ACPI device for ASUS EEEPC 1101HA not added

ACPI device for ASUS EEEPC 1101HA not added

[Daniel Mack]

On my new 1101HA Eeepc, the special function keys do not work. This is
because the asus-laptop driver fails to match the ACPI device named
"ASUS010". However, acpidump tells me a device with that name does
actually exist - the (simplified) disassembly of the DSDT is below.

It looks like the gigantic parser state machine fails to add that
device for whatever reason. Any ideas how to fix that?


Thanks,
Daniel


/*
 * Intel ACPI Component Architecture
 * AML Disassembler version 20090521
 */
DefinitionBlock ("out/DSDT-0x3f660430-0.aml", "DSDT", 2, "A1359", "A1359000", 0x00000000)
{

    [...]

    Scope (_SB)
    {

        [...]

        Device (PCI0)
        {
            Name (_HID, EisaId ("PNP0A08"))
            Name (_ADR, Zero)

            [...]

            Device (SBRG)
            {

                [...]

                Scope (\_SB)
                {
                    Name (ATKP, Zero)
                    Device (ATKD)
                    {
                        Name (_HID, "ASUS010")
                        Name (_UID, 0x01010100)

                        [...]


The full version of the dump is here:

  http://caiaq.de/download/tmp/DSDT-0x3f660430-0.dsl

Re: ACPI device for ASUS EEEPC 1101HA not added

[andrej.gelenberg-KJIyc1CJxjQ]

Hi,

you need to whitelist your eee pc for OSI(Linux) in drivers/acpi/blacklist.c
like this:

+        /*
+        * On newer Eeepc, the interface used by eeepc-laptop (ASUS010)
+        * is disabled without _OSI(Linux)
+        */
+       {
+       .callback = dmi_enable_osi_linux,
+       .ident = "Asus Eeepc-1101HA",
+       .matches = {
+                    DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK Computer INC."),
+                    DMI_MATCH(DMI_PRODUCT_NAME, "1101HA"),
+               },
+       },

Regards,
Andrej

Daniel Mack writes:

> On my new 1101HA Eeepc, the special function keys do not work. This is
> because the asus-laptop driver fails to match the ACPI device named
> "ASUS010". However, acpidump tells me a device with that name does
> actually exist - the (simplified) disassembly of the DSDT is below.
> 
> It looks like the gigantic parser state machine fails to add that
> device for whatever reason. Any ideas how to fix that?
> 
> 
> Thanks,
> Daniel
> 
> 
> /*
>  * Intel ACPI Component Architecture
>  * AML Disassembler version 20090521
>  */
> DefinitionBlock ("out/DSDT-0x3f660430-0.aml", "DSDT", 2, "A1359", "A1359000", 0x00000000)
> {
> 
>     [...]
> 
>     Scope (_SB)
>     {
> 
>         [...]
> 
>         Device (PCI0)
>         {
>             Name (_HID, EisaId ("PNP0A08"))
>             Name (_ADR, Zero)
> 
>             [...]
> 
>             Device (SBRG)
>             {
> 
>                 [...]
> 
>                 Scope (\_SB)
>                 {
>                     Name (ATKP, Zero)
>                     Device (ATKD)
>                     {
>                         Name (_HID, "ASUS010")
>                         Name (_UID, 0x01010100)
> 
>                         [...]
> 
> 
> The full version of the dump is here:
> 
>   http://caiaq.de/download/tmp/DSDT-0x3f660430-0.dsl
> 
> 
> ------------------------------------------------------------------------------
> The Planet: dedicated and managed hosting, cloud storage, colocation
> Stay online with enterprise data centers and the best network in the business
> Choose flexible plans and management services without long-term contracts
> Personal 24x7 support from experience hosting pros just a phone call away.
> http://p.sf.net/sfu/theplanet-com
> _______________________________________________
> Acpi4asus-user mailing list
> Acpi4asus-user-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/acpi4asus-user

------------------------------------------------------------------------------
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com

Re: [Acpi4asus-user] ACPI device for ASUS EEEPC 1101HA not added

[Daniel Mack]

On Mon, Feb 01, 2010 at 12:06:40PM +0100, andrej.gelenberg@udo.edu wrote:
> you need to whitelist your eee pc for OSI(Linux) in drivers/acpi/blacklist.c
> like this:
> 
> +        /*
> +        * On newer Eeepc, the interface used by eeepc-laptop (ASUS010)
> +        * is disabled without _OSI(Linux)
> +        */
> +       {
> +       .callback = dmi_enable_osi_linux,
> +       .ident = "Asus Eeepc-1101HA",
> +       .matches = {
> +                    DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK Computer INC."),
> +                    DMI_MATCH(DMI_PRODUCT_NAME, "1101HA"),
> +               },
> +       },

Great, thanks. That worked. Is that already queued anywhere or do you
want me to send a proper patch?

