David,
Seems awful slow box you are on. Is this a z Box?
Why have a DO UNTIL when "EXECIO 5000 DISKR INPUT ( STEM REC. FINIS"
is all that's needed?
Your REXX code was 1-2 elapsed seconds on my box.
No need for the second EXECIO....
Cowlishaw BENCH:
----- REXXCPS 2.1 -- MEASURING REXX CLAUSES/SECOND -----
REXX VERSION IS: REXX370 3.48 01 May 1992
SYSTEM IS: TSO
AVERAGING: 5 MEASURES OF 3 ITERATIONS
PERFORMANCE: 21600 REXX CLAUSES PER SECOND
***
On Mon, Jul 1, 2013 at 2:51 PM, David Crayford <[email protected]> wrote:
> On 1/07/2013 1:25 AM, Thomas Conley wrote:
>>
>> On 6/30/2013 12:17 PM, Shmuel Metz , Seymour J. wrote:
>>>
>>> In
>>> <of28a24a26.61fdf4d9-on80257b99.0039d7d3-80257b99.003a9...@uk.ibm.com>,
>>> on 06/29/2013
>>> at 11:39 AM, Martin Packer <[email protected]> said:
>>>
>>>> In the Munich System z Technical University conference 2 weeks ago
>>>> Marna Walle mentioned that in z/OS 2.1 EXECIO is previewed to
>>>> perform I/O to VBS (as in "Spanned" :-) ) data sets.
>>>
>>>
>>> Have they added the language enhancements in OOREXX, or at least those
>>> in the ANSI standard?
>>>
>>
>> Shmuel,
>>
>> Unfortunately, any changes to the Rexx language standard that would also
>> require changes to the compiler preclude themselves. The compiler is in the
>> Dead Pool at IBM, functionally stabilized with no budget to enhance it.
>> I've tried and failed numerous times to get IBM to bring Rexx up to ANSI
>> standards. The enhancements you see to Rexx are all done through external
>> functions, like EXECIO. Those enhancements require no compiler updates, so
>> for now, that is all you will see as far as Rexx enhancements go.
>>
>
>
> Then the REXX compiler is more trouble than it's worth. I ran some simple
> EXECIO read I/O benchmarks and compiled REXX was the slowest of the lot
> http://users.tpg.com.au/crayford/rexx-lua-c-io-benchmark.htm.
>
>
>
>
>> Regards,
>> Tom Conley
>>
>> ----------------------------------------------------------------------
>> For IBM-MAIN subscribe / signoff / archive access instructions,
>> send email to [email protected] with the message: INFO IBM-MAIN
>
>
> ----------------------------------------------------------------------
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to [email protected] with the message: INFO IBM-MAIN
--
Wayne V. Bickerdike
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
