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 Received: from bombadil.infradead.org (bombadil.infradead.org [198.137.202.133]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id D5CFBC433FE for ; Fri, 20 May 2022 15:34:36 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20210309; h=Sender: Content-Transfer-Encoding:Content-Type:List-Subscribe:List-Help:List-Post: List-Archive:List-Unsubscribe:List-Id:In-Reply-To:MIME-Version:References: Message-ID:Subject:Cc:To:From:Date:Reply-To:Content-ID:Content-Description: Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID: List-Owner; bh=HOrqjafaOifI1KhLZkd07kBtyZrgTtIgCiFJibuHM1Q=; b=zbsdqJxWumjYfG JbPdKgOMydkL5NtHlNNxkB7XjVJsFjtKnKR52K/4W/m7j7yqQifq0btmIwUDrcVIUASJNPbVjHOl+ TYvIOEGmebz7pg+qfy69NkkM4mVKg9wC5nKcsdcmvuUFsm61SAQudtpNK14LfXhLm426+I4b6y0D7 dTdrTeKKl3D4vB0uLfJuodWDzJKD26wR9VJGHjRAVI79+VtWP/LnhANZkhr0X1DqME38KnmnL+Hju thWNAqJYlY0s/4dWYGJJBjVKvMA4YQXc4mYF5fsBuSLmzly34R9Q74Zeq/oYkobeJS6xbbgbPNkVb 0qqxhZ50zFRPD2W6ZkPA==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.94.2 #2 (Red Hat Linux)) id 1ns4d3-00DEpX-13; Fri, 20 May 2022 15:33:25 +0000 Received: from foss.arm.com ([217.140.110.172]) by bombadil.infradead.org with esmtp (Exim 4.94.2 #2 (Red Hat Linux)) id 1ns4cx-00DEnS-1i; Fri, 20 May 2022 15:33:21 +0000 Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 0C0E51477; Fri, 20 May 2022 08:33:18 -0700 (PDT) Received: from bogus (unknown [10.57.66.157]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id AADFA3F73D; Fri, 20 May 2022 08:33:15 -0700 (PDT) Date: Fri, 20 May 2022 16:33:09 +0100 From: Sudeep Holla To: Ionela Voinescu Cc: Atish Patra , linux-kernel@vger.kernel.org, Atish Patra , Vincent Guittot , Sudeep Holla , Morten Rasmussen , Dietmar Eggemann , Qing Wang , linux-arm-kernel@lists.infradead.org, linux-riscv@lists.infradead.org, Rob Herring Subject: Re: [PATCH v2 0/8] arch_topology: Updates to add socket support and fix cluster ids Message-ID: <20220520153309.rjtticnfxidz4x7p@bogus> References: <20220518093325.2070336-1-sudeep.holla@arm.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20220520_083319_236682_EE8D2377 X-CRM114-Status: GOOD ( 40.53 ) X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+linux-arm-kernel=archiver.kernel.org@lists.infradead.org On Thu, May 19, 2022 at 05:32:49PM +0100, Ionela Voinescu wrote: > Hi Sudeep, > > On Wednesday 18 May 2022 at 10:33:17 (+0100), Sudeep Holla wrote: > > Hi All, > > > > This series intends to fix some discrepancies we have in the CPU topology > > parsing from the device tree /cpu-map node. Also this diverges from the > > behaviour on a ACPI enabled platform. The expectation is that both DT > > and ACPI enabled systems must present consistent view of the CPU topology. > > > > Currently we assign generated cluster count as the physical package identifier > > for each CPU which is wrong. The device tree bindings for CPU topology supports > > sockets to infer the socket or physical package identifier for a given CPU. > > Also we don't check if all the cores/threads belong to the same cluster before > > updating their sibling masks which is fine as we don't set the cluster id yet. > > > > These changes also assigns the cluster identifier as parsed from the device tree > > cluster nodes within /cpu-map without support for nesting of the clusters. > > Finally, it also add support for socket nodes in /cpu-map. With this the > > parsing of exact same information from ACPI PPTT and /cpu-map DT node > > aligns well. > > > > The only exception is that the last level cache id information can be > > inferred from the same ACPI PPTT while we need to parse CPU cache nodes > > in the device tree. > > > > P.S: I have not cc-ed Greg and Rafael so that all the users of arch_topology > > agree with the changes first before we include them. > > > > v1[1]->v2: > > - Updated ID validity check include all non-negative value > > - Added support to get the device node for the CPU's last level cache > > - Added support to build llc_sibling on DT platforms > > > > [1] https://lore.kernel.org/lkml/20220513095559.1034633-1-sudeep.holla@arm.com > > > > Sudeep Holla (8): > > arch_topology: Don't set cluster identifier as physical package identifier > > arch_topology: Set thread sibling cpumask only within the cluster > > arch_topology: Set cluster identifier in each core/thread from /cpu-map > > arch_topology: Add support for parsing sockets in /cpu-map > > arch_topology: Check for non-negative value rather than -1 for IDs validity > > arch_topology: Avoid parsing through all the CPUs once a outlier CPU is found > > of: base: add support to get the device node for the CPU's last level cache > > arch_topology: Add support to build llc_sibling on DT platforms > > > > Just a recommendation for patch-set structure: it would be best to have > the following sequence to maintain the same scheduler topology and > behaviour when partially applying the set (currently testing this on Juno, > but should be the case for other platforms as well): > > 2/8 arch_topology: Set thread sibling cpumask only within the cluster > 5/8 arch_topology: Check for non-negative value rather than -1 for IDs validity > 6/8 arch_topology: Avoid parsing through all the CPUs once a outlier CPU is found > > --> these are only preparation/cleanup patches and don't affect current > functionality > Agreed. It was in my TODO list but I just to post this v2 for some reason. I knew the patches were more in the order of my thoughts on what all needs to be done and the order I added them. I knew it would result in issue with kernel bisection. > 7/8 of: base: add support to get the device node for the CPU's last level cache > 8/8 arch_topology: Add support to build llc_sibling on DT platforms > > --> these will populate llc siblings but this list will be equal to > core siblings (based on package_id) so nothing changes in the scheduler > topology. Even if CONFIG_SCHED_CLUSTER=y, we still have cluster_id=-1 so > nothing will change in that case either, for the patches so far. > Correct, I had worked out this much detail. > 1/8 arch_topology: Don't set cluster identifier as physical package identifier > > --> 1/8 is the trouble maker if it's the first patch as it will result > in having all CPUs in core_siblings so the topology will be flattened to > just an MC level for a typical b.L system like Juno. But if you add it after > all of the above patches, the llc_siblings will contribute to create the same > MC and DIE levels we expect. > > 3/8 arch_topology: Set cluster identifier in each core/thread from /cpu-map > 4/8 arch_topology: Add support for parsing sockets in /cpu-map > > --> Here 3/8 will start creating complications when having clusters in > DT and we have CONFIG_SCHED_CLUSTER=y. But I'll detail this in a reply > to that patch. For CONFIG_SCHED_CLUSTER=n the topology and scheduler > behaviour should be the same as before this set. > Thanks for the ordering of the last 3. I wanted to work out, but you need to have done more work than me on that and thanks for saving my time. Much appreciated. > Hope it helps, Yes definitely, thanks again. -- Regards, Sudeep _______________________________________________ linux-arm-kernel mailing list linux-arm-kernel@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-arm-kernel