On Tue, Feb 18, 2014 at 12:41 PM, John Gilmore <[email protected]> wrote:

> John [McKown]:
>
> The answer to the OP's question depends upon whether the length n
> under discussion is a length in bytes, as it is and must be in XLn, or
> a length in bits.
>

True. I assumed the length was in bits. I also assumed, due to the use of
the word "stores" and "opened or closed". that the ITMAP was, in reality,
being used as a bit string, with 8 bits in each byte, with each "bit"
representing a "store". With '1'B meaning "store open" and '0'B meaning
"store closed". I did not state, but also assumed that ITMLEN in fact
contained the number of stores in total, and so was the number of bits used
in the actual data item. I guess I should have explicitly stated this in my
original message.


>
> If the former XL2 is sufficient for either +12500 or unsigned 12500.
> If the latter a fullword will be required because 100000 is greater
> than either 32767 or 65535.
>
> Since, however, the storage is being declared/allocated in bytes, it
> seems clear to me that the length is/ought to be expressed in bytes
> too.  I do not see much rationale for your different view in this
> particular context.
>

In my view, ITMAP most likely should have been defined as a BL1250 instead
of a XL1250. But lengths in HLASM are always in bytes. The z can't allocate
a single bit in the hardware as a independently addressable value. I
haven't seen many programs which use the B (Binary). I do see a lot of A(),
F'..', H'..',XL'..', and C'..'


>
> Note also that your expression for the capacity of a signed halfword
> is slightly wrong.  Because twos-complement representations are
> used--Zero is its own twos complement--these limits are not
> symmetrical.  They are
>
> -2^15 <= H <= +2^15 - 1
> -32768 <= H <= +32767.
>

Did you ever have your fingers type the wrong word out of habit when you
meant to write a similar word? That's what happened to me with the -32_767
instead of -32_768. Muscle memory got me.


>
> John Gilmore, Ashland, MA 01721 - USA
>
> ----------------------------------------------------------------------
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to [email protected] with the message: INFO IBM-MAIN
>



-- 
Wasn't there something about a PASCAL programmer knowing the value of
everything and the Wirth of nothing?

Maranatha! <><
John McKown

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to