Re: [PATCH v2] memcg: add charging of already allocated slab objects

[Yosry Ahmed <yosryahmed@google.com>]

[..]
>
> Another reason is memory savings, if we have a small subset of objects in
> KMALLOC_NORMAL caches accounted, there might be e.g. one vector per a slab
> just to account on object while the rest is unaccounted. Separating between
> kmalloc and kmalloc-cg caches keeps the former with no vectors and the
> latter with fully used vectors.

Makes sense.

>
> > Wouldn't it be easier to special case the specific slab cache used for
> > the objcg vector or use a dedicated cache for it instead of using
> > kmalloc caches to begin with?
>
> The problem is the vector isn't a fixed size, it depends on how many objects
> a particular slab (not even a particular cache) has.

Oh right, I missed that part. Thanks for pointing it out.

>
> > Anyway, I am fine with any approach you and/or the slab maintainers
> > prefer, as long as we make things clear. If you keep the following
> > approach as-is, please expand the comment or refer to the commit you
> > just referenced.
> >
> > Personally, I prefer either explicitly special casing the slab cache
> > used for the objcgs vector, explicitly tagging KMALLOC_NORMAL
> > allocations, or having a dedicated documented helper that finds the
> > slab cache kmalloc type (if any) or checks if it is a KMALLOC_NORMAL
> > cache.
>
> A helper to check is_kmalloc_normal() would be better than defining
> KMALLOC_TYPE and using it directly, yes. We don't need to handle any other
> types now until anyone needs those.

is_kmalloc_normal() sounds good to me.

Thanks, Vlastimil.