[PATCH 6/6] net/tls: implement ->read_sock()

[Sagi Grimberg]

On 7/21/23 17:35, Hannes Reinecke wrote:
> Implement ->read_sock() function for use with nvme-tcp.
> 
> Signed-off-by: Hannes Reinecke <hare at suse.de>
> Reviewed-by: Sagi Grimberg <sagi at grimberg.me>
> Cc: Boris Pismenny <boris.pismenny at gmail.com>
> Cc: Jakub Kicinski <kuba at kernel.org>
> Cc: netdev at vger.kernel.org
> ---
>   net/tls/tls.h      |  2 ++
>   net/tls/tls_main.c |  2 ++
>   net/tls/tls_sw.c   | 89 ++++++++++++++++++++++++++++++++++++++++++++++
>   3 files changed, 93 insertions(+)
> 
> diff --git a/net/tls/tls.h b/net/tls/tls.h
> index 86cef1c68e03..7e4d45537deb 100644
> --- a/net/tls/tls.h
> +++ b/net/tls/tls.h
> @@ -110,6 +110,8 @@ bool tls_sw_sock_is_readable(struct sock *sk);
>   ssize_t tls_sw_splice_read(struct socket *sock, loff_t *ppos,
>   			   struct pipe_inode_info *pipe,
>   			   size_t len, unsigned int flags);
> +int tls_sw_read_sock(struct sock *sk, read_descriptor_t *desc,
> +		     sk_read_actor_t read_actor);
>   
>   int tls_device_sendmsg(struct sock *sk, struct msghdr *msg, size_t size);
>   void tls_device_splice_eof(struct socket *sock);
> diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c
> index b6896126bb92..7dbb8cd8f809 100644
> --- a/net/tls/tls_main.c
> +++ b/net/tls/tls_main.c
> @@ -962,10 +962,12 @@ static void build_proto_ops(struct proto_ops ops[TLS_NUM_CONFIG][TLS_NUM_CONFIG]
>   	ops[TLS_BASE][TLS_SW  ] = ops[TLS_BASE][TLS_BASE];
>   	ops[TLS_BASE][TLS_SW  ].splice_read	= tls_sw_splice_read;
>   	ops[TLS_BASE][TLS_SW  ].poll		= tls_sk_poll;
> +	ops[TLS_BASE][TLS_SW  ].read_sock	= tls_sw_read_sock;
>   
>   	ops[TLS_SW  ][TLS_SW  ] = ops[TLS_SW  ][TLS_BASE];
>   	ops[TLS_SW  ][TLS_SW  ].splice_read	= tls_sw_splice_read;
>   	ops[TLS_SW  ][TLS_SW  ].poll		= tls_sk_poll;
> +	ops[TLS_SW  ][TLS_SW  ].read_sock	= tls_sw_read_sock;
>   
>   #ifdef CONFIG_TLS_DEVICE
>   	ops[TLS_HW  ][TLS_BASE] = ops[TLS_BASE][TLS_BASE];
> diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c
> index d0636ea13009..f7ffbe7620cb 100644
> --- a/net/tls/tls_sw.c
> +++ b/net/tls/tls_sw.c
> @@ -2202,6 +2202,95 @@ ssize_t tls_sw_splice_read(struct socket *sock,  loff_t *ppos,
>   	goto splice_read_end;
>   }
>   
> +int tls_sw_read_sock(struct sock *sk, read_descriptor_t *desc,
> +		     sk_read_actor_t read_actor)
> +{
> +	struct tls_context *tls_ctx = tls_get_ctx(sk);
> +	struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx);
> +	struct strp_msg *rxm = NULL;
> +	struct sk_buff *skb = NULL;
> +	struct sk_psock *psock;
> +	struct tls_msg *tlm;
> +	ssize_t copied = 0;
> +	int err, used;
> +
> +	psock = sk_psock_get(sk);
> +	if (psock) {
> +		sk_psock_put(sk, psock);
> +		return -EINVAL;
> +	}
> +	err = tls_rx_reader_acquire(sk, ctx, true);
> +	if (err < 0)
> +		return err;
> +
> +	/* If crypto failed the connection is broken */
> +	err = ctx->async_wait.err;
> +	if (err)
> +		goto read_sock_end;
> +
> +	do {
> +		if (!skb_queue_empty(&ctx->rx_list)) {
> +			skb = __skb_dequeue(&ctx->rx_list);
> +			rxm = strp_msg(skb);
> +			tlm = tls_msg(skb);
> +		} else {
> +			struct tls_decrypt_arg darg;
> +
> +			err = tls_rx_rec_wait(sk, NULL, true, true);
> +			if (err <= 0)
> +				goto read_sock_end;
> +
> +			memset(&darg.inargs, 0, sizeof(darg.inargs));
> +
> +			rxm = strp_msg(tls_strp_msg(ctx));
> +			tlm = tls_msg(tls_strp_msg(ctx));
> +
> +			err = tls_rx_one_record(sk, NULL, &darg);
> +			if (err < 0) {
> +				tls_err_abort(sk, -EBADMSG);
> +				goto read_sock_end;
> +			}
> +
> +			sk_flush_backlog(sk);

Question,
Based on Jakub's comment, the flush is better spaced out.
Why not just do it once at the end? Or alternatively,
call tls_read_flush_backlog() ? Or just count by hand
every 4 records or 128K (and once in the end)?

I don't really know what would be the impact though, but
you are effectively releasing and re-acquiring the socket
flushing the backlog every record...