public inbox for quic@lists.linux.dev 
 help / color / mirror / Atom feed
From: Xin Long <lucien.xin@gmail.com>
To: Paolo Abeni <pabeni@redhat.com>
Cc: Simon Horman <horms@kernel.org>,
	steved@redhat.com, marcelo.leitner@gmail.com,
	 aahringo@redhat.com, alibuda@linux.alibaba.com,
	jbaron@akamai.com,  hare@suse.de, kuba@kernel.org,
	tom@talpey.com, linux-cifs@vger.kernel.org,  daniel@haxx.se,
	quic@lists.linux.dev, jlayton@kernel.org, tfanelli@redhat.com,
	 dhowells@redhat.com, linkinjeon@kernel.org,
	hepengtao@xiaomi.com,  pc@manguebit.com,
	kernel-tls-handshake@lists.linux.dev,  illiliti@protonmail.com,
	xiyou.wangcong@gmail.com,  andrew.gospodarek@broadcom.com,
	mail@johnericson.me, edumazet@google.com,  matttbe@kernel.org,
	netdev@vger.kernel.org, smfrench@gmail.com,  metze@samba.org,
	mbuhl@openbsd.org, chuck.lever@oracle.com, dreibh@simula.no,
	 davem@davemloft.net, sd@queasysnail.net
Subject: Re: [net-next,v9,02/15] net: build socket infrastructure for QUIC protocol
Date: Thu, 5 Feb 2026 14:03:52 -0500	[thread overview]
Message-ID: <CADvbK_eFXJB-JqL7w-50Jb=GNzzmSFD_cRGSmRscZ2Hq0o=6ZQ@mail.gmail.com> (raw)
In-Reply-To: <e9b073b0-60b7-4a45-9b6d-85c15b51fb4f@redhat.com>

On Thu, Feb 5, 2026 at 7:48 AM Paolo Abeni <pabeni@redhat.com> wrote:
>
> On 2/5/26 12:54 PM, Simon Horman wrote:
> > This is an AI-generated review of your patch. The human sending this
> > email has considered the AI review valid, or at least plausible.
> >
> > For local reproduction steps see https://netdev-ai.bots.linux.dev/ai-local.html
> > ---
> > net: build socket infrastructure for QUIC protocol
> >
> > This patch lays the groundwork for QUIC socket support in the kernel.
> > It defines the core structures and protocol hooks needed to create
> > QUIC sockets, without implementing any protocol behavior at this stage.
> > Basic integration is included to allow building the module via
> > CONFIG_IP_QUIC=m. This provides the scaffolding necessary for adding
> > actual QUIC socket behavior in follow-up patches.
> >
> >> diff --git a/net/quic/protocol.c b/net/quic/protocol.c
> >> new file mode 100644
> >> index 000000000000..6ab81f1c50cc
> >> --- /dev/null
> >> +++ b/net/quic/protocol.c
> >
> > [ ... ]
> >
> >> @@ -0,0 +1,369 @@
> >
> > [ ... ]
> >
> >> +static int quic_protosw_init(void)
> >> +{
> >> +    int err;
> >> +
> >> +    err = proto_register(&quic_prot, 1);
> >> +    if (err)
> >> +            return err;
> >> +
> >> +    err = proto_register(&quicv6_prot, 1);
> >> +    if (err) {
> >> +            proto_unregister(&quic_prot);
> >> +            return err;
> >> +    }
> >> +
> >> +    inet_register_protosw(&quic_stream_protosw);
> >> +    inet_register_protosw(&quic_dgram_protosw);
> >> +    inet6_register_protosw(&quicv6_stream_protosw);
> >> +    inet6_register_protosw(&quicv6_dgram_protosw);
> >> +
> >> +    return 0;
> >> +}
> >
> > Can the return values from inet6_register_protosw() be checked?
> >
> > Looking at the implementation, inet6_register_protosw() returns int and
> > can fail with -EINVAL (if p->type >= SOCK_MAX) or -EPERM (if trying to
> > override a permanent protocol). The calls at lines 234-235 don't check
> > these return values, so if registration fails, quic_protosw_init() still
> > returns 0 (success).
> >
> > More critically, quic_protosw_exit() unconditionally calls
> > inet6_unregister_protosw() on both structures:
> >
> > static void quic_protosw_exit(void)
> > {
> >       inet_unregister_protosw(&quic_dgram_protosw);
> >       inet_unregister_protosw(&quic_stream_protosw);
> >       proto_unregister(&quic_prot);
> >
> >       inet6_unregister_protosw(&quicv6_dgram_protosw);
> >       inet6_unregister_protosw(&quicv6_stream_protosw);
> >       proto_unregister(&quicv6_prot);
> > }
> >
> > If a protosw was never successfully registered, its list field remains
> > zero-initialized (NULL pointers). inet6_unregister_protosw() calls
> > list_del_rcu(&p->list), which with NULL pointers will dereference NULL
> > in __list_del_entry() -> __list_del(NULL, NULL), causing a kernel crash.
> >
> > Note that inet_register_protosw() returns void (cannot fail from caller's
> > perspective), but inet6_register_protosw() returns int (can fail). The
> > code correctly checks errors from proto_register() but not from
> > inet6_register_protosw().
> >
> > Should the return values from inet6_register_protosw() be checked, and
> > should the cleanup path track which registrations succeeded so only those
> > are unregistered?
>
> I think it's easy to infer that that such failures are impossible, but
> since a repost is needed please drop some sentence in the commit message
> to help AI see the point.
>
Will do, thanks.

