On Wed, Nov 24, 2010 at 09:42:09PM +0800, Peter Zijlstra wrote: > On Wed, 2010-11-24 at 21:20 +0800, Wu Fengguang wrote: > > > (jiffies - bdi->write_bandwidth_update_time < elapsed) > > > > this will be true if someone else has _done_ overlapped estimation, > > otherwise it will equal: > > > > jiffies - bdi->write_bandwidth_update_time == elapsed > > > > Sorry the comment needs updating. > > Right, but its racy as hell.. Yeah, for N concurrent dirtiers, plus the background flusher, only one is able to update write_bandwidth[_update_time].. -- 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 policy in Canada: sign http://dissolvethecrtc.ca/ Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>