diff mbox series

[1/1] mm/khugepaged: keep mm in mm_slot without MMF_DISABLE_THP check

Message ID 20240227035135.54593-1-ioworker0@gmail.com
State New
Headers
Series [1/1] mm/khugepaged: keep mm in mm_slot without MMF_DISABLE_THP check |

Commit Message

Lance Yang Feb. 27, 2024, 3:51 a.m. UTC 
  Previously, we removed the mm from mm_slot and dropped mm_count
if the MMF_THP_DISABLE flag was set. However, we didn't re-add
the mm back after clearing the MMF_THP_DISABLE flag. Additionally,
We add a check for the MMF_THP_DISABLE flag in hugepage_vma_revalidate().

Fixes: 879c6000e191 ("mm/khugepaged: bypassing unnecessary scans with MMF_DISABLE_THP check")

Signed-off-by: Lance Yang <ioworker0@gmail.com>
Suggested-by: Yang Shi <shy828301@gmail.com>
---
 mm/khugepaged.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

Comments

Yang Shi Feb. 27, 2024, 7:13 p.m. UTC | #1 
On Mon, Feb 26, 2024 at 7:51 PM Lance Yang <ioworker0@gmail.com> wrote:
>
> Previously, we removed the mm from mm_slot and dropped mm_count
> if the MMF_THP_DISABLE flag was set. However, we didn't re-add
> the mm back after clearing the MMF_THP_DISABLE flag. Additionally,
> We add a check for the MMF_THP_DISABLE flag in hugepage_vma_revalidate().
>
> Fixes: 879c6000e191 ("mm/khugepaged: bypassing unnecessary scans with MMF_DISABLE_THP check")

I think Andrew will fold this fix into your original patch.

>
> Signed-off-by: Lance Yang <ioworker0@gmail.com>
> Suggested-by: Yang Shi <shy828301@gmail.com>

Reviewed-by: Yang Shi <shy828301@gmail.com>

> ---
>  mm/khugepaged.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/mm/khugepaged.c b/mm/khugepaged.c
> index 2771fc043b3b..1c0073daad82 100644
> --- a/mm/khugepaged.c
> +++ b/mm/khugepaged.c
> @@ -920,7 +920,7 @@ static int hugepage_vma_revalidate(struct mm_struct *mm, unsigned long address,
>  {
>         struct vm_area_struct *vma;
>
> -       if (unlikely(hpage_collapse_test_exit(mm)))
> +       if (unlikely(hpage_collapse_test_exit_or_disable(mm)))
>                 return SCAN_ANY_PROCESS;
>
>         *vmap = vma = find_vma(mm, address);
> @@ -1428,7 +1428,7 @@ static void collect_mm_slot(struct khugepaged_mm_slot *mm_slot)
>
>         lockdep_assert_held(&khugepaged_mm_lock);
>
> -       if (hpage_collapse_test_exit_or_disable(mm)) {
> +       if (hpage_collapse_test_exit(mm)) {
>                 /* free mm_slot */
>                 hash_del(&slot->hash);
>                 list_del(&slot->mm_node);
> @@ -2456,7 +2456,7 @@ static unsigned int khugepaged_scan_mm_slot(unsigned int pages, int *result,
>          * Release the current mm_slot if this mm is about to die, or
>          * if we scanned all vmas of this mm.
>          */
> -       if (hpage_collapse_test_exit_or_disable(mm) || !vma) {
> +       if (hpage_collapse_test_exit(mm) || !vma) {
>                 /*
>                  * Make sure that if mm_users is reaching zero while
>                  * khugepaged runs here, khugepaged_exit will find
> --
> 2.33.1
>
David Hildenbrand Feb. 27, 2024, 7:18 p.m. UTC | #2 
On 27.02.24 20:13, Yang Shi wrote:
> On Mon, Feb 26, 2024 at 7:51 PM Lance Yang <ioworker0@gmail.com> wrote:
>>
>> Previously, we removed the mm from mm_slot and dropped mm_count
>> if the MMF_THP_DISABLE flag was set. However, we didn't re-add
>> the mm back after clearing the MMF_THP_DISABLE flag. Additionally,
>> We add a check for the MMF_THP_DISABLE flag in hugepage_vma_revalidate().
>>
>> Fixes: 879c6000e191 ("mm/khugepaged: bypassing unnecessary scans with MMF_DISABLE_THP check")
> 
> I think Andrew will fold this fix into your original patch.


Unlikely, it's already in mm-stable.

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

Patch

diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index 2771fc043b3b..1c0073daad82 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -920,7 +920,7 @@  static int hugepage_vma_revalidate(struct mm_struct *mm, unsigned long address,
 {
 	struct vm_area_struct *vma;
 
-	if (unlikely(hpage_collapse_test_exit(mm)))
+	if (unlikely(hpage_collapse_test_exit_or_disable(mm)))
 		return SCAN_ANY_PROCESS;
 
 	*vmap = vma = find_vma(mm, address);
@@ -1428,7 +1428,7 @@  static void collect_mm_slot(struct khugepaged_mm_slot *mm_slot)
 
 	lockdep_assert_held(&khugepaged_mm_lock);
 
-	if (hpage_collapse_test_exit_or_disable(mm)) {
+	if (hpage_collapse_test_exit(mm)) {
 		/* free mm_slot */
 		hash_del(&slot->hash);
 		list_del(&slot->mm_node);
@@ -2456,7 +2456,7 @@  static unsigned int khugepaged_scan_mm_slot(unsigned int pages, int *result,
 	 * Release the current mm_slot if this mm is about to die, or
 	 * if we scanned all vmas of this mm.
 	 */
-	if (hpage_collapse_test_exit_or_disable(mm) || !vma) {
+	if (hpage_collapse_test_exit(mm) || !vma) {
 		/*
 		 * Make sure that if mm_users is reaching zero while
 		 * khugepaged runs here, khugepaged_exit will find