Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!uwm.edu!gem.mps.ohio-state.edu!ginosko!uunet!mcsun!hp4nl!orcenl!bengsig
From: bengsig@oracle.nl (Bjorn Engsig)
Newsgroups: comp.sys.ibm.pc.rt
Subject: Re: integer alignment problems on RT
Keywords: RT 6150 032 ROMP alignment
Message-ID: <537.nlhp3@oracle.nl>
Date: 12 Oct 89 12:49:37 GMT
References: <162@eliza.edvvie.at> <2396@ibmpa.UUCP> <488@ctycal.UUCP>
Reply-To: bengsig@oracle.nl (Bjorn Engsig)
Organization: ORACLE Europe, The Netherlands
Lines: 21

Article <488@ctycal.UUCP> by ingoldsb@ctycal.UUCP (Terry Ingoldsby) says:
|
|This [using unaligned pointers] seems to me to be no less of an unusual
|event than a divide by zero
|and no more code dependent.  ie. it is possible to generate faulty code
|that at execution time (when it can't be detected by the compiler) causes
|an arithmetic exception.  Similarly addresses can be calculated incorrectly.
No, these two are clearly different, at least in the case of C, where the
language definition (both) say:

1) Only valid pointers may be used.

2) You get SIGFPE on arithmetic errors, e.g. zero divide or overflow.

There is nothing at all, that describes the behaviour of a C program using
invalid pointers.  In my opinion, the way they are handled on the ROMP
processor is not worse than the handling on other processors.

It's just a matter of writing truly portable code :-)
-- 
Bjorn Engsig, bengsig@oracle.nl, bengsig@oracle.com, mcvax!orcenl!bengsig