mm/vmstat: add events for THP max_ptes_* exceeds

There are interfaces to adjust max_ptes_none, max_ptes_swap,
max_ptes_shared values, see
  /sys/kernel/mm/transparent_hugepage/khugepaged/.

But system administrator may not know which value is the best.  So Add
those events to support adjusting max_ptes_* to suitable values.

For example, if default max_ptes_swap value causes too much failures,
and system uses zram whose IO is fast, administrator could increase
max_ptes_swap until THP_SCAN_EXCEED_SWAP_PTE not increase anymore.

Link: https://lkml.kernel.org/r/20211225094036.574157-1-yang.yang29@zte.com.cn
Signed-off-by: Yang Yang <yang.yang29@zte.com.cn>
Cc: "Huang, Ying" <ying.huang@intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Saravanan D <saravanand@fb.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

diff --git a/include/linux/vm_event_item.h b/include/linux/vm_event_item.h
index a185cc7..7b23633 100644 (file)
--- a/include/linux/vm_event_item.h
+++ b/include/linux/vm_event_item.h
@@ -98,6 +98,9 @@ enum vm_event_item { PGPGIN, PGPGOUT, PSWPIN, PSWPOUT,
                THP_SPLIT_PAGE_FAILED,
                THP_DEFERRED_SPLIT_PAGE,
                THP_SPLIT_PMD,
+               THP_SCAN_EXCEED_NONE_PTE,
+               THP_SCAN_EXCEED_SWAP_PTE,
+               THP_SCAN_EXCEED_SHARED_PTE,
 #ifdef CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD
                THP_SPLIT_PUD,
 #endif

diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index 698ea19..02071f2 100644 (file)
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -618,6 +618,7 @@ static int __collapse_huge_page_isolate(struct vm_area_struct *vma,
                                continue;
                        } else {
                                result = SCAN_EXCEED_NONE_PTE;
+                               count_vm_event(THP_SCAN_EXCEED_NONE_PTE);
                                goto out;
                        }
                }
@@ -636,6 +637,7 @@ static int __collapse_huge_page_isolate(struct vm_area_struct *vma,
                if (page_mapcount(page) > 1 &&
                                ++shared > khugepaged_max_ptes_shared) {
                        result = SCAN_EXCEED_SHARED_PTE;
+                       count_vm_event(THP_SCAN_EXCEED_SHARED_PTE);
                        goto out;
                }
 
@@ -1253,6 +1255,7 @@ static int khugepaged_scan_pmd(struct mm_struct *mm,
                                continue;
                        } else {
                                result = SCAN_EXCEED_SWAP_PTE;
+                               count_vm_event(THP_SCAN_EXCEED_SWAP_PTE);
                                goto out_unmap;
                        }
                }
@@ -1262,6 +1265,7 @@ static int khugepaged_scan_pmd(struct mm_struct *mm,
                                continue;
                        } else {
                                result = SCAN_EXCEED_NONE_PTE;
+                               count_vm_event(THP_SCAN_EXCEED_NONE_PTE);
                                goto out_unmap;
                        }
                }
@@ -1290,6 +1294,7 @@ static int khugepaged_scan_pmd(struct mm_struct *mm,
                if (page_mapcount(page) > 1 &&
                                ++shared > khugepaged_max_ptes_shared) {
                        result = SCAN_EXCEED_SHARED_PTE;
+                       count_vm_event(THP_SCAN_EXCEED_SHARED_PTE);
                        goto out_unmap;
                }
 
@@ -2000,6 +2005,7 @@ static void khugepaged_scan_file(struct mm_struct *mm,
                if (xa_is_value(page)) {
                        if (++swap > khugepaged_max_ptes_swap) {
                                result = SCAN_EXCEED_SWAP_PTE;
+                               count_vm_event(THP_SCAN_EXCEED_SWAP_PTE);
                                break;
                        }
                        continue;
@@ -2046,6 +2052,7 @@ static void khugepaged_scan_file(struct mm_struct *mm,
        if (result == SCAN_SUCCEED) {
                if (present < HPAGE_PMD_NR - khugepaged_max_ptes_none) {
                        result = SCAN_EXCEED_NONE_PTE;
+                       count_vm_event(THP_SCAN_EXCEED_NONE_PTE);
                } else {
                        node = khugepaged_find_target_node();
                        collapse_file(mm, file, start, hpage, node);

diff --git a/mm/vmstat.c b/mm/vmstat.c
index d701c33..4057372 100644 (file)
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -1353,6 +1353,9 @@ const char * const vmstat_text[] = {
        "thp_split_page_failed",
        "thp_deferred_split_page",
        "thp_split_pmd",
+       "thp_scan_exceed_none_pte",
+       "thp_scan_exceed_swap_pte",
+       "thp_scan_exceed_share_pte",
 #ifdef CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD
        "thp_split_pud",
 #endif