linux-kernel - [PATCH v2] mm: memcg: add THP swap out info for anonymous reclaim

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <20230912021727.61601-1-vernhao@tencent.com>
Date:   Tue, 12 Sep 2023 10:17:25 +0800
From:   Vern Hao <haoxing990@...il.com>
To:     hannes@...xchg.org
Cc:     mhocko@...nel.org, roman.gushchin@...ux.dev, shakeelb@...gle.com,
        akpm@...ux-foundation.org, cgroups@...r.kernel.org,
        linux-mm@...ck.org, linux-kernel@...r.kernel.org,
        haoxing990@...il.com, Xin Hao <vernhao@...cent.com>
Subject: [PATCH v2] mm: memcg: add THP swap out info for anonymous reclaim

From: Xin Hao <vernhao@...cent.com>

At present, we support per-memcg reclaim strategy, however we do not
know the number of transparent huge pages being reclaimed, as we know
the transparent huge pages need to be splited before reclaim them, and
they will bring some performance bottleneck effect. for example, when
two memcg (A & B) are doing reclaim for anonymous pages at same time,
and 'A' memcg is reclaiming a large number of transparent huge pages, we
can better analyze that the performance bottleneck will be caused by 'A'
memcg.  therefore, in order to better analyze such problems, there add
THP swap out info for per-memcg.

Signed-off-by: Xin Hao <vernhao@...cent.com>
---
v1 -> v2
	- Do some fix as Johannes Weiner suggestion.
v1: 
https://lore.kernel.org/linux-mm/20230911160824.GB103342@cmpxchg.org/T/

 mm/memcontrol.c | 2 ++
 mm/page_io.c    | 4 +++-
 mm/vmscan.c     | 1 +
 3 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index ecc07b47e813..32d50db9ea0d 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -752,6 +752,8 @@ static const unsigned int memcg_vm_event_stat[] = {
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
 	THP_FAULT_ALLOC,
 	THP_COLLAPSE_ALLOC,
+	THP_SWPOUT,
+	THP_SWPOUT_FALLBACK,
 #endif
 };
 
diff --git a/mm/page_io.c b/mm/page_io.c
index fe4c21af23f2..73b2f2846bec 100644
--- a/mm/page_io.c
+++ b/mm/page_io.c
@@ -208,8 +208,10 @@ int swap_writepage(struct page *page, struct writeback_control *wbc)
 static inline void count_swpout_vm_event(struct folio *folio)
 {
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
-	if (unlikely(folio_test_pmd_mappable(folio)))
+	if (unlikely(folio_test_pmd_mappable(folio))) {
+		count_memcg_folio_events(folio, THP_SWPOUT, 1);
 		count_vm_event(THP_SWPOUT);
+	}
 #endif
 	count_vm_events(PSWPOUT, folio_nr_pages(folio));
 }
diff --git a/mm/vmscan.c b/mm/vmscan.c
index ea57a43ebd6b..39beb0d30156 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1928,6 +1928,7 @@ static unsigned int shrink_folio_list(struct list_head *folio_list,
 								folio_list))
 						goto activate_locked;
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
+					count_memcg_folio_events(folio, THP_SWPOUT_FALLBACK, 1);
 					count_vm_event(THP_SWPOUT_FALLBACK);
 #endif
 					if (!add_to_swap(folio))
-- 
2.42.0