linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Waiman Long <longman@redhat.com>
To: Johannes Weiner <hannes@cmpxchg.org>,
	Michal Hocko <mhocko@kernel.org>,
	Vladimir Davydov <vdavydov.dev@gmail.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Petr Mladek <pmladek@suse.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Sergey Senozhatsky <senozhatsky@chromium.org>,
	Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
	Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc: linux-kernel@vger.kernel.org, cgroups@vger.kernel.org,
	linux-mm@kvack.org, Ira Weiny <ira.weiny@intel.com>,
	Mike Rapoport <rppt@kernel.org>,
	David Rientjes <rientjes@google.com>,
	Roman Gushchin <guro@fb.com>, Rafael Aquini <aquini@redhat.com>,
	Waiman Long <longman@redhat.com>
Subject: [PATCH v3 3/4] mm/page_owner: Print memcg information
Date: Mon, 31 Jan 2022 14:23:07 -0500	[thread overview]
Message-ID: <20220131192308.608837-4-longman@redhat.com> (raw)
In-Reply-To: <20220131192308.608837-1-longman@redhat.com>

It was found that a number of offlined memcgs were not freed because
they were pinned by some charged pages that were present. Even "echo
1 > /proc/sys/vm/drop_caches" wasn't able to free those pages. These
offlined but not freed memcgs tend to increase in number over time with
the side effect that percpu memory consumption as shown in /proc/meminfo
also increases over time.

In order to find out more information about those pages that pin
offlined memcgs, the page_owner feature is extended to print memory
cgroup information especially whether the cgroup is offlined or not.

Signed-off-by: Waiman Long <longman@redhat.com>
Acked-by: David Rientjes <rientjes@google.com>
---
 mm/page_owner.c | 39 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 39 insertions(+)

diff --git a/mm/page_owner.c b/mm/page_owner.c
index 28dac73e0542..a471c74c7fe0 100644
--- a/mm/page_owner.c
+++ b/mm/page_owner.c
@@ -10,6 +10,7 @@
 #include <linux/migrate.h>
 #include <linux/stackdepot.h>
 #include <linux/seq_file.h>
+#include <linux/memcontrol.h>
 #include <linux/sched/clock.h>
 
 #include "internal.h"
@@ -325,6 +326,42 @@ void pagetypeinfo_showmixedcount_print(struct seq_file *m,
 	seq_putc(m, '\n');
 }
 
+#ifdef CONFIG_MEMCG
+/*
+ * Looking for memcg information and print it out
+ */
+static inline void print_page_owner_memcg(char *kbuf, size_t count, int *pret,
+					  struct page *page)
+{
+	unsigned long memcg_data = READ_ONCE(page->memcg_data);
+	struct mem_cgroup *memcg;
+	bool onlined;
+	char name[80];
+
+	if (!memcg_data)
+		return;
+
+	if (memcg_data & MEMCG_DATA_OBJCGS)
+		*pret += scnprintf(kbuf + *pret, count - *pret,
+				"Slab cache page\n");
+
+	memcg = page_memcg_check(page);
+	if (!memcg)
+		return;
+
+	onlined = (memcg->css.flags & CSS_ONLINE);
+	cgroup_name(memcg->css.cgroup, name, sizeof(name));
+	*pret += scnprintf(kbuf + *pret, count - *pret,
+			"Charged %sto %smemcg %s\n",
+			PageMemcgKmem(page) ? "(via objcg) " : "",
+			onlined ? "" : "offlined ",
+			name);
+}
+#else /* CONFIG_MEMCG */
+static inline void print_page_owner_memcg(char *kbuf, size_t count, int *pret,
+					  struct page *page) { }
+#endif /* CONFIG_MEMCG */
+
 static ssize_t
 print_page_owner(char __user *buf, size_t count, unsigned long pfn,
 		struct page *page, struct page_owner *page_owner,
@@ -365,6 +402,8 @@ print_page_owner(char __user *buf, size_t count, unsigned long pfn,
 			migrate_reason_names[page_owner->last_migrate_reason]);
 	}
 
