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 7E5ECCD4F2E for ; Fri, 22 Sep 2023 08:19:33 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id DB9276B0294; Fri, 22 Sep 2023 04:19:32 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id D691D6B0295; Fri, 22 Sep 2023 04:19:32 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id C58C96B0297; Fri, 22 Sep 2023 04:19:32 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0012.hostedemail.com [216.40.44.12]) by kanga.kvack.org (Postfix) with ESMTP id B73AE6B0294 for ; Fri, 22 Sep 2023 04:19:32 -0400 (EDT) Received: from smtpin01.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay05.hostedemail.com (Postfix) with ESMTP id 86D71408D8 for ; Fri, 22 Sep 2023 08:19:32 +0000 (UTC) X-FDA: 81263534184.01.F36A583 Received: from out-210.mta0.migadu.com (out-210.mta0.migadu.com [91.218.175.210]) by imf21.hostedemail.com (Postfix) with ESMTP id 994E61C000B for ; Fri, 22 Sep 2023 08:19:29 +0000 (UTC) Authentication-Results: imf21.hostedemail.com; dkim=pass header.d=linux.dev header.s=key1 header.b=aaqU6sR2; dmarc=pass (policy=none) header.from=linux.dev; spf=pass (imf21.hostedemail.com: domain of muchun.song@linux.dev designates 91.218.175.210 as permitted sender) smtp.mailfrom=muchun.song@linux.dev ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1695370770; h=from:from:sender: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:dkim-signature; bh=imHpJXjN7oTKFJS9REtYSyqNmmkpWVduPz1WiGdeDhQ=; b=qH3Tv1BkRIw3+mXMPAC3diwvgT+Q3MidgCrmj0NFfoZnCfcAzy31RnHBojkDZkaHJhK/cQ BEs6rLQjmzsOaw2Jgc2ng0Xqbuv2ItLzuSZ124nTCRlSGd994zMDy8Evpwe93T8LPsH0Au aro8g16TZ8q5S/hihbslvVp8SS/dj4E= ARC-Authentication-Results: i=1; imf21.hostedemail.com; dkim=pass header.d=linux.dev header.s=key1 header.b=aaqU6sR2; dmarc=pass (policy=none) header.from=linux.dev; spf=pass (imf21.hostedemail.com: domain of muchun.song@linux.dev designates 91.218.175.210 as permitted sender) smtp.mailfrom=muchun.song@linux.dev ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1695370770; a=rsa-sha256; cv=none; b=KsPRSZNFbTprQhhpMA2dESIs6nKIhmm27PyiCQRDYjBmxDBY3pQ+hV/syxhG0JIBS+Lmdm Os1F5hL64EAwAK8jNZI6QRaJWOMykfAub+JtbbFJ1bJG+5e52KCwNRa/jC+H/f/wOlTwm/ Dr0DEvTqe68uCNjIfLCbhOJ6W2rphPk= Message-ID: <7f222dae-c256-a625-6846-dd22e16687fb@linux.dev> DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux.dev; s=key1; t=1695370767; 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=imHpJXjN7oTKFJS9REtYSyqNmmkpWVduPz1WiGdeDhQ=; b=aaqU6sR2Ls4axHOb/Ybmaf9uLXm07hG4eJMU3sKyN+oEFzI61M9Yp4tWEae7xyLZYHQXjC Ts3Y7eEeCyCQFA0hv0IgmszkamVHxn2C0wdntuxz+AybAfs3Ellb6XmL6ILM+9N/rl2Ca4 0BtKTuYV12xdb+Gu3qGbUFEdJ1rvYfw= Date: Fri, 22 Sep 2023 16:19:17 +0800 MIME-Version: 1.0 Subject: Re: [PATCH v4 4/8] hugetlb: perform vmemmap restoration on a list of pages To: Mike Kravetz Cc: Linux-MM , LKML , Muchun Song , Joao Martins , Oscar Salvador , David Hildenbrand , Miaohe Lin , David Rientjes , Anshuman Khandual , Naoya Horiguchi , Barry Song <21cnbao@gmail.com>, Michal Hocko , Matthew Wilcox , Xiongchun Duan , Andrew Morton References: <20230918230202.254631-1-mike.kravetz@oracle.com> <20230918230202.254631-5-mike.kravetz@oracle.com> <20230919205756.GB425719@monkey> <2FDB2018-74AE-4514-9B43-01664A8E5DBF@linux.dev> <20230921011223.GC4065@monkey> <306da2a1-0dd4-e858-930f-211947a466d2@linux.dev> <20230921215810.GA21193@monkey> X-Report-Abuse: Please report any abuse attempt to abuse@migadu.com and include these headers. From: Muchun Song In-Reply-To: <20230921215810.GA21193@monkey> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit X-Migadu-Flow: FLOW_OUT X-Rspam-User: X-Rspamd-Server: rspam12 X-Rspamd-Queue-Id: 994E61C000B X-Stat-Signature: rrorgpkbdiey1xegnfg8n7fm4dcafnu3 X-HE-Tag: 1695370769-850804 X-HE-Meta: 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 