Blame view
kernel/seccomp.c
60.1 KB
b24413180 License cleanup: ... |
1 |
// SPDX-License-Identifier: GPL-2.0 |
1da177e4c Linux-2.6.12-rc2 |
2 3 4 5 6 |
/* * linux/kernel/seccomp.c * * Copyright 2004-2005 Andrea Arcangeli <andrea@cpushare.com> * |
e2cfabdfd seccomp: add syst... |
7 8 9 10 11 12 13 14 |
* Copyright (C) 2012 Google, Inc. * Will Drewry <wad@chromium.org> * * This defines a simple but solid secure-computing facility. * * Mode 1 uses a fixed list of allowed system calls. * Mode 2 allows user-defined system call filters in the form * of Berkeley Packet Filters/Linux Socket Filters. |
1da177e4c Linux-2.6.12-rc2 |
15 |
*/ |
e68f9d49d seccomp: Use pr_fmt |
16 |
#define pr_fmt(fmt) "seccomp: " fmt |
1da177e4c Linux-2.6.12-rc2 |
17 |
|
0b5fa2290 seccomp: Switch f... |
18 |
#include <linux/refcount.h> |
85e7bac33 seccomp: audit ab... |
19 |
#include <linux/audit.h> |
5b1017404 x86-64: seccomp: ... |
20 |
#include <linux/compat.h> |
b25e67161 seccomp: dump cor... |
21 |
#include <linux/coredump.h> |
8e5f1ad11 seccomp: Sysctl t... |
22 |
#include <linux/kmemleak.h> |
5c3070890 seccomp: Enable s... |
23 24 |
#include <linux/nospec.h> #include <linux/prctl.h> |
e2cfabdfd seccomp: add syst... |
25 |
#include <linux/sched.h> |
68db0cf10 sched/headers: Pr... |
26 |
#include <linux/sched/task_stack.h> |
e2cfabdfd seccomp: add syst... |
27 |
#include <linux/seccomp.h> |
c8bee430d seccomp: split fi... |
28 |
#include <linux/slab.h> |
48dc92b9f seccomp: add "sec... |
29 |
#include <linux/syscalls.h> |
8e5f1ad11 seccomp: Sysctl t... |
30 |
#include <linux/sysctl.h> |
1da177e4c Linux-2.6.12-rc2 |
31 |
|
a4412fc94 seccomp,x86,arm,m... |
32 |
#ifdef CONFIG_HAVE_ARCH_SECCOMP_FILTER |
e2cfabdfd seccomp: add syst... |
33 |
#include <asm/syscall.h> |
a4412fc94 seccomp,x86,arm,m... |
34 |
#endif |
e2cfabdfd seccomp: add syst... |
35 36 |
#ifdef CONFIG_SECCOMP_FILTER |
6a21cc50f seccomp: add a re... |
37 |
#include <linux/file.h> |
e2cfabdfd seccomp: add syst... |
38 |
#include <linux/filter.h> |
c2e1f2e30 seccomp: implemen... |
39 |
#include <linux/pid.h> |
fb0fadf9b ptrace,seccomp: A... |
40 |
#include <linux/ptrace.h> |
fb14528e4 seccomp: Set PF_S... |
41 |
#include <linux/capability.h> |
e2cfabdfd seccomp: add syst... |
42 43 |
#include <linux/tracehook.h> #include <linux/uaccess.h> |
6a21cc50f seccomp: add a re... |
44 |
#include <linux/anon_inodes.h> |
9f87dcf14 seccomp: Add find... |
45 |
#include <linux/lockdep.h> |
6a21cc50f seccomp: add a re... |
46 |
|
47e33c05f seccomp: Fix ioct... |
47 48 49 50 51 52 53 |
/* * When SECCOMP_IOCTL_NOTIF_ID_VALID was first introduced, it had the * wrong direction flag in the ioctl number. This is the broken one, * which the kernel needs to keep supporting until all userspaces stop * using the wrong command number. */ #define SECCOMP_IOCTL_NOTIF_ID_VALID_WRONG_DIR SECCOMP_IOR(2, __u64) |
6a21cc50f seccomp: add a re... |
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 |
enum notify_state { SECCOMP_NOTIFY_INIT, SECCOMP_NOTIFY_SENT, SECCOMP_NOTIFY_REPLIED, }; struct seccomp_knotif { /* The struct pid of the task whose filter triggered the notification */ struct task_struct *task; /* The "cookie" for this request; this is unique for this filter. */ u64 id; /* * The seccomp data. This pointer is valid the entire time this * notification is active, since it comes from __seccomp_filter which * eclipses the entire lifecycle here. */ const struct seccomp_data *data; /* * Notification states. When SECCOMP_RET_USER_NOTIF is returned, a * struct seccomp_knotif is created and starts out in INIT. Once the * handler reads the notification off of an FD, it transitions to SENT. * If a signal is received the state transitions back to INIT and * another message is sent. When the userspace handler replies, state * transitions to REPLIED. */ enum notify_state state; /* The return values, only valid when in SECCOMP_NOTIFY_REPLIED */ int error; long val; |
fb3c5386b seccomp: add SECC... |
87 |
u32 flags; |
6a21cc50f seccomp: add a re... |
88 |
|
7cf97b125 seccomp: Introduc... |
89 90 91 92 |
/* * Signals when this has changed states, such as the listener * dying, a new seccomp addfd message, or changing to REPLIED */ |
6a21cc50f seccomp: add a re... |
93 94 95 |
struct completion ready; struct list_head list; |
7cf97b125 seccomp: Introduc... |
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 |
/* outstanding addfd requests */ struct list_head addfd; }; /** * struct seccomp_kaddfd - container for seccomp_addfd ioctl messages * * @file: A reference to the file to install in the other task * @fd: The fd number to install it at. If the fd number is -1, it means the * installing process should allocate the fd as normal. * @flags: The flags for the new file descriptor. At the moment, only O_CLOEXEC * is allowed. * @ret: The return value of the installing process. It is set to the fd num * upon success (>= 0). * @completion: Indicates that the installing process has completed fd * installation, or gone away (either due to successful * reply, or signal) * */ struct seccomp_kaddfd { struct file *file; int fd; unsigned int flags; /* To only be set on reply */ int ret; struct completion completion; struct list_head list; |
6a21cc50f seccomp: add a re... |
125 126 127 128 129 130 131 132 133 134 135 136 137 |
}; /** * struct notification - container for seccomp userspace notifications. Since * most seccomp filters will not have notification listeners attached and this * structure is fairly large, we store the notification-specific stuff in a * separate structure. * * @request: A semaphore that users of this notification can wait on for * changes. Actual reads and writes are still controlled with * filter->notify_lock. * @next_id: The id of the next request. * @notifications: A list of struct seccomp_knotif elements. |
6a21cc50f seccomp: add a re... |
138 139 140 141 142 |
*/ struct notification { struct semaphore request; u64 next_id; struct list_head notifications; |
6a21cc50f seccomp: add a re... |
143 |
}; |
e2cfabdfd seccomp: add syst... |
144 |
|
f89fef0ee UPSTREAM: seccomp... |
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 |
#ifdef SECCOMP_ARCH_NATIVE /** * struct action_cache - per-filter cache of seccomp actions per * arch/syscall pair * * @allow_native: A bitmap where each bit represents whether the * filter will always allow the syscall, for the * native architecture. * @allow_compat: A bitmap where each bit represents whether the * filter will always allow the syscall, for the * compat architecture. */ struct action_cache { DECLARE_BITMAP(allow_native, SECCOMP_ARCH_NATIVE_NR); #ifdef SECCOMP_ARCH_COMPAT DECLARE_BITMAP(allow_compat, SECCOMP_ARCH_COMPAT_NR); #endif }; #else struct action_cache { }; static inline bool seccomp_cache_check_allow(const struct seccomp_filter *sfilter, const struct seccomp_data *sd) { return false; } |
2d660e977 UPSTREAM: seccomp... |
171 172 173 174 |
static inline void seccomp_cache_prepare(struct seccomp_filter *sfilter) { } |
f89fef0ee UPSTREAM: seccomp... |
175 |
#endif /* SECCOMP_ARCH_NATIVE */ |
e2cfabdfd seccomp: add syst... |
176 177 178 |
/** * struct seccomp_filter - container for seccomp BPF programs * |
b707ddee1 seccomp: rename "... |
179 180 181 182 183 |
* @refs: Reference count to manage the object lifetime. * A filter's reference count is incremented for each directly * attached task, once for the dependent filter, and if * requested for the user notifier. When @refs reaches zero, * the filter can be freed. |
99cdb8b9a seccomp: notify a... |
184 185 186 187 188 189 190 191 |
* @users: A filter's @users count is incremented for each directly * attached task (filter installation, fork(), thread_sync), * and once for the dependent filter (tracked in filter->prev). * When it reaches zero it indicates that no direct or indirect * users of that filter exist. No new tasks can get associated with * this filter after reaching 0. The @users count is always smaller * or equal to @refs. Hence, reaching 0 for @users does not mean * the filter can be freed. |
2d660e977 UPSTREAM: seccomp... |
192 |
* @cache: cache of arch/syscall mappings to actions |
e66a39977 seccomp: Filter f... |
193 |
* @log: true if all actions except for SECCOMP_RET_ALLOW should be logged |
e2cfabdfd seccomp: add syst... |
194 |
* @prev: points to a previously installed, or inherited, filter |
285fdfc5d seccomp: Fix docu... |
195 |
* @prog: the BPF program to evaluate |
6a21cc50f seccomp: add a re... |
196 197 |
* @notif: the struct that holds all notification related information * @notify_lock: A lock for all notification-related accesses. |
76194c4e8 seccomp: Lift wai... |
198 |
* @wqh: A wait queue for poll if a notifier is in use. |
e2cfabdfd seccomp: add syst... |
199 200 201 202 203 204 205 206 207 |
* * seccomp_filter objects are organized in a tree linked via the @prev * pointer. For any task, it appears to be a singly-linked list starting * with current->seccomp.filter, the most recently attached or inherited filter. * However, multiple filters may share a @prev node, by way of fork(), which * results in a unidirectional tree existing in memory. This is similar to * how namespaces work. * * seccomp_filter objects should never be modified after being attached |
b707ddee1 seccomp: rename "... |
208 |
* to a task_struct (other than @refs). |
e2cfabdfd seccomp: add syst... |
209 210 |
*/ struct seccomp_filter { |
b707ddee1 seccomp: rename "... |
211 |
refcount_t refs; |
99cdb8b9a seccomp: notify a... |
212 |
refcount_t users; |
e66a39977 seccomp: Filter f... |
213 |
bool log; |
2d660e977 UPSTREAM: seccomp... |
214 |
struct action_cache cache; |
e2cfabdfd seccomp: add syst... |
215 |
struct seccomp_filter *prev; |
7ae457c1e net: filter: spli... |
216 |
struct bpf_prog *prog; |
6a21cc50f seccomp: add a re... |
217 218 |
struct notification *notif; struct mutex notify_lock; |
76194c4e8 seccomp: Lift wai... |
219 |
wait_queue_head_t wqh; |
e2cfabdfd seccomp: add syst... |
220 221 222 223 |
}; /* Limit any path through the tree to 256KB worth of instructions. */ #define MAX_INSNS_PER_PATH ((1 << 18) / sizeof(struct sock_filter)) |
bd4cf0ed3 net: filter: rewo... |
224 |
/* |
e2cfabdfd seccomp: add syst... |
225 226 227 |
* Endianness is explicitly ignored and left for BPF program authors to manage * as per the specific architecture. */ |
bd4cf0ed3 net: filter: rewo... |
228 |
static void populate_seccomp_data(struct seccomp_data *sd) |
e2cfabdfd seccomp: add syst... |
229 |
{ |
2d9ca267a seccomp: Use curr... |
230 231 232 233 |
/* * Instead of using current_pt_reg(), we're already doing the work * to safely fetch "current", so just use "task" everywhere below. */ |
bd4cf0ed3 net: filter: rewo... |
234 235 |
struct task_struct *task = current; struct pt_regs *regs = task_pt_regs(task); |
2eac76483 seccomp: fix popu... |
236 |
unsigned long args[6]; |
e2cfabdfd seccomp: add syst... |
237 |
|
bd4cf0ed3 net: filter: rewo... |
238 |
sd->nr = syscall_get_nr(task, regs); |
16add4116 syscall_get_arch:... |
239 |
sd->arch = syscall_get_arch(task); |
b35f549df syscalls: Remove ... |
240 |
syscall_get_arguments(task, regs, args); |
2eac76483 seccomp: fix popu... |
241 242 243 244 245 246 |
sd->args[0] = args[0]; sd->args[1] = args[1]; sd->args[2] = args[2]; sd->args[3] = args[3]; sd->args[4] = args[4]; sd->args[5] = args[5]; |
bd4cf0ed3 net: filter: rewo... |
247 |