Daniel


> Daniel Mack writes:
> 
> >On my new 1101HA Eeepc, the special function keys do not work. This is
> >because the asus-laptop driver fails to match the ACPI device named
> >"ASUS010". However, acpidump tells me a device with that name does
> >actually exist - the (simplified) disassembly of the DSDT is below.
> >
> >It looks like the gigantic parser state machine fails to add that
> >device for whatever reason. Any ideas how to fix that?
> >
> >
> >Thanks,
> >Daniel
> >

Re: ACPI device for ASUS EEEPC 1101HA not added

[andrej.gelenberg-KJIyc1CJxjQ]

Hi,

there was a patch for 1005ha and 1000h (or so), but i can't find it
in the repository anymore. I miss it in 2.6.33-rc* too. 

acpi/blacklist: enable OSI(Linux) on newer eeepc
patch from git://git.iksaif.net/acpi4asus.git.

Regards,
Andrej

Daniel Mack writes:

> On Mon, Feb 01, 2010 at 12:06:40PM +0100, andrej.gelenberg-KJIyc1CJxjQ@public.gmane.org wrote:
>> you need to whitelist your eee pc for OSI(Linux) in drivers/acpi/blacklist.c
>> like this:
>> 
>> +        /*
>> +        * On newer Eeepc, the interface used by eeepc-laptop (ASUS010)
>> +        * is disabled without _OSI(Linux)
>> +        */
>> +       {
>> +       .callback = dmi_enable_osi_linux,
>> +       .ident = "Asus Eeepc-1101HA",
>> +       .matches = {
>> +                    DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK Computer INC."),
>> +                    DMI_MATCH(DMI_PRODUCT_NAME, "1101HA"),
>> +               },
>> +       },
> 
> Great, thanks. That worked. Is that already queued anywhere or do you
> want me to send a proper patch?
> 
> Daniel
> 
> 
>> Daniel Mack writes:
>> 
>> >On my new 1101HA Eeepc, the special function keys do not work. This is
>> >because the asus-laptop driver fails to match the ACPI device named
>> >"ASUS010". However, acpidump tells me a device with that name does
>> >actually exist - the (simplified) disassembly of the DSDT is below.
>> >
>> >It looks like the gigantic parser state machine fails to add that
>> >device for whatever reason. Any ideas how to fix that?
>> >
>> >
>> >Thanks,
>> >Daniel
>> >

------------------------------------------------------------------------------
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com

Re: ACPI device for ASUS EEEPC 1101HA not added

[Zhang Rui]

On Mon, 2010-02-01 at 11:25 +0800, Daniel Mack wrote:
> On my new 1101HA Eeepc, the special function keys do not work. This is
> because the asus-laptop driver fails to match the ACPI device named
> "ASUS010". However, acpidump tells me a device with that name does
> actually exist - the (simplified) disassembly of the DSDT is below.
> 
> It looks like the gigantic parser state machine fails to add that
> device for whatever reason. Any ideas how to fix that?
> 
eeepc is handled in eeepci-laptop driver.

~/src/linux-2.6$ grep ASUS010 drivers/platform/*/*
drivers/platform/x86/eeepc-laptop.c:#define EEEPC_ACPI_HID		"ASUS010"

thanks,
rui

> 
> Thanks,
> Daniel
> 
> 
> /*
>  * Intel ACPI Component Architecture
>  * AML Disassembler version 20090521
>  */
> DefinitionBlock ("out/DSDT-0x3f660430-0.aml", "DSDT", 2, "A1359", "A1359000", 0x00000000)
> {
> 
>     [...]
> 
>     Scope (_SB)
>     {
> 
>         [...]
> 
>         Device (PCI0)
>         {
>             Name (_HID, EisaId ("PNP0A08"))
>             Name (_ADR, Zero)
> 
>             [...]
> 
>             Device (SBRG)
>             {
> 
>                 [...]
> 
>                 Scope (\_SB)
>                 {
>                     Name (ATKP, Zero)
>                     Device (ATKD)
>                     {
>                         Name (_HID, "ASUS010")
>                         Name (_UID, 0x01010100)
> 
>                         [...]
> 
> 
> The full version of the dump is here:
> 
>   http://caiaq.de/download/tmp/DSDT-0x3f660430-0.dsl
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: ACPI device for ASUS EEEPC 1101HA not added

