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 298ECC2BD09 for ; Mon, 1 Jul 2024 09:43:50 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 9DE7B6B00A6; Mon, 1 Jul 2024 05:43:49 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 98EBF6B00A8; Mon, 1 Jul 2024 05:43:49 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 82F2B6B00AA; Mon, 1 Jul 2024 05:43:49 -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 63AFC6B00A6 for ; Mon, 1 Jul 2024 05:43:49 -0400 (EDT) Received: from smtpin06.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay07.hostedemail.com (Postfix) with ESMTP id 0EDC3161A94 for ; Mon, 1 Jul 2024 09:43:49 +0000 (UTC) X-FDA: 82290696978.06.B82F372 Received: from out0-207.mail.aliyun.com (out0-207.mail.aliyun.com [140.205.0.207]) by imf11.hostedemail.com (Postfix) with ESMTP id EC60440023 for ; Mon, 1 Jul 2024 09:43:45 +0000 (UTC) Authentication-Results: imf11.hostedemail.com; dkim=pass header.d=antgroup.com header.s=default header.b=vQegFkf8; dmarc=pass (policy=quarantine) header.from=antgroup.com; spf=pass (imf11.hostedemail.com: domain of libang.li@antgroup.com designates 140.205.0.207 as permitted sender) smtp.mailfrom=libang.li@antgroup.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1719827003; 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=PKR6sxM7W5GGchw1blJuJyfZABT5gHN6PYWv8xcaq5E=; b=ZOnye7cCfMmpmhRfJ4ZA6T7EtjiVr1Uwt8fcZ1WALKLCiabMP9wVeyi3sp0XC8fkoHQ3A4 WoYT6TGHoRnTuJwotOnoSCtACflMg+d1yMf7FpzOjvcYQgNOJHoPa0pbap+jJV4rFBndz5 F2cvQDYbc4zhBZD6pKhJT0HJAdvJnQc= ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1719827003; a=rsa-sha256; cv=none; b=VNqGgxyFphY+Tq/6esszLtx0NwTYzDbMM1V/wXgAqhRO0zm0lXwfyJ0QAvKQxHjUI5QNO7 WuM5gLo3uKY03qYZo89tD8IGgKOceEUiHL+dMKS8yM72PQN0VZ4xOBNHKmLcF65luiqxQl KJkNRsKV6ehg+lW2sTEj2DuxyvHNUE0= ARC-Authentication-Results: i=1; imf11.hostedemail.com; dkim=pass header.d=antgroup.com header.s=default header.b=vQegFkf8; dmarc=pass (policy=quarantine) header.from=antgroup.com; spf=pass (imf11.hostedemail.com: domain of libang.li@antgroup.com designates 140.205.0.207 as permitted sender) smtp.mailfrom=libang.li@antgroup.com DKIM-Signature:v=1; a=rsa-sha256; c=relaxed/relaxed; d=antgroup.com; s=default; t=1719827022; h=Message-ID:Date:MIME-Version:Subject:To:From:Content-Type; bh=PKR6sxM7W5GGchw1blJuJyfZABT5gHN6PYWv8xcaq5E=; b=vQegFkf8VuHemutQ/3HI+nFLtaYR3id6R38ggVTT7vNeOL9P7ry1FqCt9ZcdRiDbLxgNXbV6UCJlDS5r7Bp4H5kGwu6xzU3L9reMaf+x0QODozRIi+ASJOzIGaEGLXcIz9FUEKcCD8GllZ+X2Po3aDAr2BNew9JiyoA81phZl5E= X-Alimail-AntiSpam:AC=PASS;BC=-1|-1;BR=01201311R191e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=maildocker-contentspam033037088118;MF=libang.li@antgroup.com;NM=1;PH=DS;RN=9;SR=0;TI=SMTPD_---.YEDLAmw_1719827021; Received: from 30.13.185.168(mailfrom:libang.li@antgroup.com fp:SMTPD_---.YEDLAmw_1719827021) by smtp.aliyun-inc.com; Mon, 01 Jul 2024 17:43:41 +0800 Message-ID: Date: Mon, 01 Jul 2024 17:43:39 +0800 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH] support "THPeligible" semantics for mTHP with anonymous shmem To: Baolin Wang , Ryan Roberts , hughd@google.com, akpm@linux-foundation.org Cc: david@redhat.com, wangkefeng.wang@huawei.com, ziy@nvidia.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org References: <20240628104926.34209-1-libang.li@antgroup.com> <4b38db15-0716-4ffb-a38b-bd6250eb93da@arm.com> <4d54880e-03f4-460a-94b9-e21b8ad13119@linux.alibaba.com> Content-Language: en-US From: "Bang Li" In-Reply-To: <4d54880e-03f4-460a-94b9-e21b8ad13119@linux.alibaba.