plc4j(ab-eth): Fix read tag after refactoring/spi3

[andvasp]

Hi @chrisdutz!

When trying to read tag values, the system is getting a timeout error. The problem occurs because the calculateTotalMessageSize is getting a big value. As a result, the driver keeps waiting for incoming data that will never arrive, since the actual data transfer has already finished.

Note: I considered whether this behavior should be configured. However, since the rest of the reading mechanism works fine without extra configuration, a direct fix in the size calculation seems to be the correct approach.

Before refactoring/spi3 it is working properly. I tried to find similar logic in the old version but It looks like it does not exist. So I believe the old mecanism approach considers that the received data will arrive in just one shot, right?

Thanks,
Anderson.

[chrisdutz]

Looking at your change it looks as if this is a Little-Endian vs. Big-Endian thing ... However looking a bit deeper into this, it seems the EIP is generally LE ... I wonder why things worked in my case. I'll look into this.

[chrisdutz]

Just noticed that it's not the EIP driver, but the AB-ETH ... sorry for the mis-understanding ... so you state that all is working with this fix? Then I'll simply merge it as I have no other way to validate it and the code changes look sound.

[andvasp]

Yes. It is working fine. If you check in the ab-eth.mspec you will notice that it specify big-endian.

I was responsible for enabling reading of this driver through #2040, Probably I am the only one using it! :)

Thanks!

[chrisdutz]

Yeah ... might be ... would really be cool, if we knwo who's using what ...

Aaah ... I had it little endian and you made it big endian to match ... perfect .. thanks for spotting this and fixing it ;-)