Received-SPF: pass (google.com: domain of linux-kernel+bounces-95104-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Precedence: bulk
MIME-Version: 1.0
References: <20240307061425.21013-1-ioworker0@gmail.com> <CAGsJ_4xcRvZGdpPh1qcFTnTnDUbwz6WreQ=L_UO+oU2iFm9EPg@mail.gmail.com>
 <CAK1f24k2G_DSEjuqqqPyY0f7+btpYbjfoyMH7btLfP8nkasCTQ@mail.gmail.com>
In-Reply-To: <CAK1f24k2G_DSEjuqqqPyY0f7+btpYbjfoyMH7btLfP8nkasCTQ@mail.gmail.com>
From: Barry Song <21cnbao@gmail.com>
Date: Thu, 7 Mar 2024 21:10:10 +1300
Message-ID: <CAGsJ_4xREM-P1mFqeM-s3-cJ9czb6PXwizb-3hOhwaF6+QM5QA@mail.gmail.com>
Subject: Re: [PATCH v2 1/1] mm/madvise: enhance lazyfreeing with mTHP in madvise_free
To: Lance Yang <ioworker0@gmail.com>, david@redhat.com, 
	Vishal Moola <vishal.moola@gmail.com>
Cc: akpm@linux-foundation.org, zokeefe@google.com, ryan.roberts@arm.com, 
	shy828301@gmail.com, mhocko@suse.com, fengwei.yin@intel.com, 
	xiehuan09@gmail.com, wangkefeng.wang@huawei.com, songmuchun@bytedance.com, 
	peterx@redhat.com, minchan@kernel.org, linux-mm@kvack.org, 
	linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Thu, Mar 7, 2024 at 9:00=E2=80=AFPM Lance Yang <ioworker0@gmail.com> wro=
te:
>
> Hey Barry,
>
> Thanks for taking time to review!
>
> On Thu, Mar 7, 2024 at 3:00=E2=80=AFPM Barry Song <21cnbao@gmail.com> wro=
te:
> >
> > On Thu, Mar 7, 2024 at 7:15=E2=80=AFPM Lance Yang <ioworker0@gmail.com>=
 wrote:
> > >
> [...]
> > > +static inline bool can_mark_large_folio_lazyfree(unsigned long addr,
> > > +                                                struct folio *folio,=
 pte_t *start_pte)
> > > +{
> > > +       int nr_pages =3D folio_nr_pages(folio);
> > > +       fpb_t flags =3D FPB_IGNORE_DIRTY | FPB_IGNORE_SOFT_DIRTY;
> > > +
> > > +       for (int i =3D 0; i < nr_pages; i++)
> > > +               if (page_mapcount(folio_page(folio, i)) !=3D 1)
> > > +                       return false;
> >
> > we have moved to folio_estimated_sharers though it is not precise, so
> > we don't do
> > this check with lots of loops and depending on the subpage's mapcount.
>
> If we don't check the subpage=E2=80=99s mapcount, and there is a cow foli=
o associated
> with this folio and the cow folio has smaller size than this folio,
> should we still
> mark this folio as lazyfree?

I agree, this is true. However, we've somehow accepted the fact that
folio_likely_mapped_shared
can result in false negatives or false positives to balance the
overhead.  So I really don't know :-)

Maybe David and Vishal can give some comments here.

