samsung-laptop: Samsung 530U3C major ACPI issues

samsung-laptop: Samsung 530U3C major ACPI issues

[Sebastian Riemer]

[-- Attachment #1: Type: text/plain, Size: 997 bytes --]

Hi Corentin,

as you're the maintainer of the samsung-laptop driver I think that you
can help with getting support for the Samsung 530U3C laptop into it.

My colleague Adrian reports the following ACPI issues:
- the closing of the lid isn't detected
- if the DC power supply is removed, this isn't detected
- Wifi LED is always turned on even if Wifi is deactivated
- high battery consumption

Attached you can find the debug data from "dmesg" and "lshw" from Fedora
17. The kernel messages already indicate ACPI issues.

How can we support you with getting the 530U3C support into the
samsung-laptop driver?

Cheers,
Sebastian


-- 
Sebastian Riemer
Linux Kernel Developer - Storage

We are looking for (SENIOR) LINUX KERNEL DEVELOPERS!

ProfitBricks GmbH • Greifswalder Str. 207 • 10405 Berlin, Germany
www.profitbricks.com • sebastian.riemer@profitbricks.com

Sitz der Gesellschaft: Berlin
Registergericht: Amtsgericht Charlottenburg, HRB 125506 B
Geschäftsführer: Andreas Gauger, Achim Weiss


[-- Attachment #2: debug.tar.gz --]
[-- Type: application/x-gzip, Size: 20327 bytes --]

Re: samsung-laptop: Samsung 530U3C major ACPI issues

[Corentin Chary]

Hi,

On Wed, Oct 24, 2012 at 3:39 PM, Sebastian Riemer
<sebastian.riemer@profitbricks.com> wrote:
> Hi Corentin,
>
> as you're the maintainer of the samsung-laptop driver I think that you
> can help with getting support for the Samsung 530U3C laptop into it.

samsung-laptop doesn't use ACPI, only SABI interface.

> My colleague Adrian reports the following ACPI issues:
> - the closing of the lid isn't detected
> - if the DC power supply is removed, this isn't detected

This looks like ACPI issues, please file a bug on bugzilla, but first
thing, check for bios updates :).

> - Wifi LED is always turned on even if Wifi is deactivated

I don't know what handle the LED here .. anything in /sys/class/led ?

> - high battery consumption

See powertop for that, may helps. Also check various graphic card
parameters for power consumption.

> Attached you can find the debug data from "dmesg" and "lshw" from Fedora
> 17. The kernel messages already indicate ACPI issues.
>
> How can we support you with getting the 530U3C support into the
> samsung-laptop driver?

Unfortunately, this doesn't seems to be related to samsung-laptop :).

-- 
Corentin Chary
http://xf.iksaif.net

Re: samsung-laptop: Samsung 530U3C major ACPI issues

[Sebastian Riemer]

Hi Corentin,

On 24.10.2012 17:24, Corentin Chary wrote:
> samsung-laptop doesn't use ACPI, only SABI interface. 

I thought it does for the video driver/backlight control. The strange
thing is that the issues start directly after loading the driver.
Perhaps it has to be blacklisted for this model.

Kernel log:

samsung_laptop: detected SABI interface: SwSmi@
samsung_laptop: Backlight controlled by ACPI video driver
ACPI Warning: 0x000000000000efa0-0x000000000000efbf SystemIO conflicts
with Region \_SB_.PCI0.SBUS.SMBI 1 (20120711/utaddress-251)
ACPI: If an ACPI driver is available for this device, you should use it
instead of the native driver

Where are the custom ACPI drivers in the kernel tree if not in
drivers/platform/x86 ?
E.g. "eeepc-laptop" registers something like a custom ACPI driver if I
see that right.

> This looks like ACPI issues, please file a bug on bugzilla, but first
> thing, check for bios updates :).

Which bugzilla? Sorry, I'm not an ACPI specialist.

Cheers,
Sebastian

Re: samsung-laptop: Samsung 530U3C major ACPI issues

[Corentin Chary]

On Wed, Oct 24, 2012 at 5:05 PM, Sebastian Riemer
<sebastian.riemer@profitbricks.com> wrote:
> Hi Corentin,
>
> On 24.10.2012 17:24, Corentin Chary wrote:
>> samsung-laptop doesn't use ACPI, only SABI interface.
>
> I thought it does for the video driver/backlight control. The strange
> thing is that the issues start directly after loading the driver.
> Perhaps it has to be blacklisted for this model.

samsung-laptop talks to the bios directly using SABI. But the BIOS is
also related to ACPI, so things can get ugly I guess...

