Received-SPF: pass (google.com: domain of linux-kernel+bounces-42607-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
From: Luis Henriques <lhenriques@suse.de>
To: Eric Biggers <ebiggers@kernel.org>
Cc: David Howells <dhowells@redhat.com>,  Jarkko Sakkinen
 <jarkko@kernel.org>,  keyrings@vger.kernel.org,
  linux-kernel@vger.kernel.org
Subject: Re: [RFC PATCH v2] keys: update key quotas in key_put()
In-Reply-To: <20240127064220.GB11935@sol.localdomain> (Eric Biggers's message
	of "Fri, 26 Jan 2024 22:42:20 -0800")
References: <2744563.1702303367@warthog.procyon.org.uk>
	<20240115120300.27606-1-lhenriques@suse.de>
	<20240124221225.GD1088@sol.localdomain> <87bk988450.fsf@suse.de>
	<20240127064220.GB11935@sol.localdomain>
Date: Mon, 29 Jan 2024 11:23:59 +0000
Message-ID: <87bk94qt68.fsf@suse.de>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

Eric Biggers <ebiggers@kernel.org> writes:

> On Fri, Jan 26, 2024 at 04:12:59PM +0000, Luis Henriques wrote:
>> Eric Biggers <ebiggers@kernel.org> writes:
>>=20
>> > On Mon, Jan 15, 2024 at 12:03:00PM +0000, Luis Henriques wrote:
>> >> Delaying key quotas update when key's refcount reaches 0 in key_put()=
 has
>> >> been causing some issues in fscrypt testing.  This patches fixes this=
 test
>> >> flakiness by dealing with the quotas immediately, but leaving all the=
 other
>> >> clean-ups to the key garbage collector.  Unfortunately, this means th=
at we
>> >> also need to switch to the irq-version of the spinlock that protects =
quota.
>> >>=20
>> >> Signed-off-by: Luis Henriques <lhenriques@suse.de>
>> >> ---
>> >> Hi David!
>> >>=20
>> >> I have these changes in my local disk for a while; I wanted to send t=
hem
>> >> before EOY break but... yeah, it didn't happen.  Anyway, I'm still se=
nding
>> >> it as an RFC as I'm probably missing something.
>> >>=20
>> >>  security/keys/gc.c     |  8 --------
>> >>  security/keys/key.c    | 32 ++++++++++++++++++++++----------
>> >>  security/keys/keyctl.c | 11 ++++++-----
>> >>  3 files changed, 28 insertions(+), 23 deletions(-)
>> >
>> > This patch seems reasonable to me, though I'm still thinking about cha=
nging
>> > fs/crypto/ to manage its key quotas itself which would avoid the issue=
 entirely.
>> >
>> > Note that as I said before, fs/crypto/ does key_put() on a whole keyri=
ng at
>> > once, in order to release the quota of the keys in the keyring.  Do yo=
u plan to
>> > also change fs/crypto/ to keyring_clear() the keyring before putting i=
t?
>> > Without that, I don't think the problem is solved, as the quota releas=
e will
>> > still happen asynchronously due to the keyring being cleared asynchron=
ously.
>>=20
>> Ah, good point.  In the meantime I had forgotten everything about this
>> code and missed that.  So, I can send another patch to fs/crypto to add
>> that extra call once (or if) this patch is accepted.
>>=20
>> If I'm reading the code correctly, the only place where this extra call =
is
>> required is on fscrypt_put_master_key():
>>=20
>> diff --git a/fs/crypto/keyring.c b/fs/crypto/keyring.c
>> index 0edf0b58daa7..4afd32f1aed9 100644
>> --- a/fs/crypto/keyring.c
>> +++ b/fs/crypto/keyring.c
>> @@ -74,6 +74,7 @@ void fscrypt_put_master_key(struct fscrypt_master_key =
*mk)
>>  	 * that concurrent keyring lookups can no longer find it.
>>  	 */
>>  	WARN_ON_ONCE(refcount_read(&mk->mk_active_refs) !=3D 0);
>> +	keyring_clear(mk->mk_users);
>>  	key_put(mk->mk_users);
>>  	mk->mk_users =3D NULL;
>
> This will need a NULL check, since keyring_clear() doesn't accept NULL:
>
> 	if (mk->mk_users) {
> 		keyring_clear(mk->mk_users);
> 		key_put(mk->mk_users);
> 		mk->mk_users =3D NULL;
> 	}
>

Ah, good point.  Makes sense.

>>  	call_rcu(&mk->mk_rcu_head, fscrypt_free_master_key);
>>=20
>> On the other hand, if you're really working towards dropping this code
>> entirely, maybe there's not point doing that.
>
> Well, it depends whether this patch (the one for security/keys/) is accep=
ted or
> not.  If it's accepted, I think it makes sense to add the keyring_clear()=
 and
> otherwise keep the fs/crypto/ code as-is.  If it's not accepted, I think =
I'll
> have to make fs/crypto/ handle the quotas itself.

Awesome, thank.

David, do you have any feedback on this patch or would you rather have me
send v3, addressing Jarkko comments regarding the commit message?

Cheers,
--=20
Lu=C3=ADs