Re: [Fwd: Re: A problem about acer-wmi]

2011-08-10 Thread joeyli 
Hi AceLan,

於 三，2011-08-10 於 10:13 +0800，joeyli 提到：
 於 二，2011-08-09 於 14:33 +0800，AceLan Kao 提到：
  Dear Joey,
  
  The current project we have right now have to follow the Acer WMI to
  handle the key events, that means no EC key event.
  And we have 3 keys that are not working now, they are touchpad toggle,
  brightness up, and brightness down.
  The touchpad toggle function is a Hotkey Break Event(function number
  0x2), and brightness key events are Brightness Change Event(function
  number 0x4). But now acer-wmi driver only handles General Hotkey
  Event(function number 0x1).
  
 
 Yes, current acer-wmi only capture the event function number 0x1, you
 can add those new function to acer_wmi_notify().
 
  I just implemented those 3 key events, so I think maybe we don't have
  to create a new acer-wmi driver.
 
 Great! Welcome for you patches, I will also test it.
 
  I'm longing for your work to clear up the acer-wmi driver, so that I
  can add the new machine id and send you the patch.
  Thanks.
  
  BTW, I'm available to help if you are too busy to do that.
  
  Best regards,
  AceLan Kao.
  
 
 I am doing the clear up, now, will send out patch (I hope today).
 
 
 Thank's
 Joey Lee
 

I add a new ACER_WMID_v2 interface flag and do some clear up in acer-wmi
initial function and get_u32 functions.

Please kindly test this patch:


Thank's
Joey Lee

From 28b2e2ebaa230d339d5749b581c667ed074bb7ea Mon Sep 17 00:00:00 2001
From: Lee, Chun-Yi j...@suse.com
Date: Wed, 10 Aug 2011 16:36:02 +0800
Subject: [PATCH] acer-wmi: add ACER_WMID_v2 interface flag to represent new 
notebooks

There have new acer notebooks' BIOS provide new WMID_GUID3 and
ACERWMID_EVENT_GUID methods.

Some of machines still keep the old WMID_GUID1 method but more and
more machines were already removed old wmi methods from DSDT.

So, this patch add a new ACER_WMID_v2 interface flag to represent
new acer notebooks, the following is definition:

 + ACER_WMID:
It means this machine only provides WMID_GUID1/2 methods.

 + ACER_WMID_v2:
It means this machine provide new WMID_GUID3 and WMID_EVENT_GUID
methods.
Some ACER_WMID_v2 machines also provide old WMID_GUID1/2 methods,
but we still query/set communication device's state by new
WMID_GUID3 method.

Tested on Acer Travelmate 8572

Signed-off-by: Lee, Chun-Yi j...@suse.com
---
 drivers/platform/x86/acer-wmi.c |  409 ---
 1 files changed, 211 insertions(+), 198 deletions(-)

diff --git a/drivers/platform/x86/acer-wmi.c b/drivers/platform/x86/acer-wmi.c
index af2bb20..712a505 100644
--- a/drivers/platform/x86/acer-wmi.c
+++ b/drivers/platform/x86/acer-wmi.c
@@ -190,6 +190,7 @@ enum interface_flags {
ACER_AMW0,
ACER_AMW0_V2,
ACER_WMID,
+   ACER_WMID_v2,
 };
 
 #define ACER_DEFAULT_WIRELESS  0
@@ -868,6 +869,176 @@ static acpi_status WMID_set_u32(u32 value, u32 cap, 
struct wmi_interface *iface)
return WMI_execute_u32(method_id, (u32)value, NULL);
 }
 
