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 lists.gnu.org (lists.gnu.org [209.51.188.17]) (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 B5199C30653 for ; Tue, 25 Jun 2024 07:22:39 +0000 (UTC) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1sM0VV-0007xC-Q0; Tue, 25 Jun 2024 03:22:25 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1sM0VT-0007wf-F3 for qemu-devel@nongnu.org; Tue, 25 Jun 2024 03:22:23 -0400 Received: from out-183.mta0.migadu.com ([91.218.175.183]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1sM0VP-0001TR-T6 for qemu-devel@nongnu.org; Tue, 25 Jun 2024 03:22:23 -0400 X-Envelope-To: ying.huang@intel.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux.dev; s=key1; t=1719300137; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=+QxrE1oC+aVRiXR0Em/+UkQz9SHDzWuJN8VEwX6wLGE=; b=nqdD5dxBF/ntSxw89rc40DPWC2mtlZe1uxRuHGSjyOZB8HUMG8K+FSCBzoXItcI8k5ZizF LYzmWSFqWSwDIsLKOPlH9KK/h0ShtLsXG9b3afTOlrOrQmzq893IMKy2MCYnWffABjEiSP jIUoUGAQPTrHMd7hy7zbrYxomS1fFMI= X-Envelope-To: jonathan.cameron@huawei.com X-Envelope-To: gourry.memverge@gmail.com X-Envelope-To: aneesh.kumar@linux.ibm.com X-Envelope-To: mhocko@suse.com X-Envelope-To: tj@kernel.org X-Envelope-To: john@jagalactic.com X-Envelope-To: emirakhur@micron.com X-Envelope-To: vtavarespetr@micron.com X-Envelope-To: ravis.opensrc@micron.com X-Envelope-To: apopple@nvidia.com X-Envelope-To: sthanneeru@micron.com X-Envelope-To: sj@kernel.org X-Envelope-To: rafael@kernel.org X-Envelope-To: lenb@kernel.org X-Envelope-To: akpm@linux-foundation.org X-Envelope-To: dave.jiang@intel.com X-Envelope-To: dan.j.williams@intel.com X-Envelope-To: linux-acpi@vger.kernel.org X-Envelope-To: linux-kernel@vger.kernel.org X-Envelope-To: linux-mm@kvack.org X-Envelope-To: horenc@vt.edu X-Envelope-To: horenchuang@bytedance.com X-Envelope-To: horenchuang@gmail.com X-Envelope-To: linux-cxl@vger.kernel.org X-Envelope-To: qemu-devel@nongnu.org MIME-Version: 1.0 Date: Tue, 25 Jun 2024 07:22:13 +0000 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable X-Report-Abuse: Please report any abuse attempt to abuse@migadu.com and include these headers. From: "Ho-Ren Chuang" Message-ID: TLS-Required: No Subject: Re: [PATCH v1] memory tier: consolidate the initialization of memory tiers To: "Huang, Ying" Cc: "Jonathan Cameron" , "Gregory Price" , aneesh.kumar@linux.ibm.com, mhocko@suse.com, tj@kernel.org, john@jagalactic.com, "Eishan Mirakhur" , "Vinicius Tavares Petrucci" , "Ravis OpenSrc" , "Alistair Popple" , "Srinivasulu Thanneeru" , "SeongJae Park" , "Rafael J. Wysocki" , "Len Brown" , "Andrew Morton" , "Dave Jiang" , "Dan Williams" , linux-acpi@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, "Ho-Ren (Jack) Chuang" , "Ho-Ren (Jack) Chuang" , "Ho-Ren (Jack) Chuang" , linux-cxl@vger.kernel.org, qemu-devel@nongnu.org In-Reply-To: <87wmmezqx7.fsf@yhuang6-desk2.ccr.corp.intel.com> References: <20240621044833.3953055-1-horen.chuang@linux.dev> <87wmmezqx7.fsf@yhuang6-desk2.ccr.corp.intel.com> X-Migadu-Flow: FLOW_OUT Received-SPF: pass client-ip=91.218.175.183; envelope-from=horen.chuang@linux.dev; helo=out-183.mta0.migadu.com X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org June 