From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christian Borntraeger Subject: Re: [GIT PULL v2 1/5] processor.h: introduce cpu_relax_yield Date: Tue, 15 Nov 2016 14:19:53 +0100 Message-ID: <7f7850e4-c7bb-9cc1-2d65-a1555e97988a@de.ibm.com> References: <1477386195-32736-1-git-send-email-borntraeger@de.ibm.com> <1477386195-32736-2-git-send-email-borntraeger@de.ibm.com> <20161115123029.GT1041@n2100.armlinux.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20161115123029.GT1041@n2100.armlinux.org.uk> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: virtualization-bounces@lists.linux-foundation.org Errors-To: virtualization-bounces@lists.linux-foundation.org To: Russell King - ARM Linux Cc: linux-arch@vger.kernel.org, linux-s390 , kvm@vger.kernel.org, Peter Zijlstra , Will Deacon , x86@kernel.org, Heiko Carstens , linux-kernel@vger.kernel.org, Nicholas Piggin , virtualization@lists.linux-foundation.org, sparclinux@vger.kernel.org, Noam Camus , Catalin Marinas , Martin Schwidefsky , xen-devel@lists.xenproject.org, linuxppc-dev@lists.ozlabs.org, Ingo Molnar List-Id: linux-arch.vger.kernel.org On 11/15/2016 01:30 PM, Russell King - ARM Linux wrote: > On Tue, Oct 25, 2016 at 11:03:11AM +0200, Christian Borntraeger wrote: >> For spinning loops people do often use barrier() or cpu_relax(). >> For most architectures cpu_relax and barrier are the same, but on >> some architectures cpu_relax can add some latency. >> For example on power,sparc64 and arc, cpu_relax can shift the CPU >> towards other hardware threads in an SMT environment. >> On s390 cpu_relax does even more, it uses an hypercall to the >> hypervisor to give up the timeslice. >> In contrast to the SMT yielding this can result in larger latencies. >> In some places this latency is unwanted, so another variant >> "cpu_relax_lowlatency" was introduced. Before this is used in more >> and more places, lets revert the logic and provide a cpu_relax_yield >> that can be called in places where yielding is more important than >> latency. By default this is the same as cpu_relax on all architectures. > > Rather than having to update all these architectures in this way, can't > we put in some linux/*.h header something like: > > #ifndef cpu_relax_yield > #define cpu_relax_yield() cpu_relax() > #endif > > so only those architectures that need to do something need to be > modified? These patches are part of linux-next since a month or so, changing that would invalidate all the next testing. If people want that, I can certainly do that, though. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx0b-001b2d01.pphosted.com ([148.163.158.5]:57396 "EHLO mx0a-001b2d01.pphosted.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S934979AbcKONU0 (ORCPT ); Tue, 15 Nov 2016 08:20:26 -0500 Received: from pps.filterd (m0098417.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.16.0.17/8.16.0.17) with SMTP id uAFDJ8jU011969 for ; Tue, 15 Nov 2016 08:20:25 -0500 Received: from e06smtp07.uk.ibm.com (e06smtp07.uk.ibm.com [195.75.94.103]) by mx0a-001b2d01.pphosted.com with ESMTP id 26qwuqr1x5-1 (version=TLSv1.2 cipher=AES256-SHA bits=256 verify=NOT) for ; Tue, 15 Nov 2016 08:20:25 -0500 Received: from localhost by e06smtp07.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Tue, 15 Nov 2016 13:20:23 -0000 Subject: Re: [GIT PULL v2 1/5] processor.h: introduce cpu_relax_yield References: <1477386195-32736-1-git-send-email-borntraeger@de.ibm.com> <1477386195-32736-2-git-send-email-borntraeger@de.ibm.com> <20161115123029.GT1041@n2100.armlinux.org.uk> From: Christian Borntraeger Date: Tue, 15 Nov 2016 14:19:53 +0100 MIME-Version: 1.0 In-Reply-To: <20161115123029.GT1041@n2100.armlinux.org.uk> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Message-ID: <7f7850e4-c7bb-9cc1-2d65-a1555e97988a@de.ibm.com> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Russell King - ARM Linux Cc: Peter Zijlstra , Ingo Molnar , Nicholas Piggin , linux-kernel@vger.kernel.org, linux-s390 , linux-arch@vger.kernel.org, linuxppc-dev@lists.ozlabs.org, Heiko Carstens , Martin Schwidefsky , Noam Camus , sparclinux@vger.kernel.org, x86@kernel.org, Will Deacon , Catalin Marinas , virtualization@lists.linux-foundation.org, xen-devel@lists.xenproject.org, kvm@vger.kernel.org Message-ID: <20161115131953.EnXc0J9tyYx9AKAhdss3qLIo3CwqqNUor-Gp_ye5KaI@z> On 11/15/2016 01:30 PM, Russell King - ARM Linux wrote: > On Tue, Oct 25, 2016 at 11:03:11AM +0200, Christian Borntraeger wrote: >> For spinning loops people do often use barrier() or cpu_relax(). >> For most architectures cpu_relax and barrier are the same, but on >> some architectures cpu_relax can add some latency. >> For example on power,sparc64 and arc, cpu_relax can shift the CPU >> towards other hardware threads in an SMT environment. >> On s390 cpu_relax does even more, it uses an hypercall to the >> hypervisor to give up the timeslice. >> In contrast to the SMT yielding this can result in larger latencies. >> In some places this latency is unwanted, so another variant >> "cpu_relax_lowlatency" was introduced. Before this is used in more >> and more places, lets revert the logic and provide a cpu_relax_yield >> that can be called in places where yielding is more important than >> latency. By default this is the same as cpu_relax on all architectures. > > Rather than having to update all these architectures in this way, can't > we put in some linux/*.h header something like: > > #ifndef cpu_relax_yield > #define cpu_relax_yield() cpu_relax() > #endif > > so only those architectures that need to do something need to be > modified? These patches are part of linux-next since a month or so, changing that would invalidate all the next testing. If people want that, I can certainly do that, though.