[PATCH v4 0/2] dt-bindings: input: Goodix GT7986U SPI HID Touchscreen

[PATCH v4 0/2] dt-bindings: input: Goodix GT7986U SPI HID Touchscreen

[Charles Wang]

The Goodix GT7986U touch controller report touch data according to the
HID protocol through the SPI bus. However, it is incompatible with
Microsoft's HID-over-SPI protocol.

The patchset introduces the following two changes:
1) Add goodix,gt7986u-spifw.yaml.
2) Modify the driver to align with the device binding file.

Signed-off-by: Charles Wang <charles.goodix@gmail.com>
---
Changes in v4:
- Fix dt build warnings.
- Modify the driver to align with the device binding file.

Changes in v3:
- Split the commit into two patches.

Changes in v2:
- Change compatible to 'goodix,gt7986u-spifw'.
- Remove 'goodix,hid-report-addr' property.
- Change additionalProperties to unevaluatedProperties.
- v1: https://lore.kernel.org/all/20241025114642.40793-2-charles.goodix@gmail.com/
---
Charles Wang (2):
  dt-bindings: input: Goodix GT7986U SPI HID Touchscreen
  HID: hid-goodix: Add OF supports

 .../bindings/input/goodix,gt7986u-spifw.yaml  | 69 +++++++++++++++++++
 drivers/hid/hid-goodix-spi.c                  | 17 +++--
 2 files changed, 80 insertions(+), 6 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/input/goodix,gt7986u-spifw.yaml

-- 
2.43.0

[PATCH v4 1/2] dt-bindings: input: Goodix GT7986U SPI HID Touchscreen

[Charles Wang]

The Goodix GT7986U touch controller report touch data according to the
HID protocol through the SPI bus. However, it is incompatible with
Microsoft's HID-over-SPI protocol.

NOTE: these bindings are distinct from the bindings used with the
GT7986U when the chip is running I2C firmware. For some background,
see discussion on the mailing lists in the thread:

https://lore.kernel.org/r/20241018020815.3098263-2-charles.goodix@gmail.com

Signed-off-by: Charles Wang <charles.goodix@gmail.com>
---
 .../bindings/input/goodix,gt7986u-spifw.yaml  | 69 +++++++++++++++++++
 1 file changed, 69 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/input/goodix,gt7986u-spifw.yaml

diff --git a/Documentation/devicetree/bindings/input/goodix,gt7986u-spifw.yaml b/Documentation/devicetree/bindings/input/goodix,gt7986u-spifw.yaml
new file mode 100644
index 000000000..92bd0041f
--- /dev/null
+++ b/Documentation/devicetree/bindings/input/goodix,gt7986u-spifw.yaml
@@ -0,0 +1,69 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/input/goodix,gt7986u-spifw.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Goodix GT7986U SPI HID Touchscreen
+
+maintainers:
+  - Charles Wang <charles.goodix@gmail.com>
+
+description: |
+  Supports the Goodix GT7986U touchscreen.
+  This touch controller reports data packaged according to the HID protocol
+  over the SPI bus, but it is incompatible with Microsoft's HID-over-SPI protocol.
+
+  NOTE: these bindings are distinct from the bindings used with the
+  GT7986U when the chip is running I2C firmware. This is because there's
+  not a single device that talks over both I2C and SPI but rather
+  distinct touchscreens that happen to be built with the same ASIC but
+  that are distinct products running distinct firmware.
+
+allOf:
+  - $ref: /schemas/spi/spi-peripheral-props.yaml#
+
+properties:
+  compatible:
+    enum:
+      - goodix,gt7986u-spifw
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    maxItems: 1
+
+  reset-gpios:
+    maxItems: 1
+
+  spi-max-frequency: true
+
+required:
+  - compatible
+  - reg
+  - interrupts
+  - reset-gpios
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/irq.h>
+    #include <dt-bindings/gpio/gpio.h>
+
+    spi {
+      #address-cells = <1>;
+      #size-cells = <0>;
+
+      touchscreen@0 {
+        compatible = "goodix,gt7986u-spifw";
+        reg = <0>;
+        interrupt-parent = <&gpio>;
+        interrupts = <25 IRQ_TYPE_LEVEL_LOW>;
+        reset-gpios = <&gpio1 1 GPIO_ACTIVE_LOW>;
+        spi-max-frequency = <10000000>;
+      };
+    };
+
+...
-- 
2.43.0

Re: [PATCH v4 1/2] dt-bindings: input: Goodix GT7986U SPI HID Touchscreen

[Doug Anderson]

