[Devel] [PATCH rh7 v2 00/10] mm/mem_cgroup_iter: Reduce the number of iterator restarts upon cgroup removals

[Kirill Tkhai]

On 24.02.2021 15:36, Kirill Tkhai wrote:
> On 24.02.2021 13:47, Konstantin Khorenko wrote:
>> May thanks to Kirill Tkhai for his bright ideas and review!
>>
>> Problem description from the user point of view:
>>   * the Node is slow
>>   * the Node has a lot of free RAM
>>   * the Node has a lot of swapin/swapout
>>   * kswapd is always running
>>
>> Problem in a nutshell from technical point of view:
>>   * kswapd is looping in shrink_zone() inside the loop
>>       do {} while ((memcg = mem_cgroup_iter(root, memcg, &reclaim)));
>>     (and never goes trough the outer loop)
>>   * there are a quite a number of memory cgroups of the Node (~1000)
>>   * some cgroups are hard to reclaim (reclaim may take ~3 seconds),
>>     this is because of very busy disk due to permanent swapin/swapout
>>   * mem_cgroup_iter() does not have success scanning all cgroups
>>     in a row, it restarts from the root cgroup one time after
>>     another (after different number of cgroups scanned)
>>
>> Q: Why does mem_cgroup_iter() restart from the root memcg?
>> A: Because it is invalidated once some memory cgroup is
>>    destroyed on the Node.
>>    Note: ANY memory cgroup destroy on the Node leads to iter
>>    restart.
>>
>> The following patchset solves this problem in the following way:
>> there is no need to restart the iter until we see the iter has
>> the position which is exactly the memory cgroup being destroyed.
>>
>> The patchset ensures the iter->last_visited is NULL-ified on
>> invalidation and thus restarts only in the unlikely case when
>> the iter points to the memcg being destroyed.
>>
>> https://jira.sw.ru/browse/PSBM-123655
>>
>> v2 changes:
>>  - reverted 2 patches in this code which were focused on syncronizing
>>    updates of iter->last_visited and ->last_dead_count
>>    (as we are getting rid of iter->last_dead_count at all)
>>  - use rcu primitives to access iter->last_visited
>>
>> Konstantin Khorenko (10):
>>   Revert "mm/memcg: fix css_tryget(),css_put() imbalance"
>>   Revert "mm/memcg: use seqlock to protect reclaim_iter updates"
>>   mm/mem_cgroup_iter: Make 'iter->last_visited' a bit more stable
>>   mm/mem_cgroup_iter: Always assign iter->last_visited under rcu
>>   mm/mem_cgroup_iter: NULL-ify 'last_visited' for invalidated iterators
>>   mm/mem_cgroup_iter: Provide _iter_invalidate() the dying memcg as an
>>     argument
>>   mm/mem_cgroup_iter: Invalidate iterators only if needed
>>   mm/mem_cgroup_iter: Don't bother checking 'dead_count' anymore
>>   mm/mem_cgroup_iter: Cleanup mem_cgroup_iter_load()
>>   mm/mem_cgroup_iter: Drop dead_count related infrastructure
>>
>>  mm/memcontrol.c | 133 +++++++++++++++++++++++++++---------------------
>>  1 file changed, 76 insertions(+), 57 deletions(-)
> 
> Somewhere we need
> 
>         struct mem_cgroup __rcu *last_visited;
> 
> with a comment that it's of RCU type because of we provide pointed memory are
> not freed one grace period after we rewrite it.

Hm, but we don't provide this, do we?