Hi, Olivier, Thanks for advice. I will modify codes according to your comments. As for the packet header, I define a simplified GTP header in patch: app/testpmd: enable GTP header parse and Tx checksum offload, in order to help forward GTP packet in testpmd. I just moved some codes to rte_ether.h according to Ferruh's comments in that patch, which may need your review as well.
Thanks. -----Original Message----- From: Olivier Matz [mailto:[email protected]] Sent: Friday, October 18, 2019 4:38 PM To: Xu, Ting <[email protected]> Cc: [email protected]; Lu, Wenzhuo <[email protected]>; Wu, Jingjing <[email protected]>; Ye, Xiaolong <[email protected]>; Yang, Qiming <[email protected]>; Mcnamara, John <[email protected]>; Kovacevic, Marko <[email protected]>; Zhang, Qi Z <[email protected]> Subject: Re: [PATCH v1 1/2] lib/mbuf: add GTP tunnel type flag Hi, On Fri, Sep 20, 2019 at 08:36:23AM +0000, Ting Xu wrote: > Add GTP tunnel type flag in mbuf for future use in GTP checksum > offload. > > Signed-off-by: Ting Xu <[email protected]> > --- > lib/librte_mbuf/rte_mbuf.h | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h > index 98225ec80..0a5d67d9a 100644 > --- a/lib/librte_mbuf/rte_mbuf.h > +++ b/lib/librte_mbuf/rte_mbuf.h > @@ -251,6 +251,8 @@ extern "C" { > /** TX packet with MPLS-in-UDP RFC 7510 header. */ #define > PKT_TX_TUNNEL_MPLSINUDP (0x5ULL << 45) #define > PKT_TX_TUNNEL_VXLAN_GPE (0x6ULL << 45) > +#define PKT_TX_TUNNEL_GTP (0x7ULL << 45) > + > /** > * Generic IP encapsulated tunnel type, used for TSO and checksum offload. > * It can be used for tunnels which are not standards or listed above. Some updates are missing, for instance in rte_get_tx_ol_flag_name(). Please look what was done for other flags. Please, can you describe in more details what is expected in the packet header? Thanks, Olivier
