Message-ID: <2c1eb3ba-3f76-4796-b076-307a72cddc76@arm.com>
Date: Mon, 15 Apr 2024 09:47:15 +0100
From: Ryan Roberts <ryan.roberts@....com>
To: Lance Yang <ioworker0@...il.com>, akpm@...ux-foundation.org
Cc: zokeefe@...gle.com, 21cnbao@...il.com, shy828301@...il.com,
 david@...hat.com, mhocko@...e.com, fengwei.yin@...el.com,
 xiehuan09@...il.com, wangkefeng.wang@...wei.com, songmuchun@...edance.com,
 peterx@...hat.com, minchan@...nel.org, linux-mm@...ck.org,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH v6 1/2] mm/madvise: optimize lazyfreeing with mTHP in
 madvise_free

On 13/04/2024 01:22, Lance Yang wrote:
> This patch optimizes lazyfreeing with PTE-mapped mTHP[1]
> (Inspired by David Hildenbrand[2]). We aim to avoid unnecessary folio
> splitting if the large folio is fully mapped within the target range.
> 
> If a large folio is locked or shared, or if we fail to split it, we just
> leave it in place and advance to the next PTE in the range. But note that
> the behavior is changed; previously, any failure of this sort would cause
> the entire operation to give up. As large folios become more common,
> sticking to the old way could result in wasted opportunities.
> 
> On an Intel I5 CPU, lazyfreeing a 1GiB VMA backed by PTE-mapped folios of
> the same size results in the following runtimes for madvise(MADV_FREE) in
> seconds (shorter is better):
> 
> Folio Size |   Old    |   New    | Change
> ------------------------------------------
>       4KiB | 0.590251 | 0.590259 |    0%
>      16KiB | 2.990447 | 0.185655 |  -94%
>      32KiB | 2.547831 | 0.104870 |  -95%
>      64KiB | 2.457796 | 0.052812 |  -97%
>     128KiB | 2.281034 | 0.032777 |  -99%
>     256KiB | 2.230387 | 0.017496 |  -99%
>     512KiB | 2.189106 | 0.010781 |  -99%
>    1024KiB | 2.183949 | 0.007753 |  -99%
>    2048KiB | 0.002799 | 0.002804 |    0%
> 
> [1] https://lkml.kernel.org/r/20231207161211.2374093-5-ryan.roberts@arm.com
> [2] https://lore.kernel.org/linux-mm/20240214204435.167852-1-david@redhat.com
> 
> Signed-off-by: Lance Yang <ioworker0@...il.com>

This is looking close IMHO. Just one bug and suggestion below.


> ---
>  include/linux/mm_types.h |   9 +++
>  include/linux/pgtable.h  |  42 +++++++++++
>  mm/internal.h            |  12 +++-
>  mm/madvise.c             | 147 ++++++++++++++++++++++-----------------
>  mm/memory.c              |   4 +-
>  5 files changed, 147 insertions(+), 67 deletions(-)
> 
> diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
> index c432add95913..3c224e25f473 100644
> --- a/include/linux/mm_types.h
> +++ b/include/linux/mm_types.h
> @@ -1367,6 +1367,15 @@ enum fault_flag {
>  
>  typedef unsigned int __bitwise zap_flags_t;
>  
> +/* Flags for clear_young_dirty_ptes(). */
> +typedef int __bitwise cydp_t;
> +
> +/* make PTEs after pte_mkold() */
> +#define CYDP_CLEAR_YOUNG		((__force cydp_t)BIT(0))
> +
> +/* make PTEs after pte_mkclean() */
> +#define CYDP_CLEAR_DIRTY		((__force cydp_t)BIT(1))
> +
>  /*
>   * FOLL_PIN and FOLL_LONGTERM may be used in various combinations with each
>   * other. Here is what they mean, and how to use them:
> diff --git a/include/linux/pgtable.h b/include/linux/pgtable.h
> index e2f45e22a6d1..d7958243f099 100644
> --- a/include/linux/pgtable.h
> +++ b/include/linux/pgtable.h
> @@ -489,6 +489,48 @@ static inline pte_t ptep_get_and_clear(struct mm_struct *mm,
>  }
>  #endif
>  
> +#ifndef clear_young_dirty_ptes
> +/**
> + * clear_young_dirty_ptes - Mark PTEs that map consecutive pages of the
> + *		same folio as old/clean.
> + * @mm: Address space the pages are mapped into.
> + * @addr: Address the first page is mapped at.
> + * @ptep: Page table pointer for the first entry.
> + * @nr: Number of entries to mark old/clean.
> + * @flags: Flags to modify the PTE batch semantics.
> + *
> + * May be overridden by the architecture; otherwise, implemented by
> + * get_and_clear/modify/set for each pte in the range.
> + *
> + * Note that PTE bits in the PTE range besides the PFN can differ. For example,
> + * some PTEs might be write-protected.
> + *
> + * Context: The caller holds the page table lock.  The PTEs map consecutive
> + * pages that belong to the same folio.  The PTEs are all in the same PMD.
> + */
> +static inline void clear_young_dirty_ptes(struct mm_struct *mm,

