> -----Original Message----- > From: Kathleen Capella <kathleen.cape...@arm.com> > Sent: Friday, March 25, 2022 6:12 AM > To: Wu, Jingjing <jingjing...@intel.com>; Xing, Beilei <beilei.x...@intel.com> > Cc: dev@dpdk.org; n...@arm.com; honnappa.nagaraha...@arm.com; > dharmik.thak...@arm.com; Kathleen Capella <kathleen.cape...@arm.com> > Subject: [PATCH] net/iavf: remove extra copy step in Rx bulk path > > In the Rx bulk path, packets which are taken from the HW ring, are first > copied to the stage data structure and then later copied from the stage to the > rx_pkts array. For the number of packets requested immediately by the > receiving function, this two-step process adds extra overhead that is not > necessary. > > Instead, put requested number of packets directly into the rx_pkts array and > only stage excess packets. On N1SDP with 1 core/port, l3fwd saw up to 4% > performance improvement. On x86, no difference in performance was > observed. > > Signed-off-by: Kathleen Capella <kathleen.cape...@arm.com> > Suggested-by: Dharmik Thakkar <dharmik.thak...@arm.com> Acked-by: Qi Zhang <qi.z.zh...@intel.com> Applied to dpdk-next-net-intel. Thanks Qi