> Kernel log:
>
> samsung_laptop: detected SABI interface: SwSmi@
> samsung_laptop: Backlight controlled by ACPI video driver
> ACPI Warning: 0x000000000000efa0-0x000000000000efbf SystemIO conflicts
> with Region \_SB_.PCI0.SBUS.SMBI 1 (20120711/utaddress-251)
> ACPI: If an ACPI driver is available for this device, you should use it
> instead of the native driver

Hum can you try to blacklist samsung-laptop, load it manually, and
send me the dmesg ?
If possible with and without acpi_backlght=vendor boot parameter.

> Where are the custom ACPI drivers in the kernel tree if not in
> drivers/platform/x86 ?

drivers/paltform/x86 contains x86 platform drivers, and some of them
are ACPI drivers, but not all of them.

>> This looks like ACPI issues, please file a bug on bugzilla, but first
>> thing, check for bios updates :).
>
> Which bugzilla? Sorry, I'm not an ACPI specialist.

The linux kernel bugzilla. But before doing that, are this issues only
hapenning when/if samsung-laptop is loaded ?
Can you send me he output of:

tree /sys/devices/platform/samsung-laptop/
tree /sys/class/backlight

Thanks


-- 
Corentin Chary
http://xf.iksaif.net

Re: samsung-laptop: Samsung 530U3C major ACPI issues

[Sebastian Riemer]

Hi Corentin,

I've found the related bugs in the kernel bugzilla. These are really
ACPI issues.

https://bugzilla.kernel.org/show_bug.cgi?id=45461
https://bugzilla.kernel.org/show_bug.cgi?id=44161

If my colleague Adrian wouldn't have removed Windows from his laptop,
then it would be possible to do some reverse engineering, I guess.
Where do vendor specific ACPI/WMI drivers hide for Windows? In the
chipset drivers? SMBus?
Special are also the HECI/MEI devices with that laptop model. Haven't
seen such drivers before - these are also in the linux staging drivers.

Cheers,
Sebastian


On 24.10.2012 20:21, Corentin Chary wrote:
> On Wed, Oct 24, 2012 at 5:05 PM, Sebastian Riemer
> <sebastian.riemer@profitbricks.com> wrote:
>> Hi Corentin,
>>
>> On 24.10.2012 17:24, Corentin Chary wrote:
>>> samsung-laptop doesn't use ACPI, only SABI interface.
>> I thought it does for the video driver/backlight control. The strange
>> thing is that the issues start directly after loading the driver.
>> Perhaps it has to be blacklisted for this model.
> samsung-laptop talks to the bios directly using SABI. But the BIOS is
> also related to ACPI, so things can get ugly I guess...
>
>> Kernel log:
>>
>> samsung_laptop: detected SABI interface: SwSmi@
>> samsung_laptop: Backlight controlled by ACPI video driver
>> ACPI Warning: 0x000000000000efa0-0x000000000000efbf SystemIO conflicts
>> with Region \_SB_.PCI0.SBUS.SMBI 1 (20120711/utaddress-251)
>> ACPI: If an ACPI driver is available for this device, you should use it
>> instead of the native driver
> Hum can you try to blacklist samsung-laptop, load it manually, and
> send me the dmesg ?
> If possible with and without acpi_backlght=vendor boot parameter.
>
>> Where are the custom ACPI drivers in the kernel tree if not in
>> drivers/platform/x86 ?
> drivers/paltform/x86 contains x86 platform drivers, and some of them
> are ACPI drivers, but not all of them.
>
>>> This looks like ACPI issues, please file a bug on bugzilla, but first
>>> thing, check for bios updates :).
>> Which bugzilla? Sorry, I'm not an ACPI specialist.
> The linux kernel bugzilla. But before doing that, are this issues only
> hapenning when/if samsung-laptop is loaded ?
> Can you send me he output of:
>
> tree /sys/devices/platform/samsung-laptop/
> tree /sys/class/backlight
>
> Thanks
>
>