1 /* SPDX-License-Identifier: GPL-2.0 */ 2 /* XDP user-space ring structure 3 * Copyright(c) 2018 Intel Corporation. 4 */ 5 6 #ifndef _LINUX_XSK_QUEUE_H 7 #define _LINUX_XSK_QUEUE_H 8 9 #include <linux/types.h> 10 #include <linux/if_xdp.h> 11 #include <net/xdp_sock.h> 12 #include <net/xsk_buff_pool.h> 13 14 #include "xsk.h" 15 16 struct xdp_ring { 17 u32 producer ____cacheline_aligned_in_smp; 18 /* Hinder the adjacent cache prefetcher to prefetch the consumer 19 * pointer if the producer pointer is touched and vice versa. 20 */ 21 u32 pad1 ____cacheline_aligned_in_smp; 22 u32 consumer ____cacheline_aligned_in_smp; 23 u32 pad2 ____cacheline_aligned_in_smp; 24 u32 flags; 25 u32 pad3 ____cacheline_aligned_in_smp; 26 }; 27 28 /* Used for the RX and TX queues for packets */ 29 struct xdp_rxtx_ring { 30 struct xdp_ring ptrs; 31 struct xdp_desc desc[] ____cacheline_aligned_in_smp; 32 }; 33 34 /* Used for the fill and completion queues for buffers */ 35 struct xdp_umem_ring { 36 struct xdp_ring ptrs; 37 u64 desc[] ____cacheline_aligned_in_smp; 38 }; 39 40 struct xsk_queue { 41 u32 ring_mask; 42 u32 nentries; 43 u32 cached_prod; 44 u32 cached_cons; 45 struct xdp_ring *ring; 46 u64 invalid_descs; 47 u64 queue_empty_descs; 48 }; 49 50 /* The structure of the shared state of the rings are a simple 51 * circular buffer, as outlined in 52 * Documentation/core-api/circular-buffers.rst. For the Rx and 53 * completion ring, the kernel is the producer and user space is the 54 * consumer. For the Tx and fill rings, the kernel is the consumer and 55 * user space is the producer. 56 * 57 * producer consumer 58 * 59 * if (LOAD ->consumer) { (A) LOAD.acq ->producer (C) 60 * STORE $data LOAD $data 61 * STORE.rel ->producer (B) STORE.rel ->consumer (D) 62 * } 63 * 64 * (A) pairs with (D), and (B) pairs with (C). 65 * 66 * Starting with (B), it protects the data from being written after 67 * the producer pointer. If this barrier was missing, the consumer 68 * could observe the producer pointer being set and thus load the data 69 * before the producer has written the new data. The consumer would in 70 * this case load the old data. 71 * 72 * (C) protects the consumer from speculatively loading the data before 73 * the producer pointer actually has been read. If we do not have this 74 * barrier, some architectures could load old data as speculative loads 75 * are not discarded as the CPU does not know there is a dependency 76 * between ->producer and data. 77 * 78 * (A) is a control dependency that separates the load of ->consumer 79 * from the stores of $data. In case ->consumer indicates there is no 80 * room in the buffer to store $data we do not. The dependency will 81 * order both of the stores after the loads. So no barrier is needed. 82 * 83 * (D) protects the load of the data to be observed to happen after the 84 * store of the consumer pointer. If we did not have this memory 85 * barrier, the producer could observe the consumer pointer being set 86 * and overwrite the data with a new value before the consumer got the 87 * chance to read the old value. The consumer would thus miss reading 88 * the old entry and very likely read the new entry twice, once right 89 * now and again after circling through the ring. 90 */ 91 92 /* The operations on the rings are the following: 93 * 94 * producer consumer 95 * 96 * RESERVE entries PEEK in the ring for entries 97 * WRITE data into the ring READ data from the ring 98 * SUBMIT entries RELEASE entries 99 * 100 * The producer reserves one or more entries in the ring. It can then 101 * fill in these entries and finally submit them so that they can be 102 * seen and read by the consumer. 103 * 104 * The consumer peeks into the ring to see if the producer has written 105 * any new entries. If so, the consumer can then read these entries 106 * and when it is done reading them release them back to the producer 107 * so that the producer can use these slots to fill in new entries. 108 * 109 * The function names below reflect these operations. 