My eyes! One of my colleagues at the Technion was called up from the reserves and the drafted me to teach his class. I told the students that I would take points of for comments that didn't make the code more readable.
I'm older and wiser now; I would have students read each others code and describe what they didn't understand, then figure out whether the problem was the code, the reader or my teaching. Shift? These days I would use MHI reg,8; I doubt that it is a performance issue. If the PL/X source is well documented, then I won't quibble if the compiler makes a dog's breakfast out of it. -- Shmuel (Seymour J.) Metz http://mason.gmu.edu/~smetz3 עַם יִשְׂרָאֵל חַי נֵ֣צַח יִשְׂרָאֵ֔ל לֹ֥א יְשַׁקֵּ֖ר ________________________________________ From: IBM Mainframe Assembler List <ASSEMBLER-LIST@LISTSERV.UGA.EDU> on behalf of Paul Gilmartin <00000014e0e4a59b-dmarc-requ...@listserv.uga.edu> Sent: Friday, July 11, 2025 11:46 AM To: ASSEMBLER-LIST@LISTSERV.UGA.EDU <ASSEMBLER-LIST@LISTSERV.UGA.EDU> Subject: Re: Food for thought. External Message: Use Caution On 7/10/25 14:25, Seymour J Metz wrote: > That still doesn't say that it requires an EQU, just that there is a use case > for which AHI with an equated symbol is appropriate. The argument about being > able to test the cc before an EX or EXRL is more compelling. > ...My manager once urged that: o We EQU all self-defining terms for doc and XREF. o we use, e.g., SLA 3 to optimize multiplication by 8. He was perplexed when I asked what name and comment to code in: ???????? EQU 3 ???????? I once saw in IBM (PL/X generated) assembler code: FIFTYSIX EQU 56 -- gil
