Blame view
init/main.c
26.8 KB
1da177e4c
|
1 2 3 4 5 6 7 8 |
/* * linux/init/main.c * * Copyright (C) 1991, 1992 Linus Torvalds * * GK 2/5/95 - Changed to support mounting root fs via NFS * Added initrd & change_root: Werner Almesberger & Hans Lermen, Feb '96 * Moan early if gcc is old, avoiding bogus kernels - Paul Gortmaker, May '96 |
dd4d9fecb
|
9 |
* Simplified starting of init: Michael A. Griffith <grif@acm.org> |
1da177e4c
|
10 |
*/ |
ea676e846
|
11 |
#define DEBUG /* Enable initcall_debug */ |
1da177e4c
|
12 |
#include <linux/types.h> |
8a293be0d
|
13 |
#include <linux/extable.h> |
1da177e4c
|
14 15 |
#include <linux/module.h> #include <linux/proc_fs.h> |
5c2c5c551
|
16 |
#include <linux/binfmts.h> |
1da177e4c
|
17 18 |
#include <linux/kernel.h> #include <linux/syscalls.h> |
9b5609fd7
|
19 |
#include <linux/stackprotector.h> |
1da177e4c
|
20 21 22 |
#include <linux/string.h> #include <linux/ctype.h> #include <linux/delay.h> |
1da177e4c
|
23 24 |
#include <linux/ioport.h> #include <linux/init.h> |
1da177e4c
|
25 |
#include <linux/initrd.h> |
1da177e4c
|
26 |
#include <linux/bootmem.h> |
4a7a16dc0
|
27 |
#include <linux/acpi.h> |
0c688614d
|
28 |
#include <linux/console.h> |
38b8d208a
|
29 |
#include <linux/nmi.h> |
1da177e4c
|
30 31 |
#include <linux/percpu.h> #include <linux/kmod.h> |
db64fe022
|
32 |
#include <linux/vmalloc.h> |
1da177e4c
|
33 |
#include <linux/kernel_stat.h> |
d7cd56111
|
34 |
#include <linux/start_kernel.h> |
1da177e4c
|
35 |
#include <linux/security.h> |
3d4422332
|
36 |
#include <linux/smp.h> |
1da177e4c
|
37 38 39 40 41 42 43 |
#include <linux/profile.h> #include <linux/rcupdate.h> #include <linux/moduleparam.h> #include <linux/kallsyms.h> #include <linux/writeback.h> #include <linux/cpu.h> #include <linux/cpuset.h> |
ddbcc7e8e
|
44 |
#include <linux/cgroup.h> |
1da177e4c
|
45 |
#include <linux/efi.h> |
906568c9c
|
46 |
#include <linux/tick.h> |
6168a702a
|
47 |
#include <linux/interrupt.h> |
c757249af
|
48 |
#include <linux/taskstats_kern.h> |
ca74e92b4
|
49 |
#include <linux/delayacct.h> |
1da177e4c
|
50 51 52 53 |
#include <linux/unistd.h> #include <linux/rmap.h> #include <linux/mempolicy.h> #include <linux/key.h> |
b6cd0b772
|
54 |
#include <linux/buffer_head.h> |
eefa864b7
|
55 |
#include <linux/page_ext.h> |
9a11b49a8
|
56 |
#include <linux/debug_locks.h> |
3ac7fe5a4
|
57 |
#include <linux/debugobjects.h> |
fbb9ce953
|
58 |
#include <linux/lockdep.h> |
3c7b4e6b8
|
59 |
#include <linux/kmemleak.h> |
84d737866
|
60 |
#include <linux/pid_namespace.h> |
1f21782e6
|
61 |
#include <linux/device.h> |
73c279927
|
62 |
#include <linux/kthread.h> |
e6fe6649b
|
63 |
#include <linux/sched.h> |
1777e4635
|
64 |
#include <linux/sched/init.h> |
a1c9eea9e
|
65 |
#include <linux/signal.h> |
199f0ca51
|
66 |
#include <linux/idr.h> |
0b4b3827d
|
67 |
#include <linux/kgdb.h> |
68bf21aa1
|
68 |
#include <linux/ftrace.h> |
22a9d6456
|
69 |
#include <linux/async.h> |
6ae6996a4
|
70 |
#include <linux/sfi.h> |
2b2af54a5
|
71 |
#include <linux/shmem_fs.h> |
5a0e3ad6a
|
72 |
#include <linux/slab.h> |
24a24bb6f
|
73 |
#include <linux/perf_event.h> |
a74fb73c1
|
74 |
#include <linux/ptrace.h> |
a4b07fb4e
|
75 |
#include <linux/pti.h> |
bb813f4c9
|
76 77 |
#include <linux/blkdev.h> #include <linux/elevator.h> |
38ff87f77
|
78 |
#include <linux/sched_clock.h> |
299300258
|
79 |
#include <linux/sched/task.h> |
68db0cf10
|
80 |
#include <linux/sched/task_stack.h> |
65f382fd0
|
81 |
#include <linux/context_tracking.h> |
47d06e532
|
82 |
#include <linux/random.h> |
7b0b73d76
|
83 |
#include <linux/list.h> |
c9cd2ce2b
|
84 |
#include <linux/integrity.h> |
e149ed2b8
|
85 |
#include <linux/proc_ns.h> |
0ddab1d2e
|
86 |
#include <linux/io.h> |
39290b389
|
87 |
#include <linux/cache.h> |
2959a5f72
|
88 |
#include <linux/rodata_test.h> |
1da177e4c
|
89 90 91 92 |
#include <asm/io.h> #include <asm/bugs.h> #include <asm/setup.h> |
a940199f2
|
93 |
#include <asm/sections.h> |
37b73c828
|
94 |
#include <asm/cacheflush.h> |
1da177e4c
|
95 |
|
aae5f662a
|
96 |
static int kernel_init(void *); |
1da177e4c
|
97 98 |
extern void init_IRQ(void); |
ff691f6e0
|
99 |
extern void fork_init(void); |
1da177e4c
|
100 |
extern void radix_tree_init(void); |
1da177e4c
|
101 |
|
2ce802f62
|
102 103 104 105 106 107 108 109 |
/* * Debug helper: via this flag we know that we are in 'early bootup code' * where only the boot processor is running with IRQ disabled. This means * two things - IRQ must not be enabled before the flag is cleared and some * operations which are not allowed with IRQ disabled are allowed while the * flag is set. */ bool early_boot_irqs_disabled __read_mostly; |
a68260483
|
110 |
enum system_states system_state __read_mostly; |
1da177e4c
|
111 112 113 114 115 116 117 118 119 120 |
