Message ID | 20190409055850.25797-1-jasowang@redhat.com |
---|---|
State | New |
Headers | show |
Series | [V3] vhost_net: don't set backend for the uninitialized virtqueue | expand |
On 2019/4/9 下午1:58, Jason Wang wrote: > We used to set backend unconditionally, this won't work for some > guests (e.g windows driver) who may not initialize all virtqueues. For > kernel backend, this will fail since it may try to validate the rings > during setting backend. > > Fixing this by simply skipping the backend set when we find desc is > not ready. > > Signed-off-by: Jason Wang <jasowang@redhat.com> > --- > Changes from V1: > - introduce a generic helper to check if a virtqueue is enabled > --- Michael, does this look good to you? Thanks > hw/net/vhost_net.c | 10 ++++++++++ > hw/virtio/virtio.c | 5 +++++ > include/hw/virtio/virtio.h | 1 + > 3 files changed, 16 insertions(+) > > diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c > index be3cc88370..a6b719035c 100644 > --- a/hw/net/vhost_net.c > +++ b/hw/net/vhost_net.c > @@ -244,6 +244,11 @@ static int vhost_net_start_one(struct vhost_net *net, > qemu_set_fd_handler(net->backend, NULL, NULL, NULL); > file.fd = net->backend; > for (file.index = 0; file.index < net->dev.nvqs; ++file.index) { > + if (!virtio_queue_enabled(dev, net->dev.vq_index + > + file.index)) { > + /* Queue might not be ready for start */ > + continue; > + } > r = vhost_net_set_backend(&net->dev, &file); > if (r < 0) { > r = -errno; > @@ -256,6 +261,11 @@ fail: > file.fd = -1; > if (net->nc->info->type == NET_CLIENT_DRIVER_TAP) { > while (file.index-- > 0) { > + if (!virtio_queue_enabled(dev, net->dev.vq_index + > + file.index)) { > + /* Queue might not be ready for start */ > + continue; > + } > int r = vhost_net_set_backend(&net->dev, &file); > assert(r >= 0); > } > diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c > index 2626a895cb..28056a7ef7 100644 > --- a/hw/virtio/virtio.c > +++ b/hw/virtio/virtio.c > @@ -2318,6 +2318,11 @@ hwaddr virtio_queue_get_desc_addr(VirtIODevice *vdev, int n) > return vdev->vq[n].vring.desc; > } > > +bool virtio_queue_enabled(VirtIODevice *vdev, int n) > +{ > + return virtio_queue_get_desc_addr(vdev, n) != 0; > +} > + > hwaddr virtio_queue_get_avail_addr(VirtIODevice *vdev, int n) > { > return vdev->vq[n].vring.avail; > diff --git a/include/hw/virtio/virtio.h b/include/hw/virtio/virtio.h > index ce9516236a..7140381e3a 100644 > --- a/include/hw/virtio/virtio.h > +++ b/include/hw/virtio/virtio.h > @@ -282,6 +282,7 @@ typedef struct VirtIORNGConf VirtIORNGConf; > VIRTIO_F_IOMMU_PLATFORM, false) > > hwaddr virtio_queue_get_desc_addr(VirtIODevice *vdev, int n); > +bool virtio_queue_enabled(VirtIODevice *vdev, int n); > hwaddr virtio_queue_get_avail_addr(VirtIODevice *vdev, int n); > hwaddr virtio_queue_get_used_addr(VirtIODevice *vdev, int n); > hwaddr virtio_queue_get_desc_size(VirtIODevice *vdev, int n);
On Tue, Apr 09, 2019 at 01:58:50PM +0800, Jason Wang wrote: > We used to set backend unconditionally, this won't work for some > guests (e.g windows driver) who may not initialize all virtqueues. For > kernel backend, this will fail since it may try to validate the rings > during setting backend. > > Fixing this by simply skipping the backend set when we find desc is > not ready. > > Signed-off-by: Jason Wang <jasowang@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> > --- > Changes from V1: > - introduce a generic helper to check if a virtqueue is enabled > --- > hw/net/vhost_net.c | 10 ++++++++++ > hw/virtio/virtio.c | 5 +++++ > include/hw/virtio/virtio.h | 1 + > 3 files changed, 16 insertions(+) > > diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c > index be3cc88370..a6b719035c 100644 > --- a/hw/net/vhost_net.c > +++ b/hw/net/vhost_net.c > @@ -244,6 +244,11 @@ static int vhost_net_start_one(struct vhost_net *net, > qemu_set_fd_handler(net->backend, NULL, NULL, NULL); > file.fd = net->backend; > for (file.index = 0; file.index < net->dev.nvqs; ++file.index) { > + if (!virtio_queue_enabled(dev, net->dev.vq_index + > + file.index)) { > + /* Queue might not be ready for start */ > + continue; > + } > r = vhost_net_set_backend(&net->dev, &file); > if (r < 0) { > r = -errno; > @@ -256,6 +261,11 @@ fail: > file.fd = -1; > if (net->nc->info->type == NET_CLIENT_DRIVER_TAP) { > while (file.index-- > 0) { > + if (!virtio_queue_enabled(dev, net->dev.vq_index + > + file.index)) { > + /* Queue might not be ready for start */ > + continue; > + } > int r = vhost_net_set_backend(&net->dev, &file); > assert(r >= 0); > } > diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c > index 2626a895cb..28056a7ef7 100644 > --- a/hw/virtio/virtio.c > +++ b/hw/virtio/virtio.c > @@ -2318,6 +2318,11 @@ hwaddr virtio_queue_get_desc_addr(VirtIODevice *vdev, int n) > return vdev->vq[n].vring.desc; > } > > +bool virtio_queue_enabled(VirtIODevice *vdev, int n) > +{ > + return virtio_queue_get_desc_addr(vdev, n) != 0; > +} > + > hwaddr virtio_queue_get_avail_addr(VirtIODevice *vdev, int n) > { > return vdev->vq[n].vring.avail; > diff --git a/include/hw/virtio/virtio.h b/include/hw/virtio/virtio.h > index ce9516236a..7140381e3a 100644 > --- a/include/hw/virtio/virtio.h > +++ b/include/hw/virtio/virtio.h > @@ -282,6 +282,7 @@ typedef struct VirtIORNGConf VirtIORNGConf; > VIRTIO_F_IOMMU_PLATFORM, false) > > hwaddr virtio_queue_get_desc_addr(VirtIODevice *vdev, int n); > +bool virtio_queue_enabled(VirtIODevice *vdev, int n); > hwaddr virtio_queue_get_avail_addr(VirtIODevice *vdev, int n); > hwaddr virtio_queue_get_used_addr(VirtIODevice *vdev, int n); > hwaddr virtio_queue_get_desc_size(VirtIODevice *vdev, int n); > -- > 2.19.1
On 2019/4/24 下午9:51, Michael S. Tsirkin wrote: > On Tue, Apr 09, 2019 at 01:58:50PM +0800, Jason Wang wrote: >> We used to set backend unconditionally, this won't work for some >> guests (e.g windows driver) who may not initialize all virtqueues. For >> kernel backend, this will fail since it may try to validate the rings >> during setting backend. >> >> Fixing this by simply skipping the backend set when we find desc is >> not ready. >> >> Signed-off-by: Jason Wang<jasowang@redhat.com> > Reviewed-by: Michael S. Tsirkin<mst@redhat.com> > Applied. Thanks
diff --git a/hw/net/vhost_net.c b/hw/net/vhost_net.c index be3cc88370..a6b719035c 100644 --- a/hw/net/vhost_net.c +++ b/hw/net/vhost_net.c @@ -244,6 +244,11 @@ static int vhost_net_start_one(struct vhost_net *net, qemu_set_fd_handler(net->backend, NULL, NULL, NULL); file.fd = net->backend; for (file.index = 0; file.index < net->dev.nvqs; ++file.index) { + if (!virtio_queue_enabled(dev, net->dev.vq_index + + file.index)) { + /* Queue might not be ready for start */ + continue; + } r = vhost_net_set_backend(&net->dev, &file); if (r < 0) { r = -errno; @@ -256,6 +261,11 @@ fail: file.fd = -1; if (net->nc->info->type == NET_CLIENT_DRIVER_TAP) { while (file.index-- > 0) { + if (!virtio_queue_enabled(dev, net->dev.vq_index + + file.index)) { + /* Queue might not be ready for start */ + continue; + } int r = vhost_net_set_backend(&net->dev, &file); assert(r >= 0); } diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c index 2626a895cb..28056a7ef7 100644 --- a/hw/virtio/virtio.c +++ b/hw/virtio/virtio.c @@ -2318,6 +2318,11 @@ hwaddr virtio_queue_get_desc_addr(VirtIODevice *vdev, int n) return vdev->vq[n].vring.desc; } +bool virtio_queue_enabled(VirtIODevice *vdev, int n) +{ + return virtio_queue_get_desc_addr(vdev, n) != 0; +} + hwaddr virtio_queue_get_avail_addr(VirtIODevice *vdev, int n) { return vdev->vq[n].vring.avail; diff --git a/include/hw/virtio/virtio.h b/include/hw/virtio/virtio.h index ce9516236a..7140381e3a 100644 --- a/include/hw/virtio/virtio.h +++ b/include/hw/virtio/virtio.h @@ -282,6 +282,7 @@ typedef struct VirtIORNGConf VirtIORNGConf; VIRTIO_F_IOMMU_PLATFORM, false) hwaddr virtio_queue_get_desc_addr(VirtIODevice *vdev, int n); +bool virtio_queue_enabled(VirtIODevice *vdev, int n); hwaddr virtio_queue_get_avail_addr(VirtIODevice *vdev, int n); hwaddr virtio_queue_get_used_addr(VirtIODevice *vdev, int n); hwaddr virtio_queue_get_desc_size(VirtIODevice *vdev, int n);
We used to set backend unconditionally, this won't work for some guests (e.g windows driver) who may not initialize all virtqueues. For kernel backend, this will fail since it may try to validate the rings during setting backend. Fixing this by simply skipping the backend set when we find desc is not ready. Signed-off-by: Jason Wang <jasowang@redhat.com> --- Changes from V1: - introduce a generic helper to check if a virtqueue is enabled --- hw/net/vhost_net.c | 10 ++++++++++ hw/virtio/virtio.c | 5 +++++ include/hw/virtio/virtio.h | 1 + 3 files changed, 16 insertions(+)