[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,
	SK_PSOCK_RX_STRP_ENABLED,
};

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

struct sk_psock_work_state {
	u32				len;
	u32				off;
};

struct sk_psock {
	struct sock			*sk;
	struct sock			*sk_redir;
	u32				apply_bytes;
	u32				cork_bytes;
	u32				eval;
	bool				redir_ingress; /* undefined if sk_redir is null */
	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);
	/* psock_update_sk_prot may be called with restore=false many times
	 * so the handler must be safe for this case. It will be called
	 * exactly once with restore=true when the psock is being destroyed
	 * and psock refcnt is zero, but before an RCU grace period.
	 */
	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 delayed_work		work;
	struct sock			*sk_pair;
	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);

#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)
{
	read_lock_bh(&sk->sk_callback_lock);
	if (psock->saved_data_ready)
		psock->saved_data_ready(sk);
	else
		sk->sk_data_ready(sk);
	read_unlock_bh(&sk->sk_callback_lock);
}

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;
}

#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,
809e4dc7	65	SK_PSOCK_RX_STRP_ENABLED,
604326b4 DB	66	};
	67
	68	struct sk_psock_link {
	69	struct list_head list;
	70	struct bpf_map *map;
	71	void *link_raw;
	72	};
	73
604326b4	74	struct sk_psock_work_state {
604326b4 DB	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;
a351d608	85	bool redir_ingress; /* undefined if sk_redir is null */
604326b4 DB	86	struct sk_msg *cork;
604326b4 DB	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);
d8616ee2	99	void (saved_destroy)(struct sock sk);
604326b4 DB	100	void (saved_close)(struct sock sk, long timeout);
604326b4 DB	101	void (saved_write_space)(struct sock sk);
5a685cd9	102	void (saved_data_ready)(struct sock sk);
7865dfb1 JF	103	/* psock_update_sk_prot may be called with restore=false many times
	104	* so the handler must be safe for this case. It will be called
	105	* exactly once with restore=true when the psock is being destroyed
	106	* and psock refcnt is zero, but before an RCU grace period.
	107	*/
51e0158a CW	108	int (psock_update_sk_prot)(struct sock sk, struct sk_psock *psock,
51e0158a CW	109	bool restore);
604326b4	110	struct proto *sk_proto;
799aa7f9	111	struct mutex work_mutex;
604326b4	112	struct sk_psock_work_state work_state;
29173d07	113	struct delayed_work work;
8866730a	114	struct sock *sk_pair;
7786dfc4	115	struct rcu_work rwork;
604326b4 DB	116	};
	117
	118	int sk_msg_alloc(struct sock sk, struct sk_msg msg, int len,
	119	int elem_first_coalesce);
d829e9c4 DB	120	int sk_msg_clone(struct sock sk, struct sk_msg dst, struct sk_msg *src,
d829e9c4 DB	121	u32 off, u32 len);
604326b4 DB	122	void sk_msg_trim(struct sock sk, struct sk_msg msg, int len);
	123	int sk_msg_free(struct sock sk, struct sk_msg msg);
	124	int sk_msg_free_nocharge(struct sock sk, struct sk_msg msg);
	125	void sk_msg_free_partial(struct sock sk, struct sk_msg msg, u32 bytes);
