[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;
	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_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_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_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 drop_sk_msg(struct sk_psock *psock, struct sk_msg *msg)
{
	if (msg->skb)
		sock_drop(psock->sk, msg->skb);
	kfree(msg);
}

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
		drop_sk_msg(psock, 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;
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;
a7ba4558	61	struct bpf_prog *skb_verdict;
604326b4 DB	62	};
	63
	64	enum sk_psock_state_bits {
	65	SK_PSOCK_TX_ENABLED,
	66	};
	67
	68	struct sk_psock_link {
	69	struct list_head list;
	70	struct bpf_map *map;
	71	void *link_raw;
	72	};
	73
604326b4 DB	74	struct sk_psock_work_state {
	75	struct sk_buff *skb;
	76	u32 len;
	77	u32 off;
	78	};
	79
	80	struct sk_psock {
	81	struct sock *sk;
	82	struct sock *sk_redir;
	83	u32 apply_bytes;
	84	u32 cork_bytes;
	85	u32 eval;
	86	struct sk_msg *cork;
	87	struct sk_psock_progs progs;
5a685cd9 CW	88	#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
	89	struct strparser strp;
	90	#endif
604326b4 DB	91	struct sk_buff_head ingress_skb;
604326b4 DB	92	struct list_head ingress_msg;
b01fd6e8	93	spinlock_t ingress_lock;
604326b4 DB	94	unsigned long state;
	95	struct list_head link;
	96	spinlock_t link_lock;
	97	refcount_t refcnt;
	98	void (saved_unhash)(struct sock sk);
	99	void (saved_close)(struct sock sk, long timeout);
	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
	174	static inline void sk_msg_clear_meta(struct sk_msg *msg)
	175	{
	176	memset(&msg->sg, 0, offsetofend(struct sk_msg_sg, copy));
	177	}
	178
	179	static inline void sk_msg_init(struct sk_msg *msg)
	180	{
031097d9	181	BUILD_BUG_ON(ARRAY_SIZE(msg->sg.data) - 1 != NR_MSG_FRAG_IDS);
604326b4	182	memset(msg, 0, sizeof(*msg));
031097d9	183	sg_init_marker(msg->sg.data, NR_MSG_FRAG_IDS);
604326b4 DB	184	}
	185
	186	static inline void sk_msg_xfer(struct sk_msg dst, struct sk_msg src,
	187	int which, u32 size)
	188	{
	189	dst->sg.data[which] = src->sg.data[which];
	190	dst->sg.data[which].length = size;
3f4c3127	191	dst->sg.size += size;
81aabbb9	192	src->sg.size -= size;
604326b4 DB	193	src->sg.data[which].length -= size;
	194	src->sg.data[which].offset += size;
	195	}
	196
d3b18ad3 JF	197	static inline void sk_msg_xfer_full(struct sk_msg dst, struct sk_msg src)
	198	{
	199	memcpy(dst, src, sizeof(*src));
	200	sk_msg_init(src);
	201	}
	202
8734a162 JF	203	static inline bool sk_msg_full(const struct sk_msg *msg)
8734a162 JF	204	{
031097d9	205	return sk_msg_iter_dist(msg->sg.start, msg->sg.end) == MAX_MSG_FRAGS;
8734a162 JF	206	}
8734a162 JF	207
604326b4 DB	208	static inline u32 sk_msg_elem_used(const struct sk_msg *msg)
604326b4 DB	209	{
683916f6	210	return sk_msg_iter_dist(msg->sg.start, msg->sg.end);
604326b4 DB	211	}
604326b4 DB	212
604326b4 DB	213	static inline struct scatterlist sk_msg_elem(struct sk_msg msg, int which)
	214	{
	215	return &msg->sg.data[which];
	216	}
	217
6fff607e JF	218	static inline struct scatterlist sk_msg_elem_cpy(struct sk_msg *msg, int which)
	219	{
	220	return msg->sg.data[which];
	221	}
	222
604326b4 DB	223	static inline struct page sk_msg_page(struct sk_msg msg, int which)
