Path: utzoo!dptcdc!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!ucbvax!pasteur!ames!sgi!arisia!quintus!ok
From: ok@quintus.UUCP (Richard A. O'Keefe)
Newsgroups: comp.arch
Subject: Re: Complex Instructions
Message-ID: <1011@quintus.UUCP>
Date: 18 Apr 89 04:36:45 GMT
References: <807@microsoft.UUCP> <92634@sun.uucp> <13322@steinmetz.ge.com> <573@loligo.cc.fsu.edu> <5064@hubcap.clemson.edu> <3169@bd.sei.cmu.edu> <38629@bbn.COM> <17167@cup.portal.com> <1003@quintus.UUCP> <17296@cup.portal.com>
Reply-To: ok@quintus.UUCP (Richard A. O'Keefe)
Organization: Quintus Computer Systems, Inc.
Lines: 34

In article <17296@cup.portal.com> bcase@cup.portal.com (Brian bcase Case) writes:
>>>And in the case of the call instruction, I find the VAX complex instruction
>>>sorta insulting:  assuming that compiler writers are too stupid to adhere to
>>>a common spec for inter-language call protocol.
>>
>>Try calling Pascal from PL/I under MVS sometime.
>
>I don't want this to get out of hand, and it might belong elsewhere anyway.
>
>BUT, was calling Pascal from PL/I under MVS a design goal?  If so, then why
>doesn't the existence of the VAX's CALLS instruction guarantee this goal
>will be met?  I guess this is yet another point:  if the CALLS instruction
>was included so that inter-language calls "would just work," then it fails
>even at that!

I am *sick* of people not reading what I wrote.  Watch carefully now.
EM.  VEE.  ESS.  I did not say "try calling Pascal from PL/I under <VMS>
sometime."  To the best of my knowledge that works perfectly well and has
done ever since it was possible to try.  And yes, the CALLS/CALLG
instructions and VIA were important in achieving that.

My point was that IBM have a software convention for System/370, and that
their Pascal compiler (which is not a fully supported product, but it is
the compiler they will provide to customers who want Pascal) was written
after that convention was well established and described in hundreds of
books, and it _does_ _not_ follow that convention.  My point, then, is in
DEC's defence:  yes there _are_ compiler writers who will not adhere to a
common spec for inter-language call protocol.  That's not to say that they
are stupid.  They had to trade off incompatibility with other languages
against inefficiency for Pascal.  DEC tried to make that a non-issue.

On the subject of calling conventions, I just had my first look at a
68030 manual.  I know, I'm way behind the times, but it was supposed to
be just more of the same.  Why did Motorola delete CALLM?