From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 23676 invoked by alias); 12 Jan 2012 22:40:59 -0000 Received: (qmail 23661 invoked by uid 22791); 12 Jan 2012 22:40:58 -0000 X-SWARE-Spam-Status: No, hits=-6.6 required=5.0 tests=AWL,BAYES_00,RCVD_IN_DNSWL_HI,SPF_HELO_PASS,TW_BJ,T_RP_MATCHES_RCVD X-Spam-Check-By: sourceware.org Received: from mx1.redhat.com (HELO mx1.redhat.com) (209.132.183.28) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Thu, 12 Jan 2012 22:40:44 +0000 Received: from int-mx11.intmail.prod.int.phx2.redhat.com (int-mx11.intmail.prod.int.phx2.redhat.com [10.5.11.24]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id q0CMehoT017151 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Thu, 12 Jan 2012 17:40:43 -0500 Received: from host2.jankratochvil.net (ovpn-116-21.ams2.redhat.com [10.36.116.21]) by int-mx11.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id q0CMed7v021427 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NO); Thu, 12 Jan 2012 17:40:42 -0500 Date: Thu, 12 Jan 2012 22:50:00 -0000 From: Jan Kratochvil To: Paul Pluzhnikov Cc: Doug Evans , gdb-patches@sourceware.org Subject: Re: [patch] Fix for PR gdb/9538 (loading of separate debuginfo and symlinks). Message-ID: <20120112224039.GA13385@host2.jankratochvil.net> References: <20120112030648.14DBE190AFD@elbrus2.mtv.corp.google.com> <20120112212959.GA24491@host2.jankratochvil.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) X-IsSubscribed: yes 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: 2012-01/txt/msg00443.txt.bz2 On Thu, 12 Jan 2012 23:26:39 +0100, Paul Pluzhnikov wrote: > Wouldn't lstat need a configury #ifdef to make it build? Yes. There exists lstat in gnulib but configury #ifdef would be more appropriate here I think. > But there is no need to do realpath on the directory again, so I just pass > dir2 as canon_dir. Still you call there: + dir1 = xstrdup (objfile->name); + canon_dir = lrealpath (dir1); and later: + dir2 = lrealpath (objfile->name); For all the directory components of objfile->name lrealpath will lstat them by syscalls twice (for NFS it may get hopefully cached, not sure). There were some complaints on GDB slowness on NFS. I do not require this optimization, there are still worse issues on NFS. Thanks, Jan