  reply	other threads:[~2026-02-05 19:04 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-02-02 14:27 [PATCH net-next v9 00/15] net: introduce QUIC infrastructure and core subcomponents Xin Long
2026-02-02 14:27 ` [PATCH net-next v9 01/15] net: define IPPROTO_QUIC and SOL_QUIC constants Xin Long
2026-02-02 14:27 ` [PATCH net-next v9 02/15] net: build socket infrastructure for QUIC protocol Xin Long
2026-02-05 11:54   ` [net-next,v9,02/15] " Simon Horman
2026-02-05 12:48     ` Paolo Abeni
2026-02-05 19:03       ` Xin Long [this message]
2026-02-02 14:27 ` [PATCH net-next v9 03/15] quic: provide common utilities and data structures Xin Long
2026-02-05 11:54   ` [net-next,v9,03/15] " Simon Horman
2026-02-05 12:51     ` Paolo Abeni
2026-02-05 19:18       ` Xin Long
2026-02-02 14:27 ` [PATCH net-next v9 04/15] quic: provide family ops for address and protocol Xin Long
2026-02-02 14:27 ` [PATCH net-next v9 05/15] quic: provide quic.h header files for kernel and userspace Xin Long
2026-02-05 11:55   ` [net-next,v9,05/15] " Simon Horman
2026-02-05 19:37     ` Xin Long
2026-02-02 14:27 ` [PATCH net-next v9 06/15] quic: add stream management Xin Long
2026-02-05 11:55   ` [net-next,v9,06/15] " Simon Horman
2026-02-05 18:56     ` Xin Long
2026-02-02 14:27 ` [PATCH net-next v9 07/15] quic: add connection id management Xin Long
2026-02-02 14:27 ` [PATCH net-next v9 08/15] quic: add path management Xin Long
2026-02-02 14:27 ` [PATCH net-next v9 09/15] quic: add congestion control Xin Long
2026-02-05 11:55   ` [net-next,v9,09/15] " Simon Horman
2026-02-05 19:00     ` Xin Long
2026-02-02 14:27 ` [PATCH net-next v9 10/15] quic: add packet number space Xin Long
2026-02-02 14:27 ` [PATCH net-next v9 11/15] quic: add crypto key derivation and installation Xin Long
2026-02-02 14:27 ` [PATCH net-next v9 12/15] quic: add crypto packet encryption and decryption Xin Long
2026-02-02 14:27 ` [PATCH net-next v9 13/15] quic: add timer management Xin Long
2026-02-02 14:27 ` [PATCH net-next v9 14/15] quic: add packet builder base Xin Long
2026-02-02 14:27 ` [PATCH net-next v9 15/15] quic: add packet parser base Xin Long
2026-02-05 11:55   ` [net-next,v9,15/15] " Simon Horman
2026-02-05 19:02     ` Xin Long

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CADvbK_eFXJB-JqL7w-50Jb=GNzzmSFD_cRGSmRscZ2Hq0o=6ZQ@mail.gmail.com' \
    --to=lucien.xin@gmail.com \
    --cc=aahringo@redhat.com \
    --cc=alibuda@linux.alibaba.com \
    --cc=andrew.gospodarek@broadcom.com \
    --cc=chuck.lever@oracle.com \
    --cc=daniel@haxx.se \
    --cc=davem@davemloft.net \
    --cc=dhowells@redhat.com \
    --cc=dreibh@simula.no \
    --cc=edumazet@google.com \
    --cc=hare@suse.de \
    --cc=hepengtao@xiaomi.com \
    --cc=horms@kernel.org \
    --cc=illiliti@protonmail.com \
    --cc=jbaron@akamai.com \
    --cc=jlayton@kernel.org \
    --cc=kernel-tls-handshake@lists.linux.dev \
    --cc=kuba@kernel.org \
    --cc=linkinjeon@kernel.org \
    --cc=linux-cifs@vger.kernel.org \
    --cc=mail@johnericson.me \
    --cc=marcelo.leitner@gmail.com \
    --cc=matttbe@kernel.org \
    --cc=mbuhl@openbsd.org \
    --cc=metze@samba.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=pc@manguebit.com \
    --cc=quic@lists.linux.dev \
    --cc=sd@queasysnail.net \
    --cc=smfrench@gmail.com \
    --cc=steved@redhat.com \
    --cc=tfanelli@redhat.com \
    --cc=tom@talpey.com \
    --cc=xiyou.wangcong@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox