home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!spool.mu.edu!darwin.sura.net!paladin.american.edu!auvm!UCLAMVS.BITNET!CSYSMAS
- Message-ID: <IBM-MAIN%92123009324574@RICEVM1.RICE.EDU>
- Newsgroups: bit.listserv.ibm-main
- Date: Wed, 30 Dec 1992 07:30:00 PST
- Sender: IBM Mainframe Discussion list <IBM-MAIN@RICEVM1.BITNET>
- From: Michael Stein <CSYSMAS@UCLAMVS.BITNET>
- Subject: Re: Is PL/S a dead language?
- Lines: 79
-
- > In article <IBM-MAIN%92122814293609@RICEVM1.RICE.EDU> Michael Stein
- > <CSYSMAS@UCLAMVS.BITNET> writes:
- > >Ha! Exactly backwards. PL/S allows writing a structure which
- > >you *know* the mapping of, while C doesn't (the C mapping depends
- >
- > Good point. But PL/S only runs on one machine, so there are no
- > endianness or word-size issues.
-
- That's a claim that PL/S is consistant with itself (since
- only one implementation exists).
-
- I claim the requirement on PL/S is more than that -- PL/S
- structures *have* to exactly match assembler DSECTs or else the
- operating system won't work. So it's an external mapping which
- PL/S must support, not just self consistancy. (You couldn't
- change the PL/S structure mapping and just recompile everything
- and expect the system to work...)
-
- > It would be possible to say that _for any given CPU_ the layout
- > of C structures would be well-defined, but the C standard
- > cannot do this.
-
- IMHO C should support an additional attribute of a structure
- which allows *absolute* structure layouts (hopefully with only
- forwards bytes?) where *all* implemetations layout the structure
- the same. This allows exportable binary files as well as binary
- network communication.
-
- "C is Portable? Ha!, It can't even talk to itself on different
- machines..."
-
- > In fact (FLAME ON) IBM should have way back said what C
- > compilers on MVS should do, how they should handle the
- > record-oriented file system, what the ASCII-to-EBCDIC
- > translation routines should be called, then we wouldn't have to
- > put up with oddball compilers.
-
- C still wouldn't be compatable with itself across a network...
-
- > Your point about SRBs etc. is a good one. But under an SRB it's
- > important to have highly optimised code.
-
- Ouch. I dislike the term "optimised code". Most compiler
- optimization techinques are designed to undo the pessimizing (?)
- introduced by the language straitjacket or compiler code
- generation.
-
- As someone said "keep it as simple as possible, but no simpler".
-
- > I use MVS TCP/IP a lot and I think many of its problems (not
- > just performance) are due to vast amounts of bloated PL/S in
- > the system (IRB, SRB, PC-call etc.) code.
-
- In the MVS TCP/IP case, I think it's just "bloat", the language
- doesn't matter (except one of us users might fix some of it if it
- wasn't in PL/S or if we had a PL/S compiler).
-
- MVS TCP/IP is not even close to being "as simple as possible".
-
- > Surely not all the assembler wizards at IBM are working on
- > VM/CMS?
-
- They may, by and large, have been raised on VM. Does anyone know
- how IBM assembly programmers are trained? (IBM seems to use VM
- internally too much to raise good MVS programmers..)
-
- In my experience, there are lots of good IBM 370/390 programmers
- around who do NOT understand the requirements of true
- multi-tasking and multi-processor coding. Not surprising, this
- includes (can't be all?) VM systems programmers as well as most
- MVS systems programmers.
-
- A flurry of CS/CDS is one of the later signs of a belated
- admission that everything doesn't happen sequentially.
-
- There are also people who have managed to get hold of a PL/S
- compiler and pretend to be assembler programmers. The programs
- of these people demonstrate that they don't know what the machine
- instructions really do...
-