From patchwork Thu Jul 1 01:47:25 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Andrew Morton X-Patchwork-Id: 12353073 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-15.8 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS autolearn=unavailable autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 6B3DFC11F69 for ; Thu, 1 Jul 2021 01:47:29 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id EDE3A6120C for ; Thu, 1 Jul 2021 01:47:28 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org EDE3A6120C Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=linux-foundation.org Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id 756CE8D01DA; Wed, 30 Jun 2021 21:47:28 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 706218D01D0; Wed, 30 Jun 2021 21:47:28 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 5A7268D01DA; Wed, 30 Jun 2021 21:47:28 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0167.hostedemail.com [216.40.44.167]) by kanga.kvack.org (Postfix) with ESMTP id 2F4768D01D0 for ; Wed, 30 Jun 2021 21:47:28 -0400 (EDT) Received: from smtpin03.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay01.hostedemail.com (Postfix) with ESMTP id E49A9180ABF62 for ; Thu, 1 Jul 2021 01:47:27 +0000 (UTC) X-FDA: 78312331734.03.E6775DE Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by imf22.hostedemail.com (Postfix) with ESMTP id 891181983 for ; Thu, 1 Jul 2021 01:47:27 +0000 (UTC) Received: by mail.kernel.org (Postfix) with ESMTPSA id C86A261469; Thu, 1 Jul 2021 01:47:25 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linux-foundation.org; s=korg; t=1625104046; bh=B/reZ1mSXT6NQEJmPCJE7MgtkRFoE/6D/4fnvFnQRoo=; h=Date:From:To:Subject:In-Reply-To:From; b=ktPjlGk8/Fo52jZ2zj6ErwyzCNpkW14fGRaEqrhGAXdvrDCLRZbnM0F7J7EwYFnS5 pV1DLJVoW3fjTP+rSznYH1pvgqxYUgHEjifzkYYaa1H++RQFRZfolJjacPtjGFW9Uh X/ZL3s6H2NL7RNEid/LJKI+6RTCk5NFtuJY0fobw= Date: Wed, 30 Jun 2021 18:47:25 -0700 From: Andrew Morton To: akpm@linux-foundation.org, almasrymina@google.com, anshuman.khandual@arm.com, bodeddub@amazon.com, bp@alien8.de, bsingharora@gmail.com, chenhuang5@huawei.com, corbet@lwn.net, dave.hansen@linux.intel.com, david@redhat.com, duanxiongchun@bytedance.com, hpa@zytor.com, joao.m.martins@oracle.com, jroedel@suse.de, linmiaohe@huawei.com, linux-mm@kvack.org, luto@kernel.org, mhocko@suse.com, mike.kravetz@oracle.com, mingo@redhat.com, mm-commits@vger.kernel.org, naoya.horiguchi@nec.com, oneukum@suse.com, osalvador@suse.de, paulmck@kernel.org, pawan.kumar.gupta@linux.intel.com, peterz@infradead.org, rdunlap@infradead.org, rientjes@google.com, song.bao.hua@hisilicon.com, songmuchun@bytedance.com, tglx@linutronix.de, torvalds@linux-foundation.org, viro@zeniv.linux.org.uk, willy@infradead.org Subject: [patch 007/192] mm: hugetlb: add a kernel parameter hugetlb_free_vmemmap Message-ID: <20210701014725.0NAzb7_it%akpm@linux-foundation.org> In-Reply-To: <20210630184624.9ca1937310b0dd5ce66b30e7@linux-foundation.org> User-Agent: s-nail v14.8.16 X-Rspamd-Server: rspam05 X-Rspamd-Queue-Id: 891181983 X-Stat-Signature: o33jx8rku5rqt5px8h65d7dca6c3gcfg Authentication-Results: imf22.hostedemail.com; dkim=pass header.d=linux-foundation.org header.s=korg