sd->instruction_pointer = KSTK_EIP(task); |
e2cfabdfd seccomp: add syst... |
248 249 250 251 252 253 254 |
} /** * seccomp_check_filter - verify seccomp filter code * @filter: filter to verify * @flen: length of filter * |
4df95ff48 net: filter: rena... |
255 |
* Takes a previously checked filter (by bpf_check_classic) and |
e2cfabdfd seccomp: add syst... |
256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 |
* redirects all filter code that loads struct sk_buff data * and related data through seccomp_bpf_load. It also * enforces length and alignment checking of those loads. * * Returns 0 if the rule set is legal or -EINVAL if not. */ static int seccomp_check_filter(struct sock_filter *filter, unsigned int flen) { int pc; for (pc = 0; pc < flen; pc++) { struct sock_filter *ftest = &filter[pc]; u16 code = ftest->code; u32 k = ftest->k; switch (code) { |
348059313 net: filter: get ... |
271 |
case BPF_LD | BPF_W | BPF_ABS: |
bd4cf0ed3 net: filter: rewo... |
272 |
ftest->code = BPF_LDX | BPF_W | BPF_ABS; |
e2cfabdfd seccomp: add syst... |
273 274 275 276 |
/* 32-bit aligned and not out of bounds. */ if (k >= sizeof(struct seccomp_data) || k & 3) return -EINVAL; continue; |
348059313 net: filter: get ... |
277 |
case BPF_LD | BPF_W | BPF_LEN: |
bd4cf0ed3 net: filter: rewo... |
278 |
ftest->code = BPF_LD | BPF_IMM; |
e2cfabdfd seccomp: add syst... |
279 280 |
ftest->k = sizeof(struct seccomp_data); continue; |
348059313 net: filter: get ... |
281 |
case BPF_LDX | BPF_W | BPF_LEN: |
bd4cf0ed3 net: filter: rewo... |
282 |
ftest->code = BPF_LDX | BPF_IMM; |
e2cfabdfd seccomp: add syst... |
283 284 285 |
ftest->k = sizeof(struct seccomp_data); continue; /* Explicitly include allowed calls. */ |
348059313 net: filter: get ... |
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 |
case BPF_RET | BPF_K: case BPF_RET | BPF_A: case BPF_ALU | BPF_ADD | BPF_K: case BPF_ALU | BPF_ADD | BPF_X: case BPF_ALU | BPF_SUB | BPF_K: case BPF_ALU | BPF_SUB | BPF_X: case BPF_ALU | BPF_MUL | BPF_K: case BPF_ALU | BPF_MUL | BPF_X: case BPF_ALU | BPF_DIV | BPF_K: case BPF_ALU | BPF_DIV | BPF_X: case BPF_ALU | BPF_AND | BPF_K: case BPF_ALU | BPF_AND | BPF_X: case BPF_ALU | BPF_OR | BPF_K: case BPF_ALU | BPF_OR | BPF_X: case BPF_ALU | BPF_XOR | BPF_K: case BPF_ALU | BPF_XOR | BPF_X: case BPF_ALU | BPF_LSH | BPF_K: case BPF_ALU | BPF_LSH | BPF_X: case BPF_ALU | BPF_RSH | BPF_K: case BPF_ALU | BPF_RSH | BPF_X: case BPF_ALU | BPF_NEG: case BPF_LD | BPF_IMM: case BPF_LDX | BPF_IMM: case BPF_MISC | BPF_TAX: case BPF_MISC | BPF_TXA: case BPF_LD | BPF_MEM: case BPF_LDX | BPF_MEM: case BPF_ST: case BPF_STX: case BPF_JMP | BPF_JA: case BPF_JMP | BPF_JEQ | BPF_K: case BPF_JMP | BPF_JEQ | BPF_X: case BPF_JMP | BPF_JGE | BPF_K: case BPF_JMP | BPF_JGE | BPF_X: case BPF_JMP | BPF_JGT | BPF_K: case BPF_JMP | BPF_JGT | BPF_X: case BPF_JMP | BPF_JSET | BPF_K: case BPF_JMP | BPF_JSET | BPF_X: |
e2cfabdfd seccomp: add syst... |
324 325 326 327 328 329 330 |
continue; default: return -EINVAL; } } return 0; } |
f89fef0ee UPSTREAM: seccomp... |
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 |
#ifdef SECCOMP_ARCH_NATIVE static inline bool seccomp_cache_check_allow_bitmap(const void *bitmap, size_t bitmap_size, int syscall_nr) { if (unlikely(syscall_nr < 0 || syscall_nr >= bitmap_size)) return false; syscall_nr = array_index_nospec(syscall_nr, bitmap_size); return test_bit(syscall_nr, bitmap); } /** * seccomp_cache_check_allow - lookup seccomp cache * @sfilter: The seccomp filter * @sd: The seccomp data to lookup the cache with * * Returns true if the seccomp_data is cached and allowed. */ static inline bool seccomp_cache_check_allow(const struct seccomp_filter *sfilter, const struct seccomp_data *sd) { int syscall_nr = sd->nr; const struct action_cache *cache = &sfilter->cache; #ifndef SECCOMP_ARCH_COMPAT /* A native-only architecture doesn't need to check sd->arch. */ return seccomp_cache_check_allow_bitmap(cache->allow_native, SECCOMP_ARCH_NATIVE_NR, syscall_nr); #else if (likely(sd->arch == SECCOMP_ARCH_NATIVE)) return seccomp_cache_check_allow_bitmap(cache->allow_native, SECCOMP_ARCH_NATIVE_NR, syscall_nr); if (likely(sd->arch == SECCOMP_ARCH_COMPAT)) return seccomp_cache_check_allow_bitmap(cache->allow_compat, SECCOMP_ARCH_COMPAT_NR, syscall_nr); #endif /* SECCOMP_ARCH_COMPAT */ WARN_ON_ONCE(true); return false; } #endif /* SECCOMP_ARCH_NATIVE */ |
e2cfabdfd seccomp: add syst... |
376 |
/** |
285fdfc5d seccomp: Fix docu... |
377 378 |
* seccomp_run_filters - evaluates all seccomp filters against @sd * @sd: optional seccomp data to be passed to filters |
deb4de8b3 seccomp: Provide ... |
379 380 381 |
* @match: stores struct seccomp_filter that resulted in the return value, * unless filter returned SECCOMP_RET_ALLOW, in which case it will * be unchanged. |
e2cfabdfd seccomp: add syst... |
382 383 384 |
* * Returns valid seccomp BPF response codes. */ |
0466bdb99 seccomp: Implemen... |
385 |
#define ACTION_ONLY(ret) ((s32)((ret) & (SECCOMP_RET_ACTION_FULL))) |
deb4de8b3 seccomp: Provide ... |
386 387 |
static u32 seccomp_run_filters(const struct seccomp_data *sd, struct seccomp_filter **match) |
e2cfabdfd seccomp: add syst... |
388 |
{ |
acf3b2c71 seccomp: add SECC... |
389 |
u32 ret = SECCOMP_RET_ALLOW; |
8225d3853 seccomp: Replace ... |
390 391 |
/* Make sure cross-thread synced filter points somewhere sane. */ struct seccomp_filter *f = |
506458efa locking/barriers:... |
392 |
READ_ONCE(current->seccomp.filter); |
acf3b2c71 seccomp: add SECC... |
393 394 |
/* Ensure unexpected behavior doesn't result in failing open. */ |
0d42d73a3 seccomp: remove u... |
395 |
if (WARN_ON(f == NULL)) |
4d3b0b05a seccomp: Introduc... |
396 |
return SECCOMP_RET_KILL_PROCESS; |
acf3b2c71 seccomp: add SECC... |
397 |
|
f89fef0ee UPSTREAM: seccomp... |
398 399 |
if (seccomp_cache_check_allow(f, sd)) return SECCOMP_RET_ALLOW; |
e2cfabdfd seccomp: add syst... |
400 401 |
/* * All filters in the list are evaluated and the lowest BPF return |
acf3b2c71 seccomp: add SECC... |
402 |
* value always takes priority (ignoring the DATA). |
e2cfabdfd seccomp: add syst... |
403 |
*/ |
3ba2530cc seccomp: allow mo... |
404 |
for (; f; f = f->prev) { |
3d9f773cf bpf: Use bpf_prog... |
405 |
u32 cur_ret = bpf_prog_run_pin_on_cpu(f->prog, sd); |
8f577cadf seccomp: JIT comp... |
406 |
|
0466bdb99 seccomp: Implemen... |
407 |
if (ACTION_ONLY(cur_ret) < ACTION_ONLY(ret)) { |
acf3b2c71 seccomp: add SECC... |
408 |
ret = cur_ret; |
deb4de8b3 seccomp: Provide ... |
409 410 |
*match = f; } |
e2cfabdfd seccomp: add syst... |
411 412 413 |
} return ret; } |
1f41b4504 seccomp: extract ... |
414 |
#endif /* CONFIG_SECCOMP_FILTER */ |
e2cfabdfd seccomp: add syst... |
415 |
|
1f41b4504 seccomp: extract ... |
416 417 |
static inline bool seccomp_may_assign_mode(unsigned long seccomp_mode) { |
69f6a34bd seccomp: Replace ... |
418 |
assert_spin_locked(¤t->sighand->siglock); |
dbd952127 seccomp: introduc... |
419 |
|
1f41b4504 seccomp: extract ... |
420 421 422 423 424 |
if (current->seccomp.mode && current->seccomp.mode != seccomp_mode) return false; return true; } |
8bf37d8c0 seccomp: Move spe... |
425 |
void __weak arch_seccomp_spec_mitigate(struct task_struct *task) { } |
5c3070890 seccomp: Enable s... |
426 |
|
3ba2530cc seccomp: allow mo... |
427 |
static inline void seccomp_assign_mode(struct task_struct *task, |
00a02d0c5 seccomp: Add filt... |
428 429 |
unsigned long seccomp_mode, unsigned long flags) |
1f41b4504 seccomp: extract ... |
430 |
{ |
69f6a34bd seccomp: Replace ... |
431 |
assert_spin_locked(&task->sighand->siglock); |
dbd952127 seccomp: introduc... |
432 |
|
3ba2530cc seccomp: allow mo... |
433 434 435 436 437 438 |
task->seccomp.mode = seccomp_mode; /* * Make sure TIF_SECCOMP cannot be set before the mode (and * filter) is set. */ smp_mb__before_atomic(); |
00a02d0c5 seccomp: Add filt... |
439 440 |
/* Assume default seccomp processes want spec flaw mitigation. */ if ((flags & SECCOMP_FILTER_FLAG_SPEC_ALLOW) == 0) |
8bf37d8c0 seccomp: Move spe... |
441 |
arch_seccomp_spec_mitigate(task); |
3ba2530cc seccomp: allow mo... |
442 |
set_tsk_thread_flag(task, TIF_SECCOMP); |
1f41b4504 seccomp: extract ... |
443 444 445 |
} #ifdef CONFIG_SECCOMP_FILTER |
c2e1f2e30 seccomp: implemen... |
446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 |
/* Returns 1 if the parent is an ancestor of the child. */ static int is_ancestor(struct seccomp_filter *parent, struct seccomp_filter *child) { /* NULL is the root ancestor. */ if (parent == NULL) return 1; for (; child; child = child->prev) if (child == parent) return 1; return 0; } /** * seccomp_can_sync_threads: checks if all threads can be synchronized * * Expects sighand and cred_guard_mutex locks to be held. * * Returns 0 on success, -ve on error, or the pid of a thread which was |
6beff00b7 seccomp: fix up g... |
465 |
* either not in the correct seccomp mode or did not have an ancestral |
c2e1f2e30 seccomp: implemen... |
466 467 468 469 470 471 472 |
* seccomp filter. */ static inline pid_t seccomp_can_sync_threads(void) { struct task_struct *thread, *caller; BUG_ON(!mutex_is_locked(¤t->signal->cred_guard_mutex)); |
69f6a34bd seccomp: Replace ... |
473 |
assert_spin_locked(¤t->sighand->siglock); |
c2e1f2e30 seccomp: implemen... |
474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 |
/* Validate all threads being eligible for synchronization. */ caller = current; for_each_thread(caller, thread) { pid_t failed; /* Skip current, since it is initiating the sync. */ if (thread == caller) continue; if (thread->seccomp.mode == SECCOMP_MODE_DISABLED || (thread->seccomp.mode == SECCOMP_MODE_FILTER && is_ancestor(thread->seccomp.filter, caller->seccomp.filter))) continue; /* Return the first thread that cannot be synchronized. */ failed = task_pid_vnr(thread); /* If the pid cannot be resolved, then return -ESRCH */ |
0d42d73a3 seccomp: remove u... |
493 |
if (WARN_ON(failed == 0)) |
c2e1f2e30 seccomp: implemen... |
494 495 496 497 498 499 |
failed = -ESRCH; return failed; } return 0; } |
3a15fb6ed seccomp: release ... |
500 501 502 503 504 505 506 |
static inline void seccomp_filter_free(struct seccomp_filter *filter) { if (filter) { bpf_prog_destroy(filter->prog); kfree(filter); } } |
99cdb8b9a seccomp: notify a... |
507 508 509 510 511 512 513 514 |
static void __seccomp_filter_orphan(struct seccomp_filter *orig) { while (orig && refcount_dec_and_test(&orig->users)) { if (waitqueue_active(&orig->wqh)) wake_up_poll(&orig->wqh, EPOLLHUP); orig = orig->prev; } } |
3a15fb6ed seccomp: release ... |
515 516 517 518 519 520 521 522 523 |
static void __put_seccomp_filter(struct seccomp_filter *orig) { /* Clean up single-reference branches iteratively. */ while (orig && refcount_dec_and_test(&orig->refs)) { struct seccomp_filter *freeme = orig; orig = orig->prev; seccomp_filter_free(freeme); } } |
99cdb8b9a seccomp: notify a... |
524 525 526 527 528 529 530 |
static void __seccomp_filter_release(struct seccomp_filter *orig) { /* Notify about any unused filters in the task's former filter tree. */ __seccomp_filter_orphan(orig); /* Finally drop all references to the task's former tree. */ __put_seccomp_filter(orig); } |
3a15fb6ed seccomp: release ... |
531 |
/** |
99cdb8b9a seccomp: notify a... |
532 533 534 |
* seccomp_filter_release - Detach the task from its filter tree, * drop its reference count, and notify * about unused filters |
