Re: ucom console again [PATCH]

[Emmanuel Dreyfus <manu%netbsd.org@localhost>]

On Sun, Dec 21, 2025 at 09:05:19AM +0000, Nick Hudson wrote:
> I've already told you where the problem is "cn_sending is racey and it's my
> fault."
> 
> Something like making ucom_cnput_bytes  loop and right after it does
> 
>             atomic_store_relaxed(&cn->cn_sending, 0);
> 
> it should re-check the ring for entries and try to set it again, etc...

If I understand you correcty, the problem is a race here (atomic
functions removed for the sake of clarity):

if (cn->cn_prod_out - cn->cn_cons_out == 0) {
	/* cn->cn_prod_out is increased between the test and set */
	cn->cn_sending = 0. 
}

Did I get you right?


-- 
Emmanuel Dreyfus
manu%netbsd.org@localhost