This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: on the speed of std::string::find
Dennis Lubert wrote:
Hi, I did some simple benchmarks with the different implementation.
yours is the "new find patch" implementation, which seems to be a bit
faster at start, but scales worse. I didn't had enough time today, but I
think within the next few days I can build up a few charts with
descriptions of what I exactly did etc. pp.
http://www.projectiwear.org/~plasmahh/string_bench_simple.png
Dennis, thanks for your help, but something is certainly wrong in the
benchmarking, we can't trust it. Likely, it's because of the
optimization level, or something else. As I said already, you must use
-O2, otherwise we are measuring something which nobody will see in
practice. I'm saying that because the "new" implementation is
*identical* to glibc' memmem, the difference should be withing a few %
(indeed, the timings I'm seeing on many different machines are
consistent with those your reported days ago, certainly about 1.5-2.0 x
faster than the current implementation). Also, please use many different
strings: either use our performance testcase or a bi amount of *real* data.
Please let me know ASAP if you are able to conduct such kind of tests,
otherwise we are delaying unnecessarily closing the issue, thanks
Paolo.