]> git.proxmox.com Git - ceph.git/blob - ceph/src/dpdk/lib/librte_ring/rte_ring.c
bump version to 12.2.12-pve1
[ceph.git] / ceph / src / dpdk / lib / librte_ring / rte_ring.c
1 /*-
2 * BSD LICENSE
3 *
4 * Copyright(c) 2010-2015 Intel Corporation. All rights reserved.
5 * All rights reserved.
6 *
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
9 * are met:
10 *
11 * * Redistributions of source code must retain the above copyright
12 * notice, this list of conditions and the following disclaimer.
13 * * Redistributions in binary form must reproduce the above copyright
14 * notice, this list of conditions and the following disclaimer in
15 * the documentation and/or other materials provided with the
16 * distribution.
17 * * Neither the name of Intel Corporation nor the names of its
18 * contributors may be used to endorse or promote products derived
19 * from this software without specific prior written permission.
20 *
21 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
22 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
23 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
24 * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
25 * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
26 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
27 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
28 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
29 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
30 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
31 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
32 */
33
34 /*
35 * Derived from FreeBSD's bufring.c
36 *
37 **************************************************************************
38 *
39 * Copyright (c) 2007,2008 Kip Macy kmacy@freebsd.org
40 * All rights reserved.
41 *
42 * Redistribution and use in source and binary forms, with or without
43 * modification, are permitted provided that the following conditions are met:
44 *
45 * 1. Redistributions of source code must retain the above copyright notice,
46 * this list of conditions and the following disclaimer.
47 *
48 * 2. The name of Kip Macy nor the names of other
49 * contributors may be used to endorse or promote products derived from
50 * this software without specific prior written permission.
51 *
52 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
53 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
54 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
55 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
56 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
57 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
58 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
59 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
60 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
61 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
62 * POSSIBILITY OF SUCH DAMAGE.
63 *
64 ***************************************************************************/
65
66 #include <stdio.h>
67 #include <stdarg.h>
68 #include <string.h>
69 #include <stdint.h>
70 #include <inttypes.h>
71 #include <errno.h>
72 #include <sys/queue.h>
73
74 #include <rte_common.h>
75 #include <rte_log.h>
76 #include <rte_memory.h>
77 #include <rte_memzone.h>
78 #include <rte_malloc.h>
79 #include <rte_launch.h>
80 #include <rte_eal.h>
81 #include <rte_eal_memconfig.h>
82 #include <rte_atomic.h>
83 #include <rte_per_lcore.h>
84 #include <rte_lcore.h>
85 #include <rte_branch_prediction.h>
86 #include <rte_errno.h>
87 #include <rte_string_fns.h>
88 #include <rte_spinlock.h>
89
90 #include "rte_ring.h"
91
92 TAILQ_HEAD(rte_ring_list, rte_tailq_entry);
93
94 static struct rte_tailq_elem rte_ring_tailq = {
95 .name = RTE_TAILQ_RING_NAME,
96 };
97 EAL_REGISTER_TAILQ(rte_ring_tailq)
98
99 /* true if x is a power of 2 */
100 #define POWEROF2(x) ((((x)-1) & (x)) == 0)
101
102 /* return the size of memory occupied by a ring */
103 ssize_t
104 rte_ring_get_memsize(unsigned count)
105 {
106 ssize_t sz;
107
108 /* count must be a power of 2 */
109 if ((!POWEROF2(count)) || (count > RTE_RING_SZ_MASK )) {
