home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!cs.utexas.edu!qt.cs.utexas.edu!yale.edu!spool.mu.edu!uwm.edu!ogicse!das-news.harvard.edu!spdcc!iecc!compilers-sender
- From: idacrd!desj@uunet.UU.NET (David desJardins)
- Newsgroups: comp.compilers
- Subject: Re: static estimation of conditional branches?
- Keywords: optimize, performance
- Message-ID: <92-12-058@comp.compilers>
- Date: 14 Dec 92 04:36:24 GMT
- Article-I.D.: comp.92-12-058
- References: <92-12-029@comp.compilers> <92-12-054@comp.compilers>
- Sender: compilers-sender@iecc.cambridge.ma.us
- Reply-To: idacrd!desj@uunet.UU.NET (David desJardins)
- Organization: IDA Center for Communications Research, Princeton
- Lines: 16
- Approved: compilers@iecc.cambridge.ma.us
-
- Henry Spencer <henry@zoo.toronto.edu> writes:
- > When people first started implementing profilers and the like, they found
- > one striking result: human intuition about the locations of the hot spots
- > in a complex program was *consistently wrong*. ...
-
- We were talking about branch prediction. It is *much* easier for a
- programmer to say how often a particular conditional branch is going to be
- taken than it is for the programmer to say how much time is going to be
- spent in a particular region of code. Among several reasons for this is
- that the former is independent of the machine, compiler, or optimizer,
- while the latter is not.
-
- David desJardins
- --
- Send compilers articles to compilers@iecc.cambridge.ma.us or
- {ima | spdcc | world}!iecc!compilers. Meta-mail to compilers-request.
-