Re: [v5][PATCH 5/6] mm: vmscan: batch shrink_page_list() lockingoperations

[Dave Hansen]

On 06/03/2013 11:02 PM, Minchan Kim wrote:
>> > Why do we need new lru list instead of using @free_pages?
> I got your point that @free_pages could have freed page by
> put_page_testzero of shrink_page_list and they don't have
> valid mapping so __remove_mapping_batch's mapping_release_page
> would access NULL pointer.
> 
> I think it would be better to mention it in comment. :(
> Otherwise, I suggest we can declare another new LIST_HEAD to
> accumulate pages freed by put_page_testzero in shrink_page_list
> so __remove_mapping_batch don't have to declare temporal LRU list
> and can remove unnecessary list_move operation.

If I respin them again, I'll add a comment.

I guess we could splice the whole list over at once instead of moving
the pages individually.  But, what are we trying to optimize here?
Saving a list_head worth of space on the stack?

--
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/