From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753822Ab1IVXA6 (ORCPT ); Thu, 22 Sep 2011 19:00:58 -0400 Received: from out2.smtp.messagingengine.com ([66.111.4.26]:60078 "EHLO out2.smtp.messagingengine.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751728Ab1IVXA4 (ORCPT ); Thu, 22 Sep 2011 19:00:56 -0400 X-Sasl-enc: INn+aJLAybcVQ0hGzSwvKjE3cFJc9HR3cOEUBpz56WUv 1316732455 Date: Thu, 22 Sep 2011 15:38:09 -0700 From: Greg KH To: Thadeu Lima de Souza Cascardo Cc: stable@kernel.org, Breno Leitao , lucaskt@linux.vnet.ibm.com, Alan Cox , linux-serial@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: jsm: remove buggy write queue Message-ID: <20110922223809.GA21159@kroah.com> References: <1314202463-12297-1-git-send-email-cascardo@linux.vnet.ibm.com> <20110913175522.GB3592@oc1711230544.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20110913175522.GB3592@oc1711230544.ibm.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Sep 13, 2011 at 02:55:23PM -0300, Thadeu Lima de Souza Cascardo wrote: > jsm uses a write queue that copies from uart_core circular buffer. This > copying however has some bugs, like not wrapping the head counter. Since > this write queue is also a circular buffer, the consumer function is > ready to use the uart_core circular buffer directly. > > This buggy copying function was making some bytes be dropped when > transmitting to a raw tty, doing something like this. > > [root@hostname ~]$ cat /dev/ttyn1 > cascardo/dump & > [1] 2658 > [root@hostname ~]$ cat /proc/tty/drivers > /dev/ttyn0 > [root@hostname ~]$ cat /proc/tty/drivers > /dev/tty /dev/tty 5 0 system:/dev/tty > /dev/console /dev/console 5 1 system:console > /dev/ptmx /dev/ptmx 5 2 system > /dev/vc/0 /dev/vc/0 4 0 system:vtmaster > jsm /dev/ttyn 250 0-31 serial > serial /dev/ttyS 4 64-95 serial > hvc /dev/hvc 229 0-7 system > pty_slave /dev/pts 136 0-1048575 pty:slave > pty_master /dev/ptm 128 0-1048575 pty:master > unknown /dev/tty 4 1-63 console > [root@hostname ~]$ cat cascardo/dump > /dev/tty /dev/tty 5 0 system:/dev/tty > /dev/console /dev/console 5 1 system:console > /dev/ptmx /dev/ptmx 5 2 system > /dev/vc/0 /dev/vc/0 4 0 system:vtmaste[root@hostname ~]$ > > This patch drops the driver write queue entirely, using the circular > buffer from uart_core only. > > Signed-off-by: Thadeu Lima de Souza Cascardo > --- > > Hello, Greg. > > This should probably go to 3.1 and stable releases, since it fixes a > bug. I already have this in my tree queued up for 3.2, so please remind me of this when it hits Linus's tree during the 3.2 merge window. thanks, greg k-h