	224	{
	225	return sg_page(sk_msg_elem(msg, which));
	226	}
	227
	228	static inline bool sk_msg_to_ingress(const struct sk_msg *msg)
	229	{
	230	return msg->flags & BPF_F_INGRESS;
	231	}
	232
	233	static inline void sk_msg_compute_data_pointers(struct sk_msg *msg)
	234	{
	235	struct scatterlist *sge = sk_msg_elem(msg, msg->sg.start);
	236
163ab96b	237	if (test_bit(msg->sg.start, &msg->sg.copy)) {
604326b4 DB	238	msg->data = NULL;
	239	msg->data_end = NULL;
	240	} else {
	241	msg->data = sg_virt(sge);
	242	msg->data_end = msg->data + sge->length;
	243	}
	244	}
	245
	246	static inline void sk_msg_page_add(struct sk_msg msg, struct page page,
	247	u32 len, u32 offset)
	248	{
	249	struct scatterlist *sge;
	250
	251	get_page(page);
	252	sge = sk_msg_elem(msg, msg->sg.end);
	253	sg_set_page(sge, page, len, offset);
	254	sg_unmark_end(sge);
	255
163ab96b	256	__set_bit(msg->sg.end, &msg->sg.copy);
604326b4 DB	257	msg->sg.size += len;
	258	sk_msg_iter_next(msg, end);
	259	}
	260
d3b18ad3 JF	261	static inline void sk_msg_sg_copy(struct sk_msg *msg, u32 i, bool copy_state)
	262	{
	263	do {
163ab96b JK	264	if (copy_state)
	265	__set_bit(i, &msg->sg.copy);
	266	else
	267	__clear_bit(i, &msg->sg.copy);
d3b18ad3 JF	268	sk_msg_iter_var_next(i);
	269	if (i == msg->sg.end)
	270	break;
	271	} while (1);
	272	}
	273
	274	static inline void sk_msg_sg_copy_set(struct sk_msg *msg, u32 start)
	275	{
	276	sk_msg_sg_copy(msg, start, true);
	277	}
	278
	279	static inline void sk_msg_sg_copy_clear(struct sk_msg *msg, u32 start)
	280	{
	281	sk_msg_sg_copy(msg, start, false);
	282	}
	283
604326b4 DB	284	static inline struct sk_psock sk_psock(const struct sock sk)
	285	{
	286	return rcu_dereference_sk_user_data(sk);
	287	}
	288
9635720b JF	289	static inline void sk_psock_set_state(struct sk_psock *psock,
	290	enum sk_psock_state_bits bit)
	291	{
	292	set_bit(bit, &psock->state);
	293	}
	294
	295	static inline void sk_psock_clear_state(struct sk_psock *psock,
	296	enum sk_psock_state_bits bit)
	297	{
	298	clear_bit(bit, &psock->state);
	299	}
	300
	301	static inline bool sk_psock_test_state(const struct sk_psock *psock,
	302	enum sk_psock_state_bits bit)
	303	{
	304	return test_bit(bit, &psock->state);
	305	}
	306
	307	static inline void sock_drop(struct sock sk, struct sk_buff skb)
	308	{
	309	sk_drops_add(sk, skb);
	310	kfree_skb(skb);
	311	}
	312
	313	static inline void drop_sk_msg(struct sk_psock psock, struct sk_msg msg)
	314	{
	315	if (msg->skb)
	316	sock_drop(psock->sk, msg->skb);
	317	kfree(msg);
	318	}
	319
604326b4 DB	320	static inline void sk_psock_queue_msg(struct sk_psock *psock,
	321	struct sk_msg *msg)
	322	{
b01fd6e8	323	spin_lock_bh(&psock->ingress_lock);
9635720b JF	324	if (sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED))
	325	list_add_tail(&msg->list, &psock->ingress_msg);
	326	else
	327	drop_sk_msg(psock, msg);
b01fd6e8 CW	328	spin_unlock_bh(&psock->ingress_lock);
	329	}
	330
	331	static inline struct sk_msg sk_psock_dequeue_msg(struct sk_psock psock)
	332	{
	333	struct sk_msg *msg;
	334
	335	spin_lock_bh(&psock->ingress_lock);
	336	msg = list_first_entry_or_null(&psock->ingress_msg, struct sk_msg, list);
	337	if (msg)
	338	list_del(&msg->list);
	339	spin_unlock_bh(&psock->ingress_lock);
	340	return msg;
	341	}
	342
