[Letux-kernel] [RFC v2] iio: input-bridge: optionally bridge iio acceleometers to create a /dev/input interface

[H. Nikolaus Schaller]

Hi Jonathan,

> Am 08.06.2019 um 12:52 schrieb Jonathan Cameron <jic23 at kernel.org>:
> 
>> So the key difference is about the question if user-space needs to tell the kernel first
>> that it wants to see a specific interface or just makes use of it if present.
> 
> Absolutely. Good summary, but I don't think either of us is going
> to persuade the other.

Well, IMHO it should less be a matter of persuading each other but one of
clear rules derived from basic system architecture principles. Unfortunately
I am not aware of a rule that could help here. KISS maybe?

Is there some general rule for Linux? Or is it decided for each subsystem individually?

It seems to be a little mixed if I compare some other subsystems... On one end
udev creates and modifies /dev entries on demand (but triggered by kernel events)
while on the other end some "aplay -l" simply lists all sound cards that exist or
"cat /proc/partitions" lists all partitions known by the kernel and lets the user-space
run mount commands if it wants to.

Maybe it is something you could bring up on some Linux core developer meeting.

> I've started work on my proposal but things have been 'interesting' in the
> last few weeks so it may be a little while yet before I have anything
> to share.

Yes, let's wait for this and then I hope others (especially users) will
join the discussion of the pros and cons of both proposals.

BR and thanks,
Nikolaus