EXPORT_SYMBOL(system_state); /* * Boot command-line arguments */ #define MAX_INIT_ARGS CONFIG_INIT_ENV_ARG_LIMIT #define MAX_INIT_ENVS CONFIG_INIT_ENV_ARG_LIMIT extern void time_init(void); /* Default late time init is NULL. archs can override this later. */ |
d2e3192b6
|
121 |
void (*__initdata late_time_init)(void); |
1da177e4c
|
122 |
|
30d7e0d46
|
123 124 125 126 127 128 |
/* Untouched command line saved by arch-specific code. */ char __initdata boot_command_line[COMMAND_LINE_SIZE]; /* Untouched saved command line (eg. for /proc) */ char *saved_command_line; /* Command line for parameter parsing */ static char *static_command_line; |
08746a65c
|
129 130 |
/* Command line for per-initcall parameter parsing */ static char *initcall_command_line; |
1da177e4c
|
131 132 |
static char *execute_command; |
ffdfc4097
|
133 |
static char *ramdisk_execute_command; |
1da177e4c
|
134 |
|
8b3b29550
|
135 |
/* |
c4b2c0c5f
|
136 137 138 |
* Used to generate warnings if static_key manipulation functions are used * before jump_label_init is called. */ |
dd4d9fecb
|
139 |
bool static_key_initialized __read_mostly; |
c4b2c0c5f
|
140 141 142 |
EXPORT_SYMBOL_GPL(static_key_initialized); /* |
8b3b29550
|
143 144 145 146 147 |
* If set, this is an indication to the drivers that reset the underlying * device before going ahead with the initialization otherwise driver might * rely on the BIOS and skip the reset operation. * * This is useful if kernel is booting in an unreliable environment. |
fc454fdc1
|
148 |
* For ex. kdump situation where previous kernel has crashed, BIOS has been |
8b3b29550
|
149 150 151 152 |
* skipped and devices will be in unknown state. */ unsigned int reset_devices; EXPORT_SYMBOL(reset_devices); |
1da177e4c
|
153 |
|
7e96287dd
|
154 155 156 157 158 159 160 |
static int __init set_reset_devices(char *str) { reset_devices = 1; return 1; } __setup("reset_devices", set_reset_devices); |
dd4d9fecb
|
161 162 |
static const char *argv_init[MAX_INIT_ARGS+2] = { "init", NULL, }; const char *envp_init[MAX_INIT_ENVS+2] = { "HOME=/", "TERM=linux", NULL, }; |
1da177e4c
|
163 |
static const char *panic_later, *panic_param; |
914dcaa84
|
164 |
extern const struct obs_kernel_param __setup_start[], __setup_end[]; |
1da177e4c
|
165 |
|
31c025b5f
|
166 |
static bool __init obsolete_checksetup(char *line) |
1da177e4c
|
167 |
{ |
914dcaa84
|
168 |
const struct obs_kernel_param *p; |
31c025b5f
|
169 |
bool had_early_param = false; |
1da177e4c
|
170 171 172 173 |
p = __setup_start; do { int n = strlen(p->str); |
b1e4d20cb
|
174 |
if (parameqn(line, p->str, n)) { |
1da177e4c
|
175 |
if (p->early) { |
33df0d19e
|
176 177 178 179 |
/* Already done in parse_early_param? * (Needs exact match on param part). * Keep iterating, as we can have early * params and __setups of same names 8( */ |
1da177e4c
|
180 |
if (line[n] == '\0' || line[n] == '=') |
31c025b5f
|
181 |
had_early_param = true; |
1da177e4c
|
182 |
} else if (!p->setup_func) { |
ea676e846
|
183 184 185 |
pr_warn("Parameter %s is obsolete, ignored ", p->str); |
31c025b5f
|
186 |
return true; |
1da177e4c
|
187 |
} else if (p->setup_func(line + n)) |
31c025b5f
|
188 |
return true; |
1da177e4c
|
189 190 191 |
} p++; } while (p < __setup_end); |
33df0d19e
|
192 193 |
return had_early_param; |
1da177e4c
|
194 195 196 197 198 199 200 |
} /* * This should be approx 2 Bo*oMips to start (note initial shift), and will * still work even if initially too large, it will just take slightly longer */ unsigned long loops_per_jiffy = (1<<12); |
1da177e4c
|
201 202 203 204 |
EXPORT_SYMBOL(loops_per_jiffy); static int __init debug_kernel(char *str) { |
a8fe19ebf
|
205 |
console_loglevel = CONSOLE_LOGLEVEL_DEBUG; |
f6f21c814
|
206 |
return 0; |
1da177e4c
|
207 208 209 210 |
} static int __init quiet_kernel(char *str) { |
a8fe19ebf
|
211 |
console_loglevel = CONSOLE_LOGLEVEL_QUIET; |
f6f21c814
|
212 |
return 0; |
1da177e4c
|
213 |
} |
f6f21c814
|
214 215 |
early_param("debug", debug_kernel); early_param("quiet", quiet_kernel); |
1da177e4c
|
216 217 218 |
static int __init loglevel(char *str) { |
808bf29b9
|
219 220 221 222 223 224 225 226 227 228 229 230 231 |
int newlevel; /* * Only update loglevel value when a correct setting was passed, * to prevent blind crashes (when loglevel being set to 0) that * are quite hard to debug */ if (get_option(&str, &newlevel)) { console_loglevel = newlevel; return 0; } return -EINVAL; |
1da177e4c
|
232 |
} |
f6f21c814
|
233 |
early_param("loglevel", loglevel); |
1da177e4c
|
234 |
|
a99cd1125
|
235 |
/* Change NUL term back to "=", to make "param" the whole string. */ |
ecc861705
|
236 237 |
static int __init repair_env_string(char *param, char *val, const char *unused, void *arg) |
1da177e4c
|
238 |
{ |
1da177e4c
|
239 240 241 242 243 244 245 246 247 248 249 |
