Re: [patch 2/4] mm: PageLRU no testset

[Nikita Danilov <nikita@clusterfs.com>]

Nick Piggin writes:
 > PG_lru is protected by zone->lru_lock. It does not need TestSet/TestClear
 > operations.
 > 
 > Signed-off-by: Nick Piggin <npiggin@suse.de>
 > 
 > Index: linux-2.6/mm/vmscan.c
 > ===================================================================
 > --- linux-2.6.orig/mm/vmscan.c
 > +++ linux-2.6/mm/vmscan.c
 > @@ -775,9 +775,10 @@ int isolate_lru_page(struct page *page)
 >  	if (PageLRU(page)) {
 >  		struct zone *zone = page_zone(page);
 >  		spin_lock_irq(&zone->lru_lock);
 > -		if (TestClearPageLRU(page)) {
 > +		if (PageLRU(page)) {
 >  			ret = 1;
 >  			get_page(page);
 > +			ClearPageLRU(page);

Why is that better? ClearPageLRU() is also "atomic" operation (in the
sense of using LOCK_PREFIX on x86), so it seems this change strictly
adds cycles to the hot-path when page is on LRU.

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