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=-2.3 required=3.0 tests=DKIM_SIGNED,DKIM_VALID, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS, USER_AGENT_MUTT autolearn=ham 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 B5C3BC43218 for ; Mon, 10 Jun 2019 21:26:26 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 7D4522082E for ; Mon, 10 Jun 2019 21:26:26 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=codeblueprint-co-uk.20150623.gappssmtp.com header.i=@codeblueprint-co-uk.20150623.gappssmtp.com header.b="uFpTPEcn" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S2390146AbfFJV0Z (ORCPT ); Mon, 10 Jun 2019 17:26:25 -0400 Received: from mail-wm1-f65.google.com ([209.85.128.65]:36805 "EHLO mail-wm1-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2390113AbfFJV0Y (ORCPT ); Mon, 10 Jun 2019 17:26:24 -0400 Received: by mail-wm1-f65.google.com with SMTP id u8so738330wmm.1 for ; Mon, 10 Jun 2019 14:26:23 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=codeblueprint-co-uk.20150623.gappssmtp.com; s=20150623; h=date:from:to:cc:subject:message-id:references:mime-version :content-disposition:in-reply-to:user-agent; bh=Mc4UMWOXRpniRfg/YA6rdVKg83xokm7a/gCcz99OO9Y=; b=uFpTPEcnhoeUfN6R5mS/lOy4jy1oPL6nbblkSqTL3TG6ZYKl4Pxl1LdjYlVGHE4QBh 7cobvOiTm4Dt7Qxl5Wsm3bfUlRnGLQy++U5KUM2lb2RdGwjgtEUOTXKcJ/qaAdwKdIn5 MigvAWwh4TWhqmQVABnzUWlYcGGL4jRKXiAUU115FuXk7VhXUn4aBBg251VqgU29hXob qbohthz5As7xrUJR2P6wDk6eviIerj1oIFJnSjgX+lq2S2z5rcKFPZ3mjNSFvqS0SJ+F YCawArjFy69uVVtkh42D83IbsOqKwQT10kjuIjbitXErUWL6s2RQ3PZCMCLO+9VCrpxR kv4Q== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:date:from:to:cc:subject:message-id:references :mime-version:content-disposition:in-reply-to:user-agent; bh=Mc4UMWOXRpniRfg/YA6rdVKg83xokm7a/gCcz99OO9Y=; b=XbrIlEjNJZCw3FtQU9ZCl5F9KE3Y6qmMjTrxsDjqKX0G7qe8Plb71Iq8SdHFXNKvdj 3nZP+3/gh7TnyM8a+BMKf28xEyR3w5sXMP9O579COaDUkxSXugkqYx63SFkiG3cmpNTb H7qyoOgE20TQoX4Rld6Qq/pCGKYny7Hb7fpi0epK4gboJLu+MNzmq9B/8tgzMMTZHzeX vaxQ74N5UqpQzfg938j5HkuwBORcnY/cZU6DNcFJ9RbqIhhsjTi2u4x/taq3/vPoIZ5l jkGBP0+8x51JZBoXi8rIkOWSRAN9UM0XiEB5qv3FbBhommK7+nBjA2RydPUKYjtwZgji UPqw== X-Gm-Message-State: APjAAAU1sxvTFPtymLwAER2o6quS7zYnTRQzrcGAeApqoc0dlQOfPPme A/ota2GvprovTnqeRx9BendpIX0O45k= X-Google-Smtp-Source: APXvYqyeuGvHOcZX+1y8oIDrqD5qVR9OcS4PL/eW8U0fC1mdpO/PrNdlwlgxUsIg+qiq3RWx1Xo/GA== X-Received: by 2002:a1c:108:: with SMTP id 8mr13793453wmb.159.1560201982401; Mon, 10 Jun 2019 14:26:22 -0700 (PDT) Received: from localhost ([94.1.151.203]) by smtp.gmail.com with ESMTPSA id r5sm22420399wrg.10.2019.06.10.14.26.21 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Mon, 10 Jun 2019 14:26:21 -0700 (PDT) Date: Mon, 10 Jun 2019 22:26:20 +0100 From: Matt Fleming To: Peter Zijlstra Cc: linux-kernel@vger.kernel.org, "Suthikulpanit, Suravee" , Mel Gorman , "Lendacky, Thomas" , Borislav Petkov Subject: Re: [PATCH] sched/topology: Improve load balancing on AMD EPYC Message-ID: <20190610212620.GA4772@codeblueprint.co.uk> References: <20190605155922.17153-1-matt@codeblueprint.co.uk> <20190605180035.GA3402@hirez.programming.kicks-ass.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20190605180035.GA3402@hirez.programming.kicks-ass.net> User-Agent: Mutt/1.5.24 (2015-08-30) Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 05 Jun, at 08:00:35PM, Peter Zijlstra wrote: > > And then we had two magic values :/ > > Should we not 'fix' RECLAIM_DISTANCE for EPYC or something? Because > surely, if we want to load-balance agressively over 30, then so too > should we do node_reclaim() I'm thikning. Yeah we can fix it just for EPYC, Mel suggested that approach originally. Suravee, Tom, what's the best way to detect these EPYC machines that need to override RECLAIM_DISTANCE?