Re: [PATCH 8/9] mm: Convert collapse_huge_page() to use a folio

[David Hildenbrand <david@redhat.com>]

On 11.12.23 17:22, Matthew Wilcox (Oracle) wrote:
> Replace three calls to compound_head() with one.
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> ---
>   mm/khugepaged.c | 15 ++++++++-------
>   1 file changed, 8 insertions(+), 7 deletions(-)
> 
> diff --git a/mm/khugepaged.c b/mm/khugepaged.c
> index e722f754797f..f7ec73976c38 100644
> --- a/mm/khugepaged.c
> +++ b/mm/khugepaged.c
> @@ -1087,6 +1087,7 @@ static int collapse_huge_page(struct mm_struct *mm, unsigned long address,
>   	pmd_t *pmd, _pmd;
>   	pte_t *pte;
>   	pgtable_t pgtable;
> +	struct folio *folio;
>   	struct page *hpage;
>   	spinlock_t *pmd_ptl, *pte_ptl;
>   	int result = SCAN_FAIL;
> @@ -1209,13 +1210,13 @@ static int collapse_huge_page(struct mm_struct *mm, unsigned long address,
>   	if (unlikely(result != SCAN_SUCCEED))
>   		goto out_up_write;
>   
> +	folio = page_folio(hpage);
>   	/*
> -	 * spin_lock() below is not the equivalent of smp_wmb(), but
> -	 * the smp_wmb() inside __SetPageUptodate() can be reused to
> -	 * avoid the copy_huge_page writes to become visible after
> -	 * the set_pmd_at() write.
> +	 * The smp_wmb() inside __folio_mark_uptodate() ensures the
> +	 * copy_huge_page writes become visible before the set_pmd_at()
> +	 * write.
>   	 */
> -	__SetPageUptodate(hpage);
> +	__folio_mark_uptodate(folio);
>   	pgtable = pmd_pgtable(_pmd);
>   
>   	_pmd = mk_huge_pmd(hpage, vma->vm_page_prot);
> @@ -1223,8 +1224,8 @@ static int collapse_huge_page(struct mm_struct *mm, unsigned long address,
>   
>   	spin_lock(pmd_ptl);
>   	BUG_ON(!pmd_none(*pmd));
> -	page_add_new_anon_rmap(hpage, vma, address);
> -	lru_cache_add_inactive_or_unevictable(hpage, vma);
> +	folio_add_new_anon_rmap(folio, vma, address);
> +	folio_add_lru_vma(folio, vma);
>   	pgtable_trans_huge_deposit(mm, pmd, pgtable);
>   	set_pmd_at(mm, address, pmd, _pmd);
>   	update_mmu_cache_pmd(vma, address, pmd);

Reviewed-by: David Hildenbrand <david@redhat.com>

-- 
Cheers,

David / dhildenb