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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 51FA3C32771 for ; Tue, 27 Sep 2022 02:50:52 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230163AbiI0Cuv (ORCPT ); Mon, 26 Sep 2022 22:50:51 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:58834 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230191AbiI0Cs4 (ORCPT ); Mon, 26 Sep 2022 22:48:56 -0400 Received: from ams.source.kernel.org (ams.source.kernel.org [IPv6:2604:1380:4601:e00::1]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 2D828642DD for ; Mon, 26 Sep 2022 19:48:23 -0700 (PDT) Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ams.source.kernel.org (Postfix) with ESMTPS id DCE31B81910 for ; Tue, 27 Sep 2022 02:48:21 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 83AB9C433D6; Tue, 27 Sep 2022 02:48:20 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linux-foundation.org; s=korg; t=1664246900; bh=6+pDATDx/RTJ4Lmv1+6QTAw6G82pnyVho3CnZXdEZZY=; h=Date:To:From:Subject:From; b=iQRBRo88VZHhtWjnLOclJHWnDIT184XyvU7jfZdIKc8GLnrGPRXX+KzWdAi9Seoak EmBlq1YF01NrP0VGeCMGfPEjwSgEBq88MS9v/FgCuiFqYdqpBqrdUYbRU5JRLMZmAA 2Ml0HDw8XYxVY1katpXbJIxbBY7E7Tmr3Xnl+JAc= Date: Mon, 26 Sep 2022 19:48:19 -0700 To: mm-commits@vger.kernel.org, ying.huang@intel.com, weixugc@google.com, tim.c.chen@intel.com, sj@kernel.org, shy828301@gmail.com, mhocko@kernel.org, jvgediya.oss@gmail.com, Jonathan.Cameron@huawei.com, hesham.almatary@huawei.com, hannes@cmpxchg.org, dave@stgolabs.net, dave.hansen@intel.com, dan.j.williams@intel.com, bharata@amd.com, apopple@nvidia.com, aneesh.kumar@linux.ibm.com, akpm@linux-foundation.org From: Andrew Morton Subject: [merged mm-stable] mm-demotion-add-pg_data_t-member-to-track-node-memory-tier-details.patch removed from -mm tree Message-Id: <20220927024820.83AB9C433D6@smtp.kernel.org> Precedence: bulk Reply-To: linux-kernel@vger.kernel.org List-ID: X-Mailing-List: mm-commits@vger.kernel.org The quilt patch titled Subject: mm/demotion: add pg_data_t member to track node memory tier details has been removed from the -mm tree. Its filename was mm-demotion-add-pg_data_t-member-to-track-node-memory-tier-details.patch This patch was dropped because it was merged into the mm-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm ------------------------------------------------------ From: "Aneesh Kumar K.V" Subject: mm/demotion: add pg_data_t member to track node memory tier details Date: Thu, 18 Aug 2022 18:40:38 +0530 Also update different helpes to use NODE_DATA()->memtier. Since node specific memtier can change based on the reassignment of NUMA node to a different memory tiers, accessing NODE_DATA()->memtier needs to happen under an rcu read lock or memory_tier_lock. Link: https://lkml.kernel.org/r/20220818131042.113280-7-aneesh.kumar@linux.ibm.com Signed-off-by: Aneesh Kumar K.V Reviewed-by: "Huang, Ying" Acked-by: Wei Xu Cc: Alistair Popple Cc: Bharata B Rao Cc: Dan Williams Cc: Dave Hansen Cc: Davidlohr Bueso Cc: Hesham Almatary Cc: Jagdish Gediya Cc: Johannes Weiner Cc: Jonathan Cameron Cc: Michal Hocko Cc: Tim Chen Cc: Yang Shi Cc: SeongJae Park Signed-off-by: Andrew Morton --- include/linux/mmzone.h | 3 ++ mm/memory-tiers.c | 40 ++++++++++++++++++++++++++++++++++----- 2 files changed, 38 insertions(+), 5 deletions(-) --- a/include/linux/mmzone.h~mm-demotion-add-pg_data_t-member-to-track-node-memory-tier-details +++ a/include/linux/mmzone.h @@ -1246,6 +1246,9 @@ typedef struct pglist_data { /* Per-node vmstats */ struct per_cpu_nodestat __percpu *per_cpu_nodestats; atomic_long_t vm_stat[NR_VM_NODE_STAT_ITEMS]; +#ifdef CONFIG_NUMA + struct memory_tier __rcu *memtier; +#endif } pg_data_t; #define node_present_pages(nid) (NODE_DATA(nid)->node_present_pages) --- a/mm/memory-tiers.c~mm-demotion-add-pg_data_t-member-to-track-node-memory-tier-details +++ a/mm/memory-tiers.c @@ -4,6 +4,7 @@ #include #include #include +#include #include #include "internal.h" @@ -141,12 +142,18 @@ static struct memory_tier *find_create_m static struct memory_tier *__node_get_memory_tier(int node) { - struct memory_dev_type *memtype; + pg_data_t *pgdat; - memtype = node_memory_types[node]; - if (memtype && node_isset(node, memtype->nodes)) - return memtype->memtier; - return NULL; + pgdat = NODE_DATA(node); + if (!pgdat) + return NULL; + /* + * Since we hold memory_tier_lock, we can avoid + * RCU read locks when accessing the details. No + * parallel updates are possible here. + */ + return rcu_dereference_check(pgdat->memtier, + lockdep_is_held(&memory_tier_lock)); } #ifdef CONFIG_MIGRATION @@ -309,6 +316,8 @@ static struct memory_tier *set_node_memo { struct memory_tier *memtier; struct memory_dev_type *memtype; + pg_data_t *pgdat = NODE_DATA(node); + lockdep_assert_held_once(&memory_tier_lock); @@ -320,24 +329,45 @@ static struct memory_tier *set_node_memo memtype = node_memory_types[node].memtype; node_set(node, memtype->nodes); memtier = find_create_memory_tier(memtype); + if (!IS_ERR(memtier)) + rcu_assign_pointer(pgdat->memtier, memtier); return memtier; } static void destroy_memory_tier(struct memory_tier *memtier) { list_del(&memtier->list); + /* + * synchronize_rcu in clear_node_memory_tier makes sure + * we don't have rcu access to this memory tier. + */ kfree(memtier); } static bool clear_node_memory_tier(int node) { bool cleared = false; + pg_data_t *pgdat; struct memory_tier *memtier; + pgdat = NODE_DATA(node); + if (!pgdat) + return false; + + /* + * Make sure that anybody looking at NODE_DATA who finds + * a valid memtier finds memory_dev_types with nodes still + * linked to the memtier. We achieve this by waiting for + * rcu read section to finish using synchronize_rcu. + * This also enables us to free the destroyed memory tier + * with kfree instead of kfree_rcu + */ memtier = __node_get_memory_tier(node); if (memtier) { struct memory_dev_type *memtype; + rcu_assign_pointer(pgdat->memtier, NULL); + synchronize_rcu(); memtype = node_memory_types[node].memtype; node_clear(node, memtype->nodes); if (nodes_empty(memtype->nodes)) { _ Patches currently in -mm which might be from aneesh.kumar@linux.ibm.com are