netdev - Re: [RFC PATCH 1/9] virtio_net: disable the hole mechanism for xdp

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: <CACGkMEvLbpNry+ROQof=tPOoX0W3-qths6493uvjBpb0nNinBQ@mail.gmail.com>
Date:   Tue, 6 Dec 2022 13:20:49 +0800
From:   Jason Wang <jasowang@...hat.com>
To:     Heng Qi <hengqi@...ux.alibaba.com>
Cc:     netdev@...r.kernel.org, bpf@...r.kernel.org,
        "Michael S. Tsirkin" <mst@...hat.com>,
        Paolo Abeni <pabeni@...hat.com>,
        Jakub Kicinski <kuba@...nel.org>,
        John Fastabend <john.fastabend@...il.com>,
        "David S. Miller" <davem@...emloft.net>,
        Daniel Borkmann <daniel@...earbox.net>,
        Alexei Starovoitov <ast@...nel.org>,
        Eric Dumazet <edumazet@...gle.com>
Subject: Re: [RFC PATCH 1/9] virtio_net: disable the hole mechanism for xdp

On Tue, Nov 22, 2022 at 3:44 PM Heng Qi <hengqi@...ux.alibaba.com> wrote:
>
> XDP core assumes that the frame_size of xdp_buff and the length of
> the frag are PAGE_SIZE. But before xdp is set, the length of the prefilled
> buffer may exceed PAGE_SIZE, which may cause the processing of xdp to fail,
> so we disable the hole mechanism when xdp is loaded.
>
> Signed-off-by: Heng Qi <hengqi@...ux.alibaba.com>
> Reviewed-by: Xuan Zhuo <xuanzhuo@...ux.alibaba.com>
> ---
>  drivers/net/virtio_net.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index 9cce7dec7366..c5046d21b281 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -1419,8 +1419,11 @@ static int add_recvbuf_mergeable(struct virtnet_info *vi,
>                 /* To avoid internal fragmentation, if there is very likely not
>                  * enough space for another buffer, add the remaining space to
>                  * the current buffer.
> +                * XDP core assumes that frame_size of xdp_buff and the length
> +                * of the frag are PAGE_SIZE, so we disable the hole mechanism.
>                  */
> -               len += hole;
> +               if (!vi->xdp_enabled)

How is this synchronized with virtnet_xdp_set()?

I think we need to use headroom here since it did:

static unsigned int virtnet_get_headroom(struct virtnet_info *vi)
{
        return vi->xdp_enabled ? VIRTIO_XDP_HEADROOM : 0;
}

Otherwise xdp_enabled could be re-read which may lead bugs.

Thanks

> +                       len += hole;
>                 alloc_frag->offset += hole;
>         }
>
> --
> 2.19.1.6.gb485710b
>