Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932639AbVJGOGp (ORCPT ); Fri, 7 Oct 2005 10:06:45 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932648AbVJGOGp (ORCPT ); Fri, 7 Oct 2005 10:06:45 -0400 Received: from 223-177.adsl.pool.ew.hu ([193.226.223.177]:10249 "EHLO dorka.pomaz.szeredi.hu") by vger.kernel.org with ESMTP id S932634AbVJGOGo (ORCPT ); Fri, 7 Oct 2005 10:06:44 -0400 To: trond.myklebust@fys.uio.no CC: akpm@osdl.org, viro@ftp.linux.org.uk, linux-kernel@vger.kernel.org In-reply-to: <1128693175.8519.84.camel@lade.trondhjem.org> (message from Trond Myklebust on Fri, 07 Oct 2005 09:52:55 -0400) Subject: Re: [PATCH] nfs: don't drop dentry in d_revalidate References: <1128693175.8519.84.camel@lade.trondhjem.org> Message-Id: From: Miklos Szeredi Date: Fri, 07 Oct 2005 16:04:53 +0200 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 952 Lines: 26 > > NFS d_revalidate() is doing things that are supposed to be done by > > d_invalidate(). > > > > Dropping the dentry is especially bad, since it will make > > d_invalidate() bypass all checks. > > NAK! > > Bypassing the stupid d_invalidate checks is precisely the point here. > > Unlike local filesystems, we have to deal with the case where someone > deletes a file on the server and then creates a new one with the same > name. The d_invalidate checks will keep the wrong dentry hashed for as > long as some borken process has the file open. > d_invalidate() only makes sense for network filesystems like NFS. If it's broken, fix it. But bypassing it is definitely the wrong thing to do. Miklos - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/