+static acpi_status wmid3_get_device_status(u32 *value, u16 device)
+{
+   struct wmid3_gds_return_value return_value;
+   acpi_status status;
+   union acpi_object *obj;
+   struct wmid3_gds_input_param params = {
+   .function_num = 0x1,
+   .hotkey_number = 0x01,
+   .devices = device,
+   };
+   struct acpi_buffer input = {
+   sizeof(struct wmid3_gds_input_param),
+   params
+   };
+   struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL };
+
+   status = wmi_evaluate_method(WMID_GUID3, 0, 0x2, input, output);
+   if (ACPI_FAILURE(status))
+   return status;
+
+   obj = output.pointer;
+
+   if (!obj)
+   return AE_ERROR;
+   else if (obj-type != ACPI_TYPE_BUFFER) {
+   kfree(obj);
+   return AE_ERROR;
+   }
+   if (obj-buffer.length != 8) {
+   pr_warn(Unknown buffer length %d\n, obj-buffer.length);
+   kfree(obj);
+   return AE_ERROR;
+   }
+
+   return_value = *((struct wmid3_gds_return_value *)obj-buffer.pointer);
+   kfree(obj);
+
+   if (return_value.error_code || return_value.ec_return_value)
+   pr_warn(Get Device Status failed: 0x%x - 0x%x\n,
+   return_value.error_code,
+   return_value.ec_return_value);
+   else
+   *value = !!(return_value.devices  device);
+
+   return status;
+}
+
+static acpi_status wmid_v2_get_u32(u32 *value, u32 cap)
+{
+   u16 device;
+
+   switch (cap) {
+   case ACER_CAP_WIRELESS:
+   device = ACER_WMID3_GDS_WIRELESS;
+   break;
+   case ACER_CAP_BLUETOOTH:
+   device = ACER_WMID3_GDS_BLUETOOTH;
+   break;
+   case ACER_CAP_THREEG:
+   device = ACER_WMID3_GDS_THREEG;
+   break;
+   default:
+

Re: [Fwd: Re: A problem about acer-wmi]

2011-08-10 Thread Joey Lee 
Hi AceLan,

於 三，2011-08-10 於 10:13 +0800，joeyli 提到：
 於 二，2011-08-09 於 14:33 +0800，AceLan Kao 提到：
  Dear Joey,
  
  The current project we have right now have to follow the Acer WMI to
  handle the key events, that means no EC key event.
  And we have 3 keys that are not working now, they are touchpad
toggle,
  brightness up, and brightness down.
  The touchpad toggle function is a Hotkey Break Event(function number
  0x2), and brightness key events are Brightness Change Event(function
  number 0x4). But now acer-wmi driver only handles General Hotkey
  Event(function number 0x1).
  
 
 Yes, current acer-wmi only capture the event function number 0x1, you
 can add those new function to acer_wmi_notify().
 
  I just implemented those 3 key events, so I think maybe we don't
have
  to create a new acer-wmi driver.
 
 Great! Welcome for you patches, I will also test it.
 
  I'm longing for your work to clear up the acer-wmi driver, so that I
  can add the new machine id and send you the patch.
  Thanks.
  
  BTW, I'm available to help if you are too busy to do that.
  
  Best regards,
  AceLan Kao.
  
 
 I am doing the clear up, now, will send out patch (I hope today).
 
 
 Thank's
 Joey Lee
 

I add a new ACER_WMID_v2 interface flag and do some clear up in acer-wmi
initial function and get_u32 functions.

Please kindly test this patch:


Thank's
Joey Lee


From 28b2e2ebaa230d339d5749b581c667ed074bb7ea Mon Sep 17 00:00:00 2001
From: Lee, Chun-Yi j...@suse.com
Date: Wed, 10 Aug 2011 16:36:02 +0800
Subject: [PATCH] acer-wmi: add ACER_WMID_v2 interface flag to represent new 
notebooks

There have new acer notebooks' BIOS provide new WMID_GUID3 and
ACERWMID_EVENT_GUID methods.

Some of machines still keep the old WMID_GUID1 method but more and
more machines were already removed old wmi methods from DSDT.

So, this patch add a new ACER_WMID_v2 interface flag to represent
new acer notebooks, the following is definition:

 + ACER_WMID:
It means this machine only provides WMID_GUID1/2 methods.

 + ACER_WMID_v2:
It means this machine provide new WMID_GUID3 and WMID_EVENT_GUID
methods.
Some ACER_WMID_v2 machines also provide old WMID_GUID1/2 methods,
but we still query/set communication device's state by new
WMID_GUID3 method.

