Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <3dec29bf-b772-d82d-fff9-6c8bcca5f464@redhat.com>
Date:   Wed, 29 Mar 2023 11:22:18 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.7.1
Subject: Re: [REGRESSION] wrong coord from Thinkpad TrackPoint since 6.2
 kernel
To:     Takashi Iwai <tiwai@suse.de>,
        =?UTF-8?Q?Pali_Roh=c3=a1r?= <pali@kernel.org>,
        Dmitry Torokhov <dmitry.torokhov@gmail.com>
Cc:     regressions@lists.linux.dev, linux-input@vger.kernel.org,
        linux-kernel@vger.kernel.org
References: <87h6u4otuc.wl-tiwai@suse.de>
Content-Language: en-US, nl
From:   Hans de Goede <hdegoede@redhat.com>
In-Reply-To: <87h6u4otuc.wl-tiwai@suse.de>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

Hi Takashi,

On 3/29/23 10:16, Takashi Iwai wrote:
> Hi,
> 
> we've received a bug report about Thinkpad TrackPoint (ALPS DualPoint
> Stick) on 6.2 kernel:
>   https://bugzilla.opensuse.org/show_bug.cgi?id=1209805
> 
> The device reports the wrong values as the movements, e.g. sometimes a
> value such as 255, 254 or -255 is returned while usually it should be
> a smaller value like -1 or 2.
> 
> The evtest on 6.2.x kernel shows the wrong values like:
> 
> Event: time 1680037542.898747, type 2 (EV_REL), code 0 (REL_X), value 255
> Event: time 1680037542.898747, -------------- SYN_REPORT ------------
> Event: time 1680037543.145196, type 2 (EV_REL), code 0 (REL_X), value 1
> Event: time 1680037543.145196, -------------- SYN_REPORT ------------
> Event: time 1680037543.175087, type 2 (EV_REL), code 1 (REL_Y), value -255
> Event: time 1680037543.175087, -------------- SYN_REPORT ------------
> Event: time 1680037543.185421, type 2 (EV_REL), code 0 (REL_X), value 1
> Event: time 1680037543.185421, type 2 (EV_REL), code 1 (REL_Y), value -255
> Event: time 1680037543.185421, -------------- SYN_REPORT ------------
> 
> while 6.1.x kernel shows the correct values like:
> 
> Event: time 1680037386.318058, type 2 (EV_REL), code 0 (REL_X), value -1
> Event: time 1680037386.318058, type 2 (EV_REL), code 1 (REL_Y), value -1
> Event: time 1680037386.318058, -------------- SYN_REPORT ------------
> Event: time 1680037386.328087, type 2 (EV_REL), code 0 (REL_X), value -1
> Event: time 1680037386.328087, type 2 (EV_REL), code 1 (REL_Y), value -1
> Event: time 1680037386.328087, -------------- SYN_REPORT ------------
> Event: time 1680037386.338046, type 2 (EV_REL), code 0 (REL_X), value -1
> Event: time 1680037386.338046, type 2 (EV_REL), code 1 (REL_Y), value -2
> Event: time 1680037386.338046, -------------- SYN_REPORT ------------
>   
> I couldn't see any relevant changes in alps.c between those versions,
> so this is likely a breakage in a lower layer.
> 
> Could you guys take a look?

I believe this is caused by the kernel now using -funsigned-char
everywhere and this should be fixed by this commit:

https://git.kernel.org/pub/scm/linux/kernel/git/dtor/input.git/commit/?h=for-linus&id=754ff5060daf5a1cf4474eff9b4edeb6c17ef7ab

And there is a similar issue in the focaltech touchpad driver:

https://git.kernel.org/pub/scm/linux/kernel/git/dtor/input.git/commit/?h=for-linus&id=8980f190947ba29f23110408e712444884b74251

Dmitry, since this is hitting 6.2 users, perhaps you can send
a pull-req for your current for-linus branch to get the fix
on its way to stable ?

Regards,

Hans