	343	static inline struct sk_msg sk_psock_peek_msg(struct sk_psock psock)
	344	{
	345	struct sk_msg *msg;
	346
	347	spin_lock_bh(&psock->ingress_lock);
	348	msg = list_first_entry_or_null(&psock->ingress_msg, struct sk_msg, list);
	349	spin_unlock_bh(&psock->ingress_lock);
	350	return msg;
	351	}
	352
	353	static inline struct sk_msg sk_psock_next_msg(struct sk_psock psock,
	354	struct sk_msg *msg)
	355	{
	356	struct sk_msg *ret;
	357
	358	spin_lock_bh(&psock->ingress_lock);
	359	if (list_is_last(&msg->list, &psock->ingress_msg))
	360	ret = NULL;
	361	else
	362	ret = list_next_entry(msg, list);
	363	spin_unlock_bh(&psock->ingress_lock);
	364	return ret;
604326b4 DB	365	}
604326b4 DB	366
d3b18ad3 JF	367	static inline bool sk_psock_queue_empty(const struct sk_psock *psock)
	368	{
	369	return psock ? list_empty(&psock->ingress_msg) : true;
	370	}
	371
b01fd6e8 CW	372	static inline void kfree_sk_msg(struct sk_msg *msg)
	373	{
	374	if (msg->skb)
	375	consume_skb(msg->skb);
	376	kfree(msg);
	377	}
	378
604326b4 DB	379	static inline void sk_psock_report_error(struct sk_psock *psock, int err)
	380	{
	381	struct sock *sk = psock->sk;
	382
	383	sk->sk_err = err;
e3ae2365	384	sk_error_report(sk);
604326b4 DB	385	}
	386
	387	struct sk_psock sk_psock_init(struct sock sk, int node);
799aa7f9	388	void sk_psock_stop(struct sk_psock *psock, bool wait);
604326b4	389
88759609	390	#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
604326b4 DB	391	int sk_psock_init_strp(struct sock sk, struct sk_psock psock);
	392	void sk_psock_start_strp(struct sock sk, struct sk_psock psock);
	393	void sk_psock_stop_strp(struct sock sk, struct sk_psock psock);
88759609 CW	394	#else
	395	static inline int sk_psock_init_strp(struct sock sk, struct sk_psock psock)
	396	{
	397	return -EOPNOTSUPP;
	398	}
	399
	400	static inline void sk_psock_start_strp(struct sock sk, struct sk_psock psock)
	401	{
	402	}
	403
	404	static inline void sk_psock_stop_strp(struct sock sk, struct sk_psock psock)
	405	{
	406	}
	407	#endif
	408
ef565928 JF	409	void sk_psock_start_verdict(struct sock sk, struct sk_psock psock);
ef565928 JF	410	void sk_psock_stop_verdict(struct sock sk, struct sk_psock psock);
604326b4 DB	411
	412	int sk_psock_msg_verdict(struct sock sk, struct sk_psock psock,
	413	struct sk_msg *msg);
	414
	415	static inline struct sk_psock_link *sk_psock_init_link(void)
	416	{
	417	return kzalloc(sizeof(struct sk_psock_link),
	418	GFP_ATOMIC \| __GFP_NOWARN);
	419	}
	420
	421	static inline void sk_psock_free_link(struct sk_psock_link *link)
	422	{
	423	kfree(link);
	424	}
	425
	426	struct sk_psock_link sk_psock_link_pop(struct sk_psock psock);
604326b4	427
604326b4 DB	428	static inline void sk_psock_cork_free(struct sk_psock *psock)
	429	{
	430	if (psock->cork) {
	431	sk_msg_free(psock->sk, psock->cork);
	432	kfree(psock->cork);
	433	psock->cork = NULL;
	434	}
	435	}
	436
604326b4 DB	437	static inline void sk_psock_restore_proto(struct sock *sk,
	438	struct sk_psock *psock)
	439	{
8a59f9d1	440	if (psock->psock_update_sk_prot)
51e0158a	441	psock->psock_update_sk_prot(sk, psock, true);
604326b4 DB	442	}
604326b4 DB	443
604326b4 DB	444	static inline struct sk_psock sk_psock_get(struct sock sk)
	445	{
	446	struct sk_psock *psock;
	447
	448	rcu_read_lock();
	449	psock = sk_psock(sk);
	450	if (psock && !refcount_inc_not_zero(&psock->refcnt))
