1/*
2 *  linux/init/main.c
3 *
4 *  Copyright (C) 1991, 1992  Linus Torvalds
5 *
6 *  GK 2/5/95  -  Changed to support mounting root fs via NFS
7 *  Added initrd & change_root: Werner Almesberger & Hans Lermen, Feb '96
8 *  Moan early if gcc is old, avoiding bogus kernels - Paul Gortmaker, May '96
9 *  Simplified starting of init:  Michael A. Griffith <grif@acm.org>
10 */
11
12#define DEBUG		/* Enable initcall_debug */
13
14#include <linux/types.h>
15#include <linux/module.h>
16#include <linux/proc_fs.h>
17#include <linux/kernel.h>
18#include <linux/syscalls.h>
19#include <linux/stackprotector.h>
20#include <linux/string.h>
21#include <linux/ctype.h>
22#include <linux/delay.h>
23#include <linux/ioport.h>
24#include <linux/init.h>
25#include <linux/initrd.h>
26#include <linux/bootmem.h>
27#include <linux/acpi.h>
28#include <linux/tty.h>
29#include <linux/percpu.h>
30#include <linux/kmod.h>
31#include <linux/vmalloc.h>
32#include <linux/kernel_stat.h>
33#include <linux/start_kernel.h>
34#include <linux/security.h>
35#include <linux/smp.h>
36#include <linux/profile.h>
37#include <linux/rcupdate.h>
38#include <linux/moduleparam.h>
39#include <linux/kallsyms.h>
40#include <linux/writeback.h>
41#include <linux/cpu.h>
42#include <linux/cpuset.h>
43#include <linux/cgroup.h>
44#include <linux/efi.h>
45#include <linux/tick.h>
46#include <linux/interrupt.h>
47#include <linux/taskstats_kern.h>
48#include <linux/delayacct.h>
49#include <linux/unistd.h>
50#include <linux/rmap.h>
51#include <linux/mempolicy.h>
52#include <linux/key.h>
53#include <linux/buffer_head.h>
54#include <linux/page_cgroup.h>
55#include <linux/debug_locks.h>
56#include <linux/debugobjects.h>
57#include <linux/lockdep.h>
58#include <linux/kmemleak.h>
59#include <linux/pid_namespace.h>
60#include <linux/device.h>
61#include <linux/kthread.h>
62#include <linux/sched.h>
63#include <linux/signal.h>
64#include <linux/idr.h>
65#include <linux/kgdb.h>
66#include <linux/ftrace.h>
67#include <linux/async.h>
68#include <linux/kmemcheck.h>
69#include <linux/sfi.h>
70#include <linux/shmem_fs.h>
71#include <linux/slab.h>
72#include <linux/perf_event.h>
73#include <linux/file.h>
74#include <linux/ptrace.h>
75#include <linux/blkdev.h>
76#include <linux/elevator.h>
77#include <linux/sched_clock.h>
78#include <linux/context_tracking.h>
79#include <linux/random.h>
80#include <linux/list.h>
81
82#include <asm/io.h>
83#include <asm/bugs.h>
84#include <asm/setup.h>
85#include <asm/sections.h>
86#include <asm/cacheflush.h>
87
88#ifdef CONFIG_X86_LOCAL_APIC
89#include <asm/smp.h>
90#endif
91
92static int kernel_init(void *);
93
94extern void init_IRQ(void);
95extern void fork_init(unsigned long);
96extern void radix_tree_init(void);
97#ifndef CONFIG_DEBUG_RODATA
98static inline void mark_rodata_ro(void) { }
99#endif
100
101/*
102 * Debug helper: via this flag we know that we are in 'early bootup code'
103 * where only the boot processor is running with IRQ disabled.  This means
104 * two things - IRQ must not be enabled before the flag is cleared and some
105 * operations which are not allowed with IRQ disabled are allowed while the
106 * flag is set.
