summaryrefslogtreecommitdiffstats
path: root/src/basic/cgroup-util.h
blob: df6d5b7bbb423ee9612dc19ab5df0666ea6e1d66 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
/* SPDX-License-Identifier: LGPL-2.1-or-later */
#pragma once

#include <dirent.h>
#include <fcntl.h>
#include <stdbool.h>
#include <stdint.h>
#include <stdio.h>
#include <sys/statfs.h>
#include <sys/types.h>

#include "def.h"
#include "set.h"

#define SYSTEMD_CGROUP_CONTROLLER_LEGACY "name=systemd"
#define SYSTEMD_CGROUP_CONTROLLER_HYBRID "name=unified"
#define SYSTEMD_CGROUP_CONTROLLER "_systemd"

/* An enum of well known cgroup controllers */
typedef enum CGroupController {
        /* Original cgroup controllers */
        CGROUP_CONTROLLER_CPU,
        CGROUP_CONTROLLER_CPUACCT,    /* v1 only */
        CGROUP_CONTROLLER_CPUSET,     /* v2 only */
        CGROUP_CONTROLLER_IO,         /* v2 only */
        CGROUP_CONTROLLER_BLKIO,      /* v1 only */
        CGROUP_CONTROLLER_MEMORY,
        CGROUP_CONTROLLER_DEVICES,    /* v1 only */
        CGROUP_CONTROLLER_PIDS,

        /* BPF-based pseudo-controllers, v2 only */
        CGROUP_CONTROLLER_BPF_FIREWALL,
        CGROUP_CONTROLLER_BPF_DEVICES,
        CGROUP_CONTROLLER_BPF_FOREIGN,
        CGROUP_CONTROLLER_BPF_SOCKET_BIND,
        CGROUP_CONTROLLER_BPF_RESTRICT_NETWORK_INTERFACES,
        /* The BPF hook implementing RestrictFileSystems= is not defined here.
         * It's applied as late as possible in exec_child() so we don't block
         * our own unit setup code. */

        _CGROUP_CONTROLLER_MAX,
        _CGROUP_CONTROLLER_INVALID = -EINVAL,
} CGroupController;

#define CGROUP_CONTROLLER_TO_MASK(c) (1U << (c))

/* A bit mask of well known cgroup controllers */
typedef enum CGroupMask {
        CGROUP_MASK_CPU = CGROUP_CONTROLLER_TO_MASK(CGROUP_CONTROLLER_CPU),
        CGROUP_MASK_CPUACCT = CGROUP_CONTROLLER_TO_MASK(CGROUP_CONTROLLER_CPUACCT),
        CGROUP_MASK_CPUSET = CGROUP_CONTROLLER_TO_MASK(CGROUP_CONTROLLER_CPUSET),
        CGROUP_MASK_IO = CGROUP_CONTROLLER_TO_MASK(CGROUP_CONTROLLER_IO),
        CGROUP_MASK_BLKIO = CGROUP_CONTROLLER_TO_MASK(CGROUP_CONTROLLER_BLKIO),
        CGROUP_MASK_MEMORY = CGROUP_CONTROLLER_TO_MASK(CGROUP_CONTROLLER_MEMORY),
        CGROUP_MASK_DEVICES = CGROUP_CONTROLLER_TO_MASK(CGROUP_CONTROLLER_DEVICES),
        CGROUP_MASK_PIDS = CGROUP_CONTROLLER_TO_MASK(CGROUP_CONTROLLER_PIDS),
        CGROUP_MASK_BPF_FIREWALL = CGROUP_CONTROLLER_TO_MASK(CGROUP_CONTROLLER_BPF_FIREWALL),
        CGROUP_MASK_BPF_DEVICES = CGROUP_CONTROLLER_TO_MASK(CGROUP_CONTROLLER_BPF_DEVICES),
        CGROUP_MASK_BPF_FOREIGN = CGROUP_CONTROLLER_TO_MASK(CGROUP_CONTROLLER_BPF_FOREIGN),
        CGROUP_MASK_BPF_SOCKET_BIND = CGROUP_CONTROLLER_TO_MASK(CGROUP_CONTROLLER_BPF_SOCKET_BIND),
        CGROUP_MASK_BPF_RESTRICT_NETWORK_INTERFACES = CGROUP_CONTROLLER_TO_MASK(CGROUP_CONTROLLER_BPF_RESTRICT_NETWORK_INTERFACES),

        /* All real cgroup v1 controllers */
        CGROUP_MASK_V1 = CGROUP_MASK_CPU|CGROUP_MASK_CPUACCT|CGROUP_MASK_BLKIO|CGROUP_MASK_MEMORY|CGROUP_MASK_DEVICES|CGROUP_MASK_PIDS,

