linux-kernel - Re: [PATCH] HID: multitouch: Add required quirk for Synaptics 0xcd7e device

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 05 Oct 2023 02:29:09 +0000
From:   Rahul Rameshbabu <sergeantsagara@...tonmail.com>
To:     Rain <rain@...showers.io>
Cc:     Jiri Kosina <jikos@...nel.org>,
        Benjamin Tissoires <benjamin.tissoires@...hat.com>,
        Henrik Rydberg <rydberg@...math.org>,
        linux-input@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] HID: multitouch: Add required quirk for Synaptics 0xcd7e device


On Wed, 04 Oct, 2023 12:01:27 -0700 "Rain" <rain@...showers.io> wrote:
> On Sun, Sep 17, 2023, at 09:18, Rahul Rameshbabu wrote:
>> Register the Synaptics device as a special multitouch device with certain
>> quirks that may improve usability of the touchpad device.
>>
>> Reported-by: Rain <rain@...showers.io>
>> Closes:
>> https://lore.kernel.org/linux-input/2bbb8e1d-1793-4df1-810f-cb0137341ff4@app.fastmail.com/
>> Signed-off-by: Rahul Rameshbabu <sergeantsagara@...tonmail.com>
>> ---
>>
>> Notes:
>>     Theory:
>>
>>       I think the Synaptics device in the related email to the linux-input
>>       mailing list requires certain quirks like MT_QUIRK_HOVERING to correctly
>>       reconfigure the distance configuration for multitouch events. This might
>>       explain why light touches were not registered originally when
>>       MT_CLS_DEFAULT was used by default for the device. Would like to have
>>       this patch tested before being merged. A Tested-by: git trailer can then
>>       be appended.
>>
>>  drivers/hid/hid-multitouch.c | 4 ++++
>>  1 file changed, 4 insertions(+)
>>
>> diff --git a/drivers/hid/hid-multitouch.c b/drivers/hid/hid-multitouch.c
>> index 521b2ffb4244..8db4ae05febc 100644
>> --- a/drivers/hid/hid-multitouch.c
>> +++ b/drivers/hid/hid-multitouch.c
>> @@ -2144,6 +2144,10 @@ static const struct hid_device_id mt_devices[] = {
>>  			USB_DEVICE_ID_MTP_STM)},
>>
>>  	/* Synaptics devices */
>> +	{ .driver_data = MT_CLS_WIN_8_FORCE_MULTI_INPUT,
>> +		HID_DEVICE(BUS_I2C, HID_GROUP_MULTITOUCH_WIN_8,
>> +			USB_VENDOR_ID_SYNAPTICS, 0xcd7e) },
>
> Thanks for the patch! I haven't tested it yet but it looks promising.
>
> The vendor ID isn't USB_VENDOR_ID_SYNAPTICS (0x06cb),
> however -- it's SYNA7DB5 (0x7db5) which I guess is an alternative vendor
> ID for Synaptics. Would be worth fixing that.

Hi Rain,

I think you might be confusing the device name string with the vendor
id, based on my interpretation of the output you shared.

  [    2.034760] input: SYNA7DB5:00 06CB:CD7E Mouse as /devices/platform/AMDI0010:00/i2c-0/i2c-SYNA7DB5:00/0018:06CB:CD7E.0001/input/input1
  [    2.034865] input: SYNA7DB5:00 06CB:CD7E Touchpad as /devices/platform/AMDI0010:00/i2c-0/i2c-SYNA7DB5:00/0018:06CB:CD7E.0001/input/input2

The first column has a device name but what we are interested in is the
second column, '06CB:CD7E'. 0x06CB is the vendor id and 0xCD7E is the
device id. Hope that makes sense and let me know if you think this is
off.

In general for these types of issues, I think the right direction would
be to make quirks loadable in userspace through HID-BPF where DEs or
some systemd subsystem can help users load quirks needed for their HID
devices rather than building a list of quirks in the driver. This is
something I am interested in exploring.

Thanks,

Rahul Rameshbabu

>
>> +
>>  	{ .driver_data = MT_CLS_WIN_8_FORCE_MULTI_INPUT,
>>  		HID_DEVICE(BUS_I2C, HID_GROUP_MULTITOUCH_WIN_8,
>>  			USB_VENDOR_ID_SYNAPTICS, 0xce08) },
>> --
>> 2.40.1