[IP-SFS] Re: Changes to IP over Semaphore Signals for April 1

[forum::für::umläute]

hallo

das ist jetzt nur an diese liste, und nicht and bob und die rfc-editoren:
(habe jetzt auch keine zeit, dass ins svn einzuschecken)


mfga.sdr
IOhannes


Bob Braden wrote:
> How can SUN be one or more?  I don't understand how you keep SFS (flag)
> synchronization
> after an error.


in "4.5.  State Transmitting" eine erweiterte fehlerbehandlung:

If an error occurs by transmitting a wrong Data SFS, the interface MAY
invalidate the last Data SFS by transmitting SUN followed by the correct
signal. A series of incorrectly transmitted Data SFS MAY be invalidated
by sending SUN for each invalid SFS.
Control SFS cannot be invalidated.
If an error occurs, the interface MAY also transmit FUN and restart
transmission of the active IP-SFS frame.
Whether the interfaces choses SUN or FUN for error correction is up to
the interface, but it is suggested to use SUN for single invalid SFS,
and FUN whenever the interface failed to transmit several SFS in a row.

in "4.6.  State Receiving"

[...]

If the interface receives SUN from the link partner, it MUST discard the
last received data SFS (if any). If several SUNs are received in a row,
then the last n data SFS MUST be discarded unless there are no more data
SFS in the frame. If there are no more data SFS in the frame to be
discarded, the SUN signal MUST be ignored by the interface.

If the interface receives FUN from the link partner, it is free to
discard the frame received thus far.  We suggest honoring FUN since
discarding the signal will decrease bandwidth.