	126	void sk_msg_free_partial_nocharge(struct sock sk, struct sk_msg msg,
	127	u32 bytes);
	128
	129	void sk_msg_return(struct sock sk, struct sk_msg msg, int bytes);
d3b18ad3	130	void sk_msg_return_zero(struct sock sk, struct sk_msg msg, int bytes);
604326b4 DB	131
	132	int sk_msg_zerocopy_from_iter(struct sock sk, struct iov_iter from,
	133	struct sk_msg *msg, u32 bytes);
	134	int sk_msg_memcopy_from_iter(struct sock sk, struct iov_iter from,
	135	struct sk_msg *msg, u32 bytes);
2bc793e3 CW	136	int sk_msg_recvmsg(struct sock sk, struct sk_psock psock, struct msghdr *msg,
2bc793e3 CW	137	int len, int flags);
fb4e0a5e	138	bool sk_msg_is_readable(struct sock *sk);
604326b4 DB	139
	140	static inline void sk_msg_check_to_free(struct sk_msg *msg, u32 i, u32 bytes)
	141	{
	142	WARN_ON(i == msg->sg.end && bytes);
	143	}
	144
	145	static inline void sk_msg_apply_bytes(struct sk_psock *psock, u32 bytes)
	146	{
	147	if (psock->apply_bytes) {
	148	if (psock->apply_bytes < bytes)
	149	psock->apply_bytes = 0;
	150	else
	151	psock->apply_bytes -= bytes;
	152	}
	153	}
	154
683916f6 JK	155	static inline u32 sk_msg_iter_dist(u32 start, u32 end)
683916f6 JK	156	{
031097d9	157	return end >= start ? end - start : end + (NR_MSG_FRAG_IDS - start);
683916f6 JK	158	}
683916f6 JK	159
604326b4 DB	160	#define sk_msg_iter_var_prev(var) \
	161	do { \
	162	if (var == 0) \
031097d9	163	var = NR_MSG_FRAG_IDS - 1; \
604326b4 DB	164	else \
	165	var--; \
	166	} while (0)
	167
	168	#define sk_msg_iter_var_next(var) \
	169	do { \
	170	var++; \
031097d9	171	if (var == NR_MSG_FRAG_IDS) \
604326b4 DB	172	var = 0; \
	173	} while (0)
	174
	175	#define sk_msg_iter_prev(msg, which) \
	176	sk_msg_iter_var_prev(msg->sg.which)
	177
	178	#define sk_msg_iter_next(msg, which) \
	179	sk_msg_iter_var_next(msg->sg.which)
	180
604326b4 DB	181	static inline void sk_msg_init(struct sk_msg *msg)
604326b4 DB	182	{
031097d9	183	BUILD_BUG_ON(ARRAY_SIZE(msg->sg.data) - 1 != NR_MSG_FRAG_IDS);
604326b4	184	memset(msg, 0, sizeof(*msg));
031097d9	185	sg_init_marker(msg->sg.data, NR_MSG_FRAG_IDS);
604326b4 DB	186	}
	187
	188	static inline void sk_msg_xfer(struct sk_msg dst, struct sk_msg src,
	189	int which, u32 size)
	190	{
	191	dst->sg.data[which] = src->sg.data[which];
	192	dst->sg.data[which].length = size;
3f4c3127	193	dst->sg.size += size;
81aabbb9	194	src->sg.size -= size;
604326b4 DB	195	src->sg.data[which].length -= size;
	196	src->sg.data[which].offset += size;
	197	}
	198
d3b18ad3 JF	199	static inline void sk_msg_xfer_full(struct sk_msg dst, struct sk_msg src)
	200	{
	201	memcpy(dst, src, sizeof(*src));
	202	sk_msg_init(src);
	203	}
	204
8734a162 JF	205	static inline bool sk_msg_full(const struct sk_msg *msg)
8734a162 JF	206	{
031097d9	207	return sk_msg_iter_dist(msg->sg.start, msg->sg.end) == MAX_MSG_FRAGS;
8734a162 JF	208	}
8734a162 JF	209
604326b4 DB	210	static inline u32 sk_msg_elem_used(const struct sk_msg *msg)
604326b4 DB	211	{
683916f6	212	return sk_msg_iter_dist(msg->sg.start, msg->sg.end);
604326b4 DB	213	}
604326b4 DB	214
