|
From: | Clemens Hintze |
Subject: | [Monotone-devel] Copy-on-write not mandated! (Was: Typesafe VA_ARGS replacement for database::execute/fetch) |
Date: | Sun, 22 Jan 2006 08:53:45 +0100 |
User-agent: | Mozilla Thunderbird 1.0 (X11/20050213) |
Nathaniel Smith wrote:
On Sat, Jan 21, 2006 at 03:11:51AM -0600, Timothy Brownawell wrote:
(...)
Aren't strings copy-on-write anyway, so this shouldn't be a problem?Oh, hrm. That's a clever point.
Sorry for bothering, but - while it is true, that the C++ standard was written in a manner that allow copy-on-write technics for implementation of basic_string, it does not mandate to do so!
Further it seems, that there are situations, where copy-on-write will behave worser than a simple plain implementation - thread safety comes to mind. See for example
http://www.gotw.ca/publications/optimizations.htmIn this light it could not being impossible, that STL vendors could leave the copy-on-write path.
Best regards, Clemens.
[Prev in Thread] | Current Thread | [Next in Thread] |