From: "Lever, Charles" Subject: RE: [PATCH] NFS client doesn't flush dirty mmap'd pages during fsync/msync Date: Tue, 20 Jan 2004 05:55:53 -0800 Sender: nfs-admin@lists.sourceforge.net Message-ID: <482A3FA0050D21419C269D13989C611302B07B38@lavender-fe.eng.netapp.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Cc: Return-path: Received: from sc8-sf-mx2-b.sourceforge.net ([10.3.1.12] helo=sc8-sf-mx2.sourceforge.net) by sc8-sf-list2.sourceforge.net with esmtp (Exim 4.30) id 1AiwMU-0000XO-Qk for nfs@lists.sourceforge.net; Tue, 20 Jan 2004 05:56:10 -0800 Received: from mx01.netapp.com ([198.95.226.53]) by sc8-sf-mx2.sourceforge.net with esmtp (Exim 4.30) id 1AiwMU-0008Jb-70 for nfs@lists.sourceforge.net; Tue, 20 Jan 2004 05:56:10 -0800 To: , Errors-To: nfs-admin@lists.sourceforge.net List-Unsubscribe: , List-Id: Discussion of NFS under Linux development, interoperability, and testing. List-Post: List-Help: List-Subscribe: , List-Archive: > > Given that we have the inode for these dirty pages,=20 > > couldn't we at least > > compare the req->wb_inode to filp->f_dentry->d_inode? >=20 > I believe that is indeed what Chuck's patch does. no, my patch simply removes the filp check. the inode is not available in nfs_scan_list, and there was never a check there to see if the inode pointers match. > > With your patch, an fsync degenerates to a global sync > > on the mounted file system, if I'm not mistaken. the patch does not change which list is passed to nfs_scan_list. nfs_scan_list is used by nfs_scan_dirty and nfs_scan_commit -- so it always works on a single file at a time. ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ NFS maillist - NFS@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs