-
Notifications
You must be signed in to change notification settings - Fork 5
/
seccomp-sandbox.c
242 lines (222 loc) · 7.72 KB
/
seccomp-sandbox.c
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
/* $OpenBSD: sandbox.c,v 1.9 2015/08/23 18:31:41 guenther Exp $ */
/*
* Copyright (c) 2012 Will Drewry <wad@dataspill.org>
* Copyright (c) 2015 Nicholas Marriott <nicm@openbsd.org>
*
* Permission to use, copy, modify, and distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
* WHATSOEVER RESULTING FROM LOSS OF MIND, USE, DATA OR PROFITS, WHETHER
* IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING
* OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
/* Linux seccomp_filter sandbox */
#ifdef HAVE_PRCTL
#include <sys/types.h>
#include <sys/ioctl.h>
#include <sys/syscall.h>
#include <sys/resource.h>
#include <sys/prctl.h>
#include <sys/mman.h>
#include <linux/net.h> /* SYS_RECVMSG/SYS_SENDMSG socketcall(2) */
#include <linux/audit.h>
#include <linux/filter.h>
#include <linux/seccomp.h>
#include <stddef.h> /* offsetof */
#include <stdint.h>
#include <errno.h>
#include <err.h>
#include "file.h"
/* XXX: */
#ifndef SECCOMP_AUDIT_ARCH
#if defined __arm__
#define SECCOMP_AUDIT_ARCH AUDIT_ARCH_ARM
#elif defined __i386__
#define SECCOMP_AUDIT_ARCH AUDIT_ARCH_I386
#elif defined __x86_64__ || defined __amd64__
#define SECCOMP_AUDIT_ARCH AUDIT_ARCH_X86_64
#endif
#endif /* SECCOMP_AUDIT_ARCH */
#ifdef DEBUG
#define SECCOMP_FILTER_FAIL SECCOMP_RET_TRAP
#else
#ifdef SECCOMP_RET_KILL_PROCESS
#define SECCOMP_FILTER_FAIL SECCOMP_RET_KILL_PROCESS
#else
#define SECCOMP_FILTER_FAIL SECCOMP_RET_KILL
#endif
#endif /* DEBUG */
#if __BYTE_ORDER == __LITTLE_ENDIAN
# define ARG_LO_OFFSET 0
# define ARG_HI_OFFSET sizeof(uint32_t)
#elif __BYTE_ORDER == __BIG_ENDIAN
# define ARG_LO_OFFSET sizeof(uint32_t)
# define ARG_HI_OFFSET 0
#else
#error "Unknown endianness"
#endif
/* Simple helpers to avoid manual errors (but larger BPF programs). */
#define SC_DENY(_nr, _errno) \
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, (_nr), 0, 1), \
BPF_STMT(BPF_RET+BPF_K, SECCOMP_RET_ERRNO|(_errno))
#define SC_ALLOW(_nr) \
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, (_nr), 0, 1), \
BPF_STMT(BPF_RET+BPF_K, SECCOMP_RET_ALLOW)
#define SC_ALLOW_ARG(_nr, _arg_nr, _arg_val) \
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, (_nr), 0, 6), \
/* load and test syscall argument, low word */ \
BPF_STMT(BPF_LD+BPF_W+BPF_ABS, \
offsetof(struct seccomp_data, args[(_arg_nr)]) + ARG_LO_OFFSET), \
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, \
((_arg_val) & 0xFFFFFFFF), 0, 3), \
/* load and test syscall argument, high word */ \
BPF_STMT(BPF_LD+BPF_W+BPF_ABS, \
offsetof(struct seccomp_data, args[(_arg_nr)]) + ARG_HI_OFFSET), \
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, \
(((uint32_t)((uint64_t)(_arg_val) >> 32)) & 0xFFFFFFFF), 0, 1), \
BPF_STMT(BPF_RET+BPF_K, SECCOMP_RET_ALLOW), \
/* reload syscall number; all rules expect it in accumulator */ \
BPF_STMT(BPF_LD+BPF_W+BPF_ABS, \
offsetof(struct seccomp_data, nr))
/* Allow if syscall argument contains only values in mask */
#define SC_ALLOW_ARG_MASK(_nr, _arg_nr, _arg_mask) \
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, (_nr), 0, 8), \
/* load, mask and test syscall argument, low word */ \
BPF_STMT(BPF_LD+BPF_W+BPF_ABS, \
offsetof(struct seccomp_data, args[(_arg_nr)]) + ARG_LO_OFFSET), \
BPF_STMT(BPF_ALU+BPF_AND+BPF_K, ~((_arg_mask) & 0xFFFFFFFF)), \
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0, 0, 4), \
/* load, mask and test syscall argument, high word */ \
BPF_STMT(BPF_LD+BPF_W+BPF_ABS, \
offsetof(struct seccomp_data, args[(_arg_nr)]) + ARG_HI_OFFSET), \
BPF_STMT(BPF_ALU+BPF_AND+BPF_K, \
~(((uint32_t)((uint64_t)(_arg_mask) >> 32)) & 0xFFFFFFFF)), \
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0, 0, 1), \
BPF_STMT(BPF_RET+BPF_K, SECCOMP_RET_ALLOW), \
