[linux-block.git] / include / linux / skmsg.h

/* SPDX-License-Identifier: GPL-2.0 */
/* Copyright (c) 2017 - 2018 Covalent IO, Inc. http://covalent.io */

#ifndef _LINUX_SKMSG_H
#define _LINUX_SKMSG_H

#include <linux/bpf.h>
#include <linux/filter.h>
#include <linux/scatterlist.h>
#include <linux/skbuff.h>

#include <net/sock.h>
#include <net/tcp.h>
#include <net/strparser.h>

#define MAX_MSG_FRAGS			MAX_SKB_FRAGS
#define NR_MSG_FRAG_IDS			(MAX_MSG_FRAGS + 1)

enum __sk_action {
	__SK_DROP = 0,
	__SK_PASS,
	__SK_REDIRECT,
	__SK_NONE,
};

struct sk_msg_sg {
	u32				start;
	u32				curr;
	u32				end;
	u32				size;
	u32				copybreak;
	unsigned long			copy;
	/* The extra two elements:
	 * 1) used for chaining the front and sections when the list becomes
	 *    partitioned (e.g. end < start). The crypto APIs require the
	 *    chaining;
	 * 2) to chain tailer SG entries after the message.
	 */
	struct scatterlist		data[MAX_MSG_FRAGS + 2];
};
static_assert(BITS_PER_LONG >= NR_MSG_FRAG_IDS);

/* UAPI in filter.c depends on struct sk_msg_sg being first element. */
struct sk_msg {
	struct sk_msg_sg		sg;
	void				*data;
	void				*data_end;
	u32				apply_bytes;
	u32				cork_bytes;
	u32				flags;
	struct sk_buff			*skb;
	struct sock			*sk_redir;
	struct sock			*sk;
	struct list_head		list;
};

struct sk_psock_progs {
	struct bpf_prog			*msg_parser;
	struct bpf_prog			*stream_parser;
	struct bpf_prog			*stream_verdict;
};

enum sk_psock_state_bits {
	SK_PSOCK_TX_ENABLED,
};

struct sk_psock_link {
	struct list_head		list;
	struct bpf_map			*map;
	void				*link_raw;
};

struct sk_psock_work_state {
	struct sk_buff			*skb;
	u32				len;
	u32				off;
};

struct sk_psock {
	struct sock			*sk;
	struct sock			*sk_redir;
	u32				apply_bytes;
	u32				cork_bytes;
	u32				eval;
	struct sk_msg			*cork;
	struct sk_psock_progs		progs;
#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
	struct strparser		strp;
#endif
	struct sk_buff_head		ingress_skb;
	struct list_head		ingress_msg;
	spinlock_t			ingress_lock;
	unsigned long			state;
	struct list_head		link;
	spinlock_t			link_lock;
	refcount_t			refcnt;
	void (*saved_unhash)(struct sock *sk);
	void (*saved_close)(struct sock *sk, long timeout);
	void (*saved_write_space)(struct sock *sk);
	void (*saved_data_ready)(struct sock *sk);
	struct proto			*sk_proto;
	struct mutex			work_mutex;
	struct sk_psock_work_state	work_state;
	struct work_struct		work;
	union {
		struct rcu_head		rcu;
		struct work_struct	gc;
	};
};

int sk_msg_alloc(struct sock *sk, struct sk_msg *msg, int len,
		 int elem_first_coalesce);
int sk_msg_clone(struct sock *sk, struct sk_msg *dst, struct sk_msg *src,
		 u32 off, u32 len);
void sk_msg_trim(struct sock *sk, struct sk_msg *msg, int len);
int sk_msg_free(struct sock *sk, struct sk_msg *msg);
int sk_msg_free_nocharge(struct sock *sk, struct sk_msg *msg);
void sk_msg_free_partial(struct sock *sk, struct sk_msg *msg, u32 bytes);
void sk_msg_free_partial_nocharge(struct sock *sk, struct sk_msg *msg,
				  u32 bytes);

void sk_msg_return(struct sock *sk, struct sk_msg *msg, int bytes);
void sk_msg_return_zero(struct sock *sk, struct sk_msg *msg, int bytes);

int sk_msg_zerocopy_from_iter(struct sock *sk, struct iov_iter *from,
			      struct sk_msg *msg, u32 bytes);
int sk_msg_memcopy_from_iter(struct sock *sk, struct iov_iter *from,
			     struct sk_msg *msg, u32 bytes);

static inline void sk_msg_check_to_free(struct sk_msg *msg, u32 i, u32 bytes)
{
	WARN_ON(i == msg->sg.end && bytes);
}

static inline void sk_msg_apply_bytes(struct sk_psock *psock, u32 bytes)
{
	if (psock->apply_bytes) {
		if (psock->apply_bytes < bytes)
			psock->apply_bytes = 0;
		else
			psock->apply_bytes -= bytes;
	}
}

static inline u32 sk_msg_iter_dist(u32 start, u32 end)
{
	return end >= start ? end - start : end + (NR_MSG_FRAG_IDS - start);
}

