• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1 // SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause)
2 /*
3  * Ring buffer operations.
4  *
5  * Copyright (C) 2020 Facebook, Inc.
6  */
7 #ifndef _GNU_SOURCE
8 #define _GNU_SOURCE
9 #endif
10 #include <stdlib.h>
11 #include <stdio.h>
12 #include <errno.h>
13 #include <unistd.h>
14 #include <linux/err.h>
15 #include <linux/bpf.h>
16 #include <asm/barrier.h>
17 #include <sys/mman.h>
18 #include <sys/epoll.h>
19 
20 #include "libbpf.h"
21 #include "libbpf_internal.h"
22 #include "bpf.h"
23 
24 struct ring {
25 	ring_buffer_sample_fn sample_cb;
26 	void *ctx;
27 	void *data;
28 	unsigned long *consumer_pos;
29 	unsigned long *producer_pos;
30 	unsigned long mask;
31 	int map_fd;
32 };
33 
34 struct ring_buffer {
35 	struct epoll_event *events;
36 	struct ring *rings;
37 	size_t page_size;
38 	int epoll_fd;
39 	int ring_cnt;
40 };
41 
ringbuf_unmap_ring(struct ring_buffer * rb,struct ring * r)42 static void ringbuf_unmap_ring(struct ring_buffer *rb, struct ring *r)
43 {
44 	if (r->consumer_pos) {
45 		munmap(r->consumer_pos, rb->page_size);
46 		r->consumer_pos = NULL;
47 	}
48 	if (r->producer_pos) {
49 		munmap(r->producer_pos, rb->page_size + 2 * (r->mask + 1));
50 		r->producer_pos = NULL;
51 	}
52 }
53 
54 /* Add extra RINGBUF maps to this ring buffer manager */
ring_buffer__add(struct ring_buffer * rb,int map_fd,ring_buffer_sample_fn sample_cb,void * ctx)55 int ring_buffer__add(struct ring_buffer *rb, int map_fd,
56 		     ring_buffer_sample_fn sample_cb, void *ctx)
57 {
58 	struct bpf_map_info info;
59 	__u32 len = sizeof(info);
60 	struct epoll_event *e;
61 	struct ring *r;
62 	__u64 mmap_sz;
63 	void *tmp;
64 	int err;
65 
66 	memset(&info, 0, sizeof(info));
67 
68 	err = bpf_obj_get_info_by_fd(map_fd, &info, &len);
69 	if (err) {
70 		err = -errno;
71 		pr_warn("ringbuf: failed to get map info for fd=%d: %d\n",
72 			map_fd, err);
73 		return libbpf_err(err);
74 	}
75 
76 	if (info.type != BPF_MAP_TYPE_RINGBUF) {
77 		pr_warn("ringbuf: map fd=%d is not BPF_MAP_TYPE_RINGBUF\n",
78 			map_fd);
79 		return libbpf_err(-EINVAL);
80 	}
81 
82 	tmp = libbpf_reallocarray(rb->rings, rb->ring_cnt + 1, sizeof(*rb->rings));
83 	if (!tmp)
84 		return libbpf_err(-ENOMEM);
85 	rb->rings = tmp;
86 
87 	tmp = libbpf_reallocarray(rb->events, rb->ring_cnt + 1, sizeof(*rb->events));
88 	if (!tmp)
89 		return libbpf_err(-ENOMEM);
90 	rb->events = tmp;
91 
92 	r = &rb->rings[rb->ring_cnt];
93 	memset(r, 0, sizeof(*r));
94 
95 	r->map_fd = map_fd;
96 	r->sample_cb = sample_cb;
97 	r->ctx = ctx;
98 	r->mask = info.max_entries - 1;
99 
100 	/* Map writable consumer page */
101 	tmp = mmap(NULL, rb->page_size, PROT_READ | PROT_WRITE, MAP_SHARED, map_fd, 0);
102 	if (tmp == MAP_FAILED) {
103 		err = -errno;
104 		pr_warn("ringbuf: failed to mmap consumer page for map fd=%d: %d\n",
105 			map_fd, err);
106 		return libbpf_err(err);
107 	}
108 	r->consumer_pos = tmp;
109 
110 	/* Map read-only producer page and data pages. We map twice as big
111 	 * data size to allow simple reading of samples that wrap around the
112 	 * end of a ring buffer. See kernel implementation for details.