107 */
108bool early_boot_irqs_disabled __read_mostly;
109
110enum system_states system_state __read_mostly;
111EXPORT_SYMBOL(system_state);
112
113/*
114 * Boot command-line arguments
115 */
116#define MAX_INIT_ARGS CONFIG_INIT_ENV_ARG_LIMIT
117#define MAX_INIT_ENVS CONFIG_INIT_ENV_ARG_LIMIT
118
119extern void time_init(void);
120/* Default late time init is NULL. archs can override this later. */
121void (*__initdata late_time_init)(void);
122
123/* Untouched command line saved by arch-specific code. */
124char __initdata boot_command_line[COMMAND_LINE_SIZE];
125/* Untouched saved command line (eg. for /proc) */
126char *saved_command_line;
127/* Command line for parameter parsing */
128static char *static_command_line;
129/* Command line for per-initcall parameter parsing */
130static char *initcall_command_line;
131
132static char *execute_command;
133static char *ramdisk_execute_command;
134
135/*
136 * Used to generate warnings if static_key manipulation functions are used
137 * before jump_label_init is called.
138 */
139bool static_key_initialized __read_mostly;
140EXPORT_SYMBOL_GPL(static_key_initialized);
141
142/*
143 * If set, this is an indication to the drivers that reset the underlying
144 * device before going ahead with the initialization otherwise driver might
145 * rely on the BIOS and skip the reset operation.
146 *
147 * This is useful if kernel is booting in an unreliable environment.
148 * For ex. kdump situaiton where previous kernel has crashed, BIOS has been
149 * skipped and devices will be in unknown state.
150 */
151unsigned int reset_devices;
152EXPORT_SYMBOL(reset_devices);
153
154static int __init set_reset_devices(char *str)
155{
156	reset_devices = 1;
157	return 1;
158}
159
160__setup("reset_devices", set_reset_devices);
161
162static const char *argv_init[MAX_INIT_ARGS+2] = { "init", NULL, };
163const char *envp_init[MAX_INIT_ENVS+2] = { "HOME=/", "TERM=linux", NULL, };
164static const char *panic_later, *panic_param;
165
166extern const struct obs_kernel_param __setup_start[], __setup_end[];
167
168static int __init obsolete_checksetup(char *line)
169{
170	const struct obs_kernel_param *p;
171	int had_early_param = 0;
172
173	p = __setup_start;
174	do {
175		int n = strlen(p->str);
176		if (parameqn(line, p->str, n)) {
177			if (p->early) {
178				/* Already done in parse_early_param?
179				 * (Needs exact match on param part).
180				 * Keep iterating, as we can have early
181				 * params and __setups of same names 8( */
182				if (line[n] == '\0' || line[n] == '=')
183					had_early_param = 1;
184			} else if (!p->setup_func) {
185				pr_warn("Parameter %s is obsolete, ignored\n",
186					p->str);
187				return 1;
188			} else if (p->setup_func(line + n))
189				return 1;
190		}
191		p++;
192	} while (p < __setup_end);
193
194	return had_early_param;
195}
196
197/*
198 * This should be approx 2 Bo*oMips to start (note initial shift), and will
199 * still work even if initially too large, it will just take slightly longer
200 */
201unsigned long loops_per_jiffy = (1<<12);
202EXPORT_SYMBOL(loops_per_jiffy);
203
204static int __init debug_kernel(char *str)
205{
206	console_loglevel = CONSOLE_LOGLEVEL_DEBUG;
207	return 0;
208}
209
210static int __init quiet_kernel(char *str)
211{
212	console_loglevel = CONSOLE_LOGLEVEL_QUIET;
213	return 0;
214}
215
216early_param("debug", debug_kernel);
217early_param("quiet", quiet_kernel);
218
219static int __init loglevel(char *str)
220{
221	int newlevel;
222
223	/*
224	 * Only update loglevel value when a correct setting was passed,
225	 * to prevent blind crashes (when loglevel being set to 0) that
226	 * are quite hard to debug
227	 */
228	if (get_option(&str, &newlevel)) {
229		console_loglevel = newlevel;
230		return 0;
231	}
232
233	return -EINVAL;
234}
235
236early_param("loglevel", loglevel);
237
238/* Change NUL term back to "=", to make "param" the whole string. */
239static int __init repair_env_string(char *param, char *val, const char *unused)
240{
241	if (val) {
242		/* param=val or param="val"? */
243		if (val == param+strlen(param)+1)
244			val[-1] = '=';
245		else if (val == param+strlen(param)+2) {
246			val[-2] = '=';
247			memmove(val-1, val, strlen(val)+1);
248			val--;
249		} else
250			BUG();
251	}
252	return 0;
253}
254
255/* Anything after -- gets handed straight to init. */
256static int __init set_init_arg(char *param, char *val, const char *unused)
257{
258	unsigned int i;
259
260	if (panic_later)
261		return 0;
262
263	repair_env_string(param, val, unused);
264
265	for (i = 0; argv_init[i]; i++) {
266		if (i == MAX_INIT_ARGS) {
267			panic_later = "init";
268			panic_param = param;
269			return 0;
270		}
271	}
272	argv_init[i] = param;
273	return 0;
274}
275
276/*
277 * Unknown boot options get handed to init, unless they look like
278 * unused parameters (modprobe will find them in /proc/cmdline).
