>> Can you please tell me previous discussion url or mail subject? >> I mean, if it is costly and performance degression risk, we don't have to >> take my idea. > > Yes. You could see it by https://lkml.org/lkml/2011/4/30/81. I think Wu pointed out "lightweight vmscan could reclaim pages but stealed from another task case". It's very different with "most heavyweight vmscan still failed to reclaim any pages". The point is, IPIs cost depend on the frequency. stealing frequently occur on current logic, but vmscan priority==0 is? -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxxx 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>