On Tue, Nov 27, 2012 at 03:43:05PM -0800, Christopher Heiny wrote:

On 11/27/2012 01:21 AM, Dmitry Torokhov wrote:

quoted

To save my old fingers...

Signed-off-by: Dmitry Torokhov<dmitry.torokhov@gmail.com>
---

It looks like this driver(s) need some love and I might have some time so I
will refresh my "synaptics" branch with the patches you have sent and start
working off it. If you have updates I would appreciate if you also make them
available relative to that branch. When we are ready we'll squash them all
together and apply to the official branch.

No problem - let me know which branch/tag to work with, and we'll be
happy to patch against that for the next round.

synaptics-rmi4. It is based off 3.7-rc6.

quoted

Thanks.

 drivers/input/rmi4/rmi_driver.c | 158 +++++++++++----------
 drivers/input/rmi4/rmi_driver.h |   4 +-
 drivers/input/rmi4/rmi_f01.c    | 298 ++++++++++++++++++++--------------------
 drivers/input/rmi4/rmi_f11.c    | 258 +++++++++++++++++-----------------
 include/linux/rmi.h             |  22 ++-
 5 files changed, 368 insertions(+), 372 deletions(-)

diff --git a/drivers/input/rmi4/rmi_driver.c b/drivers/input/rmi4/rmi_driver.c
index 05a73ae..e8a4b52 100644
--- a/drivers/input/rmi4/rmi_driver.c
+++ b/drivers/input/rmi4/rmi_driver.c

@@ -594,7 +594,7 @@ static struct device_attribute bsr_attribute = __ATTR(bsr, RMI_RW_ATTR,

 static void rmi_free_function_list(struct rmi_device *rmi_dev)
 {
-	struct rmi_function_container *entry, *n;
+	struct rmi_function *entry, *n;
 	struct rmi_driver_data *data = dev_get_drvdata(&rmi_dev->dev);

 	if (!data) {

@@ -613,44 +613,44 @@ static void rmi_free_function_list(struct rmi_device *rmi_dev)
 	}
 }

-static void release_fc_device(struct device *dev)
+static void release_function_device(struct device *dev)
 {
 	dev_dbg(dev, "REMOVING KOBJ!");
 	kobject_put(&dev->kobj);
 }

Hmmm.  Since rmi_function_container has evolved into a child device
of the RMI4 module, maybe it would be better renamed
rmi_function_device or rmi_function_dev?  I find this clearer, but
can live with just rmi_function if you prefer that.

I just prefer something reasonably short so I still like rmi_function or
something of similar length.

Similarly, rmi_function_handler has evolved into a driver for such
devices, so perhaps it should be renamed rmi_function_driver?

No preference here.

Thanks.

-- 
Dmitry