mir/ovs
2
0
Fork 0
mirror of https://github.com/openvswitch/ovs synced 2025-08-31 22:35:15 +00:00
Code Issues Releases Wiki Activity

dp-packet: Refactor offloading API.

Browse Source 
1. No reason to have mbuf related APIs in a generic code.
2. Not only RSS/checksums should be invalidated in case of tunnel
   decapsulation or sending to 'ring' ports.

In order to fix two above issues, new function
'dp_packet_reset_offload' introduced. In order to clean up/unify
the code and simplify addition of new offloading features to non-DPDK
version of dp_packet, introduced 'ol_flags' bitmask. Additionally
reduced code complexity in 'dp_packet_clone_with_headroom' by using
already existent generic APIs.

Unfortunately, we still need to have a special case for mbuf
initialization inside 'dp_packet_init__()'.
'dp_packet_init_specific()' introduced for this purpose as a generic
API for initialization of the implementation-specific fields.

Acked-by: Flavio Leitner <fbl@sysclose.org>
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Signed-off-by: Ian Stokes <ian.stokes@intel.com>
This commit is contained in:
Ilya Maximets
2019-02-26 13:38:37 +03:00
committed by Ian Stokes
parent cae643534e
commit a47e2db209
4 changed files with 41 additions and 59 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
lib/netdev-dpdk.c
View File

@@ -3775,11 +3775,11 @@ netdev_dpdk_ring_send(struct netdev *netdev, int qid,
struct dp_packet *packet;
/* When using 'dpdkr' and sending to a DPDK ring, we want to ensure that
* the rss hash field is clear. This is because the same mbuf may be
* the offload fields are clear. This is because the same mbuf may be
* modified by the consumer of the ring and return into the datapath
* without recalculating the RSS hash. */
* without recalculating the RSS hash or revalidating the checksums. */
DP_PACKET_BATCH_FOR_EACH (i, packet, batch) {
dp_packet_mbuf_rss_flag_reset(packet);
dp_packet_reset_offload(packet);
}
netdev_dpdk_send__(dev, qid, batch, concurrent_txq);