net: reorganize struct sock for better data locality

Group fields used in TX path, and keep some cache lines mostly read
to permit sharing among cpus.

Gained two 4 bytes holes on 64bit arches.

Added a place holder for tcp tsq_flags, next to sk_wmem_alloc
to speed up tcp_wfree() in the following patch.

I have not added ____cacheline_aligned_in_smp, this might be done later.
I prefer doing this once inet and tcp/udp sockets reorg is also done.

Tested with both TCP and UDP.

UDP receiver performance under flood increased by ~20 % :
Accessing sk_filter/sk_wq/sk_napi_id no longer stalls because sk_drops
was moved away from a critical cache line, now mostly read and shared.

	/* --- cacheline 4 boundary (256 bytes) --- */
	unsigned int               sk_napi_id;           /* 0x100   0x4 */
	int                        sk_rcvbuf;            /* 0x104   0x4 */
	struct sk_filter *         sk_filter;            /* 0x108   0x8 */
	union {
		struct socket_wq * sk_wq;                /*         0x8 */
		struct socket_wq * sk_wq_raw;            /*         0x8 */
	};                                               /* 0x110   0x8 */
	struct xfrm_policy *       sk_policy[2];         /* 0x118  0x10 */
	struct dst_entry *         sk_rx_dst;            /* 0x128   0x8 */
	struct dst_entry *         sk_dst_cache;         /* 0x130   0x8 */
	atomic_t                   sk_omem_alloc;        /* 0x138   0x4 */
	int                        sk_sndbuf;            /* 0x13c   0x4 */
	/* --- cacheline 5 boundary (320 bytes) --- */
	int                        sk_wmem_queued;       /* 0x140   0x4 */
	atomic_t                   sk_wmem_alloc;        /* 0x144   0x4 */
	long unsigned int          sk_tsq_flags;         /* 0x148   0x8 */
	struct sk_buff *           sk_send_head;         /* 0x150   0x8 */
	struct sk_buff_head        sk_write_queue;       /* 0x158  0x18 */
	__s32                      sk_peek_off;          /* 0x170   0x4 */
	int                        sk_write_pending;     /* 0x174   0x4 */
	long int                   sk_sndtimeo;          /* 0x178   0x8 */

Signed-off-by: Eric Dumazet <edumazet@google.com>
Tested-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

include/net/sock.h

@@ -343,6 +343,9 @@ struct sock {
 #define sk_rxhash		__sk_common.skc_rxhash
 
 	socket_lock_t		sk_lock;
+	atomic_t		sk_drops;
+	int			sk_rcvlowat;
+	struct sk_buff_head	sk_error_queue;
 	struct sk_buff_head	sk_receive_queue;
 	/*
 	 * The backlog queue is special, it is always used with
@@ -359,14 +362,13 @@ struct sock {
 		struct sk_buff	*tail;
 	} sk_backlog;
 #define sk_rmem_alloc sk_backlog.rmem_alloc
-	int			sk_forward_alloc;
 
-	__u32			sk_txhash;
+	int			sk_forward_alloc;
 #ifdef CONFIG_NET_RX_BUSY_POLL
-	unsigned int		sk_napi_id;
 	unsigned int		sk_ll_usec;
+	/* ===== mostly read cache line ===== */
+	unsigned int		sk_napi_id;
 #endif
-	atomic_t		sk_drops;
 	int			sk_rcvbuf;
 
 	struct sk_filter __rcu	*sk_filter;
@@ -379,11 +381,30 @@ struct sock {
 #endif
 	struct dst_entry	*sk_rx_dst;
 	struct dst_entry __rcu	*sk_dst_cache;
-	/* Note: 32bit hole on 64bit arches */
-	atomic_t		sk_wmem_alloc;
 	atomic_t		sk_omem_alloc;
 	int			sk_sndbuf;
+
+	/* ===== cache line for TX ===== */
+	int			sk_wmem_queued;
+	atomic_t		sk_wmem_alloc;
+	unsigned long		sk_tsq_flags;
+	struct sk_buff		*sk_send_head;
 	struct sk_buff_head	sk_write_queue;
+	__s32			sk_peek_off;
+	int			sk_write_pending;
+	long			sk_sndtimeo;
+	struct timer_list	sk_timer;
+	__u32			sk_priority;
+	__u32			sk_mark;
+	u32			sk_pacing_rate; /* bytes per second */
+	u32			sk_max_pacing_rate;
+	struct page_frag	sk_frag;
+	netdev_features_t	sk_route_caps;
+	netdev_features_t	sk_route_nocaps;
+	int			sk_gso_type;
+	unsigned int		sk_gso_max_size;
+	gfp_t			sk_allocation;
+	__u32			sk_txhash;
 
 	/*
 	 * Because of non atomicity rules, all
@@ -414,42 +435,24 @@ struct sock {
 #define SK_PROTOCOL_MAX U8_MAX
 	kmemcheck_bitfield_end(flags);
 
-	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;
-	unsigned int		sk_gso_max_size;
 	u16			sk_gso_max_segs;
-	int			sk_rcvlowat;
 	unsigned long	        sk_lingertime;
-	struct sk_buff_head	sk_error_queue;
 	struct proto		*sk_prot_creator;
 	rwlock_t		sk_callback_lock;
 	int			sk_err,
 				sk_err_soft;
 	u32			sk_ack_backlog;
 	u32			sk_max_ack_backlog;
-	__u32			sk_priority;
-	__u32			sk_mark;
 	kuid_t			sk_uid;
 	struct pid		*sk_peer_pid;
 	const struct cred	*sk_peer_cred;
 	long			sk_rcvtimeo;
-	long			sk_sndtimeo;
-	struct timer_list	sk_timer;
 	ktime_t			sk_stamp;
 	u16			sk_tsflags;
 	u8			sk_shutdown;
 	u32			sk_tskey;
 	struct socket		*sk_socket;
 	void			*sk_user_data;
-	struct page_frag	sk_frag;
-	struct sk_buff		*sk_send_head;
-	__s32			sk_peek_off;
-	int			sk_write_pending;
 #ifdef CONFIG_SECURITY
 	void			*sk_security;
 #endif