#define sk_msg_iter_var_prev(var)			\
	do {						\
		if (var == 0)				\
			var = NR_MSG_FRAG_IDS - 1;	\
		else					\
			var--;				\
	} while (0)

#define sk_msg_iter_var_next(var)			\
	do {						\
		var++;					\
		if (var == NR_MSG_FRAG_IDS)		\
			var = 0;			\
	} while (0)

#define sk_msg_iter_prev(msg, which)			\
	sk_msg_iter_var_prev(msg->sg.which)

#define sk_msg_iter_next(msg, which)			\
	sk_msg_iter_var_next(msg->sg.which)

static inline void sk_msg_clear_meta(struct sk_msg *msg)
{
	memset(&msg->sg, 0, offsetofend(struct sk_msg_sg, copy));
}

static inline void sk_msg_init(struct sk_msg *msg)
{
	BUILD_BUG_ON(ARRAY_SIZE(msg->sg.data) - 1 != NR_MSG_FRAG_IDS);
	memset(msg, 0, sizeof(*msg));
	sg_init_marker(msg->sg.data, NR_MSG_FRAG_IDS);
}

static inline void sk_msg_xfer(struct sk_msg *dst, struct sk_msg *src,
			       int which, u32 size)
{
	dst->sg.data[which] = src->sg.data[which];
	dst->sg.data[which].length  = size;
	dst->sg.size		   += size;
	src->sg.size		   -= size;
	src->sg.data[which].length -= size;
	src->sg.data[which].offset += size;
}

static inline void sk_msg_xfer_full(struct sk_msg *dst, struct sk_msg *src)
{
	memcpy(dst, src, sizeof(*src));
	sk_msg_init(src);
}

static inline bool sk_msg_full(const struct sk_msg *msg)
{
	return sk_msg_iter_dist(msg->sg.start, msg->sg.end) == MAX_MSG_FRAGS;
}

static inline u32 sk_msg_elem_used(const struct sk_msg *msg)
{
	return sk_msg_iter_dist(msg->sg.start, msg->sg.end);
}

static inline struct scatterlist *sk_msg_elem(struct sk_msg *msg, int which)
{
	return &msg->sg.data[which];
}

static inline struct scatterlist sk_msg_elem_cpy(struct sk_msg *msg, int which)
{
	return msg->sg.data[which];
}

static inline struct page *sk_msg_page(struct sk_msg *msg, int which)
{
	return sg_page(sk_msg_elem(msg, which));
}

static inline bool sk_msg_to_ingress(const struct sk_msg *msg)
{
	return msg->flags & BPF_F_INGRESS;
}

static inline void sk_msg_compute_data_pointers(struct sk_msg *msg)
{
	struct scatterlist *sge = sk_msg_elem(msg, msg->sg.start);

	if (test_bit(msg->sg.start, &msg->sg.copy)) {
		msg->data = NULL;
		msg->data_end = NULL;
	} else {
		msg->data = sg_virt(sge);
		msg->data_end = msg->data + sge->length;
	}
}

static inline void sk_msg_page_add(struct sk_msg *msg, struct page *page,
				   u32 len, u32 offset)
{
	struct scatterlist *sge;

	get_page(page);
	sge = sk_msg_elem(msg, msg->sg.end);
	sg_set_page(sge, page, len, offset);
	sg_unmark_end(sge);

	__set_bit(msg->sg.end, &msg->sg.copy);
	msg->sg.size += len;
	sk_msg_iter_next(msg, end);
}

static inline void sk_msg_sg_copy(struct sk_msg *msg, u32 i, bool copy_state)
{
	do {
		if (copy_state)
			__set_bit(i, &msg->sg.copy);
		else
			__clear_bit(i, &msg->sg.copy);
		sk_msg_iter_var_next(i);
		if (i == msg->sg.end)
			break;
	} while (1);
}

static inline void sk_msg_sg_copy_set(struct sk_msg *msg, u32 start)
{
	sk_msg_sg_copy(msg, start, true);
}

static inline void sk_msg_sg_copy_clear(struct sk_msg *msg, u32 start)
{
	sk_msg_sg_copy(msg, start, false);
}

static inline struct sk_psock *sk_psock(const struct sock *sk)
{
	return rcu_dereference_sk_user_data(sk);
}

static inline void sk_psock_queue_msg(struct sk_psock *psock,
				      struct sk_msg *msg)
{
	spin_lock_bh(&psock->ingress_lock);
	list_add_tail(&msg->list, &psock->ingress_msg);
	spin_unlock_bh(&psock->ingress_lock);
}

static inline struct sk_msg *sk_psock_dequeue_msg(struct sk_psock *psock)
{
	struct sk_msg *msg;

	spin_lock_bh(&psock->ingress_lock);
	msg = list_first_entry_or_null(&psock->ingress_msg, struct sk_msg, list);
	if (msg)
		list_del(&msg->list);
	spin_unlock_bh(&psock->ingress_lock);
	return msg;
}

