeamBugPolicies#Filing_Kernel_Bug_reports
Ubuntu Community: https://help.ubuntu.com/community/ReportingBugs#Bug_reporting_etiquette

When opening up the new report, please feel free to subscribe me to it.

Thank you for your understanding.

Helpful bug reporting tips:
https://wiki.ubuntu.com/ReportingBugs

[aaron.lu (aaron.lu-linux-kernel-bugs)]

(In reply to Aaron Lu from comment #171)
> (In reply to edm from comment #170)
> > Just a stupid question: does it make any difference doing "acpidump" with
> or
> > without acpi_osi="!Windows 2012" as boot option?
>
> No difference.
> I'll add your system into the table shortly, sorry for the long delay.

Matthew sent out a patch to enable native backlight control interface by default for win8 systems so that we do not need to add more systems into the DMI table:
http://www.spinics.net/lists/dri-devel/msg57228.html

[Nils Sabelstrom (nilssab)]

Bug can be confirmed present in 14.04 with standard LTS kernel and 3.15rc2 kernel from ubuntu mainline kernel-ppa, daily build for april 24th (latest marked trusty at the time, now nonexistent, should I try a newer kernel even if marked utopic?)

I tried upgrading to bios version 2.21, but there was only an installer for win 8, and using it removes the ubuntu boot loader. Will post after getting my computer to boot ubuntu again.. (I would not recommend users without good knowledge of how UEFI bootloaders work to try upgrading the bios)

[Nils Sabelstrom (nilssab)]

will do, as soon as I get it booting into ubuntu again.
Apparently update-grub (or rather grub-probe) doesn't work when chrooting into a root partition if it's btrfs, so I'm gonna try booting through a efi booter from another computer running the same kernelversion of 14.04..
I'm quite pissed at myself for not backing up the EFI partition before doing a BIOS update from windows... <_<

[Nils Sabelstrom (nilssab)]

added a duplicate bug generated by ubuntu-bug.
is there a way to make it upload the information to a specific bug and not make a new bug entry? or is it fine to just make duplicates? should I also make one using the newest kernel?

[gabriele.mzt (gabriele.mzt-linux-kernel-bugs)]

Created attachment 137791
Dell XPS13 9333 acpidump

I attached the acpidump of my Dell XPS13 9333.

If video.use_native_backlight=1 is not added to the kernel cmdline, the Fn keys are buggy: the brightness increases or decreases by one unit (max_brigthness is 937, so a variation of 1 is hardly noticeable) and I get not DE OSD.

With video.use_native_backlight=1, I get the OSD and a brightness variations of 10% as expected.

System Information
        Manufacturer: Dell Inc.
        Product Name: XPS13 9333

[robert.zavalczki (robert.zavalczki-linux-kernel-bugs)]

Created attachment 140631
ThinkPad T530 use native backlight patch

[robert.zavalczki (robert.zavalczki-linux-kernel-bugs)]

The backlight keys do not work properly with kernel 3.15.1 and ThinkPad T530 (with Intel integrated graphics). Only 2 brightness levels are available.

I noticed that using native backlight option for Thinkpad T530 fixes the issue:

see attachment 140631

After applying the above patch 11 brightness levels + blank can be set (I am using an Xfce 4.10 desktop).

Weirdly, in pre 3.15 kernel versions adding the use native backlight option caused my monitor to become unresponsive (stay blank) after a screen lock followed by a suspend screen. I no longer get that behavior with kernel version 3.15.1. Should you need any more information I can happily provide it.

[corsac (corsac-linux-kernel-bugs)]

So I just got the 3.16 kernel with the changes for all Windows 8 laptops.

And while I bought this laptop (Thinkpad x201s) in 2010, it seems a BIOS update added the “support” for Windows 8 (at least that's a guess from http://paste.debian.net/120230/ , I can provide full acpidump).

So this laptop where brightness keys where working perfectly using ACPI, has now the same issue, and I need to have something in userspace doing the job.

That's really a regression for me.

[aaron.lu (aaron.lu-linux-kernel-bugs)]

Hi Perez,

I believe commit 5f24079b021cd3147c8d24ba65833f7a0df7e80d is meant to fix such problems:

commit 5f24079b021cd3147c8d24ba65833f7a0df7e80d
Author: Hans de Goede <email address hidden>
Date: Thu Aug 28 10:20:46 2014 +0200

    ACPI / video: Add a disable_native_backlight quirk

    Some laptops have a working acpi_video backlight control, and using native
    backlight on these causes a regression where backlight control does not work
    when userspace is not handling brightness key events. Disable native_backlight
    on these to fix this.

    Link: https://bugzilla.kernel.org/show_bug.cgi?id=81691
    Reported-and-tested-by: Andre Müller <email address hidden>
    Cc: 3.16+ <email address hidden> # 3.16+
    Signed-off-by: Hans de Goede <email address hidden>
    Signed-off-by: Rafael J. Wysocki <email address hidden>

Feel free to submit a patch to add your model into the video_dmi_table to disable native backlight control.

