Re: [PATCH RFC net-next 7/7] netdev_features: convert NETIF_F_FCOE_MTU to IFF_FCOE_MTU

[Jakub Kicinski]

On Fri,  5 Apr 2024 15:37:31 +0200 Alexander Lobakin wrote:
> Ability to handle maximum FCoE frames of 2158 bytes can never be changed
> and thus more of an attribute, not a toggleable feature.
> Move it from netdev_features_t to netdev_priv_flags and free one more
> feature bit.

> @@ -1700,6 +1701,7 @@ enum netdev_priv_flags {
>  	IFF_NETNS_LOCAL			= BIT_ULL(35),
>  	IFF_HIGHDMA			= BIT_ULL(36),
>  	IFF_VLAN_CHALLENGED		= BIT_ULL(37),
> +	IFF_FCOE_MTU			= BIT_ULL(38),
>  	IFF_LOGICAL			= IFF_NO_QUEUE | IFF_LLTX,
>  	IFF_ONE_FOR_ALL			= IFF_HIGHDMA | IFF_VLAN_CHALLENGED,
>  	IFF_ALL_FOR_ALL			= IFF_XMIT_DST_RELEASE |

Any reason not to make it a bitfield? I haven't looked at the longer
patches but this one seems to be used like a basic bool.

> diff --git a/net/ethtool/common.c b/net/ethtool/common.c
> index 2de4dd5a30de..71e36e1a1b15 100644
> --- a/net/ethtool/common.c
> +++ b/net/ethtool/common.c
> @@ -47,7 +47,6 @@ const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN] = {
>  
>  	[NETIF_F_FCOE_CRC_BIT] =         "tx-checksum-fcoe-crc",
>  	[NETIF_F_SCTP_CRC_BIT] =        "tx-checksum-sctp",
> -	[NETIF_F_FCOE_MTU_BIT] =         "fcoe-mtu",


But this definitely _is_ a uAPI change, right?
Some analysis why this is fine and why avoiding the problem isn't worth
it in the cover letter would be great.

>  	[NETIF_F_NTUPLE_BIT] =           "rx-ntuple-filter",
>  	[NETIF_F_RXHASH_BIT] =           "rx-hashing",
>  	[NETIF_F_RXCSUM_BIT] =           "rx-checksum",