+	print_page_owner_memcg(kbuf, count, &ret, page);
+
 	ret += snprintf(kbuf + ret, count - ret, "\n");
 	if (ret >= count)
 		goto err;
-- 
2.27.0



  parent reply	other threads:[~2022-01-31 19:23 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-31 19:23 [PATCH v3 0/4] mm/page_owner: Extend page_owner to show memcg information Waiman Long
2022-01-31 19:23 ` [PATCH v3 1/4] lib/vsprintf: Avoid redundant work with 0 size Waiman Long
2022-01-31 20:42   ` Mike Rapoport
2022-01-31 19:23 ` [PATCH v3 2/4] mm/page_owner: Use scnprintf() to avoid excessive buffer overrun check Waiman Long
2022-01-31 20:38   ` Roman Gushchin
2022-01-31 20:43   ` Mike Rapoport
2022-01-31 19:23 ` Waiman Long [this message]
2022-01-31 20:51   ` [PATCH v3 3/4] mm/page_owner: Print memcg information Mike Rapoport
2022-01-31 21:43     ` Waiman Long
2022-02-01  6:23       ` Mike Rapoport
2022-01-31 20:51   ` Roman Gushchin
2022-02-01 10:54   ` Michal Hocko
2022-02-01 17:04     ` Waiman Long
2022-02-02  8:49       ` Michal Hocko
2022-02-02 16:12         ` Waiman Long
2022-01-31 19:23 ` [PATCH v3 4/4] mm/page_owner: Record task command name Waiman Long
2022-01-31 20:54   ` Roman Gushchin
2022-01-31 21:46     ` Waiman Long
2022-01-31 22:03   ` [PATCH v4 " Waiman Long
2022-02-01 15:28     ` Michal Hocko
2022-02-02 16:53       ` Waiman Long
2022-02-03 12:10         ` Vlastimil Babka
2022-02-03 18:53           ` Waiman Long
2022-02-02 20:30   ` [PATCH v4 0/4] mm/page_owner: Extend page_owner to show memcg information Waiman Long
2022-02-02 23:06     ` Rafael Aquini
2022-02-02 20:30   ` [PATCH v4 1/4] lib/vsprintf: Avoid redundant work with 0 size Waiman Long
2022-02-08 10:08     ` Petr Mladek
2022-02-02 20:30   ` [PATCH v4 2/4] mm/page_owner: Use scnprintf() to avoid excessive buffer overrun check Waiman Long
2022-02-03 15:46     ` Vlastimil Babka
2022-02-03 18:49       ` Waiman Long
2022-02-08 10:51         ` Petr Mladek
2022-02-02 20:30   ` [PATCH v4 3/4] mm/page_owner: Print memcg information Waiman Long
2022-02-03  6:53     ` Mike Rapoport
2022-02-03 12:46     ` Michal Hocko
2022-02-03 19:03       ` Waiman Long
2022-02-07 17:20         ` Michal Hocko
2022-02-07 19:09           ` Andrew Morton
2022-02-07 19:33             ` Waiman Long
2022-02-02 20:30   ` [PATCH v4 4/4] mm/page_owner: Record task command name Waiman Long

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=20220131192308.608837-4-longman@redhat.com \
    --to=longman@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=andriy.shevchenko@linux.intel.com \
    --cc=aquini@redhat.com \
    --cc=cgroups@vger.kernel.org \
    --cc=guro@fb.com \
    --cc=hannes@cmpxchg.org \
    --cc=ira.weiny@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux@rasmusvillemoes.dk \
    --cc=mhocko@kernel.org \
    --cc=pmladek@suse.com \
    --cc=rientjes@google.com \
    --cc=rostedt@goodmis.org \
    --cc=rppt@kernel.org \
    --cc=senozhatsky@chromium.org \
    --cc=vdavydov.dev@gmail.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;
as well as URLs for NNTP newsgroup(s).