[PATCH resend] bcm5974: Add pointer and buttonpad properties

Signed-off-by: Chase Douglas <redacted>
---
I sent to the wrong mailing list address the first time. Sorry!

 drivers/input/mouse/bcm5974.c |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/drivers/input/mouse/bcm5974.c b/drivers/input/mouse/bcm5974.c
index 5ec617e..9faa049 100644
--- a/drivers/input/mouse/bcm5974.c
+++ b/drivers/input/mouse/bcm5974.c

@@ -434,6 +434,9 @@ static void setup_events_to_report(struct input_dev *input_dev,
 	__set_bit(BTN_LEFT, input_dev->keybit);
 
 	input_set_events_per_packet(input_dev, 60);
+
+	__set_bit(INPUT_PROP_POINTER, input->propbit);
+	__set_bit(INPUT_PROP_BUTTONPAD, input->propbit);
 }
 
 /* report button data as logical button state */

-- 
1.7.9

On 02/14/2012 12:24 PM, Chase Douglas wrote:

Signed-off-by: Chase Douglas <redacted>
---
I sent to the wrong mailing list address the first time. Sorry!

Please add some comments here.
This shouldn't be the change log of this patch.

Thanks
-Wanlong Gao

On 02/14/2012 06:07 AM, Wanlong Gao wrote:

On 02/14/2012 12:24 PM, Chase Douglas wrote:

quoted

Signed-off-by: Chase Douglas <redacted>
---
I sent to the wrong mailing list address the first time. Sorry!

Please add some comments here.
This shouldn't be the change log of this patch.

I'm not sure what you want... The driver should expose these properties
because of the intrinsic nature of the device. Documentation for the
properties is found elsewhere.

-- Chase

Hi Chase,

On Mon, Feb 13, 2012 at 08:24:08PM -0800, Chase Douglas wrote:

I can take this for 3.4, but the patch needs a commit message. "In
order to move userspace in the direction of simpler and more accurate
device setup, emit device properties..."?

Thanks,
Henrik

On 02/14/2012 04:19 PM, Henrik Rydberg wrote:

Hi Chase,

On Mon, Feb 13, 2012 at 08:24:08PM -0800, Chase Douglas wrote:

quoted

Signed-off-by: Chase Douglas <redacted>
---
I sent to the wrong mailing list address the first time. Sorry!

 drivers/input/mouse/bcm5974.c |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/drivers/input/mouse/bcm5974.c b/drivers/input/mouse/bcm5974.c
index 5ec617e..9faa049 100644
--- a/drivers/input/mouse/bcm5974.c
+++ b/drivers/input/mouse/bcm5974.c

@@ -434,6 +434,9 @@ static void setup_events_to_report(struct input_dev *input_dev,
 	__set_bit(BTN_LEFT, input_dev->keybit);
 
 	input_set_events_per_packet(input_dev, 60);
+
+	__set_bit(INPUT_PROP_POINTER, input->propbit);
+	__set_bit(INPUT_PROP_BUTTONPAD, input->propbit);
 }
 
 /* report button data as logical button state */

-- 
1.7.9

I can take this for 3.4, but the patch needs a commit message. "In
order to move userspace in the direction of simpler and more accurate
device setup, emit device properties..."?

I can add that, it's just superfluous. It doesn't really tell the reader
anything if they need to know what is going on. Unless the message is
supposed to copy the definition of the properties themselves, which is
rather pointless.

It's like documenting a function:

bool is_object_mutable();
Returns true if object is mutable;

There's really no point.

I'll send a v2 version of this patch with some kind of commit message
just to get it over with, but I don't know where the insistence of a
commit message body comes from...

-- Chase

On Tue, Feb 14, 2012 at 4:23 PM, Chase Douglas
[off-list ref] wrote:

I can add that, it's just superfluous. It doesn't really tell the reader
anything if they need to know what is going on. Unless the message is
supposed to copy the definition of the properties themselves, which is
rather pointless.

It's like documenting a function:

bool is_object_mutable();
Returns true if object is mutable;

There's really no point.

We don't always have the message+code. In the end we have only commit
msgs like the raw changelog here :
http://www.kernel.org/pub/linux/kernel/v3.0/ChangeLog-3.2.6

i

On Tuesday, February 14, 2012 04:28:18 PM Mohamed Ikbel Boulabiar wrote:

On Tue, Feb 14, 2012 at 4:23 PM, Chase Douglas

[off-list ref] wrote:

quoted

I can add that, it's just superfluous. It doesn't really tell the reader
anything if they need to know what is going on. Unless the message is
supposed to copy the definition of the properties themselves, which is
rather pointless.

It's like documenting a function:

bool is_object_mutable();
Returns true if object is mutable;

There's really no point.

We don't always have the message+code. In the end we have only commit
msgs like the raw changelog here :
http://www.kernel.org/pub/linux/kernel/v3.0/ChangeLog-3.2.6

The summary of the patch would still be in the changelog:

	bcm5974: Add pointer and buttonpad properties

and I rally do not see the point of trying to expand it into the body
of the changelog if there isn't anything more to say. It's the same
as checkpatch complaining that some Kconfig options are too short
when full description does indeed fit into 1 line (for example enabling
SPI or I2C support module for a certain device when driver split
into core and bus-specific parts - there is literally nothing else to
say).

Thanks.

-- 
Dmitry