279 */
280static int __init unknown_bootoption(char *param, char *val, const char *unused)
281{
282	repair_env_string(param, val, unused);
283
284	/* Handle obsolete-style parameters */
285	if (obsolete_checksetup(param))
286		return 0;
287
288	/* Unused module parameter. */
289	if (strchr(param, '.') && (!val || strchr(param, '.') < val))
290		return 0;
291
292	if (panic_later)
293		return 0;
294
295	if (val) {
296		/* Environment option */
297		unsigned int i;
298		for (i = 0; envp_init[i]; i++) {
299			if (i == MAX_INIT_ENVS) {
300				panic_later = "env";
301				panic_param = param;
302			}
303			if (!strncmp(param, envp_init[i], val - param))
304				break;
305		}
306		envp_init[i] = param;
307	} else {
308		/* Command line option */
309		unsigned int i;
310		for (i = 0; argv_init[i]; i++) {
311			if (i == MAX_INIT_ARGS) {
312				panic_later = "init";
313				panic_param = param;
314			}
315		}
316		argv_init[i] = param;
317	}
318	return 0;
319}
320
321static int __init init_setup(char *str)
322{
323	unsigned int i;
324
325	execute_command = str;
326	/*
327	 * In case LILO is going to boot us with default command line,
328	 * it prepends "auto" before the whole cmdline which makes
329	 * the shell think it should execute a script with such name.
330	 * So we ignore all arguments entered _before_ init=... [MJ]
331	 */
332	for (i = 1; i < MAX_INIT_ARGS; i++)
333		argv_init[i] = NULL;
334	return 1;
335}
336__setup("init=", init_setup);
337
338static int __init rdinit_setup(char *str)
339{
340	unsigned int i;
341
342	ramdisk_execute_command = str;
343	/* See "auto" comment in init_setup */
344	for (i = 1; i < MAX_INIT_ARGS; i++)
345		argv_init[i] = NULL;
346	return 1;
347}
348__setup("rdinit=", rdinit_setup);
349
350#ifndef CONFIG_SMP
351static const unsigned int setup_max_cpus = NR_CPUS;
352#ifdef CONFIG_X86_LOCAL_APIC
353static void __init smp_init(void)
354{
355	APIC_init_uniprocessor();
356}
357#else
358#define smp_init()	do { } while (0)
359#endif
360
361static inline void setup_nr_cpu_ids(void) { }
362static inline void smp_prepare_cpus(unsigned int maxcpus) { }
363#endif
364
365/*
366 * We need to store the untouched command line for future reference.
367 * We also need to store the touched command line since the parameter
368 * parsing is performed in place, and we should allow a component to
369 * store reference of name/value for future reference.
