Mel Gorman <mgorman@xxxxxxx> writes: > I think the only concrete outcome has been that userspace potentially > benefits if the total number of MADV_FREE pages is reported > globally. Even that is marginal as smaps has the information to tell > the difference between high RSS due to a memory leak and high RSS > usage due to MADV_FREE. The /proc/vmstats for MADV_FREE are of > marginal benefit given that they do not tell us much about the current > number of MADV_FREE pages in the system. We can implement a counter for MADV_FREE that increases when we ClearPageSwapBacked() and decrease when we SetPageSwapBacked() for an anonymous page. But this cannot count lazily freed pages correctly. Because when a clean MDV_FREE page becomes dirty, there's no page fault so we will not be notified. And you have never run into the MADV_FREE issues other than the memory leaking debugging... Best Regards, Huang, Ying
