[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;
	DECLARE_BITMAP(copy, MAX_MSG_FRAGS + 2);
	/* 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];
};

/* 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;
	struct bpf_prog			*skb_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_destroy)(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);
	int  (*psock_update_sk_prot)(struct sock *sk, struct sk_psock *psock,
				     bool restore);
	struct proto			*sk_proto;
	struct mutex			work_mutex;
	struct sk_psock_work_state	work_state;
	struct work_struct		work;
	struct rcu_work			rwork;
};

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);
int sk_msg_recvmsg(struct sock *sk, struct sk_psock *psock, struct msghdr *msg,
		   int len, int flags);
bool sk_msg_is_readable(struct sock *sk);

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_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_with_flags(sk,
							 SK_USER_DATA_PSOCK);
}

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 void sock_drop(struct sock *sk, struct sk_buff *skb)
{
	sk_drops_add(sk, skb);
	kfree_skb(skb);
}

static inline void sk_psock_queue_msg(struct sk_psock *psock,
				      struct sk_msg *msg)
{
	spin_lock_bh(&psock->ingress_lock);
	if (sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED))
		list_add_tail(&msg->list, &psock->ingress_msg);
	else {
		sk_msg_free(psock->sk, msg);
		kfree(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_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_restore_proto(struct sock *sk,
					  struct sk_psock *psock)
{
	if (psock->psock_update_sk_prot)
		psock->psock_update_sk_prot(sk, psock, true);
}

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);
	psock_set_prog(&progs->skb_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;
}

static inline bool sk_is_udp(const struct sock *sk)
{
	return sk->sk_type == SOCK_DGRAM &&
	       sk->sk_protocol == IPPROTO_UDP;
}

#if IS_ENABLED(CONFIG_NET_SOCK_MSG)

#define BPF_F_STRPARSER	(1UL << 1)

/* We only have two bits so far. */
#define BPF_F_PTR_MASK ~(BPF_F_INGRESS | BPF_F_STRPARSER)

