On Wed, Sep 23, 2026 at 02:11:11PM -0400, Zack Gomez wrote:
__netpoll_send_skb() parks an skb on npinfo->txq whenever the device
cannot take it at once, and once the queue is non-empty every later skb
goes straight there to keep ordering. queue_process() drains it from a
workqueue and, unlike the direct path, never polls the device for
completions: when the ring is stopped it backs off HZ/10. Nothing limits
the queue length.
...
Fixes: b6cd27ed3388 ("netpoll per device txq")
I don't think this should be a fix, honestly. I would prefer to see this
on net-next without the Fixes tag.
Assisted-by: Claude:claude-fable-5-1
Signed-off-by: Zack Gomez <redacted>
Reviewed-by: Breno Leitao <leitao@debian.org>