        /* All real cgroup v2 controllers */
        CGROUP_MASK_V2 = CGROUP_MASK_CPU|CGROUP_MASK_CPUSET|CGROUP_MASK_IO|CGROUP_MASK_MEMORY|CGROUP_MASK_PIDS,

        /* All cgroup v2 BPF pseudo-controllers */
        CGROUP_MASK_BPF = CGROUP_MASK_BPF_FIREWALL|CGROUP_MASK_BPF_DEVICES|CGROUP_MASK_BPF_FOREIGN|CGROUP_MASK_BPF_SOCKET_BIND|CGROUP_MASK_BPF_RESTRICT_NETWORK_INTERFACES,

        _CGROUP_MASK_ALL = CGROUP_CONTROLLER_TO_MASK(_CGROUP_CONTROLLER_MAX) - 1
} CGroupMask;

static inline CGroupMask CGROUP_MASK_EXTEND_JOINED(CGroupMask mask) {
        /* We always mount "cpu" and "cpuacct" in the same hierarchy. Hence, when one bit is set also set the other */

        if (mask & (CGROUP_MASK_CPU|CGROUP_MASK_CPUACCT))
                mask |= (CGROUP_MASK_CPU|CGROUP_MASK_CPUACCT);

        return mask;
}

CGroupMask get_cpu_accounting_mask(void);
bool cpu_accounting_is_cheap(void);

/* Special values for all weight knobs on unified hierarchy */
#define CGROUP_WEIGHT_INVALID UINT64_MAX
#define CGROUP_WEIGHT_IDLE UINT64_C(0)
#define CGROUP_WEIGHT_MIN UINT64_C(1)
#define CGROUP_WEIGHT_MAX UINT64_C(10000)
#define CGROUP_WEIGHT_DEFAULT UINT64_C(100)

#define CGROUP_LIMIT_MIN UINT64_C(0)
#define CGROUP_LIMIT_MAX UINT64_MAX

static inline bool CGROUP_WEIGHT_IS_OK(uint64_t x) {
        return
            x == CGROUP_WEIGHT_INVALID ||
            (x >= CGROUP_WEIGHT_MIN && x <= CGROUP_WEIGHT_MAX);
}

/* IO limits on unified hierarchy */
typedef enum CGroupIOLimitType {
        CGROUP_IO_RBPS_MAX,
        CGROUP_IO_WBPS_MAX,
        CGROUP_IO_RIOPS_MAX,
        CGROUP_IO_WIOPS_MAX,

        _CGROUP_IO_LIMIT_TYPE_MAX,
        _CGROUP_IO_LIMIT_TYPE_INVALID = -EINVAL,
} CGroupIOLimitType;

extern const uint64_t cgroup_io_limit_defaults[_CGROUP_IO_LIMIT_TYPE_MAX];

const char* cgroup_io_limit_type_to_string(CGroupIOLimitType t) _const_;
CGroupIOLimitType cgroup_io_limit_type_from_string(const char *s) _pure_;

/* Special values for the cpu.shares attribute */
#define CGROUP_CPU_SHARES_INVALID UINT64_MAX
#define CGROUP_CPU_SHARES_MIN UINT64_C(2)
#define CGROUP_CPU_SHARES_MAX UINT64_C(262144)
#define CGROUP_CPU_SHARES_DEFAULT UINT64_C(1024)

static inline bool CGROUP_CPU_SHARES_IS_OK(uint64_t x) {
        return
            x == CGROUP_CPU_SHARES_INVALID ||
            (x >= CGROUP_CPU_SHARES_MIN && x <= CGROUP_CPU_SHARES_MAX);
}

/* Special values for the special {blkio,io}.bfq.weight attribute */
#define CGROUP_BFQ_WEIGHT_INVALID UINT64_MAX
#define CGROUP_BFQ_WEIGHT_MIN UINT64_C(1)
#define CGROUP_BFQ_WEIGHT_MAX UINT64_C(1000)
#define CGROUP_BFQ_WEIGHT_DEFAULT UINT64_C(100)

/* Convert the normal io.weight value to io.bfq.weight */
static inline uint64_t BFQ_WEIGHT(uint64_t io_weight) {
        return
            io_weight <= CGROUP_WEIGHT_DEFAULT ?
            CGROUP_BFQ_WEIGHT_DEFAULT - (CGROUP_WEIGHT_DEFAULT - io_weight) * (CGROUP_BFQ_WEIGHT_DEFAULT - CGROUP_BFQ_WEIGHT_MIN) / (CGROUP_WEIGHT_DEFAULT - CGROUP_WEIGHT_MIN) :
            CGROUP_BFQ_WEIGHT_DEFAULT + (io_weight - CGROUP_WEIGHT_DEFAULT) * (CGROUP_BFQ_WEIGHT_MAX - CGROUP_BFQ_WEIGHT_DEFAULT) / (CGROUP_WEIGHT_MAX - CGROUP_WEIGHT_DEFAULT);
}

