Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753944AbdIGGQE convert rfc822-to-8bit (ORCPT ); Thu, 7 Sep 2017 02:16:04 -0400 Received: from mout.gmx.net ([212.227.15.18]:50520 "EHLO mout.gmx.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752783AbdIGGQD (ORCPT ); Thu, 7 Sep 2017 02:16:03 -0400 Message-ID: <1504764929.6355.5.camel@gmx.de> Subject: Re: Abysmal scheduler performance in Linus' tree? From: Mike Galbraith To: Peter Zijlstra , Andy Lutomirski Cc: Andy Lutomirski , Ingo Molnar , "linux-kernel@vger.kernel.org" Date: Thu, 07 Sep 2017 08:15:29 +0200 In-Reply-To: <20170906161422.kvqe5y6b3ng6ouyu@hirez.programming.kicks-ass.net> References: <20170906082520.xgvo3hewje7jvdyo@hirez.programming.kicks-ass.net> <877A43A3-AC8F-4D7C-88E4-8E3D36B1DAFA@amacapital.net> <20170906090333.7v627vdyvhe3x2cs@hirez.programming.kicks-ass.net> <20170906161422.kvqe5y6b3ng6ouyu@hirez.programming.kicks-ass.net> Content-Type: text/plain; charset="UTF-8" X-Mailer: Evolution 3.20.5 Mime-Version: 1.0 Content-Transfer-Encoding: 8BIT X-Provags-ID: V03:K0:NXoBn+QvsdpS1VP06pSC/GDbvv/qkWCuwj5sw37FuLyDjj8hfO5 7dll3xTZuMnXxd1UOXaEJc2rKHbVxYc6A/HaXyb42hljX92ai0PgC5XutAdKSfb+sUVgWAN QwiZs4yrQrQ5n39x9o8M8d+2HaW96fErE2kWTMAdQfv0FndaC4gqxDqMLqXqzhu0eq9n/YD 6L+xKaFDZ9GT+ZKbQ2fLg== X-UI-Out-Filterresults: notjunk:1;V01:K0:xq8h/Rsl2E8=:qqD2DhxLCWGPzx4276b8uh zipA1i63O13GCpwYWoqtPKf1H4VZQbeHFozRLi9OTnCOhZkdqHW9rg7MIOukmTXyMJj8K19MN rMaAaamvlMbYA5zukVpwykFaf19EDGeriHYTM9tjig95nuD6WK+Yf+XkFQhD8c8AfkIPifEb3 vIqlpveWXtYU41SMO996r1GVyxKjJEBAXD41+/Ef6SJeapJQ/HDNGkQX58MMb1hZ42xrEAm6B Y8JXkepSP3LFQXj/givdnL/2/rnrUGZs1rrcEuDwZjLTyBuC0cD9RXHWeyNvd9hCJb2h3MHHr InRmRbnB1jBquDUZY+S6DlcdNoDwqgilh7aOn6SEZK/EZeeExDsdCT7TgjxbBFV/sI/RLa06a aVBQOUpt2r7LWSRMPkb8y7yPKlzam9QoULskmQ4IsB6EVPb6e2ASWjZsr+Kt3joKh0Zf4NA4G 3Q+16Ca1h8RI5zcI220Q4q3oOO3/LBPJO00hLvkSbqFMjA0vY01hW8AnliDup1nAOBoX4Rbj1 jGpsXrxHLPLdB0EGC7NYUz4g/nKt2vTIGt7ARsPjIZqip12gbaQaM/ylOkcpysPgNjNsz/s1v VFh8TxByIiOkWUd8LqEvBPB3sNlcuf73OX4C3PVt3MrjbREuYydFu8V406Az1ztTekRYa/gec 9GXyEkBSD6PJuRvFRLLoD4BuovjjJg4ZVaWgzPY3a8rGdQVeVb4nuN5anSM2jXzCNLNnQB/Rn kEMRQrp5EXbSbUdLZeQ8CAlvrHBDzvFyS8FDpTNESlIvRdSR9JhK7tIWakWuoF1AddJk7wops vnJY/Uqd5QidiFED9okqCZOH/uwlCxxsFRXiAR0npw/lCyWFsM= Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 799 Lines: 20 On Wed, 2017-09-06 at 18:14 +0200, Peter Zijlstra wrote: > On Wed, Sep 06, 2017 at 11:03:33AM +0200, Peter Zijlstra wrote: > > On Wed, Sep 06, 2017 at 01:59:14AM -0700, Andy Lutomirski wrote: > > > > Any specifics on your setup? > > > > > > On further fiddling, I only see this after a suspend and resume cycle. > > > > Ah, ok. That's not something I otherwise test. Lets see if I can force > > this brick of mine through a suspend-resume cycle :-) > > I'd be suspicious of these here commits: > > 77d1dfda0e79 ("sched/topology, cpuset: Avoid spurious/wrong domain rebuilds") Seems to be that one.  After suspend/resume, cpu7 (i4790+SMT) is online with NULL domain.  offline/online it, it grows proper domains. -Mike (on vacation, about to go explore that unfamiliar 'outside' place)