Re: [PATCH v2 1/2] mm/huge_memory: Adjust try_to_migrate_one() and split_huge_pmd_locked()

[Zi Yan <ziy@nvidia.com>]

On 25 Apr 2025, at 6:38, Gavin Guo wrote:

> The split_huge_pmd_locked function currently performs redundant checks
> for migration entries and folio validation that are already handled by
> the page_vma_mapped_walk mechanism in try_to_migrate_one.
>
> Specifically, page_vma_mapped_walk already ensures that:
> - The folio is properly mapped in the given VMA area
> - pmd_trans_huge, pmd_devmap, and migration entry validation are
> performed
>
> To leverage page_vma_mapped_walk's work, moving TTU_SPLIT_HUGE_PMD
> handling to the while loop checking and removing these duplicate checks
> from split_huge_pmd_locked.
>
> Suggested-by: David Hildenbrand <david@redhat.com>
> Link: https://lore.kernel.org/all/98d1d195-7821-4627-b518-83103ade56c0@redhat.com/
> Link: https://lore.kernel.org/all/91599a3c-e69e-4d79-bac5-5013c96203d7@redhat.com/
> Signed-off-by: Gavin Guo <gavinguo@igalia.com>
> Acked-by: David Hildenbrand <david@redhat.com>
> ---
>  mm/huge_memory.c | 21 ++-------------------
>  mm/rmap.c        | 18 +++++++++---------
>  2 files changed, 11 insertions(+), 28 deletions(-)
>
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> index 47d76d03ce30..485a0ba011af 100644
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -3075,27 +3075,10 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
>  void split_huge_pmd_locked(struct vm_area_struct *vma, unsigned long address,
>  			   pmd_t *pmd, bool freeze, struct folio *folio)
>  {
> -	bool pmd_migration = is_pmd_migration_entry(*pmd);
> -
> -	VM_WARN_ON_ONCE(folio && !folio_test_pmd_mappable(folio));
>  	VM_WARN_ON_ONCE(!IS_ALIGNED(address, HPAGE_PMD_SIZE));
> -	VM_WARN_ON_ONCE(folio && !folio_test_locked(folio));
> -	VM_BUG_ON(freeze && !folio);
> -
> -	/*
> -	 * When the caller requests to set up a migration entry, we
> -	 * require a folio to check the PMD against. Otherwise, there
> -	 * is a risk of replacing the wrong folio.
> -	 */
> -	if (pmd_trans_huge(*pmd) || pmd_devmap(*pmd) || pmd_migration) {
> -		/*
> -		 * Do not apply pmd_folio() to a migration entry; and folio lock
> -		 * guarantees that it must be of the wrong folio anyway.
> -		 */
> -		if (folio && (pmd_migration || folio != pmd_folio(*pmd)))
> -			return;
> +	if (pmd_trans_huge(*pmd) || pmd_devmap(*pmd) ||
> +	    is_pmd_migration_entry(*pmd))
>  		__split_huge_pmd_locked(vma, pmd, address, freeze);
> -	}
>  }
>
>  void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
> diff --git a/mm/rmap.c b/mm/rmap.c
> index 67bb273dfb80..b53a4dcaeaae 100644
> --- a/mm/rmap.c
> +++ b/mm/rmap.c
> @@ -2291,13 +2291,6 @@ static bool try_to_migrate_one(struct folio *folio, struct vm_area_struct *vma,
>  	if (flags & TTU_SYNC)
>  		pvmw.flags = PVMW_SYNC;
>
> -	/*
> -	 * unmap_page() in mm/huge_memory.c is the only user of migration with
> -	 * TTU_SPLIT_HUGE_PMD and it wants to freeze.
> -	 */
> -	if (flags & TTU_SPLIT_HUGE_PMD)
> -		split_huge_pmd_address(vma, address, true, folio);
> -
>  	/*
>  	 * For THP, we have to assume the worse case ie pmd for invalidation.
>  	 * For hugetlb, it could be much worse if we need to do pud
> @@ -2323,9 +2316,16 @@ static bool try_to_migrate_one(struct folio *folio, struct vm_area_struct *vma,
>  	mmu_notifier_invalidate_range_start(&range);
>
>  	while (page_vma_mapped_walk(&pvmw)) {
> -#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
>  		/* PMD-mapped THP migration entry */

This comment should be moved along with #ifdef to avoid confusion.

>  		if (!pvmw.pte) {
> +			if (flags & TTU_SPLIT_HUGE_PMD) {
> +				split_huge_pmd_locked(vma, pvmw.address,
> +						      pvmw.pmd, true, NULL);
> +				ret = false;
> +				page_vma_mapped_walk_done(&pvmw);
> +				break;
> +			}
> +#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
>  			subpage = folio_page(folio,
>  				pmd_pfn(*pvmw.pmd) - folio_pfn(folio));
>  			VM_BUG_ON_FOLIO(folio_test_hugetlb(folio) ||
> @@ -2337,8 +2337,8 @@ static bool try_to_migrate_one(struct folio *folio, struct vm_area_struct *vma,
>  				break;
>  			}
>  			continue;
> -		}
>  #endif

I wonder if we need a WARN here to make sure when THP migration support is not
present all PMDs are split in try_to_migrate_one().

> +		}
>
>  		/* Unexpected PMD-mapped THP? */
>  		VM_BUG_ON_FOLIO(!pvmw.pte, folio);
> -- 
> 2.43.0

Otherwise, looks good to me. Thanks. Reviewed-by: Zi Yan <ziy@nvidia.com>

--
Best Regards,
Yan, Zi