[PATCH 0/5] WMI

[PATCH 0/5] WMI

[Carlos Corbacho]

Another WMI release.

Patch #1: (WMI - initial driver and in kernel interface)

Big change here is putting in a basic EC region handler - which is
mostly a shameless rip off of the EC driver's acpi_ec_space_handler,
with the exception that WMI doesn't need to care about the EC device,
and just calls ec_{read,write} instead.

Therefore, CC'ing Alexey for the EC bit.

For review, and hopefully to go up to 2.6.25

Patch #2: (acer-wmi) - No Change

For review, and (rather tentatively here) 2.6.25 material?

Patch #3: (tc1100-wmi)

RFC only, needs actual testing on the hardware, and probably very broken. 
Waiting on Matthew Garrett to get some free time to test this.

Bluetooth renamed to jogdial, as per comments from Joshua Wise.

Patch #4: (WMI sysfs interface) - No Change

RFC only (see patch #5 for reason). Adds interface 
under /sys/devices/virtual/wmi

Changes: Added 'string' file to each GUID, which reports whether the GUID has 
the 'string' flag set (userspace needs to know for converting from its' 
format to ASCII and back).

Patch #5: (WMI sysfs workaround) - No change

Temporary hack, needed to get patch #4 working, due to a limitation on bus_id 
length (Kay Sievers is apparently working on this, ref Greg KH[1]). This 
should _NOT_ be applied to any upstream kernel.

-Carlos

[1] http://lkml.org/lkml/2007/12/4/30

[PATCH 0/5] WMI

[Carlos Corbacho]

ANYWR - Another New Year WMI release

Len,

Can you please review at least the first two patches in this series, as I
would still really like to try and get them into 2.6.25.

-Carlos

===

Patch #1: (WMI - driver and in kernel interface) - Fix ACPI debug issues

Fix the compile errors with ACPI debugging enabled.

Len: For you to review, and to go upstream.

====

Patch #2: (acer-wmi) - No change

Len: For you to review, and to go upstream.

====

Patch #3: (tc1100-wmi) - No Change

RFC only, needs actual testing on the hardware, and probably very broken.
Waiting on Matthew Garrett to get some free time to test this.

====

Patch #4: (WMI sysfs interface) - Added new in-kernel function

RFC only (see patch #5 for reason and discussion). Adds interface under
/sys/devices/virtual/wmi

Also adds a function to return the (virtual) device associated with a GUID for
use by other parts of the kernel that want to set a parent device.

====

Patch #5: (WMI sysfs workaround) - No change

Temporary hack, needed to get patch #4 working, due to a limitation on bus_id
length (Kay Sievers is apparently working on this, ref Greg KH[1]).

Apparently, the necessary fixes _might_ go into 2.6.25 as well, or worst case
scenario is 2.6.26[2]. So we can either:

1) Hold off on patch #4 until the necessary fixes go in (this might be 2.6.25
if we're lucky, or we end up putting off WMI sysfs support until 2.6.26)

or

2) Apply patch #4, and #5 on top as a temporary hack, and then revert #5 as
soon as the necessary changes go in. (For bisection purposes, #4 and #5 should
be merged if we go ahead with this).

I don't know what would be the best solution here.

[1] http://lkml.org/lkml/2007/12/4/30
[2] http://www.nabble.com/Re:-Fix-Firmware-class-name-collision-p14352966.html
---

Carlos Corbacho (5):
      [WIP] ACPI: WMI: Limit size of device string to 19 characters
      [RFC] ACPI: WMI: Add sysfs userspace interface
      [RFC] tc1100-wmi: Add driver for HP Compaq TC1100 Tablets
      acer-wmi: Add driver for newer Acer laptops
      ACPI: WMI: Add ACPI-WMI mapping driver


 MAINTAINERS               |    7 
 drivers/acpi/Kconfig      |   11 
 drivers/acpi/Makefile     |    1 
 drivers/acpi/wmi.c        | 1190 +++++++++++++++++++++++++++++++++++++++++++++
 drivers/misc/Kconfig      |   25 +
 drivers/misc/Makefile     |    2 
 drivers/misc/acer-wmi.c   | 1100 ++++++++++++++++++++++++++++++++++++++++++
 drivers/misc/tc1100-wmi.c |  290 +++++++++++
 include/linux/acpi.h      |   21 +
 9 files changed, 2647 insertions(+), 0 deletions(-)
 create mode 100644 drivers/acpi/wmi.c
 create mode 100644 drivers/misc/acer-wmi.c
 create mode 100644 drivers/misc/tc1100-wmi.c

Re: [PATCH 0/5] WMI

[Carlos Corbacho]

On Friday 18 January 2008 23:58:18 Carlos Corbacho wrote:
> ANYWR - Another New Year WMI release
>
> Len,
>
> Can you please review at least the first two patches in this series, as I
> would still really like to try and get them into 2.6.25.

