Re: [PATCH -V12 1/3] NUMA Balancing: add page promotion counter

[Huang, Ying]

Oscar Salvador <osalvador@xxxxxxx> writes:

> On Wed, Feb 16, 2022 at 03:38:13PM +0800, Huang Ying wrote:
>> In a system with multiple memory types, e.g. DRAM and PMEM, the CPU
>> and DRAM in one socket will be put in one NUMA node as before, while
>> the PMEM will be put in another NUMA node as described in the
>> description of the commit c221c0b0308f ("device-dax: "Hotplug"
>> persistent memory for use like normal RAM").  So, the NUMA balancing
>> mechanism will identify all PMEM accesses as remote access and try to
>> promote the PMEM pages to DRAM.
>> 
>> To distinguish the number of the inter-type promoted pages from that
>> of the inter-socket migrated pages.  A new vmstat count is added.  The
>> counter is per-node (count in the target node).  So this can be used
>> to identify promotion imbalance among the NUMA nodes.
>> 
>> Signed-off-by: "Huang, Ying" <ying.huang@xxxxxxxxx>
>> Reviewed-by: Yang Shi <shy828301@xxxxxxxxx>
>> Tested-by: Baolin Wang <baolin.wang@xxxxxxxxxxxxxxxxx>
>> Reviewed-by: Baolin Wang <baolin.wang@xxxxxxxxxxxxxxxxx>
>> Acked-by: Johannes Weiner <hannes@xxxxxxxxxxx>
>> Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
>> Cc: Michal Hocko <mhocko@xxxxxxxx>
>> Cc: Rik van Riel <riel@xxxxxxxxxxx>
>> Cc: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>
>> Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
>> Cc: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>
>> Cc: Zi Yan <ziy@xxxxxxxxxx>
>> Cc: Wei Xu <weixugc@xxxxxxxxxx>
>> Cc: osalvador <osalvador@xxxxxxx>
>> Cc: Shakeel Butt <shakeelb@xxxxxxxxxx>
>> Cc: zhongjiang-ali <zhongjiang-ali@xxxxxxxxxxxxxxxxx>
>> Cc: linux-kernel@xxxxxxxxxxxxxxx
>> Cc: linux-mm@xxxxxxxxx
>> ---
> ...
>
>> @@ -2072,6 +2072,7 @@ int migrate_misplaced_page(struct page *page, struct vm_area_struct *vma,
>>  	pg_data_t *pgdat = NODE_DATA(node);
>>  	int isolated;
>>  	int nr_remaining;
>> +	int nr_succeeded;
>
> I think we should make this consistent and make it "unsigned int".
> That is what migrate_pages() expects, and what the other caller using
> nr_succeeded (demote_page_list()) already uses.
> Unless there is a strong reason not to do so.

Yes.  Will do that in the next version.

> Reviewed-by: Oscar Salvador <osalvador@xxxxxxx>

Thanks!

Best Regards,
Huang, Ying