linux-master/mm/vmscan.c

282 	 * overestimating the reclaimed amount (potentially under-reclaiming).
284 	 * Only count such pages for global reclaim to prevent under-reclaiming
1182 		 * Before reclaiming the folio, try to relocate
1681 	 * this disrupts the LRU order when reclaiming for lower zones but
1948 	 * pressure reclaiming all the clean cache. And in some cases,
2397 	 * proportional to the cost of reclaiming each list, as
3124 static int folio_inc_gen(struct lruvec *lruvec, struct folio *folio, bool reclaiming)
3144 		if (reclaiming)
5663 	 * Global reclaiming within direct reclaim at DEF_PRIORITY is a normal
5668 	 * reclaiming implies that kswapd is not keeping up and it is best to
5701 		 * stop reclaiming one LRU and reduce the amount scanning
5775  * It will give up earlier than that if there is difficulty reclaiming pages.
5819 	 * inactive lists are large enough, continue reclaiming
6097 		 * Take care memory controller reclaiming has small influence
6223 		 * If we're getting trouble reclaiming, start doing
6756  * Returns the order kswapd finished reclaiming at.
6816 		 * then consider reclaiming from all zones. This has a dual
6871 		 * referenced before reclaiming. All pages are rotated
6877 		 * If we're getting trouble reclaiming, start doing writepage
6916 		 * progress in reclaiming pages
6977 	 * Return the order kswapd stopped reclaiming at as
7156 		 * reclaim fails then kswapd falls back to reclaiming for
7158 		 * for the order it finished reclaiming at (reclaim_order)
7178  * pgdat.  It will wake up kcompactd after reclaiming memory.  If kswapd reclaim
7233  * LRU order by reclaiming preferentially