net: introduce SO_MAX_PACING_RATE

As mentioned in commit afe4fd0624 ("pkt_sched: fq: Fair Queue packet
scheduler"), this patch adds a new socket option.

SO_MAX_PACING_RATE offers the application the ability to cap the
rate computed by transport layer. Value is in bytes per second.

u32 val = 1000000;
setsockopt(sockfd, SOL_SOCKET, SO_MAX_PACING_RATE, &val, sizeof(val));

To be effectively paced, a flow must use FQ packet scheduler.

Note that a packet scheduler takes into account the headers for its
computations. The effective payload rate depends on MSS and retransmits
if any.

I chose to make this pacing rate a SOL_SOCKET option instead of a
TCP one because this can be used by other protocols.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Steinar H. Gunderson <sesse@google.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Change-Id: Iea51b6104a5420d5c9ed1ea7382cbd53bdb8f4be

arch/alpha/include/asm/socket.h

@@ -76,6 +76,8 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	44
+
 #ifdef __KERNEL__
 /* O_NONBLOCK clashes with the bits used for socket types.  Therefore we
  * have to define SOCK_NONBLOCK to a different value here.

arch/arm/include/asm/socket.h

@@ -69,4 +69,6 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	44
+
 #endif /* _ASM_SOCKET_H */

arch/avr32/include/asm/socket.h

@@ -69,4 +69,6 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	44
+
 #endif /* __ASM_AVR32_SOCKET_H */

arch/cris/include/asm/socket.h

@@ -71,6 +71,8 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	44
+
 #endif /* _ASM_SOCKET_H */
 
 

arch/frv/include/asm/socket.h

@@ -69,5 +69,7 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	44
+
 #endif /* _ASM_SOCKET_H */
 

arch/h8300/include/asm/socket.h

@@ -69,4 +69,6 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	44
+
 #endif /* _ASM_SOCKET_H */

arch/ia64/include/asm/socket.h

@@ -78,4 +78,6 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	44
+
 #endif /* _ASM_IA64_SOCKET_H */

arch/m32r/include/asm/socket.h

@@ -69,4 +69,6 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	44
+
 #endif /* _ASM_M32R_SOCKET_H */

arch/m68k/include/asm/socket.h

@@ -69,4 +69,6 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	44
+
 #endif /* _ASM_SOCKET_H */

arch/mips/include/asm/socket.h

@@ -89,6 +89,8 @@ To add: #define SO_REUSEPORT 0x0200	/* Allow local address and port reuse.  */
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	44
+
 #ifdef __KERNEL__
 
 /** sock_type - Socket types

arch/mn10300/include/asm/socket.h

@@ -69,4 +69,6 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	44
+
 #endif /* _ASM_SOCKET_H */

arch/parisc/include/asm/socket.h

@@ -68,6 +68,7 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		0x4024
 
+#define SO_MAX_PACING_RATE	0x4025
 
 /* O_NONBLOCK clashes with the bits used for socket types.  Therefore we
  * have to define SOCK_NONBLOCK to a different value here.

arch/powerpc/include/asm/socket.h

@@ -76,4 +76,6 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	44
+
 #endif	/* _ASM_POWERPC_SOCKET_H */

arch/s390/include/asm/socket.h

@@ -77,4 +77,6 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	44
+
 #endif /* _ASM_SOCKET_H */

arch/sparc/include/asm/socket.h

@@ -65,6 +65,7 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		0x0027
 
+#define SO_MAX_PACING_RATE	0x0028
 
 /* Security levels - as per NRL IPv6 - don't actually do anything */
 #define SO_SECURITY_AUTHENTICATION		0x5001

arch/xtensa/include/asm/socket.h

@@ -80,4 +80,6 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	47
+
 #endif	/* _XTENSA_SOCKET_H */

include/asm-generic/socket.h

@@ -72,4 +72,6 @@
 /* Instruct lower device to use last 4-bytes of skb data as FCS */
 #define SO_NOFCS		43
 
+#define SO_MAX_PACING_RATE	44
+
 #endif /* __ASM_GENERIC_SOCKET_H */

include/net/sock.h

@@ -335,6 +335,7 @@ struct sock {
 	int			sk_wmem_queued;
 	gfp_t			sk_allocation;
 	u32			sk_pacing_rate; /* bytes per second */
+	u32			sk_max_pacing_rate;
 	netdev_features_t	sk_route_caps;
 	netdev_features_t	sk_route_nocaps;
 	int			sk_gso_type;

net/core/sock.c

@@ -792,6 +792,12 @@ set_rcvbuf:
 		sock_valbool_flag(sk, SOCK_NOFCS, valbool);
 		break;
 
+	case SO_MAX_PACING_RATE:
+		sk->sk_max_pacing_rate = val;
+		sk->sk_pacing_rate = min(sk->sk_pacing_rate,
+					 sk->sk_max_pacing_rate);
+		break;
+
 	default:
 		ret = -ENOPROTOOPT;
 		break;
@@ -1032,6 +1038,11 @@ int sock_getsockopt(struct socket *sock, int level, int optname,
 	case SO_NOFCS:
 		v.val = !!sock_flag(sk, SOCK_NOFCS);
 		break;
+
+	case SO_MAX_PACING_RATE:
+		v.val = sk->sk_max_pacing_rate;
+		break;
+
 	default:
 		return -ENOPROTOOPT;
 	}
@@ -2109,6 +2120,8 @@ void sock_init_data(struct socket *sock, struct sock *sk)
 
 	sk->sk_stamp = ktime_set(-1L, 0);
 
+	sk->sk_max_pacing_rate = ~0U;
+
 	/*
 	 * Before updating sk_refcnt, we must commit prior changes to memory
 	 * (Documentation/RCU/rculist_nulls.txt for details)

net/ipv4/tcp_input.c

@@ -725,7 +725,7 @@ static void tcp_update_pacing_rate(struct sock *sk)
 	if (tp->srtt > 8 + 2)
 		do_div(rate, tp->srtt);
 
-	sk->sk_pacing_rate = min_t(u64, rate, ~0U);
+	sk->sk_pacing_rate = min_t(u64, rate, sk->sk_max_pacing_rate);
 }
 
 /* Calculate rto without backoff.  This is the second half of Van Jacobson's