/* reload syscall number; all rules expect it in accumulator */ \
BPF_STMT(BPF_LD+BPF_W+BPF_ABS, \
offsetof(struct seccomp_data, nr))
/* Deny unless syscall argument contains only values in mask */
#define SC_DENY_UNLESS_ARG_MASK(_nr, _arg_nr, _arg_mask, _errno) \
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, (_nr), 0, 8), \
/* load, mask and test syscall argument, low word */ \
BPF_STMT(BPF_LD+BPF_W+BPF_ABS, \
offsetof(struct seccomp_data, args[(_arg_nr)]) + ARG_LO_OFFSET), \
BPF_STMT(BPF_ALU+BPF_AND+BPF_K, ~((_arg_mask) & 0xFFFFFFFF)), \
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0, 0, 3), \
/* load, mask and test syscall argument, high word */ \
BPF_STMT(BPF_LD+BPF_W+BPF_ABS, \
offsetof(struct seccomp_data, args[(_arg_nr)]) + ARG_HI_OFFSET), \
BPF_STMT(BPF_ALU+BPF_AND+BPF_K, \
~(((uint32_t)((uint64_t)(_arg_mask) >> 32)) & 0xFFFFFFFF)), \
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0, 1, 0), \
BPF_STMT(BPF_RET+BPF_K, SECCOMP_RET_ERRNO|(_errno)), \
/* reload syscall number; all rules expect it in accumulator */ \
BPF_STMT(BPF_LD+BPF_W+BPF_ABS, \
offsetof(struct seccomp_data, nr))
#define SC_DENY_UNLESS_MASK(_nr, _arg_nr, _arg_val, _errno) \
#if defined(__NR_mmap) || defined(__NR_mmap2)
# ifdef MAP_FIXED_NOREPLACE
# define SC_MMAP_FLAGS MAP_PRIVATE|MAP_ANONYMOUS|MAP_FIXED|MAP_FIXED_NOREPLACE
# else
# define SC_MMAP_FLAGS MAP_PRIVATE|MAP_ANONYMOUS|MAP_FIXED
# endif /* MAP_FIXED_NOREPLACE */
#endif /* __NR_mmap || __NR_mmap2 */
/* Syscall filtering set */
static const struct sock_filter filt_insns[] = {
/* Ensure the syscall arch convention is as expected. */
BPF_STMT(BPF_LD+BPF_W+BPF_ABS,
offsetof(struct seccomp_data, arch)),
BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, SECCOMP_AUDIT_ARCH, 1, 0),
BPF_STMT(BPF_RET+BPF_K, SECCOMP_FILTER_FAIL),
/* Load the syscall number for checking. */
BPF_STMT(BPF_LD+BPF_W+BPF_ABS,
offsetof(struct seccomp_data, nr)),
/* Syscalls to non-fatally deny. */
#ifdef __NR_open
SC_DENY(__NR_open, EACCES),
#endif
#ifdef __NR_openat
SC_DENY(__NR_openat, EACCES),
#endif
#ifdef __NR_newfstatat
SC_DENY(__NR_newfstatat, EACCES),
#endif
#ifdef __NR_statx
SC_DENY(__NR_statx, EACCES),
#endif
/* Syscalls to permit. */
#ifdef __NR_brk
SC_ALLOW(__NR_brk),
#endif
#ifdef __NR_close
SC_ALLOW(__NR_close),
#endif
#ifdef __NR_exit_group
SC_ALLOW(__NR_exit_group),
#endif
#ifdef __NR_fstat
SC_ALLOW(__NR_fstat),
#endif
#ifdef __NR_fstat64
SC_ALLOW(__NR_fstat64),
#endif
#ifdef __NR_getpid
SC_ALLOW(__NR_getpid),
#endif
#ifdef __NR_mmap
SC_DENY_UNLESS_ARG_MASK(__NR_mmap, 3, SC_MMAP_FLAGS, EINVAL),
SC_ALLOW_ARG_MASK(__NR_mmap, 2, PROT_READ|PROT_WRITE|PROT_NONE),
#endif
#ifdef __NR_mmap2
SC_DENY_UNLESS_ARG_MASK(__NR_mmap2, 3, SC_MMAP_FLAGS, EINVAL),
SC_ALLOW_ARG_MASK(__NR_mmap2, 2, PROT_READ|PROT_WRITE|PROT_NONE),
#endif
#ifdef __NR_mprotect
SC_ALLOW_ARG_MASK(__NR_mprotect, 2, PROT_READ|PROT_WRITE|PROT_NONE),
#endif
#ifdef __NR_munmap
SC_ALLOW(__NR_munmap),
#endif
#ifdef __NR_read
SC_ALLOW(__NR_read),
#endif
#ifdef __NR_recvmsg
SC_ALLOW(__NR_recvmsg),
#endif
#ifdef __NR_sendmsg
SC_ALLOW(__NR_sendmsg),
#endif
/* Linux Multilib 32-on-64 binary nonsense */
#ifdef __NR_socketcall
SC_ALLOW_ARG(__NR_socketcall, 0, SYS_RECVMSG),
SC_ALLOW_ARG(__NR_socketcall, 0, SYS_SENDMSG),
#endif
#ifdef __NR_wait
SC_ALLOW(__NR_wait),
#endif
#ifdef __NR_write
SC_ALLOW(__NR_write),
#endif
/* Default deny. */
BPF_STMT(BPF_RET+BPF_K, SECCOMP_FILTER_FAIL),
};
static const struct sock_fprog filt_program = {
.len = (unsigned short)(sizeof(filt_insns)/sizeof(filt_insns[0])),
.filter = (struct sock_filter *)filt_insns,
};
void
sandbox_seccomp(void)
{
if (prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0) == -1)
err(1, "prctl(PR_SET_NO_NEW_PRIVS)");
if (prctl(PR_SET_SECCOMP, SECCOMP_MODE_FILTER,
&filt_program) == -1)
err(1, "prctl(PR_SET_SECCOMP/SECCOMP_MODE_FILTER)");
}
#endif /* HAVE_PRCTL */