Tested on Acer Travelmate 8572

Signed-off-by: Lee, Chun-Yi j...@suse.com
---
 drivers/platform/x86/acer-wmi.c |  409 ---
 1 files changed, 211 insertions(+), 198 deletions(-)

diff --git a/drivers/platform/x86/acer-wmi.c b/drivers/platform/x86/acer-wmi.c
index af2bb20..712a505 100644
--- a/drivers/platform/x86/acer-wmi.c
+++ b/drivers/platform/x86/acer-wmi.c
@@ -190,6 +190,7 @@ enum interface_flags {
ACER_AMW0,
ACER_AMW0_V2,
ACER_WMID,
+   ACER_WMID_v2,
 };
 
 #define ACER_DEFAULT_WIRELESS  0
@@ -868,6 +869,176 @@ static acpi_status WMID_set_u32(u32 value, u32 cap, 
struct wmi_interface *iface)
return WMI_execute_u32(method_id, (u32)value, NULL);
 }
 
+static acpi_status wmid3_get_device_status(u32 *value, u16 device)
+{
+   struct wmid3_gds_return_value return_value;
+   acpi_status status;
+   union acpi_object *obj;
+   struct wmid3_gds_input_param params = {
+   .function_num = 0x1,
+   .hotkey_number = 0x01,
+   .devices = device,
+   };
+   struct acpi_buffer input = {
+   sizeof(struct wmid3_gds_input_param),
+   params
+   };
+   struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL };
+
+   status = wmi_evaluate_method(WMID_GUID3, 0, 0x2, input, output);
+   if (ACPI_FAILURE(status))
+   return status;
+
+   obj = output.pointer;
+
+   if (!obj)
+   return AE_ERROR;
+   else if (obj-type != ACPI_TYPE_BUFFER) {
+   kfree(obj);
+   return AE_ERROR;
+   }
+   if (obj-buffer.length != 8) {
+   pr_warn(Unknown buffer length %d\n, obj-buffer.length);
+   kfree(obj);
+   return AE_ERROR;
+   }
+
+   return_value = *((struct wmid3_gds_return_value *)obj-buffer.pointer);
+   kfree(obj);
+
+   if (return_value.error_code || return_value.ec_return_value)
+   pr_warn(Get Device Status failed: 0x%x - 0x%x\n,
+   return_value.error_code,
+   return_value.ec_return_value);
+   else
+   *value = !!(return_value.devices  device);
+
+   return status;
+}
+
+static acpi_status wmid_v2_get_u32(u32 *value, u32 cap)
+{
+   u16 device;
+
+   switch (cap) {
+   case ACER_CAP_WIRELESS:
+   device = ACER_WMID3_GDS_WIRELESS;
+   break;
+   case ACER_CAP_BLUETOOTH:
+   device = ACER_WMID3_GDS_BLUETOOTH;
+   break;
+   case ACER_CAP_THREEG:
+   device = ACER_WMID3_GDS_THREEG;
+   break;
+   default:
+

test mail, please ignore

2011-08-10 Thread joeyli 

test mail, please ignore

--
To unsubscribe from this list: send the line unsubscribe platform-driver-x86 
in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [Fwd: Re: A problem about acer-wmi]

2011-08-10 Thread Joey Lee 
Hi AceLan,

於 三，2011-08-10 於 10:13 +0800，joeyli 提到：
 於 二，2011-08-09 於 14:33 +0800，AceLan Kao 提到：
  Dear Joey,
  
  The current project we have right now have to follow the Acer WMI to
  handle the key events, that means no EC key event.
  And we have 3 keys that are not working now, they are touchpad
toggle,
  brightness up, and brightness down.
  The touchpad toggle function is a Hotkey Break Event(function number
  0x2), and brightness key events are Brightness Change Event(function
  number 0x4). But now acer-wmi driver only handles General Hotkey
  Event(function number 0x1).
  
 
 Yes, current acer-wmi only capture the event function number 0x1, you
 can add those new function to acer_wmi_notify().
 
  I just implemented those 3 key events, so I think maybe we don't
have
  to create a new acer-wmi driver.
 
 Great! Welcome for you patches, I will also test it.
 
