This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
[(nowhere)] Thought you might want to see this...
- From: Gabriel Dos Reis <gdr at integrable-solutions dot net>
- To: libstdc++ at gcc dot gnu dot org
- Date: 05 May 2004 15:32:43 +0200
- Subject: [(nowhere)] Thought you might want to see this...
- Organization: Integrable Solutions
A friend sent this to me and since we now have performance testsuite,
I thought you'll be interested too. The downside is that the original
poster did not provide a testcase. However, I thought V3 might want
to get into contact with him and try to sort out what is going on.
-- Gaby
Subject: Topics
Topics:
Thought you might want to see this...
--- Begin Message ---
- From: David Abrahams <dave at boost-consulting dot com>
- To: Gabriel Dos Reis <gdr at acm dot org>
- Date: Wed, 05 May 2004 10:28:27 -0400
- Subject: Thought you might want to see this...
--- Begin Message ---
- From: loewe at elvis dot com (Bernd)
- Date: 5 May 2004 06:51:33 -0400
- Subject: poor performance of gcc 3.4 stl::multimap
- Approved: francis.moderator@robinton.demon.co.uk
- Newsgroups: comp.lang.c++.moderated
- Organization: http://groups.google.com
- Xref: news.rcn.com comp.lang.c++.moderated:165209
I recognized very poor performance in using the gcc 3.4 multimap. When
builing a big map and iterating very often over the ranges delivered
by equal_range for specific keys the gcc implementation is about 30
times slower than the implementation shipped with SUN Studio 8. This I
think is very strange, because all other code of our application is
normally about 10 % faster when using gcc.
Regards,
Bernd
[ See http://www.gotw.ca/resources/clcm.htm for info about ]
[ comp.lang.c++.moderated. First time posters: Do this! ]
--- End Message ---
--
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com
--- End Message ---
--
Gabriel Dos Reis
gdr@integrable-solutions.net