static inline struct sk_msg *sk_psock_peek_msg(struct sk_psock *psock)
{
	struct sk_msg *msg;

	spin_lock_bh(&psock->ingress_lock);
	msg = list_first_entry_or_null(&psock->ingress_msg, struct sk_msg, list);
	spin_unlock_bh(&psock->ingress_lock);
	return msg;
}

static inline struct sk_msg *sk_psock_next_msg(struct sk_psock *psock,
					       struct sk_msg *msg)
{
	struct sk_msg *ret;

	spin_lock_bh(&psock->ingress_lock);
	if (list_is_last(&msg->list, &psock->ingress_msg))
		ret = NULL;
	else
		ret = list_next_entry(msg, list);
	spin_unlock_bh(&psock->ingress_lock);
	return ret;
}

static inline bool sk_psock_queue_empty(const struct sk_psock *psock)
{
	return psock ? list_empty(&psock->ingress_msg) : true;
}

static inline void kfree_sk_msg(struct sk_msg *msg)
{
	if (msg->skb)
		consume_skb(msg->skb);
	kfree(msg);
}

static inline void sk_psock_report_error(struct sk_psock *psock, int err)
{
	struct sock *sk = psock->sk;

	sk->sk_err = err;
	sk->sk_error_report(sk);
}

struct sk_psock *sk_psock_init(struct sock *sk, int node);
void sk_psock_stop(struct sk_psock *psock, bool wait);

#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
int sk_psock_init_strp(struct sock *sk, struct sk_psock *psock);
void sk_psock_start_strp(struct sock *sk, struct sk_psock *psock);
void sk_psock_stop_strp(struct sock *sk, struct sk_psock *psock);
#else
static inline int sk_psock_init_strp(struct sock *sk, struct sk_psock *psock)
{
	return -EOPNOTSUPP;
}

static inline void sk_psock_start_strp(struct sock *sk, struct sk_psock *psock)
{
}

static inline void sk_psock_stop_strp(struct sock *sk, struct sk_psock *psock)
{
}
#endif

void sk_psock_start_verdict(struct sock *sk, struct sk_psock *psock);
void sk_psock_stop_verdict(struct sock *sk, struct sk_psock *psock);

int sk_psock_msg_verdict(struct sock *sk, struct sk_psock *psock,
			 struct sk_msg *msg);

static inline struct sk_psock_link *sk_psock_init_link(void)
{
	return kzalloc(sizeof(struct sk_psock_link),
		       GFP_ATOMIC | __GFP_NOWARN);
}

static inline void sk_psock_free_link(struct sk_psock_link *link)
{
	kfree(link);
}

struct sk_psock_link *sk_psock_link_pop(struct sk_psock *psock);

static inline void sk_psock_cork_free(struct sk_psock *psock)
{
	if (psock->cork) {
		sk_msg_free(psock->sk, psock->cork);
		kfree(psock->cork);
		psock->cork = NULL;
	}
}

static inline void sk_psock_update_proto(struct sock *sk,
					 struct sk_psock *psock,
					 struct proto *ops)
{
	/* Pairs with lockless read in sk_clone_lock() */
	WRITE_ONCE(sk->sk_prot, ops);
}

static inline void sk_psock_restore_proto(struct sock *sk,
					  struct sk_psock *psock)
{
	sk->sk_prot->unhash = psock->saved_unhash;
	if (inet_csk_has_ulp(sk)) {
		tcp_update_ulp(sk, psock->sk_proto, psock->saved_write_space);
	} else {
		sk->sk_write_space = psock->saved_write_space;
		/* Pairs with lockless read in sk_clone_lock() */
		WRITE_ONCE(sk->sk_prot, psock->sk_proto);
	}
}

static inline void sk_psock_set_state(struct sk_psock *psock,
				      enum sk_psock_state_bits bit)
{
	set_bit(bit, &psock->state);
}

static inline void sk_psock_clear_state(struct sk_psock *psock,
					enum sk_psock_state_bits bit)
{
	clear_bit(bit, &psock->state);
}

static inline bool sk_psock_test_state(const struct sk_psock *psock,
				       enum sk_psock_state_bits bit)
{
	return test_bit(bit, &psock->state);
}

static inline struct sk_psock *sk_psock_get(struct sock *sk)
{
	struct sk_psock *psock;

	rcu_read_lock();
	psock = sk_psock(sk);
	if (psock && !refcount_inc_not_zero(&psock->refcnt))
		psock = NULL;
	rcu_read_unlock();
	return psock;
}

void sk_psock_drop(struct sock *sk, struct sk_psock *psock);

static inline void sk_psock_put(struct sock *sk, struct sk_psock *psock)
{
	if (refcount_dec_and_test(&psock->refcnt))
		sk_psock_drop(sk, psock);
}

