MIME-Version: 1.0
References: <20230208200957.14073-1-djeffery@redhat.com>
In-Reply-To: <20230208200957.14073-1-djeffery@redhat.com>
From:   Maurizio Lombardi <mlombard@redhat.com>
Date:   Mon, 13 Feb 2023 12:59:07 +0100
Message-ID: <CAFL455kEoLQF+vc2MGmtLdrQ-=U+HJzqgknZmM54iCPJD1p_mA@mail.gmail.com>
Subject: Re: [PATCH] scsi: target: iscsi: set memalloc_noio with loopback
 network connections
To:     David Jeffery <djeffery@redhat.com>
Cc:     target-devel@vger.kernel.org,
        "Martin K . Petersen" <martin.petersen@oracle.com>,
        Mike Christie <michael.christie@oracle.com>,
        Laurence Oberman <loberman@redhat.com>,
        linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

st 8. 2. 2023 v 21:10 odes=C3=ADlatel David Jeffery <djeffery@redhat.com> n=
apsal:
>
>
> +       /*
> +        * If the iscsi connection is over a loopback device from using
> +        * iscsi and iscsit on the same system, we need to set memalloc_n=
oio to
> +        * prevent memory allocation deadlocks between target and initiat=
or.
> +        */
> +       rcu_read_lock();
> +       dst =3D rcu_dereference(conn->sock->sk->sk_dst_cache);
> +       if (dst && dst->dev && dst->dev->flags & IFF_LOOPBACK)
> +               loopback =3D true;
> +       rcu_read_unlock();

Hi Mike,
I tested it, it works. The customer also confirmed that it fixes the
deadlock on his setup.

Maurizio