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

[Luben Tuikov <ltuikov@yahoo.com>]

--- Paul Koning <ni1d@arrl.net> wrote:
> 
> I agree that the current description is not consistent
> and is
> confusing.  That's why I proposed replacement text.  I'd
> be interested
> in comments on that text.

I gave you a comment in my last letter:

  Please also note that the Ethernet spec CRC algorithm,
  is an optimized version of a basic
  prefix-postfix-initialize-the-
  CRC-register-to-some-constant algoritm. BUT as SMD it
  operates on NON-AUGMENTED messages. FOR THAT REASON
  you cannot say that we have to multiply M(x) by x^32!!!

You see, you cannot say that the message has to be
multiplied by x^32, if you are describing the Ethernet
CRC algo.

>  Luben> Please also note that the Ethernet spec CRC
> algorithm, is an
>  Luben> optimized version of a basic
> prefix-postfix-initialize-the-
>  Luben> CRC-register-to-some-constant algoritm.
> 
> I'm not sure what spec you're talking about.

The one you always mention, the one you quote in you
messages. The one iSCSI is trying to present.

> The algorithm in the Ethernet spec (section 6.2.4) which
> was copied
> into the 802.3 spec (section 3.2.8) pretty much verbatim
> except for a
> typo, is not a prefix-postfix-whatever algorithm at all. 

Yes, it is. I'll send you the math when I finish it (later
this week).

I'll also send you 2 algorithms, one simplest division, and
another as per Ethernet, which produce identical results.
(I can send it now, but no time.)

> Instead, it
> is a mathematical definition of the CRC value.  The
> phrase "initialize
> the CRC register" does not occur at all in any form.

So it doesn't in the definition I gave you.
If it is a formal, mathematical definition, then you should
see the same initial constant as I found: 0x2a26f826.

If it mentiones 0xFFFFFFFF, then it means that SMD
(Simultaneous Divide and Multiply) algo is described, which
is hardly a formal mathematical definition.
 
> If I understand right, this is a mathematically
> equivalent way of
> describing an Ethernet-style CRC.

Yep. Mathematically, and when implemented, by hand, by a
program, by a handheld calculator, by a mathematics
analysis program, it will yield identical results as in
``Ethernet-style CRC''.
 
> Fine, no problem.  But what's the point?  All we need is
> a single
> formal definition.

The point? You and Pat wanted a mathematical definition, so
here it is!

This is a ``single formal definition''.

>  And it would make the most sense for
> that
> definition to look like the ones found in other
> standards, so it will
> be clear to the reader that the CRC in iSCSI is the same
> as the one
> used in many other places except for the G(x).

Then lets forget about trying to REPHRASE Ethernet and just
mention that: that it is equivalent to Ethernet, but just
that it uses a different polynomial.

> The
> description you gave may be mathematically equivalent,
> but that is not
> at all obvious unless you're fluent in abstract math.

True.
 
> In any case, no mathematical description by itself
> translates easily
> to an implementation (again, unless you're fluent in
> abstract math).

True.

> Most specs disregard that concern and leave it up to the
> implementer
> to search the literature.  My current proposal is to do
> likewise for
> iSCSI.  The only exception I've seen is the Ethernet
> spec, which gives
> a single example implementation in its appendix.  

Example implementation is good, as are examples.
I'd vote for both to be included in iSCSI.

-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