From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-5.2 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS, USER_AGENT_SANE_1 autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 6EA2FC433DF for ; Mon, 12 Oct 2020 11:31:16 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 250912080D for ; Mon, 12 Oct 2020 11:31:16 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S2388018AbgJLLbO (ORCPT ); Mon, 12 Oct 2020 07:31:14 -0400 Received: from mail2-relais-roc.national.inria.fr ([192.134.164.83]:23655 "EHLO mail2-relais-roc.national.inria.fr" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2387743AbgJLLbO (ORCPT ); Mon, 12 Oct 2020 07:31:14 -0400 X-IronPort-AV: E=Sophos;i="5.77,366,1596492000"; d="scan'208";a="472127745" Received: from abo-173-121-68.mrs.modulonet.fr (HELO hadrien) ([85.68.121.173]) by mail2-relais-roc.national.inria.fr with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 12 Oct 2020 13:31:12 +0200 Date: Mon, 12 Oct 2020 13:31:12 +0200 (CEST) From: Julia Lawall X-X-Sender: jll@hadrien To: Peter Zijlstra cc: Julia Lawall , Valentin Schneider , Ingo Molnar , Juri Lelli , Vincent Guittot , linux-kernel@vger.kernel.org, Gilles Muller , "Rafael J. Wysocki" Subject: Re: SD_LOAD_BALANCE In-Reply-To: <20201012112124.GT2628@hirez.programming.kicks-ass.net> Message-ID: References: <20201012112124.GT2628@hirez.programming.kicks-ass.net> User-Agent: Alpine 2.22 (DEB 394 2020-01-19) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 12 Oct 2020, Peter Zijlstra wrote: > On Sat, Oct 10, 2020 at 06:14:23PM +0200, Julia Lawall wrote: > > Prior to v5.8 on my machine this was a rare event, because there were not > > many of these background processes. But in v5.8, the default governor for > > Intel machines without the HWP feature was changed from intel_pstate to > > intel_cpufreq. The use of intel_cpufreq triggers very frequent kworkers on > > all cores, which makes it much more likely that cores that are currently > > idle, and are overall not at all overloaded, will have a higher load > > average even with the waking thread deducted, than the core managing the > > wakeup of the threads. > > Rafael, any idea what those kworkers are for, and can we get rid of > them? They execute the function intel_cpufreq_target defined in drivers/cpufreq/intel_pstate.c julia