Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Wed, 23 Sep 2020 09:01:38 -0400
From:   Rafael Aquini <aquini@redhat.com>
To:     "Huang, Ying" <ying.huang@intel.com>
Cc:     linux-mm@kvack.org, linux-kernel@vger.kernel.org,
        akpm@linux-foundation.org
Subject: Re: [PATCH] mm: swapfile: avoid split_swap_cluster() NULL pointer
 dereference
Message-ID: <20200923130138.GM795820@optiplex-lnx>
References: <20200922184838.978540-1-aquini@redhat.com>
 <878sd1qllb.fsf@yhuang-dev.intel.com>
 <20200923043459.GL795820@optiplex-lnx>
 <87sgb9oz1u.fsf@yhuang-dev.intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <87sgb9oz1u.fsf@yhuang-dev.intel.com>
Precedence: bulk

On Wed, Sep 23, 2020 at 01:13:49PM +0800, Huang, Ying wrote:
> Rafael Aquini <aquini@redhat.com> writes:
> 
> > On Wed, Sep 23, 2020 at 10:21:36AM +0800, Huang, Ying wrote:
> >> Hi, Rafael,
> >> 
> >> Rafael Aquini <aquini@redhat.com> writes:
> >> 
> >> > The swap area descriptor only gets struct swap_cluster_info *cluster_info
> >> > allocated if the swapfile is backed by non-rotational storage.
> >> > When the swap area is laid on top of ordinary disk spindles, lock_cluster()
> >> > will naturally return NULL.
> >> 
> >> Thanks for reporting.  But the bug looks strange.  Because in a system
> >> with only HDD swap devices, during THP swap out, the swap cluster
> >> shouldn't be allocated, as in
> >> 
> >> shrink_page_list()
> >>   add_to_swap()
> >>     get_swap_page()
> >>       get_swap_pages()
> >>         swap_alloc_cluster()
> >>
> >
> > The underlying problem is that swap_info_struct.cluster_info is always NULL 
> > on the rotational storage case.
> 
> Yes.
> 
> > So, it's very easy to follow that constructions 
> > like this one, in split_swap_cluster 
> >
> > ...
> >         ci = lock_cluster(si, offset);
> >         cluster_clear_huge(ci);
> > ...
> >
> > will go for a NULL pointer dereference, in that case, given that lock_cluster 
> > reads:
> >
> > ...
> > 	struct swap_cluster_info *ci;
> >         ci = si->cluster_info;
> >         if (ci) {
> >                 ci += offset / SWAPFILE_CLUSTER;
> >                 spin_lock(&ci->lock);
> >         }
> >         return ci;
> > ...
> 
> But on HDD, we shouldn't call split_swap_cluster() at all, because we
> will not allocate swap cluster firstly.  So, if we run into this,
> there should be some other bug, we need to figure it out.
>

split_swap_cluster() gets called by split_huge_page_to_list(),
if the page happens to be in the swapcache, and it will always
go that way, regardless the backing storage type:

...
            __split_huge_page(page, list, end, flags);
            if (PageSwapCache(head)) {
                    swp_entry_t entry = { .val = page_private(head) };

                    ret = split_swap_cluster(entry);
            } else
                    ret = 0;
...

The problem is not about allocating the swap_cluster -- it's obviously
not allocated in these cases. The problem is that on rotational
storage you don't even have the base structure that allows you to
keep the swap clusters (cluster_info does not get allocated, at all,
so si->cluster_info is always NULL)

You can argue about other bugs all you want, it doesn't change
the fact that this code is incomplete as it sits, because it 
misses checking for a real case where lock_cluster() will return NULL.