From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from merlin.infradead.org (merlin.infradead.org [205.233.59.134]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTPS id E7EECB6FA5 for ; Wed, 15 Feb 2012 02:21:39 +1100 (EST) Message-ID: <1329232814.2293.7.camel@twins> Subject: Re: smp: Start up non-boot CPUs asynchronously From: Peter Zijlstra To: Arjan van de Ven Date: Tue, 14 Feb 2012 16:20:14 +0100 In-Reply-To: References: <20120130205444.22f5e26a@infradead.org> <20120131125232.GD4408@elte.hu> <20120131054155.371e8307@infradead.org> <20120131143130.GF13676@elte.hu> <20120131072216.1ce78e50@infradead.org> <20120131161207.GA18357@elte.hu> <20120131082439.575978c0@infradead.org> <4F3A1891.8060001@linux.vnet.ibm.com> <4F3A2DFB.5000209@linux.vnet.ibm.com> Content-Type: text/plain; charset="ISO-8859-1" Mime-Version: 1.0 Cc: Stephen Rothwell , mikey@neuling.org, gregkh@linuxfoundation.org, Ingo Molnar , linux-kernel@vger.kernel.org, Milton Miller , Srivatsa Vaddagiri , Linus Torvalds , Arjan van de Ven , "Srivatsa S. Bhat" , "H. Peter Anvin" , Thomas Gleixner , "Paul E. McKenney" , ppc-dev , Andrew Morton , Arjan van de Ven List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Tue, 2012-02-14 at 06:31 -0800, Arjan van de Ven wrote: >=20 > frankly, such code HAS to worry about cpus going online and offline > even today; the firmware, at least on X86, can start taking cores > offline/online once ACPI is initialized.... (as controlled by a data > center manager from outside the box, usually done based on thermal or > power conditions on a datacenter level). Now, no doubt that we have > bugs in this space, since this only happened very rarely before.=20 Which frankly is an utter piece of crap, that ACPI spec is total garbage and completely useless. You might have noticed that the ACPI code supporting that failure carries a big nacked-by from me. That's not to say we shouldn't try to fix hotplug, but bringing that ACPI nonsense to the table makes me care less, not more. I mean, really, that spec is broken, the support is worse. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760685Ab2BNPVr (ORCPT ); Tue, 14 Feb 2012 10:21:47 -0500 Received: from merlin.infradead.org ([205.233.59.134]:52516 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751236Ab2BNPVp convert rfc822-to-8bit (ORCPT ); Tue, 14 Feb 2012 10:21:45 -0500 Message-ID: <1329232814.2293.7.camel@twins> Subject: Re: smp: Start up non-boot CPUs asynchronously From: Peter Zijlstra To: Arjan van de Ven Cc: "Srivatsa S. Bhat" , Arjan van de Ven , Stephen Rothwell , mikey@neuling.org, "Paul E. McKenney" , gregkh@linuxfoundation.org, ppc-dev , linux-kernel@vger.kernel.org, Milton Miller , Srivatsa Vaddagiri , Thomas Gleixner , "H. Peter Anvin" , Andrew Morton , Linus Torvalds , Ingo Molnar , Arjan van de Ven Date: Tue, 14 Feb 2012 16:20:14 +0100 In-Reply-To: References: <20120130205444.22f5e26a@infradead.org> <20120131125232.GD4408@elte.hu> <20120131054155.371e8307@infradead.org> <20120131143130.GF13676@elte.hu> <20120131072216.1ce78e50@infradead.org> <20120131161207.GA18357@elte.hu> <20120131082439.575978c0@infradead.org> <4F3A1891.8060001@linux.vnet.ibm.com> <4F3A2DFB.5000209@linux.vnet.ibm.com> Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT X-Mailer: Evolution 3.2.2- Mime-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2012-02-14 at 06:31 -0800, Arjan van de Ven wrote: > > frankly, such code HAS to worry about cpus going online and offline > even today; the firmware, at least on X86, can start taking cores > offline/online once ACPI is initialized.... (as controlled by a data > center manager from outside the box, usually done based on thermal or > power conditions on a datacenter level). Now, no doubt that we have > bugs in this space, since this only happened very rarely before. Which frankly is an utter piece of crap, that ACPI spec is total garbage and completely useless. You might have noticed that the ACPI code supporting that failure carries a big nacked-by from me. That's not to say we shouldn't try to fix hotplug, but bringing that ACPI nonsense to the table makes me care less, not more. I mean, really, that spec is broken, the support is worse.