[corsac (corsac-linux-kernel-bugs)]

(In reply to Aaron Lu from comment #179)
> Hi Perez,
>
> I believe commit 5f24079b021cd3147c8d24ba65833f7a0df7e80d is meant to fix
> such problems:
>
> commit 5f24079b021cd3147c8d24ba65833f7a0df7e80d
> Author: Hans de Goede <email address hidden>
> Date: Thu Aug 28 10:20:46 2014 +0200
>
> ACPI / video: Add a disable_native_backlight quirk
>
> Some laptops have a working acpi_video backlight control, and using
> native
> backlight on these causes a regression where backlight control does not
> work
> when userspace is not handling brightness key events. Disable
> native_backlight
> on these to fix this.
>
> Link: https://bugzilla.kernel.org/show_bug.cgi?id=81691
> Reported-and-tested-by: Andre Müller <email address hidden>
> Cc: 3.16+ <email address hidden> # 3.16+
> Signed-off-by: Hans de Goede <email address hidden>
> Signed-off-by: Rafael J. Wysocki <email address hidden>
>
> Feel free to submit a patch to add your model into the video_dmi_table to
> disable native backlight control.

Thanks. So the “DMI table” approach was considered bad because not sustainable for this bug, but ok for bug #81691 ?

Anyway, will provide the correct DMI information on that bug.

[aaron.lu (aaron.lu-linux-kernel-bugs)]

(In reply to Yves-Alexis Perez from comment #180)
> (In reply to Aaron Lu from comment #179)
> > Hi Perez,
> >
> > I believe commit 5f24079b021cd3147c8d24ba65833f7a0df7e80d is meant to fix
> > such problems:
> >
> > commit 5f24079b021cd3147c8d24ba65833f7a0df7e80d
> > Author: Hans de Goede <email address hidden>
> > Date: Thu Aug 28 10:20:46 2014 +0200
> >
> > ACPI / video: Add a disable_native_backlight quirk
> >
> > Some laptops have a working acpi_video backlight control, and using
> > native
> > backlight on these causes a regression where backlight control does not
> > work
> > when userspace is not handling brightness key events. Disable
> > native_backlight
> > on these to fix this.
> >
> > Link: https://bugzilla.kernel.org/show_bug.cgi?id=81691
> > Reported-and-tested-by: Andre Müller <email address hidden>
> > Cc: 3.16+ <email address hidden> # 3.16+
> > Signed-off-by: Hans de Goede <email address hidden>
> > Signed-off-by: Rafael J. Wysocki <email address hidden>
> >
> > Feel free to submit a patch to add your model into the video_dmi_table to
> > disable native backlight control.
>
> Thanks. So the “DMI table” approach was considered bad because not
> sustainable for this bug, but ok for bug #81691 ?

The systems that claim to be Win8 but have a working ACPI video interface are supposed to be not common(hopefully this is true), so should be OK to keep a DMI table for them. While the systems that claim to be Win8 and don't have a working ACPI video interface are supposed to be the common case, so keeping a DMI table for all of them is not feasible.

>
> Anyway, will provide the correct DMI information on that bug.

OK, thanks.

[ucelsanicin (ucelsanicin-linux-kernel-bugs)]

With this patch (not sure yet whether it's relevant) in place:
... http://www-look-4.com/category/computers/
diff --git a/gdb/testsuite/lib/gdbserver-support.exp b/gdb/testsuite/lib/gdbserver-support.
Exp https://komiya-dental.com/health/healthy-foods/
index a2cc80f28d..7b9c0eef6e 100644
--- a/gdb/testsuite/lib/gdbserver-support.exp http://www.iu-bloomington.com/services/travel-services/
+++ b/gdb/testsuite/lib/gdbserver-support.exp
@@ -451,8 +451,10 @@ proc gdbserver_exit { is_mi } { https://waytowhatsnext.com/technology/korean-technology/
            # We use expect rather than gdb_expect because
            # we want to suppress printing exception messages, otherwise, http://www.wearelondonmade.com/category/health/
            # remote_expect, invoked by gdb_expect, prints the exceptions.
+ set read_prompt 0
            expect { http://www.jopspeech.com/category/property/
                -i "$gdb_spawn_id" -re "$gdb_prompt $" {
+ set read_prompt 1 http://joerg.li/category/technology/
                    exp_continue
                }
                -i "$server_spawn_id" eof { http://connstr.net/property/mars-researches/
@@ -463,6 +465,7 @@ proc gdbserver_exit { is_mi } {
                    warning "Timed out waiting for EOF in server after $monitor_exit"
                }
}
+ gdb_assert {$read_prompt}
        }
     } http://embermanchester.uk/health/social-privacy/
     close_gdbserver
...
and running in parallel with:
...
$ stress -c 5 http://www.slipstone.co.uk/category/services/
...
I ran into:
...
(gdb) PASS: gdb.multi/multi-target.exp: continue: non-stop=on: inferior 2 http://www.logoarts.co.uk/category/travel/
Remote debugging from host ::1, port 34088^M
Process build/gdb/testsuite/outputs/gdb.multi/multi-target/multi-target created; pid = 8649^M
monitor exit^M
(gdb) Killing process(es): 8649^M http://www.acpirateradio.co.uk/technology/facetime/
#9 0x16a2c57 in pop_all_targets_at_and_above(strata) /home/vries/gdb_versions/devel/src/gdb/target.c:678
    #10 0x1442749 in remote_unpush_target http://www.compilatori.com/tech/xiaomi/ /home/vries/gdb_versions/devel/src/gdb/remote.c:5522
    #11 0x1458c16 in remote_target::readchar(int) /home/vries/gdb_versions/devel/src/gdb/remote.c:9137 https://www.webb-dev.co.uk/sports/how-to-choose-sportwear/
    #12 0x145b25b in remote_target::getpkt_or_notif_sane_1(std::vector<char, gdb::default_init_allocator<char, std::allocator<char> > >*, int, int, int*)