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

	return sk_redir & BPF_F_STRPARSER;
}

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

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;
5a8fb33e	32	DECLARE_BITMAP(copy, MAX_MSG_FRAGS + 2);
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 DB	40	};
604326b4 DB	41
7a69c0f2	42	/* UAPI in filter.c depends on struct sk_msg_sg being first element. */
604326b4 DB	43	struct sk_msg {
	44	struct sk_msg_sg sg;
	45	void *data;
	46	void *data_end;
	47	u32 apply_bytes;
	48	u32 cork_bytes;
	49	u32 flags;
	50	struct sk_buff *skb;
	51	struct sock *sk_redir;
	52	struct sock *sk;
	53	struct list_head list;
	54	};
	55
	56	struct sk_psock_progs {
	57	struct bpf_prog *msg_parser;
ae8b8332 CW	58	struct bpf_prog *stream_parser;
ae8b8332 CW	59	struct bpf_prog *stream_verdict;
a7ba4558	60	struct bpf_prog *skb_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);
d8616ee2	98	void (saved_destroy)(struct sock sk);
604326b4 DB	99	void (saved_close)(struct sock sk, long timeout);
604326b4 DB	100	void (saved_write_space)(struct sock sk);
5a685cd9	101	void (saved_data_ready)(struct sock sk);
51e0158a CW	102	int (psock_update_sk_prot)(struct sock sk, struct sk_psock *psock,
51e0158a CW	103	bool restore);
604326b4	104	struct proto *sk_proto;
799aa7f9	105	struct mutex work_mutex;
604326b4 DB	106	struct sk_psock_work_state work_state;
604326b4 DB	107	struct work_struct work;
7786dfc4	108	struct rcu_work rwork;
604326b4 DB	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);
2bc793e3 CW	129	int sk_msg_recvmsg(struct sock sk, struct sk_psock psock, struct msghdr *msg,
2bc793e3 CW	130	int len, int flags);
fb4e0a5e	131	bool sk_msg_is_readable(struct sock *sk);
604326b4 DB	132
	133	static inline void sk_msg_check_to_free(struct sk_msg *msg, u32 i, u32 bytes)
	134	{
	135	WARN_ON(i == msg->sg.end && bytes);
	136	}
	137
	138	static inline void sk_msg_apply_bytes(struct sk_psock *psock, u32 bytes)
	139	{
	140	if (psock->apply_bytes) {
	141	if (psock->apply_bytes < bytes)
	142	psock->apply_bytes = 0;
	143	else
	144	psock->apply_bytes -= bytes;
	145	}
	146	}
	147
683916f6 JK	148	static inline u32 sk_msg_iter_dist(u32 start, u32 end)
683916f6 JK	149	{
031097d9	150	return end >= start ? end - start : end + (NR_MSG_FRAG_IDS - start);
683916f6 JK	151	}
683916f6 JK	152
604326b4 DB	153	#define sk_msg_iter_var_prev(var) \
	154	do { \
	155	if (var == 0) \
031097d9	156	var = NR_MSG_FRAG_IDS - 1; \
604326b4 DB	157	else \
	158	var--; \
	159	} while (0)
	160
	161	#define sk_msg_iter_var_next(var) \
	162	do { \
	163	var++; \
031097d9	164	if (var == NR_MSG_FRAG_IDS) \
604326b4 DB	165	var = 0; \
	166	} while (0)
	167
	168	#define sk_msg_iter_prev(msg, which) \
	169	sk_msg_iter_var_prev(msg->sg.which)
	170
	171	#define sk_msg_iter_next(msg, which) \
	172	sk_msg_iter_var_next(msg->sg.which)
	173
604326b4 DB	174	static inline void sk_msg_init(struct sk_msg *msg)
604326b4 DB	175	{
031097d9	176	BUILD_BUG_ON(ARRAY_SIZE(msg->sg.data) - 1 != NR_MSG_FRAG_IDS);
604326b4	177	memset(msg, 0, sizeof(*msg));
031097d9	178	sg_init_marker(msg->sg.data, NR_MSG_FRAG_IDS);
604326b4 DB	179	}
	180
	181	static inline void sk_msg_xfer(struct sk_msg dst, struct sk_msg src,
	182	int which, u32 size)
	183	{
	184	dst->sg.data[which] = src->sg.data[which];
	185	dst->sg.data[which].length = size;
3f4c3127	186	dst->sg.size += size;
81aabbb9	187	src->sg.size -= size;
604326b4 DB	188	src->sg.data[which].length -= size;
	189	src->sg.data[which].offset += size;
	190	}
	191
d3b18ad3 JF	192	static inline void sk_msg_xfer_full(struct sk_msg dst, struct sk_msg src)