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-Rspamd-Server: rspam07 X-Rspamd-Queue-Id: EC60440023 X-Stat-Signature: tz7waccqd6yprj88rqu1883umoekcuri X-Rspam-User: X-HE-Tag: 1719827025-822625 X-HE-Meta: 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 ZblwtsDI sGjlrVkgGNTCqW3M5R14N3M69RxJSdvxN1Vrfn4OFj+HyZIjESfAtMsG7ZvxJ1TTYMoR5elDhmaHIdiLGiplxiZMXRTA3/hg982j1fraG0B/O2Nzes59kqjVvgIoyLLcgM28y+C82PvVkpNr61FZ6vuKeGO4JoTOBmEHBON0INObriDb6+61Kcsk0y35rDe3mP3V0xh4p6eCuBA1ZqhSKVBYhn5OSp4uswWLKsqbkGj3QHcxE9wRwwxSDvE0q2VmgbIYsamMT1g0rIyc1JB4/k7h0aFHsz8e9ukjzHCAH9vRuSdjYZ+6B1n7NAkwWEfGJLLQM38K/73KwhwAOu2GEW+9M8QRaG7G8afeEjz3SL5hlLQ0= 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: List-Subscribe: List-Unsubscribe: Hi, Baolin On 2024/7/1 16:33, Baolin Wang wrote: > > > On 2024/7/1 15:55, Ryan Roberts wrote: >> On 28/06/2024 11:49, Bang Li wrote: >>> After the commit 7fb1b252afb5 ("mm: shmem: add mTHP support for >>> anonymous shmem"), we can configure different policies through >>> the multi-size THP sysfs interface for anonymous shmem. But >>> currently "THPeligible" indicates only whether the mapping is >>> eligible for allocating THP-pages as well as the THP is PMD >>> mappable or not for anonymous shmem, we need to support semantics >>> for mTHP with anonymous shmem similar to those for mTHP with >>> anonymous memory. >>> >>> Signed-off-by: Bang Li >>> --- >>>   fs/proc/task_mmu.c      | 10 +++++++--- >>>   include/linux/huge_mm.h | 11 +++++++++++ >>>   mm/shmem.c              |  9 +-------- >>>   3 files changed, 19 insertions(+), 11 deletions(-) >>> >>> diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c >>> index 93fb2c61b154..09b5db356886 100644 >>> --- a/fs/proc/task_mmu.c >>> +++ b/fs/proc/task_mmu.c >>> @@ -870,6 +870,7 @@ static int show_smap(struct seq_file *m, void *v) >>>   { >>>       struct vm_area_struct *vma = v; >>>       struct mem_size_stats mss = {}; >>> +    bool thp_eligible; >>>       smap_gather_stats(vma, &mss, 0); >>> @@ -882,9 +883,12 @@ static int show_smap(struct seq_file *m, void *v) >>>       __show_smap(m, &mss, false); >>> -    seq_printf(m, "THPeligible:    %8u\n", >>> -           !!thp_vma_allowable_orders(vma, vma->vm_flags, >>> -               TVA_SMAPS | TVA_ENFORCE_SYSFS, THP_ORDERS_ALL)); >>> +    thp_eligible = !!thp_vma_allowable_orders(vma, vma->vm_flags, >>> +                        TVA_SMAPS | TVA_ENFORCE_SYSFS, THP_ORDERS_ALL); >>> +    if (vma_is_anon_shmem(vma)) >>> +        thp_eligible = >>> !!shmem_allowable_huge_orders(file_inode(vma->vm_file), >>> +                            vma, vma->vm_pgoff, thp_eligible); >> >> Afraid I haven't been following the shmem mTHP support work as much as >> I would >> have liked, but is there a reason why we need a separate function for >> shmem? > > Since shmem_allowable_huge_orders() only uses shmem specific logic to > determine if huge orders are allowable, there is no need to complicate > the thp_vma_allowable_orders() function by adding more shmem related > logic, making it more bloated. In my view, providing a dedicated helper > shmem_allowable_huge_orders(), specifically for shmem, simplifies the > logic. > > IIUC, I agree with David's suggestion that the > shmem_allowable_huge_orders() helper function could be used in > thp_vma_allowable_orders() to support shmem mTHP. Something like: > > diff --git a/mm/huge_memory.c b/mm/huge_memory.c > index c7ce28f6b7f3..9677fe6cf478 100644 > --- a/mm/huge_memory.c > +++ b/mm/huge_memory.c > @@ -151,10 +151,13 @@ unsigned long __thp_vma_allowable_orders(struct > vm_area_struct *vma, >          * Must be done before hugepage flags check since shmem has its >          * own flags. >          */ > -       if (!in_pf && shmem_file(vma->vm_file)) > -               return shmem_is_huge(file_inode(vma->vm_file), > vma->vm_pgoff, > -                                    !enforce_sysfs, vma->vm_mm, vm_flags) > -                       ? orders : 0; > +       if (!in_pf && shmem_file(vma->vm_file)) { > +               bool