From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 2738 invoked by alias); 17 Aug 2010 17:40:44 -0000 Received: (qmail 2729 invoked by uid 22791); 17 Aug 2010 17:40:43 -0000 X-SWARE-Spam-Status: No, hits=-1.4 required=5.0 tests=AWL,BAYES_00,KAM_STOCKGEN,T_RP_MATCHES_RCVD X-Spam-Check-By: sourceware.org Received: from mail.codesourcery.com (HELO mail.codesourcery.com) (38.113.113.100) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Tue, 17 Aug 2010 17:40:37 +0000 Received: (qmail 13097 invoked from network); 17 Aug 2010 17:40:35 -0000 Received: from unknown (HELO caradoc.them.org) (dan@127.0.0.2) by mail.codesourcery.com with ESMTPA; 17 Aug 2010 17:40:35 -0000 Date: Tue, 17 Aug 2010 17:40:00 -0000 From: Daniel Jacobowitz To: sami wagiaalla Cc: gdb-patches@sourceware.org Subject: Re: [RFC] Use custom hash function with bcache Message-ID: <20100817174023.GA26338@caradoc.them.org> Mail-Followup-To: sami wagiaalla , gdb-patches@sourceware.org References: <4C6946E1.6000709@redhat.com> <20100816191348.GA16221@caradoc.them.org> <4C699679.6090209@redhat.com> <20100816200325.GA19823@caradoc.them.org> <4C699B55.5000601@redhat.com> <20100816204947.GA22993@caradoc.them.org> <4C6AC07A.1040303@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4C6AC07A.1040303@redhat.com> User-Agent: Mutt/1.5.20 (2009-06-14) Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org X-SW-Source: 2010-08/txt/msg00262.txt.bz2 On Tue, Aug 17, 2010 at 01:01:46PM -0400, sami wagiaalla wrote: > Hmm... this is probably a separate patch, but how about this: > symbol_set_names sets the mangled and demangled names we need at > least the mangled name to find out wether there is already a name > version in the bcache. So maybe we could only set the mangled name, > check the bcache and only set the demangled name after we have > consulted the bcache. This only works of course if we ignore the > demangled name as is in this patch (Do you agree that this is OK btw > ?). I think it'll be OK. As for SYMBOL_SET_NAMES, I worry that will lose the optimization we got by doing them both at once; I don't remember how it works (it's been five years), but it made a big difference to startup time. -- Daniel Jacobowitz CodeSourcery