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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416
|
/*
* This file is part of PowerDNS or dnsdist.
* Copyright -- PowerDNS.COM B.V. and its contributors
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of version 2 of the GNU General Public License as
* published by the Free Software Foundation.
*
* In addition, for the avoidance of any doubt, permission is granted to
* link this program with OpenSSL and to (re)distribute the binaries
* produced as the result of such linking.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
*/
#include "bpf-filter.hh"
#ifdef HAVE_EBPF
#include <sys/syscall.h>
#include <linux/bpf.h>
#include "ext/libbpf/libbpf.h"
static __u64 ptr_to_u64(void *ptr)
{
return (__u64) (unsigned long) ptr;
}
int bpf_create_map(enum bpf_map_type map_type, int key_size, int value_size,
int max_entries)
{
union bpf_attr attr = { 0 };
attr.map_type = map_type;
attr.key_size = key_size;
attr.value_size = value_size;
attr.max_entries = max_entries;
return syscall(SYS_bpf, BPF_MAP_CREATE, &attr, sizeof(attr));
}
int bpf_update_elem(int fd, void *key, void *value, unsigned long long flags)
{
union bpf_attr attr = { 0 };
attr.map_fd = fd;
attr.key = ptr_to_u64(key);
attr.value = ptr_to_u64(value);
attr.flags = flags;
return syscall(SYS_bpf, BPF_MAP_UPDATE_ELEM, &attr, sizeof(attr));
}
int bpf_lookup_elem(int fd, void *key, void *value)
{
union bpf_attr attr = { 0 };
attr.map_fd = fd;
attr.key = ptr_to_u64(key);
attr.value = ptr_to_u64(value);
return syscall(SYS_bpf, BPF_MAP_LOOKUP_ELEM, &attr, sizeof(attr));
}
int bpf_delete_elem(int fd, void *key)
{
union bpf_attr attr = { 0 };
attr.map_fd = fd;
attr.key = ptr_to_u64(key);
return syscall(SYS_bpf, BPF_MAP_DELETE_ELEM, &attr, sizeof(attr));
}
int bpf_get_next_key(int fd, void *key, void *next_key)
{
union bpf_attr attr = { 0 };
attr.map_fd = fd;
attr.key = ptr_to_u64(key);
attr.next_key = ptr_to_u64(next_key);
return syscall(SYS_bpf, BPF_MAP_GET_NEXT_KEY, &attr, sizeof(attr));
}
int bpf_prog_load(enum bpf_prog_type prog_type,
const struct bpf_insn *insns, int prog_len,
const char *license, int kern_version)
{
char log_buf[65535];
union bpf_attr attr = { 0 };
attr.prog_type = prog_type;
attr.insns = ptr_to_u64((void *) insns);
attr.insn_cnt = prog_len / sizeof(struct bpf_insn);
attr.license = ptr_to_u64((void *) license);
attr.log_buf = ptr_to_u64(log_buf);
attr.log_size = sizeof(log_buf);
attr.log_level = 1;
/* assign one field outside of struct init to make sure any
* padding is zero initialized
*/
attr.kern_version = kern_version;
long res = syscall(SYS_bpf, BPF_PROG_LOAD, &attr, sizeof(attr));
if (res == -1) {
if (errno == ENOSPC) {
/* not enough space in the log buffer */
attr.log_level = 0;
attr.log_size = 0;
attr.log_buf = ptr_to_u64(nullptr);
res = syscall(SYS_bpf, BPF_PROG_LOAD, &attr, sizeof(attr));
if (res != -1) {
return res;
}
}
throw std::runtime_error("Error loading BPF program: (" + std::string(strerror(errno)) + "):\n" + std::string(log_buf));
}
return res;
}
struct KeyV6
{
uint8_t src[16];
};
struct QNameKey
{
uint8_t qname[255];
};
struct QNameValue
{
uint64_t counter;
uint16_t qtype;
};
BPFFilter::BPFFilter(uint32_t maxV4Addresses, uint32_t maxV6Addresses, uint32_t maxQNames): d_maxV4(maxV4Addresses), d_maxV6(maxV6Addresses), d_maxQNames(maxQNames)
{
d_v4map.fd = bpf_create_map(BPF_MAP_TYPE_HASH, sizeof(uint32_t), sizeof(uint64_t), (int) maxV4Addresses);
if (d_v4map.fd == -1) {
throw std::runtime_error("Error creating a BPF v4 map of size " + std::to_string(maxV4Addresses) + ": " + std::string(strerror(errno)));
}
d_v6map.fd = bpf_create_map(BPF_MAP_TYPE_HASH, sizeof(struct KeyV6), sizeof(uint64_t), (int) maxV6Addresses);
if (d_v6map.fd == -1) {
throw std::runtime_error("Error creating a BPF v6 map of size " + std::to_string(maxV6Addresses) + ": " + std::string(strerror(errno)));
