aboutsummaryrefslogtreecommitdiffstats
path: root/net/xdp/xsk_queue.h
blob: 1b9a350f2e66323480ce39969de5ffbbac5f47ad (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
/* SPDX-License-Identifier: GPL-2.0 */
/* XDP user-space ring structure
 * Copyright(c) 2018 Intel Corporation.
 */

#ifndef _LINUX_XSK_QUEUE_H
#define _LINUX_XSK_QUEUE_H

#include <linux/types.h>
#include <linux/if_xdp.h>
#include <net/xdp_sock.h>

#define RX_BATCH_SIZE 16

struct xdp_ring {
	u32 producer ____cacheline_aligned_in_smp;
	u32 consumer ____cacheline_aligned_in_smp;
	u32 flags;
};

/* Used for the RX and TX queues for packets */
struct xdp_rxtx_ring {
	struct xdp_ring ptrs;
	struct xdp_desc desc[0] ____cacheline_aligned_in_smp;
};

/* Used for the fill and completion queues for buffers */
struct xdp_umem_ring {
	struct xdp_ring ptrs;
	u64 desc[0] ____cacheline_aligned_in_smp;
};

struct xsk_queue {
	u64 chunk_mask;
	u64 size;
	u32 ring_mask;
	u32 nentries;
	u32 cached_prod;
	u32 cons_head;
	u32 cons_tail;
	struct xdp_ring *ring;
	u64 invalid_descs;
};

/* The structure of the shared state of the rings are the same as the
 * ring buffer in kernel/events/ring_buffer.c. For the Rx and completion
 * ring, the kernel is the producer and user space is the consumer. For
 * the Tx and fill rings, the kernel is the consumer and user space is
 * the producer.
 *
 * producer                         consumer
 *
 * if (LOAD ->consumer) {           LOAD ->producer
 *                    (A)           smp_rmb()       (C)
 *    STORE $data                   LOAD $data
 *    smp_wmb()       (B)           smp_mb()        (D)
 *    STORE ->producer              STORE ->consumer
 * }
 *
 * (A) pairs with (D), and (B) pairs with (C).
 *
 * Starting with (B), it protects the data from being written after
 * the producer pointer. If this barrier was missing, the consumer
 * could observe the producer pointer being set and thus load the data
 * before the producer has written the new data. The consumer would in
 * this case load the old data.
 *
 * (C) protects the consumer from speculatively loading the data before
 * the producer pointer actually has been read. If we do not have this
 * barrier, some architectures could load old data as speculative loads
 * are not discarded as the CPU does not know there is a dependency
 * between ->producer and data.
 *
 * (A) is a control dependency that separates the load of ->consumer
 * from the stores of $data. In case ->consumer indicates there is no
 * room in the buffer to store $data we do not. So no barrier is needed.
 *
 * (D) protects the load of the data to be observed to happen after the
 * store of the consumer pointer. If we did not have this memory
 * barrier, the producer could observe the consumer pointer being set
 * and overwrite the data with a new value before the consumer got the
 * chance to read the old value. The consumer would thus miss reading
 * the old entry and very likely read the new entry twice, once right
 * now and again after circling through the ring.
 */

/* Common functions operating for both RXTX and umem queues */

static inline u64 xskq_nb_invalid_descs(struct xsk_queue *q)
{
	return q ? q->invalid_descs : 0;
}

static inline u32 xskq_nb_avail(struct xsk_queue *q, u32 dcnt)
{
	u32 entries = q->cached_prod - q->cons_tail;

	if (entries == 0) {
		/* Refresh the local pointer */
		q->cached_prod = READ_ONCE(q->ring->producer);
		entries = q->cached_prod - q->cons_tail;
	}

	return (entries > dcnt) ? dcnt : entries;
}

static inline u32 xskq_nb_free(struct xsk_queue *q, u32 dcnt)
{
	u32 free_entries = q->nentries - (q->cached_prod - q->cons_tail);

	if (free_entries >= dcnt)
		return free_entries;

