From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Thu, 15 Sep 2005 17:40:35 +0100 From: Russell King Subject: Re: [RFC] killing linux/irq.h Message-ID: <20050915174035.B26124@flint.arm.linux.org.uk> References: <20050909184254.GT9623@ZenIV.linux.org.uk> <20050915163455.GD16698@parisc-linux.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20050915163455.GD16698@parisc-linux.org>; from matthew@wil.cx on Thu, Sep 15, 2005 at 10:34:55AM -0600 Sender: Russell King To: Matthew Wilcox Cc: Geert Uytterhoeven , viro@ZenIV.linux.org.uk, Linus Torvalds , linux-arch@vger.kernel.org, Linux Kernel Development List-ID: On Thu, Sep 15, 2005 at 10:34:55AM -0600, Matthew Wilcox wrote: > On Sun, Sep 11, 2005 at 09:50:38AM +0200, Geert Uytterhoeven wrote: > > On Fri, 9 Sep 2005 viro@ZenIV.linux.org.uk wrote: > > > We get regular portability bugs when somebody decides to include > > > linux/irq.h into a driver instead of asm/irq.h. It's almost always a > > > wrong thing to do and, in fact, causes immediate breakage on e.g. arm. > > > > Wouldn't it be more logical to make linux/irq.h the preferred include? > > Usually the linux/* versions are preferred over the asm/* versions. > > There's almost no reason to want <*/irq.h> in the first place. Almost > all drivers really want The only exception I can think of is for ARM where we supplement the Linux interrupt API to deal with our configurable interrupt sources (high level/low level/rising edge/falling edge triggers) on certain platform groups. -- Russell King Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/ maintainer of: 2.6 Serial core