Re: [PATCH net-next 5/9] net-timestamp: ready to turn on the button to generate tx timestamps

[Vadim Fedorenko <vadim.fedorenko@xxxxxxxxx>]

On 08/10/2024 10:51, Jason Xing wrote:
> From: Jason Xing <kernelxing@xxxxxxxxxxx>
>
> Once we set BPF_SOCK_OPS_TX_TIMESTAMP_OPT_CB_FLAG flag here, there
> are three points in the previous patches where generating timestamps
> works. Let us make the basic bpf mechanism for timestamping feature
>   work finally.
>
> We can use like this as a simple example in bpf program:
> __section("sockops")
>
> case BPF_SOCK_OPS_TX_TIMESTAMP_OPT_CB:
> 	dport = bpf_ntohl(skops->remote_port);
> 	sport = skops->local_port;
> 	skops->reply = SOF_TIMESTAMPING_TX_SCHED;
> 	bpf_sock_ops_cb_flags_set(skops, BPF_SOCK_OPS_TX_TIMESTAMP_OPT_CB_FLAG);
> case BPF_SOCK_OPS_TS_SCHED_OPT_CB:
> 	bpf_printk(...);
>
> Signed-off-by: Jason Xing <kernelxing@xxxxxxxxxxx>
> ---
>   include/uapi/linux/bpf.h       |  8 ++++++++
>   net/ipv4/tcp.c                 | 27 ++++++++++++++++++++++++++-
>   tools/include/uapi/linux/bpf.h |  8 ++++++++
>   3 files changed, 42 insertions(+), 1 deletion(-)
>
> diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
> index 1b478ec18ac2..6bf3f2892776 100644
> --- a/include/uapi/linux/bpf.h
> +++ b/include/uapi/linux/bpf.h
> @@ -7034,6 +7034,14 @@ enum {
>   					 * feature is on. It indicates the
>   					 * recorded timestamp.
>   					 */
> +	BPF_SOCK_OPS_TX_TS_OPT_CB,	/* Called when the last skb from
> +					 * sendmsg is going to push when
> +					 * SO_TIMESTAMPING feature is on.
> +					 * Let user have a chance to switch
> +					 * on BPF_SOCK_OPS_TX_TIMESTAMPING_OPT_CB_FLAG
> +					 * flag for other three tx timestamp
> +					 * use.
> +					 */
>   };
>   
>   /* List of TCP states. There is a build check in net/ipv4/tcp.c to detect
> diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
> index 82cc4a5633ce..ddf4089779b5 100644
> --- a/net/ipv4/tcp.c
> +++ b/net/ipv4/tcp.c
> @@ -477,12 +477,37 @@ void tcp_init_sock(struct sock *sk)
>   }
>   EXPORT_SYMBOL(tcp_init_sock);
>   
> +static u32 bpf_tcp_tx_timestamp(struct sock *sk)
> +{
> +	u32 flags;
> +
> +	flags = tcp_call_bpf(sk, BPF_SOCK_OPS_TX_TS_OPT_CB, 0, NULL);
> +	if (flags <= 0)
> +		return 0;
> +
> +	if (flags & ~SOF_TIMESTAMPING_MASK)
> +		return 0;
> +
> +	if (!(flags & SOF_TIMESTAMPING_TX_RECORD_MASK))
> +		return 0;
> +
> +	return flags;
> +}
> +
>   static void tcp_tx_timestamp(struct sock *sk, struct sockcm_cookie *sockc)
>   {
>   	struct sk_buff *skb = tcp_write_queue_tail(sk);
>   	u32 tsflags = sockc->tsflags;
> +	u32 flags;
> +
> +	if (!skb)
> +		return;
> +
> +	flags = bpf_tcp_tx_timestamp(sk);
> +	if (flags)
> +		tsflags = flags;

In this case it's impossible to clear timestamping flags from bpf
program, but it may be very useful. Consider providing flags from
socket cookie to the program or maybe add an option to combine them?

>   
> -	if (tsflags && skb) {
> +	if (tsflags) {
>   		struct skb_shared_info *shinfo = skb_shinfo(skb);
>   		struct tcp_skb_cb *tcb = TCP_SKB_CB(skb);
>   
> diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
> index fc9b94de19f2..d3bf538846da 100644
> --- a/tools/include/uapi/linux/bpf.h
> +++ b/tools/include/uapi/linux/bpf.h
> @@ -7033,6 +7033,14 @@ enum {
>   					 * feature is on. It indicates the
>   					 * recorded timestamp.
>   					 */
> +	BPF_SOCK_OPS_TX_TS_OPT_CB,	/* Called when the last skb from
> +					 * sendmsg is going to push when
> +					 * SO_TIMESTAMPING feature is on.
> +					 * Let user have a chance to switch
> +					 * on BPF_SOCK_OPS_TX_TIMESTAMPING_OPT_CB_FLAG
> +					 * flag for other three tx timestamp
> +					 * use.
> +					 */
>   };
>   
>   /* List of TCP states. There is a build check in net/ipv4/tcp.c to detect