Re: [patch] mm, thp: copying user pages must schedule on collapse

From: Vlastimil Babka
Date: Thu May 11 2017 - 05:51:44 EST


On 05/10/2017 11:27 PM, David Rientjes wrote:
> We have encountered need_resched warnings in __collapse_huge_page_copy()
> while doing {clear,copy}_user_highpage() over HPAGE_PMD_NR source pages.
>
> mm->mmap_sem is held for write, but the iteration is well bounded.
>
> Reschedule as needed.
>
> Signed-off-by: David Rientjes <rientjes@xxxxxxxxxx>

Acked-by: Vlastimil Babka <vbabka@xxxxxxx>

> ---
> mm/khugepaged.c | 7 +++----
> 1 file changed, 3 insertions(+), 4 deletions(-)
>
> diff --git a/mm/khugepaged.c b/mm/khugepaged.c
> --- a/mm/khugepaged.c
> +++ b/mm/khugepaged.c
> @@ -612,7 +612,8 @@ static void __collapse_huge_page_copy(pte_t *pte, struct page *page,
> spinlock_t *ptl)
> {
> pte_t *_pte;
> - for (_pte = pte; _pte < pte+HPAGE_PMD_NR; _pte++) {
> + for (_pte = pte; _pte < pte + HPAGE_PMD_NR;
> + _pte++, page++, address += PAGE_SIZE) {
> pte_t pteval = *_pte;
> struct page *src_page;
>
> @@ -651,9 +652,7 @@ static void __collapse_huge_page_copy(pte_t *pte, struct page *page,
> spin_unlock(ptl);
> free_page_and_swap_cache(src_page);
> }
> -
> - address += PAGE_SIZE;
> - page++;
> + cond_resched();
> }
> }
>
>