Re: [PATCH 2/2] mm/mm_init.c: remove reset_node_present_pages()

[David Hildenbrand]

On 01.06.23 18:26, Haifeng Xu wrote:
> reset_node_present_pages() only get called in hotadd_init_pgdat(), move
> the action that clear present pages to free_area_init_core_hotplug(), so
> the helper can be removed.
>
> Signed-off-by: Haifeng Xu <haifeng.xu@xxxxxxxxxx>
> ---

[...]

> diff --git a/mm/mm_init.c b/mm/mm_init.c
> index 78e67041ae9f..e24ea1db9e26 100644
> --- a/mm/mm_init.c
> +++ b/mm/mm_init.c
> @@ -1491,6 +1491,7 @@ void __ref free_area_init_core_hotplug(struct pglist_data *pgdat)
>   {
>   	int nid = pgdat->node_id;
>   	enum zone_type z;
> +	struct zone *zone;

You can declare that within the for loop.

>   	int cpu;
>   
>   	pgdat_init_internals(pgdat);
> @@ -1507,6 +1508,8 @@ void __ref free_area_init_core_hotplug(struct pglist_data *pgdat)
>   	pgdat->kswapd_order = 0;
>   	pgdat->kswapd_highest_zoneidx = 0;
>   	pgdat->node_start_pfn = 0;
> +	pgdat->node_present_pages = 0;
> +
>   	for_each_online_cpu(cpu) {
>   		struct per_cpu_nodestat *p;
>   
> @@ -1514,8 +1517,15 @@ void __ref free_area_init_core_hotplug(struct pglist_data *pgdat)
>   		memset(p, 0, sizeof(*p));
>   	}
>   
> -	for (z = 0; z < MAX_NR_ZONES; z++)
> -		zone_init_internals(&pgdat->node_zones[z], z, nid, 0);
> +	/* When memory is hot-added, all the memory is in offline state. So

/*
 * When ...


> +	 * clear all zones' present_pages and managed_pages because they will
> +	 * be updated in online_pages() and offline_pages().
> +	 */
> +	for (z = 0; z < MAX_NR_ZONES; z++) {
> +		zone = &pgdat->node_zones[z];

Too bad we don't have something like node_zone(pgdat, z)

Might be a reasonable cleanup staring at "git grep "pgdat->node_zones"" 
output ...

> +		zone->present_pages = 0;
> +		zone_init_internals(zone, z, nid, 0);
> +	}
>   }
>   #endif
>   


In general,

Acked-by: David Hildenbrand <david@xxxxxxxxxx>

--
Thanks,

David / dhildenb