Len,

I have a small update for acer-wmi to add another DMI entry queued up - before 
I submit that and re-submit the whole WMI series, is there anything else I 
can do that would help expedite the review of WMI?

I accept that at this stage, acceptance into 2.6.25 is unlikely - but I don't 
think aiming for 2.6.26 at least is entirely unrealistic (I've been posting 
the WMI patches for 'upstream review' since the end of October now, and 
earlier works-in-progress since the beginning of October). The sooner this 
gets looked at, the sooner I can get any problems addressed and this can be 
in a mergeable state for 2.6.26 (and can hopefully sit in -mm for a bit as 
well that way).

-Carlos
-- 
E-Mail: carlos@strangeworlds.co.uk
Web: strangeworlds.co.uk
GPG Key ID: 0x23EE722D

[PATCH 0/5] WMI

[Carlos Corbacho]

ANYWR - A New Year WMI release

Len,

Can you please review at least the first two patches in this series, as I
would really like to try and get them into 2.6.25.

-Carlos

===

Patch #1: (WMI - driver and in kernel interface) - No change

Len: For you to review, and to go upstream.

====

Patch #2: (acer-wmi)

Added more device autodetection. Also, fixed setting bluetooth and wireless
on older AMW0 V2 laptops, and improved the error handling code on driver
registration.

(I am currently working on rfkill support, but have been running into some
strange problems with rfkill-input, so I'll release that as a future patch
on top of this when it's done. Until then, though, this patch still works
as-is).

Len: For you to review, and to go upstream.

====

Patch #3: (tc1100-wmi) - No Change

RFC only, needs actual testing on the hardware, and probably very broken.
Waiting on Matthew Garrett to get some free time to test this.

====

Patch #4: (WMI sysfs interface) - No Change

RFC only (see patch #5 for reason). Adds interface under
/sys/devices/virtual/wmi

Matthew: Do you want me to fold patch #5 into this?

====

Patch #5: (WMI sysfs workaround) - No change

Temporary hack, needed to get patch #4 working, due to a limitation on bus_id
length (Kay Sievers is apparently working on this, ref Greg KH[1]).

[1] http://lkml.org/lkml/2007/12/4/30
---

Carlos Corbacho (5):
      [WIP] ACPI: WMI: Limit size of device string to 19 characters
      [RFC] ACPI: WMI: Add sysfs userspace interface
      [RFC] tc1100-wmi: Add driver for HP Compaq TC1100 Tablets
      acer-wmi: Add driver for newer Acer laptops
      ACPI: WMI: Add ACPI-WMI mapping driver


 MAINTAINERS               |    7 
 drivers/acpi/Kconfig      |   11 
 drivers/acpi/Makefile     |    1 
 drivers/acpi/wmi.c        | 1171 +++++++++++++++++++++++++++++++++++++++++++++
 drivers/misc/Kconfig      |   25 +
 drivers/misc/Makefile     |    2 
 drivers/misc/acer-wmi.c   | 1100 ++++++++++++++++++++++++++++++++++++++++++
 drivers/misc/tc1100-wmi.c |  290 +++++++++++
 include/linux/acpi.h      |   20 +
 9 files changed, 2627 insertions(+), 0 deletions(-)
 create mode 100644 drivers/acpi/wmi.c
 create mode 100644 drivers/misc/acer-wmi.c
 create mode 100644 drivers/misc/tc1100-wmi.c

[PATCH 0/5] WMI

[Carlos Corbacho]

ACSWR - A Christmas Sale WMI release

Patch #1: (WMI - driver and in kernel interface)

Added per GUID event handling callbacks, as per Matthew's suggestion.

Len:
For review, and to go upstream.

(Bar the EC stuff, this code has also been inflic^^^^tested by acer_acpi users
with my 0.10 releases, so I'm confident on Acer hardware at least that it won't
break horribly).

Patch #2: (acer-wmi)

Add device detection for bluetooth and maximum brightness (removes more hard
coded EC quirks). Also simplified some of the code.

Len:
For review, and to go upstream.

Patch #3: (tc1100-wmi) - No Change

RFC only, needs actual testing on the hardware, and probably very broken.
Waiting on Matthew Garrett to get some free time to test this.

Patch #4: (WMI sysfs interface) - No Change

RFC only (see patch #5 for reason). Adds interface under
/sys/devices/virtual/wmi

Matthew - Do you want me to fold patch #5 into this?

Patch #5: (WMI sysfs workaround) - No change

Temporary hack, needed to get patch #4 working, due to a limitation on bus_id
length (Kay Sievers is apparently working on this, ref Greg KH[1]).

-Carlos

[1] http://lkml.org/lkml/2007/12/4/30

Re: [PATCH 0/5] WMI

[Carlos Corbacho]

On Thursday 27 December 2007 15:38:04 Carlos Corbacho wrote:
> Patch #1: (WMI - driver and in kernel interface)
>
> Added per GUID event handling callbacks, as per Matthew's suggestion.
>
> Len:
> For review, and to go upstream.
>
> (Bar the EC stuff, this code has also been inflic^^^^tested by acer_acpi
> users with my 0.10 releases, so I'm confident on Acer hardware at least
> that it won't break horribly).
>
> Patch #2: (acer-wmi)
>
> Add device detection for bluetooth and maximum brightness (removes more
> hard coded EC quirks). Also simplified some of the code.
>
> Len:
> For review, and to go upstream.