  I'm longing for your work to clear up the acer-wmi driver, so that I
  can add the new machine id and send you the patch.
  Thanks.
  
  BTW, I'm available to help if you are too busy to do that.
  
  Best regards,
  AceLan Kao.
  
 
 I am doing the clear up, now, will send out patch (I hope today).
 
 
 Thank's
 Joey Lee
 

I add a new ACER_WMID_v2 interface flag and do some clear up in acer-wmi
initial function and get_u32 functions.

Please kindly test this patch:


Thank's
Joey Lee

From 28b2e2ebaa230d339d5749b581c667ed074bb7ea Mon Sep 17 00:00:00 2001
From: Lee, Chun-Yi j...@suse.com
Date: Wed, 10 Aug 2011 16:36:02 +0800
Subject: [PATCH] acer-wmi: add ACER_WMID_v2 interface flag to represent new 
notebooks

There have new acer notebooks' BIOS provide new WMID_GUID3 and
ACERWMID_EVENT_GUID methods.

Some of machines still keep the old WMID_GUID1 method but more and
more machines were already removed old wmi methods from DSDT.

So, this patch add a new ACER_WMID_v2 interface flag to represent
new acer notebooks, the following is definition:

 + ACER_WMID:
It means this machine only provides WMID_GUID1/2 methods.

 + ACER_WMID_v2:
It means this machine provide new WMID_GUID3 and WMID_EVENT_GUID
methods.
Some ACER_WMID_v2 machines also provide old WMID_GUID1/2 methods,
but we still query/set communication device's state by new
WMID_GUID3 method.

Tested on Acer Travelmate 8572

Signed-off-by: Lee, Chun-Yi j...@suse.com
---
 drivers/platform/x86/acer-wmi.c |  409 ---
 1 files changed, 211 insertions(+), 198 deletions(-)

diff --git a/drivers/platform/x86/acer-wmi.c b/drivers/platform/x86/acer-wmi.c
index af2bb20..712a505 100644
--- a/drivers/platform/x86/acer-wmi.c
+++ b/drivers/platform/x86/acer-wmi.c
@@ -190,6 +190,7 @@ enum interface_flags {
ACER_AMW0,
ACER_AMW0_V2,
ACER_WMID,
+   ACER_WMID_v2,
 };
 
 #define ACER_DEFAULT_WIRELESS  0
@@ -868,6 +869,176 @@ static acpi_status WMID_set_u32(u32 value, u32 cap, 
struct wmi_interface *iface)
return WMI_execute_u32(method_id, (u32)value, NULL);
 }
 
+static acpi_status wmid3_get_device_status(u32 *value, u16 device)
+{
+   struct wmid3_gds_return_value return_value;
+   acpi_status status;
+   union acpi_object *obj;
+   struct wmid3_gds_input_param params = {
+   .function_num = 0x1,
+   .hotkey_number = 0x01,
+   .devices = device,
+   };
+   struct acpi_buffer input = {
+   sizeof(struct wmid3_gds_input_param),
+   params
+   };
+   struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL };
+
+   status = wmi_evaluate_method(WMID_GUID3, 0, 0x2, input, output);
+   if (ACPI_FAILURE(status))
+   return status;
+
+   obj = output.pointer;
+
+   if (!obj)
+   return AE_ERROR;
+   else if (obj-type != ACPI_TYPE_BUFFER) {
+   kfree(obj);
+   return AE_ERROR;
+   }
+   if (obj-buffer.length != 8) {
+   pr_warn(Unknown buffer length %d\n, obj-buffer.length);
+   kfree(obj);
+   return AE_ERROR;
+   }
+
+   return_value = *((struct wmid3_gds_return_value *)obj-buffer.pointer);
+   kfree(obj);
+
+   if (return_value.error_code || return_value.ec_return_value)
+   pr_warn(Get Device Status failed: 0x%x - 0x%x\n,
+   return_value.error_code,
+   return_value.ec_return_value);
+   else
+   *value = !!(return_value.devices  device);
+
+   return status;
+}
+
+static acpi_status wmid_v2_get_u32(u32 *value, u32 cap)
+{
+   u16 device;
+
+   switch (cap) {
+   case ACER_CAP_WIRELESS:
+   device = ACER_WMID3_GDS_WIRELESS;
+   break;
+   case ACER_CAP_BLUETOOTH:
+   device = ACER_WMID3_GDS_BLUETOOTH;
+   break;
+   case ACER_CAP_THREEG:
+   device = ACER_WMID3_GDS_THREEG;
+   break;
+   default:
+

