From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by smtp.subspace.kernel.org (Postfix) with ESMTP id 612DF155C97 for ; Wed, 17 Dec 2025 07:23:55 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=217.140.110.172 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1765956238; cv=none; b=uk4D90h5/IjevijcDqJ+sLOzxriNiUyF2RtSLFV75wEqkS6pNTZyJpgo16Rc96av8yrmjk6MYmh8+BrMTze/+wlTee9o1UcnQlwD5UG+Is4iR4/XE9BIHfyhAHwgoc7SB/csYf0O2tDQamhg7fOSguJYGd+bnBOE3pB8b+48DA0= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1765956238; c=relaxed/simple; bh=vgRpvePxGoVV4YlwIyS2Oquq2eRhgUkEpMbAKegj3fo=; h=Message-ID:Date:MIME-Version:Subject:To:Cc:References:From: In-Reply-To:Content-Type; b=I/KPgzyDQ3l0evkr+idP34LmUir010Yn1Wq+tkpYs53E8Wdg8DU03zlusrA3prIu35/nog4TNeEj0q0U/Yfl3mS7kVs7qmgWgf+ytEoPaFiwQZxqh3E+OFwMkrNqqTsFdOQ8FwQJEE6KIjtORhkl4N6tAY+66g2OVJvBqUU8Bq4= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com; spf=pass smtp.mailfrom=arm.com; arc=none smtp.client-ip=217.140.110.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=arm.com Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 1C4ADFEC; Tue, 16 Dec 2025 23:23:47 -0800 (PST) Received: from [10.164.18.63] (MacBook-Pro.blr.arm.com [10.164.18.63]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id 1803C3F73F; Tue, 16 Dec 2025 23:23:48 -0800 (PST) Message-ID: Date: Wed, 17 Dec 2025 12:53:46 +0530 Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH v2 2/3] mm: rmap: support batched checks of the references for large folios To: Baolin Wang , akpm@linux-foundation.org, david@kernel.org, catalin.marinas@arm.com, will@kernel.org Cc: lorenzo.stoakes@oracle.com, ryan.roberts@arm.com, Liam.Howlett@oracle.com, vbabka@suse.cz, rppt@kernel.org, surenb@google.com, mhocko@suse.com, riel@surriel.com, harry.yoo@oracle.com, jannh@google.com, willy@infradead.org, baohua@kernel.org, linux-mm@kvack.org, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org References: <545dba5e899634bc6c8ca782417d16fef3bd049f.1765439381.git.baolin.wang@linux.alibaba.com> <52ed7e12-b32e-4a73-ba55-eed993b930b9@arm.com> <753ee7bd-8c9a-4242-a216-98defcd8280f@linux.alibaba.com> Content-Language: en-US From: Dev Jain In-Reply-To: <753ee7bd-8c9a-4242-a216-98defcd8280f@linux.alibaba.com> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On 17/12/25 12:39 pm, Baolin Wang wrote: > > > On 2025/12/17 14:49, Dev Jain wrote: >> >> On 11/12/25 1:46 pm, Baolin Wang wrote: >>> Currently, folio_referenced_one() always checks the young flag for each PTE >>> sequentially, which is inefficient for large folios. This inefficiency is >>> especially noticeable when reclaiming clean file-backed large folios, where >>> folio_referenced() is observed as a significant performance hotspot. >>> >>> Moreover, on Arm architecture, which supports contiguous PTEs, there is already >>> an optimization to clear the young flags for PTEs within a contiguous range. >>> However, this is not sufficient. We can extend this to perform batched >>> operations >>> for the entire large folio (which might exceed the contiguous range: >>> CONT_PTE_SIZE). >>> >>> Introduce a new API: clear_flush_young_ptes() to facilitate batched checking >>> of the young flags and flushing TLB entries, thereby improving performance >>> during large folio reclamation. >>> >>> Performance testing: >>> Allocate 10G clean file-backed folios by mmap() in a memory cgroup, and try to >>> reclaim 8G file-backed folios via the memory.reclaim interface. I can observe >>> 33% performance improvement on my Arm64 32-core server (and 10%+ improvement >>> on my X86 machine). Meanwhile, the hotspot folio_check_references() dropped >>> from approximately 35% to around 5%. >>> >>> W/o patchset: >>> real    0m1.518s >>> user    0m0.000s >>> sys    0m1.518s >>> >>> W/ patchset: >>> real    0m1.018s >>> user    0m0.000s >>> sys    0m1.018s >>> >>> Signed-off-by: Baolin Wang >>> --- >>>   arch/arm64/include/asm/pgtable.h | 11 +++++++++++ >>>   include/linux/mmu_notifier.h     |  9 +++++---- >>>   include/linux/pgtable.h          | 19 +++++++++++++++++++ >>>   mm/rmap.c                        | 22 ++++++++++++++++++++-- >>>   4 files changed, 55 insertions(+), 6 deletions(-) >>> >>> diff --git a/arch/arm64/include/asm/pgtable.h >>> b/arch/arm64/include/asm/pgtable.h >>> index e03034683156..a865bd8c46a3 100644 >>> --- a/arch/arm64/include/asm/pgtable.h >>> +++ b/arch/arm64/include/asm/pgtable.h >>> @@ -1869,6 +1869,17 @@ static inline int ptep_clear_flush_young(struct >>> vm_area_struct *vma, >>>       return contpte_clear_flush_young_ptes(vma, addr, ptep, CONT_PTES); >>>   } >>>   +#define clear_flush_young_ptes clear_flush_young_ptes >>> +static inline int clear_flush_young_ptes(struct vm_area_struct *vma, >>> +                    unsigned long addr, pte_t *ptep, >>> +                    unsigned int nr) >>> +{ >>> +    if (likely(nr == 1)) >>> +        return __ptep_clear_flush_young(vma, addr, ptep); >>> + >>> +    