From: Yosry Ahmed <yosryahmed@google.com>
To: Andrew Morton <akpm@linux-foundation.org>,
Shakeel Butt <shakeelb@google.com>
Cc: "Johannes Weiner" <hannes@cmpxchg.org>,
"Michal Hocko" <mhocko@kernel.org>,
"Roman Gushchin" <roman.gushchin@linux.dev>,
"Muchun Song" <muchun.song@linux.dev>,
"Michal Koutný" <mkoutny@suse.com>,
linux-mm@kvack.org, cgroups@vger.kernel.org,
linux-kernel@vger.kernel.org,
"Yosry Ahmed" <yosryahmed@google.com>
Subject: [PATCH v2 2/2] mm: memcg: normalize the value passed into memcg_rstat_updated()
Date: Fri, 22 Sep 2023 17:57:40 +0000 [thread overview]
Message-ID: <20230922175741.635002-3-yosryahmed@google.com> (raw)
In-Reply-To: <20230922175741.635002-1-yosryahmed@google.com>
memcg_rstat_updated() uses the value of the state update to keep track
of the magnitude of pending updates, so that we only do a stats flush
when it's worth the work. Most values passed into memcg_rstat_updated()
are in pages, however, a few of them are actually in bytes or KBs.
To put this into perspective, a 512 byte slab allocation today would
look the same as allocating 512 pages. This may result in premature
flushes, which means unnecessary work and latency.
Normalize all the state values passed into memcg_rstat_updated() to
pages. Round up non-zero sub-page to 1 page, because
memcg_rstat_updated() ignores 0 page updates.
Fixes: 5b3be698a872 ("memcg: better bounds on the memcg stats updates")
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
---
mm/memcontrol.c | 20 ++++++++++++++++++--
1 file changed, 18 insertions(+), 2 deletions(-)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 308cc7353ef0..d1a322a75172 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -763,6 +763,22 @@ unsigned long memcg_page_state(struct mem_cgroup *memcg, int idx)
return x;
}
+static int memcg_page_state_unit(int item);
+
+/*
+ * Normalize the value passed into memcg_rstat_updated() to be in pages. Round
+ * up non-zero sub-page updates to 1 page as zero page updates are ignored.
+ */
+static int memcg_state_val_in_pages(int idx, int val)
+{
+ int unit = memcg_page_state_unit(idx);
+
+ if (!val || unit == PAGE_SIZE)
+ return val;
+ else
+ return max(val * unit / PAGE_SIZE, 1UL);
+}
+
/**
* __mod_memcg_state - update cgroup memory statistics
* @memcg: the memory cgroup
@@ -775,7 +791,7 @@ void __mod_memcg_state(struct mem_cgroup *memcg, int idx, int val)
return;
__this_cpu_add(memcg->vmstats_percpu->state[idx], val);
- memcg_rstat_updated(memcg, val);
+ memcg_rstat_updated(memcg, memcg_state_val_in_pages(idx, val));
}
/* idx can be of type enum memcg_stat_item or node_stat_item. */
@@ -826,7 +842,7 @@ void __mod_memcg_lruvec_state(struct lruvec *lruvec, enum node_stat_item idx,
/* Update lruvec */
__this_cpu_add(pn->lruvec_stats_percpu->state[idx], val);
- memcg_rstat_updated(memcg, val);
+ memcg_rstat_updated(memcg, memcg_state_val_in_pages(idx, val));
memcg_stats_unlock();
}
--
2.42.0.515.g380fc7ccd1-goog
next prev parent reply other threads:[~2023-09-22 17:57 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-22 17:57 [PATCH v2 0/2] mm: memcg: fix tracking of pending stats updates values Yosry Ahmed
2023-09-22 17:57 ` [PATCH v2 1/2] mm: memcg: refactor page state unit helpers Yosry Ahmed
2023-10-03 13:03 ` Johannes Weiner
2023-10-03 18:11 ` Michal Koutný
2023-10-03 19:47 ` Yosry Ahmed
2023-10-04 9:02 ` Michal Koutný
2023-10-04 16:58 ` Yosry Ahmed
2023-10-04 18:36 ` Johannes Weiner
2023-10-05 9:06 ` Michal Koutný
2023-10-05 9:31 ` Yosry Ahmed
2023-10-05 16:30 ` Michal Koutný
2023-10-05 17:30 ` Yosry Ahmed
2023-10-18 19:27 ` Andrew Morton
2023-09-22 17:57 ` Yosry Ahmed [this message]
2023-10-03 13:13 ` [PATCH v2 2/2] mm: memcg: normalize the value passed into memcg_rstat_updated() Johannes Weiner
2023-10-03 15:53 ` Yosry Ahmed
2023-10-03 18:22 ` Michal Koutný
2023-10-03 19:51 ` Yosry Ahmed
2023-09-25 13:50 ` [PATCH v2 0/2] mm: memcg: fix tracking of pending stats updates values Michal Hocko
2023-09-25 17:11 ` Yosry Ahmed
2023-10-03 7:57 ` Michal Hocko
2023-10-03 8:03 ` Yosry Ahmed
2023-10-03 8:09 ` Michal Hocko
2023-10-03 8:49 ` Yosry Ahmed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230922175741.635002-3-yosryahmed@google.com \
--to=yosryahmed@google.com \
--cc=akpm@linux-foundation.org \
--cc=cgroups@vger.kernel.org \
--cc=hannes@cmpxchg.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=mkoutny@suse.com \
--cc=muchun.song@linux.dev \
--cc=roman.gushchin@linux.dev \
--cc=shakeelb@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox