From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ns.suse.de ([195.135.220.2]:43157 "EHLO Cantor.suse.de") by vger.kernel.org with ESMTP id S263238AbUCTHMJ (ORCPT ); Sat, 20 Mar 2004 02:12:09 -0500 Date: Sat, 20 Mar 2004 05:57:53 +0100 From: Andi Kleen Subject: Re: compat-signal-noarch-2004-01-29.patch again Message-Id: <20040320055753.27517d88.ak@suse.de> In-Reply-To: <20040319223744.1f003f9f.akpm@osdl.org> References: <20040319214949.6b7d50bc.akpm@osdl.org> <20040320051519.780318b8.ak@suse.de> <20040319223744.1f003f9f.akpm@osdl.org> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit To: Andrew Morton Cc: linux-arch@vger.kernel.org, carlos@baldric.uwo.ca List-ID: On Fri, 19 Mar 2004 22:37:44 -0800 Andrew Morton wrote: > Andi Kleen wrote: > > > > On Fri, 19 Mar 2004 21:49:49 -0800 > > Andrew Morton wrote: > > > > > > > > This patch continues to hang around at Morton manor. last time it was > > > discussed Andi had issues with it and thing petered out. > > > > > > Can we please wrap this up one ay or the other? > > > > My only issue was the exporting of is_compat_task() to generic linux/compat.h > > If that is fixed (the code that does the is_compat_task is moved to arch/* > > again) then the change is fine and I would also likely use it for x86-64. > > Cannot you define HAVE_ARCH_IS_COMPAT_TASK? No. The problem is not in a particular implementation of this for a specific architecture, but in the generic design - it is the wrong direction and will cause hard to fix later design problems. -Andi