LKML Archive on
help / color / mirror / Atom feed
From: Peter Xu <>
To: David Hildenbrand <>
Cc: Tiberiu A Georgescu <>,,,,,,,,,,,,,,,,,,,,,
Subject: Re: [PATCH 0/1] pagemap: swap location for shared pages
Date: Wed, 11 Aug 2021 14:25:52 -0400	[thread overview]
Message-ID: <YRQWMIBwkdBK12Z3@t490s> (raw)
In-Reply-To: <>

On Wed, Aug 11, 2021 at 06:15:37PM +0200, David Hildenbrand wrote:
> On 04.08.21 21:17, Peter Xu wrote:
> > On Wed, Aug 04, 2021 at 08:49:14PM +0200, David Hildenbrand wrote:
> > > TBH, I tend to really dislike the PTE marker idea. IMHO, we shouldn't store
> > > any state information regarding shared memory in per-process page tables: it
> > > just doesn't make too much sense.
> > > 
> > > And this is similar to SOFTDIRTY or UFFD_WP bits: this information actually
> > > belongs to the shared file ("did *someone* write to this page", "is
> > > *someone* interested into changes to that page", "is there something"). I
> > > know, that screams for a completely different design in respect to these
> > > features.
> > > 
> > > I guess we start learning the hard way that shared memory is just different
> > > and requires different interfaces than per-process page table interfaces we
> > > have (pagemap, userfaultfd).
> > > 
> > > I didn't have time to explore any alternatives yet, but I wonder if tracking
> > > such stuff per an actual fd/memfd and not via process page tables is
> > > actually the right and clean approach. There are certainly many issues to
> > > solve, but conceptually to me it feels more natural to have these shared
> > > memory features not mangled into process page tables.
> > 
> > Yes, we can explore all the possibilities, I'm totally fine with it.
> > 
> > I just want to say I still don't think when there's page cache then we must put
> > all the page-relevant things into the page cache.
> [sorry for the late reply]
> Right, but for the case of shared, swapped out pages, the information is
> already there, in the page cache :)
> > 
> > They're shared by processes, but process can still have its own way to describe
> > the relationship to that page in the cache, to me it's as simple as "we allow
> > process A to write to page cache P", while "we don't allow process B to write
> > to the same page" like the write bit.
> The issue I'm having uffd-wp as it was proposed for shared memory is that
> there is hardly a sane use case where we would *want* it to work that way.
> A UFFD-WP flag in a page table for shared memory means "please notify once
> this process modifies the shared memory (via page tables, not via any other
> fd modification)". Do we have an example application where these semantics
> makes sense and don't over-complicate the whole approach? I don't know any,
> thus I'm asking dumb questions :)
> For background snapshots in QEMU the flow would currently be like this,
> assuming all processes have the shared guest memory mapped.
> 1. Background snapshot preparation: QEMU requests all processes
>    to uffd-wp the range
> a) All processes register a uffd handler on guest RAM

To be explicit: not a handler; just register with uffd-wp and pass over the fd
to the main process.

> b) All processes fault in all guest memory (essentially populating all
>    memory): with a uffd-WP extensions we might be able to get rid of
>    that, I remember you were working on that.
> c) All processes uffd-WP the range to set the bit in their page table
> 2. Background snapshot runs:
> a) A process either receives a UFFD-WP event and forwards it to QEMU or
>    QEMU polls all other processes for UFFD events.
> b) QEMU writes the to-be-changed page to the migration stream.
> c) QEMU triggers all processes to un-protect the page and wake up any
>    waiters. All processes clear the uffd-WP bit in their page tables.
> 3. Background snapshot completes:
> a) All processes unregister the uffd handler
> Now imagine something like this:
> 1. Background snapshot preparation:
> a) QEMU registers a UFFD-WP handler on a *memfd file* that corresponds
>    to guest memory.
> b) QEMU uffd-wp's the whole file
> 2. Background snapshot runs:
> a) QEMU receives a UFFD-WP event.
> b) QEMU writes the to-be-changed page to the migration stream.
> c) QEMU un-protect the page and wake up any waiters.
> 3. Background snapshot completes:
> a) QEMU unregister the uffd handler
> Wouldn't that be much nicer and much easier to handle? Yes, it is much
> harder to implement because such an infrastructure does not exist yet, and
> it most probably wouldn't be called uffd anymore, because we are dealing
> with file access. But this way, it would actually be super easy to use the
> feature across multiple processes and eventually to even catch other file
> modifications.

I can totally understand how you see this.  We've discussed about that, isn't
it? About the ideal worlds. :)

It would be great if this can work out, I hope so.  So far I'm not that
ambicious, and as I said, I don't know whether there will be other concerns
when it goes into the page cache layer, and when it's a behavior of multiple
processes where one of them can rule others without others being notice of it.

Even if we want to go that way, I think we should first come up with some way
to describe the domains that one uffd-wp registered file should behave upon.
It shouldn't be "any process touching this file".

One quick example in my mind is when a malicious process wants to stop another
daemon process, it'll be easier as long as the malicious process can delete a
file that the daemon used to read/write, replace it with a shmem with uffd-wp
registered (or maybe just a regular file on file systems, if your proposal will
naturally work on them).  The problem is, is it really "legal" to be able to
stop the daemon running like that?

I also don't know the initial concept when uffd is designed and why it's
designed at pte level.  Avoid vma manipulation should be a major factor, but I
can't say I understand all of them.  Not sure whether Andrea has any input here.

That's why I think current uffd can still make sense with per-process concepts
and keep it that way.  When register uffd-wp yes we need to do that for
multiple processes, but it also means each process is fully aware that this is
happening so it's kind of verified that this is wanted behavior for that
process.  It'll happen with less "surprises", and smells safer.

I don't think that will not work out.  It may require all the process to
support uffd-wp apis and cooperate, but that's so far how it should work for me
in a safe and self-contained way.  Say, every process should be aware of what's
going to happen on blocked page faults.

> Again, I am not sure if uffd-wp or softdirty make too much sense in general
> when applied to shmem. But I'm happy to learn more.

Me too, I'm more than glad to know whether the page cache idea could be
welcomed or am I just wrong about it.  Before I understand more things around
this, so far I still think the per-process based and fd-based solution of uffd
still makes sense.


Peter Xu

  parent reply	other threads:[~2021-08-11 18:25 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-30 16:08 Tiberiu A Georgescu
2021-07-30 16:08 ` [PATCH 1/1] pagemap: report " Tiberiu A Georgescu
2021-07-30 17:28 ` [PATCH 0/1] pagemap: " Eric W. Biederman
2021-08-02 12:20   ` Tiberiu Georgescu
2021-08-04 18:33 ` Peter Xu
2021-08-04 18:49   ` David Hildenbrand
2021-08-04 19:17     ` Peter Xu
2021-08-11 16:15       ` David Hildenbrand
2021-08-11 16:17         ` David Hildenbrand
2021-08-11 18:25         ` Peter Xu [this message]
2021-08-11 18:41           ` David Hildenbrand
2021-08-11 19:54             ` Peter Xu
2021-08-11 20:13               ` David Hildenbrand

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:

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

  git send-email \
    --in-reply-to=YRQWMIBwkdBK12Z3@t490s \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \
    --subject='Re: [PATCH 0/1] pagemap: swap location for shared pages' \

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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).