[Daniel Mack]

On Tue, Feb 02, 2010 at 09:19:22AM +0800, Zhang Rui wrote:
> On Mon, 2010-02-01 at 11:25 +0800, Daniel Mack wrote:
> > On my new 1101HA Eeepc, the special function keys do not work. This is
> > because the asus-laptop driver fails to match the ACPI device named
> > "ASUS010". However, acpidump tells me a device with that name does
> > actually exist - the (simplified) disassembly of the DSDT is below.
> > 
> > It looks like the gigantic parser state machine fails to add that
> > device for whatever reason. Any ideas how to fix that?
> > 
> eeepc is handled in eeepci-laptop driver.
> 
> ~/src/linux-2.6$ grep ASUS010 drivers/platform/*/*
> drivers/platform/x86/eeepc-laptop.c:#define EEEPC_ACPI_HID		"ASUS010"

Yes, that I know. The driver was just unable to match the device because
it was not detected by the ACPI core. Works now with the patch Andrej
sent. I would still like to make sure it's getting merged mainline
sooner or later.

Thanks,
Daniel

RE: ACPI device for ASUS EEEPC 1101HA not added

[Zhang, Rui]

> -----Original Message-----
> From: Daniel Mack [mailto:daniel@caiaq.de]
> Sent: Tuesday, February 02, 2010 10:10 AM
> To: Zhang, Rui
> Cc: linux-acpi@vger.kernel.org; linux-kernel@vger.kernel.org;
> acpi4asus-user@lists.sourceforge.net
> Subject: Re: ACPI device for ASUS EEEPC 1101HA not added
> 
> On Tue, Feb 02, 2010 at 09:19:22AM +0800, Zhang Rui wrote:
> > On Mon, 2010-02-01 at 11:25 +0800, Daniel Mack wrote:
> > > On my new 1101HA Eeepc, the special function keys do not work. This
> is
> > > because the asus-laptop driver fails to match the ACPI device named
> > > "ASUS010". However, acpidump tells me a device with that name does
> > > actually exist - the (simplified) disassembly of the DSDT is below.
> > >
> > > It looks like the gigantic parser state machine fails to add that
> > > device for whatever reason. Any ideas how to fix that?
> > >
> > eeepc is handled in eeepci-laptop driver.
> >
> > ~/src/linux-2.6$ grep ASUS010 drivers/platform/*/*
> > drivers/platform/x86/eeepc-laptop.c:#define EEEPC_ACPI_HID
> 	"ASUS010"
> 
> Yes, that I know. The driver was just unable to match the device
> because
> it was not detected by the ACPI core. Works now with the patch Andrej
> sent. I would still like to make sure it's getting merged mainline
> sooner or later.
>
Oh, can you give a link to this patch please?

Thanks,
rui
 
> Thanks,
> Daniel

Re: ACPI device for ASUS EEEPC 1101HA not added

[Daniel Mack]

On Tue, Feb 02, 2010 at 10:17:17AM +0800, Zhang, Rui wrote:
> > On Tue, Feb 02, 2010 at 09:19:22AM +0800, Zhang Rui wrote:
> > > eeepc is handled in eeepci-laptop driver.
> > >
> > > ~/src/linux-2.6$ grep ASUS010 drivers/platform/*/*
> > > drivers/platform/x86/eeepc-laptop.c:#define EEEPC_ACPI_HID
> > 	"ASUS010"
> > 
> > Yes, that I know. The driver was just unable to match the device
> > because
> > it was not detected by the ACPI core. Works now with the patch Andrej
> > sent. I would still like to make sure it's getting merged mainline
> > sooner or later.
> >
> Oh, can you give a link to this patch please?

http://permalink.gmane.org/gmane.linux.acpi.acpi4asus.user/3781

Re: [Acpi4asus-user] ACPI device for ASUS EEEPC 1101HA not added

[Len Brown]

On Mon, 1 Feb 2010, andrej.gelenberg@udo.edu wrote:

> Hi,
> 
> you need to whitelist your eee pc for OSI(Linux) in drivers/acpi/blacklist.c
> like this:
> 
> +        /*
> +        * On newer Eeepc, the interface used by eeepc-laptop (ASUS010)
> +        * is disabled without _OSI(Linux)
> +        */
> +       {
> +       .callback = dmi_enable_osi_linux,
> +       .ident = "Asus Eeepc-1101HA",
> +       .matches = {
> +                    DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK Computer INC."),
> +                    DMI_MATCH(DMI_PRODUCT_NAME, "1101HA"),
> +               },
> +       },

Not necessarily the right fix.  We have gone to a lot of trouble
to discourage BIOS vendors from depending on the ill-defined OSI(Linux),
so I hesitate to invoke it -- even for a workaround.

