Re: [PATCH v6 2/4] mm: add per-order mTHP anon_swpout and anon_swpout_fallback counters

From: David Hildenbrand
Date: Tue Apr 16 2024 - 04:17:50 EST


On 16.04.24 10:14, David Hildenbrand wrote:
On 12.04.24 13:48, Barry Song wrote:
From: Barry Song <v-songbaohua@xxxxxxxx>

This helps to display the fragmentation situation of the swapfile, knowing
the proportion of how much we haven't split large folios. So far, we only
support non-split swapout for anon memory, with the possibility of
expanding to shmem in the future. So, we add the "anon" prefix to the
counter names.

Signed-off-by: Barry Song <v-songbaohua@xxxxxxxx>
Reviewed-by: Ryan Roberts <ryan.roberts@xxxxxxx>
Cc: Chris Li <chrisl@xxxxxxxxxx>
Cc: David Hildenbrand <david@xxxxxxxxxx>
Cc: Domenico Cerasuolo <cerasuolodomenico@xxxxxxxxx>
Cc: Kairui Song <kasong@xxxxxxxxxxx>
Cc: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx>
Cc: Peter Xu <peterx@xxxxxxxxxx>
Cc: Ryan Roberts <ryan.roberts@xxxxxxx>
Cc: Suren Baghdasaryan <surenb@xxxxxxxxxx>
Cc: Yosry Ahmed <yosryahmed@xxxxxxxxxx>
Cc: Yu Zhao <yuzhao@xxxxxxxxxx>
---
include/linux/huge_mm.h | 2 ++
mm/huge_memory.c | 4 ++++
mm/page_io.c | 1 +
mm/vmscan.c | 3 +++
4 files changed, 10 insertions(+)

diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
index d4fdb2641070..7cd07b83a3d0 100644
--- a/include/linux/huge_mm.h
+++ b/include/linux/huge_mm.h
@@ -268,6 +268,8 @@ enum mthp_stat_item {
MTHP_STAT_ANON_FAULT_ALLOC,
MTHP_STAT_ANON_FAULT_FALLBACK,
MTHP_STAT_ANON_FAULT_FALLBACK_CHARGE,
+ MTHP_STAT_ANON_SWPOUT,
+ MTHP_STAT_ANON_SWPOUT_FALLBACK,
__MTHP_STAT_COUNT
};
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index dfc38cc83a04..58f2c4745d80 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -555,11 +555,15 @@ static struct kobj_attribute _name##_attr = __ATTR_RO(_name)
DEFINE_MTHP_STAT_ATTR(anon_fault_alloc, MTHP_STAT_ANON_FAULT_ALLOC);
DEFINE_MTHP_STAT_ATTR(anon_fault_fallback, MTHP_STAT_ANON_FAULT_FALLBACK);
DEFINE_MTHP_STAT_ATTR(anon_fault_fallback_charge, MTHP_STAT_ANON_FAULT_FALLBACK_CHARGE);
+DEFINE_MTHP_STAT_ATTR(anon_swpout, MTHP_STAT_ANON_SWPOUT);
+DEFINE_MTHP_STAT_ATTR(anon_swpout_fallback, MTHP_STAT_ANON_SWPOUT_FALLBACK);
static struct attribute *stats_attrs[] = {
&anon_fault_alloc_attr.attr,
&anon_fault_fallback_attr.attr,
&anon_fault_fallback_charge_attr.attr,
+ &anon_swpout_attr.attr,
+ &anon_swpout_fallback_attr.attr,
NULL,
};
diff --git a/mm/page_io.c b/mm/page_io.c
index a9a7c236aecc..46c603dddf04 100644
--- a/mm/page_io.c
+++ b/mm/page_io.c
@@ -217,6 +217,7 @@ static inline void count_swpout_vm_event(struct folio *folio)
count_memcg_folio_events(folio, THP_SWPOUT, 1);
count_vm_event(THP_SWPOUT);
}
+ count_mthp_stat(folio_order(folio), MTHP_STAT_ANON_SWPOUT);
#endif
count_vm_events(PSWPOUT, folio_nr_pages(folio));
}
diff --git a/mm/vmscan.c b/mm/vmscan.c
index bca2d9981c95..49bd94423961 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1231,6 +1231,8 @@ static unsigned int shrink_folio_list(struct list_head *folio_list,
goto activate_locked;
}
if (!add_to_swap(folio)) {
+ int __maybe_unused order = folio_order(folio);
+
if (!folio_test_large(folio))
goto activate_locked_split;
/* Fallback to swap normal pages */
@@ -1242,6 +1244,7 @@ static unsigned int shrink_folio_list(struct list_head *folio_list,
THP_SWPOUT_FALLBACK, 1);
count_vm_event(THP_SWPOUT_FALLBACK);
}
+ count_mthp_stat(order, MTHP_STAT_ANON_SWPOUT_FALLBACK);

Why the temporary variable for order?

count_mthp_stat(folio_order(order),
MTHP_STAT_ANON_SWPOUT_FALLBACK);

... but now I do wonder if we want to pass the folio to count_mthp_stat() ?

.. and now realizing, that that doesn't make sense if we fail to allocate the folio in the first place. So all good.

--
Cheers,

David / dhildenb