Re: [mm, net-next v2] mm: net: memcg accounting for TCP rx zerocopy

[Michal Hocko]

On Wed 24-03-21 13:53:34, Shakeel Butt wrote:
[...]
> > Given that's the case, the options seem to be:
> > 1) Use a page flag - with the downside that they are a severely
> > limited resource,
> > 2) Use some bits inside page->memcg_data - this I believe Johannes had
> > reasons against, and it isn't always the case that MEMCG support is
> > enabled.
> > 3) Use compound_dtor - but I think this would have problems for the
> > prior reasons.
> 
> I don't think Michal is suggesting to use PageCompound() or
> PageHead(). He is suggesting to add a more general page flag
> (PageHasDestructor) and corresponding page->dtor, so other potential
> users can use it too.

Yes, that is eaxactly my point. If there is a page flag to use for a
specific destruction then we can use an already existing scheme. I have
fully digested Johannes' other reply so I might be still missing
something but fundamentally if sombody knows that the particular part of
the page is not used (most really should) then the page can claim
destructor by a flag and the freeing routine would just call that
callback. Or is there any reason why othe subsystems outside of
networking couldn't claim their own callback?
-- 
Michal Hocko
SUSE Labs