@Scott, there are two ways to compile Rexx: in one you end up with something that behaves like "normal" Rexx, but performs a little better in you have the library licensed* on the target machine. I am not familiar with that beyond the sentence I just wrote. The other way produces object code that you then linkedit. I am fairly familiar with that latter way. For that, the resulting load module is just like any other load module -- you would call/link to it just like you would if it were written in assembler. IIRC in that situation the parm passing is also just like for an assembler program. The first parm pointed to by R1 becomes Arg(1) and so forth.
*For a vendor, compiled Rexx is less of a benefit than you might imagine. You will gain performance improvements only at customer sites that license the Rexx compiler run-time library. At other sites your compiled Rexx program will run, but it will run interpreted with "normal" performance. You also gain only very slight "source code obfuscation" benefits because all of your source code is sitting in plain EBCDIC inside the resulting load module. It is not perfectly editable source, but it would not take a hacking genius to turn it back into "real" source code. Charles -----Original Message----- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Steve Thompson Sent: Friday, November 25, 2016 2:44 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Compiled Rexx Question Sorry, I read your initial post backwards -- REXX calling COBOL. So far I haven't called REXX from COBOL. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
