On Wed, 27 Mar 2013 07:06:58 -0500, Tom Marchant wrote:
>
>>When long PARMS have been discussed here previously, the reactionaries
>>and IBM loyalists have objected, "Oh, no!  We can't have that!  Suddenly
>>programs that clearly failed on JCL errors will instead program check
>>on buffer overruns and our help desk bandwidth will be strained."
>
>Really?  I don't remember anyone saying that.  What I remember seeing is 
>that any program that accepts a PARM should be coded so that it can be 
>called from a program that might pass a longer PARM. 
> 
Non-zero, but not as many as I thought I remembered.  Searching the
archives for May, 2005, I find, e.g.:

On Fri, 13 May 2005 15:56:50 -0400, Arthur T. wrote:
>    ...
>      There *must* be a method to make sure the system does
>not pass more than 100 bytes to a program which isn't
>prepared to receive them.  Merely expanding PARM= or
>changing to ZPARM= does not provide that protection.
>    ...

And Peter Relson mentioned it as a concern when he originated
the thread.

It's interesting how many features coming in z/OS 2.1 were
mentioned in that thread:

o Long PARMs (of course).

o A load module attribute indicating a program's willingness
  to accept long PARMs.

o PARM generated from a data set (e.g. PARMDD).

o Instream data sets in PROCs

o Symbol substitution in instream data sets.

o System symbol substitution in batch JCL.

-- gil

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

Reply via email to