Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Tue, 12 Mar 2002 23:12:43 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Tue, 12 Mar 2002 23:12:34 -0500 Received: from gate.tuxia.com ([213.209.134.221]:46075 "EHLO exchange1.win.agb.tuxia") by vger.kernel.org with ESMTP id convert rfc822-to-8bit; Tue, 12 Mar 2002 23:12:24 -0500 X-MimeOLE: Produced By Microsoft Exchange V6.0.4418.65 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Subject: 2.4.1 Block Driver and mmap Date: Wed, 13 Mar 2002 05:12:23 +0100 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: SCSI errors with Philips MOs Thread-Index: AcHKPT0ZWt7w9vZOTYi2JNvzuCBT2AABiY2w From: "Tim McDaniel" To: Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org I'm porting a working 2.2 driver to 2.4 . Unfortunately, the mmap entry point is not defined in the block_device_operations. I've tried forcing filp->f_op to an initialized file operations structure while also registering the driver with block operations (as does the ramdisk driver). My mmap routine is entered and appears to work, but now the rest of the driver is experiencing strange failures such as: 1) errno 25 (decimal) on an ioctl that used to work even in 2.4 2) fsck fails with a short read error Without the change for filep->f_op, the mmap fails with errno 19 decimal. My driver sits between the file system and the low level block driver and funnels ata commands in a manner that is more fair and thus more appropriate for a PVR implementation. Any help would be greatly appreciated. Thanks, Tim - 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/