From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261347AbUF3Tay (ORCPT ); Wed, 30 Jun 2004 15:30:54 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261602AbUF3Tax (ORCPT ); Wed, 30 Jun 2004 15:30:53 -0400 Received: from caramon.arm.linux.org.uk ([212.18.232.186]:45576 "EHLO caramon.arm.linux.org.uk") by vger.kernel.org with ESMTP id S261347AbUF3Taj (ORCPT ); Wed, 30 Jun 2004 15:30:39 -0400 Date: Wed, 30 Jun 2004 20:30:30 +0100 From: Russell King To: Pavel Machek Cc: Linux Kernel List Subject: Re: [PATCH 2/3] Provide console_suspend() and console_resume() Message-ID: <20040630203030.B1496@flint.arm.linux.org.uk> Mail-Followup-To: Pavel Machek , Linux Kernel List References: <20040614151217.H14403@flint.arm.linux.org.uk> <20040614151307.I14403@flint.arm.linux.org.uk> <20040630123727.GA16409@elf.ucw.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5.1i In-Reply-To: <20040630123727.GA16409@elf.ucw.cz>; from pavel@ucw.cz on Wed, Jun 30, 2004 at 02:37:29PM +0200 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Jun 30, 2004 at 02:37:29PM +0200, Pavel Machek wrote: > Hi! > > > Add console_suspend() and console_resume() methods so the serial drivers > > can disable console output before suspending a port, and re-enable output > > afterwards. > > Could it be called console_stop()/console_start()? suspend/resume > sounds like power managment, and it is unrelated.... It is power management related! However, we decided that console_start/ console_stop was a better name for it. That said, it is _NOT_ intended for general use since there is NO counting of how many times console_stop() is called. It is merely used by the serial driver to prevent further serial console output before shutting the console port down for power management. This avoids the problem where, on resume, we haven't even enabled the UART - if we were to attempt to write to the UART in this circumstance, it would not be setup in any way - no baud rate, and in fact, since it is disabled, we'd spin for ever waiting for the port to say its completed transmission of the first byte. So yes, it most definitely _is_ power management related. -- Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of: 2.6 PCMCIA - http://pcmcia.arm.linux.org.uk/ 2.6 Serial core