Re: [External] Re: [PATCH] mm/memcontrol: make the slab calculation consistent

[Muchun Song <songmuchun@bytedance.com>]

On Thu, Dec 3, 2020 at 11:21 AM Roman Gushchin <guro@fb.com> wrote:
>
> On Thu, Dec 03, 2020 at 10:53:33AM +0800, Muchun Song wrote:
> > On Thu, Dec 3, 2020 at 5:16 AM Roman Gushchin <guro@fb.com> wrote:
> > >
> > > On Wed, Dec 02, 2020 at 08:14:34PM +0800, Muchun Song wrote:
> > > > Although the ratio of the slab is one, we also should read the ratio
> > > > from the related memory_stats instead of hard-coding. And the local
> > > > variable of size is already the value of slab_unreclaimable. So we
> > > > do not need to read again. Simplify the code here.
> > > >
> > > > Signed-off-by: Muchun Song <songmuchun@bytedance.com>
> > > > ---
> > > >  mm/memcontrol.c | 22 +++++++++++++++++-----
> > > >  1 file changed, 17 insertions(+), 5 deletions(-)
> > >
> > > Hi Muchun!
> > >
> > > >
> > > > diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> > > > index 9922f1510956..03a9c64560f6 100644
> > > > --- a/mm/memcontrol.c
> > > > +++ b/mm/memcontrol.c
> > > > @@ -1545,12 +1545,22 @@ static int __init memory_stats_init(void)
> > > >       int i;
> > > >
> > > >       for (i = 0; i < ARRAY_SIZE(memory_stats); i++) {
> > > > +             switch (memory_stats[i].idx) {
> > > >  #ifdef CONFIG_TRANSPARENT_HUGEPAGE
> > > > -             if (memory_stats[i].idx == NR_ANON_THPS ||
> > > > -                 memory_stats[i].idx == NR_FILE_THPS ||
> > > > -                 memory_stats[i].idx == NR_SHMEM_THPS)
> > > > +             case NR_ANON_THPS:
> > > > +             case NR_FILE_THPS:
> > > > +             case NR_SHMEM_THPS:
> > > >                       memory_stats[i].ratio = HPAGE_PMD_SIZE;
> > > > +                     break;
> > > >  #endif
> > > > +             case NR_SLAB_UNRECLAIMABLE_B:
> > > > +                     VM_BUG_ON(i < 1);
> > > > +                     VM_BUG_ON(memory_stats[i - 1].idx != NR_SLAB_RECLAIMABLE_B);
> > >
> > > Please, convert these to BUILD_BUG_ON(), they don't have to be runtime checks.
> >
> > Agree. But here we cannot use BUILD_BUG_ON(). The compiler will
> > complain about it.
>
> We can!
>
> We just need to change the condition. All we really need to check is that
> NR_SLAB_UNRECLAIMABLE_B immediately following NR_SLAB_RECLAIMABLE_B.

But I think that we need to check that memory_stats[i] immediately following
memory_stats[j] where i is the index of NR_SLAB_UNRECLAIMABLE_B and
j is the index of NR_SLAB_RECLAIMABLE_B.

>
> Something like BUILD_BUG_ON(NR_SLAB_UNRECLAIMABLE_B != NR_SLAB_RECLAIMABLE_B + 1)

So this cannot work. Thanks.

> should work (completely untested).

>
> >
> > >
> > >
> > > > +                     break;
> > > > +             default:
> > > > +                     break;
> > > > +             }
> > > > +
> > > >               VM_BUG_ON(!memory_stats[i].ratio);
> > > >               VM_BUG_ON(memory_stats[i].idx >= MEMCG_NR_STAT);
> > > >       }
> > > > @@ -1587,8 +1597,10 @@ static char *memory_stat_format(struct mem_cgroup *memcg)
> > > >               seq_buf_printf(&s, "%s %llu\n", memory_stats[i].name, size);
> > > >
> > >
> > > Can you, please, add a small comment here stating that we're printing
> > > unreclaimable, reclaimable and the sum of both? It will simplify the reading of the code.
> >
> > Will do.
>
> Thank you!



-- 
Yours,
Muchun