@@ -81,4 +81,6 @@
#define SO_SELECT_ERR_QUEUE 45
+#define SO_LL 46
+
#endif /* _UAPI_ASM_SOCKET_H */
@@ -74,4 +74,6 @@
#define SO_SELECT_ERR_QUEUE 45
+#define SO_LL 46
+
#endif /* __ASM_AVR32_SOCKET_H */
@@ -76,6 +76,8 @@
#define SO_SELECT_ERR_QUEUE 45
+#define SO_LL 46
+
#endif /* _ASM_SOCKET_H */
@@ -74,5 +74,7 @@
#define SO_SELECT_ERR_QUEUE 45
+#define SO_LL 46
+
#endif /* _ASM_SOCKET_H */
@@ -74,4 +74,6 @@
#define SO_SELECT_ERR_QUEUE 45
+#define SO_LL 46
+
#endif /* _ASM_SOCKET_H */
@@ -83,4 +83,6 @@
#define SO_SELECT_ERR_QUEUE 45
+#define SO_LL 46
+
#endif /* _ASM_IA64_SOCKET_H */
@@ -74,4 +74,6 @@
#define SO_SELECT_ERR_QUEUE 45
+#define SO_LL 46
+
#endif /* _ASM_M32R_SOCKET_H */
@@ -92,4 +92,6 @@
#define SO_SELECT_ERR_QUEUE 45
+#define SO_LL 46
+
#endif /* _UAPI_ASM_SOCKET_H */
@@ -74,4 +74,6 @@
#define SO_SELECT_ERR_QUEUE 45
+#define SO_LL 46
+
#endif /* _ASM_SOCKET_H */
@@ -73,6 +73,8 @@
#define SO_SELECT_ERR_QUEUE 0x4026
+#define SO_LL 0x4027
+
/* O_NONBLOCK clashes with the bits used for socket types. Therefore we
* have to define SOCK_NONBLOCK to a different value here.
*/
@@ -81,4 +81,6 @@
#define SO_SELECT_ERR_QUEUE 45
+#define SO_LL 46
+
#endif /* _ASM_POWERPC_SOCKET_H */
@@ -80,4 +80,6 @@
#define SO_SELECT_ERR_QUEUE 45
+#define SO_LL 46
+
#endif /* _ASM_SOCKET_H */
@@ -70,6 +70,8 @@
#define SO_SELECT_ERR_QUEUE 0x0029
+#define SO_LL 0x0030
+
/* Security levels - as per NRL IPv6 - don't actually do anything */
#define SO_SECURITY_AUTHENTICATION 0x5001
#define SO_SECURITY_ENCRYPTION_TRANSPORT 0x5002
@@ -85,4 +85,6 @@
#define SO_SELECT_ERR_QUEUE 45
+#define SO_LL 46
+
#endif /* _XTENSA_SOCKET_H */
@@ -39,12 +39,12 @@ extern unsigned int sysctl_net_ll_poll __read_mostly;
/* we can use sched_clock() because we don't care much about precision
* we only care that the average is bounded
*/
-static inline u64 ll_end_time(void)
+static inline u64 ll_end_time(struct sock *sk)
{
- u64 end_time = ACCESS_ONCE(sysctl_net_ll_poll);
+ u64 end_time = ACCESS_ONCE(sk->sk_ll_usec);
/* we don't mind a ~2.5% imprecision
- * sysctl_net_ll_poll is a u_int so this can't overflow
+ * sk->sk_ll_usec is a u_int so this can't overflow
*/
end_time = (end_time << 10) + sched_clock();
@@ -53,7 +53,7 @@ static inline u64 ll_end_time(void)
static inline bool sk_valid_ll(struct sock *sk)
{
- return sysctl_net_ll_poll && sk->sk_napi_id &&
+ return sk->sk_ll_usec && sk->sk_napi_id &&
!need_resched() && !signal_pending(current);
}
@@ -65,7 +65,7 @@ static inline bool can_poll_ll(u64 end_time)
static inline bool sk_poll_ll(struct sock *sk, int nonblock)
{
const struct net_device_ops *ops;
- u64 end_time = ll_end_time();
+ u64 end_time = ll_end_time(sk);
struct napi_struct *napi;
int rc = false;
@@ -118,7 +118,7 @@ static inline void sk_mark_ll(struct sock *sk, struct sk_buff *skb)
#else /* CONFIG_NET_LL_RX_POLL */
-static inline u64 ll_end_time(void)
