Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 13 Nov 2019 10:22:37 -0500
From:   "Michael S. Tsirkin" <mst@redhat.com>
To:     Laurent Vivier <lvivier@redhat.com>
Cc:     linux-kernel@vger.kernel.org, Arnd Bergmann <arnd@arndb.de>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Amit Shah <amit@kernel.org>,
        virtualization@lists.linux-foundation.org, stable@vger.kernel.org
Subject: Re: [PATCH v2] virtio_console: allocate inbufs in add_port() only if
 it is needed
Message-ID: <20191113102126-mutt-send-email-mst@kernel.org>
References: <20191113150056.9990-1-lvivier@redhat.com>
 <20191113101929-mutt-send-email-mst@kernel.org>
MIME-Version: 1.0
In-Reply-To: <20191113101929-mutt-send-email-mst@kernel.org>
Content-Type: text/plain; charset=WINDOWS-1252
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, Nov 13, 2019 at 10:21:11AM -0500, Michael S. Tsirkin wrote:
> On Wed, Nov 13, 2019 at 04:00:56PM +0100, Laurent Vivier wrote:
> > When we hot unplug a virtserialport and then try to hot plug again,
> > it fails:
> >=20
> > (qemu) chardev-add socket,id=3Dserial0,path=3D/tmp/serial0,server,nowai=
t
> > (qemu) device_add virtserialport,bus=3Dvirtio-serial0.0,nr=3D2,\
> >                   chardev=3Dserial0,id=3Dserial0,name=3Dserial0
> > (qemu) device_del serial0
> > (qemu) device_add virtserialport,bus=3Dvirtio-serial0.0,nr=3D2,\
> >                   chardev=3Dserial0,id=3Dserial0,name=3Dserial0
> > kernel error:
> >   virtio-ports vport2p2: Error allocating inbufs
> > qemu error:
> >   virtio-serial-bus: Guest failure in adding port 2 for device \
> >                      virtio-serial0.0
> >=20
> > This happens because buffers for the in_vq are allocated when the port =
is
> > added but are not released when the port is unplugged.
> >=20
> > They are only released when virtconsole is removed (see a7a69ec0d8e4)
> >=20
> > To avoid the problem and to be symmetric, we could allocate all the buf=
fers
> > in init_vqs() as they are released in remove_vqs(), but it sounds like
> > a waste of memory.
> >=20
> > Rather than that, this patch changes add_port() logic to ignore ENOSPC
> > error in fill_queue(), which means queue has already been filled.
> >=20
> > Fixes: a7a69ec0d8e4 ("virtio_console: free buffers after reset")
> > Cc: mst@redhat.com
> > Cc: stable@vger.kernel.org
> > Signed-off-by: Laurent Vivier <lvivier@redhat.com>
> > ---
> >=20
> > Notes:
> >     v2: making fill_queue return int and testing return code for -ENOSP=
C
> >=20
> >  drivers/char/virtio_console.c | 24 +++++++++---------------
> >  1 file changed, 9 insertions(+), 15 deletions(-)
> >=20
> > diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_consol=
e.c
> > index 7270e7b69262..9e6534fd1aa4 100644
> > --- a/drivers/char/virtio_console.c
> > +++ b/drivers/char/virtio_console.c
> > @@ -1325,24 +1325,24 @@ static void set_console_size(struct port *port,=
 u16 rows, u16 cols)
> >  =09port->cons.ws.ws_col =3D cols;
> >  }
> > =20
> > -static unsigned int fill_queue(struct virtqueue *vq, spinlock_t *lock)
> > +static int fill_queue(struct virtqueue *vq, spinlock_t *lock)
> >  {
> >  =09struct port_buffer *buf;
> > -=09unsigned int nr_added_bufs;
> > +=09int nr_added_bufs;
> >  =09int ret;
> > =20
> >  =09nr_added_bufs =3D 0;
> >  =09do {
> >  =09=09buf =3D alloc_buf(vq->vdev, PAGE_SIZE, 0);
> >  =09=09if (!buf)
> > -=09=09=09break;
> > +=09=09=09return -ENOMEM;
> > =20
> >  =09=09spin_lock_irq(lock);
> >  =09=09ret =3D add_inbuf(vq, buf);
> >  =09=09if (ret < 0) {
> >  =09=09=09spin_unlock_irq(lock);
> >  =09=09=09free_buf(buf, true);
> > -=09=09=09break;
> > +=09=09=09return ret;
> >  =09=09}
> >  =09=09nr_added_bufs++;
> >  =09=09spin_unlock_irq(lock);

So actually, how about handling ENOSPC specially here, and
returning success? After all queue is full as requested ...


> > @@ -1362,7 +1362,6 @@ static int add_port(struct ports_device *portdev,=
 u32 id)
> >  =09char debugfs_name[16];
> >  =09struct port *port;
> >  =09dev_t devt;
> > -=09unsigned int nr_added_bufs;
> >  =09int err;
> > =20
> >  =09port =3D kmalloc(sizeof(*port), GFP_KERNEL);
> > @@ -1421,11 +1420,9 @@ static int add_port(struct ports_device *portdev=
, u32 id)
> >  =09spin_lock_init(&port->outvq_lock);
> >  =09init_waitqueue_head(&port->waitqueue);
> > =20
> > -=09/* Fill the in_vq with buffers so the host can send us data. */
> > -=09nr_added_bufs =3D fill_queue(port->in_vq, &port->inbuf_lock);
> > -=09if (!nr_added_bufs) {
> > +=09err =3D fill_queue(port->in_vq, &port->inbuf_lock);
> > +=09if (err < 0 && err !=3D -ENOSPC) {
> >  =09=09dev_err(port->dev, "Error allocating inbufs\n");
> > -=09=09err =3D -ENOMEM;
> >  =09=09goto free_device;
> >  =09}
> > =20
>=20
> Pls add a comment explaining that -ENOSPC happens when
> queue already has buffers (e.g. from previous detach).
>=20
>=20
> > @@ -2059,14 +2056,11 @@ static int virtcons_probe(struct virtio_device =
*vdev)
> >  =09INIT_WORK(&portdev->control_work, &control_work_handler);
> > =20
> >  =09if (multiport) {
> > -=09=09unsigned int nr_added_bufs;
> > -
> >  =09=09spin_lock_init(&portdev->c_ivq_lock);
> >  =09=09spin_lock_init(&portdev->c_ovq_lock);
> > =20
> > -=09=09nr_added_bufs =3D fill_queue(portdev->c_ivq,
> > -=09=09=09=09=09   &portdev->c_ivq_lock);
> > -=09=09if (!nr_added_bufs) {
> > +=09=09err =3D fill_queue(portdev->c_ivq, &portdev->c_ivq_lock);
> > +=09=09if (err < 0) {
> >  =09=09=09dev_err(&vdev->dev,
> >  =09=09=09=09"Error allocating buffers for control queue\n");
> >  =09=09=09/*
> > @@ -2077,7 +2071,7 @@ static int virtcons_probe(struct virtio_device *v=
dev)
> >  =09=09=09=09=09   VIRTIO_CONSOLE_DEVICE_READY, 0);
> >  =09=09=09/* Device was functional: we need full cleanup. */
> >  =09=09=09virtcons_remove(vdev);
> > -=09=09=09return -ENOMEM;
> > +=09=09=09return err;
> >  =09=09}
> >  =09} else {
> >  =09=09/*
> > --=20
> > 2.23.0