My suggestion to introduce clear_young_dirty_ptes() was so that we could
*remove* mkold_ptes(). So I think it would be good to split that to a separate
preparatory patch, where you change all the callers to call the new function and
remove mkold_ptes().

Additionally since many arches already override ptep_test_and_clear_young() (and
that's what the default mkold_ptes() did, you might want to call that in the
below loop if *only* CYDP_CLEAR_YOUNG is set to avoid the possibility of any
regression. I know I only made this change for the swap-out series so what you
have done may well be ok in practice - and certainly cleaner. It would be good
to hear others' opinions.

Note the existing mkold_ptes() takes a vma instead of mm (because that's what
ptep_test_and_clear_young() takes). So suggest passing vma. You can get mm from
vma->mm.

> +					  unsigned long addr, pte_t *ptep,
> +					  unsigned int nr, cydp_t flags)
> +{
> +	pte_t pte;
> +
> +	for (;;) {

Suggestion:
		if (flags == CYDP_CLEAR_YOUNG) {
			ptep_test_and_clear_young(vma, addr, ptep);
		else {

> +		pte = ptep_get_and_clear(mm, addr, ptep);
> +
> +		if (flags | CYDP_CLEAR_YOUNG)

bug: this needs to be bitwise and (&). Currently it will always evaluate to
true. Same for next one.

> +			pte = pte_mkold(pte);
> +		if (flags | CYDP_CLEAR_DIRTY)
> +			pte = pte_mkclean(pte);
> +
> +		set_pte_at(mm, addr, ptep, pte);

