Re: [v6 PATCH 07/11] mm: vmscan: add per memcg shrinker nr_deferred

2021-02-05 Thread Kirill Tkhai
On 04.02.2021 20:17, Yang Shi wrote: > On Thu, Feb 4, 2021 at 12:31 AM Kirill Tkhai wrote: >> >> On 03.02.2021 20:20, Yang Shi wrote: >>> Currently the number of deferred objects are per shrinker, but some slabs, >>> for example, >>> vfs inode/dentry cache are per memcg, this would result in poor

Re: [v6 PATCH 07/11] mm: vmscan: add per memcg shrinker nr_deferred

2021-02-05 Thread Yang Shi
On Fri, Feb 5, 2021 at 6:38 AM Kirill Tkhai wrote: > > On 04.02.2021 20:17, Yang Shi wrote: > > On Thu, Feb 4, 2021 at 12:31 AM Kirill Tkhai wrote: > >> > >> On 03.02.2021 20:20, Yang Shi wrote: > >>> Currently the number of deferred objects are per shrinker, but some > >>> slabs, for example, >

Re: [v6 PATCH 07/11] mm: vmscan: add per memcg shrinker nr_deferred

2021-02-04 Thread Yang Shi
On Thu, Feb 4, 2021 at 12:31 AM Kirill Tkhai wrote: > > On 03.02.2021 20:20, Yang Shi wrote: > > Currently the number of deferred objects are per shrinker, but some slabs, > > for example, > > vfs inode/dentry cache are per memcg, this would result in poor isolation > > among memcgs. > > > > The

Re: [v6 PATCH 07/11] mm: vmscan: add per memcg shrinker nr_deferred

2021-02-04 Thread Kirill Tkhai
On 03.02.2021 20:20, Yang Shi wrote: > Currently the number of deferred objects are per shrinker, but some slabs, > for example, > vfs inode/dentry cache are per memcg, this would result in poor isolation > among memcgs. > > The deferred objects typically are generated by __GFP_NOFS allocations,

[v6 PATCH 07/11] mm: vmscan: add per memcg shrinker nr_deferred

2021-02-03 Thread Yang Shi
Currently the number of deferred objects are per shrinker, but some slabs, for example, vfs inode/dentry cache are per memcg, this would result in poor isolation among memcgs. The deferred objects typically are generated by __GFP_NOFS allocations, one memcg with excessive __GFP_NOFS allocations