From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261188AbVDSIre (ORCPT ); Tue, 19 Apr 2005 04:47:34 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261168AbVDSIre (ORCPT ); Tue, 19 Apr 2005 04:47:34 -0400 Received: from dspnet.fr.eu.org ([213.186.44.138]:45574 "EHLO dspnet.fr.eu.org") by vger.kernel.org with ESMTP id S261188AbVDSIrc (ORCPT ); Tue, 19 Apr 2005 04:47:32 -0400 Date: Tue, 19 Apr 2005 10:47:30 +0200 From: Olivier Galibert To: "Hack inc." Subject: Can a non-sg scsi write command be more than PAGE_SIZE length? Message-ID: <20050419084730.GA96767@dspnet.fr.eu.org> Mail-Followup-To: Olivier Galibert , "Hack inc." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.4.2.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org ...or more importantly, is it allowed. Kernel is FC3 2.6.10-1.766. The latest iscsi driver[1] blows on a 32K-long request for a tape write which followed this path: [] iscsi_queuecommand+0x161/0x2f1 [iscsi_tcp] [] scsi_dispatch_cmd+0x1e9/0x24f [scsi_mod] [] scsi_request_fn+0x29a/0x310 [scsi_mod] [] blk_insert_request+0x8b/0x9e [] scsi_insert_special_req+0x20/0x26 [scsi_mod] [] st_do_scsi+0x111/0x147 [st] [] st_sleep_done+0x0/0xc1 [st] [] st_write+0x3eb/0x671 [st] [] vfs_write+0xb6/0xe2 [] sys_write+0x3c/0x62 [] syscall_call+0x7/0xb The command is not using scatter-gather (sc->use_sg is 0) but is more than PAGE_SIZE, and the driver hates that. So my question is whether it's allowed (then I'll have to fix the driver, not sure many people have virtual tapes over iscsi) or whether st_write has to be fixed. OG. [1] 5.0.0.2.