Re: [PATCH 5/6] platform/x86: dell-smo8800: Instantiate an i2c_client for the IIO st_accel driver

[Hans de Goede <hdegoede@redhat.com>]

Hi,

On 1/5/24 20:20, Pali Rohár wrote:
> On Friday 05 January 2024 21:04:59 Andy Shevchenko wrote:
>> On Fri, Jan 5, 2024 at 8:37 PM Andy Shevchenko
>> <andy.shevchenko@gmail.com> wrote:
>>> On Fri, Jan 5, 2024 at 6:34 PM Hans de Goede <hdegoede@redhat.com> wrote:
>>>> On 12/24/23 23:03, Pali Rohár wrote:
>>>>> On Sunday 24 December 2023 22:36:21 Hans de Goede wrote:
>>
>> ...
>>
>>>> But AFAIK / AFAICT there are no actual userspace consumers of
>>>> /dev/freefall so removing it should not be an issue.
>>>
>>> IIRC/AFAIK there is at least one (simple) computer game using it as a joystick.
>>
>> Okay, I can't google for it and now I realised that it was my x60s,
>> which has no freefall, but another interface to it. In any case the
>> side effect of that googling is this (maybe more, I just took this one
>> as example):
>> https://github.com/linux-thinkpad/hdapsd/blob/master/README.md
>>
>> So, dropping it will break at least this tool.
>>
>> -- 
>> With Best Regards,
>> Andy Shevchenko
> 
> Yes, this is that correct one. I forget the name of this daemon.
> 
> Just to note /dev/freefall does not provide axes state, it just send
> signal to process when interrupt is triggered. Process than park disk
> heads.
> 
> Axes state are/were exported throw /dev/js* interface and those games
> uses just js interface. I remember Tux Racer.
> 
> Interrupt on HP and Dell is triggered only when laptop fall is detected,
> so games did not used it (hopefully!)

Ok, so I clearly need to change the module parameter so that we stick with
the drivers/char/misc/lis3lv02d driver as default and offer using
the i2c-client-id which results in loading the iio driver instead as
an option enabled by a module parameter.

I'll fix this for v2.

Regards,

Hans