From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751422Ab0JOW3Z (ORCPT ); Fri, 15 Oct 2010 18:29:25 -0400 Received: from sj-iport-5.cisco.com ([171.68.10.87]:8914 "EHLO sj-iport-5.cisco.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751071Ab0JOW3X (ORCPT ); Fri, 15 Oct 2010 18:29:23 -0400 Authentication-Results: sj-iport-5.cisco.com; dkim=neutral (message not signed) header.i=none X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: Av0EAE9yuEyrR7Hu/2dsb2JhbACDH54RcaROih2SLIEigzN0BIRUhXY X-IronPort-AV: E=Sophos;i="4.57,338,1283731200"; d="scan'208";a="270312135" From: Tom Lyon Organization: Cisco Systems, Inc. To: Alex Williamson Subject: Re: [PATCH 3/3] VFIO V4: VFIO driver: Non-privileged user level PCI drivers Date: Fri, 15 Oct 2010 15:29:13 -0700 User-Agent: KMail/1.13.5 (Linux/2.6.34.7-0.3-desktop; KDE/4.4.4; x86_64; ; ) Cc: linux-pci@vger.kernel.org, jbarnes@virtuousgeek.org, linux-kernel@vger.kernel.org, kvm@vger.kernel.org, randy.dunlap@oracle.com, arnd@arndb.de, joro@8bytes.org, hjk@linutronix.de, avi@redhat.com, gregkh@suse.de, chrisw@sous-sol.org, mst@redhat.com References: <4c9a72a0.u2cnjUB7QZ91tLeo%pugs@cisco.com> <1285620073.4951.44.camel@x201> In-Reply-To: <1285620073.4951.44.camel@x201> MIME-Version: 1.0 Content-Type: Text/Plain; charset="utf-8" Content-Transfer-Encoding: 7bit Message-Id: <201010151529.13571.pugs@cisco.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Michael & Alex et al - Sorry for going quiet; I've been digesting the comments and researching a lot more stuff. I plan to release V5 shortly after 2.6.36 is out, highlights will be: 1. Re-written pci config tables - using approach suggested by MST to clean things up. Looking much better. Also fixed endian issues. 2. Clean up ROM bar handling. Now only read() system call can access rom bar; it is always disabled afterwards. 3. Clean up pci_iomap and pci_request_regions handling. Allocates on demand but frees all on close. 4. Resets device on open. Disables, but does not do full reset on close due to lock problem with remove() callers. 5. Fixed up memlock rlimit accounting. 6. Lots of small cleanups. Stay tuned. I'm also looking at adding PCIe extended capabilities - got a request from a cisco project.