	451	psock = NULL;
	452	rcu_read_unlock();
	453	return psock;
	454	}
	455
604326b4 DB	456	void sk_psock_drop(struct sock sk, struct sk_psock psock);
	457
	458	static inline void sk_psock_put(struct sock sk, struct sk_psock psock)
	459	{
	460	if (refcount_dec_and_test(&psock->refcnt))
	461	sk_psock_drop(sk, psock);
	462	}
	463
552de910 JF	464	static inline void sk_psock_data_ready(struct sock sk, struct sk_psock psock)
552de910 JF	465	{
5a685cd9 CW	466	if (psock->saved_data_ready)
5a685cd9 CW	467	psock->saved_data_ready(sk);
552de910 JF	468	else
	469	sk->sk_data_ready(sk);
	470	}
	471
604326b4 DB	472	static inline void psock_set_prog(struct bpf_prog **pprog,
	473	struct bpf_prog *prog)
	474	{
	475	prog = xchg(pprog, prog);
	476	if (prog)
	477	bpf_prog_put(prog);
	478	}
	479
bb0de313 LB	480	static inline int psock_replace_prog(struct bpf_prog **pprog,
	481	struct bpf_prog *prog,
	482	struct bpf_prog *old)
	483	{
	484	if (cmpxchg(pprog, old, prog) != old)
	485	return -ENOENT;
	486
	487	if (old)
	488	bpf_prog_put(old);
	489
	490	return 0;
	491	}
	492
604326b4 DB	493	static inline void psock_progs_drop(struct sk_psock_progs *progs)
	494	{
	495	psock_set_prog(&progs->msg_parser, NULL);
ae8b8332 CW	496	psock_set_prog(&progs->stream_parser, NULL);
ae8b8332 CW	497	psock_set_prog(&progs->stream_verdict, NULL);
a7ba4558	498	psock_set_prog(&progs->skb_verdict, NULL);
604326b4 DB	499	}
604326b4 DB	500
e91de6af JF	501	int sk_psock_tls_strp_read(struct sk_psock psock, struct sk_buff skb);
	502
	503	static inline bool sk_psock_strp_enabled(struct sk_psock *psock)
	504	{
	505	if (!psock)
	506	return false;
5a685cd9	507	return !!psock->saved_data_ready;
e91de6af	508	}
e3526bb9	509
40a34121 JF	510	static inline bool sk_is_udp(const struct sock *sk)
	511	{
	512	return sk->sk_type == SOCK_DGRAM &&
	513	sk->sk_protocol == IPPROTO_UDP;
	514	}
	515
e3526bb9 CW	516	#if IS_ENABLED(CONFIG_NET_SOCK_MSG)
e3526bb9 CW	517
7303524e LJ	518	#define BPF_F_STRPARSER (1UL << 1)
	519
	520	/* We only have two bits so far. */
	521	#define BPF_F_PTR_MASK ~(BPF_F_INGRESS \| BPF_F_STRPARSER)
	522
	523	static inline bool skb_bpf_strparser(const struct sk_buff *skb)
	524	{
	525	unsigned long sk_redir = skb->_sk_redir;
	526
	527	return sk_redir & BPF_F_STRPARSER;
	528	}
	529
	530	static inline void skb_bpf_set_strparser(struct sk_buff *skb)
	531	{
	532	skb->_sk_redir \|= BPF_F_STRPARSER;
	533	}
e3526bb9 CW	534
	535	static inline bool skb_bpf_ingress(const struct sk_buff *skb)
	536	{
	537	unsigned long sk_redir = skb->_sk_redir;
	538
	539	return sk_redir & BPF_F_INGRESS;
	540	}
	541
	542	static inline void skb_bpf_set_ingress(struct sk_buff *skb)
	543	{
	544	skb->_sk_redir \|= BPF_F_INGRESS;
	545	}
	546
	547	static inline void skb_bpf_set_redir(struct sk_buff skb, struct sock sk_redir,
	548	bool ingress)
	549	{
	550	skb->_sk_redir = (unsigned long)sk_redir;
	551	if (ingress)
	552	skb->_sk_redir \|= BPF_F_INGRESS;
	553	}
	554
	555	static inline struct sock skb_bpf_redirect_fetch(const struct sk_buff skb)
	556	{
	557	unsigned long sk_redir = skb->_sk_redir;
	558
	559	return (struct sock *)(sk_redir & BPF_F_PTR_MASK);
	560	}
	561
	562	static inline void skb_bpf_redirect_clear(struct sk_buff *skb)
	563	{
	564	skb->_sk_redir = 0;
	565	}
	566	#endif /* CONFIG_NET_SOCK_MSG */
604326b4	567	#endif /* _LINUX_SKMSG_H */