4 * Copyright (C) 1991, 1992 Linus Torvalds
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>
12 #include <linux/types.h>
13 #include <linux/module.h>
14 #include <linux/proc_fs.h>
15 #include <linux/kernel.h>
16 #include <linux/syscalls.h>
17 #include <linux/string.h>
18 #include <linux/ctype.h>
19 #include <linux/delay.h>
20 #include <linux/utsname.h>
21 #include <linux/ioport.h>
22 #include <linux/init.h>
23 #include <linux/smp_lock.h>
24 #include <linux/initrd.h>
25 #include <linux/hdreg.h>
26 #include <linux/bootmem.h>
27 #include <linux/tty.h>
28 #include <linux/gfp.h>
29 #include <linux/percpu.h>
30 #include <linux/kmod.h>
31 #include <linux/kernel_stat.h>
32 #include <linux/start_kernel.h>
33 #include <linux/security.h>
34 #include <linux/workqueue.h>
35 #include <linux/profile.h>
36 #include <linux/rcupdate.h>
37 #include <linux/moduleparam.h>
38 #include <linux/kallsyms.h>
39 #include <linux/writeback.h>
40 #include <linux/cpu.h>
41 #include <linux/cpuset.h>
42 #include <linux/cgroup.h>
43 #include <linux/efi.h>
44 #include <linux/tick.h>
45 #include <linux/interrupt.h>
46 #include <linux/taskstats_kern.h>
47 #include <linux/delayacct.h>
48 #include <linux/unistd.h>
49 #include <linux/rmap.h>
50 #include <linux/mempolicy.h>
51 #include <linux/key.h>
52 #include <linux/unwind.h>
53 #include <linux/buffer_head.h>
54 #include <linux/debug_locks.h>
55 #include <linux/lockdep.h>
56 #include <linux/pid_namespace.h>
57 #include <linux/device.h>
58 #include <linux/kthread.h>
59 #include <linux/sched.h>
60 #include <linux/signal.h>
64 #include <asm/setup.h>
65 #include <asm/sections.h>
66 #include <asm/cacheflush.h>
68 #ifdef CONFIG_X86_LOCAL_APIC
73 * This is one of the first .c files built. Error out early if we have compiler
77 #if __GNUC__ == 4 && __GNUC_MINOR__ == 1 && __GNUC_PATCHLEVEL__ == 0
78 #warning gcc-4.1.0 is known to miscompile the kernel. A different compiler version is recommended.
81 static int kernel_init(void *);
83 extern void init_IRQ(void);
84 extern void fork_init(unsigned long);
85 extern void mca_init(void);
86 extern void sbus_init(void);
87 extern void pidhash_init(void);
88 extern void pidmap_init(void);
89 extern void prio_tree_init(void);
90 extern void radix_tree_init(void);
91 extern void free_initmem(void);
93 extern void acpi_early_init(void);
95 static inline void acpi_early_init(void) { }
97 #ifndef CONFIG_DEBUG_RODATA
98 static inline void mark_rodata_ro(void) { }
102 extern void tc_init(void);
105 enum system_states system_state
;
106 EXPORT_SYMBOL(system_state
);
109 * Boot command-line arguments
111 #define MAX_INIT_ARGS CONFIG_INIT_ENV_ARG_LIMIT
112 #define MAX_INIT_ENVS CONFIG_INIT_ENV_ARG_LIMIT
114 extern void time_init(void);
115 /* Default late time init is NULL. archs can override this later. */
116 void (*late_time_init
)(void);
117 extern void softirq_init(void);
119 /* Untouched command line saved by arch-specific code. */
120 char __initdata boot_command_line
[COMMAND_LINE_SIZE
];
121 /* Untouched saved command line (eg. for /proc) */
122 char *saved_command_line
;
123 /* Command line for parameter parsing */
124 static char *static_command_line
;
126 static char *execute_command
;
127 static char *ramdisk_execute_command
;
130 /* Setup configured maximum number of CPUs to activate */
131 static unsigned int __initdata max_cpus
= NR_CPUS
;
134 * Setup routine for controlling SMP activation
136 * Command-line option of "nosmp" or "maxcpus=0" will disable SMP
137 * activation entirely (the MPS table probe still happens, though).
