home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!darwin.sura.net!mips!swrinde!elroy.jpl.nasa.gov!nntp-server.caltech.edu!brooke
- From: brooke@cco.caltech.edu (Brooke Paul Anderson)
- Newsgroups: comp.os.os2.programmer
- Subject: Weird gcc effect when compiling a benchmark prog.
- Message-ID: <1992Jul21.110954.4347@cco.caltech.edu>
- Date: 21 Jul 92 11:09:54 GMT
- Sender: news@cco.caltech.edu
- Organization: California Institute of Technology, Pasadena
- Lines: 8
- Nntp-Posting-Host: kanga
-
- I was compiling a simple floating-point benchmark program, and I
- noticed that it gave me different results depending on whether or
- not I had a printf statement in the code. The odd thing is that the
- printf statement was entirely out of the timing loop, so it shouldn't
- have any effect at all on the performance measure spit out by the
- program. Nevertheless, it has an effect. Has anyone ever seen something
- like this? I can supply the code if people would like, so they can
- test it -- the program is about 100 lines long.
-