110 */ 111 112 /* Functions that read and validate content from consumer rings. */ 113 114 static inline void __xskq_cons_read_addr_unchecked(struct xsk_queue *q, u32 cached_cons, u64 *addr) 115 { 116 struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring; 117 u32 idx = cached_cons & q->ring_mask; 118 119 *addr = ring->desc[idx]; 120 } 121 122 static inline bool xskq_cons_read_addr_unchecked(struct xsk_queue *q, u64 *addr) 123 { 124 if (q->cached_cons != q->cached_prod) { 125 __xskq_cons_read_addr_unchecked(q, q->cached_cons, addr); 126 return true; 127 } 128 129 return false; 130 } 131 132 static inline bool xp_aligned_validate_desc(struct xsk_buff_pool *pool, 133 struct xdp_desc *desc) 134 { 135 u64 chunk, chunk_end; 136 137 chunk = xp_aligned_extract_addr(pool, desc->addr); 138 if (likely(desc->len)) { 139 chunk_end = xp_aligned_extract_addr(pool, desc->addr + desc->len - 1); 140 if (chunk != chunk_end) 141 return false; 142 } 143 144 if (chunk >= pool->addrs_cnt) 145 return false; 146 147 if (desc->options) 148 return false; 149 return true; 150 } 151 152 static inline bool xp_unaligned_validate_desc(struct xsk_buff_pool *pool, 153 struct xdp_desc *desc) 154 { 155 u64 addr, base_addr; 156 157 base_addr = xp_unaligned_extract_addr(desc->addr); 158 addr = xp_unaligned_add_offset_to_addr(desc->addr); 159 160 if (desc->len > pool->chunk_size) 161 return false; 162 163 if (base_addr >= pool->addrs_cnt || addr >= pool->addrs_cnt || 164 xp_desc_crosses_non_contig_pg(pool, addr, desc->len)) 165 return false; 166 167 if (desc->options) 168 return false; 169 return true; 170 } 171 172 static inline bool xp_validate_desc(struct xsk_buff_pool *pool, 173 struct xdp_desc *desc) 174 { 175 return pool->unaligned ? xp_unaligned_validate_desc(pool, desc) : 176 xp_aligned_validate_desc(pool, desc); 177 } 178 179 static inline bool xskq_cons_is_valid_desc(struct xsk_queue *q, 180 struct xdp_desc *d, 181 struct xsk_buff_pool *pool) 182 { 183 if (!xp_validate_desc(pool, d)) { 184 q->invalid_descs++; 185 return false; 186 } 187 return true; 188 } 189 190 static inline bool xskq_cons_read_desc(struct xsk_queue *q, 191 struct xdp_desc *desc, 192 struct xsk_buff_pool *pool) 193 { 194 while (q->cached_cons != q->cached_prod) { 195 struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring; 196 u32 idx = q->cached_cons & q->ring_mask; 197 198 *desc = ring->desc[idx]; 199 if (xskq_cons_is_valid_desc(q, desc, pool)) 200 return true; 201 202 q->cached_cons++; 203 } 204 205 return false; 206 } 207 208 static inline void xskq_cons_release_n(struct xsk_queue *q, u32 cnt) 209 { 210 q->cached_cons += cnt; 211 } 212 213 static inline u32 xskq_cons_read_desc_batch(struct xsk_queue *q, struct xsk_buff_pool *pool, 214 u32 max) 215 { 216 u32 cached_cons = q->cached_cons, nb_entries = 0; 217 struct xdp_desc *descs = pool->tx_descs; 218 219 while (cached_cons != q->cached_prod && nb_entries < max) { 220 struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring; 221 u32 idx = cached_cons & q->ring_mask; 222 223 descs[nb_entries] = ring->desc[idx]; 224 if (unlikely(!xskq_cons_is_valid_desc(q, &descs[nb_entries], pool))) { 225 /* Skip the entry */ 226 cached_cons++; 227 continue; 228 } 229 230 nb_entries++; 231 cached_cons++; 232 } 233 234 /* Release valid plus any invalid entries */ 235 xskq_cons_release_n(q, cached_cons - q->cached_cons); 236 return nb_entries; 237 } 238 239 /* Functions for consumers */ 240 241 static inline void __xskq_cons_release(struct xsk_queue *q) 242 { 243 smp_store_release(&q->ring->consumer, q->cached_cons); /* D, matchees A */ 244 } 245 246 static inline void __xskq_cons_peek(struct xsk_queue *q) 247 { 248 /* Refresh the local pointer */ 249 q->cached_prod = smp_load_acquire(&q->ring->producer); /* C, matches B */ 250 } 251 252 static inline void xskq_cons_get_entries(struct xsk_queue *q) 253 { 254 __xskq_cons_release(q); 255 __xskq_cons_peek(q); 256 } 257 258 static inline u32 xskq_cons_nb_entries(struct xsk_queue *q, u32 max) 259 { 260 u32 entries = q->cached_prod - q->cached_cons; 261 262 if (entries >= max) 263 return max; 264 265 __xskq_cons_peek(q); 266 entries = q->cached_prod - q->cached_cons; 267 268 return entries >= max ? max : entries; 269 } 270 271 static inline bool xskq_cons_has_entries(struct xsk_queue *q, u32 cnt) 272 { 273 return xskq_cons_nb_entries(q, cnt) >= cnt; 274 } 275 276 static inline bool xskq_cons_peek_addr_unchecked(struct xsk_queue *q, u64 *addr) 277 { 278 if (q->cached_prod == q->cached_cons) 279 xskq_cons_get_entries(q); 