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 kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id E8142C43334 for ; Mon, 6 Jun 2022 08:06:53 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 7B4F46B0073; Mon, 6 Jun 2022 04:06:53 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 760888D0003; Mon, 6 Jun 2022 04:06:53 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 5FF006B0078; Mon, 6 Jun 2022 04:06:53 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0010.hostedemail.com [216.40.44.10]) by kanga.kvack.org (Postfix) with ESMTP id 516206B0073 for ; Mon, 6 Jun 2022 04:06:53 -0400 (EDT) Received: from smtpin13.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay01.hostedemail.com (Postfix) with ESMTP id 278A360791 for ; Mon, 6 Jun 2022 08:06:53 +0000 (UTC) X-FDA: 79547079906.13.5C56372 Received: from mga02.intel.com (mga02.intel.com [134.134.136.20]) by imf12.hostedemail.com (Postfix) with ESMTP id 66FD040038 for ; Mon, 6 Jun 2022 08:06:06 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1654502812; x=1686038812; h=message-id:subject:from:to:cc:date:in-reply-to: references:mime-version:content-transfer-encoding; bh=owmFIl1Z/QChDPV4ZrzI5MhEJW35PIhDJJr8K+89Odw=; b=QmYNjR92sPLcgHwrldv2ymdu/rRclONnWdawp0sG4Z6sSgtmzkLcppkr oFAWSs5sRIS70xPh6T2xhf5aJbKhw+WV77ljOFA4DdFDdhxwGL6m/ofG/ ZAfcArPD2fCmH7gFEJgxq3QDePkesLZN8LDJgMoKxPvslc69Ihyduq1Ah 9mb2ao1kQbylgPH/4VriJzxOnFT67D0A0p8VTWZfK746r1E5+91BnrpK7 8d1GjaOPCooji9bSWkTjv+arVx4ym47FxW7vUD2Hq067yVmyH6hD/FuoT mfD3DuppIGWzYyRhRFMIJFYVwPKJoHgQjUfL6kmIcDyidZUkAhzueLxPh Q==; X-IronPort-AV: E=McAfee;i="6400,9594,10369"; a="264472468" X-IronPort-AV: E=Sophos;i="5.91,280,1647327600"; d="scan'208";a="264472468" Received: from fmsmga007.fm.intel.com ([10.253.24.52]) by orsmga101.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 06 Jun 2022 01:06:47 -0700 X-IronPort-AV: E=Sophos;i="5.91,280,1647327600"; d="scan'208";a="583507338" Received: from xingguom-mobl.ccr.corp.intel.com ([10.254.213.116]) by fmsmga007-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 06 Jun 2022 01:06:38 -0700 Message-ID: Subject: Re: [RFC PATCH v4 7/7] mm/demotion: Demote pages according to allocation fallback order From: Ying Huang To: Aneesh Kumar K V , linux-mm@kvack.org, akpm@linux-foundation.org Cc: Greg Thelen , Yang Shi , Davidlohr Bueso , Tim C Chen , Brice Goglin , Michal Hocko , Linux Kernel Mailing List , Hesham Almatary , Dave Hansen , Jonathan Cameron , Alistair Popple , Dan Williams , Feng Tang , Jagdish Gediya , Baolin Wang , David Rientjes Date: Mon, 06 Jun 2022 16:06:34 +0800 In-Reply-To: <3d361a74-c333-da7f-9578-8774f6943d32@linux.ibm.com> References: <20220527122528.129445-1-aneesh.kumar@linux.ibm.com> <20220527122528.129445-8-aneesh.kumar@linux.ibm.com> <046c373a-f30b-091d-47a1-e28bfb7e9394@linux.ibm.com> <9f6e60cc8be3cbde4871458c612c5c31d2a9e056.camel@intel.com> <65919df6b3302741780ff6fa69e497af06a9825e.camel@intel.com> <87leua8g7s.fsf@linux.ibm.com> <3d361a74-c333-da7f-9578-8774f6943d32@linux.ibm.com> Content-Type: text/plain; charset="UTF-8" User-Agent: Evolution 3.38.3-1 MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Stat-Signature: tr7uk6x9xbur1c1sfq8xe5s9hhp9tc5q X-Rspam-User: Authentication-Results: imf12.hostedemail.com; dkim=pass header.d=intel.com header.s=Intel header.b=QmYNjR92; dmarc=pass (policy=none) header.from=intel.com; spf=none (imf12.hostedemail.com: domain of ying.huang@intel.com has no SPF policy when checking 134.134.136.20) smtp.mailfrom=ying.huang@intel.com X-Rspamd-Server: rspam09 X-Rspamd-Queue-Id: 66FD040038 X-HE-Tag: 1654502766-255208 