static inline void sk_psock_data_ready(struct sock *sk, struct sk_psock *psock)
{
	if (psock->saved_data_ready)
		psock->saved_data_ready(sk);
	else
		sk->sk_data_ready(sk);
}

static inline void psock_set_prog(struct bpf_prog **pprog,
				  struct bpf_prog *prog)
{
	prog = xchg(pprog, prog);
	if (prog)
		bpf_prog_put(prog);
}

static inline int psock_replace_prog(struct bpf_prog **pprog,
				     struct bpf_prog *prog,
				     struct bpf_prog *old)
{
	if (cmpxchg(pprog, old, prog) != old)
		return -ENOENT;

	if (old)
		bpf_prog_put(old);

	return 0;
}

static inline void psock_progs_drop(struct sk_psock_progs *progs)
{
	psock_set_prog(&progs->msg_parser, NULL);
	psock_set_prog(&progs->stream_parser, NULL);
	psock_set_prog(&progs->stream_verdict, NULL);
}

int sk_psock_tls_strp_read(struct sk_psock *psock, struct sk_buff *skb);

static inline bool sk_psock_strp_enabled(struct sk_psock *psock)
{
	if (!psock)
		return false;
	return !!psock->saved_data_ready;
}

#if IS_ENABLED(CONFIG_NET_SOCK_MSG)

/* We only have one bit so far. */
#define BPF_F_PTR_MASK ~(BPF_F_INGRESS)

static inline bool skb_bpf_ingress(const struct sk_buff *skb)
{
	unsigned long sk_redir = skb->_sk_redir;

	return sk_redir & BPF_F_INGRESS;
}

static inline void skb_bpf_set_ingress(struct sk_buff *skb)
{
	skb->_sk_redir |= BPF_F_INGRESS;
}

static inline void skb_bpf_set_redir(struct sk_buff *skb, struct sock *sk_redir,
				     bool ingress)
{
	skb->_sk_redir = (unsigned long)sk_redir;
	if (ingress)
		skb->_sk_redir |= BPF_F_INGRESS;
}

static inline struct sock *skb_bpf_redirect_fetch(const struct sk_buff *skb)
{
	unsigned long sk_redir = skb->_sk_redir;

