RE: effect of initializing CRC reg to 1's depends on implementati on? iSCSI

["THALER,PAT (A-Roseville,ex1)" <pat_thaler@agilent.com>]

Luben,

Your statement "That is, it has to have the correct CRC _after_ _each_ _bit_
has arrived." is incorrect for Ethernet. If the frame was not an integral
number of bytes, then the excess bits (the ones that are not part of a
complete byte) are not to be used in the CRC calculation. The IEEE 802.3
spec (and the
earlier DIX Ethernet spec) requires that the CRC checking algorithm only
process whole bytes. Therefore, a good CRC algorithm for Ethernet would only
process bits once a whole byte worth of bits had arrived.

This really isn't relevant to the iSCSI reflector so I suggest we take this
offline if you want to discuss it further. 

IEEE 802.3 is about 1500 pages in the 2000 edition so if I am making a point
with respect to it, I think it is polite to provide a reference to where
that information can be found so that people can read it for themselves.
Certainly no offense was intended by making a specific citation.

Regards,
Pat Thaler

-----Original Message-----
From: Luben Tuikov [mailto:ltuikov@yahoo.com]
Sent: Tuesday, December 18, 2001 2:57 PM
To: pat_thaler@agilent.com; ni1d@arrl.net; vince_cavanna@agilent.com
Cc: ips@ece.cmu.edu
Subject: RE: effect of initializing CRC reg to 1's depends on
implementati on? iSCSI


--- pat_thaler@agilent.com wrote:
>
> method. However, that is irrelavent to what should go
> into the standard.

I never posted a message of the sort: 
``Here guys put this into the draft: ...''. And I don't
want to post one.

All I've been doing is mathematical analysis of the
algorithm(s). And all I've been looking for is an
unambiguious definition.

> By the way, there is an error in your statement below
> with respect to
> Ethernet behavior: "A good CRC algorithm will have to be
> as strong as being
> able to produce the CRC at any time. That is, it has to
> have the correct CRC _after_ _each_ _bit_ has arrived.
> 
> I.e. in any moment in time, if I were to say STOP, I
> would
> have to be able to ``produce'' the right CRC."

Read the sentence again please: indefinite article ``a''.
This means ``any'' CRC algorithm.

No there is no error.

> Actually and Ethernet implementation is specifically
> required not to behave
> this way. If you will refer to IEEE 802.3 4.2.9 procedure
> ReceiveLinkMgmt
... [cut]

Don't you think I can read this myself?

Then again: indefinite article ``a''.

-l



=====
--

__________________________________________________
Do You Yahoo!?
Check out Yahoo! Shopping and Yahoo! Auctions for all of
your unique holiday gifts! Buy at http://shopping.yahoo.com
or bid at http://auctions.yahoo.com