>
> > BTW, do we need to rebase our work against David's changes[1]?
> > [1] https://lore.kernel.org/linux-mm/20240227201548.857831-1-david@redh=
at.com/
>
> Yes, we should rebase our work against David=E2=80=99s changes.
>
> >
> > > +
> > > +       return nr_pages =3D=3D folio_pte_batch(folio, addr, start_pte=
,
> > > +                                        ptep_get(start_pte), nr_page=
s, flags, NULL);
> > > +}
> > > +
> > >  static int madvise_free_pte_range(pmd_t *pmd, unsigned long addr,
> > >                                 unsigned long end, struct mm_walk *wa=
lk)
> > >
> > > @@ -676,11 +690,45 @@ static int madvise_free_pte_range(pmd_t *pmd, u=
nsigned long addr,
> > >                  */
> > >                 if (folio_test_large(folio)) {
> > >                         int err;
> > > +                       unsigned long next_addr, align;
> > >
> > > -                       if (folio_estimated_sharers(folio) !=3D 1)
> > > -                               break;
> > > -                       if (!folio_trylock(folio))
> > > -                               break;
> > > +                       if (folio_estimated_sharers(folio) !=3D 1 ||
> > > +                           !folio_trylock(folio))
> > > +                               goto skip_large_folio;
> >
> >
> > I don't think we can skip all the PTEs for nr_pages, as some of them mi=
ght be
> > pointing to other folios.
> >
> > for example, for a large folio with 16PTEs, you do MADV_DONTNEED(15-16)=
,
> > and write the memory of PTE15 and PTE16, you get page faults, thus PTE1=
5
> > and PTE16 will point to two different small folios. We can only skip wh=
en we
> > are sure nr_pages =3D=3D folio_pte_batch() is sure.
>
> Agreed. Thanks for pointing that out.
>
> >
> > > +
> > > +                       align =3D folio_nr_pages(folio) * PAGE_SIZE;
> > > +                       next_addr =3D ALIGN_DOWN(addr + align, align)=
;
> > > +
> > > +                       /*
> > > +                        * If we mark only the subpages as lazyfree, =
or
> > > +                        * cannot mark the entire large folio as lazy=
free,
> > > +                        * then just split it.
> > > +                        */
> > > +                       if (next_addr > end || next_addr - addr !=3D =
align ||
> > > +                           !can_mark_large_folio_lazyfree(addr, foli=
o, pte))
> > > +                               goto split_large_folio;
> > > +
> > > +                       /*
> > > +                        * Avoid unnecessary folio splitting if the l=
arge
> > > +                        * folio is entirely within the given range.
> > > +                        */
> > > +                       folio_clear_dirty(folio);
> > > +                       folio_unlock(folio);
> > > +                       for (; addr !=3D next_addr; pte++, addr +=3D =
PAGE_SIZE) {
> > > +                               ptent =3D ptep_get(pte);
> > > +                               if (pte_young(ptent) || pte_dirty(pte=
nt)) {
> > > +                                       ptent =3D ptep_get_and_clear_=
full(
> > > +                                               mm, addr, pte, tlb->f=
ullmm);
> > > +                                       ptent =3D pte_mkold(ptent);
> > > +                                       ptent =3D pte_mkclean(ptent);
> > > +                                       set_pte_at(mm, addr, pte, pte=
nt);
> > > +                                       tlb_remove_tlb_entry(tlb, pte=
, addr);
> > > +                               }
> >
> > Can we do this in batches? for a CONT-PTE mapped large folio, you are u=
nfolding
> > and folding again. It seems quite expensive.
>
> Thanks for your suggestion. I'll do this in batches in v3.
>
> Thanks again for your time!
>
> Best,
> Lance
>
> >
> > > +                       }
> > > +                       folio_mark_lazyfree(folio);
> > > +                       goto next_folio;
> > > +
> > > +split_large_folio:
> > >                         folio_get(folio);
> > >                         arch_leave_lazy_mmu_mode();
> > >                         pte_unmap_unlock(start_pte, ptl);
> > > @@ -688,13 +736,28 @@ static int madvise_free_pte_range(pmd_t *pmd, u=
nsigned long addr,
> > >                         err =3D split_folio(folio);
> > >                         folio_unlock(folio);
> > >                         folio_put(folio);
> > > -                       if (err)
> > > -                               break;
> > > -                       start_pte =3D pte =3D
> > > -                               pte_offset_map_lock(mm, pmd, addr, &p=
tl);
> > > -                       if (!start_pte)
> > > -                               break;
> > > -                       arch_enter_lazy_mmu_mode();
> > > +
> > > +                       /*
> > > +                        * If the large folio is locked or cannot be =
split,
> > > +                        * we just skip it.
> > > +                        */
> > > +                       if (err) {
> > > +skip_large_folio:
> > > +                               if (next_addr >=3D end)
> > > +                                       break;
> > > +                               pte +=3D (next_addr - addr) / PAGE_SI=
ZE;
> > > +                               addr =3D next_addr;
> > > +                       }
> > > +
> > > +                       if (!start_pte) {
> > > +                               start_pte =3D pte =3D pte_offset_map_=
lock(
> > > +                                       mm, pmd, addr, &ptl);
> > > +                               if (!start_pte)
> > > +                                       break;
> > > +                               arch_enter_lazy_mmu_mode();
> > > +                       }
> > > +
> > > +next_folio:
> > >                         pte--;
> > >                         addr -=3D PAGE_SIZE;
> > >                         continue;
> > > --
> > > 2.33.1
> > >

Thanks
Barry