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: On Mon, 2022-06-06 at 13:32 +0530, Aneesh Kumar K V wrote: > On 6/6/22 1:12 PM, Ying Huang wrote: > > On Mon, 2022-06-06 at 11:51 +0530, Aneesh Kumar K.V wrote: > > > Ying Huang writes: > > > > > > ..... > > > > > > > > > > > > > > > > > > https://lore.kernel.org/lkml/69f2d063a15f8c4afb4688af7b7890f32af55391.camel@intel.com/ > > > > > > > > > > > > > > > > That is, something like below, > > > > > > > > > > > > > > > > static struct page *alloc_demote_page(struct page *page, unsigned long node) > > > > > > > > { > > > > > > > > struct page *page; > > > > > > > > nodemask_t allowed_mask; > > > > > > > > struct migration_target_control mtc = { > > > > > > > > /* > > > > > > > > * Allocate from 'node', or fail quickly and quietly. > > > > > > > > * When this happens, 'page' will likely just be discarded > > > > > > > > * instead of migrated. > > > > > > > > */ > > > > > > > > .gfp_mask = (GFP_HIGHUSER_MOVABLE & ~__GFP_RECLAIM) | > > > > > > > > __GFP_THISNODE | __GFP_NOWARN | > > > > > > > > __GFP_NOMEMALLOC | GFP_NOWAIT, > > > > > > > > .nid = node > > > > > > > > }; > > > > > > > > > > > > > > > > page = alloc_migration_target(page, (unsigned long)&mtc); > > > > > > > > if (page) > > > > > > > > return page; > > > > > > > > > > > > > > > > mtc.gfp_mask &= ~__GFP_THISNODE; > > > > > > > > mtc.nmask = &allowed_mask; > > > > > > > > > > > > > > > > return alloc_migration_target(page, (unsigned long)&mtc); > > > > > > > > } > > > > > > > > > > > > > > I skipped doing this in v5 because I was not sure this is really what we > > > > > > > want. > > > > > > > > > > > > I think so. And this is the original behavior. We should keep the > > > > > > original behavior as much as possible, then make changes if necessary. > > > > > > > > > > > > > > > > That is the reason I split the new page allocation as a separate patch. > > > > > Previous discussion on this topic didn't conclude on whether we really > > > > > need to do the above or not > > > > > https://lore.kernel.org/lkml/CAAPL-u9endrWf_aOnPENDPdvT-2-YhCAeJ7ONGckGnXErTLOfQ@mail.gmail.com/ > > > > > > > > Please check the later email in the thread you referenced. Both Wei and > > > > me agree that the use case needs to be supported. We just didn't reach > > > > concensus about how to implement it. If you think Wei's solution is > > > > better (referenced as below), you can try to do that too. Although I > > > > think my proposed implementation is much simpler. > > > > > > How about the below details > > > > > > diff --git a/include/linux/memory-tiers.h b/include/linux/memory-tiers.h > > > index 79bd8d26feb2..cd6e71f702ad 100644 > > > --- a/include/linux/memory-tiers.h > > > +++ b/include/linux/memory-tiers.h > > > @@ -21,6 +21,7 @@ void node_remove_from_memory_tier(int node); > > >   int node_get_memory_tier_id(int node); > > >   int node_set_memory_tier(int node, int tier); > > >   int node_reset_memory_tier(int node, int tier); > > > +void node_get_allowed_targets(int node, nodemask_t *targets); > > >   #else > > >   #define numa_demotion_enabled false > > >   static inline int next_demotion_node(int node) > > > @@ -28,6 +29,10 @@ static inline int next_demotion_node(int node) > > >    return NUMA_NO_NODE; > > >   } > > >    > > > > > > > > > > > > > > > > > > > > > +static inline void node_get_allowed_targets(int node, nodemask_t *targets) > > > +{ > > > + *targets = NODE_MASK_NONE; > > > +} > > >   #endif /* CONFIG_TIERED_MEMORY */ > > >    > > > > > > > > > > > > > > > > > > > > >   #endif > > > diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c > > > index b4e72b672d4d..592d939ec28d 100644 > > > --- a/mm/memory-tiers.c > > > +++ b/mm/memory-tiers.c > > > @@ -18,6 +18,7 @@ struct memory_tier { > > >    > > > > > > > > > > > > > > > > > > > > >   struct demotion_nodes { > > >    nodemask_t preferred; > > > + nodemask_t allowed; > > >   }; > > >    > > > > > > > > > > > > > > > > > > > > >   #define to_memory_tier(device) container_of(device, struct memory_tier, dev) > > > @@ -378,6 +379,25 @@ int node_set_memory_tier(int node, int tier) > > >   } > > >   EXPORT_SYMBOL_GPL(node_set_memory_tier); > > >    > > > > > > > > > > > > > > > > > > > > > +void node_get_allowed_targets(int node, nodemask_t *targets) > > > +{ > > > + /* > > > + * node_demotion[] is updated without excluding this > > > + * function from running. > > > + * > > > + * If any node is moving to lower tiers then modifications > > > + * in node_demotion[] are still valid for this node, if any > > > + * node is moving to higher tier then moving node may be > > > + * used once for demotion which should be ok so rcu should > > > + * be enough here. > > > + */ > > > + rcu_read_lock(); > > > + > > > + *targets = node_demotion[node].allowed; > > > + > > > + rcu_read_unlock(); > > > +} > > > + > > >   /** > > >    * next_demotion_node() - Get the next node in the demotion path > > >    * @node: The starting node to lookup the next node > > > @@ -437,8 +457,10 @@ static void __disable_all_migrate_targets(void) > > >   { > > >    int node; > > >    > > > > > > > > > > > > > > > > > > > > > - for_each_node_mask(node, node_states[N_MEMORY]) > > > + for_each_node_mask(node, node_states[N_MEMORY]) { > > >    node_demotion[node].preferred = NODE_MASK_NONE; > > > + node_demotion[node].allowed = NODE_MASK_NONE; > > > + } > > >   } > > >    > > > > > > > > > > > > > > > > > > > > >   static void disable_all_migrate_targets(void) > > > @@ -465,7 +487,7 @@ static void establish_migration_targets(void) > > >    struct demotion_nodes *nd; > > >    int target = NUMA_NO_NODE, node; > > >    int distance, best_distance; > > > - nodemask_t used; > > > + nodemask_t used, allowed = NODE_MASK_NONE; > > >    > > > > > > > > > > > > > > > > > > > > >    if (!node_demotion) > > >    return; > > > @@ -511,6 +533,29 @@ static void establish_migration_targets(void) > > >    } > > >    } while (1); > > >    } > > > + /* > > > + * Now build the allowed mask for each node collecting node mask from > > > + * all memory tier below it. This allows us to fallback demotion page > > > + * allocation to a set of nodes that is closer the above selected > > > + * perferred node. > > > + */ > > > + list_for_each_entry(memtier, &memory_tiers, list) > > > + nodes_or(allowed, allowed, memtier->nodelist); > > > + /* > > > + * Removes nodes not yet in N_MEMORY. > > > + */ > > > + nodes_and(allowed, node_states[N_MEMORY], allowed); > > > + > > > + list_for_each_entry(memtier, &memory_tiers, list) { > > > + /* > > > + * Keep removing current tier from allowed nodes, > > > + * This will remove all nodes in current and above > > > + * memory tier from the allowed mask. > > > + */ > > > + nodes_andnot(allowed, allowed, memtier->nodelist); > > > + for_each_node_mask(node, memtier->nodelist) > > > + node_demotion[node].allowed = allowed; > > > + } > > >   } > > >    > > > > > > > > > > > > > > > > > > > > >   /* > > > diff --git a/mm/vmscan.c b/mm/vmscan.c > > > index 3a8f78277f99..b0792d838efb 100644 > > > --- a/mm/vmscan.c > > > +++ b/mm/vmscan.c > > > @@ -1460,19 +1460,32 @@ static void folio_check_dirty_writeback(struct folio *folio, > > >    mapping->a_ops->is_dirty_writeback(folio, dirty, writeback); > > >   } > > >    > > > > > > > > > > > > > > > > > > > > > -static struct page *alloc_demote_page(struct page *page, unsigned long node) > > > +static struct page *alloc_demote_page(struct page *page, unsigned long private) > > >   { > > > - struct migration_target_control mtc = { > > > - /* > > > - * Allocate from 'node', or fail quickly and quietly. > > > - * When this happens, 'page' will likely just be discarded > > > - * instead of migrated. > > > - */ > > > - .gfp_mask = (GFP_HIGHUSER_MOVABLE & ~__GFP_RECLAIM) | > > > - __GFP_THISNODE | __GFP_NOWARN | > > > - __GFP_NOMEMALLOC | GFP_NOWAIT, > > > - .nid = node > > > - }; > > > + struct page *target_page; > > > + nodemask_t *allowed_mask; > > > + struct migration_target_control *mtc; > > > + > > > + mtc = (struct migration_target_control *)private; > > > + > > > + allowed_mask = mtc->nmask; > > > + /* > > > + * make sure we allocate from the target node first also trying to > > > + * reclaim pages from the target node via kswapd if we are low on > > > + * free memory on target node. If we don't do this and if we have low > > > + * free memory on the target memtier, we would start allocating pages > > > + * from higher memory tiers without even forcing a demotion of cold > > > + * pages from the target memtier. This can result in the kernel placing > > > + * hotpages in higher memory tiers. > > > + */ > > > + mtc->nmask = NULL; > > > + mtc->gfp_mask |= __GFP_THISNODE; > > > + target_page = alloc_migration_target(page, (unsigned long)&mtc); > > > + if (target_page) > > > + return target_page; > > > + > > > + mtc->gfp_mask &= ~__GFP_THISNODE; > > > + mtc->nmask = allowed_mask; > > >    > > > > > > > > > > > > > > > > > > > > >    return alloc_migration_target(page, (unsigned long)&mtc); > > >   } > > > @@ -1487,6 +1500,19 @@ static unsigned int demote_page_list(struct list_head *demote_pages, > > >   { > > >    int target_nid = next_demotion_node(pgdat->node_id); > > >    unsigned int nr_succeeded; > > > + nodemask_t allowed_mask; > > > + > > > + struct migration_target_control mtc = { > > > + /* > > > + * Allocate from 'node', or fail quickly and quietly. > > > + * When this happens, 'page' will likely just be discarded > > > + * instead of migrated. > > > + */ > > > + .gfp_mask = (GFP_HIGHUSER_MOVABLE & ~__GFP_RECLAIM) | __GFP_NOWARN | > > > + __GFP_NOMEMALLOC | GFP_NOWAIT, > > > + .nid = target_nid, > > > + .nmask = &allowed_mask > > > + }; > > >    > > > > > > > > > > > > > > > > > > > > >    if (list_empty(demote_pages)) > > >    return 0; > > > @@ -1494,10 +1520,12 @@ static unsigned int demote_page_list(struct list_head *demote_pages, > > >    if (target_nid == NUMA_NO_NODE) > > >    return 0; > > >    > > > > > > > > > > > > > > > > > > > > > + node_get_allowed_targets(pgdat->node_id, &allowed_mask); > > > + > > >    /* Demotion ignores all cpuset and mempolicy settings */ > > >    migrate_pages(demote_pages, alloc_demote_page, NULL, > > > - target_nid, MIGRATE_ASYNC, MR_DEMOTION, > > > - &nr_succeeded); > > > + (unsigned long)&mtc, MIGRATE_ASYNC, MR_DEMOTION, > > > + &nr_succeeded); > > > > Firstly, it addressed my requirement, Thanks! And, I'd prefer to put > > mtc definition in alloc_demote_page(). Because that makes all page > > allocation logic in one function. Thus the readability of code is > > better. > > The challenge is in allowed_mask computation. That is based on the > src_node and not target_node. > How about passing the src_node to alloc_demote_page()? Best Regards, Huang, Ying