Len,

Could you please bump up reviewing patches #1 and #2 on your to-do list?

-Carlos
-- 
E-Mail: carlos@strangeworlds.co.uk
Web: strangeworlds.co.uk
GPG Key ID: 0x23EE722D

[PATCH 0/5] WMI

[Carlos Corbacho]

YAWR - Yet Another WMI release

Patch #1: (WMI - driver and in kernel interface)

Updated EC region handling as per Alexey's comments

Len:
For review, and hopefully to go up to 2.6.25

(Bar the EC stuff, this code has also been inflic^^^^tested by acer_acpi users
with the 0.10 RCs and 0.10 releases, so I'm confident on Acer hardware at least
that it won't break horribly).

Patch #2: (acer-wmi) - No Change

For review, and (rather tentatively here) 2.6.25 material?

Patch #3: (tc1100-wmi) - No Change

RFC only, needs actual testing on the hardware, and probably very broken.
Waiting on Matthew Garrett to get some free time to test this.

Patch #4: (WMI sysfs interface) - No Change

RFC only (see patch #5 for reason). Adds interface under
/sys/devices/virtual/wmi

Patch #5: (WMI sysfs workaround) - No change

Temporary hack, needed to get patch #4 working, due to a limitation on bus_id
length (Kay Sievers is apparently working on this, ref Greg KH[1]). This
should _NOT_ be applied to any upstream kernel.

-Carlos

[1] http://lkml.org/lkml/2007/12/4/30

[PATCH 0/5] WMI

[Carlos Corbacho]

Another WMI release.

Patch #1: (WMI - initial driver and in kernel interface)

Big change here is putting in a basic EC region handler - which is
mostly a shameless rip off of the EC driver's acpi_ec_space_handler,
with the exception that WMI doesn't need to care about the EC device,
and just calls ec_{read,write} instead.

Therefore, CC'ing Alexey for the EC bit.

For review, and hopefully to go up to 2.6.25

Patch #2: (acer-wmi) - No Change

For review, and (rather tentatively here) 2.6.25 material?

Patch #3: (tc1100-wmi)

RFC only, needs actual testing on the hardware, and probably very broken. 
Waiting on Matthew Garrett to get some free time to test this.

Bluetooth renamed to jogdial, as per comments from Joshua Wise.

Patch #4: (WMI sysfs interface) - No Change

RFC only (see patch #5 for reason). Adds interface 
under /sys/devices/virtual/wmi

Changes: Added 'string' file to each GUID, which reports whether the GUID has 
the 'string' flag set (userspace needs to know for converting from its' 
format to ASCII and back).

Patch #5: (WMI sysfs workaround) - No change

Temporary hack, needed to get patch #4 working, due to a limitation on bus_id 
length (Kay Sievers is apparently working on this, ref Greg KH[1]). This 
should _NOT_ be applied to any upstream kernel.

-Carlos

[1] http://lkml.org/lkml/2007/12/4/30

[PATCH 0/5] WMI

[Carlos Corbacho]

So, here comes the next release of WMI.

Patch #1: (WMI - initial driver and in kernel interface)

One big change - add a fix to handle querying data GUIDs that violate the WMI 
spec with setting the 'expensive' flag (one case seen in the wild on a newer 
Acer laptop) - GUID declares the block as expensive, but there is no 
accompanying WCxx method.

A few minor taste cleanups thrown in (replacing (a == NULL) with (!a)).

For review, and hopefully to go up to 2.6.25 

Patch #2: (acer-wmi)

Add a few more DMI matched EC quirks in since last time. No other code 
changes. 

For review, and (rather tentatively here) 2.6.25 material?

Patch #3: (tc1100-wmi)

RFC only, needs actual testing on the hardware, and probably very broken. 
Waiting on Matthew Garrett to get some free time to test this.

Patch #4: (WMI sysfs interface)

RFC only (see patch #5 for reason). Adds interface 
under /sys/devices/virtual/wmi

Changes: Added 'string' file to each GUID, which reports whether the GUID has 
the 'string' flag set (userspace needs to know for converting from its' 
format to ASCII and back).

Patch #5: (WMI sysfs workaround)

Temporary hack, needed to get patch #4 working, due to a limitation on bus_id 
length (Kay Sievers is apparently working on this, ref Greg KH[1]). This 
should _NOT_ be applied to any upstream kernel.

-Carlos

[1] http://lkml.org/lkml/2007/12/4/30
-- 
E-Mail: carlos@strangeworlds.co.uk
Web: strangeworlds.co.uk
GPG Key ID: 0x23EE722D