[PATCH vhost v6 5/6] virtio: vring_new_virtqueue(): pass struct instead of multi parameters

[Jason Wang]

On Wed, Mar 27, 2024 at 5:58 PM Xuan Zhuo <xuanzhuo at linux.alibaba.com> wrote:
>
> Now, we pass multi parameters to vring_new_virtqueue. These parameters
> may from transport or from driver.
>
> vring_new_virtqueue is called by many places.
> Every time, we try to add a new parameter, that is difficult.
>
> If parameters from the driver, that should directly be passed to vring.
> Then the vring can access the config from driver directly.
>
> If parameters from the transport, we squish the parameters to a
> structure. That will be helpful to add new parameter.
>
> Signed-off-by: Xuan Zhuo <xuanzhuo at linux.alibaba.com>
> Reviewed-by: Ilpo Järvinen <ilpo.jarvinen at linux.intel.com>
> ---
>  drivers/platform/mellanox/mlxbf-tmfifo.c | 12 ++++---
>  drivers/remoteproc/remoteproc_virtio.c   | 11 ++++---
>  drivers/virtio/virtio_ring.c             | 29 +++++++++++-----
>  include/linux/virtio_ring.h              | 42 +++++++++++++++++++-----
>  tools/virtio/virtio_test.c               |  4 +--
>  tools/virtio/vringh_test.c               | 28 ++++++++--------
>  6 files changed, 84 insertions(+), 42 deletions(-)
>
> diff --git a/drivers/platform/mellanox/mlxbf-tmfifo.c b/drivers/platform/mellanox/mlxbf-tmfifo.c
> index 4252388f52a2..d2e871fad8b4 100644
> --- a/drivers/platform/mellanox/mlxbf-tmfifo.c
> +++ b/drivers/platform/mellanox/mlxbf-tmfifo.c
> @@ -1059,6 +1059,7 @@ static int mlxbf_tmfifo_virtio_find_vqs(struct virtio_device *vdev,
>                                         struct virtio_vq_config *cfg)
>  {
>         struct mlxbf_tmfifo_vdev *tm_vdev = mlxbf_vdev_to_tmfifo(vdev);
> +       struct vq_transport_config tp_cfg = {};
>         struct virtqueue **vqs = cfg->vqs;
>         struct mlxbf_tmfifo_vring *vring;
>         unsigned int nvqs = cfg->nvqs;
> @@ -1078,10 +1079,13 @@ static int mlxbf_tmfifo_virtio_find_vqs(struct virtio_device *vdev,
>                 /* zero vring */
>                 size = vring_size(vring->num, vring->align);
>                 memset(vring->va, 0, size);
> -               vq = vring_new_virtqueue(i, vring->num, vring->align, vdev,
> -                                        false, false, vring->va,
> -                                        mlxbf_tmfifo_virtio_notify,
> -                                        cfg->callbacks[i], cfg->names[i]);
> +
> +               tp_cfg.num = vring->num;
> +               tp_cfg.vring_align = vring->align;
> +               tp_cfg.weak_barriers = false;
> +               tp_cfg.notify = mlxbf_tmfifo_virtio_notify;
> +
> +               vq = vring_new_virtqueue(vdev, i, vring->va, &tp_cfg, cfg);
>                 if (!vq) {
>                         dev_err(&vdev->dev, "vring_new_virtqueue failed\n");
>                         ret = -ENOMEM;
> diff --git a/drivers/remoteproc/remoteproc_virtio.c b/drivers/remoteproc/remoteproc_virtio.c
> index 489fea1d41c0..2319c2007833 100644
> --- a/drivers/remoteproc/remoteproc_virtio.c
> +++ b/drivers/remoteproc/remoteproc_virtio.c
> @@ -106,6 +106,7 @@ static struct virtqueue *rp_find_vq(struct virtio_device *vdev,
>  {
>         struct rproc_vdev *rvdev = vdev_to_rvdev(vdev);
>         struct rproc *rproc = vdev_to_rproc(vdev);
> +       struct vq_transport_config tp_cfg;

Should we zero this structure?

Thanks