On Fri, Jul 15, 2022 at 6:13 AM Jason Wang <jasow...@redhat.com> wrote:
>
> On Fri, Jul 15, 2022 at 12:32 AM Eugenio Pérez <epere...@redhat.com> wrote:
> >
> > Finally offering the possibility to enable SVQ from the command line.
> >
> > Signed-off-by: Eugenio Pérez <epere...@redhat.com>
> > Acked-by: Markus Armbruster <arm...@redhat.com>
> > ---
> > qapi/net.json | 9 +++++-
> > net/vhost-vdpa.c | 72 ++++++++++++++++++++++++++++++++++++++++++++++--
> > 2 files changed, 77 insertions(+), 4 deletions(-)
> >
> > diff --git a/qapi/net.json b/qapi/net.json
> > index 9af11e9a3b..75ba2cb989 100644
> > --- a/qapi/net.json
> > +++ b/qapi/net.json
> > @@ -445,12 +445,19 @@
> > # @queues: number of queues to be created for multiqueue vhost-vdpa
> > # (default: 1)
> > #
> > +# @x-svq: Start device with (experimental) shadow virtqueue. (Since 7.1)
> > +# (default: false)
> > +#
> > +# Features:
> > +# @unstable: Member @x-svq is experimental.
> > +#
> > # Since: 5.1
> > ##
> > { 'struct': 'NetdevVhostVDPAOptions',
> > 'data': {
> > '*vhostdev': 'str',
> > - '*queues': 'int' } }
> > + '*queues': 'int',
> > + '*x-svq': {'type': 'bool', 'features' : [ 'unstable'] } } }
> >
> > ##
> > # @NetdevVmnetHostOptions:
> > diff --git a/net/vhost-vdpa.c b/net/vhost-vdpa.c
> > index 7ccf9eaf4d..85148a5114 100644
> > --- a/net/vhost-vdpa.c
> > +++ b/net/vhost-vdpa.c
> > @@ -75,6 +75,28 @@ const int vdpa_feature_bits[] = {
> > VHOST_INVALID_FEATURE_BIT
> > };
> >
> > +/** Supported device specific feature bits with SVQ */
> > +static const uint64_t vdpa_svq_device_features =
> > + BIT_ULL(VIRTIO_NET_F_CSUM) |
> > + BIT_ULL(VIRTIO_NET_F_GUEST_CSUM) |
> > + BIT_ULL(VIRTIO_NET_F_MTU) |
> > + BIT_ULL(VIRTIO_NET_F_MAC) |
> > + BIT_ULL(VIRTIO_NET_F_GUEST_TSO4) |
> > + BIT_ULL(VIRTIO_NET_F_GUEST_TSO6) |
> > + BIT_ULL(VIRTIO_NET_F_GUEST_ECN) |
> > + BIT_ULL(VIRTIO_NET_F_GUEST_UFO) |
> > + BIT_ULL(VIRTIO_NET_F_HOST_TSO4) |
> > + BIT_ULL(VIRTIO_NET_F_HOST_TSO6) |
> > + BIT_ULL(VIRTIO_NET_F_HOST_ECN) |
> > + BIT_ULL(VIRTIO_NET_F_HOST_UFO) |
> > + BIT_ULL(VIRTIO_NET_F_MRG_RXBUF) |
> > + BIT_ULL(VIRTIO_NET_F_STATUS) |
> > + BIT_ULL(VIRTIO_NET_F_CTRL_VQ) |
> > + BIT_ULL(VIRTIO_F_ANY_LAYOUT) |
> > + BIT_ULL(VIRTIO_NET_F_CTRL_MAC_ADDR) |
> > + BIT_ULL(VIRTIO_NET_F_RSC_EXT) |
> > + BIT_ULL(VIRTIO_NET_F_STANDBY);
>
> We need to have a plan for the full feature support like
>
> indirect, event_index, and packed.
>
Event idx is almost straightforward to develop. Packed has more code
but it's equally doable. In order it should not be hard too.
Indirect is a little bit more complicated because of the indirect
table. I guess we will need to either allocate a big buffer where we
can obtain indirect tables and cvq buffers, or allocate & map them
individually.
Note that we can half-support them. To enable them in the guest's
vring is as easy as to accept that feature in SVQ, and SVQ can easily
translate one format to another. I know the interesting part is the
shadow vring to speed the communication with the device, but it's
still a first step in that direction if needed.
> I can help in developing some of these if you wish.
>
We could plan for the next release cycle for sure.
Thanks!