604326b4 DB	215	static inline struct scatterlist sk_msg_elem(struct sk_msg msg, int which)
	216	{
	217	return &msg->sg.data[which];
	218	}
	219
6fff607e JF	220	static inline struct scatterlist sk_msg_elem_cpy(struct sk_msg *msg, int which)
	221	{
	222	return msg->sg.data[which];
	223	}
	224
604326b4 DB	225	static inline struct page sk_msg_page(struct sk_msg msg, int which)
	226	{
	227	return sg_page(sk_msg_elem(msg, which));
	228	}
	229
	230	static inline bool sk_msg_to_ingress(const struct sk_msg *msg)
	231	{
	232	return msg->flags & BPF_F_INGRESS;
	233	}
	234
	235	static inline void sk_msg_compute_data_pointers(struct sk_msg *msg)
	236	{
	237	struct scatterlist *sge = sk_msg_elem(msg, msg->sg.start);
	238
5a8fb33e	239	if (test_bit(msg->sg.start, msg->sg.copy)) {
604326b4 DB	240	msg->data = NULL;
	241	msg->data_end = NULL;
	242	} else {
	243	msg->data = sg_virt(sge);
	244	msg->data_end = msg->data + sge->length;
	245	}
	246	}
	247
	248	static inline void sk_msg_page_add(struct sk_msg msg, struct page page,
	249	u32 len, u32 offset)
	250	{
	251	struct scatterlist *sge;
	252
	253	get_page(page);
	254	sge = sk_msg_elem(msg, msg->sg.end);
	255	sg_set_page(sge, page, len, offset);
	256	sg_unmark_end(sge);
	257
5a8fb33e	258	__set_bit(msg->sg.end, msg->sg.copy);
604326b4 DB	259	msg->sg.size += len;
	260	sk_msg_iter_next(msg, end);
	261	}
	262
d3b18ad3 JF	263	static inline void sk_msg_sg_copy(struct sk_msg *msg, u32 i, bool copy_state)
	264	{
	265	do {
163ab96b	266	if (copy_state)
5a8fb33e	267	__set_bit(i, msg->sg.copy);
163ab96b	268	else
5a8fb33e	269	__clear_bit(i, msg->sg.copy);
d3b18ad3 JF	270	sk_msg_iter_var_next(i);
	271	if (i == msg->sg.end)
	272	break;
	273	} while (1);
	274	}
	275
	276	static inline void sk_msg_sg_copy_set(struct sk_msg *msg, u32 start)
	277	{
	278	sk_msg_sg_copy(msg, start, true);
	279	}
	280
	281	static inline void sk_msg_sg_copy_clear(struct sk_msg *msg, u32 start)
	282	{
	283	sk_msg_sg_copy(msg, start, false);
	284	}
	285
604326b4 DB	286	static inline struct sk_psock sk_psock(const struct sock sk)
604326b4 DB	287	{
2a013372 HJ	288	return __rcu_dereference_sk_user_data_with_flags(sk,
2a013372 HJ	289	SK_USER_DATA_PSOCK);
604326b4 DB	290	}
604326b4 DB	291
9635720b JF	292	static inline void sk_psock_set_state(struct sk_psock *psock,
	293	enum sk_psock_state_bits bit)
	294	{
	295	set_bit(bit, &psock->state);
	296	}
	297
	298	static inline void sk_psock_clear_state(struct sk_psock *psock,
	299	enum sk_psock_state_bits bit)
	300	{
	301	clear_bit(bit, &psock->state);
	302	}
	303
	304	static inline bool sk_psock_test_state(const struct sk_psock *psock,
	305	enum sk_psock_state_bits bit)
	306	{
	307	return test_bit(bit, &psock->state);
	308	}
	309
	310	static inline void sock_drop(struct sock sk, struct sk_buff skb)
	311	{
	312	sk_drops_add(sk, skb);
	313	kfree_skb(skb);
	314	}
	315
604326b4 DB	316	static inline void sk_psock_queue_msg(struct sk_psock *psock,
	317	struct sk_msg *msg)
	318	{
b01fd6e8	319	spin_lock_bh(&psock->ingress_lock);
9635720b JF	320	if (sk_psock_test_state(psock, SK_PSOCK_TX_ENABLED))
