Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262690AbTI1TPX (ORCPT ); Sun, 28 Sep 2003 15:15:23 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262691AbTI1TPX (ORCPT ); Sun, 28 Sep 2003 15:15:23 -0400 Received: from caramon.arm.linux.org.uk ([212.18.232.186]:12560 "EHLO caramon.arm.linux.org.uk") by vger.kernel.org with ESMTP id S262690AbTI1TPU (ORCPT ); Sun, 28 Sep 2003 15:15:20 -0400 Date: Sun, 28 Sep 2003 20:15:16 +0100 From: Russell King To: Linus Torvalds Cc: Linux Kernel List , Roman Zippel Subject: Re: CONFIG_I8042 Message-ID: <20030928201516.D1428@flint.arm.linux.org.uk> Mail-Followup-To: Linus Torvalds , Linux Kernel List , Roman Zippel References: <20030928194511.C1428@flint.arm.linux.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5.1i In-Reply-To: ; from torvalds@osdl.org on Sun, Sep 28, 2003 at 11:49:45AM -0700 X-Message-Flag: Your copy of Microsoft Outlook is vulnerable to viruses. See www.mutt.org for more details. Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sun, Sep 28, 2003 at 11:49:45AM -0700, Linus Torvalds wrote: > > On Sun, 28 Sep 2003, Russell King wrote: > > > > It appears that "select" doesn't accept conditionals as the kconfig > > language stands - jejb also ran into this issue, and tried various > > ways around. The only solution which seems to work is to remove this > > select line entirely. > > That is WRONG. I don't think you've interpreted what I've said correctly. I'm not arguing at all about SERIO itself. In fact, I completely agree that selecting KEYBOARD_ATKBD should automatically select SERIO since atkbd.c uses serio.c. The information I received today from James Bottomley, who is also seeing this issue, is that the following construct resulted in I8042 being unconditionally selected: config KEYBOARD_ATKBD ... select SERIO select SERIO_I8042 if X86 However, I've just decided to try it myself, and it does indeed work as expected. Here's a (tested on non-x86) patch which fixes this issue and gives us the correct behaviour for non-x86 platforms, and should also give the desired behaviour for x86 and embedded platforms. --- orig/drivers/input/keyboard/Kconfig Sun Sep 28 09:54:29 2003 +++ linux/drivers/input/keyboard/Kconfig Sun Sep 28 20:06:40 2003 @@ -15,7 +15,8 @@ tristate "AT keyboard support" if EMBEDDED || !X86 default y depends on INPUT && INPUT_KEYBOARD - select SERIO_I8042 + select SERIO + select SERIO_I8042 if !EMBEDDED && X86 help Say Y here if you want to use a standard AT or PS/2 keyboard. Usually you'll need this, unless you have a different type keyboard (USB, ADB -- Russell King (rmk@arm.linux.org.uk) http://www.arm.linux.org.uk/personal/ 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 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/