370 */
371static void __init setup_command_line(char *command_line)
372{
373	saved_command_line =
374		memblock_virt_alloc(strlen(boot_command_line) + 1, 0);
375	initcall_command_line =
376		memblock_virt_alloc(strlen(boot_command_line) + 1, 0);
377	static_command_line = memblock_virt_alloc(strlen(command_line) + 1, 0);
378	strcpy(saved_command_line, boot_command_line);
379	strcpy(static_command_line, command_line);
380}
381
382/*
383 * We need to finalize in a non-__init function or else race conditions
384 * between the root thread and the init thread may cause start_kernel to
385 * be reaped by free_initmem before the root thread has proceeded to
386 * cpu_idle.
387 *
388 * gcc-3.4 accidentally inlines this function, so use noinline.
389 */
390
391static __initdata DECLARE_COMPLETION(kthreadd_done);
392
393static noinline void __init_refok rest_init(void)
394{
395	int pid;
396
397	rcu_scheduler_starting();
398	/*
399	 * We need to spawn init first so that it obtains pid 1, however
400	 * the init task will end up wanting to create kthreads, which, if
401	 * we schedule it before we create kthreadd, will OOPS.
402	 */
403	kernel_thread(kernel_init, NULL, CLONE_FS);
404	numa_default_policy();
405	pid = kernel_thread(kthreadd, NULL, CLONE_FS | CLONE_FILES);
406	rcu_read_lock();
407	kthreadd_task = find_task_by_pid_ns(pid, &init_pid_ns);
408	rcu_read_unlock();
409	complete(&kthreadd_done);
410
411	/*
412	 * The boot idle thread must execute schedule()
413	 * at least once to get things moving:
414	 */
415	init_idle_bootup_task(current);
416	schedule_preempt_disabled();
417	/* Call into cpu_idle with preempt disabled */
418	cpu_startup_entry(CPUHP_ONLINE);
419}
420
421/* Check for early params. */
422static int __init do_early_param(char *param, char *val, const char *unused)
423{
424	const struct obs_kernel_param *p;
425
426	for (p = __setup_start; p < __setup_end; p++) {
427		if ((p->early && parameq(param, p->str)) ||
428		    (strcmp(param, "console") == 0 &&
429		     strcmp(p->str, "earlycon") == 0)
430		) {
431			if (p->setup_func(val) != 0)
432				pr_warn("Malformed early option '%s'\n", param);
433		}
434	}
435	/* We accept everything at this stage. */
436	return 0;
437}
438
439void __init parse_early_options(char *cmdline)
440{
441	parse_args("early options", cmdline, NULL, 0, 0, 0, do_early_param);
442}
443
444/* Arch code calls this early on, or if not, just before other parsing. */
445void __init parse_early_param(void)
446{
447	static int done __initdata;
448	static char tmp_cmdline[COMMAND_LINE_SIZE] __initdata;
449
450	if (done)
451		return;
452
453	/* All fall through to do_early_param. */
454	strlcpy(tmp_cmdline, boot_command_line, COMMAND_LINE_SIZE);
455	parse_early_options(tmp_cmdline);
456	done = 1;
457}
458
459/*
460 *	Activate the first processor.
461 */
462
463static void __init boot_cpu_init(void)
464{
465	int cpu = smp_processor_id();
466	/* Mark the boot cpu "present", "online" etc for SMP and UP case */
467	set_cpu_online(cpu, true);
468	set_cpu_active(cpu, true);
469	set_cpu_present(cpu, true);
470	set_cpu_possible(cpu, true);
471}
472
473void __init __weak smp_setup_processor_id(void)
474{
475}
476
477# if THREAD_SIZE >= PAGE_SIZE
478void __init __weak thread_info_cache_init(void)
479{
480}
481#endif
482
483/*
484 * Set up kernel memory allocators
485 */
486static void __init mm_init(void)
487{
488	/*
489	 * page_cgroup requires contiguous pages,
490	 * bigger than MAX_ORDER unless SPARSEMEM.
