[spp] [PATCH 01/11] docs: add support for rte_flow

x-fn-spp-ml at ntt-tx.co.jp x-fn-spp-ml at ntt-tx.co.jp
Wed Feb 26 12:43:25 CET 2020


From: Hideyuki Yamashita<yamashita.hideyuki at ntt-tx.co.jp>

Packet classification and distribution based on destination MAC address
can be done using NIC capability. By doing so, resource consumption of
host can be reduced. VLAN tag manipulation(entag/detag) can be done
using NIC capability in some NICs.
Primary process is enhanced to invoke such hardware capability using
DPDK rte_flow APIs and description is added in design section.

Signed-off-by: Hideyuki Yamashita <yamashita.hideyuki at ntt-tx.co.jp>
Signed-off-by: Naoki Takada <ntakada14 at gmail.com>
---
 docs/guides/design/spp_primary.rst | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/docs/guides/design/spp_primary.rst b/docs/guides/design/spp_primary.rst
index 35e4bbb..9ad3cc8 100644
--- a/docs/guides/design/spp_primary.rst
+++ b/docs/guides/design/spp_primary.rst
@@ -25,6 +25,9 @@ the server process of the sample application.
 It does not work for packet forwaring without in some usecases, but just
 provide rings and memory pools for secondary processes.
 
+Primary process supports ``rte_flow`` of DPDK for hardware offloading.
+Packet distribution based on dst MAC address and/or VLAN ID
+is supported. Entag/detag of VLAN is also supported.
 
 Master and Worker Threads
 -------------------------
-- 
2.17.1



More information about the spp mailing list