		}

> +		if (--nr == 0)
> +			break;
> +		ptep++;
> +		addr += PAGE_SIZE;
> +	}
> +}
> +#endif
> +
>  static inline void ptep_clear(struct mm_struct *mm, unsigned long addr,
>  			      pte_t *ptep)
>  {
> diff --git a/mm/internal.h b/mm/internal.h
> index 3c0f3e3f9d99..ab8fcdeaf6eb 100644
> --- a/mm/internal.h
> +++ b/mm/internal.h
> @@ -134,6 +134,8 @@ static inline pte_t __pte_batch_clear_ignored(pte_t pte, fpb_t flags)
>   *		  first one is writable.
>   * @any_young: Optional pointer to indicate whether any entry except the
>   *		  first one is young.
> + * @any_dirty: Optional pointer to indicate whether any entry except the
> + *		  first one is dirty.
>   *
>   * Detect a PTE batch: consecutive (present) PTEs that map consecutive
>   * pages of the same large folio.
> @@ -149,18 +151,20 @@ static inline pte_t __pte_batch_clear_ignored(pte_t pte, fpb_t flags)
>   */
>  static inline int folio_pte_batch(struct folio *folio, unsigned long addr,
>  		pte_t *start_ptep, pte_t pte, int max_nr, fpb_t flags,
> -		bool *any_writable, bool *any_young)
> +		bool *any_writable, bool *any_young, bool *any_dirty)
>  {
>  	unsigned long folio_end_pfn = folio_pfn(folio) + folio_nr_pages(folio);
>  	const pte_t *end_ptep = start_ptep + max_nr;
>  	pte_t expected_pte, *ptep;
> -	bool writable, young;
> +	bool writable, young, dirty;
>  	int nr;
>  
>  	if (any_writable)
>  		*any_writable = false;
>  	if (any_young)
>  		*any_young = false;
> +	if (any_dirty)
> +		*any_dirty = false;
>  
>  	VM_WARN_ON_FOLIO(!pte_present(pte), folio);
>  	VM_WARN_ON_FOLIO(!folio_test_large(folio) || max_nr < 1, folio);
> @@ -176,6 +180,8 @@ static inline int folio_pte_batch(struct folio *folio, unsigned long addr,
>  			writable = !!pte_write(pte);
>  		if (any_young)
>  			young = !!pte_young(pte);
> +		if (any_dirty)
> +			dirty = !!pte_dirty(pte);
>  		pte = __pte_batch_clear_ignored(pte, flags);
>  
>  		if (!pte_same(pte, expected_pte))
> @@ -193,6 +199,8 @@ static inline int folio_pte_batch(struct folio *folio, unsigned long addr,
>  			*any_writable |= writable;
>  		if (any_young)
>  			*any_young |= young;
> +		if (any_dirty)
> +			*any_dirty |= dirty;
>  
>  		nr = pte_batch_hint(ptep, pte);
>  		expected_pte = pte_advance_pfn(expected_pte, nr);
> diff --git a/mm/madvise.c b/mm/madvise.c
> index d34ca6983227..b4103e2df346 100644
> --- a/mm/madvise.c
> +++ b/mm/madvise.c
> @@ -321,6 +321,39 @@ static inline bool can_do_file_pageout(struct vm_area_struct *vma)
>  	       file_permission(vma->vm_file, MAY_WRITE) == 0;
>  }
>  
> +static inline int madvise_folio_pte_batch(unsigned long addr, unsigned long end,
> +					  struct folio *folio, pte_t *ptep,
> +					  pte_t pte, bool *any_young,
> +					  bool *any_dirty)
> +{
> +	int max_nr = (end - addr) / PAGE_SIZE;
> +	const fpb_t fpb_flags = FPB_IGNORE_DIRTY | FPB_IGNORE_SOFT_DIRTY;
> +
> +	return folio_pte_batch(folio, addr, ptep, pte, max_nr, fpb_flags, NULL,
> +			       any_young, any_dirty);
> +}
> +
> +static inline bool madvise_pte_split_folio(struct mm_struct *mm, pmd_t *pmd,
> +					   unsigned long addr,
> +					   struct folio *folio, pte_t **pte,
> +					   spinlock_t **ptl)
> +{
> +	int err;
> +
