Hi Dmitry,
I am really confused about this distinction. Could you please elaborate why wacom driver only supports 0x056a (and, curiously, some Lenovo) devices.
We want 0x2d1F to work with ***ONLY*** with the generic HID driver because the every recent firmware is designed for that; no help is needed from other sides.
In contrast, the most of the devices is required help from other sides. Therefore, the devices with VID 0x2D1F devices are supposed to run alone and separated out from our brand shipped products.
I don't know how much we can go further to tell you about this -because of our business, but 0x2D1F was obtained such purpose. Your understanding is much appreciated.
Thanks,
Tats
-----Original Message-----
From: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Sent: Tuesday, October 19, 2021 10:51 AM
To: Ping Cheng <redacted>
Cc: Alistair Francis <redacted>; shawnguo@kernel.org; s.hauer@pengutronix.de; <redacted>; Jiri Kosina <jikos@kernel.org>; Benjamin Tissoires <redacted>; linux-input <redacted>; <redacted>; LKML <redacted>; <redacted>; <redacted>
Subject: Re: [PATCH v11 1/4] HID: wacom_sys: Add support for flipping the data values
[EXTERNAL]
Hi Ping,
On Mon, Oct 18, 2021 at 10:41:55AM -0700, Ping Cheng wrote:
Hi Alistair,
On Sat, Oct 9, 2021, 4:44 AM Alistair Francis [off-list ref]
wrote:
quoted
Add support to the Wacom HID device for flipping the values based on
device tree settings. This allows us to support devices where the
panel is installed in a different orientation, such as the reMarkable2.
This device was designed for hid-generic driver, if it's not driven by
wacom_i2c.c or an out of tree driver.
wacom.ko doesn't support vid 0x2d1f devices.
I am really confused about this distinction. Could you please elaborate why wacom driver only supports 0x056a (and, curiously, some Lenovo) devices.
Thanks.
Nacked-by: Ping Cheng [off-list ref]
Sorry about that,
Ping
Signed-off-by: Alistair Francis <redacted>
quoted
---
.../bindings/input/hid-over-i2c.txt | 20 ++++++
drivers/hid/wacom_sys.c | 25 ++++++++
drivers/hid/wacom_wac.c | 61 +++++++++++++++++++
drivers/hid/wacom_wac.h | 13 ++++
4 files changed, 119 insertions(+)
a/Documentation/devicetree/bindings/input/hid-over-i2c.txt
b/Documentation/devicetree/bindings/input/hid-over-i2c.txt
index c76bafaf98d2..16ebd7c46049 100644
--- a/Documentation/devicetree/bindings/input/hid-over-i2c.txt
+++ b/Documentation/devicetree/bindings/input/hid-over-i2c.txt
@@ -33,6 +33,26 @@ device-specific compatible properties, which
should be used in addition to the
- post-power-on-delay-ms: time required by the device after
enabling its regulators
or powering it on, before it is ready for communication.
+ flip-tilt-x:
+ type: boolean
+ description: Flip the tilt X values read from device
+
+ flip-tilt-y:
+ type: boolean
+ description: Flip the tilt Y values read from device
Do these really need to be controlled separately from the main touchcsreen orientation?
quoted
+
+ flip-pos-x:
+ type: boolean
+ description: Flip the X position values read from device
+
+ flip-pos-y:
+ type: boolean
+ description: Flip the Y position values read from device
We already have touchscreen-inverted-x/y defined in Documentation/devicetree/bindings/input/touchscreen/touchscreen.yaml,
why are they not sufficient?
quoted
+
+ flip-distance:
+ type: boolean
+ description: Flip the distance values read from device
I am still confused of the notion of flipped distance.
quoted
+
Example:
i2c-hid-dev@2c {diff --git a/drivers/hid/wacom_sys.c b/drivers/hid/wacom_sys.c index
93f49b766376..47d9590b10bd 100644
--- a/drivers/hid/wacom_sys.c
+++ b/drivers/hid/wacom_sys.c
@@ -10,6 +10,7 @@
#include "wacom_wac.h"
#include "wacom.h"
+#include <linux/of.h>
#include <linux/input/mt.h>
#define WAC_MSG_RETRIES 5
@@ -2730,6 +2731,28 @@ static void wacom_mode_change_work(struct
work_struct *work)
return;
}
+static void wacom_of_read(struct hid_device *hdev, struct wacom_wac
*wacom_wac)
+{
+ if (IS_ENABLED(CONFIG_OF)) {
+ wacom_wac->flip_tilt_x =
of_property_read_bool(hdev->dev.parent->of_node,
+ "flip-tilt-x");
+ wacom_wac->flip_tilt_y =
of_property_read_bool(hdev->dev.parent->of_node,
+ "flip-tilt-y");
+ wacom_wac->flip_pos_x =
of_property_read_bool(hdev->dev.parent->of_node,
+ "flip-pos-x");
+ wacom_wac->flip_pos_y =
of_property_read_bool(hdev->dev.parent->of_node,
+ "flip-pos-y");
+ wacom_wac->flip_distance =
of_property_read_bool(hdev->dev.parent->of_node,
+ "flip-distance");
53 more quoted lines
You are still parsing raw data. The point of HID is to provide common framework for scaling raw values.
Thanks.
--
Dmitry