24, 2024 at 1:27 AM, "Huang, Ying" wrote: Hi Huang, Ying, Thanks for your feedback. Replies inlined. >=20 >=20Hi, Jack, >=20 >=20Thanks for patch! >=20 >=20"Ho-Ren (Jack) Chuang" writes: >=20 >=20>=20 >=20> If we simply move the set_node_memory_tier() from memory_tier_init(= ) to > >=20 >=20> late_initcall(), it will result in HMAT not registering the > >=20 >=20> mt_adistance_algorithm callback function, because set_node_memory_= tier() > >=20 >=20> is not performed during the memory tiering initialization phase, > >=20 >=20> leading to a lack of correct default_dram information. > >=20 >=20> Therefore, we introduced a nodemask to pass the information of the > >=20 >=20> default DRAM nodes. The reason for not choosing to reuse > >=20 >=20> default_dram_type->nodes is that it is not clean enough. So in the= end, > >=20 >=20> we use a __initdata variable, which is a variable that is released= once > >=20 >=20> initialization is complete, including both CPU and memory nodes fo= r HMAT > >=20 >=20> to iterate through. > >=20 >=20> Besides, since default_dram_type may be checked/used during the > >=20 >=20> initialization process of HMAT and drivers, it is better to keep t= he > >=20 >=20> allocation of default_dram_type in memory_tier_init(). > >=20 >=20> Signed-off-by: Ho-Ren (Jack) Chuang > >=20 >=20 > Suggested-by: Jonathan Cameron >=20 Thank=20you for your help with the input. Will add it in the v2. > >=20 >=20> --- > >=20 >=20> Hi all, > >=20 >=20> The current memory tier initialization process is distributed acro= ss two > >=20 >=20> different functions, memory_tier_init() and memory_tier_late_init(= ). This > >=20 >=20> design is hard to maintain. Thus, this patch is proposed to reduce= the > >=20 >=20> possible code paths by consolidating different initialization patc= hes into one. > >=20 >=20> The earlier discussion with Jonathan and Ying is listed here: > >=20 >=20> https://lore.kernel.org/lkml/20240405150244.00004b49@Huawei.com/ > >=20 >=20> If we want to put these two initializations together, they must be= placed > >=20 >=20> together in the later function. Because only at that time, the HMA= T information > >=20 >=20> will be ready, adist between nodes can be calculated, and memory t= iering can be > >=20 >=20> established based on the adist. So we position the initialization = at > >=20 >=20> memory_tier_init() to the memory_tier_late_init() call. > >=20 >=20> Moreover, it's natural to keep memory_tier initialization in drive= rs at > >=20 >=20> device_initcall() level. > >=20 >=20> This patchset is based on commits cf93be18fa1b and a72a30af550c: > >=20 >=20> [0/2] https://lkml.kernel.org/r/20240405000707.2670063-1-horenchua= ng@bytedance.com > >=20 >=20> [1/2] https://lkml.kernel.org/r/20240405000707.2670063-2-horenchua= ng@bytedance.com > >=20 >=20> [1/2] https://lkml.kernel.org/r/20240405000707.2670063-3-horenchua= ng@bytedance.com > >=20 >=20 > It appears that you should switch the parts before and after "---". >=20 >=20This is the real patch description, as pointed out by Andrew too. >=20 Thank=20you for the suggestion. I plan to write the real patch descriptio= n in the cover letter in the next version to avoid any