9635720b JF	321	list_add_tail(&msg->list, &psock->ingress_msg);
938d3480 WY	322	else {
	323	sk_msg_free(psock->sk, msg);
	324	kfree(msg);
	325	}
b01fd6e8 CW	326	spin_unlock_bh(&psock->ingress_lock);
	327	}
	328
	329	static inline struct sk_msg sk_psock_dequeue_msg(struct sk_psock psock)
	330	{
	331	struct sk_msg *msg;
	332
	333	spin_lock_bh(&psock->ingress_lock);
	334	msg = list_first_entry_or_null(&psock->ingress_msg, struct sk_msg, list);
	335	if (msg)
	336	list_del(&msg->list);
	337	spin_unlock_bh(&psock->ingress_lock);
	338	return msg;
	339	}
	340
	341	static inline struct sk_msg sk_psock_peek_msg(struct sk_psock psock)
	342	{
	343	struct sk_msg *msg;
	344
	345	spin_lock_bh(&psock->ingress_lock);
	346	msg = list_first_entry_or_null(&psock->ingress_msg, struct sk_msg, list);
	347	spin_unlock_bh(&psock->ingress_lock);
	348	return msg;
	349	}
	350
	351	static inline struct sk_msg sk_psock_next_msg(struct sk_psock psock,
	352	struct sk_msg *msg)
	353	{
	354	struct sk_msg *ret;
	355
	356	spin_lock_bh(&psock->ingress_lock);
	357	if (list_is_last(&msg->list, &psock->ingress_msg))
	358	ret = NULL;
	359	else
	360	ret = list_next_entry(msg, list);
	361	spin_unlock_bh(&psock->ingress_lock);
	362	return ret;
604326b4 DB	363	}
604326b4 DB	364
d3b18ad3 JF	365	static inline bool sk_psock_queue_empty(const struct sk_psock *psock)
	366	{
	367	return psock ? list_empty(&psock->ingress_msg) : true;
	368	}
	369
b01fd6e8 CW	370	static inline void kfree_sk_msg(struct sk_msg *msg)
	371	{
	372	if (msg->skb)
	373	consume_skb(msg->skb);
	374	kfree(msg);
	375	}
	376
604326b4 DB	377	static inline void sk_psock_report_error(struct sk_psock *psock, int err)
	378	{
	379	struct sock *sk = psock->sk;
	380
	381	sk->sk_err = err;
e3ae2365	382	sk_error_report(sk);
604326b4 DB	383	}
	384
	385	struct sk_psock sk_psock_init(struct sock sk, int node);
8bbabb3f	386	void sk_psock_stop(struct sk_psock *psock);
604326b4	387
88759609	388	#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
604326b4 DB	389	int sk_psock_init_strp(struct sock sk, struct sk_psock psock);
	390	void sk_psock_start_strp(struct sock sk, struct sk_psock psock);
	391	void sk_psock_stop_strp(struct sock sk, struct sk_psock psock);
88759609 CW	392	#else
	393	static inline int sk_psock_init_strp(struct sock sk, struct sk_psock psock)
	394	{
	395	return -EOPNOTSUPP;
	396	}
	397
	398	static inline void sk_psock_start_strp(struct sock sk, struct sk_psock psock)
	399	{
	400	}
	401
	402	static inline void sk_psock_stop_strp(struct sock sk, struct sk_psock psock)
	403	{
	404	}
	405	#endif
	406
ef565928 JF	407	void sk_psock_start_verdict(struct sock sk, struct sk_psock psock);
ef565928 JF	408	void sk_psock_stop_verdict(struct sock sk, struct sk_psock psock);
604326b4 DB	409
	410	int sk_psock_msg_verdict(struct sock sk, struct sk_psock psock,
	411	struct sk_msg *msg);
	412
	413	static inline struct sk_psock_link *sk_psock_init_link(void)
	414	{
	415	return kzalloc(sizeof(struct sk_psock_link),
	416	GFP_ATOMIC \| __GFP_NOWARN);
	417	}
	418
	419	static inline void sk_psock_free_link(struct sk_psock_link *link)
