Messages in this thread | | | Subject | Re: [PATCH] mm: allow to decrease swap.max below actual swap usage | From | Rik van Riel <> | Date | Mon, 16 Apr 2018 11:07:22 -0400 |
| |
On Sun, 2018-04-15 at 18:39 -0700, Tejun Heo wrote: > Hello, Roman. > > The reclaim behavior is a bit worrisome. > > * It disables an entire swap area while reclaim is in progress. Most > systems only have one swap area, so this would disable allocating > new swap area for everyone.
That could easily cause OOM kills on systems.
I prefer Tejun's simple approach, of having the system slowly reduce swap use below the new maximum limit.
> * The reclaim seems very inefficient. IIUC, it has to read every > swap > page to see whether the page belongs to the target memcg and for > each matching page, which involves walking page mm's and page > tables.
One of my Outreachy interns, Kelley Nielsen, worked on making swap reclaim more efficient by scanning the virtual address space of processes.
Unfortunately, we ran into some unforseen issues with that approach that we never managed to sort out :(
> Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Rik van Riel <riel@surriel.com>
-- All Rights Reversed.[unhandled content-type:application/pgp-signature] | |