Messages in this thread | | | Date | Fri, 10 Dec 2010 10:42:54 +0000 | From | Mel Gorman <> | Subject | Re: [PATCH 4/6] mm: kswapd: Reset kswapd_max_order and classzone_idx after reading |
| |
On Fri, Dec 10, 2010 at 01:03:01AM +0900, Minchan Kim wrote: > On Fri, Dec 10, 2010 at 12:59 AM, Minchan Kim <minchan.kim@gmail.com> wrote: > > On Thu, Dec 09, 2010 at 11:18:18AM +0000, Mel Gorman wrote: > >> When kswapd wakes up, it reads its order and classzone from pgdat and > >> calls balance_pgdat. While its awake, it potentially reclaimes at a high > >> order and a low classzone index. This might have been a once-off that > >> was not required by subsequent callers. However, because the pgdat > >> values were not reset, they remain artifically high while > >> balance_pgdat() is running and potentially kswapd enters a second > >> unnecessary reclaim cycle. Reset the pgdat order and classzone index > >> after reading. > >> > >> Signed-off-by: Mel Gorman <mel@csn.ul.ie> > > Reviewed-by: Minchan Kim <minchan.kim@gmail.com> > > Nitpick. > Shouldn't this patch be merged with 1/6? >
I don't think so as it's a standalone fix. For example, if this was merged on its own, the "order" should still be reset after reading.
-- Mel Gorman Part-time Phd Student Linux Technology Center University of Limerick IBM Dublin Software Lab
| |