Re: [PATCH mptcp-next 2/2] Squash to "mptcp: multi subflows subflow_push_pending" v5

[Mat Martineau <mathew.j.martineau@linux.intel.com>]

On Sun, 9 Oct 2022, Geliang Tang wrote:

> Update the first chunk of data pushing.
>
> Signed-off-by: Geliang Tang <geliang.tang@suse.com>
> ---
> net/mptcp/protocol.c | 18 +++++++-----------
> 1 file changed, 7 insertions(+), 11 deletions(-)
>
> diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
> index ff2c1bbf4484..acc220bfb34e 100644
> --- a/net/mptcp/protocol.c
> +++ b/net/mptcp/protocol.c
> @@ -1614,17 +1614,8 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk,
> 		/* check for a different subflow usage only after
> 		 * spooling the first chunk of data
> 		 */
> -		if (first) {
> -			ret = __subflow_push_pending(sk, ssk, &info);
> -			if (ret <= 0) {
> -				if (ret == -EAGAIN)
> -					goto again;
> -				break;
> -			}
> -			first = false;
> -			msk->last_snd = ssk;
> -			continue;
> -		}
> +		if (first)
> +			goto push;

Is the only goal here to reduce duplicate code? It looks like the 
functionality is equivalent, but I'd rather not use 'goto' to jump into 
the middle of a loop.

The 'Squash to "mptcp: simplify push_pending"' looks fine to include in 
v6.

- Mat

>
> 		if (mptcp_sched_get_send(msk))
> 			goto out;
> @@ -1643,6 +1634,7 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk,
> 					continue;
> 				}
>
> +push:
> 				ret = __subflow_push_pending(sk, ssk, &info);
> 				if (ret <= 0) {
> 					if (ret == -EAGAIN)
> @@ -1650,6 +1642,10 @@ static void __mptcp_subflow_push_pending(struct sock *sk, struct sock *ssk,
> 					goto out;
> 				}
> 				msk->last_snd = ssk;
> +				if (first) {
> +					first = false;
> +					goto again;
> +				}
> 				mptcp_subflow_set_scheduled(subflow, false);
> 			}
> 		}
> -- 
> 2.35.3
>
>
>

--
Mat Martineau
Intel