Linux-mm Archive on lore.kernel.org
 help / color / Atom feed
From: Jann Horn <jannh@google.com>
To: Michal Hocko <mhocko@kernel.org>
Cc: Minchan Kim <minchan@kernel.org>, Linux-MM <linux-mm@kvack.org>,
	 kernel list <linux-kernel@vger.kernel.org>,
	Daniel Colascione <dancol@google.com>,
	 Dave Hansen <dave.hansen@intel.com>,
	"Joel Fernandes (Google)" <joel@joelfernandes.org>
Subject: Re: interaction of MADV_PAGEOUT with CoW anonymous mappings?
Date: Tue, 10 Mar 2020 20:11:45 +0100
Message-ID: <CAG48ez2pNSKL9ZTH-PQ93+Kc6ObH6Pa1vVg3OS85WT0TB8m3=A@mail.gmail.com> (raw)
In-Reply-To: <20200310184814.GA8447@dhcp22.suse.cz>

On Tue, Mar 10, 2020 at 7:48 PM Michal Hocko <mhocko@kernel.org> wrote:
> On Tue 10-03-20 19:08:28, Jann Horn wrote:
> > Hi!
> >
> > >From looking at the source code, it looks to me as if using
> > MADV_PAGEOUT on a CoW anonymous mapping will page out the page if
> > possible, even if other processes still have the same page mapped. Is
> > that correct?
> >
> > If so, that's probably bad in environments where many processes (with
> > different privileges) are forked from a single zygote process (like
> > Android and Chrome), I think? If you accidentally call it on a CoW
> > anonymous mapping with shared pages, you'll degrade the performance of
> > other processes. And if an attacker does it intentionally, they could
> > use that to aid with exploiting race conditions or weird
> > microarchitectural stuff (e.g. the new https://lviattack.eu/lvi.pdf
> > talks about "the assumption that attackers can provoke page faults or
> > microcode assists for (arbitrary) load operations in the victim
> > domain").
> >
> > Should madvise_cold_or_pageout_pte_range() maybe refuse to operate on
> > pages with mapcount>1, or something like that? Or does it already do
> > that, and I just missed the check?
>
> I have brought up side channel attacks earlier [1] but only in the
> context of shared page cache pages. I didn't really consider shared
> anonymous pages to be a real problem. I was under impression that CoW
> pages shouldn't be a real problem because any security sensible
> applications shouldn't allow untrusted code to be forked and CoW
> anything really important. I believe we have made this assumption
> in other places - IIRC on gup with FOLL_FORCE but I admit I have
> very happily forgot most details.

Android has a "zygote" process that starts up the whole Java
environment with a bunch of libraries before entering into a loop that
fork()s off a child every time the user wants to launch an app. So all
the apps, and even browser renderer processes, on the device share
many CoW VMAs. See
<https://developer.android.com/topic/performance/memory-overview#SharingRAM>.

I think Chrome on Linux desktop systems also forks off renderers from
a common zygote process after initializing libraries and so on. See
<https://chromium.googlesource.com/chromium/src.git/+/master/docs/linux/zygote.md>.
(But they use a relatively strict seccomp sandbox that e.g. doesn't
permit MADV_PAGEOUT.)


  reply index

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-10 18:08 Jann Horn
2020-03-10 18:48 ` Michal Hocko
2020-03-10 19:11   ` Jann Horn [this message]
2020-03-10 21:09     ` Michal Hocko
2020-03-10 22:48       ` Dave Hansen
2020-03-11  8:45         ` Michal Hocko
2020-03-11 22:02           ` Minchan Kim
2020-03-11 23:53           ` Shakeel Butt
2020-03-12  0:18             ` Minchan Kim
2020-03-12  2:03               ` Daniel Colascione
2020-03-12 15:15                 ` Shakeel Butt
2020-03-10 20:19   ` Daniel Colascione
2020-03-10 21:40     ` Jann Horn
2020-03-10 21:52       ` Daniel Colascione
2020-03-10 22:14 ` Minchan Kim
2020-03-12  8:22 ` Michal Hocko
2020-03-12 15:40   ` Vlastimil Babka
2020-03-12 20:16   ` Minchan Kim
2020-03-12 20:26     ` Dave Hansen
2020-03-12 20:41     ` Michal Hocko
2020-03-13  2:08       ` Minchan Kim
2020-03-13  8:05         ` Michal Hocko
2020-03-13 20:59           ` Minchan Kim
2020-03-16  9:20             ` Michal Hocko
2020-03-17  1:43               ` Minchan Kim
2020-03-17  7:12                 ` Michal Hocko
2020-03-17 15:00                   ` Minchan Kim
2020-03-17 15:58                     ` Michal Hocko
2020-03-17 17:20                       ` Minchan Kim
2020-03-12 21:41     ` Dave Hansen
2020-03-13  2:00       ` Minchan Kim
2020-03-13 16:59         ` Dave Hansen
2020-03-13 21:13           ` Minchan Kim
2020-03-12 23:29     ` Jann Horn

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAG48ez2pNSKL9ZTH-PQ93+Kc6ObH6Pa1vVg3OS85WT0TB8m3=A@mail.gmail.com' \
    --to=jannh@google.com \
    --cc=dancol@google.com \
    --cc=dave.hansen@intel.com \
    --cc=joel@joelfernandes.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@kernel.org \
    --cc=minchan@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

Linux-mm Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-mm/0 linux-mm/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-mm linux-mm/ https://lore.kernel.org/linux-mm \
		linux-mm@kvack.org
	public-inbox-index linux-mm

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kvack.linux-mm


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git