| Message ID | 20230328221644.803272-6-yosryahmed@google.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | memcg: make rstat flushing irq and sleep | expand |
On Tue, Mar 28, 2023 at 3:17 PM Yosry Ahmed <yosryahmed@google.com> wrote: > > As Johannes notes in [1], stats_flush_lock is currently used to: > (a) Protect updated to stats_flush_threshold. > (b) Protect updates to flush_next_time. > (c) Serializes calls to cgroup_rstat_flush() based on those ratelimits. > > However: > > 1. stats_flush_threshold is already an atomic > > 2. flush_next_time is not atomic. The writer is locked, but the reader > is lockless. If the reader races with a flush, you could see this: > > if (time_after(jiffies, flush_next_time)) > spin_trylock() > flush_next_time = now + delay > flush() > spin_unlock() > spin_trylock() > flush_next_time = now + delay > flush() > spin_unlock() > > which means we already can get flushes at a higher frequency than > FLUSH_TIME during races. But it isn't really a problem. > > The reader could also see garbled partial updates, so it needs at > least READ_ONCE and WRITE_ONCE protection. > > 3. Serializing cgroup_rstat_flush() calls against the ratelimit > factors is currently broken because of the race in 2. But the race > is actually harmless, all we might get is the occasional earlier > flush. If there is no delta, the flush won't do much. And if there > is, the flush is justified. > > So the lock can be removed all together. However, the lock also served > the purpose of preventing a thundering herd problem for concurrent > flushers, see [2]. Use an atomic instead to serve the purpose of > unifying concurrent flushers. > > [1]https://lore.kernel.org/lkml/20230323172732.GE739026@cmpxchg.org/ > [2]https://lore.kernel.org/lkml/20210716212137.1391164-2-shakeelb@google.com/ > > Signed-off-by: Yosry Ahmed <yosryahmed@google.com> > Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Shakeel Butt <shakeelb@google.com>
On Tue 28-03-23 22:16:40, Yosry Ahmed wrote: > As Johannes notes in [1], stats_flush_lock is currently used to: > (a) Protect updated to stats_flush_threshold. > (b) Protect updates to flush_next_time. > (c) Serializes calls to cgroup_rstat_flush() based on those ratelimits. > > However: > > 1. stats_flush_threshold is already an atomic > > 2. flush_next_time is not atomic. The writer is locked, but the reader > is lockless. If the reader races with a flush, you could see this: > > if (time_after(jiffies, flush_next_time)) > spin_trylock() > flush_next_time = now + delay > flush() > spin_unlock() > spin_trylock() > flush_next_time = now + delay > flush() > spin_unlock() > > which means we already can get flushes at a higher frequency than > FLUSH_TIME during races. But it isn't really a problem. > > The reader could also see garbled partial updates, so it needs at > least READ_ONCE and WRITE_ONCE protection. Just a nit. Sounds more serious than it is actually. This would only happen if compiler decides to split the write. > 3. Serializing cgroup_rstat_flush() calls against the ratelimit > factors is currently broken because of the race in 2. But the race > is actually harmless, all we might get is the occasional earlier > flush. If there is no delta, the flush won't do much. And if there > is, the flush is justified. > > So the lock can be removed all together. However, the lock also served > the purpose of preventing a thundering herd problem for concurrent > flushers, see [2]. Use an atomic instead to serve the purpose of > unifying concurrent flushers. > > [1]https://lore.kernel.org/lkml/20230323172732.GE739026@cmpxchg.org/ > [2]https://lore.kernel.org/lkml/20210716212137.1391164-2-shakeelb@google.com/ > > Signed-off-by: Yosry Ahmed <yosryahmed@google.com> > Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Michal Hocko <mhocko@suse.com> > --- > mm/memcontrol.c | 18 +++++++++++------- > 1 file changed, 11 insertions(+), 7 deletions(-) > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > index ff39f78f962e..65750f8b8259 100644 > --- a/mm/memcontrol.c > +++ b/mm/memcontrol.c > @@ -585,8 +585,8 @@ mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_node *mctz) > */ > static void flush_memcg_stats_dwork(struct work_struct *w); > static DECLARE_DEFERRABLE_WORK(stats_flush_dwork, flush_memcg_stats_dwork); > -static DEFINE_SPINLOCK(stats_flush_lock); > static DEFINE_PER_CPU(unsigned int, stats_updates); > +static atomic_t stats_flush_ongoing = ATOMIC_INIT(0); > static atomic_t stats_flush_threshold = ATOMIC_INIT(0); > static u64 flush_next_time; > > @@ -636,15 +636,19 @@ static inline void memcg_rstat_updated(struct mem_cgroup *memcg, int val) > > static void __mem_cgroup_flush_stats(void) > { > - unsigned long flag; > - > - if (!spin_trylock_irqsave(&stats_flush_lock, flag)) > + /* > + * We always flush the entire tree, so concurrent flushers can just > + * skip. This avoids a thundering herd problem on the rstat global lock > + * from memcg flushers (e.g. reclaim, refault, etc). > + */ > + if (atomic_read(&stats_flush_ongoing) || > + atomic_xchg(&stats_flush_ongoing, 1)) > return; > > - flush_next_time = jiffies_64 + 2*FLUSH_TIME; > + WRITE_ONCE(flush_next_time, jiffies_64 + 2*FLUSH_TIME); > cgroup_rstat_flush_atomic(root_mem_cgroup->css.cgroup); > atomic_set(&stats_flush_threshold, 0); > - spin_unlock_irqrestore(&stats_flush_lock, flag); > + atomic_set(&stats_flush_ongoing, 0); > } > > void mem_cgroup_flush_stats(void) > @@ -655,7 +659,7 @@ void mem_cgroup_flush_stats(void) > > void mem_cgroup_flush_stats_ratelimited(void) > { > - if (time_after64(jiffies_64, flush_next_time)) > + if (time_after64(jiffies_64, READ_ONCE(flush_next_time))) > mem_cgroup_flush_stats(); > } > > -- > 2.40.0.348.gf938b09366-goog
On Wed, Mar 29, 2023 at 8:58 AM Michal Hocko <mhocko@suse.com> wrote: > > On Tue 28-03-23 22:16:40, Yosry Ahmed wrote: > > As Johannes notes in [1], stats_flush_lock is currently used to: > > (a) Protect updated to stats_flush_threshold. > > (b) Protect updates to flush_next_time. > > (c) Serializes calls to cgroup_rstat_flush() based on those ratelimits. > > > > However: > > > > 1. stats_flush_threshold is already an atomic > > > > 2. flush_next_time is not atomic. The writer is locked, but the reader > > is lockless. If the reader races with a flush, you could see this: > > > > if (time_after(jiffies, flush_next_time)) > > spin_trylock() > > flush_next_time = now + delay > > flush() > > spin_unlock() > > spin_trylock() > > flush_next_time = now + delay > > flush() > > spin_unlock() > > > > which means we already can get flushes at a higher frequency than > > FLUSH_TIME during races. But it isn't really a problem. > > > > The reader could also see garbled partial updates, so it needs at > > least READ_ONCE and WRITE_ONCE protection. > > Just a nit. Sounds more serious than it is actually. This would only > happen if compiler decides to split the write. Thanks for the note, Michal. I honestly quoted Johannes here as I do not have much expertise when it comes to this. I will add "if the compiler decides to split the write" to the commit log if I respin. > > > 3. Serializing cgroup_rstat_flush() calls against the ratelimit > > factors is currently broken because of the race in 2. But the race > > is actually harmless, all we might get is the occasional earlier > > flush. If there is no delta, the flush won't do much. And if there > > is, the flush is justified. > > > > So the lock can be removed all together. However, the lock also served > > the purpose of preventing a thundering herd problem for concurrent > > flushers, see [2]. Use an atomic instead to serve the purpose of > > unifying concurrent flushers. > > > > [1]https://lore.kernel.org/lkml/20230323172732.GE739026@cmpxchg.org/ > > [2]https://lore.kernel.org/lkml/20210716212137.1391164-2-shakeelb@google.com/ > > > > Signed-off-by: Yosry Ahmed <yosryahmed@google.com> > > Acked-by: Johannes Weiner <hannes@cmpxchg.org> > > Acked-by: Michal Hocko <mhocko@suse.com> > > > --- > > mm/memcontrol.c | 18 +++++++++++------- > > 1 file changed, 11 insertions(+), 7 deletions(-) > > > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c > > index ff39f78f962e..65750f8b8259 100644 > > --- a/mm/memcontrol.c > > +++ b/mm/memcontrol.c > > @@ -585,8 +585,8 @@ mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_node *mctz) > > */ > > static void flush_memcg_stats_dwork(struct work_struct *w); > > static DECLARE_DEFERRABLE_WORK(stats_flush_dwork, flush_memcg_stats_dwork); > > -static DEFINE_SPINLOCK(stats_flush_lock); > > static DEFINE_PER_CPU(unsigned int, stats_updates); > > +static atomic_t stats_flush_ongoing = ATOMIC_INIT(0); > > static atomic_t stats_flush_threshold = ATOMIC_INIT(0); > > static u64 flush_next_time; > > > > @@ -636,15 +636,19 @@ static inline void memcg_rstat_updated(struct mem_cgroup *memcg, int val) > > > > static void __mem_cgroup_flush_stats(void) > > { > > - unsigned