113 	 * */
114 	mmap_sz = rb->page_size + 2 * (__u64)info.max_entries;
115 	if (mmap_sz != (__u64)(size_t)mmap_sz) {
116 		pr_warn("ringbuf: ring buffer size (%u) is too big\n", info.max_entries);
117 		return libbpf_err(-E2BIG);
118 	}
119 	tmp = mmap(NULL, (size_t)mmap_sz, PROT_READ, MAP_SHARED, map_fd, rb->page_size);
120 	if (tmp == MAP_FAILED) {
121 		err = -errno;
122 		ringbuf_unmap_ring(rb, r);
123 		pr_warn("ringbuf: failed to mmap data pages for map fd=%d: %d\n",
124 			map_fd, err);
125 		return libbpf_err(err);
126 	}
127 	r->producer_pos = tmp;
128 	r->data = tmp + rb->page_size;
129 
130 	e = &rb->events[rb->ring_cnt];
131 	memset(e, 0, sizeof(*e));
132 
133 	e->events = EPOLLIN;
134 	e->data.fd = rb->ring_cnt;
135 	if (epoll_ctl(rb->epoll_fd, EPOLL_CTL_ADD, map_fd, e) < 0) {
136 		err = -errno;
137 		ringbuf_unmap_ring(rb, r);
138 		pr_warn("ringbuf: failed to epoll add map fd=%d: %d\n",
139 			map_fd, err);
140 		return libbpf_err(err);
141 	}
142 
143 	rb->ring_cnt++;
144 	return 0;
145 }
146 
ring_buffer__free(struct ring_buffer * rb)147 void ring_buffer__free(struct ring_buffer *rb)
148 {
149 	int i;
150 
151 	if (!rb)
152 		return;
153 
154 	for (i = 0; i < rb->ring_cnt; ++i)
155 		ringbuf_unmap_ring(rb, &rb->rings[i]);
156 	if (rb->epoll_fd >= 0)
157 		close(rb->epoll_fd);
158 
159 	free(rb->events);
160 	free(rb->rings);
161 	free(rb);
162 }
163 
164 struct ring_buffer *
ring_buffer__new(int map_fd,ring_buffer_sample_fn sample_cb,void * ctx,const struct ring_buffer_opts * opts)165 ring_buffer__new(int map_fd, ring_buffer_sample_fn sample_cb, void *ctx,
166 		 const struct ring_buffer_opts *opts)
167 {
168 	struct ring_buffer *rb;
169 	int err;
170 
171 	if (!OPTS_VALID(opts, ring_buffer_opts))
172 		return errno = EINVAL, NULL;
173 
174 	rb = calloc(1, sizeof(*rb));
175 	if (!rb)
176 		return errno = ENOMEM, NULL;
177 
178 	rb->page_size = getpagesize();
179 
180 	rb->epoll_fd = epoll_create1(EPOLL_CLOEXEC);
181 	if (rb->epoll_fd < 0) {
182 		err = -errno;
183 		pr_warn("ringbuf: failed to create epoll instance: %d\n", err);
184 		goto err_out;
185 	}
186 
187 	err = ring_buffer__add(rb, map_fd, sample_cb, ctx);
188 	if (err)
189 		goto err_out;
190 
191 	return rb;
192 
193 err_out:
194 	ring_buffer__free(rb);
195 	return errno = -err, NULL;
196 }
197 
roundup_len(__u32 len)198 static inline int roundup_len(__u32 len)
199 {
200 	/* clear out top 2 bits (discard and busy, if set) */
201 	len <<= 2;
202 	len >>= 2;
203 	/* add length prefix */
204 	len += BPF_RINGBUF_HDR_SZ;
205 	/* round up to 8 byte alignment */
206 	return (len + 7) / 8 * 8;
207 }
208 
ringbuf_process_ring(struct ring * r)209 static int64_t ringbuf_process_ring(struct ring* r)
210 {
211 	int *len_ptr, len, err;
212 	/* 64-bit to avoid overflow in case of extreme application behavior */
