Hi Johannes, On 4/19/22 14:46, Johannes Weiner wrote: > Hi Nico, > > On Tue, Apr 19, 2022 at 02:11:53PM -0400, Nico Pache wrote: >> I think its is important to note the issue we are seeing has greatly improved >> since the initial posting. However we have noticed that the issue is still >> present (and significantly worse) when cgroupV1 is set. >> >> We were initially testing with CgroupV1 and later found that the issue was not >> as bad in CgroupV2 (but was still an noticeable issue). This is also resulting >> in the splitting of THPs in the host kernel. > > When swappiness is 0, cgroup limit reclaim has a fixed SCAN_FILE > branch, so it shouldn't ever look at anon. I'm assuming you're getting > global reclaim mixed in. Indeed, I think we can try harder not to swap > for global reclaim if the user asks for that. We aren't actually utilizing the cgroup mechanism; however, switching between the two has a noticeable affect on the global reclaim of the system. This is not a writeback case either-- The reproducer simply reads. So I think we can rule out the v2 writeback controller being involved. My initial patch was also targeting swappiness=0 but this also occurs when >0. > > Can you try the below patch? of course thanks for that :) I'll let you know how it goes! Cheers, -- Nico