139 * Command-line option of "maxcpus=<NUM>", where <NUM> is an integer
140 * greater than 0, limits the maximum number of CPUs activated in
143 #ifndef CONFIG_X86_IO_APIC
144 static inline void disable_ioapic_setup(void) {};
147 static int __init
nosmp(char *str
)
150 disable_ioapic_setup();
154 early_param("nosmp", nosmp
);
156 static int __init
maxcpus(char *str
)
158 get_option(&str
, &max_cpus
);
160 disable_ioapic_setup();
165 early_param("maxcpus", maxcpus
);
167 #define max_cpus NR_CPUS
171 * If set, this is an indication to the drivers that reset the underlying
172 * device before going ahead with the initialization otherwise driver might
173 * rely on the BIOS and skip the reset operation.
175 * This is useful if kernel is booting in an unreliable environment.
176 * For ex. kdump situaiton where previous kernel has crashed, BIOS has been
177 * skipped and devices will be in unknown state.
179 unsigned int reset_devices
;
180 EXPORT_SYMBOL(reset_devices
);
182 static int __init
set_reset_devices(char *str
)
188 __setup("reset_devices", set_reset_devices
);
190 static char * argv_init
[MAX_INIT_ARGS
+2] = { "init", NULL
, };
191 char * envp_init
[MAX_INIT_ENVS
+2] = { "HOME=/", "TERM=linux", NULL
, };
192 static const char *panic_later
, *panic_param
;
194 extern struct obs_kernel_param __setup_start
[], __setup_end
[];
196 static int __init
obsolete_checksetup(char *line
)
198 struct obs_kernel_param
*p
;
199 int had_early_param
= 0;
203 int n
= strlen(p
->str
);
204 if (!strncmp(line
, p
->str
, n
)) {
206 /* Already done in parse_early_param?
207 * (Needs exact match on param part).
208 * Keep iterating, as we can have early
209 * params and __setups of same names 8( */
210 if (line
[n
] == '\0' || line
[n
] == '=')
212 } else if (!p
->setup_func
) {
213 printk(KERN_WARNING
"Parameter %s is obsolete,"
214 " ignored\n", p
->str
);
216 } else if (p
->setup_func(line
+ n
))
220 } while (p
< __setup_end
);
222 return had_early_param
;
226 * This should be approx 2 Bo*oMips to start (note initial shift), and will
227 * still work even if initially too large, it will just take slightly longer
229 unsigned long loops_per_jiffy
= (1<<12);
231 EXPORT_SYMBOL(loops_per_jiffy
);
233 static int __init
debug_kernel(char *str
)
237 console_loglevel
= 10;
241 static int __init
quiet_kernel(char *str
)
245 console_loglevel
= 4;
249 __setup("debug", debug_kernel
);
250 __setup("quiet", quiet_kernel
);
252 static int __init
loglevel(char *str
)
254 get_option(&str
, &console_loglevel
);
258 __setup("loglevel=", loglevel
);
261 * Unknown boot options get handed to init, unless they look like
264 static int __init
unknown_bootoption(char *param
, char *val
)
266 /* Change NUL term back to "=", to make "param" the whole string. */
268 /* param=val or param="val"? */
269 if (val
== param
+strlen(param
)+1)
271 else if (val
== param
+strlen(param
)+2) {
273 memmove(val
-1, val
, strlen(val
)+1);
279 /* Handle obsolete-style parameters */
280 if (obsolete_checksetup(param
))
284 * Preemptive maintenance for "why didn't my misspelled command
287 if (strchr(param
, '.') && (!val
|| strchr(param
, '.') < val
)) {
288 printk(KERN_ERR
"Unknown boot option `%s': ignoring\n", param
);
296 /* Environment option */
298 for (i
= 0; envp_init
[i
]; i
++) {
299 if (i
== MAX_INIT_ENVS
) {
300 panic_later
= "Too many boot env vars at `%s'";
303 if (!strncmp(param
, envp_init
[i
], val
- param
))
306 envp_init
[i
] = param
;
308 /* Command line option */
310 for (i
= 0; argv_init
[i
]; i
++) {
311 if (i
== MAX_INIT_ARGS
) {
312 panic_later
= "Too many boot init vars at `%s'";
316 argv_init
[i
] = param
;
321 static int __init
init_setup(char *str
)
325 execute_command
= str
;
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]
332 for (i
= 1; i
