netdev - Re: [PATCH net-next 21/33] virtio_net: add XDP frame size in two code paths

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: <d939445b-9713-2b2f-9830-38c2867bb963@redhat.com>
Date:   Tue, 28 Apr 2020 17:50:11 +0800
From:   Jason Wang <jasowang@...hat.com>
To:     Jesper Dangaard Brouer <brouer@...hat.com>
Cc:     sameehj@...zon.com, netdev@...r.kernel.org, bpf@...r.kernel.org,
        zorik@...zon.com, akiyano@...zon.com, gtzalik@...zon.com,
        Toke Høiland-Jørgensen <toke@...hat.com>,
        Daniel Borkmann <borkmann@...earbox.net>,
        Alexei Starovoitov <alexei.starovoitov@...il.com>,
        John Fastabend <john.fastabend@...il.com>,
        Alexander Duyck <alexander.duyck@...il.com>,
        Jeff Kirsher <jeffrey.t.kirsher@...el.com>,
        David Ahern <dsahern@...il.com>,
        Willem de Bruijn <willemdebruijn.kernel@...il.com>,
        Ilias Apalodimas <ilias.apalodimas@...aro.org>,
        Lorenzo Bianconi <lorenzo@...nel.org>,
        Saeed Mahameed <saeedm@...lanox.com>,
        steffen.klassert@...unet.com
Subject: Re: [PATCH net-next 21/33] virtio_net: add XDP frame size in two code
 paths


On 2020/4/27 下午10:32, Jesper Dangaard Brouer wrote:
> On Mon, 27 Apr 2020 15:21:02 +0800
> Jason Wang<jasowang@...hat.com>  wrote:
>
>> On 2020/4/23 上午12:09, Jesper Dangaard Brouer wrote:
>>> The virtio_net driver is running inside the guest-OS. There are two
>>> XDP receive code-paths in virtio_net, namely receive_small() and
>>> receive_mergeable(). The receive_big() function does not support XDP.
>>>
>>> In receive_small() the frame size is available in buflen. The buffer
>>> backing these frames are allocated in add_recvbuf_small() with same
>>> size, except for the headroom, but tailroom have reserved room for
>>> skb_shared_info. The headroom is encoded in ctx pointer as a value.
>>>
>>> In receive_mergeable() the frame size is more dynamic. There are two
>>> basic cases: (1) buffer size is based on a exponentially weighted
>>> moving average (see DECLARE_EWMA) of packet length. Or (2) in case
>>> virtnet_get_headroom() have any headroom then buffer size is
>>> PAGE_SIZE. The ctx pointer is this time used for encoding two values;
>>> the buffer len "truesize" and headroom. In case (1) if the rx buffer
>>> size is underestimated, the packet will have been split over more
>>> buffers (num_buf info in virtio_net_hdr_mrg_rxbuf placed in top of
>>> buffer area). If that happens the XDP path does a xdp_linearize_page
>>> operation.
>>>
>>> Cc: Jason Wang<jasowang@...hat.com>
>>> Signed-off-by: Jesper Dangaard Brouer<brouer@...hat.com>
>>> ---
>>>    drivers/net/virtio_net.c |   15 ++++++++++++---
>>>    1 file changed, 12 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
>>> index 11f722460513..1df3676da185 100644
>>> --- a/drivers/net/virtio_net.c
>>> +++ b/drivers/net/virtio_net.c
>>> @@ -689,6 +689,7 @@ static struct sk_buff *receive_small(struct net_device *dev,
>>>    		xdp.data_end = xdp.data + len;
>>>    		xdp.data_meta = xdp.data;
>>>    		xdp.rxq = &rq->xdp_rxq;
>>> +		xdp.frame_sz = buflen;
>>>    		orig_data = xdp.data;
>>>    		act = bpf_prog_run_xdp(xdp_prog, &xdp);
>>>    		stats->xdp_packets++;
>>> @@ -797,10 +798,11 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
>>>    	int offset = buf - page_address(page);
>>>    	struct sk_buff *head_skb, *curr_skb;
>>>    	struct bpf_prog *xdp_prog;
>>> -	unsigned int truesize;
>>> +	unsigned int truesize = mergeable_ctx_to_truesize(ctx);
>>>    	unsigned int headroom = mergeable_ctx_to_headroom(ctx);
>>> -	int err;
>>>    	unsigned int metasize = 0;
>>> +	unsigned int frame_sz;
>>> +	int err;
>>>    
>>>    	head_skb = NULL;
>>>    	stats->bytes += len - vi->hdr_len;
>>> @@ -821,6 +823,11 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
>>>    		if (unlikely(hdr->hdr.gso_type))
>>>    			goto err_xdp;
>>>    
>>> +		/* Buffers with headroom use PAGE_SIZE as alloc size,
>>> +		 * see add_recvbuf_mergeable() + get_mergeable_buf_len()
>>> +		 */
>>> +		frame_sz = headroom ? PAGE_SIZE : truesize;
>>> +
>>>    		/* This happens when rx buffer size is underestimated
>>>    		 * or headroom is not enough because of the buffer
>>>    		 * was refilled before XDP is set. This should only
>>> @@ -834,6 +841,8 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
>>>    						      page, offset,
>>>    						      VIRTIO_XDP_HEADROOM,
>>>    						      &len);
>>> +			frame_sz = PAGE_SIZE;
>> Should this be PAGE_SIZE -  SKB_DATA_ALIGN(sizeof(struct skb_shared_info))?
> No, frame_sz include the SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) length.


Ok, consider mergeable buffer path depends on the truesize which is 
encoded in ctx.

It looks to the the calculation in add_recvfbuf_mergeable() is wrong, we 
need count both headroom and tailroom there.

We probably need the attached 2 patches to fix this.

(untested, will test it tomorrow).

Thanks



View attachment "0002-virtio-net-fix-the-XDP-truesize-calculation-for-merg.patch" of type "text/x-patch" (1826 bytes)

View attachment "0001-virtio-net-don-t-reserve-space-for-vnet-header-for-X.patch" of type "text/x-patch" (1740 bytes)