From: Pavel Machek Subject: Re: [RFC PATCH 17/35] Segment register changes for Xen Date: Wed, 10 May 2006 22:30:15 +0200 Message-ID: <20060510203015.GA13949@elf.ucw.cz> References: <20060509084945.373541000@sous-sol.org> <20060509085154.802230000@sous-sol.org> <20060509071640.GA4150@ucw.cz> <200605102209.05004.ak@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <200605102209.05004.ak@suse.de> Sender: linux-kernel-owner@vger.kernel.org To: Andi Kleen Cc: virtualization@lists.osdl.org, Chris Wright , xen-devel@lists.xensource.com, linux-kernel@vger.kernel.org, Ian Pratt List-Id: xen-devel@lists.xenproject.org On St 10-05-06 22:09:04, Andi Kleen wrote: > On Tuesday 09 May 2006 09:16, Pavel Machek wrote: > > Hi! > > > > > --- linus-2.6.orig/include/asm-i386/mach-default/mach_system.h > > > +++ linus-2.6/include/asm-i386/mach-default/mach_system.h > > > @@ -1,6 +1,8 @@ > > > #ifndef __ASM_MACH_SYSTEM_H > > > #define __ASM_MACH_SYSTEM_H > > > > > > +#define clearsegment(seg) > > > > do {} while (0), please. > > It's not needed. Think about it. Really? If someone does if (something) clearsegment(seg) somethingelse(); ... he'll get very confusing behaviour instead of compile error. Okay, that's weaker argument than expected... Also clearsegment(x) clearsegment(y); will compile when it should not. Also clearsegment(i++) will behave strangely. So perhaps #define clearsegment(seg) do { seg; } while (0) is best variant? Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html