From mboxrd@z Thu Jan 1 00:00:00 1970 From: Zach Brown Subject: Re: git (commit|tag) atomicity Date: Tue, 28 Feb 2012 13:57:21 -0500 Message-ID: <4F4D2391.9020701@zabbo.net> References: <4F4D04F4.80905@ira.uka.de> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit Cc: Jon Jagger , git@vger.kernel.org, Holger Hellmuth To: Jakub Narebski X-From: git-owner@vger.kernel.org Tue Feb 28 20:04:43 2012 Return-path: Envelope-to: gcvg-git-2@plane.gmane.org Received: from vger.kernel.org ([209.132.180.67]) by plane.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1S2SLk-0003Hk-T1 for gcvg-git-2@plane.gmane.org; Tue, 28 Feb 2012 20:04:37 +0100 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965995Ab2B1TEc (ORCPT ); Tue, 28 Feb 2012 14:04:32 -0500 Received: from lulu.zabbo.net ([69.168.54.52]:51459 "EHLO lulu.zabbo.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S964824Ab2B1TEc (ORCPT ); Tue, 28 Feb 2012 14:04:32 -0500 X-Greylist: delayed 429 seconds by postgrey-1.27 at vger.kernel.org; Tue, 28 Feb 2012 14:04:32 EST Received: from [192.168.1.133] (unknown [66.207.222.38]) (using TLSv1 with cipher DHE-RSA-CAMELLIA256-SHA (256/256 bits)) (No client certificate requested) by lulu.zabbo.net (Postfix) with ESMTPSA id 7FD2222161A; Tue, 28 Feb 2012 10:57:22 -0800 (PST) User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20120131 Thunderbird/10.0 In-Reply-To: Sender: git-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: git@vger.kernel.org Archived-At: It's a bit of a tangent, but just to be sure people don't get the wrong impression.. > But I am not sure... that probably depends on how opendir(3) and > readdir(3) works on given filesystem wrt. updates to opened directory. > I think VFS on Linux ensures that you see view of filesystem as it was > on opendir(). No, readdir() does not give you a static view of the entries in a directory as it was on opendir(). readdir() will reflect modifications that are done after opendir(). The specifics for a given situation depend on how the file system maps the readdir position (f_pos) to directory entries. You can see very different results when comparing, say, stock ext2, indexed ext[34], and btrfs. - z (your message probably caught my eye because telldir()/seekdir() is *loathed* by file system designers)