On Wed, Oct 02, 2019 at 08:31:59AM +0200, Magnus Karlsson wrote:
> Fixes a crash in poll() when an AF_XDP socket is opened in copy mode
> and the bound device does not have ndo_xsk_wakeup defined. Avoid
> trying to call the non-existing ndo and instead call the internal xsk
> sendmsg function to send packets in the same way (from the
> application's point of view) as calling sendmsg() in any mode or
> poll() in zero-copy mode would have done. The application should
> behave in the same way independent on if zero-copy mode or copy mode
> is used.
> 
> Fixes: 77cd0d7b3f25 ("xsk: add support for need_wakeup flag in AF_XDP rings")
> Reported-by: syzbot+a5765ed8cdb1cca4d...@syzkaller.appspotmail.com
> Signed-off-by: Magnus Karlsson <magnus.karls...@intel.com>

Applied, thanks!

[...]
> +static int xsk_generic_xmit(struct sock *sk)
>  {
> -     u32 max_batch = TX_BATCH_SIZE;
>       struct xdp_sock *xs = xdp_sk(sk);
> +     u32 max_batch = TX_BATCH_SIZE;
>       bool sent_frame = false;
>       struct xdp_desc desc;
>       struct sk_buff *skb;
> @@ -394,6 +392,18 @@ static int xsk_generic_xmit(struct sock *sk, struct 
> msghdr *m,
>       return err;
>  }
>  
> +static int __xsk_sendmsg(struct sock *sk)

Bit unclear why you don't just pass xs directly in here from both call
sites ...

> +{
> +     struct xdp_sock *xs = xdp_sk(sk);
> +
> +     if (unlikely(!(xs->dev->flags & IFF_UP)))
> +             return -ENETDOWN;
> +     if (unlikely(!xs->tx))
> +             return -ENOBUFS;
> +
> +     return xs->zc ? xsk_zc_xmit(xs) : xsk_generic_xmit(sk);

... and for the xsk_generic_xmit() pass in &xs->sk. Presumably generated
code should be the same, but maybe small cleanup for next batch of AF_XDP
patches.

> +}
> +
>  static int xsk_sendmsg(struct socket *sock, struct msghdr *m, size_t 
> total_len)
>  {
>       bool need_wait = !(m->msg_flags & MSG_DONTWAIT);
> @@ -402,21 +412,18 @@ static int xsk_sendmsg(struct socket *sock, struct 
> msghdr *m, size_t total_len)
>  
>       if (unlikely(!xsk_is_bound(xs)))
>               return -ENXIO;
> -     if (unlikely(!(xs->dev->flags & IFF_UP)))
> -             return -ENETDOWN;
> -     if (unlikely(!xs->tx))
> -             return -ENOBUFS;
> -     if (need_wait)
> +     if (unlikely(need_wait))
>               return -EOPNOTSUPP;
>  
> -     return (xs->zc) ? xsk_zc_xmit(sk) : xsk_generic_xmit(sk, m, total_len);
> +     return __xsk_sendmsg(sk);
>  }

Thanks,
Daniel

Reply via email to