lkml.org 
[lkml]   [2022]   [Apr]   [14]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH v2 1/5] mm: demotion: Set demotion list differently
On Thu, Apr 14, 2022 at 03:09:42PM +0800, ying.huang@intel.com wrote:
> On Wed, 2022-04-13 at 14:52 +0530, Jagdish Gediya wrote:
> > Sharing used_targets between multiple nodes in a single
> > pass limits some of the opportunities for demotion target
> > sharing.
> >
> > Don't share the used targets between multiple nodes in a
> > single pass, instead accumulate all the used targets in
> > source nodes shared by all pass, and reset 'used_targets'
> > to source nodes while finding demotion targets for any new
> > node.
> >
> > This results into some more opportunities to share demotion
> > targets between multiple source nodes, e.g. with below NUMA
> > topology, where node 0 & 1 are cpu + dram nodes, node 2 & 3
> > are equally slower memory only nodes, and node 4 is slowest
> > memory only node,
> >
> > available: 5 nodes (0-4)
> > node 0 cpus: 0 1
> > node 0 size: n MB
> > node 0 free: n MB
> > node 1 cpus: 2 3
> > node 1 size: n MB
> > node 1 free: n MB
> > node 2 cpus:
> > node 2 size: n MB
> > node 2 free: n MB
> > node 3 cpus:
> > node 3 size: n MB
> > node 3 free: n MB
> > node 4 cpus:
> > node 4 size: n MB
> > node 4 free: n MB
> > node distances:
> > node 0 1 2 3 4
> >   0: 10 20 40 40 80
> >   1: 20 10 40 40 80
> >   2: 40 40 10 40 80
> >   3: 40 40 40 10 80
> >   4: 80 80 80 80 10
> >
> > The existing implementation gives below demotion targets,
> >
> > node demotion_target
> >  0 3, 2
> >  1 4
> >  2 X
> >  3 X
> >  4 X
> >
> > With this patch applied, below are the demotion targets,
> >
> > node demotion_target
> >  0 3, 2
> >  1 3, 2
> >  2 4
> >  3 4
> >  4 X
> >
> > e.g. with below NUMA topology, where node 0, 1 & 2 are
> > cpu + dram nodes and node 3 is slow memory node,
> >
> > available: 4 nodes (0-3)
> > node 0 cpus: 0 1
> > node 0 size: n MB
> > node 0 free: n MB
> > node 1 cpus: 2 3
> > node 1 size: n MB
> > node 1 free: n MB
> > node 2 cpus: 4 5
> > node 2 size: n MB
> > node 2 free: n MB
> > node 3 cpus:
> > node 3 size: n MB
> > node 3 free: n MB
> > node distances:
> > node 0 1 2 3
> >   0: 10 20 20 40
> >   1: 20 10 20 40
> >   2: 20 20 10 40
> >   3: 40 40 40 10
> >
> > The existing implementation gives below demotion targets,
> >
> > node demotion_target
> >  0 3
> >  1 X
> >  2 X
> >  3 X
> >
> > With this patch applied, below are the demotion targets,
> >
> > node demotion_target
> >  0 3
> >  1 3
> >  2 3
> >  3 X
> >
>
> With the [PATCH v1], you have describe the demotion order changes for
> the following system, I guess there's no change with [PATCH v2]?

Yes, there is no change with v2.

> With below NUMA topology, where node 0 & 2 are cpu + dram
> nodes and node 1 & 3 are slow memory nodes,
>
> available: 4 nodes (0-3)
> node 0 cpus: 0 1
> node 0 size: n MB
> node 0 free: n MB
> node 1 cpus:
> node 1 size: n MB
> node 1 free: n MB
> node 2 cpus: 2 3
> node 2 size: n MB
> node 2 free: n MB
> node 3 cpus:
> node 3 size: n MB
> node 3 free: n MB
> node distances:
> node 0 1 2 3
> 0: 10 40 20 80
> 1: 40 10 80 80
> 2: 20 80 10 40
> 3: 80 80 40 10
>
> And, what is the demotion order for the following system with [PATCH
> v2]?
>
> Node 0 & 2 are cpu + dram nodes and node 1 are slow
> memory node near node 0,
>
> available: 3 nodes (0-2)
> node 0 cpus: 0 1
> node 0 size: n MB
> node 0 free: n MB
> node 1 cpus:
> node 1 size: n MB
> node 1 free: n MB
> node 2 cpus: 2 3
> node 2 size: n MB
> node 2 free: n MB
> node distances:
> node 0 1 2
> 0: 10 40 20
> 1: 40 10 80
> 2: 20 80 10

node 1 is demotion target for both node 0 and node 2 with this patch.
node 1 is demotion target only for node 0 with existing implementation,
however if node 1 is near to node 2 instead of node 0, still existing
implementation will give node 1 as demotion target only for node 0 which
is not the correct behavior.

for both the scenario, with this patch applied, node 1 will be demotion
target for both 0 and 2.

> Best Regards,
> Huang, Ying
>
>
> [snip]
>
>
>
Best regards,
Jagdish

\
 
 \ /
  Last update: 2022-04-14 10:49    [W:0.105 / U:0.300 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site