213 	int64_t cnt = 0;
214 	unsigned long cons_pos, prod_pos;
215 	bool got_new_data;
216 	void *sample;
217 
218 	cons_pos = smp_load_acquire(r->consumer_pos);
219 	do {
220 		got_new_data = false;
221 		prod_pos = smp_load_acquire(r->producer_pos);
222 		while (cons_pos < prod_pos) {
223 			len_ptr = r->data + (cons_pos & r->mask);
224 			len = smp_load_acquire(len_ptr);
225 
226 			/* sample not committed yet, bail out for now */
227 			if (len & BPF_RINGBUF_BUSY_BIT)
228 				goto done;
229 
230 			got_new_data = true;
231 			cons_pos += roundup_len(len);
232 
233 			if ((len & BPF_RINGBUF_DISCARD_BIT) == 0) {
234 				sample = (void *)len_ptr + BPF_RINGBUF_HDR_SZ;
235 				err = r->sample_cb(r->ctx, sample, len);
236 				if (err < 0) {
237 					/* update consumer pos and bail out */
238 					smp_store_release(r->consumer_pos,
239 							  cons_pos);
240 					return err;
241 				}
242 				cnt++;
243 			}
244 
245 			smp_store_release(r->consumer_pos, cons_pos);
246 		}
247 	} while (got_new_data);
248 done:
249 	return cnt;
250 }
251 
252 /* Consume available ring buffer(s) data without event polling.
253  * Returns number of records consumed across all registered ring buffers (or
254  * INT_MAX, whichever is less), or negative number if any of the callbacks
255  * return error.
256  */
ring_buffer__consume(struct ring_buffer * rb)257 int ring_buffer__consume(struct ring_buffer *rb)
258 {
259 	int64_t err, res = 0;
260 	int i;
261 
262 	for (i = 0; i < rb->ring_cnt; i++) {
263 		struct ring *ring = &rb->rings[i];
264 
265 		err = ringbuf_process_ring(ring);
266 		if (err < 0)
267 			return libbpf_err(err);
268 		res += err;
269 	}
270 	if (res > INT_MAX)
271 		return INT_MAX;
272 	return res;
273 }
274 
275 /* Poll for available data and consume records, if any are available.
276  * Returns number of records consumed (or INT_MAX, whichever is less), or
277  * negative number, if any of the registered callbacks returned error.
278  */
ring_buffer__poll(struct ring_buffer * rb,int timeout_ms)279 int ring_buffer__poll(struct ring_buffer *rb, int timeout_ms)
280 {
281 	int i, cnt;
282 	int64_t err, res = 0;
283 
284 	cnt = epoll_wait(rb->epoll_fd, rb->events, rb->ring_cnt, timeout_ms);
285 	if (cnt < 0)
286 		return libbpf_err(-errno);
287 
288 	for (i = 0; i < cnt; i++) {
289 		__u32 ring_id = rb->events[i].data.fd;
290 		struct ring *ring = &rb->rings[ring_id];
291 
292 		err = ringbuf_process_ring(ring);
293 		if (err < 0)
294 			return libbpf_err(err);
295 		res += err;
296 	}
297 	if (res > INT_MAX)
298 		return INT_MAX;
299 	return res;
300 }
301 
302 /* Get an fd that can be used to sleep until data is available in the ring(s) */
ring_buffer__epoll_fd(const struct ring_buffer * rb)303 int ring_buffer__epoll_fd(const struct ring_buffer *rb)
304 {
305 	return rb->epoll_fd;
306 }
307