> I believe we do need some background compaction, especially > to help allocations from network interrupts. I completely agree. > If you believe the compaction is better done from some > other thread, I guess we could do that, but truthfully, if > kswapd spends a lot of time doing compaction, I made a > mistake somewhere :) I don't have much experience of compaction on real production systems. but I have a few bad experience of background lumpy reclaim. If much network allocation is happen when kswapd get stucked large order lumpy reclaim, kswapd can't work for making order-0 job.it was bad. I'm only worry about similar issue will occur. But, ok, we can fix it when we actually observed such thing. So, please go ahead. -- 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>