From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932547AbcFTVNo (ORCPT ); Mon, 20 Jun 2016 17:13:44 -0400 Received: from mga02.intel.com ([134.134.136.20]:62052 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751596AbcFTVMe (ORCPT ); Mon, 20 Jun 2016 17:12:34 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.26,500,1459839600"; d="scan'208";a="1001834933" From: Andi Kleen To: Borislav Petkov Cc: lkml , Thomas Gleixner , Ingo Molnar , "H. Peter Anvin" , Greg Kroah-Hartman , Thomas Renninger , Kan Liang , "Peter Zijlstra \(Intel\)" , "Rafael J. Wysocki" , Len Brown , linux-pm@vger.kernel.org Subject: Re: [RFC PATCH] x86: Move away from /dev/cpu/*/msr References: <20160615100029.GB32588@pd.tnic> Date: Mon, 20 Jun 2016 14:12:32 -0700 In-Reply-To: <20160615100029.GB32588@pd.tnic> (Borislav Petkov's message of "Wed, 15 Jun 2016 12:00:29 +0200") Message-ID: <87inx3bm4f.fsf@tassilo.jf.intel.com> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.5 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Borislav Petkov writes: > Comments are, as always, appreciated. Seems like a waste of kernel code to me. The MSR interface works perfectly fine. There are potentially hundreds of useful MSRs, are you going to add new sysfs for each of them? Even the more obscure ones can be very useful for debugging and monitoring. Most MSRs are model specific so this would end up with tons of switch (x86_model) ... which are always difficult to maintain and need to be updated all the time when new CPUs come out. This will likely generate a really large ongoing number of patches, and to solve what problem exactly? The whole thing doesn't make any sense to me. It's just a waste of code, maintainer time, patch review capacity, which all could be far more usefully employed to do something that actually solves real problems. -Andi -- ak@linux.intel.com -- Speaking for myself only From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andi Kleen Subject: Re: [RFC PATCH] x86: Move away from /dev/cpu/*/msr Date: Mon, 20 Jun 2016 14:12:32 -0700 Message-ID: <87inx3bm4f.fsf@tassilo.jf.intel.com> References: <20160615100029.GB32588@pd.tnic> Mime-Version: 1.0 Content-Type: text/plain Return-path: In-Reply-To: <20160615100029.GB32588@pd.tnic> (Borislav Petkov's message of "Wed, 15 Jun 2016 12:00:29 +0200") Sender: linux-kernel-owner@vger.kernel.org To: Borislav Petkov Cc: lkml , Thomas Gleixner , Ingo Molnar , "H. Peter Anvin" , Greg Kroah-Hartman , Thomas Renninger , Kan Liang , "Peter Zijlstra (Intel)" , "Rafael J. Wysocki" , Len Brown , linux-pm@vger.kernel.org List-Id: linux-pm@vger.kernel.org Borislav Petkov writes: > Comments are, as always, appreciated. Seems like a waste of kernel code to me. The MSR interface works perfectly fine. There are potentially hundreds of useful MSRs, are you going to add new sysfs for each of them? Even the more obscure ones can be very useful for debugging and monitoring. Most MSRs are model specific so this would end up with tons of switch (x86_model) ... which are always difficult to maintain and need to be updated all the time when new CPUs come out. This will likely generate a really large ongoing number of patches, and to solve what problem exactly? The whole thing doesn't make any sense to me. It's just a waste of code, maintainer time, patch review capacity, which all could be far more usefully employed to do something that actually solves real problems. -Andi -- ak@linux.intel.com -- Speaking for myself only