On Mon, Apr 9, 2012 at 2:23 PM, Ying Han <yinghan@xxxxxxxxxx> wrote: > On Mon, Apr 9, 2012 at 12:50 PM, Andrew Morton > <akpm@xxxxxxxxxxxxxxxxxxxx> wrote: >> On Mon, 9 Apr 2012 12:42:04 -0700 >> Ying Han <yinghan@xxxxxxxxxx> wrote: >> >>> This reverts commit c38446cc65e1f2b3eb8630c53943b94c4f65f670. >>> >>> Before the commit, the code makes senses to me but not after the commit. The >>> "nr_reclaimed" is the number of pages reclaimed by scanning through the memcg's >>> lru lists. The "nr_to_reclaim" is the target value for the whole function. For >>> example, we like to early break the reclaim if reclaimed 32 pages under direct >>> reclaim (not DEF_PRIORITY). >>> >>> After the reverted commit, the target "nr_to_reclaim" is decremented each time >>> by "nr_reclaimed" but we still use it to compare the "nr_reclaimed". It just >>> doesn't make sense to me... >>> >>> Signed-off-by: Ying Han <yinghan@xxxxxxxxxx> >>> --- >>> mm/vmscan.c | 7 +------ >>> 1 files changed, 1 insertions(+), 6 deletions(-) >>> >>> diff --git a/mm/vmscan.c b/mm/vmscan.c >>> index 33c332b..1a51868 100644 >>> --- a/mm/vmscan.c >>> +++ b/mm/vmscan.c >>> @@ -2107,12 +2107,7 @@ restart: >>> * with multiple processes reclaiming pages, the total >>> * freeing target can get unreasonably large. >>> */ >>> - if (nr_reclaimed >= nr_to_reclaim) >>> - nr_to_reclaim = 0; >>> - else >>> - nr_to_reclaim -= nr_reclaimed; >>> - >>> - if (!nr_to_reclaim && priority < DEF_PRIORITY) >>> + if (nr_reclaimed >= nr_to_reclaim && priority < DEF_PRIORITY) >>> break; >>> } >>> blk_finish_plug(&plug); >> >> This code is all within a loop: the "goto restart" thing. We reset >> nr_reclaimed to zero each time around that loop. nr_to_reclaim is (or >> rather, was) constant throughout the entire function. >> >> Comparing nr_reclaimed (whcih is reset each time around the loop) to >> nr_to_reclaim made no sense. >> >> I think the code as it stands is ugly. It would be better to make >> nr_to_reclaim a const and to add another local total_reclaimed, and >> compare that with nr_to_reclaim. > > Ok, I will resend the patch w/ the "total_reclaimed" change. > > --Ying I have the patch ready but I am not sure if that is what we want. If we use total_reclaimed to compare w/ nr_to_reclaim, we end up reducing the amount of work to reclaim before compaction(should_continue_reclaim() is true case). --Ying > > Or just stop resetting nr_reclaimed >> each time around the loop. >> -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href