umRNsFex PCD0/fNwYB1tHbTf/6HdWCrMFAJVY4Tq6mU20Yojfv0MTe+A3o4gBwxvfoThCq5sYU0y7p91xtYgxYbs6vLk1UK4Sv2TuAyqOwrZbhiWc7tj6/+C4OywcBXeifw== 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 2023/9/22 05:58, Mike Kravetz wrote: > On 09/21/23 17:31, Muchun Song wrote: >> >> On 2023/9/21 09:12, Mike Kravetz wrote: >>> On 09/20/23 11:03, Muchun Song wrote: >>>>> On Sep 20, 2023, at 10:56, Muchun Song wrote: >>>>>> On Sep 20, 2023, at 04:57, Mike Kravetz wrote: >>>>>> On 09/19/23 17:52, Muchun Song wrote: >>>>>>> On 2023/9/19 07:01, Mike Kravetz wrote: >>> +/** >>> + * hugetlb_vmemmap_restore_folios - restore vmemmap for every folio on the list. >>> + * @h: struct hstate. >>> + * @folio_list: list of folios. >>> + * @restored: Set to number of folios for which vmemmap was restored >>> + * successfully if caller passes a non-NULL pointer. >>> + * >>> + * Return: %0 if vmemmap exists for all folios on the list. If an error is >>> + * encountered restoring vmemmap for ANY folio, an error code >>> + * will be returned to the caller. It is then the responsibility >>> + * of the caller to check the hugetlb vmemmap optimized flag of >>> + * each folio to determine if vmemmap was actually restored. >>> + * Note that processing is stopped when first error is encountered. >>> + */ >>> +int hugetlb_vmemmap_restore_folios(const struct hstate *h, >>> + struct list_head *folio_list, >>> + unsigned long *restored) >> How about changing parameter of @restored to a list_head type which >> returns the non-optimized (previously) or vmemmap-restored-sucessful huge >> pages? >> In which case, the caller could traverse this returned list to free >> them first like you have implemented in bulk_vmemmap_restore_enomem(), >> it will be more efficient. The meaning of returned value should also >> be changed accordingly since update_and_free_pages_bulk() wants to >> whether there is a vmemmap-optimized huge page being restored sucessfully >> to determine if it should clear hugetlb flag. So >> hugetlb_vmemmap_restore_folios() >> could return how many huge pages being restored successful, if a negative >> number is returned meaning there is some error in the process of restoring >> of vmemmap. >> > I had similar thoughts. An updated patch based on this approach is below. > When creating the patch, I discovered that using the function return code > for both number of vmemmap restored pages as well as error code was > unnecessary. Just checking !list_empty() of non-optimized pages tells us > if any were restored or could be freed. I also thought about this. But there is a little different. If HVO is disabled, we will always clear the hugetlb flag twice since the list couldn't be empty, I thought it is an optimization for HVO-disabled case. > > From b79f6eeb7a11644830bddfc43d2219c149d26405 Mon Sep 17 00:00:00 2001 > From: Mike Kravetz > Date: Sun, 10 Sep 2023 16:14:50 -0700 > Subject: [PATCH] hugetlb: perform vmemmap restoration on a list of pages > > The routine update_and_free_pages_bulk already performs vmemmap > restoration on the list of hugetlb pages in a separate step. In > preparation for more functionality to be added in this step, create a > new routine hugetlb_vmemmap_restore_folios() that will restore > vmemmap for a list of folios. > > This new routine must provide sufficient feedback about errors and > actual restoration performed so that update_and_free_pages_bulk can > perform optimally. > > Special care must be taken when encountering an error from > hugetlb_vmemmap_restore_folios. We want to continue making as much > forward progress as possible. A new routine bulk_vmemmap_restore_error > handles this specific situation. > > Signed-off-by: Mike Kravetz > --- > mm/hugetlb.c | 94 +++++++++++++++++++++++++++++++------------- > mm/hugetlb_vmemmap.c | 36 +++++++++++++++++ > mm/hugetlb_vmemmap.h | 10 +++++ > 3 files changed, 112 insertions(+), 28 deletions(-) > > diff --git a/mm/hugetlb.c b/mm/hugetlb.c > index 70fedf8682c4..11de3f885065 100644 > --- a/mm/hugetlb.c > +++ b/mm/hugetlb.c > @@ -1834,50 +1834,88 @@ static void update_and_free_hugetlb_folio(struct hstate *h, struct folio *folio, > schedule_work(&free_hpage_work); > } > > -static void update_and_free_pages_bulk(struct hstate *h, struct list_head *list) > +static void bulk_vmemmap_restore_error(struct hstate *h, > + struct list_head *list, > + struct list_head *non_op_folios) > { > struct folio *folio, *t_folio; > - bool clear_dtor = false; > > - /* > - * First allocate required vmemmmap (if necessary) for all folios on > - * list. If vmemmap can not be allocated, we can not free folio to > - * lower level allocator, so add back as hugetlb surplus page. > - * add_hugetlb_folio() removes the page from THIS list. > - * Use clear_dtor to note if vmemmap was successfully allocated for > - * ANY page on the list. > - */ > - list_for_each_entry_safe(folio, t_folio, list, lru) { > - if (folio_test_hugetlb_vmemmap_optimized(folio)) { > + if (!list_empty(non_op_folios)) { > + /* > + * Free any restored hugetlb pages so that restore of the > + * entire list can be retried. > + * The idea is that in the common case of ENOMEM errors freeing > + * hugetlb pages with vmemmap we will free up memory so that we > + * can allocate vmemmap for more hugetlb pages. > + */ > + list_for_each_entry_safe(folio, t_folio, non_op_folios, lru) { > + list_del(&folio->lru); > + spin_lock_irq(&hugetlb_lock); > + __clear_hugetlb_destructor(h, folio); > + spin_unlock_irq(&hugetlb_lock); > + update_and_free_hugetlb_folio(h, folio, false); > + cond_resched(); > + } > + } else { > + /* > + * In the case where vmemmap was not restored for ANY folios, > + * we loop through them trying to restore individually in the > + * hope that someone elsewhere may have done something to cause > + * success (such as freeing some memory). > + * If unable to restore a hugetlb page, the hugetlb page is > + * made a surplus page and removed from the list. > + * If are able to restore vmemmap for one hugetlb page, we free > + * it and quit processing the list to retry the bulk operation. > + */ > + list_for_each_entry_safe(folio, t_folio, list, lru) > if (hugetlb_vmemmap_restore(h, &folio->page)) { > spin_lock_irq(&hugetlb_lock); > add_hugetlb_folio(h, folio, true); > spin_unlock_irq(&hugetlb_lock); > - } else > - clear_dtor = true; > - } > + } else { > + list_del(&folio->lru); > + spin_lock_irq(&hugetlb_lock); > + __clear_hugetlb_destructor(h, folio); > + spin_unlock_irq(&hugetlb_lock); > + update_and_free_hugetlb_folio(h, folio, false); > + cond_resched(); > + break; > + } > } > +} > + > +static void update_and_free_pages_bulk(struct hstate *h, struct list_head *list) > +{ > + int ret; > + LIST_HEAD(non_op_folio); > + struct folio *folio, *t_folio; > > /* > - * If vmemmmap allocation was performed on any folio above, take lock > - * to clear destructor of all folios on list. This avoids the need to > - * lock/unlock for each individual folio. > - * The assumption is vmemmap allocation was performed on all or none > - * of the folios on the list. This is true expect in VERY rare cases. > + * First allocate required vmemmmap (if necessary) for all folios. > + * Carefully handle errors and free up any available hugetlb pages > + * in an effort to make forward progress. > */ > - if (clear_dtor) { > +retry: > + ret = hugetlb_vmemmap_restore_folios(h, list, &non_op_folio); > + if (ret < 0) { > + bulk_vmemmap_restore_error(h, list, &non_op_folio); > + goto retry; > + } > + > + /* > + * At this point, list should be empty, and there should only be > + * pages on the non_op_folio