}
d_qnamemap.fd = bpf_create_map(BPF_MAP_TYPE_HASH, sizeof(struct QNameKey), sizeof(struct QNameValue), (int) maxQNames);
if (d_qnamemap.fd == -1) {
throw std::runtime_error("Error creating a BPF qname map of size " + std::to_string(maxQNames) + ": " + std::string(strerror(errno)));
}
d_filtermap.fd = bpf_create_map(BPF_MAP_TYPE_PROG_ARRAY, sizeof(uint32_t), sizeof(uint32_t), 1);
if (d_filtermap.fd == -1) {
throw std::runtime_error("Error creating a BPF program map of size 1: " + std::string(strerror(errno)));
}
struct bpf_insn main_filter[] = {
#include "bpf-filter.main.ebpf"
};
d_mainfilter.fd = bpf_prog_load(BPF_PROG_TYPE_SOCKET_FILTER,
main_filter,
sizeof(main_filter),
"GPL",
0);
if (d_mainfilter.fd == -1) {
throw std::runtime_error("Error loading BPF main filter: " + std::string(strerror(errno)));
}
struct bpf_insn qname_filter[] = {
#include "bpf-filter.qname.ebpf"
};
d_qnamefilter.fd = bpf_prog_load(BPF_PROG_TYPE_SOCKET_FILTER,
qname_filter,
sizeof(qname_filter),
"GPL",
0);
if (d_qnamefilter.fd == -1) {
throw std::runtime_error("Error loading BPF qname filter: " + std::string(strerror(errno)));
}
uint32_t key = 0;
int res = bpf_update_elem(d_filtermap.fd, &key, &d_qnamefilter.fd, BPF_ANY);
if (res != 0) {
throw std::runtime_error("Error updating BPF filters map: " + std::string(strerror(errno)));
}
}
void BPFFilter::addSocket(int sock)
{
int res = setsockopt(sock, SOL_SOCKET, SO_ATTACH_BPF, &d_mainfilter.fd, sizeof(d_mainfilter.fd));
if (res != 0) {
throw std::runtime_error("Error attaching BPF filter to this socket: " + std::string(strerror(errno)));
}
}
void BPFFilter::removeSocket(int sock)
{
int res = setsockopt(sock, SOL_SOCKET, SO_DETACH_BPF, &d_mainfilter.fd, sizeof(d_mainfilter.fd));
if (res != 0) {
throw std::runtime_error("Error detaching BPF filter from this socket: " + std::string(strerror(errno)));
}
}
void BPFFilter::block(const ComboAddress& addr)
{
std::unique_lock<std::mutex> lock(d_mutex);
uint64_t counter = 0;
int res = 0;
if (addr.sin4.sin_family == AF_INET) {
uint32_t key = htonl(addr.sin4.sin_addr.s_addr);
if (d_v4Count >= d_maxV4) {
throw std::runtime_error("Table full when trying to block " + addr.toString());
}
res = bpf_lookup_elem(d_v4map.fd, &key, &counter);
if (res != -1) {
throw std::runtime_error("Trying to block an already blocked address: " + addr.toString());
}
res = bpf_update_elem(d_v4map.fd, &key, &counter, BPF_NOEXIST);
if (res == 0) {
d_v4Count++;
}
}
else if (addr.sin4.sin_family == AF_INET6) {
uint8_t key[16];
static_assert(sizeof(addr.sin6.sin6_addr.s6_addr) == sizeof(key), "POSIX mandates s6_addr to be an array of 16 uint8_t");
for (size_t idx = 0; idx < sizeof(key); idx++) {
key[idx] = addr.sin6.sin6_addr.s6_addr[idx];
}
if (d_v6Count >= d_maxV6) {
throw std::runtime_error("Table full when trying to block " + addr.toString());
}
res = bpf_lookup_elem(d_v6map.fd, &key, &counter);
if (res != -1) {
throw std::runtime_error("Trying to block an already blocked address: " + addr.toString());
}
res = bpf_update_elem(d_v6map.fd, key, &counter, BPF_NOEXIST);
if (res == 0) {
d_v6Count++;
}
}
if (res != 0) {
throw std::runtime_error("Error adding blocked address " + addr.toString() + ": " + std::string(strerror(errno)));
}
}
void BPFFilter::unblock(const ComboAddress& addr)
{
std::unique_lock<std::mutex> lock(d_mutex);
int res = 0;
if (addr.sin4.sin_family == AF_INET) {
uint32_t key = htonl(addr.sin4.sin_addr.s_addr);
res = bpf_delete_elem(d_v4map.fd, &key);
if (res == 0) {
d_v4Count--;
}
}
else if (addr.sin4.sin_family == AF_INET6) {
uint8_t key[16];
static_assert(sizeof(addr.sin6.sin6_addr.s6_addr) == sizeof(key), "POSIX mandates s6_addr to be an array of 16 uint8_t");
for (size_t idx = 0; idx < sizeof(key); idx++) {
key[idx] = addr.sin6.sin6_addr.s6_addr[idx];