if (val) { /* param=val or param="val"? */ if (val == param+strlen(param)+1) val[-1] = '='; else if (val == param+strlen(param)+2) { val[-2] = '='; memmove(val-1, val, strlen(val)+1); val--; } else BUG(); } |
a99cd1125
|
250 251 |
return 0; } |
51e158c12
|
252 |
/* Anything after -- gets handed straight to init. */ |
ecc861705
|
253 254 |
static int __init set_init_arg(char *param, char *val, const char *unused, void *arg) |
51e158c12
|
255 256 257 258 259 |
{ unsigned int i; if (panic_later) return 0; |
ecc861705
|
260 |
repair_env_string(param, val, unused, NULL); |
51e158c12
|
261 262 263 264 265 266 267 268 269 270 271 |
for (i = 0; argv_init[i]; i++) { if (i == MAX_INIT_ARGS) { panic_later = "init"; panic_param = param; return 0; } } argv_init[i] = param; return 0; } |
a99cd1125
|
272 273 274 275 |
/* * Unknown boot options get handed to init, unless they look like * unused parameters (modprobe will find them in /proc/cmdline). */ |
ecc861705
|
276 277 |
static int __init unknown_bootoption(char *param, char *val, const char *unused, void *arg) |
a99cd1125
|
278 |
{ |
ecc861705
|
279 |
repair_env_string(param, val, unused, NULL); |
1da177e4c
|
280 281 282 283 |
/* Handle obsolete-style parameters */ if (obsolete_checksetup(param)) return 0; |
f066a4f6d
|
284 285 |
/* Unused module parameter. */ if (strchr(param, '.') && (!val || strchr(param, '.') < val)) |
1da177e4c
|
286 |
return 0; |
1da177e4c
|
287 288 289 290 291 292 293 294 295 |
if (panic_later) return 0; if (val) { /* Environment option */ unsigned int i; for (i = 0; envp_init[i]; i++) { if (i == MAX_INIT_ENVS) { |
499a4584d
|
296 |
panic_later = "env"; |
1da177e4c
|
297 298 299 300 301 302 303 304 305 306 307 |
panic_param = param; } if (!strncmp(param, envp_init[i], val - param)) break; } envp_init[i] = param; } else { /* Command line option */ unsigned int i; for (i = 0; argv_init[i]; i++) { if (i == MAX_INIT_ARGS) { |
499a4584d
|
308 |
panic_later = "init"; |
1da177e4c
|
309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 |
panic_param = param; } } argv_init[i] = param; } return 0; } static int __init init_setup(char *str) { unsigned int i; execute_command = str; /* * In case LILO is going to boot us with default command line, * it prepends "auto" before the whole cmdline which makes * the shell think it should execute a script with such name. * So we ignore all arguments entered _before_ init=... [MJ] */ for (i = 1; i < MAX_INIT_ARGS; i++) argv_init[i] = NULL; return 1; } __setup("init=", init_setup); |
ffdfc4097
|
333 334 335 336 337 338 339 340 341 342 343 |
static int __init rdinit_setup(char *str) { unsigned int i; ramdisk_execute_command = str; /* See "auto" comment in init_setup */ for (i = 1; i < MAX_INIT_ARGS; i++) argv_init[i] = NULL; return 1; } __setup("rdinit=", rdinit_setup); |
1da177e4c
|
344 |
#ifndef CONFIG_SMP |
34db18a05
|
345 |
static const unsigned int setup_max_cpus = NR_CPUS; |
e0982e90c
|
346 |
static inline void setup_nr_cpu_ids(void) { } |
1da177e4c
|
347 |
static inline void smp_prepare_cpus(unsigned int maxcpus) { } |
1da177e4c
|
348 349 350 |
#endif /* |
30d7e0d46
|
351 352 353 354 355 356 357 |
* We need to store the untouched command line for future reference. * We also need to store the touched command line since the parameter * parsing is performed in place, and we should allow a component to * store reference of name/value for future reference. */ static void __init setup_command_line(char *command_line) { |
098b081b5
|
358 359 360 361 362 |
saved_command_line = memblock_virt_alloc(strlen(boot_command_line) + 1, 0); initcall_command_line = memblock_virt_alloc(strlen(boot_command_line) + 1, 0); static_command_line = memblock_virt_alloc(strlen(command_line) + 1, 0); |
dd4d9fecb
|
363 364 |
strcpy(saved_command_line, boot_command_line); strcpy(static_command_line, command_line); |
30d7e0d46
|
365 366 367 |
} /* |
1da177e4c
|
368 369 370 371 372 373 374 |
* We need to finalize in a non-__init function or else race conditions * between the root thread and the init thread may cause start_kernel to * be reaped by free_initmem before the root thread has proceeded to * cpu_idle. * * gcc-3.4 accidentally inlines this function, so use noinline. */ |
b433c3d45
|
375 |
static __initdata DECLARE_COMPLETION(kthreadd_done); |
bd721ea73
|
376 |
static noinline void __ref rest_init(void) |
1da177e4c
|
377 |
{ |
8fb12156b
|
378 |
struct task_struct *tsk; |
73c279927
|
379 |
int pid; |
7db905e63
|
380 |
rcu_scheduler_starting(); |
b433c3d45
|
381 |
/* |
971585692
|
382 |
* We need to spawn init first so that it obtains pid 1, however |
b433c3d45
|
383 384 385 |
* the init task will end up wanting to create kthreads, which, if * we schedule it before we create kthreadd, will OOPS. */ |
8fb12156b
|
386 387 388 389 390 391 392 393 394 395 |
