Re: [RFC PATCH v3 00/11] KEYS: Sanitize the garbage collector

From: Jarkko Sakkinen <jarkko@kernel.org>
To: David Howells <dhowells@redhat.com>
Cc: keyrings@vger.kernel.org, linux-afs@lists.infradead.org,
	linux-cifs@vger.kernel.org, linux-integrity@vger.kernel.org,
	linux-kernel@vger.kernel.org, linux-nfs@vger.kernel.org,
	linux-s390@vger.kernel.org,
	linux-security-module@vger.kernel.org, netdev@vger.kernel.org,
	samba-technical@lists.samba.org,
	Alexander Gordeev <agordeev@linux.ibm.com>,
	Anna Schumaker <anna@kernel.org>,
	Bharath SM <bharathsm@microsoft.com>,
	Christian Borntraeger <borntraeger@linux.ibm.com>,
	"David S. Miller" <davem@davemloft.net>,
	Eric Biggers <ebiggers@kernel.org>,
	Eric Dumazet <edumazet@kernel.org>,
	Heiko Carstens <hca@linux.ibm.com>,
	Jakub Kicinski <kuba@kernel.org>,
	James Bottomley <James.Bottomley@hansenpartnership.com>,
	James Morris <jmorris@namei.org>,
	Marc Dionne <marc.dionne@auristor.com>,
	Mimi Zohar <zohar@linux.ibm.com>,
	Namjae Jeon <linkinjeon@kernel.org>,
	Paolo Abeni <pabeni@redhat.com>, Paul Moore <paul@paul-moore.com>,
	Paulo Alcantara <pc@manguebit.org>,
	Ronnie Sahlberg <ronniesahlberg@gmail.com>,
	"Serge E. Hallyn" <serge@hallyn.com>,
	Shyam Prasad N <sprasad@microsoft.com>,
	Simon Horman <horms@kernel.org>,
	Sumit Garg <sumit.garg@kernel.org>,
	Sven Schnelle <svens@linux.ibm.com>, Tom Talpey <tom@talpey.com>,
	Trond Myklebust <trondmy@kernel.org>,
	Vasily Gorbik <gor@linux.ibm.com>
Subject: Re: [RFC PATCH v3 00/11] KEYS: Sanitize the garbage collector
Date: Thu, 5 Nov 2026 03:04:51 +0200	[thread overview]
Message-ID: <auvWM6zn1MSP5C5b@kernel.org> (raw)
In-Reply-To: <20261105005311.804098-1-jarkko@kernel.org>

On Thu, Nov 05, 2026 at 02:52:59AM +0200, Jarkko Sakkinen wrote:
> 1. Replace key_type_dead retyping with module reference pinning so live
>    keys and restrictions keep their backing key_type module loaded until
>    key_gc_unused_keys() or keyring_destroy() completes.
> 2. Process serial-tree scans and graveyard destruction in fixed 64-key
>    chunks to a bound workqueue execution latency without hurting overall
>    throughput for a significant amount.
> 
> The first part has a user-space visible effect: rmmod returns -EBUSY while
> keys or keyring restrictions still pin the backing module.
> 
> Internal key caches or transport state bottlenecks are addressed by either
> clearing their private caches on last-client teardown or splitting key
> types into dedicated provider modules.

<samba-technical@lists.samba.org>: host
    lists-mx.samba.org[2a01:4f8:192:486::1:0] said: 550-Please subscribe the
    address jarkko@kernel.org to the samba-technical 550-mailing list before
    posting. See <https://lists.samba.org/mailman/listinfo/samba-technical>.
    550-You can also write a mail to samba-technical-owner@lists.samba.org and
    ask 550 to get on the list of accepted non-members to be able to send mail
    here. (in reply to end of DATA command)

This was accidentally pulled by:

	./scripts/get_maintainer.pl \
		--nogit \
		--nogit-fallback \
		--norolestats v3-00?[1-9]-*.patch | \
		awk '{print "Cc: " $0}' | \
		sed -i '/^Subject:/r /dev/stdin' v3-0000-cover-letter.patch

I'll address this in the next version but perhaps that email should be
removed from MAINTAINERS? 

Br, Jarkko