linux-kernel - Re: [PATCH v2 2/3] mm: use folio_add_new_anon_rmap() if folio_test

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date: Tue, 25 Jun 2024 11:42:53 +1200
From: Barry Song <21cnbao@...il.com>
To: akpm@...ux-foundation.org
Cc: baolin.wang@...ux.alibaba.com,
	chrisl@...nel.org,
	david@...hat.com,
	linux-kernel@...r.kernel.org,
	linux-mm@...ck.org,
	mhocko@...e.com,
	ryan.roberts@....com,
	shy828301@...il.com,
	surenb@...gle.com,
	v-songbaohua@...o.com,
	willy@...radead.org,
	ying.huang@...el.com,
	yosryahmed@...gle.com,
	yuanshuai@...o.com,
	yuzhao@...gle.com
Subject: Re: [PATCH v2 2/3] mm: use folio_add_new_anon_rmap() if folio_test_anon(folio)==false

On Tue, Jun 25, 2024 at 11:25 AM Andrew Morton <akpm@...ux-foundation.org> wrote:
>
> On Sat, 22 Jun 2024 15:20:02 +1200 Barry Song <21cnbao@...il.com> wrote:
>
> > > >
> > > > Since this is primarily a documentation update, I'll wait for two or
> > > > three days to see if
> > > > there are any more Linux-next reports before sending v3 combining these fixes
> > > > together(I've already fixed another doc warn reported by lkp) and seek Andrew's
> > > > assistance to drop v2 and apply v3.
> > >
> > > Feel free to send fixup patches for such small stuff (for example, as
> > > reply to this mail). Usually, no need for a new series. Andrew will
> > > squash all fixups before merging it to mm-stable.
> >
> > Hi Andrew,
> >
> > Can you please squash this change(another one suggested by David)?
>
> sure, but...
>
> > From: Barry Song <v-songbaohua@...o.com>
> > Date: Sat, 22 Jun 2024 15:14:53 +1200
> > Subject: [PATCH] enhance doc- mm: use folio_add_new_anon_rmap() if
> >  folio_test_anon(folio)==false
>
> The only description we have here is "enhance doc"
>
> > --- a/mm/memory.c
> > +++ b/mm/memory.c
> > @@ -4346,6 +4346,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
> >                * here, we have to be careful.
> >                */
> >               VM_WARN_ON_ONCE(folio_test_large(folio));
> > +             VM_WARN_ON_FOLIO(!folio_test_locked(folio), folio);
>
> And these aren't documentation changes.  Please send along a small
> changelog for this patch.

Thanks for the suggestion.
Could we have this in changelog?

For new anon(!anon), there's a possibility that multiple concurrent threads
might execute "if (!anon) folio_add_new_anon_rmap()" in parallel. In such
cases, the threads should lock the folio before executing this sequence.
We use VM_WARN_ON_FOLIO() to verify if this condition holds true.

>

Thanks
Barry