On Mon, Feb 24, 2020 at 12:49:41AM +0100, Florian Westphal wrote:
> Silly, since its easy to fetch this via libmnl.
> Unfortunately there is a large number of software that uses the old
> API, so add a helper to return the attribute.
>
> Signed-off-by: Florian Westphal <fw@xxxxxxxxx>
> ---
> fixmanpages.sh | 6 ++--
> .../libnetfilter_queue/libnetfilter_queue.h | 1 +
> src/libnetfilter_queue.c | 31 +++++++++++++++++++
> 3 files changed, 36 insertions(+), 2 deletions(-)
>
> diff --git a/fixmanpages.sh b/fixmanpages.sh
> index 897086bad6df..4d12247d14f6 100755
> --- a/fixmanpages.sh
> +++ b/fixmanpages.sh
> @@ -11,8 +11,10 @@ function main
> add2group nfq_get_nfmark nfq_get_timestamp nfq_get_indev nfq_get_physindev
> add2group nfq_get_outdev nfq_get_physoutdev nfq_get_indev_name
> add2group nfq_get_physindev_name nfq_get_outdev_name
> - add2group nfq_get_physoutdev_name nfq_get_packet_hw nfq_get_uid
> - add2group nfq_get_gid nfq_get_secctx nfq_get_payload
> + add2group nfq_get_physoutdev_name nfq_get_packet_hw
> + add2group nfq_get_skbinfo
> + add2group nfq_get_uid nfq_get_gid
> + add2group nfq_get_secctx nfq_get_payload
> setgroup Queue nfq_fd
> add2group nfq_create_queue nfq_destroy_queue nfq_handle_packet nfq_set_mode
> add2group nfq_set_queue_flags nfq_set_queue_maxlen nfq_set_verdict
> diff --git a/include/libnetfilter_queue/libnetfilter_queue.h b/include/libnetfilter_queue/libnetfilter_queue.h
> index 092c57d07451..46e14e135458 100644
> --- a/include/libnetfilter_queue/libnetfilter_queue.h
> +++ b/include/libnetfilter_queue/libnetfilter_queue.h
> @@ -103,6 +103,7 @@ extern uint32_t nfq_get_indev(struct nfq_data *nfad);
> extern uint32_t nfq_get_physindev(struct nfq_data *nfad);
> extern uint32_t nfq_get_outdev(struct nfq_data *nfad);
> extern uint32_t nfq_get_physoutdev(struct nfq_data *nfad);
> +extern uint32_t nfq_get_skbinfo(struct nfq_data *nfad);
> extern int nfq_get_uid(struct nfq_data *nfad, uint32_t *uid);
> extern int nfq_get_gid(struct nfq_data *nfad, uint32_t *gid);
> extern int nfq_get_secctx(struct nfq_data *nfad, unsigned char **secdata);
> diff --git a/src/libnetfilter_queue.c b/src/libnetfilter_queue.c
> index 3cf9653393e6..f5462a374b80 100644
> --- a/src/libnetfilter_queue.c
> +++ b/src/libnetfilter_queue.c
> @@ -1210,6 +1210,37 @@ struct nfqnl_msg_packet_hw *nfq_get_packet_hw(struct nfq_data *nfad)
> struct nfqnl_msg_packet_hw);
> }
>
> +/**
> + * nfq_get_skbinfo - return the NFQA_SKB_INFO meta information
> + * \param nfad Netlink packet data handle passed to callback function
> + *
> + * This can be used to obtain extra information about a packet by testing
> + * the returned integer for any of the following bit flags:
> + *
> + * - NFQA_SKB_CSUMNOTREADY
> + * packet header checksums will be computed by hardware later on, i.e.
> + * tcp/ip checksums in the packet must not be validated, application
> + * should pretend they are correct.
> + * - NFQA_SKB_GSO
> + * packet is an aggregated super-packet. It exceeds device mtu and will
> + * be (re-)split on transmit by hardware.
> + * - NFQA_SKB_CSUM_NOTVERIFIED
> + * packet checksum was not yet verified by the kernel/hardware, for
> + * example because this is an incoming packet and the NIC does not
> + * perform checksum validation at hardware level.
> + * See nfq_set_queue_flags() documentation for more information.
> + *
> + * \return the skbinfo value
> + */
> +EXPORT_SYMBOL
> +uint32_t nfq_get_skbinfo(struct nfq_data *nfad)
> +{
> + if (!nfnl_attr_present(nfad->data, NFQA_SKB_INFO))
> + return 0;
> +
> + return ntohl(nfnl_get_data(nfad->data, NFQA_SKB_INFO, uint32_t));
> +}
> +
> /**
> * nfq_get_uid - get the UID of the user the packet belongs to
> * \param nfad Netlink packet data handle passed to callback function
> --
> 2.24.1
>
Can I suggest:
> + * example because this is an incoming packet and the NIC does not
> + * perform checksum validation at hardware level.
- > + * See nfq_set_queue_flags() documentation for more information.
> + *
> + * \return the skbinfo value
+ > + * \sa __nfq_set_queue_flags__(3)
> + */
> +EXPORT_SYMBOL
I think this will look better, especially on the man page.
Cheers ... Duncan.