Re: [RESEND][PATCH 0/5] Support for the TS-5500 board

2011-08-10 Thread Vivien Didelot 
Hi,

Le Wed, 10 Aug 2011 08:18:28 +0200,
Corentin Chary corentin.ch...@gmail.com a écrit :

 On Wed, Aug 10, 2011 at 7:53 AM, Ike Panhc ike@canonical.com
 wrote:
  I think platform-driver-x86 may not be the best mailing list for
  your patches.
 
  Please try to use scripts/get_maintainer.pl and it can tell you who
  you shall ask for reviewing.
 
 Like Ike said, this is the mainling list for drivers/platform/x86, not
 arch/x86, and running get_maintainer.pl on your patchs will give you
 the rights people to CC.

Thanks for this information, it can be useful indeed.

 
 And btw, I'm not sure we really want to put all these drivers in
 arch/x86/platform/ts5500/ a re-create the arm driver mess.
 drivers/leds/, drivers/gpio/  and drivers/mtd/ are probably a best
 choice.

In fact, this set of patches is a rewrite of the first RFC set, where
we discussed that arch/x86/platform/ could be a good place, as the
TS-5500 is a specific platform with its platform devices (the
corresponding thread is here: https://lkml.org/lkml/2011/5/3/60).

Regards,
Vivien.
--
To unsubscribe from this list: send the line unsubscribe platform-driver-x86 
in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [RESEND][PATCH 0/5] Support for the TS-5500 board

2011-08-10 Thread David Woodhouse 
On Wed, 2011-08-10 at 10:06 -0400, Vivien Didelot wrote:
  And btw, I'm not sure we really want to put all these drivers in
  arch/x86/platform/ts5500/ a re-create the arm driver mess.
  drivers/leds/, drivers/gpio/  and drivers/mtd/ are probably a best
  choice.
 
 In fact, this set of patches is a rewrite of the first RFC set, where
 we discussed that arch/x86/platform/ could be a good place, as the
 TS-5500 is a specific platform with its platform devices (the
 corresponding thread is here: https://lkml.org/lkml/2011/5/3/60). 

Just expose these devices in the device-tree. You shouldn't *need* a new
MTD device driver; physmap_of should be just fine.

If your bootloader doesn't pass one, your platform-specific code should
just register its own device-tree blob during early boot. And your
device drivers should just match against the devices therein.

-- 
dwmw2

--
To unsubscribe from this list: send the line unsubscribe platform-driver-x86 
in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [RESEND][PATCH 0/5] Support for the TS-5500 board

2011-08-10 Thread Vivien Didelot 
On Wed, 10 Aug 2011 15:20:46 +0100,
David Woodhouse dw...@infradead.org wrote:
 Just expose these devices in the device-tree. You shouldn't *need* a
 new MTD device driver; physmap_of should be just fine.

I didn't write a new MTD device driver, but just moved the already
existing one to the ts5500 platform directory. The point is that my
TS-5500 doesn't have this flash, so I couldn't try. The fifth patch
from this set could just be ignored.

 If your bootloader doesn't pass one, your platform-specific code
 should just register its own device-tree blob during early boot. And
 your device drivers should just match against the devices therein.

I'm a bit confused with what should be done for the support for an
embedded platform (X86-based) like this TS-5500. Is the actual code
hierarchy OK? What is the impact of adding a device-tree to this code?

Thanks,
Vivien.
--
To unsubscribe from this list: send the line unsubscribe platform-driver-x86 
in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [Fwd: Re: A problem about acer-wmi]

2011-08-10 Thread AceLan Kao 
Dear Joey,

This is the dmesg log.
[ 9010.944053] acer_wmi: Acer Laptop ACPI-WMI Extras
[ 9010.944071] acer_wmi: Function bitmap for Communication Button: 0x1
[ 9010.944075] acer_wmi: Brightness must be controlled by generic video driver
[ 9010.944996] input: Acer WMI hotkeys as /devices/virtual/input/input21
[ 9010.947095] acer_wmi: Set Device Status failed: 0xe2 - 0x0

And after adding my quirk, the set device status still failed, but
except that, everything works well.

