Re: [patch v2] mm, tmp: only collapse hugepages to nodes with affinity for zone_reclaim_mode

[Vlastimil Babka]

On 07/16/2014 03:22 AM, Bob Liu wrote:
>> @@ -2545,6 +2571,11 @@ static int khugepaged_scan_pmd(struct mm_struct *mm,
>>  		 * hit record.
>>  		 */
>>  		node = page_to_nid(page);
>> +		if (node != last_node) {
>> +			if (khugepaged_scan_abort(node))
>> +				goto out_unmap;
> 
> Nitpick: How about not break the loop but only reset the related
> khugepaged_node_load[] to zero. E.g. modify khugepaged_scan_abort() like
> this:
> if (node_distance(nid, i) > RECLAIM_DISTANCE)
>    khugepaged_node_load[i] = 0;
> 
> By this way, we may have a chance to find a more suitable node.

Hm theoretically there might be a suitable node, but this approach wouldn't
work. By resetting it to zero you forget that there ever was node 'i'. If there
is no more base page from node 'i', the load remains zero and the next call with
'nid' will think that 'nid' is OK.

So the correct way would be more complex but I wonder if it's worth the trouble...

>> +			last_node = node;
>> +		}
>>  		khugepaged_node_load[node]++;
>>  		VM_BUG_ON_PAGE(PageCompound(page), page);
>>  		if (!PageLRU(page) || PageLocked(page) || !PageAnon(page))
>> 
> 

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/