Re: [PATCH 2/5] mm: vmscan: make memcg slab shrink lockless

[Qi Zheng <zhengqi.arch@bytedance.com>]

On 2023/2/23 03:58, Kirill Tkhai wrote:
> On 22.02.2023 10:32, Qi Zheng wrote:
>>
>>
>> On 2023/2/22 05:28, Kirill Tkhai wrote:
>>> On 20.02.2023 12:16, Qi Zheng wrote:
>> <...>
>>>>      void reparent_shrinker_deferred(struct mem_cgroup *memcg)
>>>>    {
>>>> -    int i, nid;
>>>> +    int i, nid, srcu_idx;
>>>>        long nr;
>>>>        struct mem_cgroup *parent;
>>>>        struct shrinker_info *child_info, *parent_info;
>>>> @@ -429,16 +443,16 @@ void reparent_shrinker_deferred(struct mem_cgroup *memcg)
>>>>            parent = root_mem_cgroup;
>>>>          /* Prevent from concurrent shrinker_info expand */
>>>> -    down_read(&shrinker_rwsem);
>>>> +    srcu_idx = srcu_read_lock(&shrinker_srcu);
>>>
>>> Don't we still have to be protected against parallel expand_one_shrinker_info()?
>>>
>>> It looks like parent->nodeinfo[nid]->shrinker_info pointer may be changed in expand*
>>> right after we've dereferenced it here.
>>
>> Hi Kirill,
>>
>> Oh, indeed. We may wrongly reparent the child's nr_deferred to the old
>> parent's nr_deferred (it is about to be freed by call_srcu).
>>
>> The reparent_shrinker_deferred() will only be called on the offline
>> path (not a hotspot path), so we may be able to use shrinker_mutex
>> introduced later for protection. What do you think?
> 
> It looks good for me. One more thing I'd analyzed is whether we want to have
> is two reparent_shrinker_deferred() are executing in parallel.

I see that mem_cgroup_css_offline() is already protected by
cgroup_mutex, so maybe shrinker_mutex is enough here. :)

> 
> Possible, we should leave rwsem there as it was used before..
>   
>>>
>>>>        for_each_node(nid) {
>>>> -        child_info = shrinker_info_protected(memcg, nid);
>>>> -        parent_info = shrinker_info_protected(parent, nid);
>>>> +        child_info = shrinker_info_srcu(memcg, nid);
>>>> +        parent_info = shrinker_info_srcu(parent, nid);
>>>>            for (i = 0; i < shrinker_nr_max; i++) {
>>>>                nr = atomic_long_read(&child_info->nr_deferred[i]);
>>>>                atomic_long_add(nr, &parent_info->nr_deferred[i]);
>>>>            }
>>>>        }
>>>> -    up_read(&shrinker_rwsem);
>>>> +    srcu_read_unlock(&shrinker_srcu, srcu_idx);
>>>>    }
>>>>      static bool cgroup_reclaim(struct scan_control *sc)
>>>> @@ -891,15 +905,14 @@ static unsigned long shrink_slab_memcg(gfp_t gfp_mask, int nid,
>>>>    {
>>>>        struct shrinker_info *info;
>>>>        unsigned long ret, freed = 0;
>>>> +    int srcu_idx;
>>>>        int i;
>>>>          if (!mem_cgroup_online(memcg))
>>>>            return 0;
>>>>    -    if (!down_read_trylock(&shrinker_rwsem))
>>>> -        return 0;
>>>> -
>>>> -    info = shrinker_info_protected(memcg, nid);
>>>> +    srcu_idx = srcu_read_lock(&shrinker_srcu);
>>>> +    info = shrinker_info_srcu(memcg, nid);
>>>>        if (unlikely(!info))
>>>>            goto unlock;
>>>>    @@ -949,14 +962,9 @@ static unsigned long shrink_slab_memcg(gfp_t gfp_mask, int nid,
>>>>                    set_shrinker_bit(memcg, nid, i);
>>>>            }
>>>>            freed += ret;
>>>> -
>>>> -        if (rwsem_is_contended(&shrinker_rwsem)) {
>>>> -            freed = freed ? : 1;
>>>> -            break;
>>>> -        }
>>>>        }
>>>>    unlock:
>>>> -    up_read(&shrinker_rwsem);
>>>> +    srcu_read_unlock(&shrinker_srcu, srcu_idx);
>>>>        return freed;
>>>>    }
>>>>    #else /* CONFIG_MEMCG */
>>>
>>
> 

-- 
Thanks,
Qi