Jon's right. The tools should behave rationally and consistently even if the usefulness is not immediately obvious.
We see it in listserve posts. Someone will say "I coded blah blah" and it didn't work and someone will respond "well it's stupid, why would you code that?" (or words to that effect) and it turns out that there is a good reason, it was just complex and would have complicated the original question. I remember coding this kind of thing as an experiment when I was learning the architecture. I recall that not only did it assemble, I saw that the assembler generated two RLDs and the link editor and loader handled it correctly: when the program was loaded ADDR2 contained the address of DWD x 2. Charles -----Original Message----- From: IBM Mainframe Assembler List <ASSEMBLER-LIST@LISTSERV.UGA.EDU> On Behalf Of Jon Perryman Sent: Wednesday, September 3, 2025 11:08 AM To: ASSEMBLER-LIST@LISTSERV.UGA.EDU Subject: Re: Question about HLASM complex relocatable expressions On Wed, 3 Sep 2025 12:34:05 -0400, John Ganci <jgan...@verizon.net> wrote: >Why would anyone use the definition of ADDR2 shown below? > >MYCSECT CSECT >DWD DS D >DWD2 DS D >ADDR2 DC A(DWD+DWD) Complex relocatable expression Consider this from a different perspective. Is there a time when this becomes useful? This specific use case does not seem useful. However, variations on this might be useful. E.g. A(DWD+DWD-MYCSECT). DWD+DWD is resolvable so why make an exception because we don't see a use case.
