[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <7f45aef5-c55b-875f-3458-c34a9b90b564@redhat.com>
Date: Thu, 9 May 2019 11:16:13 +0800
From: Jason Wang <jasowang@...hat.com>
To: David Miller <davem@...emloft.net>
Cc: netdev@...r.kernel.org, mst@...hat.com, yuehaibing@...wei.com,
xiyou.wangcong@...il.com, weiyongjun1@...wei.com,
eric.dumazet@...il.com
Subject: Re: [PATCH net V2] tuntap: synchronize through tfiles array instead
of tun->numqueues
On 2019/5/9 上午1:36, David Miller wrote:
> From: Jason Wang <jasowang@...hat.com>
> Date: Tue, 7 May 2019 00:03:36 -0400
>
>> @@ -1313,6 +1315,10 @@ static int tun_xdp_xmit(struct net_device *dev, int n,
>>
>> tfile = rcu_dereference(tun->tfiles[smp_processor_id() %
>> numqueues]);
>> + if (!tfile) {
>> + rcu_read_unlock();
>> + return -ENXIO; /* Caller will free/return all frames */
>> + }
>>
>> spin_lock(&tfile->tx_ring.producer_lock);
>> for (i = 0; i < n; i++) {
> The only way we can see a NULL here is if a detach happened in parallel,
> and if that happens we should retry the tfile[] indexing after resampling
> numqueues rather than dropping the packet.
Ok, will fix this in V3.
Thanks
Powered by blists - more mailing lists