On Tue, Apr 24, 2018 at 11:10 AM, Jason Wang <jasow...@redhat.com> wrote: > > > On 2018年04月24日 16:44, Magnus Karlsson wrote: >>>> >>>> We have run some benchmarks on a dual socket system with two Broadwell >>>> E5 2660 @ 2.0 GHz with hyperthreading turned off. Each socket has 14 >>>> cores which gives a total of 28, but only two cores are used in these >>>> experiments. One for TR/RX and one for the user space application. The >>>> memory is DDR4 @ 2133 MT/s (1067 MHz) and the size of each DIMM is >>>> 8192MB and with 8 of those DIMMs in the system we have 64 GB of total >>>> memory. The compiler used is gcc version 5.4.0 20160609. The NIC is an >>>> Intel I40E 40Gbit/s using the i40e driver. >>>> >>>> Below are the results in Mpps of the I40E NIC benchmark runs for 64 >>>> and 1500 byte packets, generated by commercial packet generator HW that >>>> is >>>> generating packets at full 40 Gbit/s line rate. >>>> >>>> AF_XDP performance 64 byte packets. Results from RFC V2 in parenthesis. >>>> Benchmark XDP_SKB XDP_DRV >>>> rxdrop 2.9(3.0) 9.4(9.3) >>>> txpush 2.5(2.2) NA* >>>> l2fwd 1.9(1.7) 2.4(2.4) (TX using XDP_SKB in both cases) >>> >>> This number looks not very exciting. I can get ~3Mpps when using testpmd >>> in >>> a guest with xdp_redirect.sh on host between ixgbe and TAP/vhost. I >>> believe >>> we can even better performance without virt. It would be interesting to >>> compare this performance with e.g testpmd + virito_user(vhost_kernel) + >>> XDP. >> >> Note that all the XDP_SKB numbers plus the TX part of XDP_DRV for l2fwd >> uses SKBs and the generic XDP path in the kernel. I am not surprised those >> numbers are lower than what you are seeing with XDP_DRV support. >> (If that is what you are running? Unsure about your setup). > > > Yes, I'm using haswell E5-2630 v3 @ 2.40GHz and ixgbe. > >> The >> 9.4 Mpps for RX is what you get with the XDP_DRV support and copies >> out to user space. Or is it this number you think is low? > > > No rxdrop looks ok. I mean for l2fwd only.
OK, sounds good. l2fwd will get much better once we add XDP_DRV support for TX. Thanks: Magnus >> Zerocopy will be added >> in later patch sets. >> >> With that said, both XDP_SKB and XDP_DRV can be optimized. We >> have not spent that much time on optimizations at this point. >> > > Yes, and it is interesting to compare the performance numbers between AF_XDP > and TAP XDP + vhost_net since their functions are almost equivalent. > > Thanks
