[PATCH v2 2/7] mm/khugepaged: stop swapping in page when VM_FAULT_RETRY occurs

From: Miaohe Lin
Date: Sat Jun 25 2022 - 05:28:29 EST


When do_swap_page returns VM_FAULT_RETRY, we do not retry here and thus
swap entry will remain in pagetable. This will result in later failure.
So stop swapping in pages in this case to save cpu cycles. As A further
optimization, mmap_lock is released when __collapse_huge_page_swapin()
fails to avoid relocking mmap_lock. And "swapped_in++" is moved after
error handling to make it more accurate.

Signed-off-by: Miaohe Lin <linmiaohe@xxxxxxxxxx>
---
mm/khugepaged.c | 32 ++++++++++++++------------------
1 file changed, 14 insertions(+), 18 deletions(-)

diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index 8a103e0f8d2b..c6fc4eb8d77b 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -940,8 +940,8 @@ static int hugepage_vma_revalidate(struct mm_struct *mm, unsigned long address,
* Bring missing pages in from swap, to complete THP collapse.
* Only done if khugepaged_scan_pmd believes it is worthwhile.
*
- * Called and returns without pte mapped or spinlocks held,
- * but with mmap_lock held to protect against vma changes.
+ * Called and returns without pte mapped or spinlocks held.
+ * Note that if false is returned, mmap_lock will be released.
*/

static bool __collapse_huge_page_swapin(struct mm_struct *mm,
@@ -968,27 +968,24 @@ static bool __collapse_huge_page_swapin(struct mm_struct *mm,
pte_unmap(vmf.pte);
continue;
}
- swapped_in++;
ret = do_swap_page(&vmf);

- /* do_swap_page returns VM_FAULT_RETRY with released mmap_lock */
+ /*
+ * do_swap_page returns VM_FAULT_RETRY with released mmap_lock.
+ * Note we treat VM_FAULT_RETRY as VM_FAULT_ERROR here because
+ * we do not retry here and swap entry will remain in pagetable
+ * resulting in later failure.
+ */
if (ret & VM_FAULT_RETRY) {
- mmap_read_lock(mm);
- if (hugepage_vma_revalidate(mm, haddr, &vma)) {
- /* vma is no longer available, don't continue to swapin */
- trace_mm_collapse_huge_page_swapin(mm, swapped_in, referenced, 0);
- return false;
- }
- /* check if the pmd is still valid */
- if (mm_find_pmd(mm, haddr) != pmd) {
- trace_mm_collapse_huge_page_swapin(mm, swapped_in, referenced, 0);
- return false;
- }
+ trace_mm_collapse_huge_page_swapin(mm, swapped_in, referenced, 0);
+ return false;
}
if (ret & VM_FAULT_ERROR) {
+ mmap_read_unlock(mm);
trace_mm_collapse_huge_page_swapin(mm, swapped_in, referenced, 0);
return false;
}
+ swapped_in++;
}

/* Drain LRU add pagevec to remove extra pin on the swapped in pages */
@@ -1054,13 +1051,12 @@ static void collapse_huge_page(struct mm_struct *mm,
}

/*
- * __collapse_huge_page_swapin always returns with mmap_lock locked.
- * If it fails, we release mmap_lock and jump out_nolock.
+ * __collapse_huge_page_swapin will return with mmap_lock released
+ * when it fails. So we jump out_nolock directly in that case.
* Continuing to collapse causes inconsistency.
*/
if (unmapped && !__collapse_huge_page_swapin(mm, vma, address,
pmd, referenced)) {
- mmap_read_unlock(mm);
goto out_nolock;
}

--
2.23.0