misunderstanding. > >=20 >=20> Thanks, > >=20 >=20> Ho-Ren (Jack) Chuang > >=20 >=20> drivers/acpi/numa/hmat.c | 4 ++- > >=20 >=20> include/linux/memory-tiers.h | 6 ++++ > >=20 >=20> mm/memory-tiers.c | 70 ++++++++++++++++++------------------ > >=20 >=20> 3 files changed, 44 insertions(+), 36 deletions(-) > >=20 >=20> diff --git a/drivers/acpi/numa/hmat.c b/drivers/acpi/numa/hmat.c > >=20 >=20> index 2c8ccc91ebe6..31a77a3324a8 100644 > >=20 >=20> --- a/drivers/acpi/numa/hmat.c > >=20 >=20> +++ b/drivers/acpi/numa/hmat.c > >=20 >=20> @@ -939,11 +939,13 @@ static int hmat_set_default_dram_perf(void) > >=20 >=20> int nid, pxm; > >=20 >=20> struct memory_target *target; > >=20 >=20> struct access_coordinate *attrs; > >=20 >=20> + nodemask_t default_dram_nodes; > >=20 >=20>=20=20 >=20>=20 >=20> if (!default_dram_type) > >=20 >=20> return -EIO; > >=20 >=20>=20=20 >=20>=20 >=20> - for_each_node_mask(nid, default_dram_type->nodes) { > >=20 >=20> + default_dram_nodes =3D mt_get_default_dram_nodemask(); > >=20 >=20> + for_each_node_mask(nid, default_dram_nodes) { > >=20 >=20 > We don't need 'default_dram_type' in the function actually. It appears >=20 >=20that we can hide it in memory-tiers.c now? >=20 Do=20you mean to remove the "if (!default_dram_type) return -EIO;" here? If so, I agree, it's not used anymore here. > >=20 >=20> pxm =3D node_to_pxm(nid); > >=20 >=20> target =3D find_mem_target(pxm); > >=20 >=20> if (!target) > >=20 >=20> diff --git a/include/linux/memory-tiers.h b/include/linux/memory-t= iers.h > >=20 >=20> index 0d70788558f4..1567db7bd40e 100644 > >=20 >=20> --- a/include/linux/memory-tiers.h > >=20 >=20> +++ b/include/linux/memory-tiers.h > >=20 >=20> @@ -51,6 +51,7 @@ int mt_perf_to_adistance(struct access_coordinat= e *perf, int *adist); > >=20 >=20> struct memory_dev_type *mt_find_alloc_memory_type(int adist, > >=20 >=20> struct list_head *memory_types); > >=20 >=20> void mt_put_memory_types(struct list_head *memory_types); > >=20 >=20> +nodemask_t mt_get_default_dram_nodemask(void); > >=20 >=20> #ifdef CONFIG_MIGRATION > >=20 >=20> int next_demotion_node(int node); > >=20 >=20> void node_get_allowed_targets(pg_data_t *pgdat, nodemask_t *target= s); > >=20 >=20> @@ -149,5 +150,10 @@ static inline struct memory_dev_type *mt_find= _alloc_memory_type(int adist, > >=20 >=20> static inline void mt_put_memory_types(struct list_head *memory_ty= pes) > >=20 >=20> { > >=20 >=20> } > >=20 >=20> + > >=20 >=20> +static inline nodemask_t mt_get_default_dram_nodemask(void) > >=20 >=20> +{ > >=20 >=20> + return NODE_MASK_NONE; > >=20 >=20> +} > >=20 >=20> #endif /* CONFIG_NUMA */ > >=20 >=20> #endif /* _LINUX_MEMORY_TIERS_H */ > >=20 >=20> diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c > >=20 >=20> index 6632102bd5c9..7d4b7f53dd8f 100644 > >=20 >=20> --- a/mm/memory-tiers.c > >=20 >=20> +++ b/mm/memory-tiers.c > >=20 >=20> @@ -43,6 +43,7 @@ static LIST_HEAD(memory_tiers); > >=20 >=20> static LIST_HEAD(default_memory_types); > >=20 >=20> static struct node_memory_type_map node_memory_types[MAX_NUMNODES]= ; > >=20 >=20> struct memory_dev_type *default_dram_type; > >=20 >=20> +static nodemask_t default_dram_nodes __initdata =3D NODE_MASK_NON= E; > >=20 >=20>=20=20 >=20>=20 >=20> static const struct bus_type memory_tier_subsys =3D { > >=20 >=20> .name =3D "memory_tiering", > >=20 >=20> @@ -125,6 +126,11 @@ static inline struct memory_tier *to_memory_t= ier(struct device *device) > >=20 >=20> return container_of(device, struct memory_tier, dev); > >=20 >=20> } > >=20 >=20>=20=20 >=20>=20 >=20> +nodemask_t __init mt_get_default_dram_nodemask(void) > >=20 >=20> +{ > >=20 >=20> + return default_dram_nodes; > >=20 >=20> +} > >=20 >=20> + > >=20 >=20 > Why not just expose 'default_dram_nodes'? >=20 I=20was thinking encapsulating it should be more systematic/structural. Do you think exposing it is better? > >=20 >=20> static __always_inline nodemask_t get_memtier_nodemask(struct memor= y_tier *memtier) > >=20 >=20> { > >=20 >=20> nodemask_t nodes =3D NODE_MASK_NONE; > >=20 >=20> @@ -671,27 +677,38 @@ EXPORT_SYMBOL_GPL(mt_put_memory_types); > >=20 >=20>=20=20 >=20>=20 >=20> /* > >=20 >=20> * This is invoked via `late_initcall()` to initialize memory tiers= for > >=20 >=20> - * CPU-less memory nodes after driver initialization, which is > >=20 >=20> - * expected to provide `adistance` algorithms. > >=20 >=20> + * memory nodes, both with and without CPUs. After the initializa= tion of > >=20 > > + * firmware and devices, adistance algorithms are expected to be pr= ovided. > >=20 >=20> */ > >=20 >=20> static int __init memory_tier_late_init(void) > >=20 >=20> { > >=20 >=20> int nid; > >=20 >=20> + struct memory_tier *memtier; > >=20 >=20>=20=20 >=20>=20 >=20> guard(mutex)(&memory_tier_lock); > >=20 >=20> + /* > >=20 >=20> + * Look at all the existing and uninitialized N_MEMORY nodes and > >=20 >=20> + * add them to default memory tier or to a tier if we already hav= e > >=20 >=20> + * memory types assigned. > >=20 >=20> + */ > >=20 >=20> for_each_node_state(nid, N_MEMORY) { > >=20 >=20 > During the function run, the node may change between N_MEMORY=20and >=20 >=20!N_MEMORY in theory. So, it appears necessary to get/put_online_mems(= ) >=20 >=20in the function? >=20 Thanks=20for the catch. I will add get/put_online_mems(). > >=20 >=20> - /* > >=20 >=20> - * Some device drivers may have initialized memory tiers > >=20 >=20> - * between `memory_tier_init()` and `memory_tier_late_init()`, > >=20 >=20> - * potentially bringing online memory nodes and > >=20 >=20> - * configuring memory tiers. Exclude them here. > >=20 >=20> - */ > >=20 >=20> - if (node_memory_types[nid].memtype) > >=20 >=20> - continue; > >=20 >=20> + if (!node_state(nid, N_CPU)) > >=20 >=20 > Why? I think that we should "continue" here even if node_state(nid, >=20 >=20N_CPU). >=20 Do=20you mean no matter node_state(nid, N_CPU) or !node_state(nid, N_CPU)= , as long as if (node_memory_types[nid].memtype) is true, we should "continue"? I think you are right, at this moment, we only care if the node_memory_types[nid].memtype is set or not. If not, we should set it here. If yes, we should continue. If my understanding is correct, I will fix it in the v2. > >=20 >=20> + /* > >=20 >=20> + * Some device drivers may have initialized > >=20 >=20> + * memory tiers, potentially bringing memory nodes > >=20 >=20> + * online and configuring memory