home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!ogicse!das-news.harvard.edu!spdcc!iecc!compilers-sender
- From: byron@netapp.com (Byron Rakitzis)
- Newsgroups: comp.compilers
- Subject: Re: Is this a new idea?
- Keywords: parse, performance
- Message-ID: <92-11-017@comp.compilers>
- Date: 4 Nov 92 08:57:30 GMT
- Article-I.D.: comp.92-11-017
- References: <92-10-113@comp.compilers>
- Sender: compilers-sender@iecc.cambridge.ma.us
- Reply-To: Byron Rakitzis <byron@netapp.com>
- Organization: Netcom - Online Communication Services (408 241-9760 guest)
- Lines: 18
- Approved: compilers@iecc.cambridge.ma.us
-
- >[Scanning and parsing can be as much as half of the total time that a
- >compiler takes (so says Ken Thompson of his Plan 9 C compiler) and that is
- >quite amenable to incremental precalculation. -John]
-
- Well, that's Ken's compiler, which doesn't do hairy optimizations, leaves
- a lot of work for the loader, and has a fast in-core implementation of
- tree passes to boot. ("Turbo C for Unix")
-
- I think the %-age for gcc (and I am assuming that most commercial
- multi-pass, aggressively optimizing compilers show similar performance) is
- considerably lower, around 20%. I am not sure of the figure, it is from
- memory from around the time that gcc2 was being released. This would have
- been with the optimizer turned on, of course.
-
- Sorry I can't be more specific right now.
- --
- Send compilers articles to compilers@iecc.cambridge.ma.us or
- {ima | spdcc | world}!iecc!compilers. Meta-mail to compilers-request.
-