Re: ✘n2k and issue #176

[Gary E. Miller]

Yo Reinhard.Arlt@t-online.de!

On Wed, 15 Dec 2021 01:36:28 +0100 (CET)
"Reinhard.Arlt@t-online.de" <Reinhard.Arlt@t-online.de> wrote:

> Hello,

I can't tell where my comments are, and where yours are.  All mushed
together.  So I may have missed something.

> I expect no functionality was lost, otherwise the regression tests
> would have caught it?  Except nmea2000 is the only input source with
> no regression tests.  That needs to get fixed.
> 
> The n2k regression test are a bit different from the "normal" serial
> devices.
> 
> A testcase and a checkfile is already in the repository
> "gpsd/test/nmea2000/"

News to me.  I learn something about gpsd every day.
I do not see that being run with "scons check".

> And i have posted, how to use this files.

Clearly I missed it, and I find not README or similar.  How do I run them?
When I figure out how to run them, I'll add them to "scons check".


> > I have no experience with the ISO bus, but i assume, that changing
> > the devices on the bus will happen here more frequently.  
> 
> So we are talking about two different things?  NMEA2000 and ISO XXX?
> 
> With ISO, i mean ISO 11783. In the n2k standard, there are references
> to this standard. As much as i know, ISO 11783 inherits n2k in a way.

Hmm: ISO 11783-1:2017 Tractors and machinery for agriculture and forestry 

That does not sound right.  Issue #176 is about cars.  Same thing?

And, another expensive, proprietary standard...

From what I can see n2k begat SAE J1939 which begat ISO 11783.  So
maybe ISO is one standard further thaan we need to go.

> As i have been aware of this problem, I have implemented the n2k
> driver  in gpsd as a "passive" device, that do not need to send
> information on the n2k bus. A passive do device not need  a vendor
> and device id.

I believe the suggested patch in #176 keeps gpsd a passive driver.

I don't understand the patch, but it appears to just be more agressive
about getting, and saving source_prio, daddr, source_pgn and source_unit.

Can you look at the patch and comment on it specifically?

> Refer as as "commanded address".

And I see no reference to that in the patch...

> As i am not a farmer, i do not access do ISO 11873 devices, i can not
> provide log files.

The application in question is SAE J1939 and cars.  And the patch looks
to have negative impact on the use case that you can test.

> But i have many real world log files from n2k bus system, and even a
> test bench, to generate specific ones.

Anything that should be added to gpsd/test/nmea2000?  Other sensors?
Other pgns

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
        gem@rellim.com  Tel:+1 541 382 8588

            Veritas liberabit vos. -- Quid est veritas?
    "If you can't measure it, you can't improve it." - Lord Kelvin

pgpXcBGcuynq2.pgp
Description: OpenPGP digital signature