From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Hurley Subject: Re: [PATCH 1/2] serial: 8250: Fix autoconfig_irq() to avoid race conditions Date: Wed, 08 Jul 2015 19:35:26 -0400 Message-ID: <559DB3BE.5080306@hurleysoftware.com> References: <55717224.9060104@cp.jp.nec.com> <55717294.8020905@cp.jp.nec.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <55717294.8020905@cp.jp.nec.com> Sender: linux-kernel-owner@vger.kernel.org To: Taichi Kageyama , "gregkh@linuxfoundation.org" Cc: "linux-serial@vger.kernel.org" , "jslaby@suse.cz" , "linux-kernel@vger.kernel.org" , "prarit@redhat.com" , Naoya Horiguchi List-Id: linux-serial@vger.kernel.org On 06/05/2015 05:57 AM, Taichi Kageyama wrote: > The following race conditions can happen if a serial is used as console. > Case1. CPU_B handles an interrupt from a serial > autoconfig_irq() fails whether the interrupt is raised or not > if CPU_B is disabled to handle interrupts for longer than it expects. > Case2. CPU_B clears UART_IER just after CPU_A sets UART_IER > A serial may not make an interrupt. > autoconfig_irq() can fail if the interrupt is not raised. > Case3. CPU_A sets UART_IER just after CPU_B clears UART_IER > This is an unexpected behavior for uart_console_write(). > > CPU_A [autoconfig_irq] CPU_B [serial8250_console_write] > ----------------------------------------------------------------- > probe_irq_on() spin_lock_irqsave(&port->lock,) > serial_outp(,UART_IER,0x0f) serial_out(,UART_IER,0) > udelay(20); uart_console_write() > probe_irq_off() > spin_unlock_irqrestore(&port->lock,) > ----------------------------------------------------------------- > > If autoconfig_irq() fails, the console doesn't work in interrupt mode, > the mode cannot be changed anymore, and "input overrun" > (which can make operation mistakes) happens easily. > This problem happens with high rate every boot once it occurs > because the boot sequence is always almost same. Reviewed-by: Peter Hurley