[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [lwip-users] lwip + profinet ... possible?
From: |
Simon Goldschmidt |
Subject: |
Re: [lwip-users] lwip + profinet ... possible? |
Date: |
Thu, 20 Feb 2014 21:39:34 +0100 |
"Claudius Zingerli" <address@hidden>
> On 2014-02-14 18:44, Pomeroy, Marty wrote:
>> Bandu
>>> Is there anyone successfully implement profinet with lwip?
That depends on what you exactly want. You can't use lwIP to implement profinet
since it is implemented
on a different level, but you can run lwIP for UDP/TCP things in parallel to
profinet. This works.
>>> (couldn't find any opensource on google related to profinet stack except
>>> expensive 3rd party stacks)
Of course not. You won't. No need in searching on.
>> From http://us.profinet.com/technology/profinet/: "Real Time (PROFINET
>> RT): Here the TCP/IP layers are bypassed in order to give deterministic
>> performance...".
>
> IMHO, there is no point in that. The same stuff is claimed by Ethercat,
> but as I understand it, any protocol - including TCP - would be
> deterministic, given there is a full duplex path, no packet loss and no
> congestion...
Exactly. But as soon as it comes to packet loss, TCP fails while UDP-based
protocols (or Profinet RT) still work. This *is* the point (keeping easier
implementation of the cyclic traffic aside if UDP/TCP/IP is not used).
Simon