Re: is C necessarily faster than C++

Urs Hoelzle <urs@engineering.ucsb.edu> writes:


> Actually, I don't buy that - I believe the proper conclusion from the
> paper is "current C++ compilers don't optimize C++ programs as well as
> current C compilers optimize C programs". In other words, much of the
> difference is in the compilers and not the languages.


C and C++ are going to have performance differences because the
languages use differing constructs (e.g. the high usage of indirect
calls in C++). Though, I do agree that with the proper optimizations
the performance differences will diminish. This is one of the points
we had in our paper, that C++ programs require different compiler
optimizations than C in order to bring its performance closer to C.
Our paper tried to stress that the performance of C++ can be improved
via compiler optimizations and certain optimizations will have a
greater benefit for C++ than C. A new version of the paper can be
found in:


http://www.cs.colorado.edu/homes/calder/public_html/


in the file called JplVersion.ps


This is the version of the paper that will appear this month (I believe)
in the Journal of Programming Languages.




Brad Calder
University of Colorado
--

Return to the comp.compilers page.
Search the comp.compilers archives again.