< MAX_INIT_ARGS
; i
++)
336 __setup("init=", init_setup
);
338 static int __init
rdinit_setup(char *str
)
342 ramdisk_execute_command
= str
;
343 /* See "auto" comment in init_setup */
344 for (i
= 1; i
< MAX_INIT_ARGS
; i
++)
348 __setup("rdinit=", rdinit_setup
);
352 #ifdef CONFIG_X86_LOCAL_APIC
353 static void __init
smp_init(void)
355 APIC_init_uniprocessor();
358 #define smp_init() do { } while (0)
361 static inline void setup_per_cpu_areas(void) { }
362 static inline void smp_prepare_cpus(unsigned int maxcpus
) { }
366 #ifdef __GENERIC_PER_CPU
367 unsigned long __per_cpu_offset
[NR_CPUS
] __read_mostly
;
369 EXPORT_SYMBOL(__per_cpu_offset
);
371 static void __init
setup_per_cpu_areas(void)
373 unsigned long size
, i
;
375 unsigned long nr_possible_cpus
= num_possible_cpus();
377 /* Copy section for each CPU (we discard the original) */
378 size
= ALIGN(PERCPU_ENOUGH_ROOM
, PAGE_SIZE
);
379 ptr
= alloc_bootmem_pages(size
* nr_possible_cpus
);
381 for_each_possible_cpu(i
) {
382 __per_cpu_offset
[i
] = ptr
- __per_cpu_start
;
383 memcpy(ptr
, __per_cpu_start
, __per_cpu_end
- __per_cpu_start
);
387 #endif /* !__GENERIC_PER_CPU */
389 /* Called by boot processor to activate the rest. */
390 static void __init
smp_init(void)
394 /* FIXME: This should be done in userspace --RR */
395 for_each_present_cpu(cpu
) {
396 if (num_online_cpus() >= max_cpus
)
398 if (!cpu_online(cpu
))
402 /* Any cleanup work */
403 printk(KERN_INFO
"Brought up %ld CPUs\n", (long)num_online_cpus());
404 smp_cpus_done(max_cpus
);
410 * We need to store the untouched command line for future reference.
411 * We also need to store the touched command line since the parameter
412 * parsing is performed in place, and we should allow a component to
413 * store reference of name/value for future reference.
415 static void __init
setup_command_line(char *command_line
)
417 saved_command_line
= alloc_bootmem(strlen (boot_command_line
)+1);
418 static_command_line
= alloc_bootmem(strlen (command_line
)+1);
419 strcpy (saved_command_line
, boot_command_line
);
420 strcpy (static_command_line
, command_line
);
424 * We need to finalize in a non-__init function or else race conditions
425 * between the root thread and the init thread may cause start_kernel to
426 * be reaped by free_initmem before the root thread has proceeded to
429 * gcc-3.4 accidentally inlines this function, so use noinline.
432 static void noinline __init_refok
rest_init(void)
433 __releases(kernel_lock
)
437 kernel_thread(kernel_init
, NULL
, CLONE_FS
| CLONE_SIGHAND
);
438 numa_default_policy();
439 pid
= kernel_thread(kthreadd
, NULL
, CLONE_FS
| CLONE_FILES
);
440 kthreadd_task
= find_task_by_pid(pid
);
444 * The boot idle thread must execute schedule()
445 * at least once to get things moving:
447 init_idle_bootup_task(current
);
448 preempt_enable_no_resched();
452 /* Call into cpu_idle with preempt disabled */
456 /* Check for early params. */
457 static int __init
do_early_param(char *param
, char *val
)
459 struct obs_kernel_param
*p
;
461 for (p
= __setup_start
; p
< __setup_end
; p
++) {
462 if ((p
->early
&& strcmp(param
, p
->str
) == 0) ||
463 (strcmp(param
, "console") == 0 &&
464 strcmp(p
->str
, "earlycon") == 0)
466 if (p
->setup_func(val
) != 0)
468 "Malformed early option '%s'\n", param
);
471 /* We accept everything at this stage. */
475 /* Arch code calls this early on, or if not, just before other parsing. */
476 void __init
parse_early_param(void)
478 static __initdata
int done
= 0;
479 static __initdata
char tmp_cmdline
[COMMAND_LINE_SIZE
];
484 /* All fall through to do_early_param. */
485 strlcpy(tmp_cmdline
, boot_command_line
, COMMAND_LINE_SIZE
);
486 parse_args("early options", tmp_cmdline
, NULL
, 0, do_early_param
);
491 * Activate the first processor.
