This is the mail archive of the
mailing list for the GCC project.
Re: gcc compile-time performance
- From: Gabriel Dos Reis <gdr at codesourcery dot com>
- To: dewar at gnat dot com (Robert Dewar)
- Cc: jss at ast dot cam dot ac dot uk, mark at codesourcery dot com, gcc at gcc dot gnu dot org
- Date: 18 May 2002 23:27:33 +0200
- Subject: Re: gcc compile-time performance
- Organization: CodeSourcery, LLC
- References: <20020518211930.64063F28D4@nile.gnat.com>
email@example.com (Robert Dewar) writes:
| > Remember that the whole technique involved here is to force the compiler
| > to do a variety of domain-specific optimizations that it would not
| > otherwise do. The reason your front end/back end distinction doesn't
| > work here is that the front end is doing as much optimization as the
| > back end.
| I understand this approach, but it seems far from mainstream C++ usage
| to me.
Most of STL implementations use such techniques. Boost
(http://www.boost.org) is another area where such techniques are
widespread. POOMA is another good example. MTL is. I can multiply
examples. Use of C++ has changed over these last 7-8 years.
| Are you sure that complaints about C++ compilation speed are really
| specifically with respect to this kind of paradigm?
Most of the slow down complaints about g++ are about stuff related to
templates. Any modern C++ programs that uses the standard library has
to deal with templates.