home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.compilers
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!eff!world!iecc!compilers-sender
- From: iwm@doc.ic.ac.uk (Ian W Moor)
- Subject: Re: static estimation of conditional branches?
- Reply-To: iwm@doc.ic.ac.uk (Ian W Moor)
- Organization: Dept. of Computing, Imperial College, University of London, UK.
- Date: Mon, 14 Dec 1992 13:43:04 GMT
- Approved: compilers@iecc.cambridge.ma.us
- Message-ID: <92-12-061@comp.compilers>
- Keywords: optimize, performance
- References: <92-12-029@comp.compilers> <92-12-058@comp.compilers>
- Sender: compilers-sender@iecc.cambridge.ma.us
- Lines: 21
-
- 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*. ...
- |>
-
- To be worthwhile, any compiler which accepts user hints like frequency or
- register has to have the following property:
-
- A correct hint must make an improvement that is worth gathering the
- information, and an incorrect hint must either be detectable or must not
- reduce performance much.
-
- This seems contradictory.. I'd rather rely on a profiler or register
- allocator.
- --
- Ian W. Moor +44 71 589 5111 x 5039
- iwm@doc.ic.ac.uk Department of Computing, Imperial College.
- --
- Send compilers articles to compilers@iecc.cambridge.ma.us or
- {ima | spdcc | world}!iecc!compilers. Meta-mail to compilers-request.
-