Hi,

On Sun, Nov 10, 2024 at 11:50 PM Charles Wang <charles.goodix@gmail.com> wrote:
>
> The Goodix GT7986U touch controller report touch data according to the
> HID protocol through the SPI bus. However, it is incompatible with
> Microsoft's HID-over-SPI protocol.
>
> NOTE: these bindings are distinct from the bindings used with the
> GT7986U when the chip is running I2C firmware. For some background,
> see discussion on the mailing lists in the thread:
>
> https://lore.kernel.org/r/20241018020815.3098263-2-charles.goodix@gmail.com
>
> Signed-off-by: Charles Wang <charles.goodix@gmail.com>
> ---
>  .../bindings/input/goodix,gt7986u-spifw.yaml  | 69 +++++++++++++++++++
>  1 file changed, 69 insertions(+)

As far as I can tell this looks fine now. Thanks!

Reviewed-by: Douglas Anderson <dianders@chromium.org>

Re: [PATCH v4 1/2] dt-bindings: input: Goodix GT7986U SPI HID Touchscreen

[Rob Herring (Arm)]

On Mon, 11 Nov 2024 15:49:59 +0800, Charles Wang wrote:
> The Goodix GT7986U touch controller report touch data according to the
> HID protocol through the SPI bus. However, it is incompatible with
> Microsoft's HID-over-SPI protocol.
> 
> NOTE: these bindings are distinct from the bindings used with the
> GT7986U when the chip is running I2C firmware. For some background,
> see discussion on the mailing lists in the thread:
> 
> https://lore.kernel.org/r/20241018020815.3098263-2-charles.goodix@gmail.com
> 
> Signed-off-by: Charles Wang <charles.goodix@gmail.com>
> ---
>  .../bindings/input/goodix,gt7986u-spifw.yaml  | 69 +++++++++++++++++++
>  1 file changed, 69 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/input/goodix,gt7986u-spifw.yaml
> 

Reviewed-by: Rob Herring (Arm) <robh@kernel.org>

[PATCH v4 2/2] HID: hid-goodix: Add OF supports

[Charles Wang]

This patch introduces the following changes:
- Adds OF match table.
- Hardcodes hid-report-addr in the driver rather than fetching it
  from the device property.

Signed-off-by: Charles Wang <charles.goodix@gmail.com>
---
 drivers/hid/hid-goodix-spi.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/hid/hid-goodix-spi.c b/drivers/hid/hid-goodix-spi.c
index 6ae2300a6..80c0288a3 100644
--- a/drivers/hid/hid-goodix-spi.c
+++ b/drivers/hid/hid-goodix-spi.c
@@ -20,6 +20,7 @@
 #define GOODIX_HID_REPORT_DESC_ADDR	0x105AA
 #define GOODIX_HID_SIGN_ADDR		0x10D32
 #define GOODIX_HID_CMD_ADDR		0x10364
+#define GOODIX_HID_REPORT_ADDR		0x22C8C
 
 #define GOODIX_HID_GET_REPORT_CMD	0x02
 #define GOODIX_HID_SET_REPORT_CMD	0x03
@@ -701,12 +702,7 @@ static int goodix_spi_probe(struct spi_device *spi)
 		return dev_err_probe(dev, PTR_ERR(ts->reset_gpio),
 				     "failed to request reset gpio\n");
 
-	error = device_property_read_u32(dev, "goodix,hid-report-addr",
-					 &ts->hid_report_addr);
-	if (error)
-		return dev_err_probe(dev, error,
-				     "failed get hid report addr\n");
-
+	ts->hid_report_addr = GOODIX_HID_REPORT_ADDR;
 	error = goodix_dev_confirm(ts);
 	if (error)
 		return error;
@@ -790,6 +786,14 @@ static const struct acpi_device_id goodix_spi_acpi_match[] = {
 MODULE_DEVICE_TABLE(acpi, goodix_spi_acpi_match);
 #endif
 
+#ifdef CONFIG_OF
+static const struct of_device_id goodix_spi_of_match[] = {
+	{ .compatible = "goodix,gt7986u-spifw", },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, goodix_spi_of_match);
+#endif
+
 static const struct spi_device_id goodix_spi_ids[] = {
 	{ "gt7986u" },
 	{ },
@@ -800,6 +804,7 @@ static struct spi_driver goodix_spi_driver = {
 	.driver = {
 		.name = "goodix-spi-hid",
 		.acpi_match_table = ACPI_PTR(goodix_spi_acpi_match),
+		.of_match_table = of_match_ptr(goodix_spi_of_match),
 		.pm = pm_sleep_ptr(&goodix_spi_pm_ops),
 	},
 	.probe =	goodix_spi_probe,
-- 
2.43.0

