Re: [PATCH net-next] net: tls: make the offload check helper take skb not socket

[Maxim Mikityanskiy <maxtram95@gmail.com>]

On Tue, 13 Jun 2023 at 13:50:06 -0700, Jakub Kicinski wrote:
> All callers of tls_is_sk_tx_device_offloaded() currently do
> an equivalent of:
> 
>  if (skb->sk && tls_is_skb_tx_device_offloaded(skb->sk))
> 
> Have the helper accept skb and do the skb->sk check locally.
> Two drivers have local static inlines with similar wrappers
> already.
> 
> While at it change the ifdef condition to TLS_DEVICE.
> Only TLS_DEVICE selects SOCK_VALIDATE_XMIT, so the two are
> equivalent. This makes removing the duplicated IS_ENABLED()
> check in funeth more obviously correct.
> 
> Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Acked-by: Maxim Mikityanskiy <maxtram95@gmail.com>

Nice cleanup, thanks!

[...]

> diff --git a/include/net/tls.h b/include/net/tls.h
> index b7d0f1e3058b..5e71dd3df8ca 100644
> --- a/include/net/tls.h
> +++ b/include/net/tls.h
> @@ -370,10 +370,12 @@ struct sk_buff *
>  tls_validate_xmit_skb_sw(struct sock *sk, struct net_device *dev,
>  			 struct sk_buff *skb);
>  
> -static inline bool tls_is_sk_tx_device_offloaded(struct sock *sk)
> +static inline bool tls_is_skb_tx_device_offloaded(const struct sk_buff *skb)
>  {
> -#ifdef CONFIG_SOCK_VALIDATE_XMIT
> -	return sk_fullsock(sk) &&
> +#ifdef CONFIG_TLS_DEVICE
> +	struct sock *sk = skb->sk;
> +
> +	return sk && sk_fullsock(sk) &&
>  	       (smp_load_acquire(&sk->sk_validate_xmit_skb) ==
>  	       &tls_validate_xmit_skb);
>  #else

After this change, the only usage of CONFIG_SOCK_VALIDATE_XMIT remains
in sk_validate_xmit_skb, which has #ifdef CONFIG_TLS_DEVICE inside
#ifdef CONFIG_SOCK_VALIDATE_XMIT. If feels a little bit weird, given
that both defines always have the same value, but maybe it's OK if we
consider that more users can start using sk_validate_xmit_skb in the
future.