From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261534AbUBUI1E (ORCPT ); Sat, 21 Feb 2004 03:27:04 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261540AbUBUI1E (ORCPT ); Sat, 21 Feb 2004 03:27:04 -0500 Received: from pr-117-210.ains.net.au ([202.147.117.210]:37317 "EHLO mail.ocs.com.au") by vger.kernel.org with ESMTP id S261541AbUBUI0w (ORCPT ); Sat, 21 Feb 2004 03:26:52 -0500 X-Mailer: exmh version 2.5 01/15/2001 with nmh-1.0.4 From: Keith Owens To: Kernel Mailing List Subject: Re: explicit dcache <-> user-space cache coherency, sys_mark_dir_clean(), O_CLEAN In-reply-to: Your message of "Sat, 21 Feb 2004 08:58:53 BST." <20040221075853.GA828@elte.hu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 21 Feb 2004 19:26:32 +1100 Message-ID: <3762.1077351992@ocs3.ocs.com.au> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sat, 21 Feb 2004 08:58:53 +0100, Ingo Molnar wrote: >* Ingo Molnar wrote: >> perhaps using a simple 64-bit generation counter would be better. >> Samba would get a new syscall to get the sum of each generation >> counter down to the root dentry - a total validation of the pathname. >> If the counter matches with that in the userspace cache entry then no >> need to re-create the cache. Such generation counters would be usable >> for multiple file servers as well. Hm? > >generation counters are problematic if they are not persistent. But >there's a pretty natural persistent 'generation counter' which could be >used for Samba's purpose: the mtime of the directory ... >... monotonity is important: two successive directory operations to not be >possible within the same nanosecond. Why do you need monotonity? Samba only cares if the dcache entry changes, the indicator from kernel to user space does not have to be monotonically increasing, just different.