hyperv/hn: Avoid mbuf cluster allocation on reception path, if the packet is small

[sepherosa_gmail.com]

Summary

This gives me ~200Mbps improvement, when running iperf3 sending test w/ 16 connections.

While I'm here, nuke the unnecessary pkthdr.csum_flags zeroing out.

Sponsored by: Microsoft OSTC

[sepherosa_gmail.com]

If no objection comes, this will be committed this week.

[adrian]

You should explain more why the change improves performance more dramatically when you commit. other than that, it's fine!

[sepherosa_gmail.com]

In D4853#103167, @adrian wrote:

You should explain more why the change improves performance more dramatically when you commit. other than that, it's fine!

Sure, thanks.

[jtl]

sys/dev/hyperv/netvsc/hv_netvsc_drv_freebsd.c

475 ↗

(On Diff #12102)

Is there a reason this is read/write?

1036 ↗

(On Diff #12102)

Do you need an atomic operation here to guarantee accurate statistics? (And, even if you do, is it that important?)