netdev - Re: [PATCH net-next] tun: fix ubuf refcount incorrectly on error path

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: <094f1828-9a73-033e-b1ca-43b73588d22b@redhat.com>
Date:   Fri, 4 Dec 2020 14:10:47 +0800
From:   Jason Wang <jasowang@...hat.com>
To:     wangyunjian <wangyunjian@...wei.com>, mst@...hat.com
Cc:     virtualization@...ts.linux-foundation.org, netdev@...r.kernel.org,
        jerry.lilijun@...wei.com, xudingke@...wei.com
Subject: Re: [PATCH net-next] tun: fix ubuf refcount incorrectly on error path


On 2020/12/3 下午4:00, wangyunjian wrote:
> From: Yunjian Wang <wangyunjian@...wei.com>
>
> After setting callback for ubuf_info of skb, the callback
> (vhost_net_zerocopy_callback) will be called to decrease
> the refcount when freeing skb. But when an exception occurs
> afterwards, the error handling in vhost handle_tx() will
> try to decrease the same refcount again. This is wrong and
> fix this by clearing ubuf_info when meeting errors.
>
> Fixes: 4477138fa0ae ("tun: properly test for IFF_UP")
> Fixes: 90e33d459407 ("tun: enable napi_gro_frags() for TUN/TAP driver")
>
> Signed-off-by: Yunjian Wang <wangyunjian@...wei.com>
> ---
>   drivers/net/tun.c | 11 +++++++++++
>   1 file changed, 11 insertions(+)
>
> diff --git a/drivers/net/tun.c b/drivers/net/tun.c
> index 2dc1988a8973..3614bb1b6d35 100644
> --- a/drivers/net/tun.c
> +++ b/drivers/net/tun.c
> @@ -1861,6 +1861,12 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
>   	if (unlikely(!(tun->dev->flags & IFF_UP))) {
>   		err = -EIO;
>   		rcu_read_unlock();
> +		if (zerocopy) {
> +			skb_shinfo(skb)->destructor_arg = NULL;
> +			skb_shinfo(skb)->tx_flags &= ~SKBTX_DEV_ZEROCOPY;
> +			skb_shinfo(skb)->tx_flags &= ~SKBTX_SHARED_FRAG;
> +		}
> +
>   		goto drop;
>   	}
>   
> @@ -1874,6 +1880,11 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
>   
>   		if (unlikely(headlen > skb_headlen(skb))) {
>   			atomic_long_inc(&tun->dev->rx_dropped);
> +			if (zerocopy) {
> +				skb_shinfo(skb)->destructor_arg = NULL;
> +				skb_shinfo(skb)->tx_flags &= ~SKBTX_DEV_ZEROCOPY;
> +				skb_shinfo(skb)->tx_flags &= ~SKBTX_SHARED_FRAG;
> +			}
>   			napi_free_frags(&tfile->napi);
>   			rcu_read_unlock();
>   			mutex_unlock(&tfile->napi_mutex);


It looks to me then we miss the failure feedback.

The issues comes from the inconsistent error handling in tun.

I wonder whether we can simply do uarg->callback(uarg, false) if 
necessary on every failture path on tun_get_user().

Note that, zerocopy has a lot of issues which makes it not good for 
production environment.

Thanks