From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 20508 invoked by alias); 4 Dec 2014 19:16:11 -0000 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 Received: (qmail 20495 invoked by uid 89); 4 Dec 2014 19:16:10 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-3.9 required=5.0 tests=AWL,BAYES_00,SPF_HELO_PASS,SPF_PASS,T_RP_MATCHES_RCVD autolearn=ham version=3.3.2 X-HELO: mx1.redhat.com Received: from mx1.redhat.com (HELO mx1.redhat.com) (209.132.183.28) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES256-GCM-SHA384 encrypted) ESMTPS; Thu, 04 Dec 2014 19:16:09 +0000 Received: from int-mx14.intmail.prod.int.phx2.redhat.com (int-mx14.intmail.prod.int.phx2.redhat.com [10.5.11.27]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id sB4JG3nu013383 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=FAIL); Thu, 4 Dec 2014 14:16:03 -0500 Received: from host2.jankratochvil.net (ovpn-116-31.ams2.redhat.com [10.36.116.31]) by int-mx14.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id sB4JFxXn029237 (version=TLSv1/SSLv3 cipher=AES128-GCM-SHA256 bits=128 verify=NO); Thu, 4 Dec 2014 14:16:02 -0500 Date: Thu, 04 Dec 2014 19:16:00 -0000 From: Jan Kratochvil To: Doug Evans Cc: gdb-patches@sourceware.org, brobecker@adacore.com Subject: Re: [RFC] symbol lookup cache Message-ID: <20141204191559.GA10848@host2.jankratochvil.net> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.23 (2014-03-12) X-IsSubscribed: yes X-SW-Source: 2014-12/txt/msg00108.txt.bz2 On Mon, 01 Dec 2014 09:23:49 +0100, Doug Evans wrote: > ..gdb_index helps speed things up for one objfile, > but when there are 100s (or 1000s) of shared > libraries, symbol lookup can still take awhile, > even with .gdb_index. > [I'm setting aside a planned change to basic > symbol lookup to use the index better. > We'll still, I think, want a cache even > with that change.] While I have no numbers personally I do not think GDB would still need a cache after that change. "Accelerate lookup_symbol_aux_objfile 85x" is so fast in a fraction of second because it traverses 146426 symtabs. So 146426 is OK but 1979341 was not (it was in some way more than 1979341 before you refactored symtabs). I was considering in the past to have some merged .gdb_index for all the shared libraries used by some app on disk (not just in memory). But that makes sense for HDD but less for SSD with the regular ~100 libraries. While it make more sense for Google 1000+ libraries I am not sure if Google always uses the same set of 1000+ libraries for a given app. Jan