>I assembled: >* test V constants > EXTRN XXX >XXX DC V(XXX+4) > END
By adding "EXTRN" the test has been corrupted. Once EXTRN is added, the rules for V apparently match the rules for A. But without EXTRN, DC V(XXX+4) is rejected. I doubt there's any case where you'd have a functional need to do a V-Con into your own csect. An A-con without EXTRN is far preferable. >For A constants, the linker adds the value assembled >into the constant (the offset). Actually, no it doesn't, to my thinking. The linker places the constant into the "word" and identifies that that word is to be relocated for the particular external reference. It is the loader that does the "add". 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
