The following data is traced by bpftrace on a desktop system. After
the system has been left idle for 10 minutes upon booting, a lot of
SCAN_PMD_MAPPED or SCAN_NO_PTE_TABLE are observed during a full scan
by khugepaged.
@scan_pmd_status[1]: 1 ## SCAN_SUCCEED
@scan_pmd_status[6]: 2 ## SCAN_EXCEED_SHARED_PTE
@scan_pmd_status[3]: 142 ## SCAN_PMD_MAPPED
@scan_pmd_status[2]: 178 ## SCAN_NO_PTE_TABLE
total progress size: 674 MB
Total time : 419 seconds ## include khugepaged_scan_sleep_millisecs
The khugepaged_scan list save all task that support collapse into hugepage,
as long as the task is not destroyed, khugepaged will not remove it from
the khugepaged_scan list. This exist a phenomenon where task has already
collapsed all memory regions into hugepage, but khugepaged continues to
scan it, which wastes CPU time and invalid, and due to
khugepaged_scan_sleep_millisecs (default 10s) causes a long wait for
scanning a large number of invalid task, so scanning really valid task
is later.
After applying this patch, when the memory is either SCAN_PMD_MAPPED or
SCAN_NO_PTE_TABLE, just skip it, as follow:
@scan_pmd_status[6]: 2
@scan_pmd_status[3]: 147
@scan_pmd_status[2]: 173
total progress size: 45 MB
Total time : 20 seconds
Signed-off-by: Vernon Yang <yanglincheng@kylinos.cn>
---
mm/khugepaged.c | 17 +++++++++++------
1 file changed, 11 insertions(+), 6 deletions(-)
diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index 5c6015ac7b5e..6df2857d94c6 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -68,7 +68,10 @@ enum scan_result {
static struct task_struct *khugepaged_thread __read_mostly;
static DEFINE_MUTEX(khugepaged_mutex);
-/* default scan 8*HPAGE_PMD_NR ptes (or vmas) every 10 second */
+/*
+ * default scan 8*HPAGE_PMD_NR ptes, pmd_mapped, no_pte_table or vmas
+ * every 10 second.
+ */
static unsigned int khugepaged_pages_to_scan __read_mostly;
static unsigned int khugepaged_pages_collapsed;
static unsigned int khugepaged_full_scans;
@@ -1267,7 +1270,7 @@ static enum scan_result hpage_collapse_scan_pmd(struct mm_struct *mm,
result = find_pmd_or_thp_or_none(mm, start_addr, &pmd);
if (result != SCAN_SUCCEED) {
if (cur_progress)
- *cur_progress = HPAGE_PMD_NR;
+ *cur_progress = 1;
goto out;
}
@@ -1276,7 +1279,7 @@ static enum scan_result hpage_collapse_scan_pmd(struct mm_struct *mm,
pte = pte_offset_map_lock(mm, pmd, start_addr, &ptl);
if (!pte) {
if (cur_progress)
- *cur_progress = HPAGE_PMD_NR;
+ *cur_progress = 1;
result = SCAN_NO_PTE_TABLE;
goto out;
}
@@ -2347,9 +2350,6 @@ static enum scan_result hpage_collapse_scan_file(struct mm_struct *mm, unsigned
continue;
}
- if (cur_progress)
- *cur_progress += folio_nr_pages(folio);
-
if (folio_order(folio) == HPAGE_PMD_ORDER &&
folio->index == start) {
/* Maybe PMD-mapped */
@@ -2361,9 +2361,14 @@ static enum scan_result hpage_collapse_scan_file(struct mm_struct *mm, unsigned
* returning.
*/
folio_put(folio);
+ if (cur_progress)
+ *cur_progress += 1;
break;
}
+ if (cur_progress)
+ *cur_progress += folio_nr_pages(folio);
+
node = folio_nid(folio);
if (hpage_collapse_scan_abort(node, cc)) {
result = SCAN_SCAN_ABORT;
--
2.51.0On 1/11/26 13:19, Vernon Yang wrote:
> The following data is traced by bpftrace on a desktop system. After
> the system has been left idle for 10 minutes upon booting, a lot of
> SCAN_PMD_MAPPED or SCAN_NO_PTE_TABLE are observed during a full scan
> by khugepaged.
>
> @scan_pmd_status[1]: 1 ## SCAN_SUCCEED
> @scan_pmd_status[6]: 2 ## SCAN_EXCEED_SHARED_PTE
> @scan_pmd_status[3]: 142 ## SCAN_PMD_MAPPED
> @scan_pmd_status[2]: 178 ## SCAN_NO_PTE_TABLE
> total progress size: 674 MB
> Total time : 419 seconds ## include khugepaged_scan_sleep_millisecs
>
> The khugepaged_scan list save all task that support collapse into hugepage,
> as long as the task is not destroyed, khugepaged will not remove it from
> the khugepaged_scan list. This exist a phenomenon where task has already
> collapsed all memory regions into hugepage, but khugepaged continues to
> scan it, which wastes CPU time and invalid, and due to
> khugepaged_scan_sleep_millisecs (default 10s) causes a long wait for
> scanning a large number of invalid task, so scanning really valid task
> is later.
>
> After applying this patch, when the memory is either SCAN_PMD_MAPPED or
> SCAN_NO_PTE_TABLE, just skip it, as follow:
>
> @scan_pmd_status[6]: 2
> @scan_pmd_status[3]: 147
> @scan_pmd_status[2]: 173
> total progress size: 45 MB
> Total time : 20 seconds
>
> Signed-off-by: Vernon Yang <yanglincheng@kylinos.cn>
> ---
> mm/khugepaged.c | 17 +++++++++++------
> 1 file changed, 11 insertions(+), 6 deletions(-)
>
> diff --git a/mm/khugepaged.c b/mm/khugepaged.c
> index 5c6015ac7b5e..6df2857d94c6 100644
> --- a/mm/khugepaged.c
> +++ b/mm/khugepaged.c
> @@ -68,7 +68,10 @@ enum scan_result {
> static struct task_struct *khugepaged_thread __read_mostly;
> static DEFINE_MUTEX(khugepaged_mutex);
>
> -/* default scan 8*HPAGE_PMD_NR ptes (or vmas) every 10 second */
> +/*
> + * default scan 8*HPAGE_PMD_NR ptes, pmd_mapped, no_pte_table or vmas
> + * every 10 second.
> + */
> static unsigned int khugepaged_pages_to_scan __read_mostly;
> static unsigned int khugepaged_pages_collapsed;
> static unsigned int khugepaged_full_scans;
> @@ -1267,7 +1270,7 @@ static enum scan_result hpage_collapse_scan_pmd(struct mm_struct *mm,
> result = find_pmd_or_thp_or_none(mm, start_addr, &pmd);
> if (result != SCAN_SUCCEED) {
> if (cur_progress)
> - *cur_progress = HPAGE_PMD_NR;
> + *cur_progress = 1;
> goto out;
> }
>
> @@ -1276,7 +1279,7 @@ static enum scan_result hpage_collapse_scan_pmd(struct mm_struct *mm,
> pte = pte_offset_map_lock(mm, pmd, start_addr, &ptl);
> if (!pte) {
> if (cur_progress)
> - *cur_progress = HPAGE_PMD_NR;
> + *cur_progress = 1;
> result = SCAN_NO_PTE_TABLE;
> goto out;
> }
The above checks are clear.
> @@ -2347,9 +2350,6 @@ static enum scan_result hpage_collapse_scan_file(struct mm_struct *mm, unsigned
> continue;
> }
>
> - if (cur_progress)
> - *cur_progress += folio_nr_pages(folio);
> -
This is a all a bit hairy.
Assume we found a single 4k folio in the xarray, but then collapse a 2M THP.
Is the progress really "1" ?
What about shmem swap entries (xa_is_value)?
So I think the whole file path needs more thought
--
Cheers
David
© 2016 - 2026 Red Hat, Inc.