Re: [PATCH v13 6/7] KEYS: Replace assoc_array with rhashtable

From: Jarkko Sakkinen

Date: Sat Nov 14 2026 - 15:38:40 EST


On Sat, Nov 14, 2026 at 08:24:27PM +0000, bot+bpf-ci@xxxxxxxxxx wrote:
> > + * The walk is RCU-protected, so the keyring's semaphore is not required. A
> > + * keyring modified concurrently may list a key twice or miss one, but it can
> > + * never have a link freed under the walk.
>
> Could the comment instead say that the caller holds the keyring's semaphore,
> which keeps the list and the link count stable across the walk and the
> length calculation, with RCU only keeping the walk itself memory-safe?

If I did not miss anything, this is the only remark.

Hmm...

I think this is enough prose to capture the preconditions:

/*
* The list and link count are stable across the walk as the caller
* holds keyring->sem.
*/

Br, Jarkko