Thread (10 messages) 10 messages, 3 authors, 2026-01-19

Re: [PATCH v2 1/3] vdpa/mlx5: update mlx_features with driver state check

From: Cindy Lu <hidden>
Date: 2026-01-19 02:12:58
Also in: kvm, lkml, virtualization

On Fri, Jan 2, 2026 at 8:09 PM Dragos Tatulea [off-list ref] wrote:

Hi Cindy,

Thanks for your patch!

On 29.12.25 08:16, Cindy Lu wrote:
quoted
Add logic in mlx5_vdpa_set_attr() to ensure the VIRTIO_NET_F_MAC
feature bit is properly set only when the device is not yet in
the DRIVER_OK (running) state.

This makes the MAC address visible in the output of:

 vdpa dev config show -jp

when the device is created without an initial MAC address.

Signed-off-by: Cindy Lu <redacted>
Having a cover letter with the summary, history and links series would
make the review process easier.
quoted
---
 drivers/vdpa/mlx5/net/mlx5_vnet.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)
diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c
index ddaa1366704b..6e42bae7c9a1 100644
--- a/drivers/vdpa/mlx5/net/mlx5_vnet.c
+++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c
@@ -4049,7 +4049,7 @@ static int mlx5_vdpa_set_attr(struct vdpa_mgmt_dev *v_mdev, struct vdpa_device *
      struct mlx5_vdpa_dev *mvdev;
      struct mlx5_vdpa_net *ndev;
      struct mlx5_core_dev *mdev;
-     int err = -EOPNOTSUPP;
+     int err = 0;

      mvdev = to_mvdev(dev);
      ndev = to_mlx5_vdpa_ndev(mvdev);
@@ -4057,13 +4057,22 @@ static int mlx5_vdpa_set_attr(struct vdpa_mgmt_dev *v_mdev, struct vdpa_device *
      config = &ndev->config;

      down_write(&ndev->reslock);
-     if (add_config->mask & (1 << VDPA_ATTR_DEV_NET_CFG_MACADDR)) {
+
+     if (add_config->mask & BIT_ULL(VDPA_ATTR_DEV_NET_CFG_MACADDR)) {
+             if (!(ndev->mvdev.status & VIRTIO_CONFIG_S_DRIVER_OK)) {
+                     ndev->mvdev.mlx_features |= BIT_ULL(VIRTIO_NET_F_MAC);
+             } else {
+                     mlx5_vdpa_warn(mvdev, "device running, skip updating MAC\n");
+                     err = -EBUSY;
+                     goto out;
+             }
              pfmdev = pci_get_drvdata(pci_physfn(mdev->pdev));
              err = mlx5_mpfs_add_mac(pfmdev, config->mac);
              if (!err)
                      ether_addr_copy(config->mac, add_config->net.mac);
      }

+out:
      up_write(&ndev->reslock);
      return err;
 }
The patch itself makes sense. For it you can add:

Reviewed-by: Dragos Tatulea <dtatulea@nvidia.com>
Thanks Dragos, will add this
Thanks,
Dragos
  
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help