3a15fb6ed seccomp: release ... |
535 536 537 538 539 540 541 542 543 544 545 |
* * This function should only be called when the task is exiting as * it detaches it from its filter tree. As such, READ_ONCE() and * barriers are not needed here, as would normally be needed. */ void seccomp_filter_release(struct task_struct *tsk) { struct seccomp_filter *orig = tsk->seccomp.filter; /* Detach task from its filter tree. */ tsk->seccomp.filter = NULL; |
99cdb8b9a seccomp: notify a... |
546 |
__seccomp_filter_release(orig); |
3a15fb6ed seccomp: release ... |
547 |
} |
c2e1f2e30 seccomp: implemen... |
548 549 550 551 552 553 554 555 |
/** * seccomp_sync_threads: sets all threads to use current's filter * * Expects sighand and cred_guard_mutex locks to be held, and for * seccomp_can_sync_threads() to have returned success already * without dropping the locks. * */ |
00a02d0c5 seccomp: Add filt... |
556 |
static inline void seccomp_sync_threads(unsigned long flags) |
c2e1f2e30 seccomp: implemen... |
557 558 559 560 |
{ struct task_struct *thread, *caller; BUG_ON(!mutex_is_locked(¤t->signal->cred_guard_mutex)); |
69f6a34bd seccomp: Replace ... |
561 |
assert_spin_locked(¤t->sighand->siglock); |
c2e1f2e30 seccomp: implemen... |
562 563 564 565 566 567 568 569 570 571 |
/* Synchronize all threads. */ caller = current; for_each_thread(caller, thread) { /* Skip current, since it needs no changes. */ if (thread == caller) continue; /* Get a task reference for the new leaf node. */ get_seccomp_filter(caller); |
99cdb8b9a seccomp: notify a... |
572 |
|
c2e1f2e30 seccomp: implemen... |
573 574 575 576 577 |
/* * Drop the task reference to the shared ancestor since * current's path will hold a reference. (This also * allows a put before the assignment.) */ |
99cdb8b9a seccomp: notify a... |
578 579 580 |
__seccomp_filter_release(thread->seccomp.filter); /* Make our new filter tree visible. */ |
c2e1f2e30 seccomp: implemen... |
581 582 |
smp_store_release(&thread->seccomp.filter, caller->seccomp.filter); |
c818c03b6 seccomp: Report n... |
583 584 |
atomic_set(&thread->seccomp.filter_count, atomic_read(&thread->seccomp.filter_count)); |
103502a35 seccomp: always p... |
585 586 587 588 589 590 591 592 593 |
/* * Don't let an unprivileged task work around * the no_new_privs restriction by creating * a thread that sets it up, enters seccomp, * then dies. */ if (task_no_new_privs(caller)) task_set_no_new_privs(thread); |
c2e1f2e30 seccomp: implemen... |
594 595 596 597 598 599 |
/* * Opt the other thread into seccomp if needed. * As threads are considered to be trust-realm * equivalent (see ptrace_may_access), it is safe to * allow one thread to transition the other. */ |
103502a35 seccomp: always p... |
600 |
if (thread->seccomp.mode == SECCOMP_MODE_DISABLED) |
00a02d0c5 seccomp: Add filt... |
601 602 |
seccomp_assign_mode(thread, SECCOMP_MODE_FILTER, flags); |
c2e1f2e30 seccomp: implemen... |
603 604 |
} } |
e2cfabdfd seccomp: add syst... |
605 |
/** |
c8bee430d seccomp: split fi... |
606 |
* seccomp_prepare_filter: Prepares a seccomp filter for use. |
e2cfabdfd seccomp: add syst... |
607 608 |
* @fprog: BPF program to install * |
c8bee430d seccomp: split fi... |
609 |
* Returns filter on success or an ERR_PTR on failure. |
e2cfabdfd seccomp: add syst... |
610 |
*/ |
c8bee430d seccomp: split fi... |
611 |
static struct seccomp_filter *seccomp_prepare_filter(struct sock_fprog *fprog) |
e2cfabdfd seccomp: add syst... |
612 |
{ |
ac67eb2c5 seccomp, filter: ... |
613 614 |
struct seccomp_filter *sfilter; int ret; |
2d660e977 UPSTREAM: seccomp... |
615 616 617 618 619 620 |
const bool save_orig = #if defined(CONFIG_CHECKPOINT_RESTORE) || defined(SECCOMP_ARCH_NATIVE) true; #else false; #endif |
e2cfabdfd seccomp: add syst... |
621 622 |
if (fprog->len == 0 || fprog->len > BPF_MAXINSNS) |
c8bee430d seccomp: split fi... |
623 |
return ERR_PTR(-EINVAL); |
d9e12f42e seccomp: simplify... |
624 |
|
c8bee430d seccomp: split fi... |
625 |
BUG_ON(INT_MAX / fprog->len < sizeof(struct sock_filter)); |
e2cfabdfd seccomp: add syst... |
626 627 |
/* |
119ce5c8b kernel/seccomp.c:... |
628 |
* Installing a seccomp filter requires that the task has |
e2cfabdfd seccomp: add syst... |
629 630 631 632 |
* CAP_SYS_ADMIN in its namespace or be running with no_new_privs. * This avoids scenarios where unprivileged tasks can affect the * behavior of privileged children. */ |
1d4457f99 sched: move no_ne... |
633 |
if (!task_no_new_privs(current) && |
fb14528e4 seccomp: Set PF_S... |
634 |
!ns_capable_noaudit(current_user_ns(), CAP_SYS_ADMIN)) |
c8bee430d seccomp: split fi... |
635 |
return ERR_PTR(-EACCES); |
e2cfabdfd seccomp: add syst... |
636 |
|
bd4cf0ed3 net: filter: rewo... |
637 |
/* Allocate a new seccomp_filter */ |
ac67eb2c5 seccomp, filter: ... |
638 639 |
sfilter = kzalloc(sizeof(*sfilter), GFP_KERNEL | __GFP_NOWARN); if (!sfilter) |
d9e12f42e seccomp: simplify... |
640 |
return ERR_PTR(-ENOMEM); |
ac67eb2c5 seccomp, filter: ... |
641 |
|
6a21cc50f seccomp: add a re... |
642 |
mutex_init(&sfilter->notify_lock); |
ac67eb2c5 seccomp, filter: ... |
643 |
ret = bpf_prog_create_from_user(&sfilter->prog, fprog, |
f8e529ed9 seccomp, ptrace: ... |
644 |
seccomp_check_filter, save_orig); |
ac67eb2c5 seccomp, filter: ... |
645 646 647 |
if (ret < 0) { kfree(sfilter); return ERR_PTR(ret); |
d9e12f42e seccomp: simplify... |
648 |
} |
bd4cf0ed3 net: filter: rewo... |
649 |
|
b707ddee1 seccomp: rename "... |
650 |
refcount_set(&sfilter->refs, 1); |
99cdb8b9a seccomp: notify a... |
651 |
refcount_set(&sfilter->users, 1); |
76194c4e8 seccomp: Lift wai... |
652 |
init_waitqueue_head(&sfilter->wqh); |
e2cfabdfd seccomp: add syst... |
653 |
|
ac67eb2c5 seccomp, filter: ... |
654 |
return sfilter; |
e2cfabdfd seccomp: add syst... |
655 656 657 |
} /** |
c8bee430d seccomp: split fi... |
658 |
* seccomp_prepare_user_filter - prepares a user-supplied sock_fprog |
e2cfabdfd seccomp: add syst... |
659 660 661 662 |
* @user_filter: pointer to the user data containing a sock_fprog. * * Returns 0 on success and non-zero otherwise. */ |
c8bee430d seccomp: split fi... |
663 664 |
static struct seccomp_filter * seccomp_prepare_user_filter(const char __user *user_filter) |
e2cfabdfd seccomp: add syst... |
665 666 |
{ struct sock_fprog fprog; |
c8bee430d seccomp: split fi... |
667 |
struct seccomp_filter *filter = ERR_PTR(-EFAULT); |
e2cfabdfd seccomp: add syst... |
668 669 |
#ifdef CONFIG_COMPAT |
5c38065e0 seccomp: check in... |
670 |
if (in_compat_syscall()) { |
e2cfabdfd seccomp: add syst... |
671 672 673 674 675 676 677 678 679 |
struct compat_sock_fprog fprog32; if (copy_from_user(&fprog32, user_filter, sizeof(fprog32))) goto out; fprog.len = fprog32.len; fprog.filter = compat_ptr(fprog32.filter); } else /* falls through to the if below. */ #endif if (copy_from_user(&fprog, user_filter, sizeof(fprog))) goto out; |
c8bee430d seccomp: split fi... |
680 |
filter = seccomp_prepare_filter(&fprog); |
e2cfabdfd seccomp: add syst... |
681 |
out: |
c8bee430d seccomp: split fi... |
682 683 |
return filter; } |
2d660e977 UPSTREAM: seccomp... |
684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 |
#ifdef SECCOMP_ARCH_NATIVE /** * seccomp_is_const_allow - check if filter is constant allow with given data * @fprog: The BPF programs * @sd: The seccomp data to check against, only syscall number and arch * number are considered constant. */ static bool seccomp_is_const_allow(struct sock_fprog_kern *fprog, struct seccomp_data *sd) { unsigned int reg_value = 0; unsigned int pc; bool op_res; if (WARN_ON_ONCE(!fprog)) return false; for (pc = 0; pc < fprog->len; pc++) { struct sock_filter *insn = &fprog->filter[pc]; u16 code = insn->code; u32 k = insn->k; switch (code) { case BPF_LD | BPF_W | BPF_ABS: switch (k) { case offsetof(struct seccomp_data, nr): reg_value = sd->nr; break; case offsetof(struct seccomp_data, arch): reg_value = sd->arch; break; default: /* can't optimize (non-constant value load) */ return false; } break; case BPF_RET | BPF_K: /* reached return with constant values only, check allow */ return k == SECCOMP_RET_ALLOW; case BPF_JMP | BPF_JA: pc += insn->k; break; case BPF_JMP | BPF_JEQ | BPF_K: case BPF_JMP | BPF_JGE | BPF_K: case BPF_JMP | BPF_JGT | BPF_K: case BPF_JMP | BPF_JSET | BPF_K: switch (BPF_OP(code)) { case BPF_JEQ: op_res = reg_value == k; break; case BPF_JGE: op_res = reg_value >= k; break; case BPF_JGT: op_res = reg_value > k; break; case BPF_JSET: op_res = !!(reg_value & k); break; default: /* can't optimize (unknown jump) */ return false; } pc += op_res ? insn->jt : insn->jf; break; case BPF_ALU | BPF_AND | BPF_K: reg_value &= k; break; default: /* can't optimize (unknown insn) */ return false; } } /* ran off the end of the filter?! */ WARN_ON(1); return false; } static void seccomp_cache_prepare_bitmap(struct seccomp_filter *sfilter, void *bitmap, const void *bitmap_prev, size_t bitmap_size, int arch) { struct sock_fprog_kern *fprog = sfilter->prog->orig_prog; struct seccomp_data sd; int nr; if (bitmap_prev) { /* The new filter must be as restrictive as the last. */ bitmap_copy(bitmap, bitmap_prev, bitmap_size); } else { /* Before any filters, all syscalls are always allowed. */ bitmap_fill(bitmap, bitmap_size); } for (nr = 0; nr < bitmap_size; nr++) { /* No bitmap change: not a cacheable action. */ if (!test_bit(nr, bitmap)) continue; sd.nr = nr; sd.arch = arch; /* No bitmap change: continue to always allow. */ if (seccomp_is_const_allow(fprog, &sd)) continue; /* * Not a cacheable action: always run filters. * atomic clear_bit() not needed, filter not visible yet. */ __clear_bit(nr, bitmap); } } /** * seccomp_cache_prepare - emulate the filter to find cachable syscalls * @sfilter: The seccomp filter * * Returns 0 if successful or -errno if error occurred. */ static void seccomp_cache_prepare(struct seccomp_filter *sfilter) { struct action_cache *cache = &sfilter->cache; const struct action_cache *cache_prev = sfilter->prev ? &sfilter->prev->cache : NULL; seccomp_cache_prepare_bitmap(sfilter, cache->allow_native, cache_prev ? cache_prev->allow_native : NULL, SECCOMP_ARCH_NATIVE_NR, SECCOMP_ARCH_NATIVE); #ifdef SECCOMP_ARCH_COMPAT seccomp_cache_prepare_bitmap(sfilter, cache->allow_compat, cache_prev ? cache_prev->allow_compat : NULL, SECCOMP_ARCH_COMPAT_NR, SECCOMP_ARCH_COMPAT); #endif /* SECCOMP_ARCH_COMPAT */ } #endif /* SECCOMP_ARCH_NATIVE */ |
c8bee430d seccomp: split fi... |
825 826 827 828 829 |
/** * seccomp_attach_filter: validate and attach filter * @flags: flags to change filter behavior * @filter: seccomp filter to add to the current process * |
dbd952127 seccomp: introduc... |
830 831 |
* Caller must be holding current->sighand->siglock lock. * |
7a0df7fbc seccomp: Make NEW... |
832 833 834 835 |
* Returns 0 on success, -ve on error, or * - in TSYNC mode: the pid of a thread which was either not in the correct * seccomp mode or did not have an ancestral seccomp filter * - in NEW_LISTENER mode: the fd of the new listener |
c8bee430d seccomp: split fi... |
836 837 838 839 840 841 |
*/ static long seccomp_attach_filter(unsigned int flags, struct seccomp_filter *filter) { unsigned long total_insns; struct seccomp_filter *walker; |
69f6a34bd seccomp: Replace ... |
842 |
assert_spin_locked(¤t->sighand->siglock); |
dbd952127 seccomp: introduc... |
843 |
|
c8bee430d seccomp: split fi... |
844 845 846 847 848 849 |