Re: [PATCH v4 2/2] HID: hid-goodix: Add OF supports

[Doug Anderson]

Hi,

On Sun, Nov 10, 2024 at 11:50 PM Charles Wang <charles.goodix@gmail.com> wrote:
>
> This patch introduces the following changes:
> - Adds OF match table.
> - Hardcodes hid-report-addr in the driver rather than fetching it
>   from the device property.
>
> Signed-off-by: Charles Wang <charles.goodix@gmail.com>
> ---
>  drivers/hid/hid-goodix-spi.c | 17 +++++++++++------
>  1 file changed, 11 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/hid/hid-goodix-spi.c b/drivers/hid/hid-goodix-spi.c
> index 6ae2300a6..80c0288a3 100644
> --- a/drivers/hid/hid-goodix-spi.c
> +++ b/drivers/hid/hid-goodix-spi.c
> @@ -20,6 +20,7 @@
>  #define GOODIX_HID_REPORT_DESC_ADDR    0x105AA
>  #define GOODIX_HID_SIGN_ADDR           0x10D32
>  #define GOODIX_HID_CMD_ADDR            0x10364
> +#define GOODIX_HID_REPORT_ADDR         0x22C8C
>
>  #define GOODIX_HID_GET_REPORT_CMD      0x02
>  #define GOODIX_HID_SET_REPORT_CMD      0x03
> @@ -701,12 +702,7 @@ static int goodix_spi_probe(struct spi_device *spi)
>                 return dev_err_probe(dev, PTR_ERR(ts->reset_gpio),
>                                      "failed to request reset gpio\n");
>
> -       error = device_property_read_u32(dev, "goodix,hid-report-addr",
> -                                        &ts->hid_report_addr);
> -       if (error)
> -               return dev_err_probe(dev, error,
> -                                    "failed get hid report addr\n");
> -
> +       ts->hid_report_addr = GOODIX_HID_REPORT_ADDR;
>         error = goodix_dev_confirm(ts);
>         if (error)
>                 return error;
> @@ -790,6 +786,14 @@ static const struct acpi_device_id goodix_spi_acpi_match[] = {
>  MODULE_DEVICE_TABLE(acpi, goodix_spi_acpi_match);
>  #endif
>
> +#ifdef CONFIG_OF
> +static const struct of_device_id goodix_spi_of_match[] = {
> +       { .compatible = "goodix,gt7986u-spifw", },
> +       { }
> +};
> +MODULE_DEVICE_TABLE(of, goodix_spi_of_match);
> +#endif
> +
>  static const struct spi_device_id goodix_spi_ids[] = {
>         { "gt7986u" },
>         { },
> @@ -800,6 +804,7 @@ static struct spi_driver goodix_spi_driver = {
>         .driver = {
>                 .name = "goodix-spi-hid",
>                 .acpi_match_table = ACPI_PTR(goodix_spi_acpi_match),
> +               .of_match_table = of_match_ptr(goodix_spi_of_match),

I can never quite remember what the current preference is in regards
to "OF" tables (whether to use #ifdef like you've done or mark them
`__maybe_unused`), so maybe someone will request you change it. ...but
IMO what you have is fine and looks to be properly guarded with
of_match_ptr(). As far as I'm concerned, this patch looks OK.

Oh, I guess the one "nit" is that I would have put "spi" in the
subject, making it "HID: hid-goodix-spi: Add OF supports". It might be
worth sending a v5 for that (after waiting a day or two) unless a
maintainer tells you not to.

In any case:

Reviewed-by: Douglas Anderson <dianders@chromium.org>

Re: [PATCH v4 2/2] HID: hid-goodix: Add OF supports

[Charles Wang]