tiers. > >=20 >=20> + * Exclude them here. > >=20 >=20> + */ > >=20 >=20> + if (node_memory_types[nid].memtype) > >=20 >=20> + continue; > >=20 >=20>=20=20 >=20>=20 >=20> - set_node_memory_tier(nid); > >=20 >=20> + memtier =3D set_node_memory_tier(nid); > >=20 >=20> + if (IS_ERR(memtier)) > >=20 >=20> + /* > >=20 >=20> + * Continue with memtiers we are able to setup. > >=20 >=20> + */ > >=20 >=20> + break; > >=20 >=20> } > >=20 >=20> - > >=20 >=20> establish_demotion_targets(); > >=20 >=20>=20=20 >=20>=20 >=20> return 0; > >=20 >=20> @@ -876,7 +893,6 @@ static int __meminit memtier_hotplug_callback(= struct notifier_block *self, > >=20 >=20> static int __init memory_tier_init(void) > >=20 >=20> { > >=20 >=20> int ret, node; > >=20 >=20> - struct memory_tier *memtier; > >=20 >=20>=20=20 >=20>=20 >=20> ret =3D subsys_virtual_register(&memory_tier_subsys, NULL); > >=20 >=20> if (ret) > >=20 >=20> @@ -887,7 +903,8 @@ static int __init memory_tier_init(void) > >=20 >=20> GFP_KERNEL); > >=20 >=20> WARN_ON(!node_demotion); > >=20 >=20> #endif > >=20 >=20> - mutex_lock(&memory_tier_lock); > >=20 >=20> + > >=20 >=20> + guard(mutex)(&memory_tier_lock); > >=20 >=20> /* > >=20 >=20> * For now we can have 4 faster memory tiers with smaller adistance > >=20 >=20> * than default DRAM tier. > >=20 >=20> @@ -898,28 +915,11 @@ static int __init memory_tier_init(void) > >=20 >=20> panic("%s() failed to allocate default DRAM tier\n", __func__); > >=20 >=20>=20=20 >=20>=20 >=20> /* > >=20 >=20> - * Look at all the existing N_MEMORY nodes and add them to > >=20 >=20> - * default memory tier or to a tier if we already have memory > >=20 >=20> - * types assigned. > >=20 >=20> + * Record nodes with memory and CPU to set default DRAM performan= ce. > >=20 >=20> */ > >=20 >=20 > For one line comments, we can use >=20 >=20 /* Record nodes with memory and CPU to set default DRAM performance.= */ >=20 Thank=20you for the guidance. Will fix in the v2. > >=20 >=20> - for_each_node_state(node, N_MEMORY) { > >=20 >=20> - if (!node_state(node, N_CPU)) > >=20 >=20> - /* > >=20 >=20> - * Defer memory tier initialization on > >=20 >=20> - * CPUless numa nodes. These will be initialized > >=20 >=20> - * after firmware and devices are initialized. > >=20 >=20> - */ > >=20 >=20> - continue; > >=20 >=20> - > >=20 >=20> - memtier =3D set_node_memory_tier(node); > >=20 >=20> - if (IS_ERR(memtier)) > >=20 >=20> - /* > >=20 >=20> - * Continue with memtiers we are able to setup > >=20 >=20> - */ > >=20 >=20> - break; > >=20 >=20> - } > >=20 >=20> - establish_demotion_targets(); > >=20 >=20> - mutex_unlock(&memory_tier_lock); > >=20 >=20> + for_each_node_state(node, N_MEMORY) > >=20 >=20> + if (node_state(node, N_CPU)) > >=20 >=20> + node_set(node, default_dram_nodes); > >=20 >=20 > Why not use >=20 >=20 nodes_andnot(default_dram_nodes, node_states[N_MEMORY], node_states[= N_CPU]); >=20 Instead=20of using nodes_andnot(), should nodes_and() be correct? because= we wanna record the nodes that are both N_MEMORY and N_CPU. > >=20 >=20> hotplug_memory_notifier(memtier_hotplug_callback, MEMTIER_HOTPLUG_P= RI); > >=20 >=20> return 0; > >=20 >=20 > -- >=20 >=20Best Regards, >=20 >=20Huang, Ying >