Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Wed, 18 Nov 2020 11:23:55 +0000
From:   Stefan Hajnoczi <stefanha@redhat.com>
To:     Stefano Garzarella <sgarzare@redhat.com>
Cc:     virtualization@lists.linux-foundation.org,
        Laurent Vivier <lvivier@redhat.com>,
        linux-kernel@vger.kernel.org, Eli Cohen <elic@nvidia.com>,
        Jason Wang <jasowang@redhat.com>,
        "Michael S. Tsirkin" <mst@redhat.com>,
        Max Gurtovoy <mgurtovoy@nvidia.com>
Subject: Re: [PATCH RFC 04/12] vdpa: add vdpa simulator for block device
Message-ID: <20201118112355.GE182763@stefanha-x1.localdomain>
References: <20201113134712.69744-1-sgarzare@redhat.com>
 <20201113134712.69744-5-sgarzare@redhat.com>
 <20201117111121.GD131917@stefanha-x1.localdomain>
 <20201117141620.ytium7r6xpxi4and@steredhat>
 <20201117164342.GT131917@stefanha-x1.localdomain>
 <20201117173811.lw4nwicoykhnn3tt@steredhat>
MIME-Version: 1.0
In-Reply-To: <20201117173811.lw4nwicoykhnn3tt@steredhat>
Content-Type: multipart/signed; micalg=pgp-sha256;
        protocol="application/pgp-signature"; boundary="AsxXAMtlQ5JHofzM"
Content-Disposition: inline
Precedence: bulk

--AsxXAMtlQ5JHofzM
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Nov 17, 2020 at 06:38:11PM +0100, Stefano Garzarella wrote:
> On Tue, Nov 17, 2020 at 04:43:42PM +0000, Stefan Hajnoczi wrote:
> > On Tue, Nov 17, 2020 at 03:16:20PM +0100, Stefano Garzarella wrote:
> > > On Tue, Nov 17, 2020 at 11:11:21AM +0000, Stefan Hajnoczi wrote:
> > > > On Fri, Nov 13, 2020 at 02:47:04PM +0100, Stefano Garzarella wrote:
> > > > > +static void vdpasim_blk_work(struct work_struct *work)
> > > > > +{
> > > > > +=09struct vdpasim *vdpasim =3D container_of(work, struct vdpasim=
, work);
> > > > > +=09u8 status =3D VIRTIO_BLK_S_OK;
> > > > > +=09int i;
> > > > > +
> > > > > +=09spin_lock(&vdpasim->lock);
> > > > > +
> > > > > +=09if (!(vdpasim->status & VIRTIO_CONFIG_S_DRIVER_OK))
> > > > > +=09=09goto out;
> > > > > +
> > > > > +=09for (i =3D 0; i < VDPASIM_BLK_VQ_NUM; i++) {
> > > > > +=09=09struct vdpasim_virtqueue *vq =3D &vdpasim->vqs[i];
> > > > > +
> > > > > +=09=09if (!vq->ready)
> > > > > +=09=09=09continue;
> > > > > +
> > > > > +=09=09while (vringh_getdesc_iotlb(&vq->vring, &vq->iov, &vq->iov=
,
> > > > > +=09=09=09=09=09    &vq->head, GFP_ATOMIC) > 0) {
> > > > > +
> > > > > +=09=09=09int write;
> > > > > +
> > > > > +=09=09=09vq->iov.i =3D vq->iov.used - 1;
> > > > > +=09=09=09write =3D vringh_iov_push_iotlb(&vq->vring, &vq->iov, &=
status, 1);
> > > > > +=09=09=09if (write <=3D 0)
> > > > > +=09=09=09=09break;
> > > >
> > > > We're lucky the guest driver doesn't crash after VIRTIO_BLK_T_GET_I=
D? :)
> > >=20
> > > The crash could happen if the simulator doesn't put the string termin=
ator,
> > > but in virtio_blk.c, the serial_show() initialize the buffer putting =
the
> > > string terminator in the VIRTIO_BLK_ID_BYTES element:
> > >=20
> > >     buf[VIRTIO_BLK_ID_BYTES] =3D '\0';
> > >     err =3D virtblk_get_id(disk, buf);
> > >=20
> > > This should prevent the issue, right?
> > >=20
> > > However in the last patch of this series I implemented VIRTIO_BLK_T_G=
ET_ID
> > > support :-)
> >=20
> > Windows, BSD, macOS, etc guest drivers aren't necessarily going to
> > terminate or initialize the serial string buffer.
>=20
> Unfortunately I discovered that VIRTIO_BLK_T_GET_ID is not in the VIRTIO
> specs, so, just for curiosity, I checked the QEMU code and I found this:
>=20
>     case VIRTIO_BLK_T_GET_ID:
>     {
>         /*
>          * NB: per existing s/n string convention the string is
>          * terminated by '\0' only when shorter than buffer.
>          */
>         const char *serial =3D s->conf.serial ? s->conf.serial : "";
>         size_t size =3D MIN(strlen(serial) + 1,
>                           MIN(iov_size(in_iov, in_num),
>                               VIRTIO_BLK_ID_BYTES));
>         iov_from_buf(in_iov, in_num, 0, serial, size);
>         virtio_blk_req_complete(req, VIRTIO_BLK_S_OK);
>         virtio_blk_free_request(req);
>         break;
>     }
>=20
> It seems that the device emulation in QEMU expects that the driver
> terminates the serial string buffer.
>=20
> Do you know why VIRTIO_BLK_T_GET_ID is not in the specs?
> Should we add it?

It's about to be merged into the VIRTIO spec:
https://github.com/oasis-tcs/virtio-spec/issues/63

Stefan

--AsxXAMtlQ5JHofzM
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iQEzBAEBCAAdFiEEhpWov9P5fNqsNXdanKSrs4Grc8gFAl+1BEsACgkQnKSrs4Gr
c8iCzgf+MQ+2r8v8e14Ettix74PLbceOdl1s1ruXCeKEWWwegoM/UZ1X0HviTaHO
jy/txmM7MpqE/QXAY+ODeiHjArVWKRDI/L8oWQiLXJKF8/Xm6ZXApG9Ca188w6Z5
JZ7+Awr7KHqdLkWgGG5d3jk0/TKin+l8g19PsiucyhpifYpHI5MNPKyhk3reXD/8
VcrVSWyWHrHjVON+kj8f1K6F8D0CDVPhT/+aJOWCjXGVRePEeGp5qNJTJMCC22Ix
OeoLj0OzncGzeX/2uaDIpNHOFPgFLSGg58kVK/HTduSPJrUZb4srRlh5HO6ajUQ0
+u6h5f9GBGkqOkbfNiondE4XYTOG3A==
=N5O5
-----END PGP SIGNATURE-----

--AsxXAMtlQ5JHofzM--