Hi Doug,

On Mon, Nov 11, 2024 at 09:24:39AM -0800, Doug Anderson wrote:
> Hi,
> 
> On Sun, Nov 10, 2024 at 11:50 PM Charles Wang <charles.goodix@gmail.com> wrote:
> >
> > This patch introduces the following changes:
> > - Adds OF match table.
> > - Hardcodes hid-report-addr in the driver rather than fetching it
> >   from the device property.
> >
> > Signed-off-by: Charles Wang <charles.goodix@gmail.com>
> > ---
> >  drivers/hid/hid-goodix-spi.c | 17 +++++++++++------
> >  1 file changed, 11 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/hid/hid-goodix-spi.c b/drivers/hid/hid-goodix-spi.c
> > index 6ae2300a6..80c0288a3 100644
> > --- a/drivers/hid/hid-goodix-spi.c
> > +++ b/drivers/hid/hid-goodix-spi.c
> > @@ -20,6 +20,7 @@
> >  #define GOODIX_HID_REPORT_DESC_ADDR    0x105AA
> >  #define GOODIX_HID_SIGN_ADDR           0x10D32
> >  #define GOODIX_HID_CMD_ADDR            0x10364
> > +#define GOODIX_HID_REPORT_ADDR         0x22C8C
> >
> >  #define GOODIX_HID_GET_REPORT_CMD      0x02
> >  #define GOODIX_HID_SET_REPORT_CMD      0x03
> > @@ -701,12 +702,7 @@ static int goodix_spi_probe(struct spi_device *spi)
> >                 return dev_err_probe(dev, PTR_ERR(ts->reset_gpio),
> >                                      "failed to request reset gpio\n");
> >
> > -       error = device_property_read_u32(dev, "goodix,hid-report-addr",
> > -                                        &ts->hid_report_addr);
> > -       if (error)
> > -               return dev_err_probe(dev, error,
> > -                                    "failed get hid report addr\n");
> > -
> > +       ts->hid_report_addr = GOODIX_HID_REPORT_ADDR;
> >         error = goodix_dev_confirm(ts);
> >         if (error)
> >                 return error;
> > @@ -790,6 +786,14 @@ static const struct acpi_device_id goodix_spi_acpi_match[] = {
> >  MODULE_DEVICE_TABLE(acpi, goodix_spi_acpi_match);
> >  #endif
> >
> > +#ifdef CONFIG_OF
> > +static const struct of_device_id goodix_spi_of_match[] = {
> > +       { .compatible = "goodix,gt7986u-spifw", },
> > +       { }
> > +};
> > +MODULE_DEVICE_TABLE(of, goodix_spi_of_match);
> > +#endif
> > +
> >  static const struct spi_device_id goodix_spi_ids[] = {
> >         { "gt7986u" },
> >         { },
> > @@ -800,6 +804,7 @@ static struct spi_driver goodix_spi_driver = {
> >         .driver = {
> >                 .name = "goodix-spi-hid",
> >                 .acpi_match_table = ACPI_PTR(goodix_spi_acpi_match),
> > +               .of_match_table = of_match_ptr(goodix_spi_of_match),
> 
> I can never quite remember what the current preference is in regards
> to "OF" tables (whether to use #ifdef like you've done or mark them
> `__maybe_unused`), so maybe someone will request you change it. ...but
> IMO what you have is fine and looks to be properly guarded with
> of_match_ptr(). As far as I'm concerned, this patch looks OK.
> 
> Oh, I guess the one "nit" is that I would have put "spi" in the
> subject, making it "HID: hid-goodix-spi: Add OF supports". It might be
> worth sending a v5 for that (after waiting a day or two) unless a
> maintainer tells you not to.
>

Ack,

> In any case:
> 
> Reviewed-by: Douglas Anderson <dianders@chromium.org>

Thanks,
Charles

Re: [PATCH v4 2/2] HID: hid-goodix: Add OF supports

[Jiri Kosina]

On Mon, 11 Nov 2024, Doug Anderson wrote:

> Oh, I guess the one "nit" is that I would have put "spi" in the
> subject, making it "HID: hid-goodix-spi: Add OF supports". It might be
> worth sending a v5 for that (after waiting a day or two) unless a
> maintainer tells you not to.

No need to resend v5 just for this, I'll adjust manually while applying.

Thanks,

-- 
Jiri Kosina
SUSE Labs

Re: [PATCH v4 0/2] dt-bindings: input: Goodix GT7986U SPI HID Touchscreen

[Jiri Kosina]

On Mon, 11 Nov 2024, Charles Wang wrote:

> The Goodix GT7986U touch controller report touch data according to the
> HID protocol through the SPI bus. However, it is incompatible with
> Microsoft's HID-over-SPI protocol.
> 
> The patchset introduces the following two changes:
> 1) Add goodix,gt7986u-spifw.yaml.
> 2) Modify the driver to align with the device binding file.
> 
> Signed-off-by: Charles Wang <charles.goodix@gmail.com>
> ---
> Changes in v4:
> - Fix dt build warnings.
> - Modify the driver to align with the device binding file.
> 
> Changes in v3:
> - Split the commit into two patches.
> 
> Changes in v2:
> - Change compatible to 'goodix,gt7986u-spifw'.
> - Remove 'goodix,hid-report-addr' property.
> - Change additionalProperties to unevaluatedProperties.
> - v1: https://lore.kernel.org/all/20241025114642.40793-2-charles.goodix@gmail.com/

Now queued in hid.git#for-6.13/goodix. Thanks,

-- 
Jiri Kosina
SUSE Labs