Re: [RFC v3] iio: input-bridge: optionally bridge iio acceleometers to create a /dev/input interface
From: Bastien Nocera
Date: Fri May 10 2019 - 04:58:20 EST
On Tue, 2019-04-16 at 21:33 +0200, H. Nikolaus Schaller wrote:
> Hi Bastien,
>
> > Am 16.04.2019 um 18:04 schrieb Bastien Nocera <hadess@xxxxxxxxxx>:
> > This can be done in user-space, reading the data from the IIO driver,
> > and using uinput to feed it back. Why is doing this at the kernel level
> > better?
>
> Well, I'd estimate that >80% of the current kernel could be done in user-space
> (but not at the same speed/quality).
>
> E.g. TCP could most likely be done by directly accessing the Ethernet layer and
> providing other processes access through named pipes instead of sockets.
>
> But usually a user-space daemon feeding things back into the kernel is slower
> (because it is scheduled differently) and needs more resources for running the
> process and IPC and is less protected against hickups and deadlocks.
This is mostly irrelevant for the amount of data we're treating, but it
doesn't matter too much.
> Two more aspects come to my mind from reading your project page:
>
> a) "It requires libgudev and systemd"
> b) "Note that a number of kernel bugs will prevent it from working correctly"
>
> a) this makes quite significant assumptions about the user-space while a kernel
> driver can be kept independent of this
It's made for modern desktop OSes/"traditional" Linux. I don't think
that those 2 libraries are problematic dependencies unless you're on
Android, where a replacement could be implemented or iio-sensor-proxy
modified for that use case.
> b) if it is in-kernel it will be kept in sync with kernel changes and such bugs
> are less likely
No they're not. This warning was because 1) drivers sometimes have bugs
2) user-space sometimes has bugs 3) user-space sometimes causes the
kernel to have bugs.
The 2 significant breakages for iio-sensor-proxy were caused by runtime
PM bugs in the hid-sensor-hub driver, and in the USB core. I doubt a
kernel-space implementation would have been able to magically fix those
bugs unfortunately.