header.b=ktPjlGk8; dmarc=none; spf=pass (imf22.hostedemail.com: domain of akpm@linux-foundation.org designates 198.145.29.99 as permitted sender) smtp.mailfrom=akpm@linux-foundation.org X-HE-Tag: 1625104047-750569 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: From: Muchun Song Subject: mm: hugetlb: add a kernel parameter hugetlb_free_vmemmap Add a kernel parameter hugetlb_free_vmemmap to enable the feature of freeing unused vmemmap pages associated with each hugetlb page on boot. We disable PMD mapping of vmemmap pages for x86-64 arch when this feature is enabled. Because vmemmap_remap_free() depends on vmemmap being base page mapped. Link: https://lkml.kernel.org/r/20210510030027.56044-8-songmuchun@bytedance.com Signed-off-by: Muchun Song Reviewed-by: Oscar Salvador Reviewed-by: Barry Song Reviewed-by: Miaohe Lin Tested-by: Chen Huang Tested-by: Bodeddula Balasubramaniam Reviewed-by: Mike Kravetz Cc: Alexander Viro Cc: Andy Lutomirski Cc: Anshuman Khandual Cc: Balbir Singh Cc: Borislav Petkov Cc: Dave Hansen Cc: David Hildenbrand Cc: David Rientjes Cc: HORIGUCHI NAOYA Cc: "H. Peter Anvin" Cc: Ingo Molnar Cc: Joao Martins Cc: Joerg Roedel Cc: Jonathan Corbet Cc: Matthew Wilcox Cc: Michal Hocko Cc: Mina Almasry Cc: Oliver Neukum Cc: Paul E. McKenney Cc: Pawan Gupta Cc: Peter Zijlstra Cc: Randy Dunlap Cc: Thomas Gleixner Cc: Xiongchun Duan Signed-off-by: Andrew Morton --- Documentation/admin-guide/kernel-parameters.txt | 17 +++++++++ Documentation/admin-guide/mm/hugetlbpage.rst | 3 + arch/x86/mm/init_64.c | 8 +++- include/linux/hugetlb.h | 19 +++++++++++ mm/hugetlb_vmemmap.c | 24 ++++++++++++++ 5 files changed, 69 insertions(+), 2 deletions(-) --- a/arch/x86/mm/init_64.c~mm-hugetlb-add-a-kernel-parameter-hugetlb_free_vmemmap +++ a/arch/x86/mm/init_64.c @@ -34,6 +34,7 @@ #include #include #include +#include #include #include @@ -1609,7 +1610,8 @@ int __meminit vmemmap_populate(unsigned VM_BUG_ON(!IS_ALIGNED(start, PAGE_SIZE)); VM_BUG_ON(!IS_ALIGNED(end, PAGE_SIZE)); - if (end - start < PAGES_PER_SECTION * sizeof(struct page)) + if ((is_hugetlb_free_vmemmap_enabled() && !altmap) || + end - start < PAGES_PER_SECTION * sizeof(struct page)) err = vmemmap_populate_basepages(start, end, node, NULL); else if (boot_cpu_has(X86_FEATURE_PSE)) err = vmemmap_populate_hugepages(start, end, node, altmap); @@ -1637,6 +1639,8 @@ void register_page_bootmem_memmap(unsign pmd_t *pmd; unsigned int nr_pmd_pages; struct page *page; + bool base_mapping = !boot_cpu_has(X86_FEATURE_PSE) || + is_hugetlb_free_vmemmap_enabled(); for (; addr < end; addr = next) { pte_t *pte = NULL; @@ -1662,7 +1666,7 @@ void register_page_bootmem_memmap(unsign } get_page_bootmem(section_nr, pud_page(*pud), MIX_SECTION_INFO); - if (!boot_cpu_has(X86_FEATURE_PSE)) { + if (base_mapping) { next = (addr + PAGE_SIZE) & PAGE_MASK; pmd = pmd_offset(pud, addr); if (pmd_none(*pmd)) --- a/Documentation/admin-guide/kernel-parameters.txt~mm-hugetlb-add-a-kernel-parameter-hugetlb_free_vmemmap +++ a/Documentation/admin-guide/kernel-parameters.txt @@ -1567,6 +1567,23 @@ Documentation/admin-guide/mm/hugetlbpage.rst. Format: size[KMG] + hugetlb_free_vmemmap= + [KNL] Reguires CONFIG_HUGETLB_PAGE_FREE_VMEMMAP + enabled. + Allows heavy hugetlb users to free up some more + memory (6 * PAGE_SIZE for each 2MB hugetlb page). + This feauture is not free though. Large page + tables are not used to back vmemmap pages which + can lead to a performance degradation for some + workloads. Also there will be memory allocation + required when hugetlb pages are freed from the + pool which can lead to corner cases under heavy + memory pressure. + Format: { on | off (default) } + + on: enable the feature + off: disable the feature + hung_task_panic= [KNL] Should the hung task detector generate panics. Format: 0 | 1 --- a/Documentation/admin-guide/mm/hugetlbpage.rst~mm-hugetlb-add-a-kernel-parameter-hugetlb_free_vmemmap +++ a/Documentation/admin-guide/mm/hugetlbpage.rst @@ -153,6 +153,9 @@ default_hugepagesz will all result in 256 2M huge pages being allocated. Valid default huge page size is architecture dependent. +hugetlb_free_vmemmap + When CONFIG_HUGETLB_PAGE_FREE_VMEMMAP is set, this enables freeing + unused vmemmap pages associated with each HugeTLB page. When multiple huge page sizes are supported, ``/proc/sys/vm/nr_hugepages`` indicates the current number of pre-allocated huge pages of the default size. --- a/include/linux/hugetlb.h~mm-hugetlb-add-a-kernel-parameter-hugetlb_free_vmemmap +++ a/include/linux/hugetlb.h @@ -892,6 +892,20 @@ static inline void huge_ptep_modify_prot } #endif +#ifdef CONFIG_HUGETLB_PAGE_FREE_VMEMMAP +extern bool hugetlb_free_vmemmap_enabled; + +static inline bool is_hugetlb_free_vmemmap_enabled(void) +{ + return hugetlb_free_vmemmap_enabled; +} +#else +static inline bool is_hugetlb_free_vmemmap_enabled(void) +{ + return false; +} +#endif + #else /* CONFIG_HUGETLB_PAGE */ struct hstate {}; @@ -1046,6 +1060,11 @@ static inline void set_huge_swap_pte_at( pte_t *ptep, pte_t pte, unsigned long sz) { } + +static inline bool is_hugetlb_free_vmemmap_enabled(void) +{ + return false; +} #endif /* CONFIG_HUGETLB_PAGE */ static inline spinlock_t *huge_pte_lock(struct hstate *h, --- a/mm/hugetlb_vmemmap.c~mm-hugetlb-add-a-kernel-parameter-hugetlb_free_vmemmap +++ a/mm/hugetlb_vmemmap.c @@ -168,6 +168,8 @@ * (last) level. So this type of HugeTLB page can be optimized only when its * size of the struct page structs is greater than 2 pages. */ +#define pr_fmt(fmt) "HugeTLB: " fmt + #include "hugetlb_vmemmap.h" /* @@ -180,6 +182,28 @@ #define RESERVE_VMEMMAP_NR 2U #define RESERVE_VMEMMAP_SIZE (RESERVE_VMEMMAP_NR << PAGE_SHIFT) +bool hugetlb_free_vmemmap_enabled; + +static int __init early_hugetlb_free_vmemmap_param(char *buf) +{ + /* We cannot optimize if a "struct page" crosses page boundaries. */ + if ((!is_power_of_2(sizeof(struct page)))) { + pr_warn("cannot free vmemmap pages because \"struct page\" crosses page boundaries\n"); + return 0; + } + + if (!buf) + return -EINVAL; + + if (!strcmp(buf, "on")) + hugetlb_free_vmemmap_enabled = true; + else if (strcmp(buf, "off")) + return -EINVAL; + + return 0; +} +early_param("hugetlb_free_vmemmap", early_hugetlb_free_vmemmap_param); + static inline unsigned long free_vmemmap_pages_size_per_hpage(struct hstate *h) { return (unsigned long)free_vmemmap_pages_per_hpage(h) << PAGE_SHIFT;