[v1] vdpa/mlx5: Pre-create HW VQs to reduce LM downtime

[PATCH vhost 23/23] vdpa/mlx5: Don't enable non-active VQs in .set_vq_ready()

Posted by Dragos Tatulea 1 year, 6 months ago

VQ indices in the range [cur_num_qps, max_vqs) represent queues that
have not yet been activated. .set_vq_ready should not activate these
VQs.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Cosmin Ratiu <cratiu@nvidia.com>
---
 drivers/vdpa/mlx5/net/mlx5_vnet.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c
index 1a5ee0d2b47f..a969a7f105a6 100644
--- a/drivers/vdpa/mlx5/net/mlx5_vnet.c
+++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c
@@ -1575,6 +1575,9 @@ static int resume_vq(struct mlx5_vdpa_net *ndev, struct mlx5_vdpa_virtqueue *mvq
 	if (!mvq->initialized)
 		return 0;
 
+	if (mvq->index >= ndev->cur_num_vqs)
+		return 0;
+
 	switch (mvq->fw_state) {
 	case MLX5_VIRTIO_NET_Q_OBJECT_STATE_INIT:
 		/* Due to a FW quirk we need to modify the VQ fields first then change state.

-- 
2.45.1

Re: [PATCH vhost 23/23] vdpa/mlx5: Don't enable non-active VQs in .set_vq_ready()

Posted by Eugenio Perez Martin 1 year, 6 months ago

On Mon, Jun 17, 2024 at 5:09 PM Dragos Tatulea <dtatulea@nvidia.com> wrote:
>
> VQ indices in the range [cur_num_qps, max_vqs) represent queues that
> have not yet been activated. .set_vq_ready should not activate these
> VQs.
>
> Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
> Reviewed-by: Cosmin Ratiu <cratiu@nvidia.com>

Acked-by: Eugenio Pérez <eperezma@redhat.com>

> ---
>  drivers/vdpa/mlx5/net/mlx5_vnet.c | 3 +++
>  1 file changed, 3 insertions(+)
>
> diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c
> index 1a5ee0d2b47f..a969a7f105a6 100644
> --- a/drivers/vdpa/mlx5/net/mlx5_vnet.c
> +++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c
> @@ -1575,6 +1575,9 @@ static int resume_vq(struct mlx5_vdpa_net *ndev, struct mlx5_vdpa_virtqueue *mvq
>         if (!mvq->initialized)
>                 return 0;
>
> +       if (mvq->index >= ndev->cur_num_vqs)
> +               return 0;
> +
>         switch (mvq->fw_state) {
>         case MLX5_VIRTIO_NET_Q_OBJECT_STATE_INIT:
>                 /* Due to a FW quirk we need to modify the VQ fields first then change state.
>
> --
> 2.45.1
>