	return (struct sock *)(sk_redir & BPF_F_PTR_MASK);
}

static inline void skb_bpf_redirect_clear(struct sk_buff *skb)
{
	skb->_sk_redir = 0;
}
#endif /* CONFIG_NET_SOCK_MSG */
#endif /* _LINUX_SKMSG_H */
Commit	Line	Data
604326b4 DB	1	/* SPDX-License-Identifier: GPL-2.0 */
	2	/* Copyright (c) 2017 - 2018 Covalent IO, Inc. http://covalent.io */
	3
	4	#ifndef _LINUX_SKMSG_H
	5	#define _LINUX_SKMSG_H
	6
	7	#include <linux/bpf.h>
	8	#include <linux/filter.h>
	9	#include <linux/scatterlist.h>
	10	#include <linux/skbuff.h>
	11
	12	#include <net/sock.h>
	13	#include <net/tcp.h>
	14	#include <net/strparser.h>
	15
	16	#define MAX_MSG_FRAGS MAX_SKB_FRAGS
031097d9	17	#define NR_MSG_FRAG_IDS (MAX_MSG_FRAGS + 1)
604326b4 DB	18
	19	enum __sk_action {
	20	__SK_DROP = 0,
	21	__SK_PASS,
	22	__SK_REDIRECT,
	23	__SK_NONE,
	24	};
	25
	26	struct sk_msg_sg {
	27	u32 start;
	28	u32 curr;
	29	u32 end;
	30	u32 size;
	31	u32 copybreak;
163ab96b	32	unsigned long copy;
031097d9 JK	33	/* The extra two elements:
	34	* 1) used for chaining the front and sections when the list becomes
	35	* partitioned (e.g. end < start). The crypto APIs require the
	36	* chaining;
	37	* 2) to chain tailer SG entries after the message.
d3b18ad3	38	*/
031097d9	39	struct scatterlist data[MAX_MSG_FRAGS + 2];
604326b4	40	};
031097d9	41	static_assert(BITS_PER_LONG >= NR_MSG_FRAG_IDS);
604326b4	42
7a69c0f2	43	/* UAPI in filter.c depends on struct sk_msg_sg being first element. */
604326b4 DB	44	struct sk_msg {
	45	struct sk_msg_sg sg;
	46	void *data;
	47	void *data_end;
	48	u32 apply_bytes;
	49	u32 cork_bytes;
	50	u32 flags;
	51	struct sk_buff *skb;
	52	struct sock *sk_redir;
	53	struct sock *sk;
	54	struct list_head list;
	55	};
	56
	57	struct sk_psock_progs {
	58	struct bpf_prog *msg_parser;
ae8b8332 CW	59	struct bpf_prog *stream_parser;
ae8b8332 CW	60	struct bpf_prog *stream_verdict;
604326b4 DB	61	};
	62
	63	enum sk_psock_state_bits {
	64	SK_PSOCK_TX_ENABLED,
	65	};
	66
	67	struct sk_psock_link {
	68	struct list_head list;
	69	struct bpf_map *map;
	70	void *link_raw;
	71	};
	72
604326b4 DB	73	struct sk_psock_work_state {
	74	struct sk_buff *skb;
	75	u32 len;
	76	u32 off;
	77	};
	78
	79	struct sk_psock {
	80	struct sock *sk;
	81	struct sock *sk_redir;
	82	u32 apply_bytes;
	83	u32 cork_bytes;
	84	u32 eval;
	85	struct sk_msg *cork;
	86	struct sk_psock_progs progs;
5a685cd9 CW	87	#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
	88	struct strparser strp;
	89	#endif
604326b4 DB	90	struct sk_buff_head ingress_skb;
604326b4 DB	91	struct list_head ingress_msg;
b01fd6e8	92	spinlock_t ingress_lock;
604326b4 DB	93	unsigned long state;
	94	struct list_head link;
	95	spinlock_t link_lock;
	96	refcount_t refcnt;
	97	void (saved_unhash)(struct sock sk);
	98	void (saved_close)(struct sock sk, long timeout);
	99	void (saved_write_space)(struct sock sk);
5a685cd9	100	void (saved_data_ready)(struct sock sk);
604326b4	101	struct proto *sk_proto;
799aa7f9	102	struct mutex work_mutex;
604326b4 DB	103	struct sk_psock_work_state work_state;
	104	struct work_struct work;
	105	union {
	106	struct rcu_head rcu;
	107	struct work_struct gc;
	108	};
	109	};
	110
	111	int sk_msg_alloc(struct sock sk, struct sk_msg msg, int len,
	112	int elem_first_coalesce);
d829e9c4 DB	113	int sk_msg_clone(struct sock sk, struct sk_msg dst, struct sk_msg *src,
d829e9c4 DB	114	u32 off, u32 len);
604326b4 DB	115	void sk_msg_trim(struct sock sk, struct sk_msg msg, int len);
	116	int sk_msg_free(struct sock sk, struct sk_msg msg);
	117	int sk_msg_free_nocharge(struct sock sk, struct sk_msg msg);
	118	void sk_msg_free_partial(struct sock sk, struct sk_msg msg, u32 bytes);
	119	void sk_msg_free_partial_nocharge(struct sock sk, struct sk_msg msg,
	120	u32 bytes);
	121
	122	void sk_msg_return(struct sock sk, struct sk_msg msg, int bytes);
d3b18ad3	123	void sk_msg_return_zero(struct sock sk, struct sk_msg msg, int bytes);
604326b4 DB	124
	125	int sk_msg_zerocopy_from_iter(struct sock sk, struct iov_iter from,
	126	struct sk_msg *msg, u32 bytes);
	127	int sk_msg_memcopy_from_iter(struct sock sk, struct iov_iter from,
