On Dec 10, 2021, at 1:38 AM, Jakub Kicinski kuba@kernel.org wrote:

[...]
I think we're on the same page, the main problem is I've not seen
anyone use the skbuff_head_cache occupancy as a signal in practice.

I'm adding a bunch of people to the CC list, hopefully someone has
an opinion one way or the other.

It looks like we won't have more opinions on that, unfortunately.

@Jakub - Should I submit it as a PATCH and see if we receive more
feedback there?