I was wrong about the
BAR DS 0D
...
END BAR
case. I had glossed over what "END BAR" meant.
I do agree with Charles and Gil from earlier:
If you have "END BAR" then the normal entry point for the module will
locate BAR, not offset 0
If you have "ALIAS BAR" then the alias will locate BAR, not offset 0.
Then, here's what I played with (my load module was FOO with alias BAR),
copying PDS to PDS:
-- ISPF copy of FOO: did only FOO (I thought ISPF COPY by default did
aliases; guess not)
-- ISPF copy of FOO and BAR: did both, but BAR was standalone, not an
alias of FOO
-- IEBCOPY with "COPY" and SELECT MEMBER=FOO: did just FOO, preserved the
EP at BAR
-- IEBCOPY with "COPY" and SELECT MEMBER=FOO and SELECT MEMBER=BAR.
Did both, BAR is an alias, preserved the EP of FOO at BAR, preserved
the EP of BAR
-- IEBCOPY with "COPYGRP" and SELECT MEMBER=FOO: did just FOO, preserved
the EP at BAR
-- IEBCOPY with "COPYGRP" and SELECT MEMBER=FOO and SELECT MEMBER=BAR.
Did both, BAR is an alias, preserved the EP of FOO at BAR, preserved
the EP of BAR
-- IEBCOPY with "COPYGROUP" and SELECT MEMBER=FOO:
Did both, BAR is an alias, preserved the EP of FOO at BAR, preserved
the EP of BAR
-- IEBCOPY with "COPYGROUP" and SELECT MEMBER=FOO and SELECT MEMBER=BAR.
Did both, BAR is an alias, preserved the EP of FOO at BAR, preserved
the EP of BAR
So I did get the described difference between COPYGRP and COPYGROUP but
not the loss of the entry point offset that Charles described.
Peter Relson
z/OS Core Technology Design
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