491	 */
492	page_cgroup_init_flatmem();
493	mem_init();
494	kmem_cache_init();
495	percpu_init_late();
496	pgtable_init();
497	vmalloc_init();
498}
499
500asmlinkage __visible void __init start_kernel(void)
501{
502	char *command_line;
503	char *after_dashes;
504
505	/*
506	 * Need to run as early as possible, to initialize the
507	 * lockdep hash:
508	 */
509	lockdep_init();
510	set_task_stack_end_magic(&init_task);
511	smp_setup_processor_id();
512	debug_objects_early_init();
513
514	/*
515	 * Set up the the initial canary ASAP:
516	 */
517	boot_init_stack_canary();
518
519	cgroup_init_early();
520
521	local_irq_disable();
522	early_boot_irqs_disabled = true;
523
524/*
525 * Interrupts are still disabled. Do necessary setups, then
526 * enable them
527 */
528	boot_cpu_init();
529	page_address_init();
530	pr_notice("%s", linux_banner);
531	setup_arch(&command_line);
532	mm_init_cpumask(&init_mm);
533	setup_command_line(command_line);
534	setup_nr_cpu_ids();
535	setup_per_cpu_areas();
536	smp_prepare_boot_cpu();	/* arch-specific boot-cpu hooks */
537
538	build_all_zonelists(NULL, NULL);
539	page_alloc_init();
540
541	pr_notice("Kernel command line: %s\n", boot_command_line);
542	parse_early_param();
543	after_dashes = parse_args("Booting kernel",
544				  static_command_line, __start___param,
545				  __stop___param - __start___param,
546				  -1, -1, &unknown_bootoption);
547	if (!IS_ERR_OR_NULL(after_dashes))
548		parse_args("Setting init args", after_dashes, NULL, 0, -1, -1,
549			   set_init_arg);
550
551	jump_label_init();
552
553	/*
554	 * These use large bootmem allocations and must precede
555	 * kmem_cache_init()
556	 */
557	setup_log_buf(0);
558	pidhash_init();
559	vfs_caches_init_early();
560	sort_main_extable();
561	trap_init();
562	mm_init();
563
564	/*
565	 * Set up the scheduler prior starting any interrupts (such as the
566	 * timer interrupt). Full topology setup happens at smp_init()
567	 * time - but meanwhile we still have a functioning scheduler.
568	 */
569	sched_init();
570	/*
571	 * Disable preemption - early bootup scheduling is extremely
572	 * fragile until we cpu_idle() for the first time.
573	 */
574	preempt_disable();
575	if (WARN(!irqs_disabled(),
576		 "Interrupts were enabled *very* early, fixing it\n"))
577		local_irq_disable();
578	idr_init_cache();
579	rcu_init();
580	context_tracking_init();
581	radix_tree_init();
582	/* init some links before init_ISA_irqs() */
583	early_irq_init();
584	init_IRQ();
585	tick_init();
586	rcu_init_nohz();
587	init_timers();
588	hrtimers_init();
589	softirq_init();
590	timekeeping_init();
591	time_init();
592	sched_clock_postinit();
593	perf_event_init();
594	profile_init();
595	call_function_init();
596	WARN(!irqs_disabled(), "Interrupts were enabled early\n");
597	early_boot_irqs_disabled = false;
598	local_irq_enable();
599
600	kmem_cache_init_late();
601
602	/*
603	 * HACK ALERT! This is early. We're enabling the console before
604	 * we've done PCI setups etc, and console_init() must be aware of
605	 * this. But we do want output early, in case something goes wrong.