pid = kernel_thread(kernel_init, NULL, CLONE_FS); /* * Pin init on the boot CPU. Task migration is not properly working * until sched_init_smp() has been run. It will set the allowed * CPUs for init to the non isolated CPUs. */ rcu_read_lock(); tsk = find_task_by_pid_ns(pid, &init_pid_ns); set_cpus_allowed_ptr(tsk, cpumask_of(smp_processor_id())); rcu_read_unlock(); |
1da177e4c
|
396 |
numa_default_policy(); |
73c279927
|
397 |
pid = kernel_thread(kthreadd, NULL, CLONE_FS | CLONE_FILES); |
d11c563dd
|
398 |
rcu_read_lock(); |
5cd204550
|
399 |
kthreadd_task = find_task_by_pid_ns(pid, &init_pid_ns); |
d11c563dd
|
400 |
rcu_read_unlock(); |
1c3c5eab1
|
401 402 403 404 405 406 407 408 409 |
/* * Enable might_sleep() and smp_processor_id() checks. * They cannot be enabled earlier because with CONFIG_PRREMPT=y * kernel_thread() would trigger might_sleep() splats. With * CONFIG_PREEMPT_VOLUNTARY=y the init task might have scheduled * already, but it's stuck on the kthreadd_done completion. */ system_state = SYSTEM_SCHEDULING; |
b433c3d45
|
410 |
complete(&kthreadd_done); |
f340c0d1a
|
411 412 413 |
/* * The boot idle thread must execute schedule() |
1df21055e
|
414 |
* at least once to get things moving: |
f340c0d1a
|
415 |
*/ |
bd2f55361
|
416 |
schedule_preempt_disabled(); |
5bfb5d690
|
417 |
/* Call into cpu_idle with preempt disabled */ |
a1a04ec3c
|
418 |
cpu_startup_entry(CPUHP_ONLINE); |
1df21055e
|
419 |
} |
1da177e4c
|
420 421 |
/* Check for early params. */ |
ecc861705
|
422 423 |
static int __init do_early_param(char *param, char *val, const char *unused, void *arg) |
1da177e4c
|
424 |
{ |
914dcaa84
|
425 |
const struct obs_kernel_param *p; |
1da177e4c
|
426 427 |
for (p = __setup_start; p < __setup_end; p++) { |
b1e4d20cb
|
428 |
if ((p->early && parameq(param, p->str)) || |
18a8bd949
|
429 430 431 |
(strcmp(param, "console") == 0 && strcmp(p->str, "earlycon") == 0) ) { |
1da177e4c
|
432 |
if (p->setup_func(val) != 0) |
ea676e846
|
433 434 |
pr_warn("Malformed early option '%s' ", param); |
1da177e4c
|
435 436 437 438 439 |
} } /* We accept everything at this stage. */ return 0; } |
13977091a
|
440 441 |
void __init parse_early_options(char *cmdline) { |
ecc861705
|
442 443 |
parse_args("early options", cmdline, NULL, 0, 0, 0, NULL, do_early_param); |
13977091a
|
444 |
} |
1da177e4c
|
445 446 447 |
/* Arch code calls this early on, or if not, just before other parsing. */ void __init parse_early_param(void) { |
dd4d9fecb
|
448 449 |
static int done __initdata; static char tmp_cmdline[COMMAND_LINE_SIZE] __initdata; |
1da177e4c
|
450 451 452 453 454 |
if (done) return; /* All fall through to do_early_param. */ |
30d7e0d46
|
455 |
strlcpy(tmp_cmdline, boot_command_line, COMMAND_LINE_SIZE); |
13977091a
|
456 |
parse_early_options(tmp_cmdline); |
1da177e4c
|
457 458 |
done = 1; } |
e7ff3a476
|
459 |
void __init __weak arch_post_acpi_subsys_init(void) { } |
839ad62e7
|
460 |
void __init __weak smp_setup_processor_id(void) |
033ab7f8e
|
461 462 |
{ } |
eded09ccf
|
463 |
# if THREAD_SIZE >= PAGE_SIZE |
b235beea9
|
464 |
void __init __weak thread_stack_cache_init(void) |
8c9843e57
|
465 466 |
{ } |
eded09ccf
|
467 |
#endif |
8c9843e57
|
468 |
|
c7753208a
|
469 |
void __init __weak mem_encrypt_init(void) { } |
444f478f6
|
470 471 472 473 474 |
/* * Set up kernel memory allocators */ static void __init mm_init(void) { |
eefa864b7
|
475 476 477 478 479 |
/* * page_ext requires contiguous pages, * bigger than MAX_ORDER unless SPARSEMEM. */ page_ext_init_flatmem(); |
444f478f6
|
480 481 |
mem_init(); kmem_cache_init(); |
b35f1819a
|
482 |
pgtable_init(); |
444f478f6
|
483 |
vmalloc_init(); |
0ddab1d2e
|
484 |
ioremap_huge_init(); |
763f7eaf6
|
485 486 |
/* Should be run before the first non-init thread is created */ init_espfix_bsp(); |
a4b07fb4e
|
487 488 |
/* Should be run after espfix64 is set up. */ pti_init(); |
444f478f6
|
489 |
} |
722a9f929
|
490 |
asmlinkage __visible void __init start_kernel(void) |
1da177e4c
|
491 |
{ |
dd4d9fecb
|
492 493 |
char *command_line; char *after_dashes; |
033ab7f8e
|
494 |
|
d4311ff1a
|
495 |
set_task_stack_end_magic(&init_task); |
73839c5b2
|
496 |
smp_setup_processor_id(); |
3ac7fe5a4
|
497 |
debug_objects_early_init(); |
420594296
|
498 |
|
ddbcc7e8e
|
499 |
cgroup_init_early(); |
fbb9ce953
|
500 501 |
local_irq_disable(); |
2ce802f62
|
502 |
early_boot_irqs_disabled = true; |
fbb9ce953
|
503 |
|
1b3b3b49b
|
504 505 506 507 |
/* * Interrupts are still disabled. Do necessary setups, then * enable them. */ |
44fd22992
|
508 |
boot_cpu_init(); |
1da177e4c
|
509 |
page_address_init(); |
ea676e846
|
510 |
pr_notice("%s", linux_banner); |
1da177e4c
|
511 |
setup_arch(&command_line); |
121388a31
|
512 513 |
/* * Set up the the initial canary and entropy after arch |
33d72f382
|
514 |
* and after adding latent and command line entropy. |
121388a31