The failed comes from acer_rfkill_init(), it'll call
acer_rfkill_register() with ACER_CAP_WIRELESS parameter without
checking the capability.
I don't know why only wifi doesn't check the capability and call
acer_rfkill_register() directly, but it doesn't hurt the system.

===
static int acer_rfkill_init(struct device *dev)
{
wireless_rfkill = acer_rfkill_register(dev, RFKILL_TYPE_WLAN,
acer-wireless, ACER_CAP_WIRELESS);
if (IS_ERR(wireless_rfkill))
return PTR_ERR(wireless_rfkill);
===

Best regards,
AceLan Kao.

2011/8/10 Joey Lee j...@novell.com:
 Hi AceLan,

 於 三，2011-08-10 於 10:13 +0800，joeyli 提到：
 於 二，2011-08-09 於 14:33 +0800，AceLan Kao 提到：
  Dear Joey,
 
  The current project we have right now have to follow the Acer WMI to
  handle the key events, that means no EC key event.
  And we have 3 keys that are not working now, they are touchpad
 toggle,
  brightness up, and brightness down.
  The touchpad toggle function is a Hotkey Break Event(function number
  0x2), and brightness key events are Brightness Change Event(function
  number 0x4). But now acer-wmi driver only handles General Hotkey
  Event(function number 0x1).
 

 Yes, current acer-wmi only capture the event function number 0x1, you
 can add those new function to acer_wmi_notify().

  I just implemented those 3 key events, so I think maybe we don't
 have
  to create a new acer-wmi driver.

 Great! Welcome for you patches, I will also test it.

  I'm longing for your work to clear up the acer-wmi driver, so that I
  can add the new machine id and send you the patch.
  Thanks.
 
  BTW, I'm available to help if you are too busy to do that.
 
  Best regards,
  AceLan Kao.
 

 I am doing the clear up, now, will send out patch (I hope today).


 Thank's
 Joey Lee


 I add a new ACER_WMID_v2 interface flag and do some clear up in acer-wmi
 initial function and get_u32 functions.

 Please kindly test this patch:


 Thank's
 Joey Lee

 From 28b2e2ebaa230d339d5749b581c667ed074bb7ea Mon Sep 17 00:00:00 2001
 From: Lee, Chun-Yi j...@suse.com
 Date: Wed, 10 Aug 2011 16:36:02 +0800
 Subject: [PATCH] acer-wmi: add ACER_WMID_v2 interface flag to represent new 
 notebooks

 There have new acer notebooks' BIOS provide new WMID_GUID3 and
 ACERWMID_EVENT_GUID methods.

 Some of machines still keep the old WMID_GUID1 method but more and
 more machines were already removed old wmi methods from DSDT.

 So, this patch add a new ACER_WMID_v2 interface flag to represent
 new acer notebooks, the following is definition:

  + ACER_WMID:
It means this machine only provides WMID_GUID1/2 methods.

  + ACER_WMID_v2:
It means this machine provide new WMID_GUID3 and WMID_EVENT_GUID
methods.
Some ACER_WMID_v2 machines also provide old WMID_GUID1/2 methods,
but we still query/set communication device's state by new
WMID_GUID3 method.

