> From: Joyce Kong [mailto:joyce.k...@arm.com] > Sent: Tuesday, 17 May 2022 12.51 > > For memif non-zero-copy mode, there is a branch to compare > the mbuf and memif buffer size during memory copying. Add > a fast memory copy path by removing this branch with mbuf > and memif buffer size defined at compile time. And for Tx > fast path, bulk free the mbufs which come from the same > mempool. > > When mbuf == memif buffer size, both Rx/Tx would choose > the fast memcpy path. When mbuf < memif buffer size, the > Rx chooses previous memcpy path while Tx chooses fast > memcpy path. When mbuf > memif buffer size, the Rx chooses > fast memcpy path while Tx chooses previous memcpy path. > > Test with 1p1q on Ampere Altra AArch64 server, > --------------------------------------------------------- > buf size | memif = mbuf | memif < mbuf | memif > mbuf > --------------------------------------------------------- > non-zc gain | 16.95% | 3.28% | 13.29% > --------------------------------------------------------- > zc gain | 19.43% | 4.62% | 18.14% > --------------------------------------------------------- > > Test with 1p1q on Cascade Lake Xeon X86server, > --------------------------------------------------------- > buf size | memif = mbuf | memif < mbuf | memif > mbuf > --------------------------------------------------------- > non-zc gain | 19.97% | 2.35% | 21.43% > --------------------------------------------------------- > zc gain | 14.30% | -1.21% | 11.98% > --------------------------------------------------------- > > Joyce Kong (2): > net/memif: add a Rx fast path > net/memif: add a Tx fast path > > drivers/net/memif/rte_eth_memif.c | 258 ++++++++++++++++++++---------- > 1 file changed, 176 insertions(+), 82 deletions(-) > > -- > 2.25.1 >
Series-Acked-by: Morten Brørup <m...@smartsharesystems.com>
