Hello, On Tue, Jun 20, 2023 at 6:44 PM Slava Ovsiienko <viachesl...@nvidia.com> wrote: > > As reported by Ilya [1], unconditionally calling > > rte_flow_get_restore_info() impacts an application performance for drivers > > that do not provide this ops. > > It could also impact processing of packets that require no call to > > rte_flow_get_restore_info() at all. > > > > Register a dynamic mbuf flag when an application negotiates tunnel metadata > > delivery (calling rte_eth_rx_metadata_negotiate() with > > RTE_ETH_RX_METADATA_TUNNEL_ID). > > > > Drivers then advertise that metadata can be extracted by setting this > > dynamic > > flag in each mbuf. > > > > The application then calls rte_flow_get_restore_info() only when required. > > > > Link: http://inbox.dpdk.org/dev/5248c2ca-f2a6-3fb0-38b8- > > 7f659bfa4...@ovn.org/ > > Signed-off-by: David Marchand <david.march...@redhat.com> > > We did not notice the degradation, so: > Acked-by: Viacheslav Ovsiienko <viachesl...@nvidia.com>
Cool, thanks for testing! Should I add some Tested-by: tag from Ali? I guess he was the one who did the testing. I'll post a non-RFC patch for the CI to pass on it. -- David Marchand
