This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: sso tradeoffs
- From: Paolo Carlini <pcarlini at suse dot de>
- To: Howard Hinnant <hhinnant at apple dot com>
- Cc: libstdc++ <libstdc++ at gcc dot gnu dot org>
- Date: Wed, 02 Nov 2005 19:34:50 +0100
- Subject: Re: sso tradeoffs
- References: <69BD9B30-F3BE-4128-997D-67B67C256FD7@apple.com>
Hi Howard,
> This is a respectable design, and according to Effective STL, very
> similar, if not a little better (imho) than implementation "D" from ESTL.
Just wanted to tell you, for now (real feedback later, of course) that I
have the book, like it a lot, generally, but haven't read that part
while designing (with Nathan's help) the very basic sso string that you
are finding in ext/ !! I don't know why!
Also wanted to tell you immediately, that indeed, swap is a concern, I
didn't expect that, and a few days ago I rewrote it from scratch because
otherwise simulated move semantics in v7-branch didn't pay off! In the
shape you see it now is decent performance-wise (together with the
trivial move constructor and move assignment that you can find in
v7-branch).
Paolo.