Re: [PATCH v2 2/3] HID: multitouch: add control of the feature "Maximum Contact Number"

[Benjamin Tissoires <benjamin.tissoires@enac.fr>]

Hi Henrik,

On Fri, Jan 20, 2012 at 17:19, Henrik Rydberg <rydberg@euromail.se> wrote:
> Hi Benjamin,
>
>> If the programmer fills the field maxcontacts in the mt_class,
>> then the driver will set the feature to this value. It is safe
>> for current drivers as the feature is read/right in the HID norm
>
> read/write, I presume.

you're right... ;-)

>
>> @@ -609,6 +612,36 @@ static void mt_set_input_mode(struct hid_device *hdev)
>>       }
>>  }
>>
>> +static void mt_set_maxcontacts(struct hid_device *hdev)
>> +{
>> +     struct mt_device *td = hid_get_drvdata(hdev);
>> +     struct hid_report *r;
>> +     struct hid_report_enum *re;
>> +     int fieldmax, max;
>> +
>> +     if (td->maxcontact_report_id < 0)
>> +             return;
>> +
>> +     if (!td->mtclass.maxcontacts)
>> +             return;
>> +
>> +     re = &(hdev->report_enum[HID_FEATURE_REPORT]);
>
> Why parenthesis here?

ok

>
>> +     r = re->report_id_hash[td->maxcontact_report_id];
>> +     if (r) {
>> +             max = td->mtclass.maxcontacts;
>> +             fieldmax = r->field[0]->logical_maximum;
>> +             hid_info(hdev, "%s: value = %d / %d / %d\n", __func__,
>> +                     r->field[0]->value[0],
>> +                     td->mtclass.maxcontacts,
>> +                     fieldmax);
>
> Looks like debug information, please downplay or remove.

I can remove them if you need.

>
>> +             max = fieldmax < max ? fieldmax : max;
>
> There are kernel variants of min/max functions to consider here.

I'll look for them.

>
>> +             if (r->field[0]->value[0] != max) {
>> +                     r->field[0]->value[0] = max;
>> +                     usbhid_submit_report(hdev, r, USB_DIR_OUT);
>> +             }
>
> So the report is in fact only set when different... Do we know for
> which current devices this is true?

Currently, I've got only the Perixx touchpad that needs this feature
to be set. The initial value is 0 whereas it should be 2.

Maybe some drivers that used the DUAL_MT_CLS_* can handle more, but I
have no idea has I don't have many of them.

Cheers,
Benjamin

>
> Thanks,
> Henrik