From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1423202AbXD3Mcn (ORCPT ); Mon, 30 Apr 2007 08:32:43 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1423200AbXD3Mcm (ORCPT ); Mon, 30 Apr 2007 08:32:42 -0400 Received: from amsfep20-int.chello.nl ([62.179.120.15]:44755 "EHLO amsfep20-int.chello.nl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1423202AbXD3Mcl (ORCPT ); Mon, 30 Apr 2007 08:32:41 -0400 Subject: Re: [PATCH] [39/40] i386: Export paravirt_ops for non GPL modules too From: Peter Zijlstra To: Andi Kleen Cc: Jan Engelhardt , Christoph Hellwig , mingo@elte.hu, patches@x86-64.org, linux-kernel@vger.kernel.org In-Reply-To: <200704301319.54291.ak@suse.de> References: <200704301227.598020000@suse.de> <200704301300.31298.ak@suse.de> <200704301319.54291.ak@suse.de> Content-Type: text/plain Date: Mon, 30 Apr 2007 14:28:59 +0200 Message-Id: <1177936139.4843.17.camel@lappy> Mime-Version: 1.0 X-Mailer: Evolution 2.8.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 2007-04-30 at 13:19 +0200, Andi Kleen wrote: > On Monday 30 April 2007 13:15:36 Jan Engelhardt wrote: > > > > On Apr 30 2007 13:00, Andi Kleen wrote: > > >On Monday 30 April 2007 12:50:09 Christoph Hellwig wrote: > > >> On Mon, Apr 30, 2007 at 12:28:14PM +0200, Andi Kleen wrote: > > >> > > > >> > Otherwise non GPL modules cannot even do basic operations > > >> > like disabling interrupts anymore, which would be excessive. > > >> > > > >> > Longer term should split the single structure up into > > >> > internal and external symbols and not export the internal > > >> > ones at all. > > >> > > > >> > Signed-off-by: Andi Kleen > > >> > > >> Ingo was dead-set against this and I kinda agree. > > > > > >The problem is that without this non GPL modules cannot even disable > > >interrupts anymore. That is imho too radical. > > > > Perhaps we can have a paravirt_ops2 that specifically deals with > > interrupt en/disable, and export that instead? > > Yes that is what the "Longer term ..." paragraph above refers to. > However it would need some restructuring in the code. FWIW I think doing this first will be better, exposing _all_ to non GNU modules will weaken whatever case we might have to take it away later. So, NACK from me too. I don't want to hear the whining; but it was allowed in .22, so why should we not be able to do this in .23.... or whatever.