2015-06-16 21:09:01

by Andrew Duggan

Subject: [PATCH] HID: rmi: Disable populating F30 when the touchpad has physical buttons

Physical buttons do not use F30 to report their state and in some cases the
data reported in F30 is incorrect and inconsistent with what is reported by
the HID descriptor. When physical buttons are present, ignore F30 and let
hid-input report buttons based on what is defined in the HID descriptor.

Signed-off-by: Andrew Duggan <[email protected]>
---
drivers/hid/hid-rmi.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/hid/hid-rmi.c b/drivers/hid/hid-rmi.c
index 368ffdf..721f7f9 100644
--- a/drivers/hid/hid-rmi.c
+++ b/drivers/hid/hid-rmi.c
@@ -1013,6 +1013,7 @@ static int rmi_populate_f30(struct hid_device *hdev)

static int rmi_populate(struct hid_device *hdev)
{
+ struct rmi_data *data = hid_get_drvdata(hdev);
int ret;

ret = rmi_scan_pdt(hdev);
@@ -1033,9 +1034,11 @@ static int rmi_populate(struct hid_device *hdev)
return ret;
}

- ret = rmi_populate_f30(hdev);
- if (ret)
- hid_warn(hdev, "Error while initializing F30 (%d).\n", ret);
+ if (!(data->device_flags & RMI_DEVICE_HAS_PHYS_BUTTONS)) {
+ ret = rmi_populate_f30(hdev);
+ if (ret)
+ hid_warn(hdev, "Error while initializing F30 (%d).\n", ret);
+ }

return 0;
}
--
2.1.4

2015-06-16 21:27:06

by Benjamin Tissoires

Subject: Re: [PATCH] HID: rmi: Disable populating F30 when the touchpad has physical buttons

Hi Andrew,

On Jun 16 2015 or thereabouts, Andrew Duggan wrote:
> Physical buttons do not use F30 to report their state and in some cases the
> data reported in F30 is incorrect and inconsistent with what is reported by
> the HID descriptor. When physical buttons are present, ignore F30 and let
> hid-input report buttons based on what is defined in the HID descriptor.
>
> Signed-off-by: Andrew Duggan <[email protected]>
> ---

Thanks for the patch.
This one is related to:
https://bugzilla.redhat.com/show_bug.cgi?id=1227891
and http://bugs.freedesktop.org/show_bug.cgi?id=90356

Andrew, can you confirm that with this patch, INPUT_PROP_BUTTONPAD is not
set for the Razer Blade touchpad?

If so, then this is:
Reviewed-by: Benjamin Tissoires <[email protected]>

Cheers,
Benjamin

> drivers/hid/hid-rmi.c | 9 ++++++---
> 1 file changed, 6 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/hid/hid-rmi.c b/drivers/hid/hid-rmi.c
> index 368ffdf..721f7f9 100644
> --- a/drivers/hid/hid-rmi.c
> +++ b/drivers/hid/hid-rmi.c
> @@ -1013,6 +1013,7 @@ static int rmi_populate_f30(struct hid_device *hdev)
>
> static int rmi_populate(struct hid_device *hdev)
> {
> + struct rmi_data *data = hid_get_drvdata(hdev);
> int ret;
>
> ret = rmi_scan_pdt(hdev);
> @@ -1033,9 +1034,11 @@ static int rmi_populate(struct hid_device *hdev)
> return ret;
> }
>
> - ret = rmi_populate_f30(hdev);
> - if (ret)
> - hid_warn(hdev, "Error while initializing F30 (%d).\n", ret);
> + if (!(data->device_flags & RMI_DEVICE_HAS_PHYS_BUTTONS)) {
> + ret = rmi_populate_f30(hdev);
> + if (ret)
> + hid_warn(hdev, "Error while initializing F30 (%d).\n", ret);
> + }
>
> return 0;
> }
> --
> 2.1.4
>

2015-06-16 21:38:06

by Andrew Duggan

Subject: Re: [PATCH] HID: rmi: Disable populating F30 when the touchpad has physical buttons

On 06/16/2015 02:26 PM, Benjamin Tissoires wrote:
> Hi Andrew,
>
> On Jun 16 2015 or thereabouts, Andrew Duggan wrote:
>> Physical buttons do not use F30 to report their state and in some cases the
>> data reported in F30 is incorrect and inconsistent with what is reported by
>> the HID descriptor. When physical buttons are present, ignore F30 and let
>> hid-input report buttons based on what is defined in the HID descriptor.
>>
>> Signed-off-by: Andrew Duggan <[email protected]>
>> ---
> Thanks for the patch.
> This one is related to:
> https://bugzilla.redhat.com/show_bug.cgi?id=1227891
> and http://bugs.freedesktop.org/show_bug.cgi?id=90356
>
> Andrew, can you confirm that with this patch, INPUT_PROP_BUTTONPAD is not
> set for the Razer Blade touchpad?

Yes, I was able to reproduce the bug and confirm the fix on a Razer
Blade system.

> If so, then this is:
> Reviewed-by: Benjamin Tissoires <[email protected]>
>
> Cheers,
> Benjamin
>
>> drivers/hid/hid-rmi.c | 9 ++++++---
>> 1 file changed, 6 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/hid/hid-rmi.c b/drivers/hid/hid-rmi.c
>> index 368ffdf..721f7f9 100644
>> --- a/drivers/hid/hid-rmi.c
>> +++ b/drivers/hid/hid-rmi.c
>> @@ -1013,6 +1013,7 @@ static int rmi_populate_f30(struct hid_device *hdev)
>>
>> static int rmi_populate(struct hid_device *hdev)
>> {
>> + struct rmi_data *data = hid_get_drvdata(hdev);
>> int ret;
>>
>> ret = rmi_scan_pdt(hdev);
>> @@ -1033,9 +1034,11 @@ static int rmi_populate(struct hid_device *hdev)
>> return ret;
>> }
>>
>> - ret = rmi_populate_f30(hdev);
>> - if (ret)
>> - hid_warn(hdev, "Error while initializing F30 (%d).\n", ret);
>> + if (!(data->device_flags & RMI_DEVICE_HAS_PHYS_BUTTONS)) {
>> + ret = rmi_populate_f30(hdev);
>> + if (ret)
>> + hid_warn(hdev, "Error while initializing F30 (%d).\n", ret);
>> + }
>>
>> return 0;
>> }
>> --
>> 2.1.4
>>

2015-06-16 21:41:48

by Jiri Kosina

Subject: Re: [PATCH] HID: rmi: Disable populating F30 when the touchpad has physical buttons

Hi Andrew, Benjamin,

applied to for-4.2/rmi. Thanks,

--
Jiri Kosina
SUSE Labs