Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1698449

Re: [PATCH net-next 3/3] tap: XDP support

From Jakub Kicinski <kubakici@wp.pl>
Newsgroups linux.kernel
Subject Re: [PATCH net-next 3/3] tap: XDP support
Date 2017-07-28 06:00 +0200
Message-ID <u84Br-6TA-7@gated-at.bofh.it> (permalink)
References <u7Nhf-4j3-1@gated-at.bofh.it> <u7Nhg-4j3-15@gated-at.bofh.it> <u848p-6HH-7@gated-at.bofh.it> <u848p-6HH-9@gated-at.bofh.it> <u84rL-6Qi-3@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


On Fri, 28 Jul 2017 06:46:40 +0300, Michael S. Tsirkin wrote:
> On Fri, Jul 28, 2017 at 11:28:54AM +0800, Jason Wang wrote:
> > > > +	old_prog = rtnl_dereference(tun->xdp_prog);
> > > > +	if (old_prog)
> > > > +		bpf_prog_put(old_prog);
> > > > +	rcu_assign_pointer(tun->xdp_prog, prog);  
> > > Is this OK?  Could this lead to the program getting freed and then
> > > datapath accessing a stale pointer?  I mean in the scenario where the
> > > process gets pre-empted between the bpf_prog_put() and
> > > rcu_assign_pointer()?  
> > 
> > Will call bpf_prog_put() after rcu_assign_pointer().  
> 
> I suspect you need to sync RCU or something before that.

I think the bpf_prog_put() will use call_rcu() to do the actual free:

static void __bpf_prog_put(struct bpf_prog *prog, bool do_idr_lock)
{
	if (atomic_dec_and_test(&prog->aux->refcnt)) {
		trace_bpf_prog_put_rcu(prog);
		/* bpf_prog_free_id() must be called first */
		bpf_prog_free_id(prog, do_idr_lock);
		bpf_prog_kallsyms_del(prog);
		call_rcu(&prog->aux->rcu, __bpf_prog_put_rcu);
	}
}

It's just that we are only under the rtnl here, RCU lock is not held, so
grace period may elapse between bpf_prog_put() and rcu_assign_pointer().

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH net-next 0/3] XDP support for tap Jason Wang <jasowang@redhat.com> - 2017-07-27 11:30 +0200
  [PATCH net-next 1/3] tap: use build_skb() for small packet Jason Wang <jasowang@redhat.com> - 2017-07-27 11:30 +0200
  [PATCH net-next 2/3] net: export some generic xdp helpers Jason Wang <jasowang@redhat.com> - 2017-07-27 11:30 +0200
  [PATCH net-next 3/3] tap: XDP support Jason Wang <jasowang@redhat.com> - 2017-07-27 11:30 +0200
    Re: [PATCH net-next 3/3] tap: XDP support Jason Wang <jasowang@redhat.com> - 2017-07-28 05:30 +0200
      Re: [PATCH net-next 3/3] tap: XDP support "Michael S. Tsirkin" <mst@redhat.com> - 2017-07-28 05:50 +0200
        Re: [PATCH net-next 3/3] tap: XDP support Jakub Kicinski <kubakici@wp.pl> - 2017-07-28 06:00 +0200
        Re: [PATCH net-next 3/3] tap: XDP support Jason Wang <jasowang@redhat.com> - 2017-07-28 06:00 +0200
          Re: [PATCH net-next 3/3] tap: XDP support "Michael S. Tsirkin" <mst@redhat.com> - 2017-07-28 06:10 +0200
        Re: [PATCH net-next 3/3] tap: XDP support Daniel Borkmann <daniel@iogearbox.net> - 2017-07-28 17:20 +0200
    Re: [PATCH net-next 3/3] tap: XDP support Jakub Kicinski <kubakici@wp.pl> - 2017-07-28 05:30 +0200

csiph-web