606	 */
607	console_init();
608	if (panic_later)
609		panic("Too many boot %s vars at `%s'", panic_later,
610		      panic_param);
611
612	lockdep_info();
613
614	/*
615	 * Need to run this when irqs are enabled, because it wants
616	 * to self-test [hard/soft]-irqs on/off lock inversion bugs
617	 * too:
618	 */
619	locking_selftest();
620
621#ifdef CONFIG_BLK_DEV_INITRD
622	if (initrd_start && !initrd_below_start_ok &&
623	    page_to_pfn(virt_to_page((void *)initrd_start)) < min_low_pfn) {
624		pr_crit("initrd overwritten (0x%08lx < 0x%08lx) - disabling it.\n",
625		    page_to_pfn(virt_to_page((void *)initrd_start)),
626		    min_low_pfn);
627		initrd_start = 0;
628	}
629#endif
630	page_cgroup_init();
631	debug_objects_mem_init();
632	kmemleak_init();
633	setup_per_cpu_pageset();
634	numa_policy_init();
635	if (late_time_init)
636		late_time_init();
637	sched_clock_init();
638	calibrate_delay();
639	pidmap_init();
640	anon_vma_init();
641	acpi_early_init();
642#ifdef CONFIG_X86
643	if (efi_enabled(EFI_RUNTIME_SERVICES))
644		efi_enter_virtual_mode();
645#endif
646#ifdef CONFIG_X86_ESPFIX64
647	/* Should be run before the first non-init thread is created */
648	init_espfix_bsp();
649#endif
650	thread_info_cache_init();
651	cred_init();
652	fork_init(totalram_pages);
653	proc_caches_init();
654	buffer_init();
655	key_init();
656	security_init();
657	dbg_late_init();
658	vfs_caches_init(totalram_pages);
659	signals_init();
660	/* rootfs populating might need page-writeback */
661	page_writeback_init();
662	proc_root_init();
663	cgroup_init();
664	cpuset_init();
665	taskstats_init_early();
666	delayacct_init();
667
668	check_bugs();
669
670	sfi_init_late();
671
672	if (efi_enabled(EFI_RUNTIME_SERVICES)) {
673		efi_late_init();
674		efi_free_boot_services();
675	}
676
677	ftrace_init();
678
679	/* Do the rest non-__init'ed, we're now alive */
680	rest_init();
681}
682
683/* Call all constructor functions linked into the kernel. */
684static void __init do_ctors(void)
685{
686#ifdef CONFIG_CONSTRUCTORS
687	ctor_fn_t *fn = (ctor_fn_t *) __ctors_start;
688
689	for (; fn < (ctor_fn_t *) __ctors_end; fn++)
690		(*fn)();
691#endif
692}
693
694bool initcall_debug;
695core_param(initcall_debug, initcall_debug, bool, 0644);
696
697#ifdef CONFIG_KALLSYMS
698struct blacklist_entry {
699	struct list_head next;
700	char *buf;
701};
702
703static __initdata_or_module LIST_HEAD(blacklisted_initcalls);
704
705static int __init initcall_blacklist(char *str)
706{
707	char *str_entry;
708	struct blacklist_entry *entry;
709
710	/* str argument is a comma-separated list of functions */
711	do {
712		str_entry = strsep(&str, ",");
713		if (str_entry) {
714			pr_debug("blacklisting initcall %s\n", str_entry);
715			entry = alloc_bootmem(sizeof(*entry));
716			entry->buf = alloc_bootmem(strlen(str_entry) + 1);
717			strcpy(entry->buf, str_entry);
718			list_add(&entry->next, &blacklisted_initcalls);
719		}
720	} while (str_entry);
721
722	return 0;
723}
724
725static bool __init_or_module initcall_blacklisted(initcall_t fn)
726{
727	struct list_head *tmp;
728	struct blacklist_entry *entry;
729	char *fn_name;
730
731	fn_name = kasprintf(GFP_KERNEL, "%pf", fn);
732	if (!fn_name)
733		return false;
734
735	list_for_each(tmp, &blacklisted_initcalls) {
736		entry = list_entry(tmp, struct blacklist_entry, next);
737		if (!strcmp(fn_name, entry->buf)) {
738			pr_debug("initcall %s blacklisted\n", fn_name);
739			kfree(fn_name);
740			return true;
741		}
742	}
743
744	kfree(fn_name);
745	return false;
746}
747#else
748static int __init initcall_blacklist(char *str)
749{
750	pr_warn("initcall_blacklist requires CONFIG_KALLSYMS\n");
751	return 0;
752}
753
754static bool __init_or_module initcall_blacklisted(initcall_t fn)
755{