|
515 516 |
*/ add_latent_entropy(); |
33d72f382
|
517 |
add_device_randomness(command_line, strlen(command_line)); |
121388a31
|
518 |
boot_init_stack_canary(); |
6345d24da
|
519 |
mm_init_cpumask(&init_mm); |
30d7e0d46
|
520 |
setup_command_line(command_line); |
e0982e90c
|
521 |
setup_nr_cpu_ids(); |
d6647bdf9
|
522 |
setup_per_cpu_areas(); |
44fd22992
|
523 |
smp_prepare_boot_cpu(); /* arch-specific boot-cpu hooks */ |
b7722f4ac
|
524 |
boot_cpu_hotplug_init(); |
1da177e4c
|
525 |
|
72675e131
|
526 |
build_all_zonelists(NULL); |
83b519e8b
|
527 |
page_alloc_init(); |
ea676e846
|
528 529 |
pr_notice("Kernel command line: %s ", boot_command_line); |
83b519e8b
|
530 |
parse_early_param(); |
51e158c12
|
531 532 533 |
after_dashes = parse_args("Booting kernel", static_command_line, __start___param, __stop___param - __start___param, |
ecc861705
|
534 |
-1, -1, NULL, &unknown_bootoption); |
3438cf549
|
535 |
if (!IS_ERR_OR_NULL(after_dashes)) |
51e158c12
|
536 |
parse_args("Setting init args", after_dashes, NULL, 0, -1, -1, |
ecc861705
|
537 |
NULL, set_init_arg); |
97ce2c88f
|
538 539 |
jump_label_init(); |
83b519e8b
|
540 541 542 543 |
/* * These use large bootmem allocations and must precede * kmem_cache_init() */ |
162a7e750
|
544 |
setup_log_buf(0); |
83b519e8b
|
545 |
pidhash_init(); |
83b519e8b
|
546 547 548 |
vfs_caches_init_early(); sort_main_extable(); trap_init(); |
444f478f6
|
549 |
mm_init(); |
de03c72cf
|
550 |
|
f631718de
|
551 |
ftrace_init(); |
e725c731e
|
552 553 |
/* trace_printk can be enabled here */ early_trace_init(); |
1da177e4c
|
554 555 556 557 558 559 560 561 562 563 564 |
/* * Set up the scheduler prior starting any interrupts (such as the * timer interrupt). Full topology setup happens at smp_init() * time - but meanwhile we still have a functioning scheduler. */ sched_init(); /* * Disable preemption - early bootup scheduling is extremely * fragile until we cpu_idle() for the first time. */ preempt_disable(); |
dd4d9fecb
|
565 566 567 |
if (WARN(!irqs_disabled(), "Interrupts were enabled *very* early, fixing it ")) |
c4a68306b
|
568 |
local_irq_disable(); |
0a835c4f0
|
569 |
radix_tree_init(); |
3347fa092
|
570 571 572 573 574 575 576 |
/* * Allow workqueue creation and work item queueing/cancelling * early. Work item execution depends on kthreads and starts after * workqueue_init(). */ workqueue_init_early(); |
1da177e4c
|
577 |
rcu_init(); |
5f893b263
|
578 |
|
e725c731e
|
579 |
/* Trace events are available after this */ |
5f893b263
|
580 |
trace_init(); |
65f382fd0
|
581 |
context_tracking_init(); |
0b8f1efad
|
582 583 |
/* init some links before init_ISA_irqs() */ early_irq_init(); |
1da177e4c
|
584 |
init_IRQ(); |
ad2b13536
|
585 |
tick_init(); |
d6dd50e07
|
586 |
rcu_init_nohz(); |
1da177e4c
|
587 |
init_timers(); |
c0a313296
|
588 |
hrtimers_init(); |
1da177e4c
|
589 |
softirq_init(); |
ad596171e
|
590 |
timekeeping_init(); |
88fecaa27
|
591 |
time_init(); |
38ff87f77
|
592 |
sched_clock_postinit(); |
f92bac3b1
|
593 |
printk_safe_init(); |
9e6302056
|
594 |
perf_event_init(); |
93e028148
|
595 |
profile_init(); |
d8ad7d112
|
596 |
call_function_init(); |
f91eb62f7
|
597 598 |
WARN(!irqs_disabled(), "Interrupts were enabled early "); |
2ce802f62
|
599 |
early_boot_irqs_disabled = false; |
93e028148
|
600 |
local_irq_enable(); |
dcce284a2
|
601 |
|
7e85ee0c1
|
602 |
kmem_cache_init_late(); |
1da177e4c
|
603 604 605 606 607 608 609 610 |
/* * HACK ALERT! This is early. We're enabling the console before * we've done PCI setups etc, and console_init() must be aware of * this. But we do want output early, in case something goes wrong. */ console_init(); if (panic_later) |
499a4584d
|
611 612 |
panic("Too many boot %s vars at `%s'", panic_later, panic_param); |
fbb9ce953
|
613 614 |
lockdep_info(); |
9a11b49a8
|
615 616 617 618 619 620 |
/* * Need to run this when irqs are enabled, because it wants * to self-test [hard/soft]-irqs on/off lock inversion bugs * too: */ locking_selftest(); |
c7753208a
|
621 622 623 624 625 626 627 |
/* * This needs to be called before any devices perform DMA * operations that might use the SWIOTLB bounce buffers. It will * mark the bounce buffers as decrypted so that their usage will * not cause "plain-text" data to be decrypted when accessed. */ mem_encrypt_init(); |
1da177e4c
|
628 629 |
#ifdef CONFIG_BLK_DEV_INITRD if (initrd_start && !initrd_below_start_ok && |
bd673c7c3
|
630 |
page_to_pfn(virt_to_page((void *)initrd_start)) < min_low_pfn) { |
ea676e846
|
631 632 |
pr_crit("initrd overwritten (0x%08lx < 0x%08lx) - disabling it. ", |
bd673c7c3
|
633 634 |
page_to_pfn(virt_to_page((void *)initrd_start)), min_low_pfn); |
1da177e4c
|
635 636 637 |
initrd_start = 0; } #endif |
fe53ca542
|
638 |
page_ext_init(); |
9b090f2da
|
639 |
kmemleak_init(); |
caba4cbbd
|
