Re: [PATCH] mm, vmscan: Clear PGDAT_WRITEBACK when zone is balanced

2017-02-05 Thread Michal Hocko
On Sun 05-02-17 11:11:13, Michal Hocko wrote: > On Fri 03-02-17 20:32:22, Mel Gorman wrote: > > Hillf Danton pointed out that since commit 1d82de618dd ("mm, vmscan: > > make kswapd reclaim in terms of nodes") that PGDAT_WRITEBACK is no longer > > cleared. It was not noticed as triggering it require

Re: [PATCH] mm, vmscan: Clear PGDAT_WRITEBACK when zone is balanced

2017-02-05 Thread Michal Hocko
On Fri 03-02-17 20:32:22, Mel Gorman wrote: > Hillf Danton pointed out that since commit 1d82de618dd ("mm, vmscan: > make kswapd reclaim in terms of nodes") that PGDAT_WRITEBACK is no longer > cleared. It was not noticed as triggering it requires pages under writeback > to cycle twice through the L

Re: [PATCH] mm, vmscan: Clear PGDAT_WRITEBACK when zone is balanced

2017-02-03 Thread Hillf Danton
On February 04, 2017 4:32 AM Mel Gorman wrote: > > Hillf Danton pointed out that since commit 1d82de618dd ("mm, vmscan: > make kswapd reclaim in terms of nodes") that PGDAT_WRITEBACK is no longer > cleared. It was not noticed as triggering it requires pages under writeback > to cycle twice throu

Re: [PATCH] mm, vmscan: Clear PGDAT_WRITEBACK when zone is balanced

2017-02-03 Thread Johannes Weiner
On Fri, Feb 03, 2017 at 08:32:22PM +, Mel Gorman wrote: > Hillf Danton pointed out that since commit 1d82de618dd ("mm, vmscan: > make kswapd reclaim in terms of nodes") that PGDAT_WRITEBACK is no longer > cleared. It was not noticed as triggering it requires pages under writeback > to cycle twi

[PATCH] mm, vmscan: Clear PGDAT_WRITEBACK when zone is balanced

2017-02-03 Thread Mel Gorman
Hillf Danton pointed out that since commit 1d82de618dd ("mm, vmscan: make kswapd reclaim in terms of nodes") that PGDAT_WRITEBACK is no longer cleared. It was not noticed as triggering it requires pages under writeback to cycle twice through the LRU and before kswapd gets stalled. Historically, suc