On Sat, Jan 14, 2023 at 1:18 PM Pietro Borrello
[off-list ref] wrote:

On Fri, 13 Jan 2023 at 13:16, Eric Dumazet [off-list ref] wrote:

quoted

1) Given this path was never really used, we have no coverage.

2) Given that we do not check inet_ehash_nolisten() return code here.

It seems there are a bunch of call sites where inet_ehash_nolisten() return
code is not checked, thus I didn't think of it to be a problem.

quoted

I would recommend _not_ adding the Fixes: tag, and target net-next tree

In fact, I would remove this dead code, and reduce complexity.

This makes a lot of sense. I can post a v3 patch completely removing
the fast path.

However, this patch's v1 was already reviewed by
Kuniyuki Iwashima [off-list ref], v2 is a nit, if posting a v3
I think I should remove the Reviewed-by: since it would completely
change the patch, but what is the preferred fix?

Yes, remove it, and Kuniyuki will review it again, thanks.