	128	struct sk_msg *msg, u32 bytes);
	129
	130	static inline void sk_msg_check_to_free(struct sk_msg *msg, u32 i, u32 bytes)
	131	{
	132	WARN_ON(i == msg->sg.end && bytes);
	133	}
	134
	135	static inline void sk_msg_apply_bytes(struct sk_psock *psock, u32 bytes)
	136	{
	137	if (psock->apply_bytes) {
	138	if (psock->apply_bytes < bytes)
	139	psock->apply_bytes = 0;
	140	else
	141	psock->apply_bytes -= bytes;
	142	}
	143	}
	144
683916f6 JK	145	static inline u32 sk_msg_iter_dist(u32 start, u32 end)
683916f6 JK	146	{
031097d9	147	return end >= start ? end - start : end + (NR_MSG_FRAG_IDS - start);
683916f6 JK	148	}
683916f6 JK	149
604326b4 DB	150	#define sk_msg_iter_var_prev(var) \
	151	do { \
	152	if (var == 0) \
031097d9	153	var = NR_MSG_FRAG_IDS - 1; \
604326b4 DB	154	else \
	155	var--; \
	156	} while (0)
	157
	158	#define sk_msg_iter_var_next(var) \
	159	do { \
	160	var++; \
031097d9	161	if (var == NR_MSG_FRAG_IDS) \
604326b4 DB	162	var = 0; \
	163	} while (0)
	164
	165	#define sk_msg_iter_prev(msg, which) \
	166	sk_msg_iter_var_prev(msg->sg.which)
	167
	168	#define sk_msg_iter_next(msg, which) \
	169	sk_msg_iter_var_next(msg->sg.which)
	170
	171	static inline void sk_msg_clear_meta(struct sk_msg *msg)
	172	{
	173	memset(&msg->sg, 0, offsetofend(struct sk_msg_sg, copy));
	174	}
	175
	176	static inline void sk_msg_init(struct sk_msg *msg)
	177	{
031097d9	178	BUILD_BUG_ON(ARRAY_SIZE(msg->sg.data) - 1 != NR_MSG_FRAG_IDS);
604326b4	179	memset(msg, 0, sizeof(*msg));
031097d9	180	sg_init_marker(msg->sg.data, NR_MSG_FRAG_IDS);
604326b4 DB	181	}
	182
	183	static inline void sk_msg_xfer(struct sk_msg dst, struct sk_msg src,
	184	int which, u32 size)
	185	{
	186	dst->sg.data[which] = src->sg.data[which];
	187	dst->sg.data[which].length = size;
3f4c3127	188	dst->sg.size += size;
81aabbb9	189	src->sg.size -= size;
604326b4 DB	190	src->sg.data[which].length -= size;
	191	src->sg.data[which].offset += size;
	192	}
	193
d3b18ad3 JF	194	static inline void sk_msg_xfer_full(struct sk_msg dst, struct sk_msg src)
	195	{
	196	memcpy(dst, src, sizeof(*src));
	197	sk_msg_init(src);
	198	}
	199
8734a162 JF	200	static inline bool sk_msg_full(const struct sk_msg *msg)
8734a162 JF	201	{
031097d9	202	return sk_msg_iter_dist(msg->sg.start, msg->sg.end) == MAX_MSG_FRAGS;
8734a162 JF	203	}
8734a162 JF	204
604326b4 DB	205	static inline u32 sk_msg_elem_used(const struct sk_msg *msg)
604326b4 DB	206	{
683916f6	207	return sk_msg_iter_dist(msg->sg.start, msg->sg.end);
604326b4 DB	208	}
604326b4 DB	209
604326b4 DB	210	static inline struct scatterlist sk_msg_elem(struct sk_msg msg, int which)
	211	{
	212	return &msg->sg.data[which];
	213	}
	214
6fff607e JF	215	static inline struct scatterlist sk_msg_elem_cpy(struct sk_msg *msg, int which)
	216	{
	217	return msg->sg.data[which];
	218	}
	219
604326b4 DB	220	static inline struct page sk_msg_page(struct sk_msg msg, int which)
	221	{
	222	return sg_page(sk_msg_elem(msg, which));
	223	}
	224
	225	static inline bool sk_msg_to_ingress(const struct sk_msg *msg)
	226	{
	227	return msg->flags & BPF_F_INGRESS;
	228	}
	229
	230	static inline void sk_msg_compute_data_pointers(struct sk_msg *msg)
	231	{
	232	struct scatterlist *sge = sk_msg_elem(msg, msg->sg.start);
	233
163ab96b	234	if (test_bit(msg->sg.start, &msg->sg.copy)) {
604326b4 DB	235	msg->data = NULL;
	236	msg->data_end = NULL;
	237	} else {
	238	msg->data = sg_virt(sge);
	239	msg->data_end = msg->data + sge->length;
	240	}
	241	}
	242
	243	static inline void sk_msg_page_add(struct sk_msg msg, struct page page,
	244	u32 len, u32 offset)
	245	{
	246	struct scatterlist *sge;
	247
	248	get_page(page);
	249	sge = sk_msg_elem(msg, msg->sg.end);
	250	sg_set_page(sge, page, len, offset);
	251	sg_unmark_end(sge);
	252
163ab96b	253	__set_bit(msg->sg.end, &msg->sg.copy);
604326b4 DB	254	msg->sg.size += len;
	255	sk_msg_iter_next(msg, end);
	256	}
	257
d3b18ad3 JF	258	static inline void sk_msg_sg_copy(struct sk_msg *msg, u32 i, bool copy_state)
	259	{
	260	do {
163ab96b JK	261	if (copy_state)
	262	__set_bit(i, &msg->sg.copy);
	263	else
