Message-ID: <cb321fb0-fde5-c78f-10b2-1d1fec7070fd@arm.com>
Date:   Tue, 18 Jul 2023 11:27:53 +0100
From:   Ryan Roberts <ryan.roberts@....com>
To:     Yin Fengwei <fengwei.yin@...el.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Matthew Wilcox <willy@...radead.org>,
        David Hildenbrand <david@...hat.com>,
        Yu Zhao <yuzhao@...gle.com>, Yang Shi <shy828301@...il.com>,
        "Huang, Ying" <ying.huang@...el.com>, Zi Yan <ziy@...dia.com>
Cc:     linux-kernel@...r.kernel.org, linux-mm@...ck.org
Subject: Re: [PATCH v1 3/3] mm: Batch-zap large anonymous folio PTE mappings

On 18/07/2023 00:27, Yin Fengwei wrote:
> 
> 
> On 7/17/23 22:31, Ryan Roberts wrote:
>> This allows batching the rmap removal with folio_remove_rmap_range(),
>> which means we avoid spuriously adding a partially unmapped folio to the
>> deferrred split queue in the common case, which reduces split queue lock
>> contention.
>>
>> Previously each page was removed from the rmap individually with
>> page_remove_rmap(). If the first page belonged to a large folio, this
>> would cause page_remove_rmap() to conclude that the folio was now
>> partially mapped and add the folio to the deferred split queue. But
>> subsequent calls would cause the folio to become fully unmapped, meaning
>> there is no value to adding it to the split queue.
>>
>> Signed-off-by: Ryan Roberts <ryan.roberts@....com>
>> ---
>>  mm/memory.c | 119 ++++++++++++++++++++++++++++++++++++++++++++++++++++
>>  1 file changed, 119 insertions(+)
>>
>> diff --git a/mm/memory.c b/mm/memory.c
>> index 01f39e8144ef..6facb8c8807a 100644
>> --- a/mm/memory.c
>> +++ b/mm/memory.c
>> @@ -1391,6 +1391,95 @@ zap_install_uffd_wp_if_needed(struct vm_area_struct *vma,
>>  	pte_install_uffd_wp_if_needed(vma, addr, pte, pteval);
>>  }
>>  
>> +static inline unsigned long page_addr(struct page *page,
>> +				struct page *anchor, unsigned long anchor_addr)
>> +{
>> +	unsigned long offset;
>> +	unsigned long addr;
>> +
>> +	offset = (page_to_pfn(page) - page_to_pfn(anchor)) << PAGE_SHIFT;
>> +	addr = anchor_addr + offset;
>> +
>> +	if (anchor > page) {
>> +		if (addr > anchor_addr)
>> +			return 0;
>> +	} else {
>> +		if (addr < anchor_addr)
>> +			return ULONG_MAX;
>> +	}
>> +
>> +	return addr;
>> +}
>> +
>> +static int calc_anon_folio_map_pgcount(struct folio *folio,
>> +				       struct page *page, pte_t *pte,
>> +				       unsigned long addr, unsigned long end)
>> +{
>> +	pte_t ptent;
>> +	int floops;
>> +	int i;
>> +	unsigned long pfn;
>> +
>> +	end = min(page_addr(&folio->page + folio_nr_pages(folio), page, addr),
>> +		  end);
>> +	floops = (end - addr) >> PAGE_SHIFT;
>> +	pfn = page_to_pfn(page);
>> +	pfn++;
>> +	pte++;
>> +
>> +	for (i = 1; i < floops; i++) {
>> +		ptent = ptep_get(pte);
>> +
>> +		if (!pte_present(ptent) ||
>> +		    pte_pfn(ptent) != pfn) {
>> +			return i;
>> +		}
>> +
>> +		pfn++;
>> +		pte++;
>> +	}
>> +
>> +	return floops;
>> +}
>> +
>> +static unsigned long zap_anon_pte_range(struct mmu_gather *tlb,
>> +					struct vm_area_struct *vma,
>> +					struct page *page, pte_t *pte,
>> +					unsigned long addr, unsigned long end,
>> +					bool *full_out)
>> +{
>> +	struct folio *folio = page_folio(page);
>> +	struct mm_struct *mm = tlb->mm;
>> +	pte_t ptent;
>> +	int pgcount;
>> +	int i;
>> +	bool full;
>> +
>> +	pgcount = calc_anon_folio_map_pgcount(folio, page, pte, addr, end);
>> +
>> +	for (i = 0; i < pgcount;) {
>> +		ptent = ptep_get_and_clear_full(mm, addr, pte, tlb->fullmm);
>> +		tlb_remove_tlb_entry(tlb, pte, addr);
>> +		full = __tlb_remove_page(tlb, page, 0);
>> +
>> +		if (unlikely(page_mapcount(page) < 1))
>> +			print_bad_pte(vma, addr, ptent, page);
>> +
>> +		i++;
>> +		page++;
>> +		pte++;
>> +		addr += PAGE_SIZE;
>> +
>> +		if (unlikely(full))
>> +			break;
>> +	}
>> +
>> +	folio_remove_rmap_range(folio, page - i, i, vma);
>> +
>> +	*full_out = full;
>> +	return i;
>> +}
>> +
>>  static unsigned long zap_pte_range(struct mmu_gather *tlb,
>>  				struct vm_area_struct *vma, pmd_t *pmd,
>>  				unsigned long addr, unsigned long end,
>> @@ -1428,6 +1517,36 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb,
>>  			page = vm_normal_page(vma, addr, ptent);
>>  			if (unlikely(!should_zap_page(details, page)))
>>  				continue;
>> +
>> +			/*
>> +			 * Batch zap large anonymous folio mappings. This allows
>> +			 * batching the rmap removal, which means we avoid
>> +			 * spuriously adding a partially unmapped folio to the
>> +			 * deferrred split queue in the common case, which
>> +			 * reduces split queue lock contention. Require the VMA
>> +			 * to be anonymous to ensure that none of the PTEs in
>> +			 * the range require zap_install_uffd_wp_if_needed().
>> +			 */
>> +			if (page && PageAnon(page) && vma_is_anonymous(vma)) {
> Why this is only for anonymous page? I suppose it can support file mapping also.

I was trying to avoid the complexity. For file-backed pages, there is a bunch of
dirty and access management stuff that needs to happen (see "if
(!PageAnon(page)) {" a bit further down). And for file-backed VMAs (even if the
page is anon, I think?) zap_install_uffd_wp_if_needed() might do some extra
work, which again I didn't want to have to drag into zap_anon_pte_range().

I guess it's implementable, but given only anon folios will be deferred-split
and anon folios in a file-backed vma will all be single page, I didn't feel that
the extra complexity would add anything performance-wise.


> 
> 
> Regards
> Yin, Fengwei
> 
>> +				bool full;
>> +				int pgcount;
>> +
>> +				pgcount = zap_anon_pte_range(tlb, vma,
>> +						page, pte, addr, end, &full);
>> +
>> +				rss[mm_counter(page)] -= pgcount;
>> +				pgcount--;
>> +				pte += pgcount;
>> +				addr += pgcount << PAGE_SHIFT;
>> +
>> +				if (unlikely(full)) {
>> +					force_flush = 1;
>> +					addr += PAGE_SIZE;
>> +					break;
>> +				}
>> +				continue;
>> +			}
>> +
>>  			ptent = ptep_get_and_clear_full(mm, addr, pte,
>>  							tlb->fullmm);
>>  			tlb_remove_tlb_entry(tlb, pte, addr);

Powered by blists - more mailing lists