110 RTE_LOG(ERR, RING,
111 "Requested size is invalid, must be power of 2, and "
112 "do not exceed the size limit %u\n", RTE_RING_SZ_MASK);
113 return -EINVAL;
114 }
115
116 sz = sizeof(struct rte_ring) + count * sizeof(void *);
117 sz = RTE_ALIGN(sz, RTE_CACHE_LINE_SIZE);
118 return sz;
119 }
120
121 int
122 rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
123 unsigned flags)
124 {
125 int ret;
126
127 /* compilation-time checks */
128 RTE_BUILD_BUG_ON((sizeof(struct rte_ring) &
129 RTE_CACHE_LINE_MASK) != 0);
130 #ifdef RTE_RING_SPLIT_PROD_CONS
131 RTE_BUILD_BUG_ON((offsetof(struct rte_ring, cons) &
132 RTE_CACHE_LINE_MASK) != 0);
133 #endif
134 RTE_BUILD_BUG_ON((offsetof(struct rte_ring, prod) &
135 RTE_CACHE_LINE_MASK) != 0);
136 #ifdef RTE_LIBRTE_RING_DEBUG
137 RTE_BUILD_BUG_ON((sizeof(struct rte_ring_debug_stats) &
138 RTE_CACHE_LINE_MASK) != 0);
139 RTE_BUILD_BUG_ON((offsetof(struct rte_ring, stats) &
140 RTE_CACHE_LINE_MASK) != 0);
141 #endif
142
143 /* init the ring structure */
144 memset(r, 0, sizeof(*r));
145 ret = snprintf(r->name, sizeof(r->name), "%s", name);
146 if (ret < 0 || ret >= (int)sizeof(r->name))
147 return -ENAMETOOLONG;
148 r->flags = flags;
149 r->prod.watermark = count;
150 r->prod.sp_enqueue = !!(flags & RING_F_SP_ENQ);
151 r->cons.sc_dequeue = !!(flags & RING_F_SC_DEQ);
152 r->prod.size = r->cons.size = count;
153 r->prod.mask = r->cons.mask = count-1;
154 r->prod.head = r->cons.head = 0;
155 r->prod.tail = r->cons.tail = 0;
156
157 return 0;
158 }
159
160 /* create the ring */
161 struct rte_ring *
162 rte_ring_create(const char *name, unsigned count, int socket_id,
163 unsigned flags)
164 {
165 char mz_name[RTE_MEMZONE_NAMESIZE];
166 struct rte_ring *r;
167 struct rte_tailq_entry *te;
168 const struct rte_memzone *mz;
169 ssize_t ring_size;
170 int mz_flags = 0;
171 struct rte_ring_list* ring_list = NULL;
172 int ret;
173
174 ring_list = RTE_TAILQ_CAST(rte_ring_tailq.head, rte_ring_list);
175
176 ring_size = rte_ring_get_memsize(count);
177 if (ring_size < 0) {
178 rte_errno = ring_size;
179 return NULL;
180 }
181
182 ret = snprintf(mz_name, sizeof(mz_name), "%s%s",
183 RTE_RING_MZ_PREFIX, name);
184 if (ret < 0 || ret >= (int)sizeof(mz_name)) {
185 rte_errno = ENAMETOOLONG;
186 return NULL;
187 }
188
189 te = rte_zmalloc("RING_TAILQ_ENTRY", sizeof(*te), 0);
190 if (te == NULL) {
191 RTE_LOG(ERR, RING, "Cannot reserve memory for tailq\n");
192 rte_errno = ENOMEM;
193 return NULL;
194 }
195
196 rte_rwlock_write_lock(RTE_EAL_TAILQ_RWLOCK);
197
198 /* reserve a memory zone for this ring. If we can't get rte_config or
199 * we are secondary process, the memzone_reserve function will set
200 * rte_errno for us appropriately - hence no check in this this function */
201 mz = rte_memzone_reserve(mz_name, ring_size, socket_id, mz_flags);
202 if (mz != NULL) {
203 r = mz->addr;
204 /* no need to check return value here, we already checked the
205 * arguments above */
206 rte_ring_init(r, name, count, flags);
207
208 te->data = (void *) r;
209 r->memzone = mz;
210
211 TAILQ_INSERT_TAIL(ring_list, te, next);
212 } else {
213 r = NULL;
214 RTE_LOG(ERR, RING, "Cannot reserve memory\n");
215 rte_free(te);
216 }
217 rte_rwlock_write_unlock(RTE_EAL_TAILQ_RWLOCK);
218
219 return r;
220 }
221
222 /* free the ring */
223 void
224 rte_ring_free(struct rte_ring *r)
225 {
226 struct rte_ring_list *ring_list = NULL;
227 struct rte_tailq_entry *te;
228
229 if (r == NULL)
230 return;
231
232 /*
233 * Ring was not created with rte_ring_create,
234 * therefore, there is no memzone to free.
235 */
236 if (r->memzone == NULL) {
237 RTE_LOG(ERR, RING, "Cannot free ring (not created with rte_ring_create()");
238 return;
239 }
240
241 if (rte_memzone_free(r->memzone) != 0) {
242 RTE_LOG(ERR, RING, "Cannot free memory\n");
243 return;
244 }
245
246 ring_list = RTE_TAILQ_CAST(rte_ring_tailq.head, rte_ring_list);
247 rte_rwlock_write_lock(RTE_EAL_TAILQ_RWLOCK);
248
249 /* find out tailq entry */
250 TAILQ_FOREACH(te, ring_list, next) {
251 if (te->data == (void *) r)
252 break;
253 }
254
255 if (te == NULL) {