/* Validate resulting filter length. */ total_insns = filter->prog->len; for (walker = current->seccomp.filter; walker; walker = walker->prev) total_insns += walker->prog->len + 4; /* 4 instr penalty */ if (total_insns > MAX_INSNS_PER_PATH) return -ENOMEM; |
c2e1f2e30 seccomp: implemen... |
850 851 852 853 854 |
/* If thread sync has been requested, check that it is possible. */ if (flags & SECCOMP_FILTER_FLAG_TSYNC) { int ret; ret = seccomp_can_sync_threads(); |
51891498f seccomp: allow TS... |
855 856 857 858 859 860 |
if (ret) { if (flags & SECCOMP_FILTER_FLAG_TSYNC_ESRCH) return -ESRCH; else return ret; } |
c2e1f2e30 seccomp: implemen... |
861 |
} |
e66a39977 seccomp: Filter f... |
862 863 864 |
/* Set log flag, if present. */ if (flags & SECCOMP_FILTER_FLAG_LOG) filter->log = true; |
c8bee430d seccomp: split fi... |
865 866 867 868 869 |
/* * If there is an existing filter, make it the prev and don't drop its * task reference. */ filter->prev = current->seccomp.filter; |
2d660e977 UPSTREAM: seccomp... |
870 |
seccomp_cache_prepare(filter); |
c8bee430d seccomp: split fi... |
871 |
current->seccomp.filter = filter; |
c818c03b6 seccomp: Report n... |
872 |
atomic_inc(¤t->seccomp.filter_count); |
c8bee430d seccomp: split fi... |
873 |
|
c2e1f2e30 seccomp: implemen... |
874 875 |
/* Now that the new filter is in place, synchronize to all threads. */ if (flags & SECCOMP_FILTER_FLAG_TSYNC) |
00a02d0c5 seccomp: Add filt... |
876 |
seccomp_sync_threads(flags); |
c2e1f2e30 seccomp: implemen... |
877 |
|
c8bee430d seccomp: split fi... |
878 |
return 0; |
e2cfabdfd seccomp: add syst... |
879 |
} |
084f5601c seccomp: make fun... |
880 |
static void __get_seccomp_filter(struct seccomp_filter *filter) |
66a733ea6 seccomp: fix the ... |
881 |
{ |
b707ddee1 seccomp: rename "... |
882 |
refcount_inc(&filter->refs); |
66a733ea6 seccomp: fix the ... |
883 |
} |
e2cfabdfd seccomp: add syst... |
884 885 886 887 888 889 |
/* get_seccomp_filter - increments the reference count of the filter on @tsk */ void get_seccomp_filter(struct task_struct *tsk) { struct seccomp_filter *orig = tsk->seccomp.filter; if (!orig) return; |
66a733ea6 seccomp: fix the ... |
890 |
__get_seccomp_filter(orig); |
99cdb8b9a seccomp: notify a... |
891 |
refcount_inc(&orig->users); |
e2cfabdfd seccomp: add syst... |
892 |
} |
ae7795bc6 signal: Distingui... |
893 |
static void seccomp_init_siginfo(kernel_siginfo_t *info, int syscall, int reason) |
b25e67161 seccomp: dump cor... |
894 |
{ |
3b10db2b0 signal: Replace m... |
895 |
clear_siginfo(info); |
b25e67161 seccomp: dump cor... |
896 897 898 899 |
info->si_signo = SIGSYS; info->si_code = SYS_SECCOMP; info->si_call_addr = (void __user *)KSTK_EIP(current); info->si_errno = reason; |
16add4116 syscall_get_arch:... |
900 |
info->si_arch = syscall_get_arch(current); |
b25e67161 seccomp: dump cor... |
901 902 |
info->si_syscall = syscall; } |
bb6ea4301 seccomp: Add SECC... |
903 904 905 906 907 908 909 910 911 |
/** * seccomp_send_sigsys - signals the task to allow in-process syscall emulation * @syscall: syscall number to send to userland * @reason: filter-supplied reason code to send to userland (via si_errno) * * Forces a SIGSYS with a code of SYS_SECCOMP and related sigsys info. */ static void seccomp_send_sigsys(int syscall, int reason) { |
ae7795bc6 signal: Distingui... |
912 |
struct kernel_siginfo info; |
b25e67161 seccomp: dump cor... |
913 |
seccomp_init_siginfo(&info, syscall, reason); |
a89e9b8ab signal: Remove th... |
914 |
force_sig_info(&info); |
bb6ea4301 seccomp: Add SECC... |
915 |
} |
e2cfabdfd seccomp: add syst... |
916 |
#endif /* CONFIG_SECCOMP_FILTER */ |
1da177e4c Linux-2.6.12-rc2 |
917 |
|
0ddec0fc8 seccomp: Sysctl t... |
918 |
/* For use with seccomp_actions_logged */ |
4d3b0b05a seccomp: Introduc... |
919 920 |
#define SECCOMP_LOG_KILL_PROCESS (1 << 0) #define SECCOMP_LOG_KILL_THREAD (1 << 1) |
0ddec0fc8 seccomp: Sysctl t... |
921 922 923 |
#define SECCOMP_LOG_TRAP (1 << 2) #define SECCOMP_LOG_ERRNO (1 << 3) #define SECCOMP_LOG_TRACE (1 << 4) |
59f5cf44a seccomp: Action t... |
924 925 |
#define SECCOMP_LOG_LOG (1 << 5) #define SECCOMP_LOG_ALLOW (1 << 6) |
6a21cc50f seccomp: add a re... |
926 |
#define SECCOMP_LOG_USER_NOTIF (1 << 7) |
0ddec0fc8 seccomp: Sysctl t... |
927 |
|
4d3b0b05a seccomp: Introduc... |
928 929 |
static u32 seccomp_actions_logged = SECCOMP_LOG_KILL_PROCESS | SECCOMP_LOG_KILL_THREAD | |
fd76875ca seccomp: Rename S... |
930 931 |
SECCOMP_LOG_TRAP | SECCOMP_LOG_ERRNO | |
6a21cc50f seccomp: add a re... |
932 |
SECCOMP_LOG_USER_NOTIF | |
fd76875ca seccomp: Rename S... |
933 |
SECCOMP_LOG_TRACE | |
59f5cf44a seccomp: Action t... |
934 |
SECCOMP_LOG_LOG; |
0ddec0fc8 seccomp: Sysctl t... |
935 |
|
e66a39977 seccomp: Filter f... |
936 937 |
static inline void seccomp_log(unsigned long syscall, long signr, u32 action, bool requested) |
0ddec0fc8 seccomp: Sysctl t... |
938 939 940 941 942 |
{ bool log = false; switch (action) { case SECCOMP_RET_ALLOW: |
e66a39977 seccomp: Filter f... |
943 |
break; |
0ddec0fc8 seccomp: Sysctl t... |
944 |
case SECCOMP_RET_TRAP: |
e66a39977 seccomp: Filter f... |
945 946 |
log = requested && seccomp_actions_logged & SECCOMP_LOG_TRAP; break; |
0ddec0fc8 seccomp: Sysctl t... |
947 |
case SECCOMP_RET_ERRNO: |
e66a39977 seccomp: Filter f... |
948 949 |
log = requested && seccomp_actions_logged & SECCOMP_LOG_ERRNO; break; |
0ddec0fc8 seccomp: Sysctl t... |
950 |
case SECCOMP_RET_TRACE: |
e66a39977 seccomp: Filter f... |
951 |
log = requested && seccomp_actions_logged & SECCOMP_LOG_TRACE; |
0ddec0fc8 seccomp: Sysctl t... |
952 |
break; |
6a21cc50f seccomp: add a re... |
953 954 955 |
case SECCOMP_RET_USER_NOTIF: log = requested && seccomp_actions_logged & SECCOMP_LOG_USER_NOTIF; break; |
59f5cf44a seccomp: Action t... |
956 957 958 |
case SECCOMP_RET_LOG: log = seccomp_actions_logged & SECCOMP_LOG_LOG; break; |
fd76875ca seccomp: Rename S... |
959 |
case SECCOMP_RET_KILL_THREAD: |
fd76875ca seccomp: Rename S... |
960 |
log = seccomp_actions_logged & SECCOMP_LOG_KILL_THREAD; |
4d3b0b05a seccomp: Introduc... |
961 962 963 964 |
break; case SECCOMP_RET_KILL_PROCESS: default: log = seccomp_actions_logged & SECCOMP_LOG_KILL_PROCESS; |
0ddec0fc8 seccomp: Sysctl t... |
965 966 967 |
} /* |
326bee028 seccomp: Don't sp... |
968 969 970 971 |
* Emit an audit message when the action is RET_KILL_*, RET_LOG, or the * FILTER_FLAG_LOG bit was set. The admin has the ability to silence * any action from being logged by removing the action name from the * seccomp_actions_logged sysctl. |
0ddec0fc8 seccomp: Sysctl t... |
972 |
*/ |
326bee028 seccomp: Don't sp... |
973 974 |
if (!log) return; |
0ddec0fc8 seccomp: Sysctl t... |
975 |
|
326bee028 seccomp: Don't sp... |
976 |
audit_seccomp(syscall, signr, action); |
0ddec0fc8 seccomp: Sysctl t... |
977 |
} |
1da177e4c Linux-2.6.12-rc2 |
978 979 980 981 982 |
/* * Secure computing mode 1 allows only read/write/exit/sigreturn. * To be fully secure this must be combined with rlimit * to limit the stack allocations too. */ |
cb4253aa0 secomp: Constify ... |
983 |
static const int mode1_syscalls[] = { |
1da177e4c Linux-2.6.12-rc2 |
984 |
__NR_seccomp_read, __NR_seccomp_write, __NR_seccomp_exit, __NR_seccomp_sigreturn, |
fe4bfff86 seccomp: Use -1 m... |
985 |
-1, /* negative terminated */ |
1da177e4c Linux-2.6.12-rc2 |
986 |
}; |
a4412fc94 seccomp,x86,arm,m... |
987 |
static void __secure_computing_strict(int this_syscall) |
1da177e4c Linux-2.6.12-rc2 |
988 |
{ |
fe4bfff86 seccomp: Use -1 m... |
989 |
const int *allowed_syscalls = mode1_syscalls; |
a4412fc94 seccomp,x86,arm,m... |
990 |
#ifdef CONFIG_COMPAT |
5c38065e0 seccomp: check in... |
991 |
if (in_compat_syscall()) |
fe4bfff86 seccomp: Use -1 m... |
992 |
allowed_syscalls = get_compat_mode1_syscalls(); |
a4412fc94 seccomp,x86,arm,m... |
993 994 |
#endif do { |
fe4bfff86 seccomp: Use -1 m... |
995 |
if (*allowed_syscalls == this_syscall) |
a4412fc94 seccomp,x86,arm,m... |
996 |
return; |
fe4bfff86 seccomp: Use -1 m... |
997 |
} while (*++allowed_syscalls != -1); |
a4412fc94 seccomp,x86,arm,m... |
998 999 1000 1001 |
#ifdef SECCOMP_DEBUG dump_stack(); #endif |
fd76875ca seccomp: Rename S... |
1002 |
seccomp_log(this_syscall, SIGKILL, SECCOMP_RET_KILL_THREAD, true); |
a4412fc94 seccomp,x86,arm,m... |
1003 1004 1005 1006 1007 1008 1009 |
do_exit(SIGKILL); } #ifndef CONFIG_HAVE_ARCH_SECCOMP_FILTER void secure_computing_strict(int this_syscall) { int mode = current->seccomp.mode; |
97f2645f3 tree-wide: replac... |
1010 |
if (IS_ENABLED(CONFIG_CHECKPOINT_RESTORE) && |
13c4a9011 seccomp: add ptra... |
1011 1012 |
unlikely(current->ptrace & PT_SUSPEND_SECCOMP)) return; |
221272f97 seccomp: swap har... |
1013 |
if (mode == SECCOMP_MODE_DISABLED) |
a4412fc94 seccomp,x86,arm,m... |
1014 1015 1016 1017 1018 1019 1020 |
return; else if (mode == SECCOMP_MODE_STRICT) __secure_computing_strict(this_syscall); else BUG(); } #else |
13aa72f0f seccomp: Refactor... |
1021 1022 |
#ifdef CONFIG_SECCOMP_FILTER |
6a21cc50f seccomp: add a re... |
1023 1024 1025 1026 1027 1028 1029 1030 1031 |
static u64 seccomp_next_notify_id(struct seccomp_filter *filter) { /* * Note: overflow is ok here, the id just needs to be unique per * filter. */ lockdep_assert_held(&filter->notify_lock); return filter->notif->next_id++; } |
7cf97b125 seccomp: Introduc... |
1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 |
static void seccomp_handle_addfd(struct seccomp_kaddfd *addfd) { /* * Remove the notification, and reset the list pointers, indicating * that it has been handled. */ list_del_init(&addfd->list); addfd->ret = receive_fd_replace(addfd->fd, addfd->file, addfd->flags); complete(&addfd->completion); } |
fb3c5386b seccomp: add SECC... |
1042 1043 1044 |
static int seccomp_do_user_notification(int this_syscall, struct seccomp_filter *match, const struct seccomp_data *sd) |
6a21cc50f seccomp: add a re... |
1045 1046 |
{ int err; |
fb3c5386b seccomp: add SECC... |
1047 |
u32 flags = 0; |
6a21cc50f seccomp: add a re... |
1048 1049 |
long ret = 0; struct seccomp_knotif n = {}; |
7cf97b125 seccomp: Introduc... |
1050 |
struct seccomp_kaddfd *addfd, *tmp; |
6a21cc50f seccomp: add a re... |
1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 |
mutex_lock(&match->notify_lock); err = -ENOSYS; if (!match->notif) goto out; n.task = current; n.state = SECCOMP_NOTIFY_INIT; n.data = sd; n.id = seccomp_next_notify_id(match); init_completion(&n.ready); list_add(&n.list, &match->notif->notifications); |
7cf97b125 seccomp: Introduc... |
1063 |
INIT_LIST_HEAD(&n.addfd); |
6a21cc50f seccomp: add a re... |
1064 1065 |
up(&match->notif->request); |
76194c4e8 seccomp: Lift wai... |
1066 |
wake_up_poll(&match->wqh, EPOLLIN | EPOLLRDNORM); |
6a21cc50f seccomp: add a re... |
1067 1068 1069 1070 1071 |
mutex_unlock(&match->notify_lock); /* * This is where we wait for a reply from userspace. */ |
7cf97b125 seccomp: Introduc... |
1072 |
wait: |
6a21cc50f seccomp: add a re... |
1073 1074 1075 |
err = wait_for_completion_interruptible(&n.ready); mutex_lock(&match->notify_lock); if (err == 0) { |
7cf97b125 seccomp: Introduc... |
1076 1077 1078 1079 1080 1081 1082 1083 |
/* Check if we were woken up by a addfd message */ addfd = list_first_entry_or_null(&n.addfd, struct seccomp_kaddfd, list); if (addfd && n.state != SECCOMP_NOTIFY_REPLIED) { seccomp_handle_addfd(addfd); mutex_unlock(&match->notify_lock); goto wait; } |
