netdev - Re: [PATCH] rxrpc_find_service_conn_rcu: use read_seqbegin() rather than read_seqbegin_or

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20231101222946.GF32034@redhat.com>
Date: Wed, 1 Nov 2023 23:29:46 +0100
From: Oleg Nesterov <oleg@...hat.com>
To: David Howells <dhowells@...hat.com>
Cc: Marc Dionne <marc.dionne@...istor.com>,
	Alexander Viro <viro@...iv.linux.org.uk>,
	"David S. Miller" <davem@...emloft.net>,
	Eric Dumazet <edumazet@...gle.com>,
	Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
	Chuck Lever <chuck.lever@...cle.com>, linux-afs@...ts.infradead.org,
	netdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] rxrpc_find_service_conn_rcu: use read_seqbegin() rather
 than read_seqbegin_or_lock()

sorry for noise, but in case I wasn't clear...

On 11/01, Oleg Nesterov wrote:
>
> On 11/01, David Howells wrote:
> >
> > However, I think just changing all of these to always-lockless isn't
> > necessarily the most optimal way.
>
> Yes, but so far I am trying to change the users which never take the
> lock for writing, so this patch doesn't change the current behaviour.
>
> > I wonder if struct seqlock would make more sense with an rwlock rather than a
> > spinlock.  As it is, it does an exclusive spinlock for the readpath which is
> > kind of overkill.
>
> Heh. Please see
>
> 	[PATCH 4/5] seqlock: introduce read_seqcount_begin_or_lock() and friends
> 	https://lore.kernel.org/all/20230913155005.GA26252@redhat.com/
>

I meant, we already have seqcount_rwlock_t, but currently you can't do
something like read_seqbegin_or_lock(&seqcount_rwlock_t).

> I am going to return to this later.

Yes.

Oleg.