	264	__clear_bit(i, &msg->sg.copy);
d3b18ad3 JF	265	sk_msg_iter_var_next(i);
	266	if (i == msg->sg.end)
	267	break;
	268	} while (1);
	269	}
	270
	271	static inline void sk_msg_sg_copy_set(struct sk_msg *msg, u32 start)
	272	{
	273	sk_msg_sg_copy(msg, start, true);
	274	}
	275
	276	static inline void sk_msg_sg_copy_clear(struct sk_msg *msg, u32 start)
	277	{
	278	sk_msg_sg_copy(msg, start, false);
	279	}
	280
604326b4 DB	281	static inline struct sk_psock sk_psock(const struct sock sk)
	282	{
	283	return rcu_dereference_sk_user_data(sk);
	284	}
	285
604326b4 DB	286	static inline void sk_psock_queue_msg(struct sk_psock *psock,
	287	struct sk_msg *msg)
	288	{
b01fd6e8	289	spin_lock_bh(&psock->ingress_lock);
604326b4	290	list_add_tail(&msg->list, &psock->ingress_msg);
b01fd6e8 CW	291	spin_unlock_bh(&psock->ingress_lock);
	292	}
	293
	294	static inline struct sk_msg sk_psock_dequeue_msg(struct sk_psock psock)
	295	{
	296	struct sk_msg *msg;
	297
	298	spin_lock_bh(&psock->ingress_lock);
	299	msg = list_first_entry_or_null(&psock->ingress_msg, struct sk_msg, list);
	300	if (msg)
	301	list_del(&msg->list);
	302	spin_unlock_bh(&psock->ingress_lock);
	303	return msg;
	304	}
	305
	306	static inline struct sk_msg sk_psock_peek_msg(struct sk_psock psock)
	307	{
	308	struct sk_msg *msg;
	309
	310	spin_lock_bh(&psock->ingress_lock);
	311	msg = list_first_entry_or_null(&psock->ingress_msg, struct sk_msg, list);
	312	spin_unlock_bh(&psock->ingress_lock);
	313	return msg;
	314	}
	315
	316	static inline struct sk_msg sk_psock_next_msg(struct sk_psock psock,
	317	struct sk_msg *msg)
	318	{
	319	struct sk_msg *ret;
	320
	321	spin_lock_bh(&psock->ingress_lock);
	322	if (list_is_last(&msg->list, &psock->ingress_msg))
	323	ret = NULL;
	324	else
	325	ret = list_next_entry(msg, list);
	326	spin_unlock_bh(&psock->ingress_lock);
	327	return ret;
604326b4 DB	328	}
604326b4 DB	329
d3b18ad3 JF	330	static inline bool sk_psock_queue_empty(const struct sk_psock *psock)
	331	{
	332	return psock ? list_empty(&psock->ingress_msg) : true;
	333	}
	334
b01fd6e8 CW	335	static inline void kfree_sk_msg(struct sk_msg *msg)
	336	{
	337	if (msg->skb)
	338	consume_skb(msg->skb);
	339	kfree(msg);
	340	}
	341
604326b4 DB	342	static inline void sk_psock_report_error(struct sk_psock *psock, int err)
	343	{
	344	struct sock *sk = psock->sk;
	345
	346	sk->sk_err = err;
	347	sk->sk_error_report(sk);
	348	}
	349
	350	struct sk_psock sk_psock_init(struct sock sk, int node);
799aa7f9	351	void sk_psock_stop(struct sk_psock *psock, bool wait);
604326b4	352
88759609	353	#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
604326b4 DB	354	int sk_psock_init_strp(struct sock sk, struct sk_psock psock);
	355	void sk_psock_start_strp(struct sock sk, struct sk_psock psock);
	356	void sk_psock_stop_strp(struct sock sk, struct sk_psock psock);
88759609 CW	357	#else
	358	static inline int sk_psock_init_strp(struct sock sk, struct sk_psock psock)
	359	{
	360	return -EOPNOTSUPP;
	361	}
	362
	363	static inline void sk_psock_start_strp(struct sock sk, struct sk_psock psock)
	364	{
	365	}
	366
	367	static inline void sk_psock_stop_strp(struct sock sk, struct sk_psock psock)
	368	{
	369	}
	370	#endif
	371
ef565928 JF	372	void sk_psock_start_verdict(struct sock sk, struct sk_psock psock);
ef565928 JF	373	void sk_psock_stop_verdict(struct sock sk, struct sk_psock psock);
604326b4 DB	374
	375	int sk_psock_msg_verdict(struct sock sk, struct sk_psock psock,
	376	struct sk_msg *msg);
	377
	378	static inline struct sk_psock_link *sk_psock_init_link(void)
	379	{
	380	return kzalloc(sizeof(struct sk_psock_link),
	381	GFP_ATOMIC \| __GFP_NOWARN);
	382	}
	383
	384	static inline void sk_psock_free_link(struct sk_psock_link *link)
	385	{
	386	kfree(link);
	387	}
	388
	389	struct sk_psock_link sk_psock_link_pop(struct sk_psock psock);
604326b4	390
604326b4 DB	391	static inline void sk_psock_cork_free(struct sk_psock *psock)
	392	{
	393	if (psock->cork) {
	394	sk_msg_free(psock->sk, psock->cork);
	395	kfree(psock->cork);
	396	psock->cork = NULL;
	397	}
	398	}
	399