256 rte_rwlock_write_unlock(RTE_EAL_TAILQ_RWLOCK);
257 return;
258 }
259
260 TAILQ_REMOVE(ring_list, te, next);
261
262 rte_rwlock_write_unlock(RTE_EAL_TAILQ_RWLOCK);
263
264 rte_free(te);
265 }
266
267 /*
268 * change the high water mark. If *count* is 0, water marking is
269 * disabled
270 */
271 int
272 rte_ring_set_water_mark(struct rte_ring *r, unsigned count)
273 {
274 if (count >= r->prod.size)
275 return -EINVAL;
276
277 /* if count is 0, disable the watermarking */
278 if (count == 0)
279 count = r->prod.size;
280
281 r->prod.watermark = count;
282 return 0;
283 }
284
285 /* dump the status of the ring on the console */
286 void
287 rte_ring_dump(FILE *f, const struct rte_ring *r)
288 {
289 #ifdef RTE_LIBRTE_RING_DEBUG
290 struct rte_ring_debug_stats sum;
291 unsigned lcore_id;
292 #endif
293
294 fprintf(f, "ring <%s>@%p\n", r->name, r);
295 fprintf(f, " flags=%x\n", r->flags);
296 fprintf(f, " size=%"PRIu32"\n", r->prod.size);
297 fprintf(f, " ct=%"PRIu32"\n", r->cons.tail);
298 fprintf(f, " ch=%"PRIu32"\n", r->cons.head);
299 fprintf(f, " pt=%"PRIu32"\n", r->prod.tail);
300 fprintf(f, " ph=%"PRIu32"\n", r->prod.head);
301 fprintf(f, " used=%u\n", rte_ring_count(r));
302 fprintf(f, " avail=%u\n", rte_ring_free_count(r));
303 if (r->prod.watermark == r->prod.size)
304 fprintf(f, " watermark=0\n");
305 else
306 fprintf(f, " watermark=%"PRIu32"\n", r->prod.watermark);
307
308 /* sum and dump statistics */
309 #ifdef RTE_LIBRTE_RING_DEBUG
310 memset(&sum, 0, sizeof(sum));
311 for (lcore_id = 0; lcore_id < RTE_MAX_LCORE; lcore_id++) {
312 sum.enq_success_bulk += r->stats[lcore_id].enq_success_bulk;
313 sum.enq_success_objs += r->stats[lcore_id].enq_success_objs;
314 sum.enq_quota_bulk += r->stats[lcore_id].enq_quota_bulk;
315 sum.enq_quota_objs += r->stats[lcore_id].enq_quota_objs;
316 sum.enq_fail_bulk += r->stats[lcore_id].enq_fail_bulk;
317 sum.enq_fail_objs += r->stats[lcore_id].enq_fail_objs;
318 sum.deq_success_bulk += r->stats[lcore_id].deq_success_bulk;
319 sum.deq_success_objs += r->stats[lcore_id].deq_success_objs;
320 sum.deq_fail_bulk += r->stats[lcore_id].deq_fail_bulk;
321 sum.deq_fail_objs += r->stats[lcore_id].deq_fail_objs;
322 }
323 fprintf(f, " size=%"PRIu32"\n", r->prod.size);
324 fprintf(f, " enq_success_bulk=%"PRIu64"\n", sum.enq_success_bulk);
325 fprintf(f, " enq_success_objs=%"PRIu64"\n", sum.enq_success_objs);
326 fprintf(f, " enq_quota_bulk=%"PRIu64"\n", sum.enq_quota_bulk);
327 fprintf(f, " enq_quota_objs=%"PRIu64"\n", sum.enq_quota_objs);
328 fprintf(f, " enq_fail_bulk=%"PRIu64"\n", sum.enq_fail_bulk);
329 fprintf(f, " enq_fail_objs=%"PRIu64"\n", sum.enq_fail_objs);
330 fprintf(f, " deq_success_bulk=%"PRIu64"\n", sum.deq_success_bulk);
331 fprintf(f, " deq_success_objs=%"PRIu64"\n", sum.deq_success_objs);
332 fprintf(f, " deq_fail_bulk=%"PRIu64"\n", sum.deq_fail_bulk);
333 fprintf(f, " deq_fail_objs=%"PRIu64"\n", sum.deq_fail_objs);
334 #else
335 fprintf(f, " no statistics available\n");
336 #endif
337 }
338
339 /* dump the status of all rings on the console */
340 void
341 rte_ring_list_dump(FILE *f)
342 {
343 const struct rte_tailq_entry *te;
344 struct rte_ring_list *ring_list;
345
346 ring_list = RTE_TAILQ_CAST(rte_ring_tailq.head, rte_ring_list);
347
348 rte_rwlock_read_lock(RTE_EAL_TAILQ_RWLOCK);
349
350 TAILQ_FOREACH(te, ring_list, next) {
351 rte_ring_dump(f, (struct rte_ring *) te->data);
352 }
353
354 rte_rwlock_read_unlock(RTE_EAL_TAILQ_RWLOCK);
355 }
356
357 /* search a ring from its name */
358 struct rte_ring *
359 rte_ring_lookup(const char *name)
360 {
361 struct rte_tailq_entry *te;
362 struct rte_ring *r = NULL;
363 struct rte_ring_list *ring_list;
364
365 ring_list = RTE_TAILQ_CAST(rte_ring_tailq.head, rte_ring_list);
366
367 rte_rwlock_read_lock(RTE_EAL_TAILQ_RWLOCK);
368
369 TAILQ_FOREACH(te, ring_list, next) {
370 r = (struct rte_ring *) te->data;
371 if (strncmp(name, r->name, RTE_RING_NAMESIZE) == 0)
372 break;
373 }
374
375 rte_rwlock_read_unlock(RTE_EAL_TAILQ_RWLOCK);
376
377 if (te == NULL) {
378 rte_errno = ENOENT;
379 return NULL;
380 }
381
382 return r;
383 }