	/* Refresh the local tail pointer */
	q->cons_tail = READ_ONCE(q->ring->consumer);
	return q->nentries - (q->cached_prod - q->cons_tail);
}

static inline bool xskq_has_addrs(struct xsk_queue *q, u32 cnt)
{
	u32 entries = q->cached_prod - q->cons_tail;

	if (entries >= cnt)
		return true;

	/* Refresh the local pointer. */
	q->cached_prod = READ_ONCE(q->ring->producer);
	entries = q->cached_prod - q->cons_tail;

	return entries >= cnt;
}

/* UMEM queue */

static inline bool xskq_crosses_non_contig_pg(struct xdp_umem *umem, u64 addr,
					      u64 length)
{
	bool cross_pg = (addr & (PAGE_SIZE - 1)) + length > PAGE_SIZE;
	bool next_pg_contig =
		(unsigned long)umem->pages[(addr >> PAGE_SHIFT)].addr &
			XSK_NEXT_PG_CONTIG_MASK;

	return cross_pg && !next_pg_contig;
}

static inline bool xskq_is_valid_addr(struct xsk_queue *q, u64 addr)
{
	if (addr >= q->size) {
		q->invalid_descs++;
		return false;
	}

	return true;
}

static inline bool xskq_is_valid_addr_unaligned(struct xsk_queue *q, u64 addr,
						u64 length,
						struct xdp_umem *umem)
{
	u64 base_addr = xsk_umem_extract_addr(addr);

	addr = xsk_umem_add_offset_to_addr(addr);
	if (base_addr >= q->size || addr >= q->size ||
	    xskq_crosses_non_contig_pg(umem, addr, length)) {
		q->invalid_descs++;
		return false;
	}

	return true;
}

static inline u64 *xskq_validate_addr(struct xsk_queue *q, u64 *addr,
				      struct xdp_umem *umem)
{
	while (q->cons_tail != q->cons_head) {
		struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
		unsigned int idx = q->cons_tail & q->ring_mask;

		*addr = READ_ONCE(ring->desc[idx]) & q->chunk_mask;

		if (umem->flags & XDP_UMEM_UNALIGNED_CHUNK_FLAG) {
			if (xskq_is_valid_addr_unaligned(q, *addr,
							 umem->chunk_size_nohr,
							 umem))
				return addr;
			goto out;
		}

		if (xskq_is_valid_addr(q, *addr))
			return addr;

out:
		q->cons_tail++;
	}

	return NULL;
}

static inline u64 *xskq_peek_addr(struct xsk_queue *q, u64 *addr,
				  struct xdp_umem *umem)
{
	if (q->cons_tail == q->cons_head) {
		smp_mb(); /* D, matches A */
		WRITE_ONCE(q->ring->consumer, q->cons_tail);
		q->cons_head = q->cons_tail + xskq_nb_avail(q, RX_BATCH_SIZE);

		/* Order consumer and data */
		smp_rmb();
	}

	return xskq_validate_addr(q, addr, umem);
}

static inline void xskq_discard_addr(struct xsk_queue *q)
{
	q->cons_tail++;
}

static inline int xskq_prod_reserve(struct xsk_queue *q)
{
	if (xskq_nb_free(q, 1) == 0)
		return -ENOSPC;

	/* A, matches D */
	q->cached_prod++;
	return 0;
}

static inline int xskq_prod_reserve_addr(struct xsk_queue *q, u64 addr)
{
	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;

	if (xskq_nb_free(q, 1) == 0)
		return -ENOSPC;

	/* A, matches D */
	ring->desc[q->cached_prod++ & q->ring_mask] = addr;
	return 0;
}

static inline void __xskq_prod_submit(struct xsk_queue *q, u32 idx)
{
	/* Order producer and data */
	smp_wmb(); /* B, matches C */

