Re: [PATCH v4 4/5] mm/zswap: Add per-memcg stat for proactive writeback

[Yosry Ahmed <yosry@kernel.org>]

[..]
>  static int zswap_writeback_entry(struct zswap_entry *entry,
> -				 swp_entry_t swpentry)
> +				 swp_entry_t swpentry,
> +				 bool proactive)

IIUC, if we refactor the code as I suggested in previous changes, we
don't really need to add an argument here..

>  {
>  	struct xarray *tree;
>  	pgoff_t offset = swp_offset(swpentry);
> @@ -1045,6 +1047,15 @@ static int zswap_writeback_entry(struct zswap_entry *entry,
>  	if (entry->objcg)
>  		count_objcg_events(entry->objcg, ZSWPWB, 1);
>  
> +	if (proactive && entry->objcg) {
> +		struct mem_cgroup *memcg;
> +
> +		rcu_read_lock();
> +		memcg = obj_cgroup_memcg(entry->objcg);
> +		mod_memcg_state(memcg, MEMCG_ZSWPWB_PROACTIVE_B, entry->length);
> +		rcu_read_unlock();
> +	}

..and this chunk of code would end up in zswap_proactive_writeback().