> +	if (!folio_trylock(folio))
> +		return false;
> +
> +	folio_get(folio);
> +	pte_unmap_unlock(*pte, *ptl);
> +	err = split_folio(folio);
> +	folio_unlock(folio);
> +	folio_put(folio);
> +
> +	*pte = pte_offset_map_lock(mm, pmd, addr, ptl);
> +
> +	return err == 0;
> +}
> +
>  static int madvise_cold_or_pageout_pte_range(pmd_t *pmd,
>  				unsigned long addr, unsigned long end,
>  				struct mm_walk *walk)
> @@ -456,41 +489,30 @@ static int madvise_cold_or_pageout_pte_range(pmd_t *pmd,
>  		 * next pte in the range.
>  		 */
>  		if (folio_test_large(folio)) {
> -			const fpb_t fpb_flags = FPB_IGNORE_DIRTY |
> -						FPB_IGNORE_SOFT_DIRTY;
> -			int max_nr = (end - addr) / PAGE_SIZE;
>  			bool any_young;
>  
> -			nr = folio_pte_batch(folio, addr, pte, ptent, max_nr,
> -					     fpb_flags, NULL, &any_young);
> -			if (any_young)
> -				ptent = pte_mkyoung(ptent);
> +			nr = madvise_folio_pte_batch(addr, end, folio, pte,
> +						     ptent, &any_young, NULL);
>  
>  			if (nr < folio_nr_pages(folio)) {
> -				int err;
> -
>  				if (folio_likely_mapped_shared(folio))
>  					continue;
>  				if (pageout_anon_only_filter && !folio_test_anon(folio))
>  					continue;
> -				if (!folio_trylock(folio))
> -					continue;
> -				folio_get(folio);
> +
>  				arch_leave_lazy_mmu_mode();
> -				pte_unmap_unlock(start_pte, ptl);
> -				start_pte = NULL;
> -				err = split_folio(folio);
> -				folio_unlock(folio);
> -				folio_put(folio);
> -				start_pte = pte =
> -					pte_offset_map_lock(mm, pmd, addr, &ptl);
> +				if (madvise_pte_split_folio(mm, pmd, addr,
> +							    folio, &start_pte, &ptl))
> +					nr = 0;
>  				if (!start_pte)
>  					break;
> +				pte = start_pte;
>  				arch_enter_lazy_mmu_mode();
> -				if (!err)
> -					nr = 0;
>  				continue;
>  			}
> +
> +			if (any_young)
> +				ptent = pte_mkyoung(ptent);
>  		}
>  
>  		/*
> @@ -507,7 +529,8 @@ static int madvise_cold_or_pageout_pte_range(pmd_t *pmd,
>  			continue;
>  
>  		if (!pageout && pte_young(ptent)) {
> -			mkold_ptes(vma, addr, pte, nr);
> +			clear_young_dirty_ptes(mm, addr, pte, nr,
> +					       CYDP_CLEAR_YOUNG);
>  			tlb_remove_tlb_entries(tlb, pte, nr, addr);
>  		}
>  
> @@ -687,44 +710,51 @@ static int madvise_free_pte_range(pmd_t *pmd, unsigned long addr,
>  			continue;
>  
>  		/*
> -		 * If pmd isn't transhuge but the folio is large and
> -		 * is owned by only this process, split it and
> -		 * deactivate all pages.
> +		 * If we encounter a large folio, only split it if it is not
> +		 * fully mapped within the range we are operating on. Otherwise
> +		 * leave it as is so that it can be marked as lazyfree. If we
> +		 * fail to split a folio, leave it in place and advance to the
> +		 * next pte in the range.
>  		 */
>  		if (folio_test_large(folio)) {
> -			int err;
> +			bool any_young, any_dirty;
>  
> -			if (folio_likely_mapped_shared(folio))
> -				break;
> -			if (!folio_trylock(folio))
> -				break;
> -			folio_get(folio);
> -			arch_leave_lazy_mmu_mode();
> -			pte_unmap_unlock(start_pte, ptl);
> -			start_pte = NULL;
> -			err = split_folio(folio);
> -			folio_unlock(folio);
> -			folio_put(folio);
> -			if (err)
> -				break;
> -			start_pte = pte =
