This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [RFC] optional fast iterators for set/map in libstdc++
- From: Benjamin Kosnik <bkoz at redhat dot com>
- To: Gawain BOLTON <gp dot bolton at computer dot org>
- Cc: janis187 at us dot ibm dot com, libstdc++ at gcc dot gnu dot org, gcc-patches at gcc dot gnu dot org
- Date: Mon, 11 May 2009 20:45:07 -0700
- Subject: Re: [RFC] optional fast iterators for set/map in libstdc++
- References: <1241744347.6152.26.camel@janis-laptop> <4A0493E9.5060308@computer.org>
> It seems that to be really useful, different container classes should
> be created. In this way, users could more easily use them based on
> their performance and memory constraints. A compile flag or switch
> just cannot give the flexibility required.
Agreed. This is an interesting idea, and seems promising, but this is
the wrong form. You'd think this would be a policy. I'll try to work up
an example of this.
And the idea of an ABI-different container with a compile flag and no
linkage difference is not the kind of thing that should be encouraged,
ever.
-benjamin