On Tue, 2011-09-27 at 19:28 +0800, Michal Hocko wrote: > On Tue 27-09-11 15:23:07, Shaohua Li wrote: > > has_under_min_watermark_zone is used to detect if there is GFP_ATOMIC allocation > > failure risk. For a high end_zone, if any zone below or equal to it has min > > matermark ok, we have no risk. But current logic is any zone has min watermark > > not ok, then we have risk. This is wrong to me. > > This, however, means that we skip congestion_wait more often as ZONE_DMA > tend to be mostly balanced, right? This would mean that kswapd could hog > CPU more. We actually might have more congestion_wait, as now if any zone can meet min watermark, we don't have has_under_min_watermark_zone set so do congestion_wait > Does this fix any particular problem you are seeing? No, just thought the logic is wrong. Thanks, Shaohua -- 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=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>