The problem at hand is that ASUS010 is not enabled for an OS
that claims compatibility with Win7 (MSOS() == MSW7) below.

                Scope (\_SB)
                {
                    Name (ATKP, Zero)
                    Device (ATKD)
                    {
                        Name (_HID, "ASUS010")
                        Name (_UID, 0x01010100)
                        Method (_STA, 0, NotSerialized)
                        {
                            If (LEqual (MSOS (), MSW7))
                            {
                                Return (Zero)
                            }
                            Else
                            {
                                Return (0x0F)
                            }

                            Return (Zero)
                            Return (0x0F)
                        }

(heh, see any indication of lack of quality in this code?:-)

MSOS() does this:

                    Scope (\)
                    {
                        Name (OSLX, 0x10)
                        Name (OSMS, 0x20)
                        Name (MS98, 0x21)
                        Name (MSME, 0x22)
                        Name (MS2K, 0x23)
                        Name (MSXP, 0x24)
                        Name (MSVT, 0x25)
                        Name (MSW7, 0x26)
                        Name (OSFG, Ones)
                        Method (MSOS, 0, NotSerialized)
                        {
                            If (LNotEqual (OSFG, Ones))
                            {
                                Return (OSFG)
                            }

                            Store (Zero, OSFG)
                            If (CondRefOf (_OSI, Local0))
                            {
                                If (_OSI ("Windows 2001"))
                                {
                                    Store (MSXP, OSFG)
                                }

                                If (_OSI ("Windows 2001 SP1"))
                                {
                                    Store (MSXP, OSFG)
                                }

                                If (_OSI ("Windows 2001 SP2"))
                                {
                                    Store (MSXP, OSFG)
                                }

                                If (_OSI ("Windows 2006"))
                                {
                                    Store (MSVT, OSFG)
                                }

                                If (_OSI ("Windows 2009"))
                                {
                                    Store (MSW7, OSFG)
                                }

                                If (_OSI ("Linux"))
                                {
                                    Store (OSLX, OSFG)
                                }

                                Return (OSFG)
                            }
                            Else


So I expect if you apply no patch, but boot with 'acpi_osi="!Windows 2009"'
then that would also work properly, as OSFG above will be set
to "MSVT" instead of "OSLX".

Looking through the DSDT, there are no references to OSLX or MSVT,
or MSXP, for that matter.  However, there is an additional reference
to MSV7 in the temperature reading part of thermal zone TZ00,
that looks like some sort of OS-specific initialization, perhaps
a workaround.  So also check that /proc/acpi/thermal_zone/*/temperature
still work before and after.

thanks,
-Len Brown, Intel Open Source Technology Center
 

Re: [Acpi4asus-user] ACPI device for ASUS EEEPC 1101HA not added

[Corentin Chary]