494 static void __init
boot_cpu_init(void)
496 int cpu
= smp_processor_id();
497 /* Mark the boot cpu "present", "online" etc for SMP and UP case */
498 cpu_set(cpu
, cpu_online_map
);
499 cpu_set(cpu
, cpu_present_map
);
500 cpu_set(cpu
, cpu_possible_map
);
503 void __init
__attribute__((weak
)) smp_setup_processor_id(void)
507 asmlinkage
void __init
start_kernel(void)
510 extern struct kernel_param __start___param
[], __stop___param
[];
512 smp_setup_processor_id();
515 * Need to run as early as possible, to initialize the
523 early_boot_irqs_off();
524 early_init_irq_lock_class();
527 * Interrupts are still disabled. Do necessary setups, then
535 printk(linux_banner
);
536 setup_arch(&command_line
);
537 setup_command_line(command_line
);
539 setup_per_cpu_areas();
540 smp_prepare_boot_cpu(); /* arch-specific boot-cpu hooks */
543 * Set up the scheduler prior starting any interrupts (such as the
544 * timer interrupt). Full topology setup happens at smp_init()
545 * time - but meanwhile we still have a functioning scheduler.
549 * Disable preemption - early bootup scheduling is extremely
550 * fragile until we cpu_idle() for the first time.
553 build_all_zonelists();
555 printk(KERN_NOTICE
"Kernel command line: %s\n", boot_command_line
);
557 parse_args("Booting kernel", static_command_line
, __start___param
,
558 __stop___param
- __start___param
,
559 &unknown_bootoption
);
560 if (!irqs_disabled()) {
561 printk(KERN_WARNING
"start_kernel(): bug: interrupts were "
562 "enabled *very* early, fixing it\n");
576 if (!irqs_disabled())
577 printk("start_kernel(): bug: interrupts were enabled early\n");
578 early_boot_irqs_on();
582 * HACK ALERT! This is early. We're enabling the console before
583 * we've done PCI setups etc, and console_init() must be aware of
584 * this. But we do want output early, in case something goes wrong.
588 panic(panic_later
, panic_param
);
593 * Need to run this when irqs are enabled, because it wants
594 * to self-test [hard/soft]-irqs on/off lock inversion bugs
599 #ifdef CONFIG_BLK_DEV_INITRD
600 if (initrd_start
&& !initrd_below_start_ok
&&
601 initrd_start
< min_low_pfn
<< PAGE_SHIFT
) {
602 printk(KERN_CRIT
"initrd overwritten (0x%08lx < 0x%08lx) - "
603 "disabling it.\n",initrd_start
,min_low_pfn
<< PAGE_SHIFT
);
607 vfs_caches_init_early();
611 setup_per_cpu_pageset();
617 pgtable_cache_init();
622 efi_enter_virtual_mode();
624 fork_init(num_physpages
);
630 vfs_caches_init(num_physpages
);
633 /* rootfs populating might need page-writeback */
634 page_writeback_init();
635 #ifdef CONFIG_PROC_FS
640 taskstats_init_early();
645 acpi_early_init(); /* before LAPIC and SMP init */
647 /* Do the rest non-__init'ed, we're now alive */
651 static int __initdata initcall_debug
;
653 static int __init
initcall_debug_setup(char *str
)
658 __setup("initcall_debug", initcall_debug_setup
);
660 extern initcall_t __initcall_start
[], __initcall_end
[];
662 static void __init
do_initcalls(void)
665 int count
= preempt_count();
667 for (call
= __initcall_start
; call
< __initcall_end
; call
++) {
668 ktime_t t0
, t1
, delta
;
673 if (initcall_debug
) {
674 printk("Calling initcall 0x%p", *call
);
675 print_fn_descriptor_symbol(": %s()",
676 (unsigned long) *call
);
683 if (initcall_debug
) {
685 delta
= ktime_sub(t1
, t0
);
687 printk("initcall 0x%p", *call
);
688 print_fn_descriptor_symbol(": %s()",
689 (unsigned long) *call
);
690 printk(" returned %d.\n", result
);
692 printk("initcall 0x%p ran for %Ld msecs: ",
693 *call
, (unsigned long long)delta
.tv64
>> 20);
694 print_fn_descriptor_symbol("%s()\n",
695 (unsigned long) *call
);
698 if (result
&& result
!= -ENODEV
&& initcall_debug
) {
699 sprintf(msgbuf
, "error code %d", result
);
702 if (preempt_count() != count
) {
703 msg
= "preemption imbalance";
704 preempt_count() = count
;
706 if (irqs_disabled()) {
707 msg
= "disabled interrupts";
711 printk(KERN_WARNING
"initcall at 0x%p", *call
);
712 print_fn_descriptor_symbol(": %s()",
713 (unsigned long) *call
);
714 printk(": returned with %s\n", msg
);
718 /* Make sure there is no pending stuff from the initcall sequence */
719 flush_scheduled_work();
723 * Ok, the machine is now initialized. None of the devices
724 * have been touched yet, but the CPU subsystem is up and
725 * running, and memory and process management works.