> Thanks
>
> > +
> > VHostNetState *vhost_vdpa_get_vhost_net(NetClientState *nc)
> > {
> > VhostVDPAState *s = DO_UPCAST(VhostVDPAState, nc, nc);
> > @@ -133,9 +155,13 @@ err_init:
> > static void vhost_vdpa_cleanup(NetClientState *nc)
> > {
> > VhostVDPAState *s = DO_UPCAST(VhostVDPAState, nc, nc);
> > + struct vhost_dev *dev = &s->vhost_net->dev;
> >
> > qemu_vfree(s->cvq_cmd_out_buffer);
> > qemu_vfree(s->cvq_cmd_in_buffer);
> > + if (dev->vq_index + dev->nvqs == dev->vq_index_end) {
> > + g_clear_pointer(&s->vhost_vdpa.iova_tree, vhost_iova_tree_delete);
> > + }
> > if (s->vhost_net) {
> > vhost_net_cleanup(s->vhost_net);
> > g_free(s->vhost_net);
> > @@ -437,7 +463,9 @@ static NetClientState
> > *net_vhost_vdpa_init(NetClientState *peer,
> > int vdpa_device_fd,
> > int queue_pair_index,
> > int nvqs,
> > - bool is_datapath)
> > + bool is_datapath,
> > + bool svq,
> > + VhostIOVATree *iova_tree)
> > {
> > NetClientState *nc = NULL;
> > VhostVDPAState *s;
> > @@ -455,6 +483,8 @@ static NetClientState
> > *net_vhost_vdpa_init(NetClientState *peer,
> >
> > s->vhost_vdpa.device_fd = vdpa_device_fd;
> > s->vhost_vdpa.index = queue_pair_index;
> > + s->vhost_vdpa.shadow_vqs_enabled = svq;
> > + s->vhost_vdpa.iova_tree = iova_tree;
> > if (!is_datapath) {
> > s->cvq_cmd_out_buffer = qemu_memalign(qemu_real_host_page_size(),
> >
> > vhost_vdpa_net_cvq_cmd_page_len());
> > @@ -465,6 +495,8 @@ static NetClientState
> > *net_vhost_vdpa_init(NetClientState *peer,
> >
> > s->vhost_vdpa.shadow_vq_ops = &vhost_vdpa_net_svq_ops;
> > s->vhost_vdpa.shadow_vq_ops_opaque = s;
> > + error_setg(&s->vhost_vdpa.migration_blocker,
> > + "Migration disabled: vhost-vdpa uses CVQ.");
> > }
> > ret = vhost_vdpa_add(nc, (void *)&s->vhost_vdpa, queue_pair_index,
> > nvqs);
> > if (ret) {
> > @@ -474,6 +506,14 @@ static NetClientState
> > *net_vhost_vdpa_init(NetClientState *peer,
> > return nc;
> > }
> >
> > +static int vhost_vdpa_get_iova_range(int fd,
> > + struct vhost_vdpa_iova_range
> > *iova_range)
> > +{
> > + int ret = ioctl(fd, VHOST_VDPA_GET_IOVA_RANGE, iova_range);
> > +
> > + return ret < 0 ? -errno : 0;
> > +}
> > +
> > static int vhost_vdpa_get_features(int fd, uint64_t *features, Error
> > **errp)
> > {
> > int ret = ioctl(fd, VHOST_GET_FEATURES, features);
> > @@ -524,6 +564,7 @@ int net_init_vhost_vdpa(const Netdev *netdev, const
> > char *name,
> > uint64_t features;
> > int vdpa_device_fd;
> > g_autofree NetClientState **ncs = NULL;
> > + g_autoptr(VhostIOVATree) iova_tree = NULL;
> > NetClientState *nc;
> > int queue_pairs, r, i, has_cvq = 0;
> >
> > @@ -551,22 +592,45 @@ int net_init_vhost_vdpa(const Netdev *netdev, const
> > char *name,
> > return queue_pairs;
> > }
> >
> > + if (opts->x_svq) {
> > + struct vhost_vdpa_iova_range iova_range;
> > +
> > + uint64_t invalid_dev_features =
> > + features & ~vdpa_svq_device_features &
> > + /* Transport are all accepted at this point */
> > + ~MAKE_64BIT_MASK(VIRTIO_TRANSPORT_F_START,
> > + VIRTIO_TRANSPORT_F_END -
> > VIRTIO_TRANSPORT_F_START);
> > +
> > + if (invalid_dev_features) {
> > + error_setg(errp, "vdpa svq does not work with features 0x%"
> > PRIx64,
> > + invalid_dev_features);
> > + goto err_svq;
> > + }
> > +
> > + vhost_vdpa_get_iova_range(vdpa_device_fd, &iova_range);
> > + iova_tree = vhost_iova_tree_new(iova_range.first, iova_range.last);
> > + }
> > +
> > ncs = g_malloc0(sizeof(*ncs) * queue_pairs);
> >
> > for (i = 0; i < queue_pairs; i++) {
> > ncs[i] = net_vhost_vdpa_init(peer, TYPE_VHOST_VDPA, name,
> > - vdpa_device_fd, i, 2, true);
> > + vdpa_device_fd, i, 2, true,
> > opts->x_svq,
> > + iova_tree);
> > if (!ncs[i])
> > goto err;
> > }
> >
> > if (has_cvq) {
> > nc = net_vhost_vdpa_init(peer, TYPE_VHOST_VDPA, name,
> > - vdpa_device_fd, i, 1, false);
> > + vdpa_device_fd, i, 1, false,
> > + opts->x_svq, iova_tree);
> > if (!nc)
> > goto err;
> > }
> >
> > + /* iova_tree ownership belongs to last NetClientState */
> > + g_steal_pointer(&iova_tree);
> > return 0;
> >
> > err:
> > @@ -575,6 +639,8 @@ err:
> > qemu_del_net_client(ncs[i]);
> > }
> > }
> > +
> > +err_svq:
> > qemu_close(vdpa_device_fd);
> >
> > return -1;
> > --
> > 2.31.1
> >
>
