linux-kernel - [PATCH v3 0/3] mm: memcg: fix protection of reclaim target memcg

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <20221202031512.1365483-1-yosryahmed@google.com>
Date:   Fri,  2 Dec 2022 03:15:09 +0000
From:   Yosry Ahmed <yosryahmed@...gle.com>
To:     Andrew Morton <akpm@...ux-foundation.org>,
        Shakeel Butt <shakeelb@...gle.com>,
        Roman Gushchin <roman.gushchin@...ux.dev>,
        Johannes Weiner <hannes@...xchg.org>,
        Michal Hocko <mhocko@...e.com>, Yu Zhao <yuzhao@...gle.com>,
        Muchun Song <songmuchun@...edance.com>,
        Tejun Heo <tj@...nel.org>
Cc:     "Matthew Wilcox (Oracle)" <willy@...radead.org>,
        Vasily Averin <vasily.averin@...ux.dev>,
        Vlastimil Babka <vbabka@...e.cz>,
        Chris Down <chris@...isdown.name>,
        David Rientjes <rientjes@...gle.com>, cgroups@...r.kernel.org,
        linux-kernel@...r.kernel.org, linux-mm@...ck.org,
        Yosry Ahmed <yosryahmed@...gle.com>
Subject: [PATCH v3 0/3] mm: memcg: fix protection of reclaim target memcg

This series fixes a bug in calculating the protection of the reclaim
target memcg where we end up using stale effective protection values from
the last reclaim operation, instead of completely ignoring the
protection of the reclaim target as intended. More detailed explanation
and examples in patch 1, which includes the fix.
Patches 2 & 3 introduce a selftest case that catches the bug.

v2 -> v3:
- Refactoring selftest helper reclaim_until (suggested by Roman).
- Collected review tags on patch 1 & 3 from Roman (Thanks!).

v1 -> v2:
- Instead of adding a new helper, extended
  mem_cgroup_supports_protection() to check if the current memcg is the
  target memcg, renamed to mem_cgroup_unprotected() which is much easier
  to reason about (suggested by Roman).
- Add a selftest case to catch the bug (suggested by Roman).

Yosry Ahmed (3):
  mm: memcg: fix stale protection of reclaim target memcg
  selftests: cgroup: refactor proactive reclaim code to reclaim_until()
  selftests: cgroup: make sure reclaim target memcg is unprotected

 include/linux/memcontrol.h                    | 31 +++++--
 mm/vmscan.c                                   | 11 ++-
 .../selftests/cgroup/test_memcontrol.c        | 91 +++++++++++--------
 3 files changed, 82 insertions(+), 51 deletions(-)

-- 
2.39.0.rc0.267.gcb52ba06e7-goog