> -				pte_offset_map_lock(mm, pmd, addr, &ptl);
> -			if (!start_pte)
> -				break;
> -			arch_enter_lazy_mmu_mode();
> -			pte--;
> -			addr -= PAGE_SIZE;
> -			continue;
> +			nr = madvise_folio_pte_batch(addr, end, folio, pte,
> +						     ptent, &any_young, &any_dirty);
> +
> +			if (nr < folio_nr_pages(folio)) {
> +				if (folio_likely_mapped_shared(folio))
> +					continue;
> +
> +				arch_leave_lazy_mmu_mode();
> +				if (madvise_pte_split_folio(mm, pmd, addr,
> +							    folio, &start_pte, &ptl))
> +					nr = 0;
> +				if (!start_pte)
> +					break;
> +				pte = start_pte;
> +				arch_enter_lazy_mmu_mode();
> +				continue;
> +			}
> +
> +			if (any_young)
> +				ptent = pte_mkyoung(ptent);
> +			if (any_dirty)
> +				ptent = pte_mkdirty(ptent);
>  		}
>  
> +		if (folio_mapcount(folio) != folio_nr_pages(folio))
> +			continue;
> +
>  		if (folio_test_swapcache(folio) || folio_test_dirty(folio)) {
>  			if (!folio_trylock(folio))
>  				continue;
>  			/*
> -			 * If folio is shared with others, we mustn't clear
> -			 * the folio's dirty flag.
> +			 * If we have a large folio at this point, we know it is
> +			 * fully mapped so if its mapcount is the same as its
> +			 * number of pages, it must be exclusive.
>  			 */
> -			if (folio_mapcount(folio) != 1) {
> +			if (folio_mapcount(folio) != folio_nr_pages(folio)) {
>  				folio_unlock(folio);
>  				continue;
>  			}
> @@ -740,19 +770,10 @@ static int madvise_free_pte_range(pmd_t *pmd, unsigned long addr,
>  		}
>  
>  		if (pte_young(ptent) || pte_dirty(ptent)) {
> -			/*
> -			 * Some of architecture(ex, PPC) don't update TLB
> -			 * with set_pte_at and tlb_remove_tlb_entry so for
> -			 * the portability, remap the pte with old|clean
> -			 * after pte clearing.
> -			 */
> -			ptent = ptep_get_and_clear_full(mm, addr, pte,
> -							tlb->fullmm);
> -
> -			ptent = pte_mkold(ptent);
> -			ptent = pte_mkclean(ptent);
> -			set_pte_at(mm, addr, pte, ptent);
> -			tlb_remove_tlb_entry(tlb, pte, addr);
> +			clear_young_dirty_ptes(mm, addr, pte, nr,
> +					       CYDP_CLEAR_YOUNG |
> +						       CYDP_CLEAR_DIRTY);
> +			tlb_remove_tlb_entries(tlb, pte, nr, addr);
>  		}
>  		folio_mark_lazyfree(folio);
>  	}
> diff --git a/mm/memory.c b/mm/memory.c
> index 76157b32faa8..b6fa5146b260 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -989,7 +989,7 @@ copy_present_ptes(struct vm_area_struct *dst_vma, struct vm_area_struct *src_vma
>  			flags |= FPB_IGNORE_SOFT_DIRTY;
>  
>  		nr = folio_pte_batch(folio, addr, src_pte, pte, max_nr, flags,
> -				     &any_writable, NULL);
> +				     &any_writable, NULL, NULL);
>  		folio_ref_add(folio, nr);
>  		if (folio_test_anon(folio)) {
>  			if (unlikely(folio_try_dup_anon_rmap_ptes(folio, page,
> @@ -1558,7 +1558,7 @@ static inline int zap_present_ptes(struct mmu_gather *tlb,
>  	 */
>  	if (unlikely(folio_test_large(folio) && max_nr != 1)) {
>  		nr = folio_pte_batch(folio, addr, pte, ptent, max_nr, fpb_flags,
> -				     NULL, NULL);
> +				     NULL, NULL, NULL);
>  
>  		zap_present_folio_ptes(tlb, vma, folio, page, pte, ptent, nr,
>  				       addr, details, rss, force_flush,

Powered by blists - more mailing lists