From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752471AbcCAIE4 (ORCPT ); Tue, 1 Mar 2016 03:04:56 -0500 Received: from www.linutronix.de ([62.245.132.108]:44442 "EHLO Galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751404AbcCAIEz (ORCPT ); Tue, 1 Mar 2016 03:04:55 -0500 Date: Tue, 1 Mar 2016 09:02:59 +0100 (CET) From: Thomas Gleixner To: "Mcfadden, Marty Jay" cc: Borislav Petkov , George Spelvin , "a.p.zijlstra@chello.nl" , "acme@infradead.org" , "ak@linux.intel.com" , "andriy.shevchenko@linux.intel.com" , "brgerst@gmail.com" , "dan.j.williams@intel.com" , "dyoung@redhat.com" , "hpa@zytor.com" , "jolsa@redhat.com" , "linux-kernel@vger.kernel.org" , "luto@kernel.org" , "mingo@kernel.org" , "mingo@redhat.com" , "pavel@ucw.cz" , "viro@zeniv.linux.org.uk" , "x86@kernel.org" , "yu.c.chen@intel.com" Subject: RE: [PATCH 0/4] MSR: MSR: MSR Whitelist and Batch Introduction In-Reply-To: Message-ID: References: <20160229145859.GA3724@pd.tnic> <20160229175318.14183.qmail@ns.horizon.com> <20160229182047.GD3724@pd.tnic> User-Agent: Alpine 2.11 (DEB 23 2013-08-11) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Linutronix-Spam-Score: -1.0 X-Linutronix-Spam-Level: - X-Linutronix-Spam-Status: No , -1.0 points, 5.0 required, ALL_TRUSTED=-1,SHORTCIRCUIT=-0.0001 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 29 Feb 2016, Mcfadden, Marty Jay wrote: > This is precisely why the whitelist approach is being proposed. The current > version of msr.ko will gladly allow userspace tools with capabilities set to > scribble all over them. With whitelists, system admins can turn off > capabilities for the apps and limit access to a very small subset of bits of > a small subset of MSRs. We don't need any of this, really. Developers who play with experimental or emerging technologies hardly need that whitelist stuff. They better know what they are doing. For normal production use we want proper interfaces/drivers for the technologies which should be made accessible to applications. msr.ko should not be available on any production machine at all. Your whitelist filter is just a sloppy hack to foster bad practices. Your arguments about emerging technologies etc. are just trying to lull us into accepting your wonderful hackery, so that you can continue to use MSRs in production environments while pretending that you provide a reasonable amount of security and sanity around it. Nice try, but it doesn't work... Thanks, tglx