756	return false;
757}
758#endif
759__setup("initcall_blacklist=", initcall_blacklist);
760
761static int __init_or_module do_one_initcall_debug(initcall_t fn)
762{
763	ktime_t calltime, delta, rettime;
764	unsigned long long duration;
765	int ret;
766
767	printk(KERN_DEBUG "calling  %pF @ %i\n", fn, task_pid_nr(current));
768	calltime = ktime_get();
769	ret = fn();
770	rettime = ktime_get();
771	delta = ktime_sub(rettime, calltime);
772	duration = (unsigned long long) ktime_to_ns(delta) >> 10;
773	printk(KERN_DEBUG "initcall %pF returned %d after %lld usecs\n",
774		 fn, ret, duration);
775
776	return ret;
777}
778
779int __init_or_module do_one_initcall(initcall_t fn)
780{
781	int count = preempt_count();
782	int ret;
783	char msgbuf[64];
784
785	if (initcall_blacklisted(fn))
786		return -EPERM;
787
788	if (initcall_debug)
789		ret = do_one_initcall_debug(fn);
790	else
791		ret = fn();
792
793	msgbuf[0] = 0;
794
795	if (preempt_count() != count) {
796		sprintf(msgbuf, "preemption imbalance ");
797		preempt_count_set(count);
798	}
799	if (irqs_disabled()) {
800		strlcat(msgbuf, "disabled interrupts ", sizeof(msgbuf));
801		local_irq_enable();
802	}
803	WARN(msgbuf[0], "initcall %pF returned with %s\n", fn, msgbuf);
804
805	return ret;
806}
807
808
809extern initcall_t __initcall_start[];
810extern initcall_t __initcall0_start[];
811extern initcall_t __initcall1_start[];
812extern initcall_t __initcall2_start[];
813extern initcall_t __initcall3_start[];
814extern initcall_t __initcall4_start[];
815extern initcall_t __initcall5_start[];
816extern initcall_t __initcall6_start[];
817extern initcall_t __initcall7_start[];
818extern initcall_t __initcall_end[];
819
820static initcall_t *initcall_levels[] __initdata = {
821	__initcall0_start,
822	__initcall1_start,
823	__initcall2_start,
824	__initcall3_start,
825	__initcall4_start,
826	__initcall5_start,
827	__initcall6_start,
828	__initcall7_start,
829	__initcall_end,
830};
831
832/* Keep these in sync with initcalls in include/linux/init.h */
833static char *initcall_level_names[] __initdata = {
834	"early",
835	"core",
836	"postcore",
837	"arch",
838	"subsys",
839	"fs",
840	"device",
841	"late",
842};
843
844static void __init do_initcall_level(int level)
845{
846	initcall_t *fn;
847
848	strcpy(initcall_command_line, saved_command_line);
849	parse_args(initcall_level_names[level],
850		   initcall_command_line, __start___param,
851		   __stop___param - __start___param,
852		   level, level,
853		   &repair_env_string);
854
855	for (fn = initcall_levels[level]; fn < initcall_levels[level+1]; fn++)
856		do_one_initcall(*fn);
857}
858
859static void __init do_initcalls(void)
860{
861	int level;
862
863	for (level = 0; level < ARRAY_SIZE(initcall_levels) - 1; level++)
864		do_initcall_level(level);
865}
866
867/*
868 * Ok, the machine is now initialized. None of the devices
869 * have been touched yet, but the CPU subsystem is up and
870 * running, and memory and process management works.
871 *
872 * Now we can finally start doing some real work..
873 */
874static void __init do_basic_setup(void)
875{
876	cpuset_init_smp();
877	usermodehelper_init();
878	shmem_init();
879	driver_init();
880	init_irq_proc();
881	do_ctors();
882	usermodehelper_enable();
883	do_initcalls();
884	random_int_secret_init();
885}
886
887static void __init do_pre_smp_initcalls(void)
888{
889	initcall_t *fn;
890
891	for (fn = __initcall_start; fn < __initcall0_start; fn++)
892		do_one_initcall(*fn);
893}
894
895/*
896 * This function requests modules which should be loaded by default and is
897 * called twice right after initrd is mounted and right before init is
898 * exec'd.  If such modules are on either initrd or rootfs, they will be
899 * loaded before control is passed to userland.