global_huge = > shmem_is_huge(file_inode(vma->vm_file), vma->vm_pgoff, > +                                    !enforce_sysfs, vma->vm_mm, vm_flags); > + > +               return > shmem_allowable_huge_orders(file_inode(vma->vm_file), > +                                       vma, vma->vm_pgoff, global_huge); > +       } > >         if (!vma_is_anonymous(vma)) { >                 /* > >> Couldn't (shouldn't) thp_vma_allowable_orders() be taught to handle >> shmem too? >> >>> +    seq_printf(m, "THPeligible:    %8u\n", thp_eligible); >>>       if (arch_pkeys_enabled()) >>>           seq_printf(m, "ProtectionKey:  %8u\n", vma_pkey(vma)); >>> diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h >>> index 212cca384d7e..f87136f38aa1 100644 >>> --- a/include/linux/huge_mm.h >>> +++ b/include/linux/huge_mm.h >>> @@ -267,6 +267,10 @@ unsigned long thp_vma_allowable_orders(struct >>> vm_area_struct *vma, >>>       return __thp_vma_allowable_orders(vma, vm_flags, tva_flags, >>> orders); >>>   } >>> +unsigned long shmem_allowable_huge_orders(struct inode *inode, >>> +                struct vm_area_struct *vma, pgoff_t index, >>> +                bool global_huge); >>> + >>>   struct thpsize { >>>       struct kobject kobj; >>>       struct list_head node; >>> @@ -460,6 +464,13 @@ static inline unsigned long >>> thp_vma_allowable_orders(struct vm_area_struct *vma, >>>       return 0; >>>   } >>> +static inline unsigned long shmem_allowable_huge_orders(struct inode >>> *inode, >>> +                struct vm_area_struct *vma, pgoff_t index, >>> +                bool global_huge) >>> +{ >>> +    return 0; >>> +} >>> + >>>   #define transparent_hugepage_flags 0UL >>>   #define thp_get_unmapped_area    NULL >>> diff --git a/mm/shmem.c b/mm/shmem.c >>> index d495c0701a83..aa85df9c662a 100644 >>> --- a/mm/shmem.c >>> +++ b/mm/shmem.c >>> @@ -1622,7 +1622,7 @@ static gfp_t limit_gfp_mask(gfp_t huge_gfp, >>> gfp_t limit_gfp) >>>   } >>>   #ifdef CONFIG_TRANSPARENT_HUGEPAGE >>> -static unsigned long shmem_allowable_huge_orders(struct inode *inode, >>> +unsigned long shmem_allowable_huge_orders(struct inode *inode, >>>                   struct vm_area_struct *vma, pgoff_t index, >>>                   bool global_huge) >>>   { >>> @@ -1707,13 +1707,6 @@ static unsigned long >>> shmem_suitable_orders(struct inode *inode, struct vm_fault >>>       return orders; >>>   } >>>   #else >>> -static unsigned long shmem_allowable_huge_orders(struct inode *inode, >>> -                struct vm_area_struct *vma, pgoff_t index, >>> -                bool global_huge) >>> -{ >>> -    return 0; >>> -} >>> - >>>   static unsigned long shmem_suitable_orders(struct inode *inode, >>> struct vm_fault *vmf, >>>                          struct address_space *mapping, pgoff_t index, >>>                          unsigned long orders) Thanks for the reference code. Currently, we only implement the mTHP of anonymous shmem, so we only need to handle anonymous shmem specially. As shown in the following code: --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -151,10 +151,14 @@ unsigned long __thp_vma_allowable_orders(struct vm_area_struct *vma, * Must be done before hugepage flags check since shmem has its * own flags. */ - if (!in_pf && shmem_file(vma->vm_file)) - return shmem_is_huge(file_inode(vma->vm_file), vma->vm_pgoff, - !enforce_sysfs, vma->vm_mm, vm_flags) - ? orders : 0; + if (!in_pf && shmem_file(vma->vm_file)) { + bool global_huge = shmem_is_huge(file_inode(vma->vm_file), vma->vm_pgoff, + !enforce_sysfs, vma->vm_mm, vm_flags); + if (!vma_is_anon_shmem(vma)) + return global_huge? orders : 0; + return shmem_allowable_huge_orders(file_inode(vma->vm_file), + vma, vma->vm_pgoff, global_huge); + } if (!vma_is_anonymous(vma)) { Thanks, Bang