 Tested on Acer Travelmate 8572

 Signed-off-by: Lee, Chun-Yi j...@suse.com
 ---
  drivers/platform/x86/acer-wmi.c |  409 
 ---
  1 files changed, 211 insertions(+), 198 deletions(-)

 diff --git a/drivers/platform/x86/acer-wmi.c b/drivers/platform/x86/acer-wmi.c
 index af2bb20..712a505 100644
 --- a/drivers/platform/x86/acer-wmi.c
 +++ b/drivers/platform/x86/acer-wmi.c
 @@ -190,6 +190,7 @@ enum interface_flags {
ACER_AMW0,
ACER_AMW0_V2,
ACER_WMID,
 +   ACER_WMID_v2,
  };

  #define ACER_DEFAULT_WIRELESS  0
 @@ -868,6 +869,176 @@ static acpi_status WMID_set_u32(u32 value, u32 cap, 
 struct wmi_interface *iface)
return WMI_execute_u32(method_id, (u32)value, NULL);
  }

 +static acpi_status wmid3_get_device_status(u32 *value, u16 device)
 +{
 +   struct wmid3_gds_return_value return_value;
 +   acpi_status status;
 +   union acpi_object *obj;
 +   struct wmid3_gds_input_param params = {
 +   .function_num = 0x1,
 +   .hotkey_number = 0x01,
 +   .devices = device,
 +   };
 +   struct acpi_buffer input = {
 +   sizeof(struct wmid3_gds_input_param),
 +   params
 +   };
 +   struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL };
 +
 +   status = wmi_evaluate_method(WMID_GUID3, 0, 0x2, input, output);
 +   if (ACPI_FAILURE(status))
 +   return status;
 +
 +   obj = output.pointer;
 +
 +   if (!obj)
 +   return AE_ERROR;
 +   else if (obj-type != ACPI_TYPE_BUFFER) {
 +

Re: [Fwd: Re: A problem about acer-wmi]

2011-08-10 Thread Joey Lee 
於 四，2011-08-11 於 10:48 +0800，AceLan Kao 提到：
 Dear Joey,
 
 This is the dmesg log.
 [ 9010.944053] acer_wmi: Acer Laptop ACPI-WMI Extras
 [ 9010.944071] acer_wmi: Function bitmap for Communication Button: 0x1
 [ 9010.944075] acer_wmi: Brightness must be controlled by generic video driver
 [ 9010.944996] input: Acer WMI hotkeys as /devices/virtual/input/input21
 [ 9010.947095] acer_wmi: Set Device Status failed: 0xe2 - 0x0
 
 And after adding my quirk, the set device status still failed, but
 except that, everything works well.
 

So, you machine have wifi hardware module or not?

 The failed comes from acer_rfkill_init(), it'll call
 acer_rfkill_register() with ACER_CAP_WIRELESS parameter without
 checking the capability.
 I don't know why only wifi doesn't check the capability and call
 acer_rfkill_register() directly, but it doesn't hurt the system.
 
 ===
 static int acer_rfkill_init(struct device *dev)
 {
 wireless_rfkill = acer_rfkill_register(dev, RFKILL_TYPE_WLAN,
 acer-wireless, ACER_CAP_WIRELESS);
 if (IS_ERR(wireless_rfkill))
 return PTR_ERR(wireless_rfkill);
 ===
 
 Best regards,
 AceLan Kao.
 

I checked the history, looks like it's just a original design in old
patch.

Yes, you are right, add ACER_CAP_WIRELESS check is better. I am doing
generate a patch to add ACER_CAP_WIRELESS check.
Will attached patch on mail.


Thank's
Joey Lee

 2011/8/10 Joey Lee j...@novell.com:
  Hi AceLan,
 
  於 三，2011-08-10 於 10:13 +0800，joeyli 提到：
  於 二，2011-08-09 於 14:33 +0800，AceLan Kao 提到：
   Dear Joey,
  
   The current project we have right now have to follow the Acer WMI to
   handle the key events, that means no EC key event.
   And we have 3 keys that are not working now, they are touchpad
  toggle,
   brightness up, and brightness down.
   The touchpad toggle function is a Hotkey Break Event(function number
   0x2), and brightness key events are Brightness Change Event(function
   number 0x4). But now acer-wmi driver only handles General Hotkey
   Event(function number 0x1).
  
 
  Yes, current acer-wmi only capture the event function number 0x1, you
  can add those new function to acer_wmi_notify().
 
   I just implemented those 3 key events, so I think maybe we don't
  have
   to create a new acer-wmi driver.
 
  Great! Welcome for you patches, I will also test it.
 
   I'm longing for your work to clear up the acer-wmi driver, so that I
   can add the new machine id and send you the patch.
   Thanks.
  