6a21cc50f seccomp: add a re... |
1084 1085 |
ret = n.val; err = n.error; |
fb3c5386b seccomp: add SECC... |
1086 |
flags = n.flags; |
6a21cc50f seccomp: add a re... |
1087 |
} |
7cf97b125 seccomp: Introduc... |
1088 1089 1090 1091 1092 1093 1094 |
/* If there were any pending addfd calls, clear them out */ list_for_each_entry_safe(addfd, tmp, &n.addfd, list) { /* The process went away before we got a chance to handle it */ addfd->ret = -ESRCH; list_del_init(&addfd->list); complete(&addfd->completion); } |
6a21cc50f seccomp: add a re... |
1095 1096 |
/* * Note that it's possible the listener died in between the time when |
7cf97b125 seccomp: Introduc... |
1097 |
* we were notified of a response (or a signal) and when we were able to |
6a21cc50f seccomp: add a re... |
1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 |
* re-acquire the lock, so only delete from the list if the * notification actually exists. * * Also note that this test is only valid because there's no way to * *reattach* to a notifier right now. If one is added, we'll need to * keep track of the notif itself and make sure they match here. */ if (match->notif) list_del(&n.list); out: mutex_unlock(&match->notify_lock); |
fb3c5386b seccomp: add SECC... |
1109 1110 1111 1112 |
/* Userspace requests to continue the syscall. */ if (flags & SECCOMP_USER_NOTIF_FLAG_CONTINUE) return 0; |
2d9ca267a seccomp: Use curr... |
1113 |
syscall_set_return_value(current, current_pt_regs(), |
6a21cc50f seccomp: add a re... |
1114 |
err, ret); |
fb3c5386b seccomp: add SECC... |
1115 |
return -1; |
6a21cc50f seccomp: add a re... |
1116 |
} |
ce6526e8a seccomp: recheck ... |
1117 1118 |
static int __seccomp_filter(int this_syscall, const struct seccomp_data *sd, const bool recheck_after_trace) |
13aa72f0f seccomp: Refactor... |
1119 1120 |
{ u32 filter_ret, action; |
deb4de8b3 seccomp: Provide ... |
1121 |
struct seccomp_filter *match = NULL; |
13aa72f0f seccomp: Refactor... |
1122 |
int data; |
db5113911 seccomp: hoist st... |
1123 |
struct seccomp_data sd_local; |
1da177e4c Linux-2.6.12-rc2 |
1124 |
|
3ba2530cc seccomp: allow mo... |
1125 1126 1127 1128 1129 |
/* * Make sure that any changes to mode from another thread have * been seen after TIF_SECCOMP was seen. */ rmb(); |
db5113911 seccomp: hoist st... |
1130 1131 1132 1133 |
if (!sd) { populate_seccomp_data(&sd_local); sd = &sd_local; } |
deb4de8b3 seccomp: Provide ... |
1134 |
filter_ret = seccomp_run_filters(sd, &match); |
13aa72f0f seccomp: Refactor... |
1135 |
data = filter_ret & SECCOMP_RET_DATA; |
0466bdb99 seccomp: Implemen... |
1136 |
action = filter_ret & SECCOMP_RET_ACTION_FULL; |
13aa72f0f seccomp: Refactor... |
1137 1138 1139 |
switch (action) { case SECCOMP_RET_ERRNO: |
580c57f10 seccomp: cap SECC... |
1140 1141 1142 |
/* Set low-order bits as an errno, capped at MAX_ERRNO. */ if (data > MAX_ERRNO) data = MAX_ERRNO; |
2d9ca267a seccomp: Use curr... |
1143 |
syscall_set_return_value(current, current_pt_regs(), |
13aa72f0f seccomp: Refactor... |
1144 1145 1146 1147 1148 |
-data, 0); goto skip; case SECCOMP_RET_TRAP: /* Show the handler the original registers. */ |
2d9ca267a seccomp: Use curr... |
1149 |
syscall_rollback(current, current_pt_regs()); |
13aa72f0f seccomp: Refactor... |
1150 1151 1152 1153 1154 |
/* Let the filter pass back 16 bits of data. */ seccomp_send_sigsys(this_syscall, data); goto skip; case SECCOMP_RET_TRACE: |
ce6526e8a seccomp: recheck ... |
1155 1156 1157 |
/* We've been put in this state by the ptracer already. */ if (recheck_after_trace) return 0; |
8112c4f14 seccomp: remove 2... |
1158 1159 1160 |
/* ENOSYS these calls if there is no tracer attached. */ if (!ptrace_event_enabled(current, PTRACE_EVENT_SECCOMP)) { syscall_set_return_value(current, |
2d9ca267a seccomp: Use curr... |
1161 |
current_pt_regs(), |
8112c4f14 seccomp: remove 2... |
1162 1163 1164 1165 1166 1167 1168 1169 |
-ENOSYS, 0); goto skip; } /* Allow the BPF to provide the event message */ ptrace_event(PTRACE_EVENT_SECCOMP, data); /* * The delivery of a fatal signal during event |
485a252a5 seccomp: Fix trac... |
1170 1171 1172 1173 1174 1175 1176 |
* notification may silently skip tracer notification, * which could leave us with a potentially unmodified * syscall that the tracer would have liked to have * changed. Since the process is about to die, we just * force the syscall to be skipped and let the signal * kill the process and correctly handle any tracer exit * notifications. |
8112c4f14 seccomp: remove 2... |
1177 1178 |
*/ if (fatal_signal_pending(current)) |
485a252a5 seccomp: Fix trac... |
1179 |
goto skip; |
8112c4f14 seccomp: remove 2... |
1180 |
/* Check if the tracer forced the syscall to be skipped. */ |
2d9ca267a seccomp: Use curr... |
1181 |
this_syscall = syscall_get_nr(current, current_pt_regs()); |
8112c4f14 seccomp: remove 2... |
1182 1183 |
if (this_syscall < 0) goto skip; |
ce6526e8a seccomp: recheck ... |
1184 1185 1186 1187 1188 1189 1190 1191 |
/* * Recheck the syscall, since it may have changed. This * intentionally uses a NULL struct seccomp_data to force * a reload of all registers. This does not goto skip since * a skip would have already been reported. */ if (__seccomp_filter(this_syscall, NULL, true)) return -1; |
8112c4f14 seccomp: remove 2... |
1192 |
return 0; |
13aa72f0f seccomp: Refactor... |
1193 |
|
6a21cc50f seccomp: add a re... |
1194 |
case SECCOMP_RET_USER_NOTIF: |
fb3c5386b seccomp: add SECC... |
1195 1196 1197 1198 |
if (seccomp_do_user_notification(this_syscall, match, sd)) goto skip; return 0; |
6a21cc50f seccomp: add a re... |
1199 |
|
59f5cf44a seccomp: Action t... |
1200 1201 1202 |
case SECCOMP_RET_LOG: seccomp_log(this_syscall, 0, action, true); return 0; |
13aa72f0f seccomp: Refactor... |
1203 |
case SECCOMP_RET_ALLOW: |
deb4de8b3 seccomp: Provide ... |
1204 1205 1206 1207 1208 |
/* * Note that the "match" filter will always be NULL for * this action since SECCOMP_RET_ALLOW is the starting * state in seccomp_run_filters(). */ |
8112c4f14 seccomp: remove 2... |
1209 |
return 0; |
13aa72f0f seccomp: Refactor... |
1210 |
|
fd76875ca seccomp: Rename S... |
1211 |
case SECCOMP_RET_KILL_THREAD: |
4d3b0b05a seccomp: Introduc... |
1212 |
case SECCOMP_RET_KILL_PROCESS: |
131b63515 seccomp: Clean up... |
1213 |
default: |
e66a39977 seccomp: Filter f... |
1214 |
seccomp_log(this_syscall, SIGSYS, action, true); |
d7276e321 seccomp: Only dum... |
1215 |
/* Dump core only if this is the last remaining thread. */ |
4d671d922 seccomp: kill pro... |
1216 |
if (action != SECCOMP_RET_KILL_THREAD || |
4d3b0b05a seccomp: Introduc... |
1217 |
get_nr_threads(current) == 1) { |
ae7795bc6 signal: Distingui... |
1218 |
kernel_siginfo_t info; |
131b63515 seccomp: Clean up... |
1219 |
|
d7276e321 seccomp: Only dum... |
1220 |
/* Show the original registers in the dump. */ |
2d9ca267a seccomp: Use curr... |
1221 |
syscall_rollback(current, current_pt_regs()); |
d7276e321 seccomp: Only dum... |
1222 1223 1224 1225 |
/* Trigger a manual coredump since do_exit skips it. */ seccomp_init_siginfo(&info, this_syscall, data); do_coredump(&info); } |
4d671d922 seccomp: kill pro... |
1226 |
if (action == SECCOMP_RET_KILL_THREAD) |
4d3b0b05a seccomp: Introduc... |
1227 |
do_exit(SIGSYS); |
4d671d922 seccomp: kill pro... |
1228 1229 |
else do_group_exit(SIGSYS); |
13aa72f0f seccomp: Refactor... |
1230 1231 1232 1233 1234 |
} unreachable(); skip: |
e66a39977 seccomp: Filter f... |
1235 |
seccomp_log(this_syscall, 0, action, match ? match->log : false); |
8112c4f14 seccomp: remove 2... |
1236 1237 1238 |
return -1; } #else |
ce6526e8a seccomp: recheck ... |
1239 1240 |
static int __seccomp_filter(int this_syscall, const struct seccomp_data *sd, const bool recheck_after_trace) |
8112c4f14 seccomp: remove 2... |
1241 1242 |
{ BUG(); |
13aa72f0f seccomp: Refactor... |
1243 |
} |
1da177e4c Linux-2.6.12-rc2 |
1244 |
#endif |
13aa72f0f seccomp: Refactor... |
1245 |
|
8112c4f14 seccomp: remove 2... |
1246 |
int __secure_computing(const struct seccomp_data *sd) |
13aa72f0f seccomp: Refactor... |
1247 1248 |
{ int mode = current->seccomp.mode; |
8112c4f14 seccomp: remove 2... |
1249 |
int this_syscall; |
13aa72f0f seccomp: Refactor... |
1250 |
|
97f2645f3 tree-wide: replac... |
1251 |
if (IS_ENABLED(CONFIG_CHECKPOINT_RESTORE) && |
13c4a9011 seccomp: add ptra... |
1252 |
unlikely(current->ptrace & PT_SUSPEND_SECCOMP)) |
8112c4f14 seccomp: remove 2... |
1253 1254 1255 |
return 0; this_syscall = sd ? sd->nr : |
2d9ca267a seccomp: Use curr... |
1256 |
syscall_get_nr(current, current_pt_regs()); |
13c4a9011 seccomp: add ptra... |
1257 |
|
13aa72f0f seccomp: Refactor... |
1258 |
switch (mode) { |
e2cfabdfd seccomp: add syst... |
1259 |
case SECCOMP_MODE_STRICT: |
13aa72f0f seccomp: Refactor... |
1260 |
__secure_computing_strict(this_syscall); /* may call do_exit */ |
8112c4f14 seccomp: remove 2... |
1261 |
return 0; |
13aa72f0f seccomp: Refactor... |
1262 |
case SECCOMP_MODE_FILTER: |
ce6526e8a seccomp: recheck ... |
1263 |
return __seccomp_filter(this_syscall, sd, false); |
1da177e4c Linux-2.6.12-rc2 |
1264 1265 1266 |
default: BUG(); } |
13aa72f0f seccomp: Refactor... |
1267 |
} |
a4412fc94 seccomp,x86,arm,m... |
1268 |
#endif /* CONFIG_HAVE_ARCH_SECCOMP_FILTER */ |
1d9d02fee move seccomp from... |
1269 1270 1271 1272 1273 |
long prctl_get_seccomp(void) { return current->seccomp.mode; } |
e2cfabdfd seccomp: add syst... |
1274 |
/** |
3b23dd128 seccomp: split mo... |
1275 |
* seccomp_set_mode_strict: internal function for setting strict seccomp |
e2cfabdfd seccomp: add syst... |
1276 1277 1278 1279 1280 |
* * Once current->seccomp.mode is non-zero, it may not be changed. * * Returns 0 on success or -EINVAL on failure. */ |
3b23dd128 seccomp: split mo... |
1281 |
static long seccomp_set_mode_strict(void) |
1d9d02fee move seccomp from... |
1282 |
{ |
3b23dd128 seccomp: split mo... |
1283 |
const unsigned long seccomp_mode = SECCOMP_MODE_STRICT; |
e2cfabdfd seccomp: add syst... |
1284 |
long ret = -EINVAL; |
1d9d02fee move seccomp from... |
1285 |
|
dbd952127 seccomp: introduc... |
1286 |
spin_lock_irq(¤t->sighand->siglock); |
1f41b4504 seccomp: extract ... |
1287 |
if (!seccomp_may_assign_mode(seccomp_mode)) |
1d9d02fee move seccomp from... |
1288 |
goto out; |
cf99abace make seccomp zero... |
1289 |
#ifdef TIF_NOTSC |
3b23dd128 seccomp: split mo... |
1290 |
disable_TSC(); |
cf99abace make seccomp zero... |
1291 |
#endif |
00a02d0c5 seccomp: Add filt... |
1292 |
seccomp_assign_mode(current, seccomp_mode, 0); |
3b23dd128 seccomp: split mo... |
1293 1294 1295 |
ret = 0; out: |
dbd952127 seccomp: introduc... |
1296 |
spin_unlock_irq(¤t->sighand->siglock); |
3b23dd128 seccomp: split mo... |
1297 1298 1299 |
return ret; } |
e2cfabdfd seccomp: add syst... |
1300 |
#ifdef CONFIG_SECCOMP_FILTER |
e83931790 seccomp: don't le... |
1301 1302 1303 1304 1305 |
static void seccomp_notify_free(struct seccomp_filter *filter) { kfree(filter->notif); filter->notif = NULL; } |
a566a9012 seccomp: don't le... |
1306 |
static void seccomp_notify_detach(struct seccomp_filter *filter) |
6a21cc50f seccomp: add a re... |
1307 |
{ |
6a21cc50f seccomp: add a re... |
1308 |
struct seccomp_knotif *knotif; |
a811dc615 seccomp: fix UAF ... |
1309 |
if (!filter) |
a566a9012 seccomp: don't le... |
1310 |
return; |
a811dc615 seccomp: fix UAF ... |
1311 |
|
6a21cc50f seccomp: add a re... |
1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 |
