From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751046AbWAIIy7 (ORCPT ); Mon, 9 Jan 2006 03:54:59 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751103AbWAIIy7 (ORCPT ); Mon, 9 Jan 2006 03:54:59 -0500 Received: from caramon.arm.linux.org.uk ([212.18.232.186]:44816 "EHLO caramon.arm.linux.org.uk") by vger.kernel.org with ESMTP id S1751046AbWAIIy7 (ORCPT ); Mon, 9 Jan 2006 03:54:59 -0500 Date: Mon, 9 Jan 2006 08:54:52 +0000 From: Russell King To: Antonio Vargas Cc: Meelis Roos , alan@lxorguk.ukuu.org.uk, linux-kernel@vger.kernel.org Subject: Re: Serial: bug in 8250.c when handling PCI or other level triggers Message-ID: <20060109085452.GA9189@flint.arm.linux.org.uk> Mail-Followup-To: Antonio Vargas , Meelis Roos , alan@lxorguk.ukuu.org.uk, linux-kernel@vger.kernel.org References: <20051222130744.GA31339@flint.arm.linux.org.uk> <20051223093343.GA22506@flint.arm.linux.org.uk> <20051223104146.GB22506@flint.arm.linux.org.uk> <20051228195509.GA12307@flint.arm.linux.org.uk> <69304d110601081524v37b15ff2tfed8341eaffbe07@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <69304d110601081524v37b15ff2tfed8341eaffbe07@mail.gmail.com> User-Agent: Mutt/1.4.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Jan 09, 2006 at 12:24:52AM +0100, Antonio Vargas wrote: > On 12/29/05, Meelis Roos wrote: > > > Can I assume that the bug has disappeared? Does the patch make it > > > disappear? > > > > Yes, seems so. > > > > -- > > Meelis Roos (mroos@linux.ee) > > Please notice official linus 2.6.15 tree doesn't have this fix... I've > just installed a virtual machine (qemu-system-i386 with linus 2.6.15 + > plain debian 3r0, console output to xterm via emulated serial console) > and trying to use any curses program (top for example) produces > exactly this type of error. That's because the patch just adds debugging seems to be sufficient to cause the bug to disappear. It isn't a fix. The only explaination I have at the moment is that it changes the timing - maybe there's a bug in these UARTs... I don't know at the moment. What I do know is that Alan's original premise seems wrong. -- Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of: 2.6 Serial core