[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id: <20190725.113915.1995894376660846014.davem@davemloft.net>
Date: Thu, 25 Jul 2019 11:39:15 -0700 (PDT)
From: David Miller <davem@...emloft.net>
To: abauvin@...leway.com
Cc: stephen@...workplumber.org, jasowang@...hat.com,
netdev@...r.kernel.org
Subject: Re: [PATCH v2] tun: mark small packets as owned by the tap sock
From: Alexis Bauvin <abauvin@...leway.com>
Date: Tue, 23 Jul 2019 16:23:01 +0200
> - v1 -> v2: Move skb_set_owner_w to __tun_build_skb to reduce patch size
>
> Small packets going out of a tap device go through an optimized code
> path that uses build_skb() rather than sock_alloc_send_pskb(). The
> latter calls skb_set_owner_w(), but the small packet code path does not.
>
> The net effect is that small packets are not owned by the userland
> application's socket (e.g. QEMU), while large packets are.
> This can be seen with a TCP session, where packets are not owned when
> the window size is small enough (around PAGE_SIZE), while they are once
> the window grows (note that this requires the host to support virtio
> tso for the guest to offload segmentation).
> All this leads to inconsistent behaviour in the kernel, especially on
> netfilter modules that uses sk->socket (e.g. xt_owner).
>
> Signed-off-by: Alexis Bauvin <abauvin@...leway.com>
> Fixes: 66ccbc9c87c2 ("tap: use build_skb() for small packet")
Applied and queued up for -stable, thanks.
Powered by blists - more mailing lists