/* Special values for the blkio.weight attribute */
#define CGROUP_BLKIO_WEIGHT_INVALID UINT64_MAX
#define CGROUP_BLKIO_WEIGHT_MIN UINT64_C(10)
#define CGROUP_BLKIO_WEIGHT_MAX UINT64_C(1000)
#define CGROUP_BLKIO_WEIGHT_DEFAULT UINT64_C(500)

static inline bool CGROUP_BLKIO_WEIGHT_IS_OK(uint64_t x) {
        return
            x == CGROUP_BLKIO_WEIGHT_INVALID ||
            (x >= CGROUP_BLKIO_WEIGHT_MIN && x <= CGROUP_BLKIO_WEIGHT_MAX);
}

typedef enum CGroupUnified {
        CGROUP_UNIFIED_UNKNOWN = -1,
        CGROUP_UNIFIED_NONE = 0,        /* Both systemd and controllers on legacy */
        CGROUP_UNIFIED_SYSTEMD = 1,     /* Only systemd on unified */
        CGROUP_UNIFIED_ALL = 2,         /* Both systemd and controllers on unified */
} CGroupUnified;

/*
 * General rules:
 *
 * We accept named hierarchies in the syntax "foo" and "name=foo".
 *
 * We expect that named hierarchies do not conflict in name with a
 * kernel hierarchy, modulo the "name=" prefix.
 *
 * We always generate "normalized" controller names, i.e. without the
 * "name=" prefix.
 *
 * We require absolute cgroup paths. When returning, we will always
 * generate paths with multiple adjacent / removed.
 */

int cg_enumerate_processes(const char *controller, const char *path, FILE **_f);
int cg_read_pid(FILE *f, pid_t *_pid);
int cg_read_event(const char *controller, const char *path, const char *event,
                  char **val);

int cg_enumerate_subgroups(const char *controller, const char *path, DIR **_d);
int cg_read_subgroup(DIR *d, char **fn);

typedef enum CGroupFlags {
        CGROUP_SIGCONT     = 1 << 0,
        CGROUP_IGNORE_SELF = 1 << 1,
        CGROUP_REMOVE      = 1 << 2,
} CGroupFlags;

typedef int (*cg_kill_log_func_t)(pid_t pid, int sig, void *userdata);

int cg_kill(const char *controller, const char *path, int sig, CGroupFlags flags, Set *s, cg_kill_log_func_t kill_log, void *userdata);
int cg_kill_kernel_sigkill(const char *controller, const char *path);
int cg_kill_recursive(const char *controller, const char *path, int sig, CGroupFlags flags, Set *s, cg_kill_log_func_t kill_log, void *userdata);

int cg_split_spec(const char *spec, char **ret_controller, char **ret_path);
int cg_mangle_path(const char *path, char **result);

int cg_get_path(const char *controller, const char *path, const char *suffix, char **fs);
int cg_get_path_and_check(const char *controller, const char *path, const char *suffix, char **fs);

int cg_pid_get_path(const char *controller, pid_t pid, char **path);

int cg_rmdir(const char *controller, const char *path);

int cg_is_threaded(const char *controller, const char *path);

typedef enum  {
        CG_KEY_MODE_GRACEFUL = 1 << 0,
} CGroupKeyMode;

int cg_set_attribute(const char *controller, const char *path, const char *attribute, const char *value);
int cg_get_attribute(const char *controller, const char *path, const char *attribute, char **ret);
int cg_get_keyed_attribute_full(const char *controller, const char *path, const char *attribute, char **keys, char **values, CGroupKeyMode mode);

static inline int cg_get_keyed_attribute(
                const char *controller,
                const char *path,
                const char *attribute,
                char **keys,
                char **ret_values) {
        return cg_get_keyed_attribute_full(controller, path, attribute, keys, ret_values, 0);
}

static inline int cg_get_keyed_attribute_graceful(
                const char *controller,
                const char *path,
                const char *attribute,
                char **keys,
                char **ret_values) {
        return cg_get_keyed_attribute_full(controller, path, attribute, keys, ret_values, CG_KEY_MODE_GRACEFUL);
}

int cg_get_attribute_as_uint64(const char *controller, const char *path, const char *attribute, uint64_t *ret);

