We have had a lot of success with COBOL 5 but have had a few "edge cases" where 
we've seen issues.

Since you mention OPT(1), one thing we've seen with COBOL 5 is that the calling 
and called program parameter lists must be identical. Check and see if there is 
a coding mistake where your calling program specifies for example the 05 level 
of a data structure and the called program specifies the 01 level.  We z/OS 
guys understand that it is just an address pointer, but the COBOL compiler will 
silently discard statements that are known to be a constant value since they 
cannot change.

01 Field-01.
  05 sub-field-02 pic x(8) value "aaaa".
  05 sub-field-03 pic x(8) value " bbbb".

If the calling program passes "sub-field-02" to the called program instead of 
"field-01", the compiler will assume the calling program cannot change 
"sub-field-03" and drop for example IF statements that refer to "sub-field-03" 
since the compiler knows what the value must be.

Valid COBOL programs work well with COBOL 5 but edge cases can be tricky to 
debug. Another option is to try OPT(0).

Brian

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

Reply via email to