RE: [RFC PATCH v2] ethdev: advertise flow restore in mbuf

2023-06-20 Thread Ali Alnubani
> -Original Message- > From: David Marchand > Sent: Thursday, June 15, 2023 4:47 PM > To: dev@dpdk.org > Cc: NBU-Contact-Thomas Monjalon (EXTERNAL) ; > i.maxim...@ovn.org; Aman Singh ; Yuying > Zhang ; Matan Azrad ; Slava > Ovsiienko ; Ori Kam ; > Suanming Mou ; David Christensen > ; Ruife

RE: [RFC PATCH v2] ethdev: advertise flow restore in mbuf

2023-06-19 Thread Slava Ovsiienko
Hi, My primary concern is vectorized rx_burst performance impact. @Ali Alnubani, could we conduct the performance check, at least with SSE (x86) and Neon (BF2/3) archs? With best regards, Slava > -Original Message- > From: David Marchand > Sent: Thursday, June 15, 2023 4:47 PM > To: de

Re: [RFC PATCH v2] ethdev: advertise flow restore in mbuf

2023-06-19 Thread Andrew Rybchenko
On 6/15/23 16:47, David Marchand 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.