[ovs-dev] [PATCH v1 05/14] dp-packet: Fix data_len handling multi-seg mbufs.

Eelco Chaudron echaudro at redhat.com
Tue Jul 3 10:22:48 UTC 2018



On 28 Jun 2018, at 17:41, Tiago Lam wrote:

> When a dp_packet is from a DPDK source, and it contains multi-segment
> mbufs, the data_len is not equal to the packet size, pkt_len. Instead,
> the data_len of each mbuf in the chain should be considered while
> distributing the new (provided) size.
>
> To account for the above dp_packet_set_size() has been changed so 
> that,
> in the multi-segment mbufs case, only the data_len on the last mbuf of
> the chain and the total size of the packet, pkt_len, are changed. The
> data_len on the intermediate mbufs preceeding the last mbuf is not
> changed by dp_packet_set_size(). Furthermore, in some cases
> dp_packet_set_size() may be used to set a smaller size than the 
> current
> packet size, thus effectively trimming the end of the packet. In the
> multi-segment mbufs case this may lead to lingering mbufs that may 
> need
> freeing.
>
> __dp_packet_set_data() now also updates an mbufs' data_len after 
> setting
> the data offset. This is so that both fields are always in sync for 
> each
> mbuf in a chain.
>
> Co-authored-by: Michael Qiu <qiudayu at chinac.com>
> Co-authored-by: Mark Kavanagh <mark.b.kavanagh at intel.com>
> Co-authored-by: Przemyslaw Lal <przemyslawx.lal at intel.com>
> Co-authored-by: Marcin Ksiadz <mksiadz at gmail.com>
> Co-authored-by: Yuanhan Liu <yliu at fridaylinux.org>
>
> Signed-off-by: Michael Qiu <qiudayu at chinac.com>
> Signed-off-by: Mark Kavanagh <mark.b.kavanagh at intel.com>
> Signed-off-by: Przemyslaw Lal <przemyslawx.lal at intel.com>
> Signed-off-by: Marcin Ksiadz <mksiadz at gmail.com>
> Signed-off-by: Yuanhan Liu <yliu at fridaylinux.org>
> Signed-off-by: Tiago Lam <tiago.lam at intel.com>
> ---
>  lib/dp-packet.h | 67 
> ++++++++++++++++++++++++++++++++++++++++++++++-----------
>  1 file changed, 55 insertions(+), 12 deletions(-)
>
> diff --git a/lib/dp-packet.h b/lib/dp-packet.h
> index c612ad4..d8cd35c 100644
> --- a/lib/dp-packet.h
> +++ b/lib/dp-packet.h
> @@ -429,17 +429,44 @@ dp_packet_size(const struct dp_packet *b)
>  static inline void
>  dp_packet_set_size(struct dp_packet *b, uint32_t v)
>  {
> -    /* netdev-dpdk does not currently support segmentation; 
> consequently, for
> -     * all intents and purposes, 'data_len' (16 bit) and 'pkt_len' 
> (32 bit) may
> -     * be used interchangably.
> -     *
> -     * On the datapath, it is expected that the size of packets
> -     * (and thus 'v') will always be <= UINT16_MAX; this means that 
> there is no
> -     * loss of accuracy in assigning 'v' to 'data_len'.
> -     */
> -    b->mbuf.data_len = (uint16_t)v;  /* Current seg length. */
> -    b->mbuf.pkt_len = v;             /* Total length of all segments 
> linked to
> -                                      * this segment. */
> +    if (b->source == DPBUF_DPDK) {
> +        struct rte_mbuf *mbuf = &b->mbuf;
> +        struct rte_mbuf *fmbuf = mbuf;
> +        uint16_t new_len = v;
> +        uint16_t data_len;
> +        uint16_t nb_segs = 0;
> +        uint16_t pkt_len = 0;
> +
> +        /* Trim 'v' length bytes from the end of the chained buffers, 
> freeing
> +           any buffers that may be left floating */
> +        while (mbuf) {
> +            data_len = MIN(new_len, mbuf->data_len);
> +            mbuf->data_len = data_len;
> +
> +            if (new_len - data_len <= 0) {
> +                /* Free the rest of chained mbufs */
> +                free_dpdk_buf(CONTAINER_OF(mbuf->next, struct 
> dp_packet,
> +                                           mbuf));
> +                mbuf->next = NULL;
> +            } else if (!mbuf->next) {
> +                /* Don't assign more than what we have available */
> +                mbuf->data_len = MIN(new_len,
> +                                     mbuf->buf_len - mbuf->data_off);
> +            }
> +
> +            new_len -= data_len;
> +            nb_segs += 1;
> +            pkt_len += mbuf->data_len;
> +            mbuf = mbuf->next;
> +        }
> +
> +        fmbuf->nb_segs = nb_segs;
> +        fmbuf->pkt_len = pkt_len;

Guess we do not need the extra variable, we can just do b->mbuf.pkt_len 
etc as below. Looks like a left over from the previous revision?

In the increase packet size case we can end up with an b->mbuf.pkt_len < 
v, maybe we should assert on this?

> +    } else {
> +        b->mbuf.data_len = v;
> +        /* Total length of all segments linked to this segment. */
> +        b->mbuf.pkt_len = v;
> +    }
>  }
>
>  static inline uint16_t
> @@ -451,7 +478,23 @@ __packet_data(const struct dp_packet *b)
>  static inline void
>  __packet_set_data(struct dp_packet *b, uint16_t v)
>  {
> -    b->mbuf.data_off = v;
> +    if (b->source == DPBUF_DPDK) {
> +        uint16_t prev_ofs = b->mbuf.data_off;
> +        b->mbuf.data_off = v;
> +        int16_t ofs_diff = prev_ofs - b->mbuf.data_off;
> +
> +        /* When dealing with DPDK mbufs, keep data_off and data_len 
> in sync.
> +         * Thus, update data_len if the length changes with the move 
> of
> +         * data_off. However, if data_len is 0, there's no data to 
> move and
> +         * data_Len should remain 0. */
> +

What if the data offset moves to the second buffer? This would probably 
never happen as all APIs assume a single address space (and they seem to 
calculate the offset like this), but an assert might help to avoid 
problems in the future.

> +        if (b->mbuf.data_len != 0) {
> +            b->mbuf.data_len = MIN(b->mbuf.data_len + ofs_diff,
> +                                   b->mbuf.buf_len - 
> b->mbuf.data_off);
> +        }
> +    } else {
> +        b->mbuf.data_off = v;
> +    }
>  }
>
>  static inline uint16_t
> -- 
> 2.7.4


More information about the dev mailing list