RE: iSCSI: Decimal encoding - why 64 bits ?

[Martins Krikis <mkrikis@yahoo.com>]

--- Black_David@emc.com wrote:

> Replying to a couple of messages on this topic.
> 
> --- Use of decimal for binary items
> 
> > There was NEVER a discussion about forbidding
> decimal for binary items.

There may not have been a big discussion about it,
since everybody was concentrating on disallowing
base64 for numbers (yes, all of them; IMO most
people seemed to think that binary strings would
suffice and that large-numerical-values aren't
necessary) or on limiting the size numerical values
encoded in decimal. But it was certainly mentioned
multiple times under various subject lines:

http://www.pdl.cmu.edu/mailinglists/ips/mail/msg09780.html

http://www.pdl.cmu.edu/mailinglists/ips/mail/msg10125.html

http://www.pdl.cmu.edu/mailinglists/ips/mail/msg10153.html

http://www.pdl.cmu.edu/mailinglists/ips/mail/msg10250.html

> > It would be counterproductive to forbid them
> > as they are so widely used in programming.

> The request to forbid them seems to have evolved
> into a request to

I'd like to disagree about the programming part,
but more importantly, why don't we reinstate 
the original request to forbid using decimal for
binary items? It is not a very useful way
of representing even short binary items because
there is no way to express the number of leading
0-bytes, for example. In case there is a vote on
this, I'm against using decimal for binary items.

> --- 64 vs. 32 bits
>
> I went back to the mailing list - and there was a 
> clear consensus to keep
> it to 64 bits (my original proposal was unlimited
> and  I suggested rhen 128
> to alleviate concerns about conversion difficulty
> for unlimited numbers).
>
> That matches what I recall from mailing list
> discussion, 32 bits was not
> considered at the time.

Getting it down from unlimited was way more
important than getting it down to 32, but it
certainly was considered:

http://www.pdl.cmu.edu/mailinglists/ips/mail/msg09804.html

http://www.pdl.cmu.edu/mailinglists/ips/mail/msg09808.html

http://www.pdl.cmu.edu/mailinglists/ips/mail/msg09810.html


> > The issue was raised without checking the
> > libraries:
>
> [... snip ...]
>
> > I don't know about Windmills - but I assume 
> > that most modern development
> > environments are supporting 64bit integers.

Some of us are working in kernel-space, and there
aren't many reasons for a kernel running on a 32-bit
architecture to provide 64-bit integer conversion
routines.

I am not saying that I don't know how to write
such a routine, just that there are normal reasons
why some people may prefer 32 to 64. In case there
is a vote, I'm for 32.

Martins Krikis, Intel Corp.

Disclaimer: these opinions are mine and may not
            be those of my employer.



__________________________________________________
Do You Yahoo!?
Sign up for SBC Yahoo! Dial - First Month Free
http://sbc.yahoo.com