home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.lang.scheme:2854 comp.lang.c:19150
- Path: sparky!uunet!spool.mu.edu!yale.edu!yale!mintaka.lcs.mit.edu!ai-lab!zurich.ai.mit.edu!gjr
- From: gjr@zurich.ai.mit.edu (Guillermo J. Rozas)
- Newsgroups: comp.lang.scheme,comp.lang.c
- Subject: Re: applying or
- Date: 3 Jan 93 08:25:03
- Organization: M.I.T. Artificial Intelligence Lab.
- Lines: 12
- Message-ID: <GJR.93Jan3082503@chamarti.ai.mit.edu>
- References: <1993Jan1.184655.7023@viewlogic.com> <GJR.93Jan2080947@chamarti.ai.mit.edu>
- Reply-To: gjr@zurich.ai.mit.edu
- NNTP-Posting-Host: chamartin.ai.mit.edu
- In-reply-to: gjr@zurich.ai.mit.edu's message of 2 Jan 93 08:09:47
-
- In article <GJR.93Jan2080947@chamarti.ai.mit.edu> gjr@zurich.ai.mit.edu (Guillermo J. Rozas) writes:
-
- | The issue here is whether the order of execution of expressions can be
- | changed by the compiler with respect to the one specified by the
- | semantics. In general compilers are allowed to do this if they can
- | prove that the effect is the same, in other words, that the user can't
- | tell what order things were done in.
-
- I meant to say that the compiler can reorder as long as the user
- cannot tell from the behavior of the resulting code that the
- expressions were not done in the order that the semantics specifies.
-
-