mutex_lock(&filter->notify_lock); /* * If this file is being closed because e.g. the task who owned it * died, let's wake everyone up who was waiting on us. */ list_for_each_entry(knotif, &filter->notif->notifications, list) { if (knotif->state == SECCOMP_NOTIFY_REPLIED) continue; knotif->state = SECCOMP_NOTIFY_REPLIED; knotif->error = -ENOSYS; knotif->val = 0; |
7cf97b125 seccomp: Introduc... |
1325 1326 1327 1328 1329 |
/* * We do not need to wake up any pending addfd messages, as * the notifier will do that for us, as this just looks * like a standard reply. */ |
6a21cc50f seccomp: add a re... |
1330 1331 |
complete(&knotif->ready); } |
e83931790 seccomp: don't le... |
1332 |
seccomp_notify_free(filter); |
6a21cc50f seccomp: add a re... |
1333 |
mutex_unlock(&filter->notify_lock); |
a566a9012 seccomp: don't le... |
1334 1335 1336 1337 1338 1339 1340 |
} static int seccomp_notify_release(struct inode *inode, struct file *file) { struct seccomp_filter *filter = file->private_data; seccomp_notify_detach(filter); |
6a21cc50f seccomp: add a re... |
1341 1342 1343 |
__put_seccomp_filter(filter); return 0; } |
9f87dcf14 seccomp: Add find... |
1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 |
/* must be called with notif_lock held */ static inline struct seccomp_knotif * find_notification(struct seccomp_filter *filter, u64 id) { struct seccomp_knotif *cur; lockdep_assert_held(&filter->notify_lock); list_for_each_entry(cur, &filter->notif->notifications, list) { if (cur->id == id) return cur; } return NULL; } |
6a21cc50f seccomp: add a re... |
1359 1360 1361 1362 1363 1364 |
static long seccomp_notify_recv(struct seccomp_filter *filter, void __user *buf) { struct seccomp_knotif *knotif = NULL, *cur; struct seccomp_notif unotif; ssize_t ret; |
2882d53c9 seccomp: Check th... |
1365 1366 1367 1368 1369 1370 |
/* Verify that we're not given garbage to keep struct extensible. */ ret = check_zeroed_user(buf, sizeof(unotif)); if (ret < 0) return ret; if (!ret) return -EINVAL; |
6a21cc50f seccomp: add a re... |
1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 |
memset(&unotif, 0, sizeof(unotif)); ret = down_interruptible(&filter->notif->request); if (ret < 0) return ret; mutex_lock(&filter->notify_lock); list_for_each_entry(cur, &filter->notif->notifications, list) { if (cur->state == SECCOMP_NOTIFY_INIT) { knotif = cur; break; } } /* * If we didn't find a notification, it could be that the task was * interrupted by a fatal signal between the time we were woken and * when we were able to acquire the rw lock. */ if (!knotif) { ret = -ENOENT; goto out; } unotif.id = knotif->id; unotif.pid = task_pid_vnr(knotif->task); unotif.data = *(knotif->data); knotif->state = SECCOMP_NOTIFY_SENT; |
76194c4e8 seccomp: Lift wai... |
1400 |
wake_up_poll(&filter->wqh, EPOLLOUT | EPOLLWRNORM); |
6a21cc50f seccomp: add a re... |
1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 |
ret = 0; out: mutex_unlock(&filter->notify_lock); if (ret == 0 && copy_to_user(buf, &unotif, sizeof(unotif))) { ret = -EFAULT; /* * Userspace screwed up. To make sure that we keep this * notification alive, let's reset it back to INIT. It * may have died when we released the lock, so we need to make * sure it's still around. */ |
6a21cc50f seccomp: add a re... |
1414 |
mutex_lock(&filter->notify_lock); |
9f87dcf14 seccomp: Add find... |
1415 |
knotif = find_notification(filter, unotif.id); |
6a21cc50f seccomp: add a re... |
1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 |
if (knotif) { knotif->state = SECCOMP_NOTIFY_INIT; up(&filter->notif->request); } mutex_unlock(&filter->notify_lock); } return ret; } static long seccomp_notify_send(struct seccomp_filter *filter, void __user *buf) { struct seccomp_notif_resp resp = {}; |
9f87dcf14 seccomp: Add find... |
1430 |
struct seccomp_knotif *knotif; |
6a21cc50f seccomp: add a re... |
1431 1432 1433 1434 |
long ret; if (copy_from_user(&resp, buf, sizeof(resp))) return -EFAULT; |
fb3c5386b seccomp: add SECC... |
1435 1436 1437 1438 1439 |
if (resp.flags & ~SECCOMP_USER_NOTIF_FLAG_CONTINUE) return -EINVAL; if ((resp.flags & SECCOMP_USER_NOTIF_FLAG_CONTINUE) && (resp.error || resp.val)) |
6a21cc50f seccomp: add a re... |
1440 1441 1442 1443 1444 |
return -EINVAL; ret = mutex_lock_interruptible(&filter->notify_lock); if (ret < 0) return ret; |
9f87dcf14 seccomp: Add find... |
1445 |
knotif = find_notification(filter, resp.id); |
6a21cc50f seccomp: add a re... |
1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 |
if (!knotif) { ret = -ENOENT; goto out; } /* Allow exactly one reply. */ if (knotif->state != SECCOMP_NOTIFY_SENT) { ret = -EINPROGRESS; goto out; } ret = 0; knotif->state = SECCOMP_NOTIFY_REPLIED; knotif->error = resp.error; knotif->val = resp.val; |
fb3c5386b seccomp: add SECC... |
1461 |
knotif->flags = resp.flags; |
6a21cc50f seccomp: add a re... |
1462 1463 1464 1465 1466 1467 1468 1469 1470 |
complete(&knotif->ready); out: mutex_unlock(&filter->notify_lock); return ret; } static long seccomp_notify_id_valid(struct seccomp_filter *filter, void __user *buf) { |
9f87dcf14 seccomp: Add find... |
1471 |
struct seccomp_knotif *knotif; |
6a21cc50f seccomp: add a re... |
1472 1473 1474 1475 1476 1477 1478 1479 1480 |
u64 id; long ret; if (copy_from_user(&id, buf, sizeof(id))) return -EFAULT; ret = mutex_lock_interruptible(&filter->notify_lock); if (ret < 0) return ret; |
9f87dcf14 seccomp: Add find... |
1481 1482 1483 1484 1485 |
knotif = find_notification(filter, id); if (knotif && knotif->state == SECCOMP_NOTIFY_SENT) ret = 0; else ret = -ENOENT; |
6a21cc50f seccomp: add a re... |
1486 |
|
6a21cc50f seccomp: add a re... |
1487 1488 1489 |
mutex_unlock(&filter->notify_lock); return ret; } |
7cf97b125 seccomp: Introduc... |
1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 |
static long seccomp_notify_addfd(struct seccomp_filter *filter, struct seccomp_notif_addfd __user *uaddfd, unsigned int size) { struct seccomp_notif_addfd addfd; struct seccomp_knotif *knotif; struct seccomp_kaddfd kaddfd; int ret; BUILD_BUG_ON(sizeof(addfd) < SECCOMP_NOTIFY_ADDFD_SIZE_VER0); BUILD_BUG_ON(sizeof(addfd) != SECCOMP_NOTIFY_ADDFD_SIZE_LATEST); if (size < SECCOMP_NOTIFY_ADDFD_SIZE_VER0 || size >= PAGE_SIZE) return -EINVAL; ret = copy_struct_from_user(&addfd, sizeof(addfd), uaddfd, size); if (ret) return ret; if (addfd.newfd_flags & ~O_CLOEXEC) return -EINVAL; if (addfd.flags & ~SECCOMP_ADDFD_FLAG_SETFD) return -EINVAL; if (addfd.newfd && !(addfd.flags & SECCOMP_ADDFD_FLAG_SETFD)) return -EINVAL; kaddfd.file = fget(addfd.srcfd); if (!kaddfd.file) return -EBADF; kaddfd.flags = addfd.newfd_flags; kaddfd.fd = (addfd.flags & SECCOMP_ADDFD_FLAG_SETFD) ? addfd.newfd : -1; init_completion(&kaddfd.completion); ret = mutex_lock_interruptible(&filter->notify_lock); if (ret < 0) goto out; knotif = find_notification(filter, addfd.id); if (!knotif) { ret = -ENOENT; goto out_unlock; } /* * We do not want to allow for FD injection to occur before the * notification has been picked up by a userspace handler, or after * the notification has been replied to. */ if (knotif->state != SECCOMP_NOTIFY_SENT) { ret = -EINPROGRESS; goto out_unlock; } list_add(&kaddfd.list, &knotif->addfd); complete(&knotif->ready); mutex_unlock(&filter->notify_lock); /* Now we wait for it to be processed or be interrupted */ ret = wait_for_completion_interruptible(&kaddfd.completion); if (ret == 0) { /* * We had a successful completion. The other side has already * removed us from the addfd queue, and * wait_for_completion_interruptible has a memory barrier upon * success that lets us read this value directly without * locking. */ ret = kaddfd.ret; goto out; } mutex_lock(&filter->notify_lock); /* * Even though we were woken up by a signal and not a successful * completion, a completion may have happened in the mean time. * * We need to check again if the addfd request has been handled, * and if not, we will remove it from the queue. */ if (list_empty(&kaddfd.list)) ret = kaddfd.ret; else list_del(&kaddfd.list); out_unlock: mutex_unlock(&filter->notify_lock); out: fput(kaddfd.file); return ret; } |
6a21cc50f seccomp: add a re... |
1585 1586 1587 1588 1589 |
static long seccomp_notify_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { struct seccomp_filter *filter = file->private_data; void __user *buf = (void __user *)arg; |
7cf97b125 seccomp: Introduc... |
1590 |
/* Fixed-size ioctls */ |
6a21cc50f seccomp: add a re... |
1591 1592 1593 1594 1595 |
switch (cmd) { case SECCOMP_IOCTL_NOTIF_RECV: return seccomp_notify_recv(filter, buf); case SECCOMP_IOCTL_NOTIF_SEND: return seccomp_notify_send(filter, buf); |
47e33c05f seccomp: Fix ioct... |
1596 |
case SECCOMP_IOCTL_NOTIF_ID_VALID_WRONG_DIR: |
6a21cc50f seccomp: add a re... |
1597 1598 |
case SECCOMP_IOCTL_NOTIF_ID_VALID: return seccomp_notify_id_valid(filter, buf); |
7cf97b125 seccomp: Introduc... |
1599 1600 1601 1602 1603 1604 1605 |
} /* Extensible Argument ioctls */ #define EA_IOCTL(cmd) ((cmd) & ~(IOC_INOUT | IOCSIZE_MASK)) switch (EA_IOCTL(cmd)) { case EA_IOCTL(SECCOMP_IOCTL_NOTIF_ADDFD): return seccomp_notify_addfd(filter, buf, _IOC_SIZE(cmd)); |
6a21cc50f seccomp: add a re... |
1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 |
default: return -EINVAL; } } static __poll_t seccomp_notify_poll(struct file *file, struct poll_table_struct *poll_tab) { struct seccomp_filter *filter = file->private_data; __poll_t ret = 0; struct seccomp_knotif *cur; |
76194c4e8 seccomp: Lift wai... |
1617 |
poll_wait(file, &filter->wqh, poll_tab); |
6a21cc50f seccomp: add a re... |
1618 |
|
319deec7d seccomp: fix poor... |
1619 |
if (mutex_lock_interruptible(&filter->notify_lock) < 0) |
6a21cc50f seccomp: add a re... |
1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 |
return EPOLLERR; list_for_each_entry(cur, &filter->notif->notifications, list) { if (cur->state == SECCOMP_NOTIFY_INIT) ret |= EPOLLIN | EPOLLRDNORM; if (cur->state == SECCOMP_NOTIFY_SENT) ret |= EPOLLOUT | EPOLLWRNORM; if ((ret & EPOLLIN) && (ret & EPOLLOUT)) break; } mutex_unlock(&filter->notify_lock); |
99cdb8b9a seccomp: notify a... |
1632 1633 |
if (refcount_read(&filter->users) == 0) ret |= EPOLLHUP; |
6a21cc50f seccomp: add a re... |
1634 1635 1636 1637 1638 1639 1640 |
return ret; } static const struct file_operations seccomp_notify_ops = { .poll = seccomp_notify_poll, .release = seccomp_notify_release, .unlocked_ioctl = seccomp_notify_ioctl, |
3db81afd9 seccomp: Add miss... |
1641 |
.compat_ioctl = seccomp_notify_ioctl, |
6a21cc50f seccomp: add a re... |
1642 1643 1644 1645 |
}; static struct file *init_listener(struct seccomp_filter *filter) { |
dfe719fef seccomp: Make dup... |
1646 |
struct file *ret; |
6a21cc50f seccomp: add a re... |
1647 1648 1649 1650 1651 1652 1653 1654 1655 |
ret = ERR_PTR(-ENOMEM); filter->notif = kzalloc(sizeof(*(filter->notif)), GFP_KERNEL); if (!filter->notif) goto out; sema_init(&filter->notif->request, 0); filter->notif->next_id = get_random_u64(); INIT_LIST_HEAD(&filter->notif->notifications); |
6a21cc50f seccomp: add a re... |
1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 |
ret = anon_inode_getfile("seccomp notify", &seccomp_notify_ops, filter, O_RDWR); if (IS_ERR(ret)) goto out_notif; /* The file has a reference to it now */ __get_seccomp_filter(filter); out_notif: if (IS_ERR(ret)) |
e83931790 seccomp: don't le... |
1667 |
seccomp_notify_free(filter); |
