netdev - Re: [PATCH net-next 2/3] virtio-net: use per-receive queue page frag alloc for mergeable bufs

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]

Message-ID: <1388123210.12212.44.camel@edumazet-glaptop2.roam.corp.google.com>
Date:	Thu, 26 Dec 2013 21:46:50 -0800
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Jason Wang <jasowang@...hat.com>
Cc:	Michael Dalton <mwdalton@...gle.com>,
	"Michael S. Tsirkin" <mst@...hat.com>,
	"David S. Miller" <davem@...emloft.net>, netdev@...r.kernel.org,
	Eric Dumazet <edumazet@...gle.com>,
	Rusty Russell <rusty@...tcorp.com.au>,
	lf-virt <virtualization@...ts.linux-foundation.org>
Subject: Re: [PATCH net-next 2/3] virtio-net: use per-receive queue page
 frag alloc for mergeable bufs

On Fri, 2013-12-27 at 12:55 +0800, Jason Wang wrote:
> On 12/27/2013 05:56 AM, Eric Dumazet wrote:
> > On Thu, 2013-12-26 at 13:28 -0800, Michael Dalton wrote:
> >> On Mon, Dec 23, 2013 at 11:37 AM, Michael S. Tsirkin <mst@...hat.com> wrote:
> >>> So there isn't a conflict with respect to locking.
> >>>
> >>> Is it problematic to use same page_frag with both GFP_ATOMIC and with
> >>> GFP_KERNEL? If yes why?
> >> I believe it is safe to use the same page_frag and I will send out a
> >> followup patchset using just the per-receive page_frags. For future
> >> consideration, Eric noted that disabling NAPI before GFP_KERNEL
> >> allocs can potentially inhibit virtio-net network processing for some
> >> time (e.g., during a blocking memory allocation or preemption).
> > Yep, using napi_disable() in the refill process looks quite inefficient
> > to me, it not buggy.
> >
> > napi_disable() is a big hammer, while whole idea of having a process to
> > block on GFP_KERNEL allocations is to allow some asynchronous behavior.
> >
> > I have hard time to convince myself virtio_net is safe anyway with this
> > work queue thing.
> >
> > virtnet_open() seems racy for example :
> >
> >         for (i = 0; i < vi->max_queue_pairs; i++) {
> >                 if (i < vi->curr_queue_pairs)
> >                         /* Make sure we have some buffers: if oom use wq. */
> >                         if (!try_fill_recv(&vi->rq[i], GFP_KERNEL))
> >                                 schedule_delayed_work(&vi->refill, 0);
> >                 virtnet_napi_enable(&vi->rq[i]);
> >
> >
> > What if the workqueue is scheduled _before_ the call to virtnet_napi_enable(&vi->rq[i]) ?
> 
> Then napi_disable() in refill_work() will busy wait until napi is
> enabled by virtnet_napi_enable() which looks safe. Looks like the real
> issue is in virtnet_restore() who calls try_fill_recv() in neither napi
> context nor napi disabled context.

I think you don't really get the race.

The issue is the following :

CPU0                                                        CPU1

schedule_delayed_work()
                                          napi_disable(&rq->napi);
                                          try_fill_recv(rq, GFP_KERNEL);

virtnet_napi_enable(&vi->rq[i]);
...
try_fill_recv(rq, GFP_ATOMIC);

                                         napi_enable();// crash on :
                                           BUG_ON(!test_bit(NAPI_STATE_SCHED, &n->state)); 




--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html