280 return xskq_cons_read_addr_unchecked(q, addr); 281 } 282 283 static inline bool xskq_cons_peek_desc(struct xsk_queue *q, 284 struct xdp_desc *desc, 285 struct xsk_buff_pool *pool) 286 { 287 if (q->cached_prod == q->cached_cons) 288 xskq_cons_get_entries(q); 289 return xskq_cons_read_desc(q, desc, pool); 290 } 291 292 /* To improve performance in the xskq_cons_release functions, only update local state here. 293 * Reflect this to global state when we get new entries from the ring in 294 * xskq_cons_get_entries() and whenever Rx or Tx processing are completed in the NAPI loop. 295 */ 296 static inline void xskq_cons_release(struct xsk_queue *q) 297 { 298 q->cached_cons++; 299 } 300 301 static inline u32 xskq_cons_present_entries(struct xsk_queue *q) 302 { 303 /* No barriers needed since data is not accessed */ 304 return READ_ONCE(q->ring->producer) - READ_ONCE(q->ring->consumer); 305 } 306 307 /* Functions for producers */ 308 309 static inline u32 xskq_prod_nb_free(struct xsk_queue *q, u32 max) 310 { 311 u32 free_entries = q->nentries - (q->cached_prod - q->cached_cons); 312 313 if (free_entries >= max) 314 return max; 315 316 /* Refresh the local tail pointer */ 317 q->cached_cons = READ_ONCE(q->ring->consumer); 318 free_entries = q->nentries - (q->cached_prod - q->cached_cons); 319 320 return free_entries >= max ? max : free_entries; 321 } 322 323 static inline bool xskq_prod_is_full(struct xsk_queue *q) 324 { 325 return xskq_prod_nb_free(q, 1) ? false : true; 326 } 327 328 static inline void xskq_prod_cancel(struct xsk_queue *q) 329 { 330 q->cached_prod--; 331 } 332 333 static inline int xskq_prod_reserve(struct xsk_queue *q) 334 { 335 if (xskq_prod_is_full(q)) 336 return -ENOSPC; 337 338 /* A, matches D */ 339 q->cached_prod++; 340 return 0; 341 } 342 343 static inline int xskq_prod_reserve_addr(struct xsk_queue *q, u64 addr) 344 { 345 struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring; 346 347 if (xskq_prod_is_full(q)) 348 return -ENOSPC; 349 350 /* A, matches D */ 351 ring->desc[q->cached_prod++ & q->ring_mask] = addr; 352 return 0; 353 } 354 355 static inline void xskq_prod_write_addr_batch(struct xsk_queue *q, struct xdp_desc *descs, 356 u32 nb_entries) 357 { 358 struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring; 359 u32 i, cached_prod; 360 361 /* A, matches D */ 362 cached_prod = q->cached_prod; 363 for (i = 0; i < nb_entries; i++) 364 ring->desc[cached_prod++ & q->ring_mask] = descs[i].addr; 365 q->cached_prod = cached_prod; 366 } 367 368 static inline int xskq_prod_reserve_desc(struct xsk_queue *q, 369 u64 addr, u32 len) 370 { 371 struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring; 372 u32 idx; 373 374 if (xskq_prod_is_full(q)) 375 return -ENOBUFS; 376 377 /* A, matches D */ 378 idx = q->cached_prod++ & q->ring_mask; 379 ring->desc[idx].addr = addr; 380 ring->desc[idx].len = len; 381 382 return 0; 383 } 384 385 static inline void __xskq_prod_submit(struct xsk_queue *q, u32 idx) 386 { 387 smp_store_release(&q->ring->producer, idx); /* B, matches C */ 388 } 389 390 static inline void xskq_prod_submit(struct xsk_queue *q) 391 { 392 __xskq_prod_submit(q, q->cached_prod); 393 } 394 395 static inline void xskq_prod_submit_addr(struct xsk_queue *q, u64 addr) 396 { 397 struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring; 398 u32 idx = q->ring->producer; 399 400 ring->desc[idx++ & q->ring_mask] = addr; 401 402 __xskq_prod_submit(q, idx); 403 } 404 405 static inline void xskq_prod_submit_n(struct xsk_queue *q, u32 nb_entries) 406 { 407 __xskq_prod_submit(q, q->ring->producer + nb_entries); 408 } 409 410 static inline bool xskq_prod_is_empty(struct xsk_queue *q) 411 { 412 /* No barriers needed since data is not accessed */ 413 return READ_ONCE(q->ring->consumer) == READ_ONCE(q->ring->producer); 414 } 415 416 /* For both producers and consumers */ 417 418 static inline u64 xskq_nb_invalid_descs(struct xsk_queue *q) 419 { 420 return q ? q->invalid_descs : 0; 421 } 422 423 static inline u64 xskq_nb_queue_empty_descs(struct xsk_queue *q) 424 { 425 return q ? q->queue_empty_descs : 0; 426 } 427 428 struct xsk_queue *xskq_create(u32 nentries, bool umem_queue); 429 void xskq_destroy(struct xsk_queue *q_ops); 430 431 #endif /* _LINUX_XSK_QUEUE_H */ 432