Revert "Bias the placement of kernel pages at lower PFNs"

This reverts commit 5adc5be7.

Alexey Dobriyan reports that it causes huge slowdowns under some loads,
in his case a "mkfs.ext2" on a 30G partition.  With the placement bias,
the mkfs took over four minutes, with it reverted it's back to about ten
seconds for Alexey.
Reported-and-tested-by: NAlexey Dobriyan <adobriyan@gmail.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

mm/page_alloc.c

@@ -749,23 +749,6 @@ int move_freepages_block(struct zone *zone, struct page *page, int migratetype)
 	return move_freepages(zone, start_page, end_page, migratetype);
 }
 
-/* Return the page with the lowest PFN in the list */
-static struct page *min_page(struct list_head *list)
-{
-	unsigned long min_pfn = -1UL;
-	struct page *min_page = NULL, *page;;
-
-	list_for_each_entry(page, list, lru) {
-		unsigned long pfn = page_to_pfn(page);
-		if (pfn < min_pfn) {
-			min_pfn = pfn;
-			min_page = page;
-		}
-	}
-
-	return min_page;
-}
-
 /* Remove an element from the buddy allocator from the fallback list */
 static struct page *__rmqueue_fallback(struct zone *zone, int order,
 						int start_migratetype)
@@ -789,11 +772,8 @@ static struct page *__rmqueue_fallback(struct zone *zone, int order,
 			if (list_empty(&area->free_list[migratetype]))
 				continue;
 
-			/* Bias kernel allocations towards low pfns */
 			page = list_entry(area->free_list[migratetype].next,
 					struct page, lru);
-			if (unlikely(start_migratetype != MIGRATE_MOVABLE))
-				page = min_page(&area->free_list[migratetype]);
 			area->nr_free--;
 
 			/*