From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752758AbbBXOnn (ORCPT ); Tue, 24 Feb 2015 09:43:43 -0500 Received: from mx1.redhat.com ([209.132.183.28]:33388 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751498AbbBXOnm (ORCPT ); Tue, 24 Feb 2015 09:43:42 -0500 Message-ID: <54EC8DFC.3030005@redhat.com> Date: Tue, 24 Feb 2015 09:43:08 -0500 From: Rik van Riel User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.3.0 MIME-Version: 1.0 To: Andy Lutomirski , "Maciej W. Rozycki" CC: Borislav Petkov , Ingo Molnar , Oleg Nesterov , X86 ML , "linux-kernel@vger.kernel.org" , Linus Torvalds Subject: Re: [RFC PATCH] x86, fpu: Use eagerfpu by default on all CPUs References: <20150221093150.GA27841@gmail.com> <20150221163840.GA32073@pd.tnic> <20150221172914.GB32073@pd.tnic> <54EB99E8.2060500@redhat.com> In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 02/23/2015 09:31 PM, Andy Lutomirski wrote: > On Mon, Feb 23, 2015 at 6:14 PM, Maciej W. Rozycki > wrote: >> That's an interesting case too, although not necessarily related. >> If you say that we always save the FP context eagerly for the >> purpose of process migration, then sure, that invalidates any >> benefit we'd have from letting the x87 proceed. >> >> However I can see different ways to address this case avoiding >> the need of eager FP context saving or an IPI: >> >> 1. We could bind any currently suspended process with an unsaved >> FP context to the CPU it last executed on. > > This would be insane. The task would only be bound to the CPU as long as nothing else ran that saved the FPU state to memory. This means the task would be bound to the CPU while the CPU is idle, or running a kernel thread. When another user space thread is about to run, the FPU state would be saved. This sounds slightly less insane already. Of course, once you figure in CPU power saving and CPU hot plug, the level of insanity is far beyond what you seem to imply above... In other words, almost certainly not worth doing :) - -- All rights reversed -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEcBAEBAgAGBQJU7I38AAoJEM553pKExN6DhKEIAKauUGBx/1sTshdYYZ1aBlLx xY7afNUjs8PIxjCbdcwrujbtNa9CFgDlRR6TegvmzQA3prXu/0XZ3vas3O/lD2lC ks8p3RBzIw4dECxZoCvTQ+VrULk07+LCI+AUNKSm/pNlBCSWeeo2nKqoTREh3oHU EWzJxn5aEfIA4vZQAnFP5TwkCwR2ob5COGx/I9l54brHEwhEqiRFrPwrIP2WJerx Lc1Wkmv2PtTN/oQkXOCVKN0hVab//eVnkUiTsY1TnfCQsZSbEMWgq6aqXlb/lhUs hhpNToBVlWF3LsCnGm6LfCrgX+VSBY9LQpYfaY1ltEmxE+nOplbI+JHQG4Yqgag= =ah5u -----END PGP SIGNATURE-----