From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Zijlstra Subject: Re: [PATCH] sched: provide scheduler_ipi() callback in response to smp_send_reschedule() Date: Mon, 17 Jan 2011 21:43:03 +0100 Message-ID: <1295296983.30950.369.camel@laptop> References: <1295262433.30950.53.camel@laptop> <1295296310.2148.29.camel@pasglop> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1295296310.2148.29.camel@pasglop> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: uclinux-dist-devel-bounces-ZG0+EudsQA8dtHy/vicBwGD2FQJk+8+b@public.gmane.org Errors-To: uclinux-dist-devel-bounces-ZG0+EudsQA8dtHy/vicBwGD2FQJk+8+b@public.gmane.org To: Benjamin Herrenschmidt Cc: linux-m32r-ja-rQhvJZKUsGBRYuoOT4C5/9i2O/JbrIOy@public.gmane.org, linux-mips-6z/3iImG2C8G8FEW9MqTrA@public.gmane.org, linux-m32r-rQhvJZKUsGBRYuoOT4C5/9i2O/JbrIOy@public.gmane.org, linux-ia64-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-sh-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Heiko Carstens , Howells , Paul Mackerras , "H. Peter Anvin" , sparclinux-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Linux-Arch , linux-s390-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Jesper Nilsson , Mikael-ZG0+EudsQA8dtHy/vicBwGD2FQJk+8+b@public.gmane.org, Russell King , Takata , x86-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org, "James E.J. Bottomley" , virtualization-qjLDD68F18O7TbgM5vRIOg@public.gmane.org, Ingo Molnar , xen-devel-GuqFBffKawuULHF6PoxzQEEOCMrvLtNR@public.gmane.org, Chris-ZG0+EudsQA8dtHy/vicBwGD2FQJk+8+b@public.gmane.org, Matt Turner , uclinux-dist-devel-ZG0+EudsQA8dtHy/vicBwGD2FQJk+8+b@public.gmane.org, Fen On Tue, 2011-01-18 at 07:31 +1100, Benjamin Herrenschmidt wrote: > > Beware of false positive, I've used "fake" reschedule IPIs in the past > for other things (like kicking a CPU out of sleep state for unrelated > reasons). Nothing that I know that is upstream today but some of that > might come back. I'd like to avoid having to add an atomic to know if > it's a real reschedule, will the scheduler be smart enough to not bother > with false positives ? Yes it can deal with that, some will be for reschedules, some will be for ttwu tail ends and x86 too uses this ipi for a few random other things like kicking kvm out of guest context.. From mboxrd@z Thu Jan 1 00:00:00 1970 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 17 Jan 2011 21:43:39 +0100 (CET) Received: from casper.infradead.org ([85.118.1.10]:42555 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S1493581Ab1AQUng convert rfc822-to-8bit (ORCPT ); Mon, 17 Jan 2011 21:43:36 +0100 Received: from j77219.upc-j.chello.nl ([24.132.77.219] helo=laptop) by casper.infradead.org with esmtpsa (Exim 4.72 #1 (Red Hat Linux)) id 1PevuB-0000p9-1r; Mon, 17 Jan 2011 20:42:23 +0000 Received: by laptop (Postfix, from userid 1000) id A8C1E1033A261; Mon, 17 Jan 2011 21:43:03 +0100 (CET) Subject: Re: [PATCH] sched: provide scheduler_ipi() callback in response to smp_send_reschedule() From: Peter Zijlstra To: Benjamin Herrenschmidt Cc: Richard Henderson , Ivan Kokshaysky , Matt Turner , Russell King , Mike Frysinger , Mikael Starvik , Jesper Nilsson , Tony Luck , Fenghua Yu , Hirokazu Takata , Ralf Baechle , David Howells , Koichi Yasutake , Kyle McMartin , Helge Deller , "James E.J. Bottomley" , Paul Mackerras , Martin Schwidefsky , Heiko Carstens , linux390@de.ibm.com, Paul Mundt , "David S. Miller" , Chris Metcalf , Jeff Dike , Thomas Gleixner , Ingo Molnar , "H. Peter Anvin" , x86@kernel.org, Jeremy Fitzhardinge , Konrad Rzeszutek Wilk , linux-alpha@vger.kernel.org, linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org, uclinux-dist-devel@blackfin.uclinux.org, linux-cris-kernel@axis.com, linux-ia64@vger.kernel.org, linux-m32r@ml.linux-m32r.org, linux-m32r-ja@ml.linux-m32r.org, linux-mips@linux-mips.org, linux-am33-list@redhat.com, linux-parisc@vger.kernel.org, linuxppc-dev@lists.ozlabs.org, linux-s390@vger.kernel.org, linux-sh@vger.kernel.org, sparclinux@vger.kernel.org, user-mode-linux-devel@lists.sourceforge.net, user-mode-linux-user@lists.sourceforge.net, xen-devel@lists.xensource.com, virtualization@lists.osdl.org, Linux-Arch In-Reply-To: <1295296310.2148.29.camel@pasglop> References: <1295262433.30950.53.camel@laptop> <1295296310.2148.29.camel@pasglop> Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 8BIT Date: Mon, 17 Jan 2011 21:43:03 +0100 Message-ID: <1295296983.30950.369.camel@laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.30.3 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 28949 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: peterz@infradead.org Precedence: bulk X-list: linux-mips On Tue, 2011-01-18 at 07:31 +1100, Benjamin Herrenschmidt wrote: > > Beware of false positive, I've used "fake" reschedule IPIs in the past > for other things (like kicking a CPU out of sleep state for unrelated > reasons). Nothing that I know that is upstream today but some of that > might come back. I'd like to avoid having to add an atomic to know if > it's a real reschedule, will the scheduler be smart enough to not bother > with false positives ? Yes it can deal with that, some will be for reschedules, some will be for ttwu tail ends and x86 too uses this ipi for a few random other things like kicking kvm out of guest context.. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from casper.infradead.org (casper.infradead.org [85.118.1.10]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTPS id DEFADB7151 for ; Tue, 18 Jan 2011 07:43:41 +1100 (EST) Subject: Re: [PATCH] sched: provide scheduler_ipi() callback in response to smp_send_reschedule() From: Peter Zijlstra To: Benjamin Herrenschmidt In-Reply-To: <1295296310.2148.29.camel@pasglop> References: <1295262433.30950.53.camel@laptop> <1295296310.2148.29.camel@pasglop> Content-Type: text/plain; charset="UTF-8" Date: Mon, 17 Jan 2011 21:43:03 +0100 Message-ID: <1295296983.30950.369.camel@laptop> Mime-Version: 1.0 Cc: linux-m32r-ja@ml.linux-m32r.org, linux-mips@linux-mips.org, linux-ia64@vger.kernel.org, linux-sh@vger.kernel.org, "H. Peter Anvin" , Heiko Carstens , Paul Mackerras , Helge Deller , sparclinux@vger.kernel.org, Linux-Arch , linux-s390@vger.kernel.org, Jesper Nilsson , Jeremy Fitzhardinge , Russell King , Hirokazu Takata , x86@kernel.org, "James E.J. Bottomley" , virtualization@lists.osdl.org, Ingo Molnar , Matt Turner , Fenghua Yu , Mike Frysinger , user-mode-linux-devel@lists.sourceforge.net, Konrad Rzeszutek Wilk , Jeff Dike , Chris Metcalf , xen-devel@lists.xensource.com, Mikael Starvik , linux-m32r@ml.linux-m32r.org, Ivan Kokshaysky , user-mode-linux-user@lists.sourceforge.net, uclinux-dist-devel@blackfin.uclinux.org, Thomas Gleixner , linux-arm-kernel@lists.infradead.org, Richard Henderson , Tony Luck , linux-parisc@vger.kernel.org, linux-cris-kernel@axis.com, linux-am33-list@redhat.com, linux-kernel@vger.kernel.org, Ralf Baechle , Kyle McMartin , Paul Mundt , linux-alpha@vger.kernel.org, Martin Schwidefsky , linux390@de.ibm.com, Koichi Yasutake , linuxppc-dev@lists.ozlabs.org, "David S. Miller" List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Tue, 2011-01-18 at 07:31 +1100, Benjamin Herrenschmidt wrote: >=20 > Beware of false positive, I've used "fake" reschedule IPIs in the past > for other things (like kicking a CPU out of sleep state for unrelated > reasons). Nothing that I know that is upstream today but some of that > might come back. I'd like to avoid having to add an atomic to know if > it's a real reschedule, will the scheduler be smart enough to not bother > with false positives ?=20 Yes it can deal with that, some will be for reschedules, some will be for ttwu tail ends and x86 too uses this ipi for a few random other things like kicking kvm out of guest context.. From mboxrd@z Thu Jan 1 00:00:00 1970 From: peterz@infradead.org (Peter Zijlstra) Date: Mon, 17 Jan 2011 21:43:03 +0100 Subject: [PATCH] sched: provide scheduler_ipi() callback in response to smp_send_reschedule() In-Reply-To: <1295296310.2148.29.camel@pasglop> References: <1295262433.30950.53.camel@laptop> <1295296310.2148.29.camel@pasglop> Message-ID: <1295296983.30950.369.camel@laptop> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Tue, 2011-01-18 at 07:31 +1100, Benjamin Herrenschmidt wrote: > > Beware of false positive, I've used "fake" reschedule IPIs in the past > for other things (like kicking a CPU out of sleep state for unrelated > reasons). Nothing that I know that is upstream today but some of that > might come back. I'd like to avoid having to add an atomic to know if > it's a real reschedule, will the scheduler be smart enough to not bother > with false positives ? Yes it can deal with that, some will be for reschedules, some will be for ttwu tail ends and x86 too uses this ipi for a few random other things like kicking kvm out of guest context..