On Sat, Jun 30, 2018 at 12:09:27PM +0200, Vlastimil Babka wrote:

On 06/29/2018 11:12 PM, Roman Gushchin wrote:

quoted

quoted

The vmstat counter NR_INDIRECTLY_RECLAIMABLE_BYTES was introduced by commit
eb59254608bc ("mm: introduce NR_INDIRECTLY_RECLAIMABLE_BYTES") with the goal of
accounting objects that can be reclaimed, but cannot be allocated via a
SLAB_RECLAIM_ACCOUNT cache. This is now possible via kmalloc() with
__GFP_RECLAIMABLE flag, and the dcache external names user is converted.

The counter is however still useful for accounting direct page allocations
(i.e. not slab) with a shrinker, such as the ION page pool. So keep it, and:

Btw, it looks like I've another example of usefulness of this counter:
dynamic per-cpu data.

Hmm, but are those reclaimable? Most likely not in general? Do you have
examples that are?

If these per-cpu data is something like per-cpu refcounters,
which are using to manage reclaimable objects (e.g. cgroup css objects).
Of course, they are not always reclaimable, but in certain states.

Thanks!