Message-ID: <CADvbK_cBBDkGt8XLJo6N5TX2YQATS+udVWm8_=8f96=0B9tnTA@mail.gmail.com>
Date:   Thu, 16 Dec 2021 11:52:56 -0500
From:   Xin Long <lucien.xin@...il.com>
To:     Lee Jones <lee.jones@...aro.org>
Cc:     Jakub Kicinski <kuba@...nel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Vlad Yasevich <vyasevich@...il.com>,
        Neil Horman <nhorman@...driver.com>,
        Marcelo Ricardo Leitner <marcelo.leitner@...il.com>,
        "David S. Miller" <davem@...emloft.net>,
        lksctp developers <linux-sctp@...r.kernel.org>,
        "H.P. Yarroll" <piggy@....org>,
        Karl Knutson <karl@...ena.chicago.il.us>,
        Jon Grimm <jgrimm@...ibm.com>,
        Xingang Guo <xingang.guo@...el.com>,
        Hui Huang <hui.huang@...ia.com>,
        Sridhar Samudrala <sri@...ibm.com>,
        Daisy Chang <daisyc@...ibm.com>,
        Ryan Layer <rmlayer@...ibm.com>,
        Kevin Gao <kevin.gao@...el.com>,
        network dev <netdev@...r.kernel.org>
Subject: Re: [RESEND 2/2] sctp: hold cached endpoints to prevent possible UAF

On Thu, Dec 16, 2021 at 11:39 AM Lee Jones <lee.jones@...aro.org> wrote:
>
> On Thu, 16 Dec 2021, Xin Long wrote:
>
> > On Wed, Dec 15, 2021 at 8:48 PM Jakub Kicinski <kuba@...nel.org> wrote:
> > >
> > > On Tue, 14 Dec 2021 21:57:32 +0000 Lee Jones wrote:
> > > > The cause of the resultant dump_stack() reported below is a
> > > > dereference of a freed pointer to 'struct sctp_endpoint' in
> > > > sctp_sock_dump().
> > > >
> > > > This race condition occurs when a transport is cached into its
> > > > associated hash table followed by an endpoint/sock migration to a new
> > > > association in sctp_assoc_migrate() prior to their subsequent use in
> > > > sctp_diag_dump() which uses sctp_for_each_transport() to walk the hash
> > > > table calling into sctp_sock_dump() where the dereference occurs.
>
> > in sctp_sock_dump():
> >         struct sock *sk = ep->base.sk;
> >         ... <--[1]
> >         lock_sock(sk);
> >
> > Do you mean in [1], the sk is peeled off and gets freed elsewhere?
>
> 'ep' and 'sk' are both switched out for new ones in sctp_sock_migrate().
>
> > if that's true, it's still late to do sock_hold(sk) in your this patch.
>
> No, that's not right.
>
> The schedule happens *inside* the lock_sock() call.
Sorry, I don't follow this.
We can't expect when the schedule happens, why do you think this
can never be scheduled before the lock_sock() call?

If the sock is peeled off or is being freed, we shouldn't dump this sock,
and it's better to skip it.

>
> So if you take the reference before it, you're good.
>
> > I talked with Marcelo about this before, if the possible UAF in [1] exists,
> > the problem also exists in the main RX path sctp_rcv().
> >
> > > >
> > > >   BUG: KASAN: use-after-free in sctp_sock_dump+0xa8/0x438 [sctp_diag]
> > > >   Call trace:
> > > >    dump_backtrace+0x0/0x2dc
> > > >    show_stack+0x20/0x2c
> > > >    dump_stack+0x120/0x144
> > > >    print_address_description+0x80/0x2f4
> > > >    __kasan_report+0x174/0x194
> > > >    kasan_report+0x10/0x18
> > > >    __asan_load8+0x84/0x8c
> > > >    sctp_sock_dump+0xa8/0x438 [sctp_diag]
> > > >    sctp_for_each_transport+0x1e0/0x26c [sctp]
> > > >    sctp_diag_dump+0x180/0x1f0 [sctp_diag]
> > > >    inet_diag_dump+0x12c/0x168
> > > >    netlink_dump+0x24c/0x5b8
> > > >    __netlink_dump_start+0x274/0x2a8
> > > >    inet_diag_handler_cmd+0x224/0x274
> > > >    sock_diag_rcv_msg+0x21c/0x230
> > > >    netlink_rcv_skb+0xe0/0x1bc
> > > >    sock_diag_rcv+0x34/0x48
> > > >    netlink_unicast+0x3b4/0x430
> > > >    netlink_sendmsg+0x4f0/0x574
> > > >    sock_write_iter+0x18c/0x1f0
> > > >    do_iter_readv_writev+0x230/0x2a8
> > > >    do_iter_write+0xc8/0x2b4
> > > >    vfs_writev+0xf8/0x184
> > > >    do_writev+0xb0/0x1a8
> > > >    __arm64_sys_writev+0x4c/0x5c
> > > >    el0_svc_common+0x118/0x250
> > > >    el0_svc_handler+0x3c/0x9c
> > > >    el0_svc+0x8/0xc
> > > >
> > > > To prevent this from happening we need to take a references to the
> > > > to-be-used/dereferenced 'struct sock' and 'struct sctp_endpoint's
> > > > until such a time when we know it can be safely released.
> > > >
> > > > When KASAN is not enabled, a similar, but slightly different NULL
> > > > pointer derefernce crash occurs later along the thread of execution in
> > > > inet_sctp_diag_fill() this time.
> > Are you able to reproduce this issue?
>
> Yes 100% of the time without this patch.
>
> 0% of the time with it applied.
>
> > What I'm thinking is to fix it by freeing sk in call_rcu() by
> > sock_set_flag(sock->sk, SOCK_RCU_FREE),
> > and add rcu_read_lock() in sctp_sock_dump().
> >
> > Thanks.
> >
> > >
> > > Are you able to identify where the bug was introduced? Fixes tag would
> > > be good to have here.
>
> It's probably been there since the code was introduced.
>
> I'll see how far back we have to go.
>
> > > You should squash the two patches together.
>
> I generally like patches to encapsulate functional changes.
>
> This one depends on the other, but they are not functionally related.
>
> You're the boss though - I'll squash them if you insist.
>
> > > > diff --git a/net/sctp/diag.c b/net/sctp/diag.c
> > > > index 760b367644c12..2029b240b6f24 100644
> > > > --- a/net/sctp/diag.c
> > > > +++ b/net/sctp/diag.c
> > > > @@ -301,6 +301,8 @@ static int sctp_sock_dump(struct sctp_transport *tsp, void *p)
> > > >       struct sctp_association *assoc;
> > > >       int err = 0;
> > > >
> > > > +     sctp_endpoint_hold(ep);
> > > > +     sock_hold(sk);
> > > >       lock_sock(sk);
> > > >       list_for_each_entry(assoc, &ep->asocs, asocs) {
> > > >               if (cb->args[4] < cb->args[1])
> > > > @@ -341,6 +343,8 @@ static int sctp_sock_dump(struct sctp_transport *tsp, void *p)
> > > >       cb->args[4] = 0;
> > > >  release:
> > > >       release_sock(sk);
> > > > +     sock_put(sk);
> > > > +     sctp_endpoint_put(ep);
> > > >       return err;
> > > >  }
> > > >
> > >
>
> --
> Lee Jones [李琼斯]
> Senior Technical Lead - Developer Services
> Linaro.org │ Open source software for Arm SoCs
> Follow Linaro: Facebook | Twitter | Blog

Powered by blists - more mailing lists