}
res = bpf_delete_elem(d_v6map.fd, key);
if (res == 0) {
d_v6Count--;
}
}
if (res != 0) {
throw std::runtime_error("Error removing blocked address " + addr.toString() + ": " + std::string(strerror(errno)));
}
}
void BPFFilter::block(const DNSName& qname, uint16_t qtype)
{
struct QNameKey key;
struct QNameValue value;
memset(&key, 0, sizeof(key));
memset(&value, 0, sizeof(value));
value.counter = 0;
value.qtype = qtype;
std::string keyStr = qname.toDNSStringLC();
if (keyStr.size() > sizeof(key.qname)) {
throw std::runtime_error("Invalid QName to block " + qname.toLogString());
}
memcpy(key.qname, keyStr.c_str(), keyStr.size());
{
std::unique_lock<std::mutex> lock(d_mutex);
if (d_qNamesCount >= d_maxQNames) {
throw std::runtime_error("Table full when trying to block " + qname.toLogString());
}
int res = bpf_lookup_elem(d_qnamemap.fd, &key, &value);
if (res != -1) {
throw std::runtime_error("Trying to block an already blocked qname: " + qname.toLogString());
}
res = bpf_update_elem(d_qnamemap.fd, &key, &value, BPF_NOEXIST);
if (res == 0) {
d_qNamesCount++;
}
if (res != 0) {
throw std::runtime_error("Error adding blocked qname " + qname.toLogString() + ": " + std::string(strerror(errno)));
}
}
}
void BPFFilter::unblock(const DNSName& qname, uint16_t qtype)
{
struct QNameKey key = { { 0 } };
std::string keyStr = qname.toDNSStringLC();
(void) qtype;
if (keyStr.size() > sizeof(key.qname)) {
throw std::runtime_error("Invalid QName to block " + qname.toLogString());
}
memcpy(key.qname, keyStr.c_str(), keyStr.size());
{
std::unique_lock<std::mutex> lock(d_mutex);
int res = bpf_delete_elem(d_qnamemap.fd, &key);
if (res == 0) {
d_qNamesCount--;
}
else {
throw std::runtime_error("Error removing qname address " + qname.toLogString() + ": " + std::string(strerror(errno)));
}
}
}
std::vector<std::pair<ComboAddress, uint64_t> > BPFFilter::getAddrStats()
{
std::vector<std::pair<ComboAddress, uint64_t> > result;
std::unique_lock<std::mutex> lock(d_mutex);
uint32_t v4Key = 0;
uint32_t nextV4Key;
uint64_t value;
int res = bpf_get_next_key(d_v4map.fd, &v4Key, &nextV4Key);
sockaddr_in v4Addr = { 0 };
v4Addr.sin_port = 0;
v4Addr.sin_family = AF_INET;
while (res == 0) {
v4Key = nextV4Key;
if (bpf_lookup_elem(d_v4map.fd, &v4Key, &value) == 0) {
v4Addr.sin_addr.s_addr = ntohl(v4Key);
result.push_back(make_pair(ComboAddress(&v4Addr), value));
}
res = bpf_get_next_key(d_v4map.fd, &v4Key, &nextV4Key);
}
uint8_t v6Key[16];
uint8_t nextV6Key[16];
sockaddr_in6 v6Addr = { 0 };
v6Addr.sin6_family = AF_INET6;
v6Addr.sin6_port = 0;
static_assert(sizeof(v6Addr.sin6_addr.s6_addr) == sizeof(v6Key), "POSIX mandates s6_addr to be an array of 16 uint8_t");
for (size_t idx = 0; idx < sizeof(v6Key); idx++) {
v6Key[idx] = 0;
}
res = bpf_get_next_key(d_v6map.fd, &v6Key, &nextV6Key);
while (res == 0) {
if (bpf_lookup_elem(d_v6map.fd, &nextV6Key, &value) == 0) {
for (size_t idx = 0; idx < sizeof(nextV6Key); idx++) {
v6Addr.sin6_addr.s6_addr[idx] = nextV6Key[idx];
}
result.push_back(make_pair(ComboAddress(&v6Addr), value));
}
res = bpf_get_next_key(d_v6map.fd, &nextV6Key, &nextV6Key);
}
return result;
}
std::vector<std::tuple<DNSName, uint16_t, uint64_t> > BPFFilter::getQNameStats()
{
std::vector<std::tuple<DNSName, uint16_t, uint64_t> > result;
std::unique_lock<std::mutex> lock(d_mutex);
struct QNameKey key = { { 0 } };
struct QNameKey nextKey = { { 0 } };
struct QNameValue value;
int res = bpf_get_next_key(d_qnamemap.fd, &key, &nextKey);
while (res == 0) {
if (bpf_lookup_elem(d_qnamemap.fd, &nextKey, &value) == 0) {
nextKey.qname[sizeof(nextKey.qname) - 1 ] = '\0';
result.push_back(std::make_tuple(DNSName((const char*) nextKey.qname, sizeof(nextKey.qname), 0, false), value.qtype, value.counter));
}
res = bpf_get_next_key(d_qnamemap.fd, &nextKey, &nextKey);
}
return result;
}
#endif /* HAVE_EBPF */
|