Re: [RFC][PATCH 0/9] [v3] Migrate Pages in lieu of discard

[Yang Shi <shy828301@gmail.com>]

On Tue, Aug 18, 2020 at 11:50 AM Dave Hansen
<dave.hansen@linux.intel.com> wrote:
>
> todo:
> Changes since (https://lwn.net/Articles/824830/):
>  * Use higher-level migrate_pages() API approach from Yang Shi's
>    earlier patches.

Thanks for incorporating in this. I believe this would be more efficient.

>  * made sure to actually check node_reclaim_mode's new bit
>  * disabled migration entirely before introducing RECLAIM_MIGRATE
>  * Replace GFP_NOWAIT with explicit __GFP_KSWAPD_RECLAIM and
>    comment why we want that.
>  * Comment on effects of that keep multiple source nodes from
>    sharing target nodes
>
> The full series is also available here:
>
>         https://github.com/hansendc/linux/tree/automigrate-20200818
>
> --
>
> We're starting to see systems with more and more kinds of memory such
> as Intel's implementation of persistent memory.
>
> Let's say you have a system with some DRAM and some persistent memory.
> Today, once DRAM fills up, reclaim will start and some of the DRAM
> contents will be thrown out.  Allocations will, at some point, start
> falling over to the slower persistent memory.
>
> That has two nasty properties.  First, the newer allocations can end
> up in the slower persistent memory.  Second, reclaimed data in DRAM
> are just discarded even if there are gobs of space in persistent
> memory that could be used.
>
> This set implements a solution to these problems.  At the end of the
> reclaim process in shrink_page_list() just before the last page
> refcount is dropped, the page is migrated to persistent memory instead
> of being dropped.
>
> While I've talked about a DRAM/PMEM pairing, this approach would
> function in any environment where memory tiers exist.
>
> This is not perfect.  It "strands" pages in slower memory and never
> brings them back to fast DRAM.  Other things need to be built to
> promote hot pages back to DRAM.
>
> This is also all based on an upstream mechanism that allows
> persistent memory to be onlined and used as if it were volatile:
>
>         http://lkml.kernel.org/r/20190124231441.37A4A305@viggo.jf.intel.com
>
> == Open Issues ==
>
>  * For cpusets and memory policies that restrict allocations
>    to PMEM, is it OK to demote to PMEM?  Do we need a cgroup-
>    level API to opt-in or opt-out of these migrations?

I'm not sure if this would work or not, but AFAICT, it is unlikely.
The nasty thing about cgroupv1 is you may end up having threads from
the same process in different cgroups although it is rare.

My initial thought is to make cpuset process only (the threads in the
same process must be in the same cpuset group), but it sounds not too
feasible either since it may break some user configurations.

>  * Migration failures will result in pages being unreclaimable.
>    Need to be able to fall back to normal reclaim.

That should be transient, shouldn't? The migration logic is supposed
to wake up kswapd on pmem nodes, then the pages should become
migratable in later retry.

>
>
> Cc: Yang Shi <yang.shi@linux.alibaba.com>
> Cc: David Rientjes <rientjes@google.com>
> Cc: Huang Ying <ying.huang@intel.com>
> Cc: Dan Williams <dan.j.williams@intel.com>
>
> --
>
> Dave Hansen (5):
>       mm/numa: node demotion data structure and lookup
>       mm/vmscan: Attempt to migrate page in lieu of discard
>       mm/numa: automatically generate node migration order
>       mm/vmscan: never demote for memcg reclaim
>       mm/numa: new reclaim mode to enable reclaim-based migration
>
> Keith Busch (2):
>       mm/migrate: Defer allocating new page until needed
>       mm/vmscan: Consider anonymous pages without swap
>
> Yang Shi (1):
>       mm/vmscan: add page demotion counter
>
>  Documentation/admin-guide/sysctl/vm.rst |    9
>  include/linux/migrate.h                 |    6
>  include/linux/node.h                    |    9
>  include/linux/vm_event_item.h           |    2
>  include/trace/events/migrate.h          |    3
>  mm/debug.c                              |    1
>  mm/internal.h                           |    1
>  mm/migrate.c                            |  400 ++++++++++++++++++++++++++------
>  mm/page_alloc.c                         |    2
>  mm/vmscan.c                             |   88 ++++++-
>  mm/vmstat.c                             |    2
>  11 files changed, 439 insertions(+), 84 deletions(-)