[PATCH 5/5] mm: spill memcg percpu stats and events before releasing

2019-03-12 Thread Roman Gushchin
Spill percpu stats and events data to corresponding before releasing percpu memory. Although per-cpu stats are never exactly precise, dropping them on floor regularly may lead to an accumulation of an error. So, it's safer to sync them before releasing. To minimize the number of atomic updates, l

Re: [PATCH 5/5] mm: spill memcg percpu stats and events before releasing

2019-03-11 Thread Roman Gushchin
On Mon, Mar 11, 2019 at 01:38:25PM -0400, Johannes Weiner wrote: > On Thu, Mar 07, 2019 at 03:00:33PM -0800, Roman Gushchin wrote: > > Spill percpu stats and events data to corresponding before releasing > > percpu memory. > > > > Although per-cpu stats are never exactly precise, dropping them on

Re: [PATCH 5/5] mm: spill memcg percpu stats and events before releasing

2019-03-11 Thread Johannes Weiner
On Thu, Mar 07, 2019 at 03:00:33PM -0800, Roman Gushchin wrote: > Spill percpu stats and events data to corresponding before releasing > percpu memory. > > Although per-cpu stats are never exactly precise, dropping them on > floor regularly may lead to an accumulation of an error. So, it's > safer

[PATCH 5/5] mm: spill memcg percpu stats and events before releasing

2019-03-07 Thread Roman Gushchin
Spill percpu stats and events data to corresponding before releasing percpu memory. Although per-cpu stats are never exactly precise, dropping them on floor regularly may lead to an accumulation of an error. So, it's safer to sync them before releasing. To minimize the number of atomic updates, l