mm-commits.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@linux-foundation.org>
To: akpm@linux-foundation.org, chris@chrisdown.name,
	hannes@cmpxchg.org, linux-mm@kvack.org, mhocko@suse.com,
	mm-commits@vger.kernel.org, torvalds@linux-foundation.org,
	xlpang@linux.alibaba.com
Subject: [patch 02/19] mm: memcg: fix memcg reclaim soft lockup
Date: Fri, 04 Sep 2020 16:35:27 -0700	[thread overview]
Message-ID: <20200904233527.1lOZT1GXC%akpm@linux-foundation.org> (raw)
In-Reply-To: <20200904163454.4db0e6ce0c4584d2653678a3@linux-foundation.org>

From: Xunlei Pang <xlpang@linux.alibaba.com>
Subject: mm: memcg: fix memcg reclaim soft lockup

We've met softlockup with "CONFIG_PREEMPT_NONE=y", when the target memcg
doesn't have any reclaimable memory.

It can be easily reproduced as below:
 watchdog: BUG: soft lockup - CPU#0 stuck for 111s![memcg_test:2204]
 CPU: 0 PID: 2204 Comm: memcg_test Not tainted 5.9.0-rc2+ #12
 Call Trace:
  shrink_lruvec+0x49f/0x640
  shrink_node+0x2a6/0x6f0
  do_try_to_free_pages+0xe9/0x3e0
  try_to_free_mem_cgroup_pages+0xef/0x1f0
  try_charge+0x2c1/0x750
  mem_cgroup_charge+0xd7/0x240
  __add_to_page_cache_locked+0x2fd/0x370
  add_to_page_cache_lru+0x4a/0xc0
  pagecache_get_page+0x10b/0x2f0
  filemap_fault+0x661/0xad0
  ext4_filemap_fault+0x2c/0x40
  __do_fault+0x4d/0xf9
  handle_mm_fault+0x1080/0x1790

It only happens on our 1-vcpu instances, because there's no chance for oom
reaper to run to reclaim the to-be-killed process.

Add a cond_resched() at the upper shrink_node_memcgs() to solve this
issue, this will mean that we will get a scheduling point for each memcg
in the reclaimed hierarchy without any dependency on the reclaimable
memory in that memcg thus making it more predictable.

Link: http://lkml.kernel.org/r/1598495549-67324-1-git-send-email-xlpang@linux.alibaba.com
Signed-off-by: Xunlei Pang <xlpang@linux.alibaba.com>
Acked-by: Chris Down <chris@chrisdown.name>
Acked-by: Michal Hocko <mhocko@suse.com>
Suggested-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 mm/vmscan.c |    8 ++++++++
 1 file changed, 8 insertions(+)

--- a/mm/vmscan.c~mm-memcg-fix-memcg-reclaim-soft-lockup
+++ a/mm/vmscan.c
@@ -2615,6 +2615,14 @@ static void shrink_node_memcgs(pg_data_t
 		unsigned long reclaimed;
 		unsigned long scanned;
 
+		/*
+		 * This loop can become CPU-bound when target memcgs
+		 * aren't eligible for reclaim - either because they
+		 * don't have any reclaimable pages, or because their
+		 * memory is explicitly protected. Avoid soft lockups.
+		 */
+		cond_resched();
+
 		mem_cgroup_calculate_protection(target_memcg, memcg);
 
 		if (mem_cgroup_below_min(memcg)) {
_

  parent reply	other threads:[~2020-09-04 23:35 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-09-04 23:34 incoming Andrew Morton
2020-09-04 23:35 ` [patch 01/19] memcg: fix use-after-free in uncharge_batch Andrew Morton
2020-09-04 23:35 ` Andrew Morton [this message]
2020-09-04 23:35 ` [patch 03/19] mm: slub: fix conversion of freelist_corrupted() Andrew Morton
2020-09-04 23:35 ` [patch 04/19] MAINTAINERS: update Cavium/Marvell entries Andrew Morton
2020-09-04 23:35 ` [patch 05/19] MAINTAINERS: add LLVM maintainers Andrew Morton
2020-09-04 23:35 ` [patch 06/19] MAINTAINERS: IA64: mark Status as Odd Fixes only Andrew Morton
2020-09-04 23:35 ` [patch 07/19] mm: track page table modifications in __apply_to_page_range() Andrew Morton
2020-09-04 23:35 ` [patch 08/19] ipc: adjust proc_ipc_sem_dointvec definition to match prototype Andrew Morton
2020-09-04 23:35 ` [patch 09/19] fork: adjust sysctl_max_threads " Andrew Morton
2020-09-04 23:35 ` [patch 10/19] checkpatch: fix the usage of capture group ( ... ) Andrew Morton
2020-09-04 23:35 ` [patch 11/19] mm: madvise: fix vma user-after-free Andrew Morton
2020-09-04 23:35 ` [patch 12/19] mm/migrate: fixup setting UFFD_WP flag Andrew Morton
2020-09-04 23:36 ` [patch 13/19] mm/rmap: fixup copying of soft dirty and uffd ptes Andrew Morton
2020-09-04 23:36 ` [patch 14/19] mm/migrate: remove unnecessary is_zone_device_page() check Andrew Morton
2020-09-04 23:36 ` [patch 15/19] mm/migrate: preserve soft dirty in remove_migration_pte() Andrew Morton
2020-09-04 23:36 ` [patch 16/19] mm/hugetlb: try preferred node first when alloc gigantic page from cma Andrew Morton
2020-09-04 23:36 ` [patch 17/19] mm/hugetlb: fix a race between hugetlb sysctl handlers Andrew Morton
2020-09-04 23:36 ` [patch 18/19] mm/khugepaged.c: fix khugepaged's request size in collapse_file Andrew Morton
2020-09-04 23:36 ` [patch 19/19] include/linux/log2.h: add missing () around n in roundup_pow_of_two() Andrew Morton

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=20200904233527.1lOZT1GXC%akpm@linux-foundation.org \
    --to=akpm@linux-foundation.org \
    --cc=chris@chrisdown.name \
    --cc=hannes@cmpxchg.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@suse.com \
    --cc=mm-commits@vger.kernel.org \
    --cc=torvalds@linux-foundation.org \
    --cc=xlpang@linux.alibaba.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).