long flag; > > - > > - if (!spin_trylock_irqsave(&stats_flush_lock, flag)) > > + /* > > + * We always flush the entire tree, so concurrent flushers can just > > + * skip. This avoids a thundering herd problem on the rstat global lock > > + * from memcg flushers (e.g. reclaim, refault, etc). > > + */ > > + if (atomic_read(&stats_flush_ongoing) || > > + atomic_xchg(&stats_flush_ongoing, 1)) > > return; > > > > - flush_next_time = jiffies_64 + 2*FLUSH_TIME; > > + WRITE_ONCE(flush_next_time, jiffies_64 + 2*FLUSH_TIME); > > cgroup_rstat_flush_atomic(root_mem_cgroup->css.cgroup); > > atomic_set(&stats_flush_threshold, 0); > > - spin_unlock_irqrestore(&stats_flush_lock, flag); > > + atomic_set(&stats_flush_ongoing, 0); > > } > > > > void mem_cgroup_flush_stats(void) > > @@ -655,7 +659,7 @@ void mem_cgroup_flush_stats(void) > > > > void mem_cgroup_flush_stats_ratelimited(void) > > { > > - if (time_after64(jiffies_64, flush_next_time)) > > + if (time_after64(jiffies_64, READ_ONCE(flush_next_time))) > > mem_cgroup_flush_stats(); > > } > > > > -- > > 2.40.0.348.gf938b09366-goog > > -- > Michal Hocko > SUSE Labs
diff --git a/mm/memcontrol.c b/mm/memcontrol.c index ff39f78f962e..65750f8b8259 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -585,8 +585,8 @@ mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_node *mctz) */ static void flush_memcg_stats_dwork(struct work_struct *w); static DECLARE_DEFERRABLE_WORK(stats_flush_dwork, flush_memcg_stats_dwork); -static DEFINE_SPINLOCK(stats_flush_lock); static DEFINE_PER_CPU(unsigned int, stats_updates); +static atomic_t stats_flush_ongoing = ATOMIC_INIT(0); static atomic_t stats_flush_threshold = ATOMIC_INIT(0); static u64 flush_next_time; @@ -636,15 +636,19 @@ static inline void memcg_rstat_updated(struct mem_cgroup *memcg, int val) static void __mem_cgroup_flush_stats(void) { - unsigned long flag; - - if (!spin_trylock_irqsave(&stats_flush_lock, flag)) + /* + * We always flush the entire tree, so concurrent flushers can just + * skip. This avoids a thundering herd problem on the rstat global lock + * from memcg flushers (e.g. reclaim, refault, etc). + */ + if (atomic_read(&stats_flush_ongoing) || + atomic_xchg(&stats_flush_ongoing, 1)) return; - flush_next_time = jiffies_64 + 2*FLUSH_TIME; + WRITE_ONCE(flush_next_time, jiffies_64 + 2*FLUSH_TIME); cgroup_rstat_flush_atomic(root_mem_cgroup->css.cgroup); atomic_set(&stats_flush_threshold, 0); - spin_unlock_irqrestore(&stats_flush_lock, flag); + atomic_set(&stats_flush_ongoing, 0); } void mem_cgroup_flush_stats(void) @@ -655,7 +659,7 @@ void mem_cgroup_flush_stats(void) void mem_cgroup_flush_stats_ratelimited(void) { - if (time_after64(jiffies_64, flush_next_time)) + if (time_after64(jiffies_64, READ_ONCE(flush_next_time))) mem_cgroup_flush_stats(); }
As Johannes notes in [1], stats_flush_lock is currently used to: (a) Protect updated to stats_flush_threshold. (b) Protect updates to flush_next_time. (c) Serializes calls to cgroup_rstat_flush() based on those ratelimits. However: 1. stats_flush_threshold is already an atomic 2. flush_next_time is not atomic. The writer is locked, but the reader is lockless. If the reader races with a flush, you could see this: if (time_after(jiffies, flush_next_time)) spin_trylock() flush_next_time = now + delay flush() spin_unlock() spin_trylock() flush_next_time = now + delay flush() spin_unlock() which means we already can get flushes at a higher frequency than FLUSH_TIME during races. But it isn't really a problem. The reader could also see garbled partial updates, so it needs at least READ_ONCE and WRITE_ONCE protection. 3. Serializing cgroup_rstat_flush() calls against the ratelimit factors is currently broken because of the race in 2. But the race is actually harmless, all we might get is the occasional earlier flush. If there is no delta, the flush won't do much. And if there is, the flush is justified. So the lock can be removed all together. However, the lock also served the purpose of preventing a thundering herd problem for concurrent flushers, see [2]. Use an atomic instead to serve the purpose of unifying concurrent flushers. [1]https://lore.kernel.org/lkml/20230323172732.GE739026@cmpxchg.org/ [2]https://lore.kernel.org/lkml/20210716212137.1391164-2-shakeelb@google.com/ Signed-off-by: Yosry Ahmed <yosryahmed@google.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> --- mm/memcontrol.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-)