640 |
debug_objects_mem_init(); |
e7c8d5c99
|
641 |
setup_per_cpu_pageset(); |
1da177e4c
|
642 643 644 645 646 |
numa_policy_init(); if (late_time_init) late_time_init(); calibrate_delay(); pidmap_init(); |
1da177e4c
|
647 |
anon_vma_init(); |
c4e1acbb3
|
648 |
acpi_early_init(); |
11520e5e7
|
649 |
#ifdef CONFIG_X86 |
83e681897
|
650 |
if (efi_enabled(EFI_RUNTIME_SERVICES)) |
11520e5e7
|
651 652 |
efi_enter_virtual_mode(); #endif |
b235beea9
|
653 |
thread_stack_cache_init(); |
d84f4f992
|
654 |
cred_init(); |
ff691f6e0
|
655 |
fork_init(); |
1da177e4c
|
656 657 |
proc_caches_init(); buffer_init(); |
1da177e4c
|
658 659 |
key_init(); security_init(); |
0b4b3827d
|
660 |
dbg_late_init(); |
4248b0da4
|
661 |
vfs_caches_init(); |
629060270
|
662 |
pagecache_init(); |
1da177e4c
|
663 |
signals_init(); |
1da177e4c
|
664 |
proc_root_init(); |
e149ed2b8
|
665 |
nsfs_init(); |
1da177e4c
|
666 |
cpuset_init(); |
695df2132
|
667 |
cgroup_init(); |
c757249af
|
668 |
taskstats_init_early(); |
ca74e92b4
|
669 |
delayacct_init(); |
1da177e4c
|
670 671 |
check_bugs(); |
b064a8fa7
|
672 |
acpi_subsystem_init(); |
e7ff3a476
|
673 |
arch_post_acpi_subsys_init(); |
6ae6996a4
|
674 |
sfi_init_late(); |
1da177e4c
|
675 |
|
83e681897
|
676 |
if (efi_enabled(EFI_RUNTIME_SERVICES)) { |
785107923
|
677 |
efi_free_boot_services(); |
2223af389
|
678 |
} |
785107923
|
679 |
|
1da177e4c
|
680 681 682 |
/* Do the rest non-__init'ed, we're now alive */ rest_init(); } |
b99b87f70
|
683 684 685 686 |
/* Call all constructor functions linked into the kernel. */ static void __init do_ctors(void) { #ifdef CONFIG_CONSTRUCTORS |
196a15b4e
|
687 |
ctor_fn_t *fn = (ctor_fn_t *) __ctors_start; |
b99b87f70
|
688 |
|
196a15b4e
|
689 690 |
for (; fn < (ctor_fn_t *) __ctors_end; fn++) (*fn)(); |
b99b87f70
|
691 692 |
#endif } |
2329abfa3
|
693 |
bool initcall_debug; |
d0ea3d7d2
|
694 |
core_param(initcall_debug, initcall_debug, bool, 0644); |
1da177e4c
|
695 |
|
7b0b73d76
|
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 |
#ifdef CONFIG_KALLSYMS struct blacklist_entry { struct list_head next; char *buf; }; static __initdata_or_module LIST_HEAD(blacklisted_initcalls); static int __init initcall_blacklist(char *str) { char *str_entry; struct blacklist_entry *entry; /* str argument is a comma-separated list of functions */ do { str_entry = strsep(&str, ","); if (str_entry) { pr_debug("blacklisting initcall %s ", str_entry); entry = alloc_bootmem(sizeof(*entry)); entry->buf = alloc_bootmem(strlen(str_entry) + 1); strcpy(entry->buf, str_entry); list_add(&entry->next, &blacklisted_initcalls); } } while (str_entry); return 0; } static bool __init_or_module initcall_blacklisted(initcall_t fn) { |
7b0b73d76
|
727 |
struct blacklist_entry *entry; |
c8cdd2be2
|
728 |
char fn_name[KSYM_SYMBOL_LEN]; |
0fd5ed8d8
|
729 |
unsigned long addr; |
7b0b73d76
|
730 |
|
c8cdd2be2
|
731 |
if (list_empty(&blacklisted_initcalls)) |
7b0b73d76
|
732 |
return false; |
0fd5ed8d8
|
733 734 |
addr = (unsigned long) dereference_function_descriptor(fn); sprint_symbol_no_offset(fn_name, addr); |
c8cdd2be2
|
735 |
|
841c06d71
|
736 737 738 739 740 |
/* * fn will be "function_name [module_name]" where [module_name] is not * displayed for built-in init functions. Strip off the [module_name]. */ strreplace(fn_name, ' ', '\0'); |
e6fd1fb3b
|
741 |
list_for_each_entry(entry, &blacklisted_initcalls, next) { |
7b0b73d76
|
742 743 744 |
if (!strcmp(fn_name, entry->buf)) { pr_debug("initcall %s blacklisted ", fn_name); |
7b0b73d76
|
745 746 747 |
return true; } } |
7b0b73d76
|
748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 |
return false; } #else static int __init initcall_blacklist(char *str) { pr_warn("initcall_blacklist requires CONFIG_KALLSYMS "); return 0; } static bool __init_or_module initcall_blacklisted(initcall_t fn) { return false; } #endif __setup("initcall_blacklist=", initcall_blacklist); |
e44612713
|
764 |
static int __init_or_module do_one_initcall_debug(initcall_t fn) |
1da177e4c
|
765 |
{ |
742390728
|
766 |
ktime_t calltime, delta, rettime; |
30dbb20e6
|
767 768 |
unsigned long long duration; int ret; |
1da177e4c
|
769 |
|
d62cf8152
|
770 771 |
printk(KERN_DEBUG "calling %pF @ %i ", fn, task_pid_nr(current)); |
22c5c03b4
|
772 |
calltime = ktime_get(); |
30dbb20e6
|
773 |
ret = fn(); |
22c5c03b4
|
774 775 776 |
rettime = ktime_get(); delta = ktime_sub(rettime, calltime); duration = (unsigned long long) ktime_to_ns(delta) >> 10; |
d62cf8152
|
777 778 |
printk(KERN_DEBUG "initcall %pF returned %d after %lld usecs ", |
ea676e846
|
779 |
fn, ret, duration); |
1da177e4c
|
780 |
|
22c5c03b4
|
781 782 |
return ret; } |
e44612713
|
783 |
int __init_or_module do_one_initcall(initcall_t fn) |
22c5c03b4
|
784 785 786 |
{ int count = preempt_count(); int ret; |
ff1c8fac8
|
787 |