	WRITE_ONCE(q->ring->producer, idx);
}

static inline void xskq_prod_submit(struct xsk_queue *q)
{
	__xskq_prod_submit(q, q->cached_prod);
}

static inline void xskq_prod_submit_addr(struct xsk_queue *q, u64 addr)
{
	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
	u32 idx = q->ring->producer;

	ring->desc[idx++ & q->ring_mask] = addr;

	__xskq_prod_submit(q, idx);
}

static inline void xskq_prod_submit_n(struct xsk_queue *q, u32 nb_entries)
{
	__xskq_prod_submit(q, q->ring->producer + nb_entries);
}

/* Rx/Tx queue */

static inline bool xskq_is_valid_desc(struct xsk_queue *q, struct xdp_desc *d,
				      struct xdp_umem *umem)
{
	if (umem->flags & XDP_UMEM_UNALIGNED_CHUNK_FLAG) {
		if (!xskq_is_valid_addr_unaligned(q, d->addr, d->len, umem))
			return false;

		if (d->len > umem->chunk_size_nohr || d->options) {
			q->invalid_descs++;
			return false;
		}

		return true;
	}

	if (!xskq_is_valid_addr(q, d->addr))
		return false;

	if (((d->addr + d->len) & q->chunk_mask) != (d->addr & q->chunk_mask) ||
	    d->options) {
		q->invalid_descs++;
		return false;
	}

	return true;
}

static inline struct xdp_desc *xskq_validate_desc(struct xsk_queue *q,
						  struct xdp_desc *desc,
						  struct xdp_umem *umem)
{
	while (q->cons_tail != q->cons_head) {
		struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring;
		unsigned int idx = q->cons_tail & q->ring_mask;

		*desc = READ_ONCE(ring->desc[idx]);
		if (xskq_is_valid_desc(q, desc, umem))
			return desc;

		q->cons_tail++;
	}

	return NULL;
}

static inline struct xdp_desc *xskq_peek_desc(struct xsk_queue *q,
					      struct xdp_desc *desc,
					      struct xdp_umem *umem)
{
	if (q->cons_tail == q->cons_head) {
		smp_mb(); /* D, matches A */
		WRITE_ONCE(q->ring->consumer, q->cons_tail);
		q->cons_head = q->cons_tail + xskq_nb_avail(q, RX_BATCH_SIZE);

		/* Order consumer and data */
		smp_rmb(); /* C, matches B */
	}

	return xskq_validate_desc(q, desc, umem);
}

static inline void xskq_discard_desc(struct xsk_queue *q)
{
	q->cons_tail++;
}

static inline int xskq_prod_reserve_desc(struct xsk_queue *q,
					 u64 addr, u32 len)
{
	struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring;
	u32 idx;

	if (xskq_nb_free(q, 1) == 0)
		return -ENOSPC;

	/* A, matches D */
	idx = q->cached_prod++ & q->ring_mask;
	ring->desc[idx].addr = addr;
	ring->desc[idx].len = len;

	return 0;
}

static inline bool xskq_full_desc(struct xsk_queue *q)
{
	/* No barriers needed since data is not accessed */
	return READ_ONCE(q->ring->producer) - READ_ONCE(q->ring->consumer) ==
		q->nentries;
}

static inline bool xskq_prod_is_empty(struct xsk_queue *q)
{
	/* No barriers needed since data is not accessed */
	return READ_ONCE(q->ring->consumer) == READ_ONCE(q->ring->producer);
}

void xskq_set_umem(struct xsk_queue *q, u64 size, u64 chunk_mask);
struct xsk_queue *xskq_create(u32 nentries, bool umem_queue);
void xskq_destroy(struct xsk_queue *q_ops);

/* Executed by the core when the entire UMEM gets freed */
void xsk_reuseq_destroy(struct xdp_umem *umem);

#endif /* _LINUX_XSK_QUEUE_H */