return contpte_clear_flush_young_ptes(vma, addr, ptep, nr); >>> +} >>> + >>>   #define wrprotect_ptes wrprotect_ptes >>>   static __always_inline void wrprotect_ptes(struct mm_struct *mm, >>>                   unsigned long addr, pte_t *ptep, unsigned int nr) >>> diff --git a/include/linux/mmu_notifier.h b/include/linux/mmu_notifier.h >>> index d1094c2d5fb6..be594b274729 100644 >>> --- a/include/linux/mmu_notifier.h >>> +++ b/include/linux/mmu_notifier.h >>> @@ -515,16 +515,17 @@ static inline void mmu_notifier_range_init_owner( >>>       range->owner = owner; >>>   } >>>   -#define ptep_clear_flush_young_notify(__vma, __address, __ptep)        \ >>> +#define ptep_clear_flush_young_notify(__vma, __address, __ptep, __nr)    \ >>>   ({                                    \ >>>       int __young;                            \ >>>       struct vm_area_struct *___vma = __vma;                \ >>>       unsigned long ___address = __address;                \ >>> -    __young = ptep_clear_flush_young(___vma, ___address, __ptep);    \ >>> +    unsigned int ___nr = __nr;                    \ >>> +    __young = clear_flush_young_ptes(___vma, ___address, __ptep, ___nr);    \ >>>       __young |= mmu_notifier_clear_flush_young(___vma->vm_mm,    \ >>>                             ___address,        \ >>>                             ___address +        \ >>> -                            PAGE_SIZE);    \ >>> +                        nr * PAGE_SIZE);    \ >>>       __young;                            \ >>>   }) >>>   @@ -650,7 +651,7 @@ static inline void >>> mmu_notifier_subscriptions_destroy(struct mm_struct *mm) >>>     #define mmu_notifier_range_update_to_read_only(r) false >>>   -#define ptep_clear_flush_young_notify ptep_clear_flush_young >>> +#define ptep_clear_flush_young_notify clear_flush_young_ptes >>>   #define pmdp_clear_flush_young_notify pmdp_clear_flush_young >>>   #define ptep_clear_young_notify ptep_test_and_clear_young >>>   #define pmdp_clear_young_notify pmdp_test_and_clear_young >>> diff --git a/include/linux/pgtable.h b/include/linux/pgtable.h >>> index b13b6f42be3c..c7d0fd228cb7 100644 >>> --- a/include/linux/pgtable.h >>> +++ b/include/linux/pgtable.h >>> @@ -947,6 +947,25 @@ static inline void wrprotect_ptes(struct mm_struct *mm, >>> unsigned long addr, >>>   } >>>   #endif >>>   +#ifndef clear_flush_young_ptes >>> +static inline int clear_flush_young_ptes(struct vm_area_struct *vma, >>> +                     unsigned long addr, pte_t *ptep, >>> +                     unsigned int nr) >>> +{ >>> +    int young = 0; >>> + >>> +    for (;;) { >>> +        young |= ptep_clear_flush_young(vma, addr, ptep); >>> +        if (--nr == 0) >>> +            break; >>> +        ptep++; >>> +        addr += PAGE_SIZE; >>> +    } >>> + >>> +    return young; >>> +} >>> +#endif >>> + >>>   /* >>>    * On some architectures hardware does not set page access bit when accessing >>>    * memory page, it is responsibility of software setting this bit. It brings >>> diff --git a/mm/rmap.c b/mm/rmap.c >>> index d6799afe1114..ec232165c47d 100644 >>> --- a/mm/rmap.c >>> +++ b/mm/rmap.c >>> @@ -827,9 +827,11 @@ static bool folio_referenced_one(struct folio *folio, >>>       struct folio_referenced_arg *pra = arg; >>>       DEFINE_FOLIO_VMA_WALK(pvmw, folio, vma, address, 0); >>>       int ptes = 0, referenced = 0; >>> +    unsigned int nr; >>>         while (page_vma_mapped_walk(&pvmw)) { >>>           address = pvmw.address; >>> +        nr = 1; >>>             if (vma->vm_flags & VM_LOCKED) { >>>               ptes++; >>> @@ -874,9 +876,21 @@ static bool folio_referenced_one(struct folio *folio, >>>               if (lru_gen_look_around(&pvmw)) >>>                   referenced++; >>>           } else if (pvmw.pte) { >>> +            if (folio_test_large(folio)) { >>> +                unsigned long end_addr = pmd_addr_end(address, vma->vm_end); >> >> I may be hallucinating here but I am just trying to recall things - is this a >> bug in >> folio_pte_batch_flags()? A folio may not be naturally aligned in virtual >> space and hence >> we may cross the PTE table while batching across it, which can be fixed by >> taking into >> account pmd_addr_end() while computing max_nr. > > IMHO, the comments for the folio_pte_batch_flags() function have already made > clear requirements for the caller to avoid such situations: > > " > * @ptep must map any page of the folio. max_nr must be at least one and > * must be limited by the caller so scanning cannot exceed a single VMA and > * a single page table. > " > > Additionally, Lance recently fixed a similar issue, see commit ddd05742b45b > ("mm/rmap: fix potential out-of-bounds page table access during batched unmap").   Ah I see, all other users of the folio_pte_batch API constrain start and end because they are already operating on a single PTE table. But for rmap code this may not be the case.