Re:An eBPF Loophole: Using XDP for Egress Traffic
by shivanshvij on Wednesday November 05, 2025 @11:54AM
(#45825086)
No this was multiple connections - and we tried with both `iperf2` and `iperf3`, UDP and TCP traffic. UDP actually does much worse on `iptables` than TCP, and I'm not sure why just yet.
[ Parent
| on HN ]
Re:An eBPF Loophole: Using XDP for Egress Traffic
by cptnntsoobv on Wednesday November 05, 2025 @12:13PM
(#45825286)
For UDP I'd look into GSO/GRO to get an upper bound on what pure kernel can do.
With performance benchmarking, specially in networking there is no end to "oh, but did you think of that?!" :)
Re:An eBPF Loophole: Using XDP for Egress Traffic
by shivanshvij on Wednesday November 05, 2025 @02:03PM
(#45826545)
That's a great point. This was one of my iPerf3 runs at one point:
`iperf3 -c 172.31.45.187 -p 5201 -P 128 -t 5 -b 512M -u -l 1448 --bidir | grep "\[SUM\]\[\(TX\|RX\)-C\].*receiver"`
We're also looking at using packet generators to test raw packet throughput. There's a lot more bottlenecks we can cleanup I'm sure.