On Mon, Feb 11, 2013 at 12:19:13PM +0100, Benjamin Tissoires wrote:

On Mon, Feb 11, 2013 at 11:13 AM, Mika Westerberg
[off-list ref] wrote:

quoted

On Fri, Feb 08, 2013 at 03:37:29PM +0100, Benjamin Tissoires wrote:

quoted

so, here is the hid drivers cleanup. The aim is to remove as much as possible
direct calls to usbhid for hid drivers. Thus, other transport layers can use
the existing hid drivers (like I2C or uhid).

We also found out that there is a dependency to usbhid in many of the
drivers, especially in sensor-hub which caused the system crash while the
driver tried to transmit over USB even though it was behind I2C.

Your series fixes that nicely :)

You can add,

Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>

for the whole series and in addition,

Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com>

for the sensor-hub part (HID over I2C). I have few more patches to the
sensor-hub myself and I'm going to rebase them on top of this series.

Thanks Mika for reviewing.

I just wanted to warn you that the HID over I2C part is still lacking
the support of requests through hid_hw_request(). So the functions
sensor_hub_set_feature(), sensor_hub_get_feature() and
sensor_hub_input_attr_get_raw_value() won't work, in a silently way
:(.

OK.

I started writing the patch but stopped at some point due to lack of
ways to test it.

Well, I have the HW here so if you have something please send to me. I can
test it.