/* Does a parse_boolean() on the attribute contents and sets ret accordingly */
int cg_get_attribute_as_bool(const char *controller, const char *path, const char *attribute, bool *ret);

int cg_set_access(const char *controller, const char *path, uid_t uid, gid_t gid);
int cg_get_owner(const char *controller, const char *path, uid_t *ret_uid);

int cg_set_xattr(const char *controller, const char *path, const char *name, const void *value, size_t size, int flags);
int cg_get_xattr(const char *controller, const char *path, const char *name, void *value, size_t size);
int cg_get_xattr_malloc(const char *controller, const char *path, const char *name, char **ret);
/* Returns negative on error, and 0 or 1 on success for the bool value */
int cg_get_xattr_bool(const char *controller, const char *path, const char *name);
int cg_remove_xattr(const char *controller, const char *path, const char *name);

int cg_install_release_agent(const char *controller, const char *agent);
int cg_uninstall_release_agent(const char *controller);

int cg_is_empty(const char *controller, const char *path);
int cg_is_empty_recursive(const char *controller, const char *path);

int cg_get_root_path(char **path);

int cg_path_get_cgroupid(const char *path, uint64_t *ret);
int cg_path_get_session(const char *path, char **session);
int cg_path_get_owner_uid(const char *path, uid_t *uid);
int cg_path_get_unit(const char *path, char **unit);
int cg_path_get_user_unit(const char *path, char **unit);
int cg_path_get_machine_name(const char *path, char **machine);
int cg_path_get_slice(const char *path, char **slice);
int cg_path_get_user_slice(const char *path, char **slice);

int cg_shift_path(const char *cgroup, const char *cached_root, const char **shifted);
int cg_pid_get_path_shifted(pid_t pid, const char *cached_root, char **cgroup);

int cg_pid_get_session(pid_t pid, char **session);
int cg_pid_get_owner_uid(pid_t pid, uid_t *uid);
int cg_pid_get_unit(pid_t pid, char **unit);
int cg_pid_get_user_unit(pid_t pid, char **unit);
int cg_pid_get_machine_name(pid_t pid, char **machine);
int cg_pid_get_slice(pid_t pid, char **slice);
int cg_pid_get_user_slice(pid_t pid, char **slice);

int cg_path_decode_unit(const char *cgroup, char **unit);

char *cg_escape(const char *p);
char *cg_unescape(const char *p) _pure_;

bool cg_controller_is_valid(const char *p);

int cg_slice_to_path(const char *unit, char **ret);

typedef const char* (*cg_migrate_callback_t)(CGroupMask mask, void *userdata);

int cg_mask_supported(CGroupMask *ret);
int cg_mask_supported_subtree(const char *root, CGroupMask *ret);
int cg_mask_from_string(const char *s, CGroupMask *ret);
int cg_mask_to_string(CGroupMask mask, char **ret);

int cg_kernel_controllers(Set **controllers);

bool cg_ns_supported(void);
bool cg_freezer_supported(void);
bool cg_kill_supported(void);

int cg_all_unified(void);
int cg_hybrid_unified(void);
int cg_unified_controller(const char *controller);
int cg_unified_cached(bool flush);
static inline int cg_unified(void) {
        return cg_unified_cached(true);
}

const char* cgroup_controller_to_string(CGroupController c) _const_;
CGroupController cgroup_controller_from_string(const char *s) _pure_;

bool is_cgroup_fs(const struct statfs *s);
bool fd_is_cgroup_fs(int fd);

typedef enum ManagedOOMMode {
        MANAGED_OOM_AUTO,
        MANAGED_OOM_KILL,
        _MANAGED_OOM_MODE_MAX,
        _MANAGED_OOM_MODE_INVALID = -EINVAL,
} ManagedOOMMode;

const char* managed_oom_mode_to_string(ManagedOOMMode m) _const_;
ManagedOOMMode managed_oom_mode_from_string(const char *s) _pure_;

typedef enum ManagedOOMPreference {
        MANAGED_OOM_PREFERENCE_NONE = 0,
        MANAGED_OOM_PREFERENCE_AVOID = 1,
        MANAGED_OOM_PREFERENCE_OMIT = 2,
        _MANAGED_OOM_PREFERENCE_MAX,
        _MANAGED_OOM_PREFERENCE_INVALID = -EINVAL,
} ManagedOOMPreference;

const char* managed_oom_preference_to_string(ManagedOOMPreference a) _const_;
ManagedOOMPreference managed_oom_preference_from_string(const char *s) _pure_;

/* The structure to pass to name_to_handle_at() on cgroupfs2 */
typedef union {
        struct file_handle file_handle;
        uint8_t space[offsetof(struct file_handle, f_handle) + sizeof(uint64_t)];
} cg_file_handle;

#define CG_FILE_HANDLE_INIT { .file_handle.handle_bytes = sizeof(uint64_t) }
#define CG_FILE_HANDLE_CGROUPID(fh) (*(uint64_t*) (fh).file_handle.f_handle)