On Fri, 20 Feb 2026 13:37:09 +0100 Maciej Fijalkowski wrote:

quoted

Personal preference perhaps but removing error checking always
gives me pause. Maybe:

	bool frag_fail;

	frag_fail = __xsk_rcv_zc(xs, xskb, len, contd);
	list_for_each...
		...
		frag_fail |= __xsk_rcv_zc(xs, xskb, len, contd);
	DEBUG_NET_WARN_ON_ONCE(frag_fail);  

error checking can be actually skipped as xskq_prod_nb_free() peeked into
xsk rx queue and told us there is enough space for descriptor production.

Understood. I was wondering whether the assert / DEBUG_NET.. may still
be worth keeping but up to you.

Nikhil, I also see you routed the set to 'net' tree, previously xsk core
was handled via bpf/bpf-next.

Dunno... We had mixed results, I think net / net-next is fine for stuff
that's purely packet movement. There is no BPF dependency here..
Unless you have a reason! I'm not feeling strongly. Just the stuff we
were fixing recently made me wonder if this code is getting sufficient
networking eyes..