Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Tue, 28 Mar 2023 06:16:29 +0000
Mime-Version: 1.0
Message-ID: <20230328061638.203420-1-yosryahmed@google.com>
Subject: [PATCH v1 0/9] memcg: make rstat flushing irq and sleep friendly
From:   Yosry Ahmed <yosryahmed@google.com>
To:     Tejun Heo <tj@kernel.org>, Josef Bacik <josef@toxicpanda.com>,
        Jens Axboe <axboe@kernel.dk>,
        Zefan Li <lizefan.x@bytedance.com>,
        Johannes Weiner <hannes@cmpxchg.org>,
        Michal Hocko <mhocko@kernel.org>,
        Roman Gushchin <roman.gushchin@linux.dev>,
        Shakeel Butt <shakeelb@google.com>,
        Muchun Song <muchun.song@linux.dev>,
        Andrew Morton <akpm@linux-foundation.org>,
        "=?UTF-8?q?Michal=20Koutn=C3=BD?=" <mkoutny@suse.com>
Cc:     Vasily Averin <vasily.averin@linux.dev>, cgroups@vger.kernel.org,
        linux-block@vger.kernel.org, linux-kernel@vger.kernel.org,
        linux-mm@kvack.org, bpf@vger.kernel.org,
        Yosry Ahmed <yosryahmed@google.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Currently, all calls to flush memcg stats use the atomic variant for
rstat flushing, cgroup_rstat_flush_irqsafe(), which keeps interrupts
disabled throughout flushing and does not sleep. Flushing stats is an
expensive operation, and we should avoid doing it atomically where
possible. Otherwise, we may end up doing a lot of work without
rescheduling and with interrupts disabled unnecessarily.

Patches 1 and 2 are cleanups requested during reviews of prior versions
of this series.

Patch 3 makes sure we never try to flush from within an irq context, and
patch 4 adds a WARN_ON_ONCE() to make sure we catch any violations.

Patches 5 to 8 introduce separate variants of mem_cgroup_flush_stats()
for atomic and non-atomic flushing, and make sure we only flush the
stats atomically when necessary.

Patch 9 is a slightly tangential optimization that limits the work done
by rstat flushing in some scenarios.

RFC -> v1:
- Dropped patch 1 that attempted to make the global rstat lock a non-irq
  lock, will follow up on that separetly (Shakeel).
- Dropped stats_flush_lock entirely, replaced by an atomic (Johannes).
- Renamed cgroup_rstat_flush_irqsafe() to cgroup_rstat_flush_atomic()
  instead of removing it (Johannes).
- Added a patch to rename mem_cgroup_flush_stats_delayed() to
  mem_cgroup_flush_stats_ratelimited() (Johannes).
- Separate APIs for flushing memcg stats in atomic and non-atomic
  contexts instead of a boolean argument (Johannes).
- Added patches 3 & 4 to make sure we never flush from irq context
  (Shakeel & Johannes).

Yosry Ahmed (9):
  cgroup: rename cgroup_rstat_flush_"irqsafe" to "atomic"
  memcg: rename mem_cgroup_flush_stats_"delayed" to "ratelimited"
  memcg: do not flush stats in irq context
  cgroup: rstat: add WARN_ON_ONCE() if flushing outside task context
  memcg: replace stats_flush_lock with an atomic
  memcg: sleep during flushing stats in safe contexts
  workingset: memcg: sleep when flushing stats in workingset_refault()
  vmscan: memcg: sleep when flushing stats during reclaim
  memcg: do not modify rstat tree for zero updates

 include/linux/cgroup.h     |  2 +-
 include/linux/memcontrol.h |  9 +++-
 kernel/cgroup/rstat.c      |  6 ++-
 mm/memcontrol.c            | 86 ++++++++++++++++++++++++++++++++------
 mm/workingset.c            |  4 +-
 5 files changed, 87 insertions(+), 20 deletions(-)

-- 
2.40.0.348.gf938b09366-goog