linux-kernel - Re: [PATCH v2] mm/gup: fix try_grab_compound_head() race with split_huge

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

Date:   Fri, 18 Jun 2021 14:50:00 +0100
From:   Matthew Wilcox <willy@...radead.org>
To:     Jason Gunthorpe <jgg@...pe.ca>
Cc:     Jann Horn <jannh@...gle.com>, John Hubbard <jhubbard@...dia.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Linux-MM <linux-mm@...ck.org>,
        kernel list <linux-kernel@...r.kernel.org>,
        "Kirill A . Shutemov" <kirill@...temov.name>,
        Jan Kara <jack@...e.cz>, stable <stable@...r.kernel.org>
Subject: Re: [PATCH v2] mm/gup: fix try_grab_compound_head() race with
 split_huge_page()

On Fri, Jun 18, 2021 at 10:25:56AM -0300, Jason Gunthorpe wrote:
> On Tue, Jun 15, 2021 at 02:09:38PM +0200, Jann Horn wrote:
> > On Tue, Jun 15, 2021 at 8:37 AM John Hubbard <jhubbard@...dia.com> wrote:
> > > On 6/14/21 6:20 PM, Jann Horn wrote:
> > > > @@ -55,8 +72,23 @@ static inline struct page *try_get_compound_head(struct page *page, int refs)
> > > >       if (WARN_ON_ONCE(page_ref_count(head) < 0))
> > > >               return NULL;
> > > >       if (unlikely(!page_cache_add_speculative(head, refs)))
> > > >               return NULL;
> > > > +
> > > > +     /*
> > > > +      * At this point we have a stable reference to the head page; but it
> > > > +      * could be that between the compound_head() lookup and the refcount
> > > > +      * increment, the compound page was split, in which case we'd end up
> > > > +      * holding a reference on a page that has nothing to do with the page
> > > > +      * we were given anymore.
> > > > +      * So now that the head page is stable, recheck that the pages still
> > > > +      * belong together.
> > > > +      */
> > > > +     if (unlikely(compound_head(page) != head)) {
> > >
> > > I was just wondering about what all could happen here. Such as: page gets split,
> > > reallocated into a different-sized compound page, one that still has page pointing
> > > to head. I think that's OK, because we don't look at or change other huge page
> > > fields.
> > >
> > > But I thought I'd mention the idea in case anyone else has any clever ideas about
> > > how this simple check might be insufficient here. It seems fine to me, but I
> > > routinely lack enough imagination about concurrent operations. :)
> > 
> > Hmmm... I think the scariest aspect here is probably the interaction
> > with concurrent allocation of a compound page on architectures with
> > store-store reordering (like ARM). *If* the page allocator handled
> > compound pages with lockless, non-atomic percpu freelists, I think it
> > might be possible that the zeroing of tail_page->compound_head in
> > put_page() could be reordered after the page has been freed,
> > reallocated and set to refcount 1 again?
> 
> Oh wow, yes, this all looks sketchy! Doing a RCU access to page->head
> is a really challenging thing :\
> 
> On the simplified store side:
> 
>   page->head = my_compound
>   *ptep = page
> 
> There must be some kind of release barrier between those two
> operations or this is all broken.. That definately deserves a comment.

set_compound_head() includes a WRITE_ONCE.  Is that enough, or does it
need an smp_wmb()?

> Ideally we'd use smp_store_release to install the *pte :\
> 
> Assuming we cover the release barrier, I would think the algorithm
> should be broadly:
> 
>  struct page *target_page = READ_ONCE(pte)
>  struct page *target_folio = READ_ONCE(target_page->head)

compound_head() includes a READ_ONCE already.

>  page_cache_add_speculative(target_folio, refs)

That's spelled folio_ref_try_add_rcu() right now.

>  if (target_folio != READ_ONCE(target_page->head) ||
>      target_page != READ_ONCE(pte))
>     goto abort
> 
> Which is what this patch does but I would like to see the
> READ_ONCE's.

... you want them to be uninlined from compound_head(), et al?

> And there possibly should be two try_grab_compound_head()'s since we
> don't need this overhead on the fully locked path, especially the
> double atomic on page_ref_add()

There's only one atomic on page_ref_add().  And you need more of this
overhead on the fully locked path than you realise; the page might be
split without holding the mmap_sem, for example.

> > I think the lockless page cache code also has to deal with somewhat
> > similar ordering concerns when it uses page_cache_get_speculative(),
> > e.g. in mapping_get_entry() - first it looks up a page pointer with
> > xas_load(), and any access to the page later on would be a _dependent
> > load_, but if the page then gets freed, reallocated, and inserted into
> > the page cache again before the refcount increment and the re-check
> > using xas_reload(), then there would be no data dependency from
> > xas_reload() to the following use of the page...
> 
> xas_store() should have the smp_store_release() inside it at least..
> 
> Even so it doesn't seem to do page->head, so this is not quite the
> same thing

The page cache only stores head pages, so it's a little simpler than
lookup from PTE.  I have ideas for making PFN->folio lookup go directly
to the folio without passing through a page on the way, but that's for
much, much later.