On Thu, Jan 22, 2015 at 4:28 AM, Sharuff Morsa3 <[email protected]> wrote:
> >Ain't progress wonderful? > > Anyone know how to stop it ? (progress that is) > > I would not rule out > 8 character mnemonics nor > 8 character HLASM > assembler directives (not that I'm currently planning any). > > Because of the very large number of mnemonics and extended mnemonics which > have been added, there are some ISPF SuperC commands to assist users in > searching their source, copybook and macro libraries to see if they may be > affected (http://www.ibm.com/support/docview.wss?uid=swg21694301). > > The new instructions have highlighted a problem for which we have to > strike a balance. Several of the new instructions have the same mnemonics > but differing instruction formats. Who can successfully execute ESA/390 > vector instructions ? But some users will have these mnemonics are coded > in their applications (anyone want to own up having some?). > > Should we always (100%) maintain the ability for users programs to > assemble programs cleanly even though they would not execute successfully? > How much can a product change (or evolve) without users having to make > some or consider those changes ? > > IBM z Systems have a very long history of minimising the affect of changes > on users - but products and their usage change over time. How customers > use our products changes over time. Is that progress ? > > Sharuff > > Sharuff Morsa IBM Hursley Labs > > One thing that I can think of which _might_ be of some help would be to have either another program, or a PARM= value for HLASM for a source validation. That is, it would act like HLASM, but would flag all opcodes which are HLASM machine opcodes and which _also_ exist as members in the SYSLIB concatenation. I don't know if HLASM does this already, but it would be nice if all machine instructions supported by HLASM, but _excluded_ by using the OPTABLE/MACHINE compile parameter, were only searched for as macros. Lastly, it might be nice to have a program which can "pseudo-compile" a source program and not only flag machine instruction opcodes which exist as members in the SYSLIB concatenation, but would also create an IEBUPDTE control deck which puts a ":MAC" on the end of the opcode. The user could then edit this and use it to more easily update their source. Just some ideas. -- While a transcendent vocabulary is laudable, one must be eternally careful so that the calculated objective of communication does not become ensconced in obscurity. In other words, eschew obfuscation. 111,111,111 x 111,111,111 = 12,345,678,987,654,321 Maranatha! <>< John McKown