	400	static inline void sk_psock_update_proto(struct sock *sk,
	401	struct sk_psock *psock,
	402	struct proto *ops)
	403	{
b8e202d1 JS	404	/* Pairs with lockless read in sk_clone_lock() */
b8e202d1 JS	405	WRITE_ONCE(sk->sk_prot, ops);
604326b4 DB	406	}
	407
	408	static inline void sk_psock_restore_proto(struct sock *sk,
	409	struct sk_psock *psock)
	410	{
4da6a196	411	sk->sk_prot->unhash = psock->saved_unhash;
7b70973d LB	412	if (inet_csk_has_ulp(sk)) {
	413	tcp_update_ulp(sk, psock->sk_proto, psock->saved_write_space);
	414	} else {
	415	sk->sk_write_space = psock->saved_write_space;
	416	/* Pairs with lockless read in sk_clone_lock() */
	417	WRITE_ONCE(sk->sk_prot, psock->sk_proto);
	418	}
604326b4 DB	419	}
	420
	421	static inline void sk_psock_set_state(struct sk_psock *psock,
	422	enum sk_psock_state_bits bit)
	423	{
	424	set_bit(bit, &psock->state);
	425	}
	426
	427	static inline void sk_psock_clear_state(struct sk_psock *psock,
	428	enum sk_psock_state_bits bit)
	429	{
	430	clear_bit(bit, &psock->state);
	431	}
	432
	433	static inline bool sk_psock_test_state(const struct sk_psock *psock,
	434	enum sk_psock_state_bits bit)
	435	{
	436	return test_bit(bit, &psock->state);
	437	}
	438
	439	static inline struct sk_psock sk_psock_get(struct sock sk)
	440	{
	441	struct sk_psock *psock;
	442
	443	rcu_read_lock();
	444	psock = sk_psock(sk);
	445	if (psock && !refcount_inc_not_zero(&psock->refcnt))
	446	psock = NULL;
	447	rcu_read_unlock();
	448	return psock;
	449	}
	450
604326b4 DB	451	void sk_psock_drop(struct sock sk, struct sk_psock psock);
	452
	453	static inline void sk_psock_put(struct sock sk, struct sk_psock psock)
	454	{
	455	if (refcount_dec_and_test(&psock->refcnt))
	456	sk_psock_drop(sk, psock);
	457	}
	458
552de910 JF	459	static inline void sk_psock_data_ready(struct sock sk, struct sk_psock psock)
552de910 JF	460	{
5a685cd9 CW	461	if (psock->saved_data_ready)
5a685cd9 CW	462	psock->saved_data_ready(sk);
552de910 JF	463	else
	464	sk->sk_data_ready(sk);
	465	}
	466
604326b4 DB	467	static inline void psock_set_prog(struct bpf_prog **pprog,
	468	struct bpf_prog *prog)
	469	{
	470	prog = xchg(pprog, prog);
	471	if (prog)
	472	bpf_prog_put(prog);
	473	}
	474
bb0de313 LB	475	static inline int psock_replace_prog(struct bpf_prog **pprog,
	476	struct bpf_prog *prog,
	477	struct bpf_prog *old)
	478	{
	479	if (cmpxchg(pprog, old, prog) != old)
	480	return -ENOENT;
	481
	482	if (old)
	483	bpf_prog_put(old);
	484
	485	return 0;
	486	}
	487
604326b4 DB	488	static inline void psock_progs_drop(struct sk_psock_progs *progs)
	489	{
	490	psock_set_prog(&progs->msg_parser, NULL);
ae8b8332 CW	491	psock_set_prog(&progs->stream_parser, NULL);
ae8b8332 CW	492	psock_set_prog(&progs->stream_verdict, NULL);
604326b4 DB	493	}
604326b4 DB	494
e91de6af JF	495	int sk_psock_tls_strp_read(struct sk_psock psock, struct sk_buff skb);
	496
	497	static inline bool sk_psock_strp_enabled(struct sk_psock *psock)
	498	{
	499	if (!psock)
	500	return false;
5a685cd9	501	return !!psock->saved_data_ready;
e91de6af	502	}
e3526bb9 CW	503
	504	#if IS_ENABLED(CONFIG_NET_SOCK_MSG)
	505
	506	/* We only have one bit so far. */
	507	#define BPF_F_PTR_MASK ~(BPF_F_INGRESS)
	508
	509	static inline bool skb_bpf_ingress(const struct sk_buff *skb)
	510	{
	511	unsigned long sk_redir = skb->_sk_redir;
	512
	513	return sk_redir & BPF_F_INGRESS;
	514	}
	515
	516	static inline void skb_bpf_set_ingress(struct sk_buff *skb)
	517	{
	518	skb->_sk_redir \|= BPF_F_INGRESS;
	519	}
	520
	521	static inline void skb_bpf_set_redir(struct sk_buff skb, struct sock sk_redir,
	522	bool ingress)
	523	{
	524	skb->_sk_redir = (unsigned long)sk_redir;
	525	if (ingress)
	526	skb->_sk_redir \|= BPF_F_INGRESS;
	527	}
	528
	529	static inline struct sock skb_bpf_redirect_fetch(const struct sk_buff skb)
	530	{
	531	unsigned long sk_redir = skb->_sk_redir;
	532
	533	return (struct sock *)(sk_redir & BPF_F_PTR_MASK);
	534	}
	535
	536	static inline void skb_bpf_redirect_clear(struct sk_buff *skb)
	537	{
	538	skb->_sk_redir = 0;
	539	}
	540	#endif /* CONFIG_NET_SOCK_MSG */
604326b4	541	#endif /* _LINUX_SKMSG_H */