Re: [PATCH v2 8/9] atomic,x86: Alternative atomic_*_overflow() scheme

[Linus Torvalds]

On Fri, Dec 10, 2021 at 8:53 AM Linus Torvalds
<torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
>
> That said - it may not matter - I'm not sure a plain "dec" is even a
> valid operation on a ref in the first place. How could you ever
> validly decrement a ref without checking for it being the last entry?

I should have checked the users - it seems to be a pattern at least in
networking where people have extra references and do

        refcount_dec(&skb->users);
        dev_kfree_skb_any(skb);

because there's no way to tell dev_kfree_skb*() to decrement more than once.

So I guess it's all good, but yes, I still think you can just do "lock
dec .. js" for this operation.

             Linus