Re: [PATCH v11 2/7] anybus-s: support HMS Anybus-S bus

[Enrico Weigelt, metux IT consult]

On 17.04.19 00:00, Sven Van Asbroeck wrote:

Hi,

>> But this port supports probing ?>> (so it's, bit like a single usb port w/o hub ;-))> > Yes, that's
exactly it.
Okay, then your approach introducing a bus_type is correct, IMHO.

>> I guess that things like introducing new bus types or device classes>> are something that subsystem maintainers wanna be aware of :)>>>
> fieldbus_dev is a new subsystem.
> At first I attempted to mainline the anybus-s drivers with a 'roll-your-own'
> userspace API, but the feedback I got was: this userspace API should
> really be generic... So this is my attempt to create a generic fieldbus_dev
> API !

Yes, it's an own class of devices, many other (semantically) simiar
devices out there. Therefore, an own subsystem for that is a good idea.

But: I'd rather call it it "iec61158" instead of fieldbus", as there're
lots of different fieldbus types (eg. MVB has pretty different semantics
than iec61158)

> It's almost certainly insufficient, but it's a start. Ideally, this
> should get lots of review from people in the industry, but response 
> to this patchset has been underwhelming. 

It's still a niche, at least for Linux. Only few people here ever had
their hands on such PLC systems.


--mtx

-- 
Enrico Weigelt, metux IT consult
Free software and Linux embedded engineering
info@xxxxxxxxx -- +49-151-27565287