list. free those entries. Do note > + * that the non_op_folio list could be empty. > + */ > + VM_WARN_ON(!list_empty(list)); > + if (!list_empty(&non_op_folio)) { > spin_lock_irq(&hugetlb_lock); > - list_for_each_entry(folio, list, lru) > + list_for_each_entry(folio, &non_op_folio, lru) > __clear_hugetlb_destructor(h, folio); > spin_unlock_irq(&hugetlb_lock); > } > > - /* > - * Free folios back to low level allocators. vmemmap and destructors > - * were taken care of above, so update_and_free_hugetlb_folio will > - * not need to take hugetlb lock. > - */ > - list_for_each_entry_safe(folio, t_folio, list, lru) { > + list_for_each_entry_safe(folio, t_folio, &non_op_folio, lru) { > update_and_free_hugetlb_folio(h, folio, false); > cond_resched(); > } > diff --git a/mm/hugetlb_vmemmap.c b/mm/hugetlb_vmemmap.c > index 4558b814ffab..f827d4efcf8e 100644 > --- a/mm/hugetlb_vmemmap.c > +++ b/mm/hugetlb_vmemmap.c > @@ -480,6 +480,42 @@ int hugetlb_vmemmap_restore(const struct hstate *h, struct page *head) > return ret; > } > > +/** > + * hugetlb_vmemmap_restore_folios - restore vmemmap for every folio on the list. > + * @h: hstate. > + * @folio_list: list of folios. > + * @non_op_list: Output list of folio for which vmemmap exists. > + * > + * Return: %0 if vmemmap exists for all folios on the list and all entries have > + * been moved to non_op_list. If an error is encountered restoring > + * vmemmap for ANY folio, an error code will be returned to the > + * caller. Processing en entries stops when the first error is > + * encountered. folios processed before the error with vmemmap > + * will reside on the non_op_list. The folio that experienced the > + * error and all non-processed folios will remain on folio_list. > + */ > +int hugetlb_vmemmap_restore_folios(const struct hstate *h, > + struct list_head *folio_list, > + struct list_head *non_op_list) non_optimized_list or vmemmap_intact_list? The abbreviation is not straightforward. > +{ > + struct folio *folio, *t_folio; > + int ret = 0; > + > + list_for_each_entry_safe(folio, t_folio, folio_list, lru) { > + if (folio_test_hugetlb_vmemmap_optimized(folio)) { hugetlb_vmemmap_restore() has this check as well, so it is unnecessary here. > + ret = hugetlb_vmemmap_restore(h, &folio->page); > + if (ret) > + goto out; Maybe we could drop the label ("out") and just breaking or returning from here is enough. > + } > + > + /* Add non-optimized folios to output list */ > + list_move(&folio->lru, non_op_list); > + } > + > +out: > + return ret; > +} > + > /* Return true iff a HugeTLB whose vmemmap should and can be optimized. */ > static bool vmemmap_should_optimize(const struct hstate *h, const struct page *head) > { > diff --git a/mm/hugetlb_vmemmap.h b/mm/hugetlb_vmemmap.h > index c512e388dbb4..e6378ae5c5b6 100644 > --- a/mm/hugetlb_vmemmap.h > +++ b/mm/hugetlb_vmemmap.h > @@ -19,6 +19,9 @@ > > #ifdef CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP > int hugetlb_vmemmap_restore(const struct hstate *h, struct page *head); > +int hugetlb_vmemmap_restore_folios(const struct hstate *h, > + struct list_head *folio_list, > + struct list_head *non_op_folios); It is better to keep 3rd name (non_op_folios) consistent with where it is defined (it is non_op_list). > void hugetlb_vmemmap_optimize(const struct hstate *h, struct page *head); > void hugetlb_vmemmap_optimize_folios(struct hstate *h, struct list_head *folio_list); > > @@ -45,6 +48,13 @@ static inline int hugetlb_vmemmap_restore(const struct hstate *h, struct page *h > return 0; > } > > +static int hugetlb_vmemmap_restore_folios(const struct hstate *h, > + struct list_head *folio_list, > + struct list_head *non_op_folios) > +{ > + return 0; > +} > + > static inline void hugetlb_vmemmap_optimize(const struct hstate *h, struct page *head) > { > }