RE: [PATCH] Input: /input/mouse/elan_i2c_core.c Fix some Asus touchapod which casue TP no funciton sometimes, the patch detect some specific touchpad and run a special initialize

From: ååæ
Date: Mon Jul 11 2016 - 08:41:11 EST


Hi Dmitry,

-----Original Message-----
From: Dmitry Torokhov [mailto:dmitry.torokhov@xxxxxxxxx]
Sent: Saturday, July 09, 2016 8:25 AM
To: KT Liao
Cc: linux-kernel@xxxxxxxxxxxxxxx; linux-input@xxxxxxxxxxxxxxx; phoenix@xxxxxxxxxx; kt.liao@xxxxxxxxxx
Subject: Re: [PATCH] Input: /input/mouse/elan_i2c_core.c Fix some Asus touchapod which casue TP no funciton sometimes, the patch detect some specific touchpad and run a special initialize

Hi KT,

On Fri, Jul 08, 2016 at 08:12:09PM +0800, KT Liao wrote:
> Signed-off-by: KT Liao <kt.liao@xxxxxxxxxx>

Please make sure you add a blan line between change subject and the rest of description. Then git send-email will compose your email properly instead of lumping everything into one long line subject.
[KT] :Sorry, I will fix it in the later upstream
> ---
> drivers/input/mouse/elan_i2c_core.c | 81
> +++++++++++++++++++++++++++++--------
> 1 file changed, 65 insertions(+), 16 deletions(-)
>
> diff --git a/drivers/input/mouse/elan_i2c_core.c
> b/drivers/input/mouse/elan_i2c_core.c
> index 2f58985..1c200fb 100644
> --- a/drivers/input/mouse/elan_i2c_core.c
> +++ b/drivers/input/mouse/elan_i2c_core.c
> @@ -3,8 +3,8 @@
> *
> * Copyright (c) 2013 ELAN Microelectronics Corp.
> *
> - * Author: ææç (Duson Lin) <dusonlin@xxxxxxxxxx>
> - * Version: 1.6.0
> + * Author: KT Liao <kt.liao@xxxxxxxxxx>
> + * Version: 1.6.2
> *
> * Based on cyapa driver:
> * copyright (c) 2011-2012 Cypress Semiconductor, Inc.
> @@ -40,7 +40,7 @@
> #include "elan_i2c.h"
>
> #define DRIVER_NAME "elan_i2c"
> -#define ELAN_DRIVER_VERSION "1.6.1"
> +#define ELAN_DRIVER_VERSION "1.6.2"
> #define ELAN_VENDOR_ID 0x04f3
> #define ETP_MAX_PRESSURE 255
> #define ETP_FWIDTH_REDUCE 90
> @@ -95,6 +95,8 @@ struct elan_tp_data {
> bool baseline_ready;
> };
>
> +static int check_ASUS_special_fw(struct elan_tp_data *data);
> +
> static int elan_get_fwinfo(u8 iap_version, u16 *validpage_count,
> u16 *signature_address)
> {
> @@ -210,21 +212,40 @@ static int __elan_initialize(struct elan_tp_data *data)
> return error;
> }
>
> - data->mode |= ETP_ENABLE_ABS;
> - error = data->ops->set_mode(client, data->mode);
> - if (error) {
> - dev_err(&client->dev,
> - "failed to switch to absolute mode: %d\n", error);
> - return error;
> - }
> + /* If it's the special FW, it need a different flow for mode change.*/
> + if (check_ASUS_special_fw(data)) {
> + error = data->ops->sleep_control(client, false);
> + if (error) {
> + dev_err(&client->dev,
> + "failed to wake device up: %d\n", error);
> + return error;
> + }
>
> - error = data->ops->sleep_control(client, false);
> - if (error) {
> - dev_err(&client->dev,
> - "failed to wake device up: %d\n", error);
> - return error;
> - }
> + msleep(200);
>
> + data->mode |= ETP_ENABLE_ABS;
> + error = data->ops->set_mode(client, data->mode);
> + if (error) {
> + dev_err(&client->dev,
> + "failed to switch to absolute mode: %d\n", error);
> + return error;
> + }
> + } else {
> + data->mode |= ETP_ENABLE_ABS;
> + error = data->ops->set_mode(client, data->mode);
> + if (error) {
> + dev_err(&client->dev,
> + "failed to switch to absolute mode: %d\n", error);
> + return error;
> + }
> +
> + error = data->ops->sleep_control(client, false);
> + if (error) {
> + dev_err(&client->dev,
> + "failed to wake device up: %d\n", error);
> + return error;
> + }
> + }
> return 0;
> }
>
> @@ -757,6 +778,34 @@ out:
> return retval;
> }
>
> +static int check_ASUS_special_fw(struct elan_tp_data *data) {
> + struct i2c_client *client = data->client;
> + int error;
> +
> + error = data->ops->get_product_id(client, &data->product_id);
> + if (error)
> + return false;
> +
> + error = data->ops->get_sm_version(client, &data->ic_type,
> + &data->sm_version);
> + if (error)
> + return false;

That means we'd be fetching product ID and IC type twice when initializing the device. Can we come with a way to do it once?
[KT]:Because the elan_query_device_info() is behind the elan_initialize(). That's why I fetching product ID and IC type in the elan_initialize()
I will discuss with FW team and then execute elan_query_device_info() first to get product_id and ic_type.
> +
> + if (data->ic_type == 0x0E) {

Maybe

if (dat->ic_type != 0x0e)
return false;

? Then you would not need to indent the switch statement.
[KT] : OK, fix in the next patch
> + switch (data->product_id) {
> + case 0x05:
> + case 0x06:
> + case 0x07:
> + case 0x09:
> + case 0x13:
> + return true;
> + default:
> + return false;
> + }
> + }
> + return false;
> +}
>
> static DEVICE_ATTR_WO(acquire);
> static DEVICE_ATTR_RO(min);
> --
> 2.7.4
>

Thanks.

--
Dmitry