	193	{
	194	memcpy(dst, src, sizeof(*src));
	195	sk_msg_init(src);
	196	}
	197
8734a162 JF	198	static inline bool sk_msg_full(const struct sk_msg *msg)
8734a162 JF	199	{
031097d9	200	return sk_msg_iter_dist(msg->sg.start, msg->sg.end) == MAX_MSG_FRAGS;
8734a162 JF	201	}
8734a162 JF	202
604326b4 DB	203	static inline u32 sk_msg_elem_used(const struct sk_msg *msg)
604326b4 DB	204	{
683916f6	205	return sk_msg_iter_dist(msg->sg.start, msg->sg.end);
604326b4 DB	206	}
604326b4 DB	207
604326b4 DB	208	static inline struct scatterlist sk_msg_elem(struct sk_msg msg, int which)
	209	{
	210	return &msg->sg.data[which];
	211	}
	212
6fff607e JF	213	static inline struct scatterlist sk_msg_elem_cpy(struct sk_msg *msg, int which)
	214	{
	215	return msg->sg.data[which];
	216	}
	217
604326b4 DB	218	static inline struct page sk_msg_page(struct sk_msg msg, int which)
	219	{
	220	return sg_page(sk_msg_elem(msg, which));
	221	}
	222
	223	static inline bool sk_msg_to_ingress(const struct sk_msg *msg)
	224	{
	225	return msg->flags & BPF_F_INGRESS;
	226	}
	227
	228	static inline void sk_msg_compute_data_pointers(struct sk_msg *msg)
	229	{
	230	struct scatterlist *sge = sk_msg_elem(msg, msg->sg.start);
	231
5a8fb33e	232	if (test_bit(msg->sg.start, msg->sg.copy)) {
604326b4 DB	233	msg->data = NULL;
	234	msg->data_end = NULL;
	235	} else {
	236	msg->data = sg_virt(sge);
	237	msg->data_end = msg->data + sge->length;
	238	}
	239	}
	240
	241	static inline void sk_msg_page_add(struct sk_msg msg, struct page page,
	242	u32 len, u32 offset)
	243	{
	244	struct scatterlist *sge;
	245
	246	get_page(page);
	247	sge = sk_msg_elem(msg, msg->sg.end);
	248	sg_set_page(sge, page, len, offset);
	249	sg_unmark_end(sge);
	250
5a8fb33e	251	__set_bit(msg->sg.end, msg->sg.copy);
604326b4 DB	252	msg->sg.size += len;
	253	sk_msg_iter_next(msg, end);
	254	}
	255
d3b18ad3 JF	256	static inline void sk_msg_sg_copy(struct sk_msg *msg, u32 i, bool copy_state)
	257	{
	258	do {
163ab96b	259	if (copy_state)
5a8fb33e	260	__set_bit(i, msg->sg.copy);
163ab96b	261	else
5a8fb33e	262	__clear_bit(i, msg->sg.copy);
d3b18ad3 JF	263	sk_msg_iter_var_next(i);
	264	if (i == msg->sg.end)
	265	break;
	266	} while (1);
	267	}
	268
	269	static inline void sk_msg_sg_copy_set(struct sk_msg *msg, u32 start)
	270	{
	271	sk_msg_sg_copy(msg, start, true);
	272	}
	273
	274	static inline void sk_msg_sg_copy_clear(struct sk_msg *msg, u32 start)
	275	{
	276	sk_msg_sg_copy(msg, start, false);
	277	}
	278
604326b4 DB	279	static inline struct sk_psock sk_psock(const struct sock sk)
604326b4 DB	280	{
2a013372 HJ	281	return __rcu_dereference_sk_user_data_with_flags(sk,
2a013372 HJ	282	SK_USER_DATA_PSOCK);
604326b4 DB	283	}
604326b4 DB	284
9635720b JF	285	static inline void sk_psock_set_state(struct sk_psock *psock,
	286	enum sk_psock_state_bits bit)
	287	{
	288	set_bit(bit, &psock->state);
	289	}
	290
	291	static inline void sk_psock_clear_state(struct sk_psock *psock,
	292	enum sk_psock_state_bits bit)
	293	{
	294	clear_bit(bit, &psock->state);
	295	}
	296
	297	static inline bool sk_psock_test_state(const struct sk_psock *psock,
	298	enum sk_psock_state_bits bit)
	299	{
	300	return test_bit(bit, &psock->state);
	301	}
	302
	303	static inline void sock_drop(struct sock sk, struct sk_buff skb)
	304	{
	305	sk_drops_add(sk, skb);
	306	kfree_skb(skb);
	307	}
	308
604326b4 DB	309	static inline void sk_psock_queue_msg(struct sk_psock *psock,
	310	struct sk_msg *msg)
	311	{
b01fd6e8	312	spin_lock_bh(&psock->ingress_lock);
9635720b JF	313	if (sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED))
