Re: [PATCH v13 0/7] KEYS: Replace assoc_array with rhashtable

From: Jarkko Sakkinen

Date: Mon Nov 16 2026 - 14:37:36 EST


On Mon, Nov 16, 2026 at 01:07:22PM +0100, Heiko Carstens wrote:
> Hi Jarkko,
>
> On Sat, Nov 14, 2026 at 09:36:24PM +0200, Jarkko Sakkinen wrote:
> > The keyrings subsystem is the sole user of assoc_array. It has over 1700
> > lines of code, no test suite, and a long history of bugs, with LLMs now
> > finding more.
> >
> > This unfortunate outcome has evolved over time by large from the shortage
> > of in-kernel callers, time bandwidth and 3rd party contributors. What the
> > keyrings subsystem needs is just a generic key-value store, which performs
> > well with sparse key sets. Replace assoc_array with an rhashtable-backed
> > keyring index and remove assoc_array from the kernel.
>
> ...
>
> > v13:
> > - s390/cert_store: Hold key reference across key_unlink(), and ignore -ENOENT.
> > - s390/cert_store: Add "s390/cert_store: Read-lock keyring semaphore in
> > invalidate_keyring_keys()".
>
> Actually I had / have a patch series of 15 small patches ready to address
> various bugs in certstore, including yours. I'll send them later - a few
> (or many) collide with your patches. Not sure how we coordinate this.

That is perfect. I don't honestly care about patch authorship in this case
too much. Do whatever you want :-) I'm just happy to get them out of the
way.


Br, Jarkko