On Tue, Feb 2, 2010 at 7:02 AM, Len Brown <lenb@kernel.org> wrote:
> On Mon, 1 Feb 2010, andrej.gelenberg@udo.edu wrote:
>
>> Hi,
>>
>> you need to whitelist your eee pc for OSI(Linux) in drivers/acpi/blacklist.c
>> like this:
>>
>> +        /*
>> +        * On newer Eeepc, the interface used by eeepc-laptop (ASUS010)
>> +        * is disabled without _OSI(Linux)
>> +        */
>> +       {
>> +       .callback = dmi_enable_osi_linux,
>> +       .ident = "Asus Eeepc-1101HA",
>> +       .matches = {
>> +                    DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK Computer INC."),
>> +                    DMI_MATCH(DMI_PRODUCT_NAME, "1101HA"),
>> +               },
>> +       },
>
> Not necessarily the right fix.  We have gone to a lot of trouble
> to discourage BIOS vendors from depending on the ill-defined OSI(Linux),
> so I hesitate to invoke it -- even for a workaround.
>
> The problem at hand is that ASUS010 is not enabled for an OS
> that claims compatibility with Win7 (MSOS() == MSW7) below.
>
>                Scope (\_SB)
>                {
>                    Name (ATKP, Zero)
>                    Device (ATKD)
>                    {
>                        Name (_HID, "ASUS010")
>                        Name (_UID, 0x01010100)
>                        Method (_STA, 0, NotSerialized)
>                        {
>                            If (LEqual (MSOS (), MSW7))
>                            {
>                                Return (Zero)
>                            }
>                            Else
>                            {
>                                Return (0x0F)
>                            }
>
>                            Return (Zero)
>                            Return (0x0F)
>                        }
>
> (heh, see any indication of lack of quality in this code?:-)
>
> MSOS() does this:
>
>                    Scope (\)
>                    {
>                        Name (OSLX, 0x10)
>                        Name (OSMS, 0x20)
>                        Name (MS98, 0x21)
>                        Name (MSME, 0x22)
>                        Name (MS2K, 0x23)
>                        Name (MSXP, 0x24)
>                        Name (MSVT, 0x25)
>                        Name (MSW7, 0x26)
>                        Name (OSFG, Ones)
>                        Method (MSOS, 0, NotSerialized)
>                        {
>                            If (LNotEqual (OSFG, Ones))
>                            {
>                                Return (OSFG)
>                            }
>
>                            Store (Zero, OSFG)
>                            If (CondRefOf (_OSI, Local0))
>                            {
>                                If (_OSI ("Windows 2001"))
>                                {
>                                    Store (MSXP, OSFG)
>                                }
>
>                                If (_OSI ("Windows 2001 SP1"))
>                                {
>                                    Store (MSXP, OSFG)
>                                }
>
>                                If (_OSI ("Windows 2001 SP2"))
>                                {
>                                    Store (MSXP, OSFG)
>                                }
>
>                                If (_OSI ("Windows 2006"))
>                                {
>                                    Store (MSVT, OSFG)
>                                }
>
>                                If (_OSI ("Windows 2009"))
>                                {
>                                    Store (MSW7, OSFG)
>                                }
>
>                                If (_OSI ("Linux"))
>                                {
>                                    Store (OSLX, OSFG)
>                                }
>
>                                Return (OSFG)
>                            }
>                            Else
>
>
> So I expect if you apply no patch, but boot with 'acpi_osi="!Windows 2009"'
> then that would also work properly, as OSFG above will be set
> to "MSVT" instead of "OSLX".
>
> Looking through the DSDT, there are no references to OSLX or MSVT,
> or MSXP, for that matter.  However, there is an additional reference
> to MSV7 in the temperature reading part of thermal zone TZ00,
> that looks like some sort of OS-specific initialization, perhaps
> a workaround.  So also check that /proc/acpi/thermal_zone/*/temperature
> still work before and after.
>
> thanks,
> -Len Brown, Intel Open Source Technology Center
>

This kind of things is present on all newer eeepc.
I tried to contact asus about that, without real success and it is
likely that the xandros distribribution
shipped with some eeepc include an acpi blacklisiting patch.

The thing is that on win7 ASUS010 is disabled, but an equivalent wmi
interface is enable, and it could
be possible to write a driver for it. Now, someone need to write it,
or send me a free 1005HA so I can do it :).

If nothing is done for 2.6.34, we may need to envisage some blacklisting :/.
Thanks
-- 
Corentin Chary
http://xf.iksaif.net

Re: [Acpi4asus-user] ACPI device for ASUS EEEPC 1101HA not added

[Matthew Garrett]

The correct solution is to implement an asus-wmi driver. The interface 
looks pretty much identical to the existing one, it's just indirected 
throguh WMI.

-- 
Matthew Garrett | mjg59@srcf.ucam.org

Re: ACPI device for ASUS EEEPC 1101HA not added

[Len Brown]

On Tue, 2 Feb 2010, Daniel Mack wrote:

> On Tue, Feb 02, 2010 at 10:17:17AM +0800, Zhang, Rui wrote:
> > > On Tue, Feb 02, 2010 at 09:19:22AM +0800, Zhang Rui wrote:
> > > > eeepc is handled in eeepci-laptop driver.
> > > >
> > > > ~/src/linux-2.6$ grep ASUS010 drivers/platform/*/*
> > > > drivers/platform/x86/eeepc-laptop.c:#define EEEPC_ACPI_HID
> > > 	"ASUS010"
> > > 
> > > Yes, that I know. The driver was just unable to match the device
> > > because
> > > it was not detected by the ACPI core. Works now with the patch Andrej
> > > sent. I would still like to make sure it's getting merged mainline
> > > sooner or later.
> > >
> > Oh, can you give a link to this patch please?
> 
> http://permalink.gmane.org/gmane.linux.acpi.acpi4asus.user/3781

boot the eeepc with acpi_osi="!Windows 2009"
until a WMI driver is written to handle this windows-specific BIOS.

thanks,
Len Brown, Intel Open Source Technology Center