From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756159Ab1HCX24 (ORCPT ); Wed, 3 Aug 2011 19:28:56 -0400 Received: from cantor2.suse.de ([195.135.220.15]:33748 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932121Ab1HCX2R (ORCPT ); Wed, 3 Aug 2011 19:28:17 -0400 X-Mailbox-Line: From gregkh@clark.kroah.org Wed Aug 3 15:12:47 2011 Message-Id: <20110803221247.121601684@clark.kroah.org> User-Agent: quilt/0.48-16.4 Date: Wed, 03 Aug 2011 15:00:51 -0700 From: Greg KH To: linux-kernel@vger.kernel.org, stable@kernel.org Cc: stable-review@kernel.org, torvalds@linux-foundation.org, akpm@linux-foundation.org, alan@lxorguk.ukuu.org.uk, Stefan Richter Subject: [031/102] firewire: cdev: prevent race between first get_info ioctl and bus reset event queuing In-Reply-To: <20110803221345.GA10346@kroah.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org 3.0-stable review patch. If anyone has any objections, please let us know. ------------------ From: Stefan Richter commit 93b37905f70083d6143f5f4dba0a45cc64379a62 upstream. Between open(2) of a /dev/fw* and the first FW_CDEV_IOC_GET_INFO ioctl(2) on it, the kernel already queues FW_CDEV_EVENT_BUS_RESET events to be read(2) by the client. The get_info ioctl is practically always issued right away after open, hence this condition only occurs if the client opens during a bus reset, especially during a rapid series of bus resets. The problem with this condition is twofold: - These bus reset events carry the (as yet undocumented) @closure value of 0. But it is not the kernel's place to choose closures; they are privat to the client. E.g., this 0 value forced from the kernel makes it unsafe for clients to dereference it as a pointer to a closure object without NULL pointer check. - It is impossible for clients to determine the relative order of bus reset events from get_info ioctl(2) versus those from read(2), except in one way: By comparison of closure values. Again, such a procedure imposes complexity on clients and reduces freedom in use of the bus reset closure. So, change the ABI to suppress queuing of bus reset events before the first FW_CDEV_IOC_GET_INFO ioctl was issued by the client. Note, this ABI change cannot be version-controlled. The kernel cannot distinguish old from new clients before the first FW_CDEV_IOC_GET_INFO ioctl. We will try to back-merge this change into currently maintained stable/ longterm series, and we only document the new behaviour. The old behavior is now considered a kernel bug, which it basically is. Signed-off-by: Stefan Richter Cc: --- drivers/firewire/core-cdev.c | 18 ++++++++++-------- include/linux/firewire-cdev.h | 3 +++ 2 files changed, 13 insertions(+), 8 deletions(-) --- a/drivers/firewire/core-cdev.c +++ b/drivers/firewire/core-cdev.c @@ -253,14 +253,11 @@ static int fw_device_op_open(struct inod init_waitqueue_head(&client->wait); init_waitqueue_head(&client->tx_flush_wait); INIT_LIST_HEAD(&client->phy_receiver_link); + INIT_LIST_HEAD(&client->link); kref_init(&client->kref); file->private_data = client; - mutex_lock(&device->client_list_mutex); - list_add_tail(&client->link, &device->client_list); - mutex_unlock(&device->client_list_mutex); - return nonseekable_open(inode, file); } @@ -451,15 +448,20 @@ static int ioctl_get_info(struct client if (ret != 0) return -EFAULT; + mutex_lock(&client->device->client_list_mutex); + client->bus_reset_closure = a->bus_reset_closure; if (a->bus_reset != 0) { fill_bus_reset_event(&bus_reset, client); - if (copy_to_user(u64_to_uptr(a->bus_reset), - &bus_reset, sizeof(bus_reset))) - return -EFAULT; + ret = copy_to_user(u64_to_uptr(a->bus_reset), + &bus_reset, sizeof(bus_reset)); } + if (ret == 0 && list_empty(&client->link)) + list_add_tail(&client->link, &client->device->client_list); - return 0; + mutex_unlock(&client->device->client_list_mutex); + + return ret ? -EFAULT : 0; } static int add_client_resource(struct client *client, --- a/include/linux/firewire-cdev.h +++ b/include/linux/firewire-cdev.h @@ -475,6 +475,9 @@ union fw_cdev_event { * of the bus. This does not cause a bus reset to happen. * @bus_reset_closure: Value of &closure in this and subsequent bus reset events * @card: The index of the card this device belongs to + * + * As a side effect, reception of %FW_CDEV_EVENT_BUS_RESET events to be read(2) + * is started by this ioctl. */ struct fw_cdev_get_info { __u32 version;