9635720b JF	314	list_add_tail(&msg->list, &psock->ingress_msg);
938d3480 WY	315	else {
	316	sk_msg_free(psock->sk, msg);
	317	kfree(msg);
	318	}
b01fd6e8 CW	319	spin_unlock_bh(&psock->ingress_lock);
	320	}
	321
	322	static inline struct sk_msg sk_psock_dequeue_msg(struct sk_psock psock)
	323	{
	324	struct sk_msg *msg;
	325
	326	spin_lock_bh(&psock->ingress_lock);
	327	msg = list_first_entry_or_null(&psock->ingress_msg, struct sk_msg, list);
	328	if (msg)
	329	list_del(&msg->list);
	330	spin_unlock_bh(&psock->ingress_lock);
	331	return msg;
	332	}
	333
	334	static inline struct sk_msg sk_psock_peek_msg(struct sk_psock psock)
	335	{
	336	struct sk_msg *msg;
	337
	338	spin_lock_bh(&psock->ingress_lock);
	339	msg = list_first_entry_or_null(&psock->ingress_msg, struct sk_msg, list);
	340	spin_unlock_bh(&psock->ingress_lock);
	341	return msg;
	342	}
	343
	344	static inline struct sk_msg sk_psock_next_msg(struct sk_psock psock,
	345	struct sk_msg *msg)
	346	{
	347	struct sk_msg *ret;
	348
	349	spin_lock_bh(&psock->ingress_lock);
	350	if (list_is_last(&msg->list, &psock->ingress_msg))
	351	ret = NULL;
	352	else
	353	ret = list_next_entry(msg, list);
	354	spin_unlock_bh(&psock->ingress_lock);
	355	return ret;
604326b4 DB	356	}
604326b4 DB	357
d3b18ad3 JF	358	static inline bool sk_psock_queue_empty(const struct sk_psock *psock)
	359	{
	360	return psock ? list_empty(&psock->ingress_msg) : true;
	361	}
	362
b01fd6e8 CW	363	static inline void kfree_sk_msg(struct sk_msg *msg)
	364	{
	365	if (msg->skb)
	366	consume_skb(msg->skb);
	367	kfree(msg);
	368	}
	369
604326b4 DB	370	static inline void sk_psock_report_error(struct sk_psock *psock, int err)
	371	{
	372	struct sock *sk = psock->sk;
	373
	374	sk->sk_err = err;
e3ae2365	375	sk_error_report(sk);
604326b4 DB	376	}
	377
	378	struct sk_psock sk_psock_init(struct sock sk, int node);
799aa7f9	379	void sk_psock_stop(struct sk_psock *psock, bool wait);
604326b4	380
88759609	381	#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
604326b4 DB	382	int sk_psock_init_strp(struct sock sk, struct sk_psock psock);
	383	void sk_psock_start_strp(struct sock sk, struct sk_psock psock);
	384	void sk_psock_stop_strp(struct sock sk, struct sk_psock psock);
88759609 CW	385	#else
	386	static inline int sk_psock_init_strp(struct sock sk, struct sk_psock psock)
	387	{
	388	return -EOPNOTSUPP;
	389	}
	390
	391	static inline void sk_psock_start_strp(struct sock sk, struct sk_psock psock)
	392	{
	393	}
	394
	395	static inline void sk_psock_stop_strp(struct sock sk, struct sk_psock psock)
	396	{
	397	}
	398	#endif
	399
ef565928 JF	400	void sk_psock_start_verdict(struct sock sk, struct sk_psock psock);
ef565928 JF	401	void sk_psock_stop_verdict(struct sock sk, struct sk_psock psock);
604326b4 DB	402
	403	int sk_psock_msg_verdict(struct sock sk, struct sk_psock psock,
	404	struct sk_msg *msg);
	405
	406	static inline struct sk_psock_link *sk_psock_init_link(void)
	407	{
	408	return kzalloc(sizeof(struct sk_psock_link),
	409	GFP_ATOMIC \| __GFP_NOWARN);
	410	}
	411
	412	static inline void sk_psock_free_link(struct sk_psock_link *link)
	413	{
	414	kfree(link);
	415	}
	416
	417	struct sk_psock_link sk_psock_link_pop(struct sk_psock psock);
604326b4	418
604326b4 DB	419	static inline void sk_psock_cork_free(struct sk_psock *psock)
	420	{
	421	if (psock->cork) {
	422	sk_msg_free(psock->sk, psock->cork);
	423	kfree(psock->cork);
	424	psock->cork = NULL;
	425	}
	426	}
	427
604326b4 DB	428	static inline void sk_psock_restore_proto(struct sock *sk,
	429	struct sk_psock *psock)
	430	{
8a59f9d1	431	if (psock->psock_update_sk_prot)
51e0158a	432	psock->psock_update_sk_prot(sk, psock, true);
604326b4 DB	433	}
604326b4 DB	434
604326b4 DB	435	static inline struct sk_psock sk_psock_get(struct sock sk)
	436	{
	437	struct sk_psock *psock;
	438
	439	rcu_read_lock();
	440	psock = sk_psock(sk);
	441	if (psock && !refcount_inc_not_zero(&psock->refcnt))
	442	psock = NULL;
	443	rcu_read_unlock();
	444	return psock;
	445	}
	446
604326b4 DB	447	void sk_psock_drop(struct sock sk, struct sk_psock psock);
	448
	449	static inline void sk_psock_put(struct sock sk, struct sk_psock psock)
	450	{
	451	if (refcount_dec_and_test(&psock->refcnt))
	452	sk_psock_drop(sk, psock);
	453	}
	454
552de910 JF	455	static inline void sk_psock_data_ready(struct sock sk, struct sk_psock psock)
552de910 JF	456	{
5a685cd9 CW	457	if (psock->saved_data_ready)
5a685cd9 CW	458	psock->saved_data_ready(sk);
552de910 JF	459	else
	460	sk->sk_data_ready(sk);
	461	}
	462
604326b4 DB	463	static inline void psock_set_prog(struct bpf_prog **pprog,
	464	struct bpf_prog *prog)
	465	{
	466	prog = xchg(pprog, prog);
	467	if (prog)
	468	bpf_prog_put(prog);
	469	}
	470
bb0de313 LB	471	static inline int psock_replace_prog(struct bpf_prog **pprog,
	472	struct bpf_prog *prog,
	473	struct bpf_prog *old)
	474	{
	475	if (cmpxchg(pprog, old, prog) != old)
	476	return -ENOENT;
	477
	478	if (old)
	479	bpf_prog_put(old);
	480
	481	return 0;
	482	}
	483
604326b4 DB	484	static inline void psock_progs_drop(struct sk_psock_progs *progs)
	485	{
	486	psock_set_prog(&progs->msg_parser, NULL);
ae8b8332 CW	487	psock_set_prog(&progs->stream_parser, NULL);
ae8b8332 CW	488	psock_set_prog(&progs->stream_verdict, NULL);
a7ba4558	489	psock_set_prog(&progs->skb_verdict, NULL);
604326b4 DB	490	}
604326b4 DB	491
e91de6af JF	492	int sk_psock_tls_strp_read(struct sk_psock psock, struct sk_buff skb);
	493
	494	static inline bool sk_psock_strp_enabled(struct sk_psock *psock)
	495	{
	496	if (!psock)
	497	return false;
5a685cd9	498	return !!psock->saved_data_ready;
e91de6af	499	}
e3526bb9	500
40a34121 JF	501	static inline bool sk_is_udp(const struct sock *sk)
	502	{
	503	return sk->sk_type == SOCK_DGRAM &&
	504	sk->sk_protocol == IPPROTO_UDP;
	505	}
	506
e3526bb9 CW	507	#if IS_ENABLED(CONFIG_NET_SOCK_MSG)
e3526bb9 CW	508
7303524e LJ	509	#define BPF_F_STRPARSER (1UL << 1)
	510
	511	/* We only have two bits so far. */
	512	#define BPF_F_PTR_MASK ~(BPF_F_INGRESS \| BPF_F_STRPARSER)
	513
	514	static inline bool skb_bpf_strparser(const struct sk_buff *skb)
	515	{
	516	unsigned long sk_redir = skb->_sk_redir;
	517
	518	return sk_redir & BPF_F_STRPARSER;
	519	}
	520
	521	static inline void skb_bpf_set_strparser(struct sk_buff *skb)
	522	{
	523	skb->_sk_redir \|= BPF_F_STRPARSER;
	524	}
e3526bb9 CW	525
	526	static inline bool skb_bpf_ingress(const struct sk_buff *skb)
	527	{
	528	unsigned long sk_redir = skb->_sk_redir;
	529
	530	return sk_redir & BPF_F_INGRESS;
	531	}
	532
	533	static inline void skb_bpf_set_ingress(struct sk_buff *skb)
	534	{
	535	skb->_sk_redir \|= BPF_F_INGRESS;
	536	}
	537
	538	static inline void skb_bpf_set_redir(struct sk_buff skb, struct sock sk_redir,
	539	bool ingress)
	540	{
	541	skb->_sk_redir = (unsigned long)sk_redir;
	542	if (ingress)
	543	skb->_sk_redir \|= BPF_F_INGRESS;
	544	}
	545
	546	static inline struct sock skb_bpf_redirect_fetch(const struct sk_buff skb)
	547	{
	548	unsigned long sk_redir = skb->_sk_redir;
	549
	550	return (struct sock *)(sk_redir & BPF_F_PTR_MASK);
	551	}
	552
	553	static inline void skb_bpf_redirect_clear(struct sk_buff *skb)
	554	{
	555	skb->_sk_redir = 0;
	556	}
	557	#endif /* CONFIG_NET_SOCK_MSG */
604326b4	558	#endif /* _LINUX_SKMSG_H */