900 */
901void __init load_default_modules(void)
902{
903	load_default_elevator_module();
904}
905
906static int run_init_process(const char *init_filename)
907{
908	argv_init[0] = init_filename;
909	return do_execve(getname_kernel(init_filename),
910		(const char __user *const __user *)argv_init,
911		(const char __user *const __user *)envp_init);
912}
913
914static int try_to_run_init_process(const char *init_filename)
915{
916	int ret;
917
918	ret = run_init_process(init_filename);
919
920	if (ret && ret != -ENOENT) {
921		pr_err("Starting init: %s exists but couldn't execute it (error %d)\n",
922		       init_filename, ret);
923	}
924
925	return ret;
926}
927
928static noinline void __init kernel_init_freeable(void);
929
930static int __ref kernel_init(void *unused)
931{
932	int ret;
933
934	kernel_init_freeable();
935	/* need to finish all async __init code before freeing the memory */
936	async_synchronize_full();
937	free_initmem();
938	mark_rodata_ro();
939	system_state = SYSTEM_RUNNING;
940	numa_default_policy();
941
942	flush_delayed_fput();
943
944	if (ramdisk_execute_command) {
945		ret = run_init_process(ramdisk_execute_command);
946		if (!ret)
947			return 0;
948		pr_err("Failed to execute %s (error %d)\n",
949		       ramdisk_execute_command, ret);
950	}
951
952	/*
953	 * We try each of these until one succeeds.
954	 *
955	 * The Bourne shell can be used instead of init if we are
956	 * trying to recover a really broken machine.
957	 */
958	if (execute_command) {
959		ret = run_init_process(execute_command);
960		if (!ret)
961			return 0;
962		pr_err("Failed to execute %s (error %d).  Attempting defaults...\n",
963			execute_command, ret);
964	}
965	if (!try_to_run_init_process("/sbin/init") ||
966	    !try_to_run_init_process("/etc/init") ||
967	    !try_to_run_init_process("/bin/init") ||
968	    !try_to_run_init_process("/bin/sh"))
969		return 0;
970
971	panic("No working init found.  Try passing init= option to kernel. "
972	      "See Linux Documentation/init.txt for guidance.");
973}
974
975static noinline void __init kernel_init_freeable(void)
976{
977	/*
978	 * Wait until kthreadd is all set-up.
979	 */
980	wait_for_completion(&kthreadd_done);
981
982	/* Now the scheduler is fully set up and can do blocking allocations */
983	gfp_allowed_mask = __GFP_BITS_MASK;
984
985	/*
986	 * init can allocate pages on any node
987	 */
988	set_mems_allowed(node_states[N_MEMORY]);
989	/*
990	 * init can run on any cpu.
991	 */
992	set_cpus_allowed_ptr(current, cpu_all_mask);
993
994	cad_pid = task_pid(current);
995
996	smp_prepare_cpus(setup_max_cpus);
997
998	do_pre_smp_initcalls();
999	lockup_detector_init();
1000
1001	smp_init();
1002	sched_init_smp();
1003
1004	do_basic_setup();
1005
1006	/* Open the /dev/console on the rootfs, this should never fail */
1007	if (sys_open((const char __user *) "/dev/console", O_RDWR, 0) < 0)
1008		pr_err("Warning: unable to open an initial console.\n");
1009
1010	(void) sys_dup(0);
1011	(void) sys_dup(0);
1012	/*
1013	 * check if there is an early userspace init.  If yes, let it do all
1014	 * the work
1015	 */
1016
1017	if (!ramdisk_execute_command)
1018		ramdisk_execute_command = "/init";
1019
1020	if (sys_access((const char __user *) ramdisk_execute_command, 0) != 0) {
1021		ramdisk_execute_command = NULL;
1022		prepare_namespace();
1023	}
1024
1025	/*
1026	 * Ok, we have completed the initial bootup, and
1027	 * we're essentially up and running. Get rid of the
1028	 * initmem segments and start the user-mode stuff..
1029	 */
1030
1031	/* rootfs is available now, try loading default modules */
1032	load_default_modules();
1033}
1034