6a21cc50f seccomp: add a re... |
1668 1669 1670 |
out: return ret; } |
dfe719fef seccomp: Make dup... |
1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 |
/* * Does @new_child have a listener while an ancestor also has a listener? * If so, we'll want to reject this filter. * This only has to be tested for the current process, even in the TSYNC case, * because TSYNC installs @child with the same parent on all threads. * Note that @new_child is not hooked up to its parent at this point yet, so * we use current->seccomp.filter. */ static bool has_duplicate_listener(struct seccomp_filter *new_child) { struct seccomp_filter *cur; /* must be protected against concurrent TSYNC */ lockdep_assert_held(¤t->sighand->siglock); if (!new_child->notif) return false; for (cur = current->seccomp.filter; cur; cur = cur->prev) { if (cur->notif) return true; } return false; } |
3b23dd128 seccomp: split mo... |
1695 1696 |
/** * seccomp_set_mode_filter: internal function for setting seccomp filter |
48dc92b9f seccomp: add "sec... |
1697 |
* @flags: flags to change filter behavior |
3b23dd128 seccomp: split mo... |
1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 |
* @filter: struct sock_fprog containing filter * * This function may be called repeatedly to install additional filters. * Every filter successfully installed will be evaluated (in reverse order) * for each system call the task makes. * * Once current->seccomp.mode is non-zero, it may not be changed. * * Returns 0 on success or -EINVAL on failure. */ |
48dc92b9f seccomp: add "sec... |
1708 1709 |
static long seccomp_set_mode_filter(unsigned int flags, const char __user *filter) |
3b23dd128 seccomp: split mo... |
1710 1711 |
{ const unsigned long seccomp_mode = SECCOMP_MODE_FILTER; |
c8bee430d seccomp: split fi... |
1712 |
struct seccomp_filter *prepared = NULL; |
3b23dd128 seccomp: split mo... |
1713 |
long ret = -EINVAL; |
6a21cc50f seccomp: add a re... |
1714 1715 |
int listener = -1; struct file *listener_f = NULL; |
3b23dd128 seccomp: split mo... |
1716 |
|
48dc92b9f seccomp: add "sec... |
1717 |
/* Validate flags. */ |
c2e1f2e30 seccomp: implemen... |
1718 |
if (flags & ~SECCOMP_FILTER_FLAG_MASK) |
dbd952127 seccomp: introduc... |
1719 |
return -EINVAL; |
48dc92b9f seccomp: add "sec... |
1720 |
|
7a0df7fbc seccomp: Make NEW... |
1721 1722 1723 1724 |
/* * In the successful case, NEW_LISTENER returns the new listener fd. * But in the failure case, TSYNC returns the thread that died. If you * combine these two flags, there's no way to tell whether something |
51891498f seccomp: allow TS... |
1725 1726 |
* succeeded or failed. So, let's disallow this combination if the user * has not explicitly requested no errors from TSYNC. |
7a0df7fbc seccomp: Make NEW... |
1727 1728 |
*/ if ((flags & SECCOMP_FILTER_FLAG_TSYNC) && |
51891498f seccomp: allow TS... |
1729 1730 |
(flags & SECCOMP_FILTER_FLAG_NEW_LISTENER) && ((flags & SECCOMP_FILTER_FLAG_TSYNC_ESRCH) == 0)) |
7a0df7fbc seccomp: Make NEW... |
1731 |
return -EINVAL; |
c8bee430d seccomp: split fi... |
1732 1733 1734 1735 |
/* Prepare the new filter before holding any locks. */ prepared = seccomp_prepare_user_filter(filter); if (IS_ERR(prepared)) return PTR_ERR(prepared); |
6a21cc50f seccomp: add a re... |
1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 |
if (flags & SECCOMP_FILTER_FLAG_NEW_LISTENER) { listener = get_unused_fd_flags(O_CLOEXEC); if (listener < 0) { ret = listener; goto out_free; } listener_f = init_listener(prepared); if (IS_ERR(listener_f)) { put_unused_fd(listener); ret = PTR_ERR(listener_f); goto out_free; } } |
c2e1f2e30 seccomp: implemen... |
1750 1751 1752 1753 1754 1755 |
/* * Make sure we cannot change seccomp or nnp state via TSYNC * while another thread is in the middle of calling exec. */ if (flags & SECCOMP_FILTER_FLAG_TSYNC && mutex_lock_killable(¤t->signal->cred_guard_mutex)) |
6a21cc50f seccomp: add a re... |
1756 |
goto out_put_fd; |
c2e1f2e30 seccomp: implemen... |
1757 |
|
dbd952127 seccomp: introduc... |
1758 |
spin_lock_irq(¤t->sighand->siglock); |
3b23dd128 seccomp: split mo... |
1759 1760 |
if (!seccomp_may_assign_mode(seccomp_mode)) goto out; |
dfe719fef seccomp: Make dup... |
1761 1762 1763 1764 |
if (has_duplicate_listener(prepared)) { ret = -EBUSY; goto out; } |
c8bee430d seccomp: split fi... |
1765 |
ret = seccomp_attach_filter(flags, prepared); |
3b23dd128 seccomp: split mo... |
1766 |
if (ret) |
e2cfabdfd seccomp: add syst... |
1767 |
goto out; |
c8bee430d seccomp: split fi... |
1768 1769 |
/* Do not free the successfully attached filter. */ prepared = NULL; |
1d9d02fee move seccomp from... |
1770 |
|
00a02d0c5 seccomp: Add filt... |
1771 |
seccomp_assign_mode(current, seccomp_mode, flags); |
e2cfabdfd seccomp: add syst... |
1772 |
out: |
dbd952127 seccomp: introduc... |
1773 |
spin_unlock_irq(¤t->sighand->siglock); |
c2e1f2e30 seccomp: implemen... |
1774 1775 |
if (flags & SECCOMP_FILTER_FLAG_TSYNC) mutex_unlock(¤t->signal->cred_guard_mutex); |
6a21cc50f seccomp: add a re... |
1776 1777 |
out_put_fd: if (flags & SECCOMP_FILTER_FLAG_NEW_LISTENER) { |
7a0df7fbc seccomp: Make NEW... |
1778 |
if (ret) { |
a811dc615 seccomp: fix UAF ... |
1779 |
listener_f->private_data = NULL; |
6a21cc50f seccomp: add a re... |
1780 1781 |
fput(listener_f); put_unused_fd(listener); |
a566a9012 seccomp: don't le... |
1782 |
seccomp_notify_detach(prepared); |
6a21cc50f seccomp: add a re... |
1783 1784 1785 1786 1787 |
} else { fd_install(listener, listener_f); ret = listener; } } |
c2e1f2e30 seccomp: implemen... |
1788 |
out_free: |
c8bee430d seccomp: split fi... |
1789 |
seccomp_filter_free(prepared); |
1d9d02fee move seccomp from... |
1790 1791 |
return ret; } |
3b23dd128 seccomp: split mo... |
1792 |
#else |
48dc92b9f seccomp: add "sec... |
1793 1794 |
static inline long seccomp_set_mode_filter(unsigned int flags, const char __user *filter) |
3b23dd128 seccomp: split mo... |
1795 1796 1797 1798 |
{ return -EINVAL; } #endif |
d78ab02c2 seccomp: create i... |
1799 |
|
d612b1fd8 seccomp: Operatio... |
1800 1801 1802 1803 1804 1805 1806 1807 |
static long seccomp_get_action_avail(const char __user *uaction) { u32 action; if (copy_from_user(&action, uaction, sizeof(action))) return -EFAULT; switch (action) { |
0466bdb99 seccomp: Implemen... |
1808 |
case SECCOMP_RET_KILL_PROCESS: |
fd76875ca seccomp: Rename S... |
1809 |
case SECCOMP_RET_KILL_THREAD: |
d612b1fd8 seccomp: Operatio... |
1810 1811 |
case SECCOMP_RET_TRAP: case SECCOMP_RET_ERRNO: |
6a21cc50f seccomp: add a re... |
1812 |
case SECCOMP_RET_USER_NOTIF: |
d612b1fd8 seccomp: Operatio... |
1813 |
case SECCOMP_RET_TRACE: |
59f5cf44a seccomp: Action t... |
1814 |
case SECCOMP_RET_LOG: |
d612b1fd8 seccomp: Operatio... |
1815 1816 1817 1818 1819 1820 1821 1822 |
case SECCOMP_RET_ALLOW: break; default: return -EOPNOTSUPP; } return 0; } |
6a21cc50f seccomp: add a re... |
1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 |
static long seccomp_get_notif_sizes(void __user *usizes) { struct seccomp_notif_sizes sizes = { .seccomp_notif = sizeof(struct seccomp_notif), .seccomp_notif_resp = sizeof(struct seccomp_notif_resp), .seccomp_data = sizeof(struct seccomp_data), }; if (copy_to_user(usizes, &sizes, sizeof(sizes))) return -EFAULT; return 0; } |
48dc92b9f seccomp: add "sec... |
1836 1837 |
/* Common entry point for both prctl and syscall. */ static long do_seccomp(unsigned int op, unsigned int flags, |
a5662e4d8 seccomp: switch s... |
1838 |
void __user *uargs) |
48dc92b9f seccomp: add "sec... |
1839 1840 1841 1842 1843 1844 1845 1846 |
{ switch (op) { case SECCOMP_SET_MODE_STRICT: if (flags != 0 || uargs != NULL) return -EINVAL; return seccomp_set_mode_strict(); case SECCOMP_SET_MODE_FILTER: return seccomp_set_mode_filter(flags, uargs); |
d612b1fd8 seccomp: Operatio... |
1847 1848 1849 1850 1851 |
case SECCOMP_GET_ACTION_AVAIL: if (flags != 0) return -EINVAL; return seccomp_get_action_avail(uargs); |
6a21cc50f seccomp: add a re... |
1852 1853 1854 1855 1856 |
case SECCOMP_GET_NOTIF_SIZES: if (flags != 0) return -EINVAL; return seccomp_get_notif_sizes(uargs); |
48dc92b9f seccomp: add "sec... |
1857 1858 1859 1860 1861 1862 |
default: return -EINVAL; } } SYSCALL_DEFINE3(seccomp, unsigned int, op, unsigned int, flags, |
a5662e4d8 seccomp: switch s... |
1863 |
void __user *, uargs) |
48dc92b9f seccomp: add "sec... |
1864 1865 1866 |
{ return do_seccomp(op, flags, uargs); } |
d78ab02c2 seccomp: create i... |
1867 1868 1869 1870 1871 1872 1873 |
/** * prctl_set_seccomp: configures current->seccomp.mode * @seccomp_mode: requested mode to use * @filter: optional struct sock_fprog for use with SECCOMP_MODE_FILTER * * Returns 0 on success or -EINVAL on failure. */ |
a5662e4d8 seccomp: switch s... |
1874 |
long prctl_set_seccomp(unsigned long seccomp_mode, void __user *filter) |
d78ab02c2 seccomp: create i... |
1875 |
{ |
48dc92b9f seccomp: add "sec... |
1876 |
unsigned int op; |
a5662e4d8 seccomp: switch s... |
1877 |
void __user *uargs; |
48dc92b9f seccomp: add "sec... |
1878 |
|
3b23dd128 seccomp: split mo... |
1879 1880 |
switch (seccomp_mode) { case SECCOMP_MODE_STRICT: |
48dc92b9f seccomp: add "sec... |
1881 1882 1883 1884 1885 1886 1887 1888 |
op = SECCOMP_SET_MODE_STRICT; /* * Setting strict mode through prctl always ignored filter, * so make sure it is always NULL here to pass the internal * check in do_seccomp(). */ uargs = NULL; break; |
3b23dd128 seccomp: split mo... |
1889 |
case SECCOMP_MODE_FILTER: |
48dc92b9f seccomp: add "sec... |
1890 1891 1892 |
op = SECCOMP_SET_MODE_FILTER; uargs = filter; break; |
3b23dd128 seccomp: split mo... |
1893 1894 1895 |
default: return -EINVAL; } |
48dc92b9f seccomp: add "sec... |
1896 1897 1898 |
/* prctl interface doesn't have flags, so they are always zero. */ return do_seccomp(op, 0, uargs); |
d78ab02c2 seccomp: create i... |
1899 |
} |
f8e529ed9 seccomp, ptrace: ... |
1900 1901 |
#if defined(CONFIG_SECCOMP_FILTER) && defined(CONFIG_CHECKPOINT_RESTORE) |
f06eae831 seccomp: hoist ou... |
1902 1903 |
static struct seccomp_filter *get_nth_filter(struct task_struct *task, unsigned long filter_off) |
f8e529ed9 seccomp, ptrace: ... |
1904 |
{ |
f06eae831 seccomp: hoist ou... |
1905 1906 |
struct seccomp_filter *orig, *filter; unsigned long count; |
f8e529ed9 seccomp, ptrace: ... |
1907 |
|
f06eae831 seccomp: hoist ou... |
1908 1909 1910 1911 |
/* * Note: this is only correct because the caller should be the (ptrace) * tracer of the task, otherwise lock_task_sighand is needed. */ |
f8e529ed9 seccomp, ptrace: ... |
1912 |
spin_lock_irq(&task->sighand->siglock); |
f06eae831 seccomp: hoist ou... |
1913 |
|
f8e529ed9 seccomp, ptrace: ... |
1914 |
if (task->seccomp.mode != SECCOMP_MODE_FILTER) { |
f06eae831 seccomp: hoist ou... |
1915 1916 |
spin_unlock_irq(&task->sighand->siglock); return ERR_PTR(-EINVAL); |
f8e529ed9 seccomp, ptrace: ... |
1917 |
} |
f06eae831 seccomp: hoist ou... |
1918 1919 1920 1921 1922 1923 |
orig = task->seccomp.filter; __get_seccomp_filter(orig); spin_unlock_irq(&task->sighand->siglock); count = 0; for (filter = orig; filter; filter = filter->prev) |
f8e529ed9 seccomp, ptrace: ... |
1924 |
count++; |
f8e529ed9 seccomp, ptrace: ... |
1925 1926 |
if (filter_off >= count) { |
f06eae831 seccomp: hoist ou... |
1927 |
filter = ERR_PTR(-ENOENT); |
f8e529ed9 seccomp, ptrace: ... |
1928 1929 |
goto out; } |
f8e529ed9 seccomp, ptrace: ... |
1930 |
|
