Re: [PATCH vhost 12/23] vdpa/mlx5: Start off rqt_size with max VQPs
On Mon, Jun 17, 2024 at 5:09 PM Dragos Tatulea wrote:
>
> Currently rqt_size is initialized during device flag configuration.
> That's because it is the earliest moment when device knows if MQ
> (multi queue) is on or off.
>
> Shift this configuration earlier to device creation time. This implies
> that non-MQ devices will have a larger RQT size. But the configuration
> will still be correct.
>
> This is done in preparation for the pre-creation of hardware virtqueues
> at device add time. When that change will be added, RQT will be created
> at device creation time so it needs to be initialized to its max size.
>
> Signed-off-by: Dragos Tatulea
> Reviewed-by: Cosmin Ratiu
Acked-by: Eugenio Pérez
> ---
> drivers/vdpa/mlx5/net/mlx5_vnet.c | 10 +-
> 1 file changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c
> b/drivers/vdpa/mlx5/net/mlx5_vnet.c
> index 1181e0ac3671..0201c6fe61e1 100644
> --- a/drivers/vdpa/mlx5/net/mlx5_vnet.c
> +++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c
> @@ -2731,10 +2731,6 @@ static int mlx5_vdpa_set_driver_features(struct
> vdpa_device *vdev, u64 features)
> return err;
>
> ndev->mvdev.actual_features = features & ndev->mvdev.mlx_features;
> - if (ndev->mvdev.actual_features & BIT_ULL(VIRTIO_NET_F_MQ))
> - ndev->rqt_size = mlx5vdpa16_to_cpu(mvdev,
> ndev->config.max_virtqueue_pairs);
> - else
> - ndev->rqt_size = 1;
>
> /* Interested in changes of vq features only. */
> if (get_features(old_features) !=
> get_features(mvdev->actual_features)) {
> @@ -3719,8 +3715,12 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev
> *v_mdev, const char *name,
> goto err_alloc;
> }
>
> - if (device_features & BIT_ULL(VIRTIO_NET_F_MQ))
> + if (device_features & BIT_ULL(VIRTIO_NET_F_MQ)) {
> config->max_virtqueue_pairs = cpu_to_mlx5vdpa16(mvdev,
> max_vqs / 2);
> + ndev->rqt_size = max_vqs / 2;
> + } else {
> + ndev->rqt_size = 1;
> + }
>
> ndev->mvdev.mlx_features = device_features;
> mvdev->vdev.dma_dev = &mdev->pdev->dev;
>
> --
> 2.45.1
>
[PATCH vhost 12/23] vdpa/mlx5: Start off rqt_size with max VQPs
Currently rqt_size is initialized during device flag configuration.
That's because it is the earliest moment when device knows if MQ
(multi queue) is on or off.
Shift this configuration earlier to device creation time. This implies
that non-MQ devices will have a larger RQT size. But the configuration
will still be correct.
This is done in preparation for the pre-creation of hardware virtqueues
at device add time. When that change will be added, RQT will be created
at device creation time so it needs to be initialized to its max size.
Signed-off-by: Dragos Tatulea
Reviewed-by: Cosmin Ratiu
---
drivers/vdpa/mlx5/net/mlx5_vnet.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c
b/drivers/vdpa/mlx5/net/mlx5_vnet.c
index 1181e0ac3671..0201c6fe61e1 100644
--- a/drivers/vdpa/mlx5/net/mlx5_vnet.c
+++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c
@@ -2731,10 +2731,6 @@ static int mlx5_vdpa_set_driver_features(struct
vdpa_device *vdev, u64 features)
return err;
ndev->mvdev.actual_features = features & ndev->mvdev.mlx_features;
- if (ndev->mvdev.actual_features & BIT_ULL(VIRTIO_NET_F_MQ))
- ndev->rqt_size = mlx5vdpa16_to_cpu(mvdev,
ndev->config.max_virtqueue_pairs);
- else
- ndev->rqt_size = 1;
/* Interested in changes of vq features only. */
if (get_features(old_features) != get_features(mvdev->actual_features))
{
@@ -3719,8 +3715,12 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev
*v_mdev, const char *name,
goto err_alloc;
}
- if (device_features & BIT_ULL(VIRTIO_NET_F_MQ))
+ if (device_features & BIT_ULL(VIRTIO_NET_F_MQ)) {
config->max_virtqueue_pairs = cpu_to_mlx5vdpa16(mvdev, max_vqs
/ 2);
+ ndev->rqt_size = max_vqs / 2;
+ } else {
+ ndev->rqt_size = 1;
+ }
ndev->mvdev.mlx_features = device_features;
mvdev->vdev.dma_dev = &mdev->pdev->dev;
--
2.45.1