	420	{
	421	kfree(link);
	422	}
	423
	424	struct sk_psock_link sk_psock_link_pop(struct sk_psock psock);
604326b4	425
604326b4 DB	426	static inline void sk_psock_cork_free(struct sk_psock *psock)
	427	{
	428	if (psock->cork) {
	429	sk_msg_free(psock->sk, psock->cork);
	430	kfree(psock->cork);
	431	psock->cork = NULL;
	432	}
	433	}
	434
604326b4 DB	435	static inline void sk_psock_restore_proto(struct sock *sk,
	436	struct sk_psock *psock)
	437	{
8a59f9d1	438	if (psock->psock_update_sk_prot)
51e0158a	439	psock->psock_update_sk_prot(sk, psock, true);
604326b4 DB	440	}
604326b4 DB	441
604326b4 DB	442	static inline struct sk_psock sk_psock_get(struct sock sk)
	443	{
	444	struct sk_psock *psock;
	445
	446	rcu_read_lock();
	447	psock = sk_psock(sk);
	448	if (psock && !refcount_inc_not_zero(&psock->refcnt))
	449	psock = NULL;
	450	rcu_read_unlock();
	451	return psock;
	452	}
	453
604326b4 DB	454	void sk_psock_drop(struct sock sk, struct sk_psock psock);
	455
	456	static inline void sk_psock_put(struct sock sk, struct sk_psock psock)
	457	{
	458	if (refcount_dec_and_test(&psock->refcnt))
	459	sk_psock_drop(sk, psock);
	460	}
	461
552de910 JF	462	static inline void sk_psock_data_ready(struct sock sk, struct sk_psock psock)
552de910 JF	463	{
6648e613	464	read_lock_bh(&sk->sk_callback_lock);
5a685cd9 CW	465	if (psock->saved_data_ready)
5a685cd9 CW	466	psock->saved_data_ready(sk);
552de910 JF	467	else
552de910 JF	468	sk->sk_data_ready(sk);
6648e613	469	read_unlock_bh(&sk->sk_callback_lock);
552de910 JF	470	}
552de910 JF	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 CW	509
	510	#if IS_ENABLED(CONFIG_NET_SOCK_MSG)
	511
7303524e LJ	512	#define BPF_F_STRPARSER (1UL << 1)
	513
	514	/* We only have two bits so far. */
	515	#define BPF_F_PTR_MASK ~(BPF_F_INGRESS \| BPF_F_STRPARSER)
	516
	517	static inline bool skb_bpf_strparser(const struct sk_buff *skb)
	518	{
	519	unsigned long sk_redir = skb->_sk_redir;
	520
	521	return sk_redir & BPF_F_STRPARSER;
	522	}
	523
	524	static inline void skb_bpf_set_strparser(struct sk_buff *skb)
	525	{
	526	skb->_sk_redir \|= BPF_F_STRPARSER;
	527	}
e3526bb9 CW	528
	529	static inline bool skb_bpf_ingress(const struct sk_buff *skb)
	530	{
	531	unsigned long sk_redir = skb->_sk_redir;
	532
	533	return sk_redir & BPF_F_INGRESS;
	534	}
	535
	536	static inline void skb_bpf_set_ingress(struct sk_buff *skb)
	537	{
	538	skb->_sk_redir \|= BPF_F_INGRESS;
	539	}
	540
	541	static inline void skb_bpf_set_redir(struct sk_buff skb, struct sock sk_redir,
	542	bool ingress)
	543	{
	544	skb->_sk_redir = (unsigned long)sk_redir;
	545	if (ingress)
	546	skb->_sk_redir \|= BPF_F_INGRESS;
	547	}
	548
	549	static inline struct sock skb_bpf_redirect_fetch(const struct sk_buff skb)
	550	{
	551	unsigned long sk_redir = skb->_sk_redir;
	552
	553	return (struct sock *)(sk_redir & BPF_F_PTR_MASK);
	554	}
	555
	556	static inline void skb_bpf_redirect_clear(struct sk_buff *skb)
	557	{
	558	skb->_sk_redir = 0;
	559	}
	560	#endif /* CONFIG_NET_SOCK_MSG */
604326b4	561	#endif /* _LINUX_SKMSG_H */