char msgbuf[64]; |
22c5c03b4
|
788 |
|
7b0b73d76
|
789 790 |
if (initcall_blacklisted(fn)) return -EPERM; |
22c5c03b4
|
791 792 793 794 |
if (initcall_debug) ret = do_one_initcall_debug(fn); else ret = fn(); |
8f0c45cdf
|
795 |
|
e0df154f4
|
796 |
msgbuf[0] = 0; |
e662e1cfd
|
797 |
|
e0df154f4
|
798 |
if (preempt_count() != count) { |
bf5d770bd
|
799 |
sprintf(msgbuf, "preemption imbalance "); |
4a2b4b222
|
800 |
preempt_count_set(count); |
1da177e4c
|
801 |
} |
e0df154f4
|
802 |
if (irqs_disabled()) { |
a76bfd0da
|
803 |
strlcat(msgbuf, "disabled interrupts ", sizeof(msgbuf)); |
e0df154f4
|
804 805 |
local_irq_enable(); } |
f91eb62f7
|
806 807 |
WARN(msgbuf[0], "initcall %pF returned with %s ", fn, msgbuf); |
59f9415ff
|
808 |
|
38addce8b
|
809 |
add_latent_entropy(); |
30dbb20e6
|
810 |
return ret; |
e0df154f4
|
811 |
} |
026cee008
|
812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 |
extern initcall_t __initcall_start[]; extern initcall_t __initcall0_start[]; extern initcall_t __initcall1_start[]; extern initcall_t __initcall2_start[]; extern initcall_t __initcall3_start[]; extern initcall_t __initcall4_start[]; extern initcall_t __initcall5_start[]; extern initcall_t __initcall6_start[]; extern initcall_t __initcall7_start[]; extern initcall_t __initcall_end[]; static initcall_t *initcall_levels[] __initdata = { __initcall0_start, __initcall1_start, __initcall2_start, __initcall3_start, __initcall4_start, __initcall5_start, __initcall6_start, __initcall7_start, __initcall_end, }; |
96263d286
|
834 |
/* Keep these in sync with initcalls in include/linux/init.h */ |
026cee008
|
835 |
static char *initcall_level_names[] __initdata = { |
9fb48c744
|
836 837 838 839 840 841 842 843 |
"early", "core", "postcore", "arch", "subsys", "fs", "device", "late", |
026cee008
|
844 |
}; |
026cee008
|
845 |
static void __init do_initcall_level(int level) |
e0df154f4
|
846 |
{ |
196a15b4e
|
847 |
initcall_t *fn; |
e0df154f4
|
848 |
|
08746a65c
|
849 |
strcpy(initcall_command_line, saved_command_line); |
026cee008
|
850 |
parse_args(initcall_level_names[level], |
08746a65c
|
851 |
initcall_command_line, __start___param, |
026cee008
|
852 853 |
__stop___param - __start___param, level, level, |
ecc861705
|
854 |
NULL, &repair_env_string); |
026cee008
|
855 856 |
for (fn = initcall_levels[level]; fn < initcall_levels[level+1]; fn++) |
196a15b4e
|
857 |
do_one_initcall(*fn); |
1da177e4c
|
858 |
} |
026cee008
|
859 860 861 |
static void __init do_initcalls(void) { int level; |
19efb72fd
|
862 |
for (level = 0; level < ARRAY_SIZE(initcall_levels) - 1; level++) |
026cee008
|
863 864 |
do_initcall_level(level); } |
1da177e4c
|
865 866 867 868 869 870 871 872 873 |
/* * Ok, the machine is now initialized. None of the devices * have been touched yet, but the CPU subsystem is up and * running, and memory and process management works. * * Now we can finally start doing some real work.. */ static void __init do_basic_setup(void) { |
759ee0915
|
874 |
cpuset_init_smp(); |
41ffe5d5c
|
875 |
shmem_init(); |
1da177e4c
|
876 |
driver_init(); |
b04c3afb2
|
877 |
init_irq_proc(); |
b99b87f70
|
878 |
do_ctors(); |
d5767c535
|
879 |
usermodehelper_enable(); |
b0f84374b
|
880 |
do_initcalls(); |
1da177e4c
|
881 |
} |
7babe8db9
|
882 |
static void __init do_pre_smp_initcalls(void) |
c2147a509
|
883 |
{ |
196a15b4e
|
884 |
initcall_t *fn; |
c2147a509
|
885 |
|
026cee008
|
886 |
for (fn = __initcall_start; fn < __initcall0_start; fn++) |
196a15b4e
|
887 |
do_one_initcall(*fn); |
c2147a509
|
888 |
} |
bb813f4c9
|
889 890 891 892 893 894 895 896 897 898 |
/* * This function requests modules which should be loaded by default and is * called twice right after initrd is mounted and right before init is * exec'd. If such modules are on either initrd or rootfs, they will be * loaded before control is passed to userland. */ void __init load_default_modules(void) { load_default_elevator_module(); } |
a74fb73c1
|
899 |
static int run_init_process(const char *init_filename) |
1da177e4c
|
900 901 |
{ argv_init[0] = init_filename; |
c4ad8f98b
|
902 |
return do_execve(getname_kernel(init_filename), |
ae903caae
|
903 904 |
(const char __user *const __user *)argv_init, (const char __user *const __user *)envp_init); |
1da177e4c
|
905 |
} |
ba24762bd
|
906 907 908 909 910 911 912 913 914 915 916 917 918 919 |
static int try_to_run_init_process(const char *init_filename) { int ret; ret = run_init_process(init_filename); if (ret && ret != -ENOENT) { pr_err("Starting init: %s exists but couldn't execute it (error %d) ", init_filename, ret); } return ret; } |
f80b0c904
|
920 |
static noinline void __init kernel_init_freeable(void); |
d6b212380
|
921 |
|
0f5bf6d0a
|
922 |
#if defined(CONFIG_STRICT_KERNEL_RWX) || defined(CONFIG_STRICT_MODULE_RWX) |
39290b389
|
923 |
bool rodata_enabled __ro_after_init = true; |
d2aa1acad
|
924 925 926 927 928 |
static int __init set_debug_rodata(char *str) { return strtobool(str, &rodata_enabled); } __setup("rodata=", set_debug_rodata); |
39290b389
|
929 |
#endif |
d2aa1acad
|
930 |
|
0f5bf6d0a
|
931 |
#ifdef CONFIG_STRICT_KERNEL_RWX |
d2aa1acad
|
932 933 |
static void mark_readonly(void) { |
2959a5f72
|
934 |
if (rodata_enabled) { |
20e557fb2
|
935 936 937 938 939 940 941 |
/* * load_module() results in W+X mappings, which are cleaned up * with call_rcu_sched(). Let's make sure that queued work is * flushed so that we don't hit false positives looking for * insecure pages which are W+X. */ rcu_barrier_sched(); |
d2aa1acad
|
942 |
mark_rodata_ro(); |
2959a5f72
|
943 944 |
rodata_test(); } else |
d2aa1acad
|
945 946 947 948 949 950 951 952 953 954 |
pr_info("Kernel memory protection disabled. "); } #else static inline void mark_readonly(void) { pr_warn("This architecture does not have kernel memory protection. "); } #endif |
d6b212380
|
955 |
static int __ref kernel_init(void *unused) |
ee5bfa642
|
956 |
{ |
ba24762bd
|
957 |
int ret; |
d6b212380
|
958 |
kernel_init_freeable(); |
22a9d6456
|
959 960 |
/* need to finish all async __init code before freeing the memory */ async_synchronize_full(); |
b80f0f6c9
|
961 |
ftrace_free_init_mem(); |
ee5bfa642
|
962 |
free_initmem(); |
d2aa1acad
|
963 |
mark_readonly(); |
ee5bfa642
|
964 965 |
system_state = SYSTEM_RUNNING; numa_default_policy(); |
967dcb8fe
|
966 |
rcu_end_inkernel_boot(); |
ee5bfa642
|
967 |
if (ramdisk_execute_command) { |
ba24762bd
|
968 969 |
ret = run_init_process(ramdisk_execute_command); if (!ret) |
a74fb73c1
|
970 |
return 0; |
ba24762bd
|
971 972 973 |
pr_err("Failed to execute %s (error %d) ", ramdisk_execute_command, ret); |
ee5bfa642
|
974 975 976 977 978 979 980 981 982 |
} /* * We try each of these until one succeeds. * * The Bourne shell can be used instead of init if we are * trying to recover a really broken machine. */ if (execute_command) { |
ba24762bd
|
983 984 |
ret = run_init_process(execute_command); if (!ret) |
a74fb73c1
|
985 |
return 0; |
6ef4536e2
|
986 987 |
panic("Requested init %s failed (error %d).", execute_command, ret); |
ee5bfa642
|
988 |
} |
ba24762bd
|
989 990 991 992 |
if (!try_to_run_init_process("/sbin/init") || !try_to_run_init_process("/etc/init") || !try_to_run_init_process("/bin/init") || !try_to_run_init_process("/bin/sh")) |
a74fb73c1
|
993 |
return 0; |
ee5bfa642
|
994 |
|
ba24762bd
|
995 |
panic("No working init found. Try passing init= option to kernel. " |
8c27ceff3
|
996 |
"See Linux Documentation/admin-guide/init.rst for guidance."); |
ee5bfa642
|
997 |
} |
f80b0c904
|
998 |
static noinline void __init kernel_init_freeable(void) |
1da177e4c
|
999 |
{ |
b433c3d45
|
1000 1001 1002 1003 |
/* * Wait until kthreadd is all set-up. */ wait_for_completion(&kthreadd_done); |
31a67102f
|
1004 1005 1006 |
/* Now the scheduler is fully set up and can do blocking allocations */ gfp_allowed_mask = __GFP_BITS_MASK; |
58568d2a8
|
1007 1008 1009 |
/* * init can allocate pages on any node */ |
3c466d46a
|
1010 |
set_mems_allowed(node_states[N_MEMORY]); |
1da177e4c
|
1011 |
|
9ec52099e
|
1012 |
cad_pid = task_pid(current); |
ca74a6f84
|
1013 |
smp_prepare_cpus(setup_max_cpus); |
1da177e4c
|
1014 |
|
3347fa092
|
1015 |
workqueue_init(); |
597b7305d
|
1016 |
init_mm_internals(); |
1da177e4c
|
1017 |
do_pre_smp_initcalls(); |
004417a6d
|
1018 |
lockup_detector_init(); |
1da177e4c
|
1019 |
|
1da177e4c
|
1020 1021 |
smp_init(); sched_init_smp(); |
0e1cc95b4
|
1022 |
page_alloc_init_late(); |
1da177e4c
|
1023 |
do_basic_setup(); |
2bd3a997b
|
1024 1025 |
/* Open the /dev/console on the rootfs, this should never fail */ if (sys_open((const char __user *) "/dev/console", O_RDWR, 0) < 0) |
ea676e846
|
1026 1027 |
pr_err("Warning: unable to open an initial console. "); |
2bd3a997b
|
1028 1029 1030 |
(void) sys_dup(0); (void) sys_dup(0); |
1da177e4c
|
1031 1032 1033 1034 |
/* * check if there is an early userspace init. If yes, let it do all * the work */ |
ffdfc4097
|
1035 1036 1037 1038 1039 1040 |
if (!ramdisk_execute_command) ramdisk_execute_command = "/init"; if (sys_access((const char __user *) ramdisk_execute_command, 0) != 0) { ramdisk_execute_command = NULL; |
1da177e4c
|
1041 |
prepare_namespace(); |
ffdfc4097
|
1042 |
} |
1da177e4c
|
1043 1044 1045 1046 1047 |
/* * Ok, we have completed the initial bootup, and * we're essentially up and running. Get rid of the * initmem segments and start the user-mode stuff.. |
c9cd2ce2b
|
1048 1049 1050 |
* * rootfs is available now, try loading the public keys * and default modules |
1da177e4c
|
1051 |
*/ |
bb813f4c9
|
1052 |
|
c9cd2ce2b
|
1053 |
integrity_load_keys(); |
bb813f4c9
|
1054 |
load_default_modules(); |
1da177e4c
|
1055 |
} |