[Guy Harris <gharris@xxxxxxxxxxxx>]

> attached the promised dissector for the Sinec H1 protocol (aka
> Siemens Industrial Ethernet) for communication with Siemens S5 PLC's
> over Ethernet.

Checked in.

I applied your subsequent "packet-osi.c" patch, to make it dissect COTP
data as the H1 protocol if the inactive subset is being used (although I
made the "inactive subset is being used" flag an argument to a new
"dissect_cotp_internal()" routine, rather than a global;
"dissect_cotp()", which is called only from the X.25 dissector, passes
FALSE).

If anything else uses COTP atop the inactive subset of CLNP, a different
scheme might have to be used.  (Presumably the inactive subset is used
when the packet doesn't need to be routed by CLNP and the receiving site
knows implicitly what protocol is to be used atop CLNP.)

> Btw.: The communication standard for the new S7 PLC's is not documented
> by Siemens but we have made some investigations and found some details.
> What is the official policy on reverse engineering of protocols?

I don't know if there's an official policy, but I don't think we have
any problem with protocol dissectors developed by doing reverse
engineering.