On Oct 12, 2022 / 01:24, Yi Zhang wrote:
> The mlx dev will be ignored in has_soft_rdma when it was used as netdev,
> add more filter keywords.
>
> link mlx5_0/1 state ACTIVE physical_state LINK_UP netdev enp33s0f0np0
> link mlx5_1/1 state DOWN physical_state DISABLED netdev enp33s0f1np1
>
> Signed-off-by: Yi Zhang <yi.zhang@xxxxxxxxxx>
> ---
> common/multipath-over-rdma | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/common/multipath-over-rdma b/common/multipath-over-rdma
> index fb820d6..6b9629a 100644
> --- a/common/multipath-over-rdma
> +++ b/common/multipath-over-rdma
> @@ -387,7 +387,7 @@ all_primary_gids() {
> # Check whether or not an rdma_rxe or siw instance has been associated with
> # network interface $1.
> has_soft_rdma() {
> - rdma link | grep -q " netdev $1[[:blank:]]*\$"
> + rdma link | grep -q "link $1.*netdev $1[[:blank:]]*\$"
> }
Question, what value do you expect in $1 here for mlx devices?
To match with the rdma link command output below, 1st $1 should be "mlx5",
and 2nd $1 should be "enp33s0f0np0", but $1 can not be both. ??
link mlx5_0/1 state ACTIVE physical_state LINK_UP netdev enp33s0f0np0
--
Shin'ichiro Kawasaki