727 * Now we can finally start doing some real work..
729 static void __init
do_basic_setup(void)
731 /* drivers will send hotplug events */
733 usermodehelper_init();
739 static int __initdata nosoftlockup
;
741 static int __init
nosoftlockup_setup(char *str
)
746 __setup("nosoftlockup", nosoftlockup_setup
);
748 static void __init
do_pre_smp_initcalls(void)
750 extern int spawn_ksoftirqd(void);
755 spawn_softlockup_task();
758 static void run_init_process(char *init_filename
)
760 argv_init
[0] = init_filename
;
761 kernel_execve(init_filename
, argv_init
, envp_init
);
765 * __init/__init_data sections are turned into normal
766 * dynamically allocated memory later in boot. When
767 * this is 0, the memory is for the __init purposes,
768 * when it it some other value, the memory is dynamic.
770 int initmem_now_dynamic
;
772 /* This is a non __init function. Force it to be noinline otherwise gcc
773 * makes it inline to init() and it becomes part of init.text section
775 static int noinline
init_post(void)
778 initmem_now_dynamic
= 1;
781 system_state
= SYSTEM_RUNNING
;
782 numa_default_policy();
784 if (sys_open((const char __user
*) "/dev/console", O_RDWR
, 0) < 0)
785 printk(KERN_WARNING
"Warning: unable to open an initial console.\n");
790 if (ramdisk_execute_command
) {
791 run_init_process(ramdisk_execute_command
);
792 printk(KERN_WARNING
"Failed to execute %s\n",
793 ramdisk_execute_command
);
797 * We try each of these until one succeeds.
799 * The Bourne shell can be used instead of init if we are
800 * trying to recover a really broken machine.
802 if (execute_command
) {
803 run_init_process(execute_command
);
804 printk(KERN_WARNING
"Failed to execute %s. Attempting "
805 "defaults...\n", execute_command
);
807 run_init_process("/sbin/init");
808 run_init_process("/etc/init");
809 run_init_process("/bin/init");
810 run_init_process("/bin/sh");
812 panic("No init found. Try passing init= option to kernel.");
815 static int __init
kernel_init(void * unused
)
819 * init can run on any cpu.
821 set_cpus_allowed(current
, CPU_MASK_ALL
);
823 * Tell the world that we're going to be the grim
824 * reaper of innocent orphaned children.
826 * We don't want people to have to make incorrect
827 * assumptions about where in the task array this
830 init_pid_ns
.child_reaper
= current
;
832 __set_special_pids(1, 1);
833 cad_pid
= task_pid(current
);
835 smp_prepare_cpus(max_cpus
);
837 do_pre_smp_initcalls();
847 * check if there is an early userspace init. If yes, let it do all
851 if (!ramdisk_execute_command
)
852 ramdisk_execute_command
= "/init";
854 if (sys_access((const char __user
*) ramdisk_execute_command
, 0) != 0) {
855 ramdisk_execute_command
= NULL
;
860 * Ok, we have completed the initial bootup, and
861 * we're essentially up and running. Get rid of the
862 * initmem segments and start the user-mode stuff..