diff mbox series

[02/11] mm: page_vma_mapped_walk(): settle PageHuge on entry

Message ID e31a483c-6d73-a6bb-26c5-43c3b880a2@google.com (mailing list archive)
State New, archived
Headers show
Series mm: page_vma_mapped_walk() cleanup and THP fixes | expand

Commit Message

Hugh Dickins June 10, 2021, 6:36 a.m. UTC
page_vma_mapped_walk() cleanup: get the hugetlbfs PageHuge case
out of the way at the start, so no need to worry about it later.

Signed-off-by: Hugh Dickins <hughd@google.com>
Cc: <stable@vger.kernel.org>
---
 mm/page_vma_mapped.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

Comments

Kirill A. Shutemov June 10, 2021, 8:57 a.m. UTC | #1
On Wed, Jun 09, 2021 at 11:36:36PM -0700, Hugh Dickins wrote:
> page_vma_mapped_walk() cleanup: get the hugetlbfs PageHuge case
> out of the way at the start, so no need to worry about it later.
> 
> Signed-off-by: Hugh Dickins <hughd@google.com>
> Cc: <stable@vger.kernel.org>

Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Peter Xu June 10, 2021, 2:17 p.m. UTC | #2
On Wed, Jun 09, 2021 at 11:36:36PM -0700, Hugh Dickins wrote:
> page_vma_mapped_walk() cleanup: get the hugetlbfs PageHuge case
> out of the way at the start, so no need to worry about it later.
> 
> Signed-off-by: Hugh Dickins <hughd@google.com>
> Cc: <stable@vger.kernel.org>
> ---
>  mm/page_vma_mapped.c | 12 ++++++++----
>  1 file changed, 8 insertions(+), 4 deletions(-)
> 
> diff --git a/mm/page_vma_mapped.c b/mm/page_vma_mapped.c
> index a6dbf714ca15..7c0504641fb8 100644
> --- a/mm/page_vma_mapped.c
> +++ b/mm/page_vma_mapped.c
> @@ -153,10 +153,11 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
>  	if (pvmw->pmd && !pvmw->pte)
>  		return not_found(pvmw);
>  
> -	if (pvmw->pte)
> -		goto next_pte;
> -
>  	if (unlikely(PageHuge(page))) {
> +		/* The only possible mapping was handled on last iteration */
> +		if (pvmw->pte)
> +			return not_found(pvmw);
> +
>  		/* when pud is not present, pte will be NULL */
>  		pvmw->pte = huge_pte_offset(mm, pvmw->address, page_size(page));
>  		if (!pvmw->pte)

Would it be even nicer to move the initial check to be after PageHuge() too?

	if (pvmw->pmd && !pvmw->pte)
		return not_found(pvmw);

It looks already better, so no strong opinion.

Reviewed-by: Peter Xu <peterx@redhat.com>

Thanks,
diff mbox series

Patch

diff --git a/mm/page_vma_mapped.c b/mm/page_vma_mapped.c
index a6dbf714ca15..7c0504641fb8 100644
--- a/mm/page_vma_mapped.c
+++ b/mm/page_vma_mapped.c
@@ -153,10 +153,11 @@  bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 	if (pvmw->pmd && !pvmw->pte)
 		return not_found(pvmw);
 
-	if (pvmw->pte)
-		goto next_pte;
-
 	if (unlikely(PageHuge(page))) {
+		/* The only possible mapping was handled on last iteration */
+		if (pvmw->pte)
+			return not_found(pvmw);
+
 		/* when pud is not present, pte will be NULL */
 		pvmw->pte = huge_pte_offset(mm, pvmw->address, page_size(page));
 		if (!pvmw->pte)
@@ -168,6 +169,9 @@  bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 			return not_found(pvmw);
 		return true;
 	}
+
+	if (pvmw->pte)
+		goto next_pte;
 restart:
 	pgd = pgd_offset(mm, pvmw->address);
 	if (!pgd_present(*pgd))
@@ -233,7 +237,7 @@  bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 			return true;
 next_pte:
 		/* Seek to next pte only makes sense for THP */
-		if (!PageTransHuge(page) || PageHuge(page))
+		if (!PageTransHuge(page))
 			return not_found(pvmw);
 		end = vma_address_end(page, pvmw->vma);
 		do {