From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 8225 invoked by alias); 5 Mar 2002 22:04:11 -0000 Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-owner@gcc.gnu.org Received: (qmail 8154 invoked from network); 5 Mar 2002 22:04:09 -0000 Received: from unknown (HELO ariane.ens-cachan.fr) (138.231.176.4) by sources.redhat.com with SMTP; 5 Mar 2002 22:04:09 -0000 Received: from mayo.cmla.ens-cachan.fr (mayo.cmla.ens-cachan.fr [138.231.64.2]) by ariane.ens-cachan.fr (8.12.0.Beta19/jtpda-5.3.3) with ESMTP id g25M43rD022584 ; Tue, 5 Mar 2002 23:04:03 +0100 Received: from jambon.cmla.ens-cachan.fr (jambon.cmla.ens-cachan.fr [138.231.64.60]) by mayo.cmla.ens-cachan.fr (8.9.1a/jtpda-5.3.2) with ESMTP id XAA08172 ; Tue, 5 Mar 2002 23:03:58 +0100 (MET) Received: from (dosreis@localhost) by jambon.cmla.ens-cachan.fr (8.9.3/jtpda-5.3.1/CL) id XAA03051 ; Tue, 5 Mar 2002 23:04:00 +0100 (MET) To: Benjamin Kosnik Cc: Gabriel Dos Reis , stephen.webb@bregmasoft.com, gcc@gcc.gnu.org, libstdc++@gcc.gnu.org Subject: Re: Results for g++ 3.1 application testing on i686-pc-linux-gnu References: From: Gabriel Dos Reis In-Reply-To: Benjamin Kosnik's message of "Tue, 5 Mar 2002 13:11:13 -0800 (PST)" Organization: CodeSourcery, LLC Mime-Version: 1.0 (generated by tm-edit 7.106) Content-Type: text/plain; charset=US-ASCII Date: Tue, 05 Mar 2002 14:04:00 -0000 Message-ID: X-Mailer: Gnus v5.6.45/Emacs 19.34 X-SW-Source: 2002-03/txt/msg00156.txt.bz2 Benjamin Kosnik writes: | > Sigh. The right fix is to replace the declaration | > | > extern const int __WORD_BIT; | > | > in bits/stl_bvector.h with | > | > const int __WORD_BIT = int(CHAR_BIT*sizeof(unsigned int)); | > | > and ditch that line in src/stl-inst.cc. | | Hmmm. I take issue with this. | | I think the real, best solution would be to remove all these goofy global | constants from the SGI STL, and put them in base classes where | possible....... Agreed. | If that's not always possible I suppose Gaby's suggestion is the | next-best thing. It's annoying to me to see, for every header with | | const int l = 5; | | in it, this in every object file | | 00000000 r l Then, may I propose to turn __WORD_BIT into an enum so as to meet the zero-overhead criterion. -- Gaby