This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: binary_search, lower_bound and upper_bound inlining
- From: Paolo Carlini <pcarlini at suse dot de>
- To: Dimitris Xochellis <jimxoch at yahoo dot gr>
- Cc: libstdc++ <libstdc++ at gcc dot gnu dot org>
- Date: Sun, 01 Apr 2007 19:45:32 +0200
- Subject: Re: binary_search, lower_bound and upper_bound inlining
- References: <655930.67722.qm@web23103.mail.ird.yahoo.com>
Dimitris Xochellis wrote:
Well, if you can provide unbiased, compelling evidence that code bloat
isn't too big...
Well, I don't feel so "lucky" today :)
(I also agree that the code bloat will probably be rather big)
As soon as I find a bit of spare time I want to investigate the issue a
bit more myself: those functions aren't *very* big and are indeed quite
important, for instance are use in our unordered_* containers. I want to
figure out how big really are in terms of instructions, etc.
That's sad. You should tell all the people, the compiler people too not
only the library developers: either the options are too hard to find or
wrongly missing. Can you send a message to gcc@gcc.gnu.org?
Good idea, I will do so.
Excellent. Your needs are certainly sensible.
Paolo.