+static inline u64 ll_end_time(struct sock *sk)
{
return 0;
}
@@ -230,6 +230,7 @@ struct cg_proto;
* @sk_wmem_queued: persistent queue size
* @sk_forward_alloc: space allocated forward
* @sk_napi_id: id of the last napi context to receive data for sk
+ * @sk_ll_usec: usecs to busypoll when there is no data
* @sk_allocation: allocation mode
* @sk_sndbuf: size of send buffer in bytes
* @sk_flags: %SO_LINGER (l_onoff), %SO_BROADCAST, %SO_KEEPALIVE,
@@ -328,6 +329,7 @@ struct sock {
#endif
#ifdef CONFIG_NET_LL_RX_POLL
unsigned int sk_napi_id;
+ unsigned int sk_ll_usec;
#endif
atomic_t sk_drops;
int sk_rcvbuf;
@@ -76,4 +76,6 @@
#define SO_SELECT_ERR_QUEUE 45
+#define SO_LL 46
+
#endif /* __ASM_GENERIC_SOCKET_H */
@@ -913,6 +913,19 @@ set_rcvbuf:
sock_valbool_flag(sk, SOCK_SELECT_ERR_QUEUE, valbool);
break;
+#ifdef CONFIG_NET_LL_RX_POLL
+ case SO_LL:
+ /* allow unprivileged users to decrease the value */
+ if ((val > sk->sk_ll_usec) && !capable(CAP_NET_ADMIN))
+ ret = -EPERM;
+ else {
+ if (val < 0)
+ ret = -EINVAL;
+ else
+ sk->sk_ll_usec = val;
+ }
+ break;
+#endif
default:
ret = -ENOPROTOOPT;
break;
@@ -1170,6 +1183,12 @@ int sock_getsockopt(struct socket *sock, int level, int optname,
v.val = sock_flag(sk, SOCK_SELECT_ERR_QUEUE);
break;
+#ifdef CONFIG_NET_LL_RX_POLL
+ case SO_LL:
+ v.val = sk->sk_ll_usec;
+ break;
+#endif
+
default:
return -ENOPROTOOPT;
}
@@ -2288,6 +2307,7 @@ void sock_init_data(struct socket *sock, struct sock *sk)
#ifdef CONFIG_NET_LL_RX_POLL
sk->sk_napi_id = 0;
+ sk->sk_ll_usec = sysctl_net_ll_poll;
#endif
/*
@@ -108,7 +108,6 @@
#ifdef CONFIG_NET_LL_RX_POLL
unsigned int sysctl_net_ll_poll __read_mostly;
-EXPORT_SYMBOL_GPL(sysctl_net_ll_poll);
#endif
static int sock_no_open(struct inode *irrelevant, struct file *dontcare);
adds a socket option for low latency polling. This allows overriding the global sysctl value with a per-socket one. Unexport sysctl_net_ll_poll since for now it's not needed in modules. Signed-off-by: Eliezer Tamir <eliezer.tamir@linux.intel.com> --- arch/alpha/include/uapi/asm/socket.h | 2 ++ arch/avr32/include/uapi/asm/socket.h | 2 ++ arch/cris/include/uapi/asm/socket.h | 2 ++ arch/frv/include/uapi/asm/socket.h | 2 ++ arch/h8300/include/uapi/asm/socket.h | 2 ++ arch/ia64/include/uapi/asm/socket.h | 2 ++ arch/m32r/include/uapi/asm/socket.h | 2 ++ arch/mips/include/uapi/asm/socket.h | 2 ++ arch/mn10300/include/uapi/asm/socket.h | 2 ++ arch/parisc/include/uapi/asm/socket.h | 2 ++ arch/powerpc/include/uapi/asm/socket.h | 2 ++ arch/s390/include/uapi/asm/socket.h | 2 ++ arch/sparc/include/uapi/asm/socket.h | 2 ++ arch/xtensa/include/uapi/asm/socket.h | 2 ++ include/net/ll_poll.h | 12 ++++++------ include/net/sock.h | 2 ++ include/uapi/asm-generic/socket.h | 2 ++ net/core/sock.c | 20 ++++++++++++++++++++ net/socket.c | 1 - 19 files changed, 58 insertions(+), 7 deletions(-) -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html