   BTW, I'm available to help if you are too busy to do that.
  
   Best regards,
   AceLan Kao.
  
 
  I am doing the clear up, now, will send out patch (I hope today).
 
 
  Thank's
  Joey Lee
 
 
  I add a new ACER_WMID_v2 interface flag and do some clear up in acer-wmi
  initial function and get_u32 functions.
 
  Please kindly test this patch:
 
 
  Thank's
  Joey Lee
 
  From 28b2e2ebaa230d339d5749b581c667ed074bb7ea Mon Sep 17 00:00:00 2001
  From: Lee, Chun-Yi j...@suse.com
  Date: Wed, 10 Aug 2011 16:36:02 +0800
  Subject: [PATCH] acer-wmi: add ACER_WMID_v2 interface flag to represent new 
  notebooks
 
  There have new acer notebooks' BIOS provide new WMID_GUID3 and
  ACERWMID_EVENT_GUID methods.
 
  Some of machines still keep the old WMID_GUID1 method but more and
  more machines were already removed old wmi methods from DSDT.
 
  So, this patch add a new ACER_WMID_v2 interface flag to represent
  new acer notebooks, the following is definition:
 
   + ACER_WMID:
 It means this machine only provides WMID_GUID1/2 methods.
 
   + ACER_WMID_v2:
 It means this machine provide new WMID_GUID3 and WMID_EVENT_GUID
 methods.
 Some ACER_ but we still query/set communication device's 
  state by new
 WMID_GUID3 method.
 
  Tested on Acer Travelmate 8572
 
  Signed-off-by: Lee, Chun-Yi j...@suse.com
  ---
   drivers/platform/x86/acer-wmi.c |  409 
  ---
   1 files changed, 211 insertions(+), 198 deletions(-)
 
  diff --git a/drivers/platform/x86/acer-wmi.c 
  b/drivers/platform/x86/acer-wmi.c
  index af2bb20..712a505 100644
  --- a/drivers/platform/x86/acer-wmi.c
  +++ b/drivers/platform/x86/acer-wmi.c
  @@ -190,6 +190,7 @@ enum interface_flags {
 ACER_AMW0,
 ACER_AMW0_V2,
 ACER_WMID,
  +   ACER_WMID_v2,
   };
 
   #define ACER_DEFAULT_WIRELESS  0
  @@ -868,6 +869,176 @@ static acpi_status WMID_set_u32(u32 value, u32 cap, 
  struct wmi_interface *iface)
 return WMI_execute_u32(method_id, (u32)value, NULL);
   }
 
  +static acpi_status wmid3_get_device_status(u32 *value, u16 device)
  +{
  +   struct wmid3_gds_return_value return_value;
  +   acpi_status status;
  +   union acpi_object *obj;
  +   struct wmid3_gds_input_param params = {
  +   .function_num = 0x1,
  +   .hotkey_number = 0x01,
  +   .devices = device,
  +   };
  +   struct acpi_buffer input = {
  +

Re: [PATCH 21/34] drivers/platform changes for SMBIOS and System Firmware

2011-08-10 Thread Jonathan Woithe 
On Mon, Jul 18, 2011 at 09:08:35AM -0400, Prarit Bhargava wrote:
 - Replace old dmi* structures and functions with new sysfw* and smbios*
 structures and functions in individual drivers
 - cleanup sysfw_id lookup tables
 - add exactmatch functionality
 - edit udev notifiers
 - cleanup of includes for dmi.h and mod_devicetable.h which were included in
 some files that did not need them
 
 [v2]: Re-exported dmi MODULE_ALIAS
 :
  drivers/platform/x86/fujitsu-laptop.c |   41 
 :

In-so-far as I can tell with a quick inspection, this looks ok for the
fujitsu-laptop.c module.

Acked-by: Jonathan Woithe jwoi...@physics.adelaide.edu.au

At the risk of looking ill-informed, I assume the DMI infrastructure has
been deprecated by the more general sysfw.

Regards
  jonathan
--
To unsubscribe from this list: send the line unsubscribe platform-driver-x86 
in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html