Messages in this thread | | | Date | Fri, 10 Dec 2010 12:53:39 +0900 | Subject | Re: [PATCH] vmscan: make kswapd use a correct order | From | Minchan Kim <> |
| |
On Fri, Dec 10, 2010 at 7:13 AM, Andrew Morton <akpm@linux-foundation.org> wrote: > On Fri, 3 Dec 2010 01:00:49 +0900 > Minchan Kim <minchan.kim@gmail.com> wrote: > >> +static bool kswapd_try_to_sleep(pg_data_t *pgdat, int order) > > OT: kswapd_try_to_sleep() does a > trace_mm_vmscan_kswapd_sleep(pgdat->node_id) if it sleeps for a long > time, but doesn't trace anything at all if it does a short sleep. > Where's the sense in that? >
AFAIU, short sleep is _sleep_ but that trace's goal is to count only long sleep. In addition, short sleep is a just ready to go or not long sleep so I think we don't need short sleep trace. And for knowing short sleep count, we can use KSWAPD_{LOW|HIGH}_WMARK_HIT_QUICKLY.
-- Kind regards, Minchan Kim -- 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/
| |