f06eae831 seccomp: hoist ou... |
1931 1932 |
count -= filter_off; for (filter = orig; filter && count > 1; filter = filter->prev) |
f8e529ed9 seccomp, ptrace: ... |
1933 |
count--; |
f8e529ed9 seccomp, ptrace: ... |
1934 1935 |
if (WARN_ON(count != 1 || !filter)) { |
f06eae831 seccomp: hoist ou... |
1936 |
filter = ERR_PTR(-ENOENT); |
f8e529ed9 seccomp, ptrace: ... |
1937 1938 |
goto out; } |
f06eae831 seccomp: hoist ou... |
1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 |
__get_seccomp_filter(filter); out: __put_seccomp_filter(orig); return filter; } long seccomp_get_filter(struct task_struct *task, unsigned long filter_off, void __user *data) { struct seccomp_filter *filter; struct sock_fprog_kern *fprog; long ret; if (!capable(CAP_SYS_ADMIN) || current->seccomp.mode != SECCOMP_MODE_DISABLED) { return -EACCES; } filter = get_nth_filter(task, filter_off); if (IS_ERR(filter)) return PTR_ERR(filter); |
f8e529ed9 seccomp, ptrace: ... |
1961 1962 |
fprog = filter->prog->orig_prog; if (!fprog) { |
470bf1f27 seccomp: Fix comm... |
1963 |
/* This must be a new non-cBPF filter, since we save |
f8e529ed9 seccomp, ptrace: ... |
1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 |
* every cBPF filter's orig_prog above when * CONFIG_CHECKPOINT_RESTORE is enabled. */ ret = -EMEDIUMTYPE; goto out; } ret = fprog->len; if (!data) goto out; |
f8e529ed9 seccomp, ptrace: ... |
1974 1975 |
if (copy_to_user(data, fprog->filter, bpf_classic_proglen(fprog))) ret = -EFAULT; |
f8e529ed9 seccomp, ptrace: ... |
1976 |
out: |
66a733ea6 seccomp: fix the ... |
1977 |
__put_seccomp_filter(filter); |
f8e529ed9 seccomp, ptrace: ... |
1978 |
return ret; |
f8e529ed9 seccomp, ptrace: ... |
1979 |
} |
f8e529ed9 seccomp, ptrace: ... |
1980 |
|
26500475a ptrace, seccomp: ... |
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 |
long seccomp_get_metadata(struct task_struct *task, unsigned long size, void __user *data) { long ret; struct seccomp_filter *filter; struct seccomp_metadata kmd = {}; if (!capable(CAP_SYS_ADMIN) || current->seccomp.mode != SECCOMP_MODE_DISABLED) { return -EACCES; } size = min_t(unsigned long, size, sizeof(kmd)); |
63bb0045b ptrace, seccomp: ... |
1994 1995 1996 1997 |
if (size < sizeof(kmd.filter_off)) return -EINVAL; if (copy_from_user(&kmd.filter_off, data, sizeof(kmd.filter_off))) |
26500475a ptrace, seccomp: ... |
1998 1999 2000 2001 2002 |
return -EFAULT; filter = get_nth_filter(task, kmd.filter_off); if (IS_ERR(filter)) return PTR_ERR(filter); |
26500475a ptrace, seccomp: ... |
2003 2004 2005 2006 2007 2008 2009 2010 |
if (filter->log) kmd.flags |= SECCOMP_FILTER_FLAG_LOG; ret = size; if (copy_to_user(data, &kmd, size)) ret = -EFAULT; __put_seccomp_filter(filter); |
f8e529ed9 seccomp, ptrace: ... |
2011 2012 2013 |
return ret; } #endif |
8e5f1ad11 seccomp: Sysctl t... |
2014 2015 2016 2017 |
#ifdef CONFIG_SYSCTL /* Human readable action names for friendly sysctl interaction */ |
0466bdb99 seccomp: Implemen... |
2018 |
#define SECCOMP_RET_KILL_PROCESS_NAME "kill_process" |
fd76875ca seccomp: Rename S... |
2019 |
#define SECCOMP_RET_KILL_THREAD_NAME "kill_thread" |
8e5f1ad11 seccomp: Sysctl t... |
2020 2021 |
#define SECCOMP_RET_TRAP_NAME "trap" #define SECCOMP_RET_ERRNO_NAME "errno" |
6a21cc50f seccomp: add a re... |
2022 |
#define SECCOMP_RET_USER_NOTIF_NAME "user_notif" |
8e5f1ad11 seccomp: Sysctl t... |
2023 |
#define SECCOMP_RET_TRACE_NAME "trace" |
59f5cf44a seccomp: Action t... |
2024 |
#define SECCOMP_RET_LOG_NAME "log" |
8e5f1ad11 seccomp: Sysctl t... |
2025 |
#define SECCOMP_RET_ALLOW_NAME "allow" |
fd76875ca seccomp: Rename S... |
2026 |
static const char seccomp_actions_avail[] = |
0466bdb99 seccomp: Implemen... |
2027 |
SECCOMP_RET_KILL_PROCESS_NAME " " |
fd76875ca seccomp: Rename S... |
2028 2029 2030 |
SECCOMP_RET_KILL_THREAD_NAME " " SECCOMP_RET_TRAP_NAME " " SECCOMP_RET_ERRNO_NAME " " |
6a21cc50f seccomp: add a re... |
2031 |
SECCOMP_RET_USER_NOTIF_NAME " " |
fd76875ca seccomp: Rename S... |
2032 2033 2034 |
SECCOMP_RET_TRACE_NAME " " SECCOMP_RET_LOG_NAME " " SECCOMP_RET_ALLOW_NAME; |
8e5f1ad11 seccomp: Sysctl t... |
2035 |
|
0ddec0fc8 seccomp: Sysctl t... |
2036 2037 2038 2039 2040 2041 |
struct seccomp_log_name { u32 log; const char *name; }; static const struct seccomp_log_name seccomp_log_names[] = { |
0466bdb99 seccomp: Implemen... |
2042 |
{ SECCOMP_LOG_KILL_PROCESS, SECCOMP_RET_KILL_PROCESS_NAME }, |
fd76875ca seccomp: Rename S... |
2043 |
{ SECCOMP_LOG_KILL_THREAD, SECCOMP_RET_KILL_THREAD_NAME }, |
0ddec0fc8 seccomp: Sysctl t... |
2044 2045 |
{ SECCOMP_LOG_TRAP, SECCOMP_RET_TRAP_NAME }, { SECCOMP_LOG_ERRNO, SECCOMP_RET_ERRNO_NAME }, |
6a21cc50f seccomp: add a re... |
2046 |
{ SECCOMP_LOG_USER_NOTIF, SECCOMP_RET_USER_NOTIF_NAME }, |
0ddec0fc8 seccomp: Sysctl t... |
2047 |
{ SECCOMP_LOG_TRACE, SECCOMP_RET_TRACE_NAME }, |
59f5cf44a seccomp: Action t... |
2048 |
{ SECCOMP_LOG_LOG, SECCOMP_RET_LOG_NAME }, |
0ddec0fc8 seccomp: Sysctl t... |
2049 2050 2051 2052 2053 |
{ SECCOMP_LOG_ALLOW, SECCOMP_RET_ALLOW_NAME }, { } }; static bool seccomp_names_from_actions_logged(char *names, size_t size, |
beb44acaf seccomp: Configur... |
2054 2055 |
u32 actions_logged, const char *sep) |
0ddec0fc8 seccomp: Sysctl t... |
2056 2057 |
{ const struct seccomp_log_name *cur; |
beb44acaf seccomp: Configur... |
2058 |
bool append_sep = false; |
0ddec0fc8 seccomp: Sysctl t... |
2059 2060 2061 2062 2063 2064 |
for (cur = seccomp_log_names; cur->name && size; cur++) { ssize_t ret; if (!(actions_logged & cur->log)) continue; |
beb44acaf seccomp: Configur... |
2065 2066 |
if (append_sep) { ret = strscpy(names, sep, size); |
0ddec0fc8 seccomp: Sysctl t... |
2067 2068 2069 2070 2071 2072 |
if (ret < 0) return false; names += ret; size -= ret; } else |
beb44acaf seccomp: Configur... |
2073 |
append_sep = true; |
0ddec0fc8 seccomp: Sysctl t... |
2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 |
ret = strscpy(names, cur->name, size); if (ret < 0) return false; names += ret; size -= ret; } return true; } static bool seccomp_action_logged_from_name(u32 *action_logged, const char *name) { const struct seccomp_log_name *cur; for (cur = seccomp_log_names; cur->name; cur++) { if (!strcmp(cur->name, name)) { *action_logged = cur->log; return true; } } return false; } static bool seccomp_actions_logged_from_names(u32 *actions_logged, char *names) { char *name; *actions_logged = 0; while ((name = strsep(&names, " ")) && *name) { u32 action_logged = 0; if (!seccomp_action_logged_from_name(&action_logged, name)) return false; *actions_logged |= action_logged; } return true; } |
6212e56b3 UPSTREAM: seccomp... |
2117 |
static int read_actions_logged(struct ctl_table *ro_table, void *buffer, |
d013db029 seccomp: Separate... |
2118 |
size_t *lenp, loff_t *ppos) |
0ddec0fc8 seccomp: Sysctl t... |
2119 2120 2121 |
{ char names[sizeof(seccomp_actions_avail)]; struct ctl_table table; |
d013db029 seccomp: Separate... |
2122 2123 2124 2125 |
memset(names, 0, sizeof(names)); if (!seccomp_names_from_actions_logged(names, sizeof(names), |
beb44acaf seccomp: Configur... |
2126 |
seccomp_actions_logged, " ")) |
d013db029 seccomp: Separate... |
2127 2128 2129 2130 2131 2132 2133 |
return -EINVAL; table = *ro_table; table.data = names; table.maxlen = sizeof(names); return proc_dostring(&table, 0, buffer, lenp, ppos); } |
6212e56b3 UPSTREAM: seccomp... |
2134 |
static int write_actions_logged(struct ctl_table *ro_table, void *buffer, |
ea6eca778 seccomp: Audit at... |
2135 |
size_t *lenp, loff_t *ppos, u32 *actions_logged) |
0ddec0fc8 seccomp: Sysctl t... |
2136 2137 2138 2139 |
{ char names[sizeof(seccomp_actions_avail)]; struct ctl_table table; int ret; |
d013db029 seccomp: Separate... |
2140 |
if (!capable(CAP_SYS_ADMIN)) |
0ddec0fc8 seccomp: Sysctl t... |
2141 2142 2143 |
return -EPERM; memset(names, 0, sizeof(names)); |
0ddec0fc8 seccomp: Sysctl t... |
2144 2145 2146 |
table = *ro_table; table.data = names; table.maxlen = sizeof(names); |
d013db029 seccomp: Separate... |
2147 |
ret = proc_dostring(&table, 1, buffer, lenp, ppos); |
0ddec0fc8 seccomp: Sysctl t... |
2148 2149 |
if (ret) return ret; |
ea6eca778 seccomp: Audit at... |
2150 |
if (!seccomp_actions_logged_from_names(actions_logged, table.data)) |
d013db029 seccomp: Separate... |
2151 |
return -EINVAL; |
0ddec0fc8 seccomp: Sysctl t... |
2152 |
|
ea6eca778 seccomp: Audit at... |
2153 |
if (*actions_logged & SECCOMP_LOG_ALLOW) |
d013db029 seccomp: Separate... |
2154 |
return -EINVAL; |
0ddec0fc8 seccomp: Sysctl t... |
2155 |
|
ea6eca778 seccomp: Audit at... |
2156 |
seccomp_actions_logged = *actions_logged; |
0ddec0fc8 seccomp: Sysctl t... |
2157 2158 |
return 0; } |
0ddec0fc8 seccomp: Sysctl t... |
2159 |
|
ea6eca778 seccomp: Audit at... |
2160 2161 2162 2163 2164 2165 2166 |
static void audit_actions_logged(u32 actions_logged, u32 old_actions_logged, int ret) { char names[sizeof(seccomp_actions_avail)]; char old_names[sizeof(seccomp_actions_avail)]; const char *new = names; const char *old = old_names; |
0ddec0fc8 seccomp: Sysctl t... |
2167 |
|
ea6eca778 seccomp: Audit at... |
2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 |
if (!audit_enabled) return; memset(names, 0, sizeof(names)); memset(old_names, 0, sizeof(old_names)); if (ret) new = "?"; else if (!actions_logged) new = "(none)"; else if (!seccomp_names_from_actions_logged(names, sizeof(names), actions_logged, ",")) new = "?"; if (!old_actions_logged) old = "(none)"; else if (!seccomp_names_from_actions_logged(old_names, sizeof(old_names), old_actions_logged, ",")) old = "?"; return audit_seccomp_actions_logged(new, old, !ret); } |
d013db029 seccomp: Separate... |
2191 |
static int seccomp_actions_logged_handler(struct ctl_table *ro_table, int write, |
32927393d sysctl: pass kern... |
2192 |
void *buffer, size_t *lenp, |
d013db029 seccomp: Separate... |
2193 2194 |
loff_t *ppos) { |
ea6eca778 seccomp: Audit at... |
2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 |
int ret; if (write) { u32 actions_logged = 0; u32 old_actions_logged = seccomp_actions_logged; ret = write_actions_logged(ro_table, buffer, lenp, ppos, &actions_logged); audit_actions_logged(actions_logged, old_actions_logged, ret); } else ret = read_actions_logged(ro_table, buffer, lenp, ppos); return ret; |
0ddec0fc8 seccomp: Sysctl t... |
2208 |
} |
8e5f1ad11 seccomp: Sysctl t... |
2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 |
static struct ctl_path seccomp_sysctl_path[] = { { .procname = "kernel", }, { .procname = "seccomp", }, { } }; static struct ctl_table seccomp_sysctl_table[] = { { .procname = "actions_avail", .data = (void *) &seccomp_actions_avail, .maxlen = sizeof(seccomp_actions_avail), .mode = 0444, .proc_handler = proc_dostring, }, |
0ddec0fc8 seccomp: Sysctl t... |
2223 2224 2225 2226 2227 |
{ .procname = "actions_logged", .mode = 0644, .proc_handler = seccomp_actions_logged_handler, }, |
8e5f1ad11 seccomp: Sysctl t... |
2228 2229 2230 2231 2232 2233 2234 2235 2236 |
{ } }; static int __init seccomp_sysctl_init(void) { struct ctl_table_header *hdr; hdr = register_sysctl_paths(seccomp_sysctl_path, seccomp_sysctl_table); if (!hdr) |
e68f9d49d seccomp: Use pr_fmt |
2237 2238 |
pr_warn("sysctl registration failed "); |
8e5f1ad11 seccomp: Sysctl t... |
2239 2240 2241 2242 2243 2244 2245 2246 2247 |
else kmemleak_not_leak(hdr); return 0; } device_initcall(seccomp_sysctl_init) #endif /* CONFIG_SYSCTL */ |