init/main.c

   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 #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>
  61
  62 #include <asm/io.h>
  63 #include <asm/bugs.h>
  64 #include <asm/setup.h>
  65 #include <asm/sections.h>
  66 #include <asm/cacheflush.h>
  67
  68 #ifdef CONFIG_X86_LOCAL_APIC
  69 #include <asm/smp.h>
  70 #endif
  71
  72 /*
  73  * This is one of the first .c files built. Error out early if we have compiler
  74  * trouble.
  75  */
  76
  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.
  79 #endif
  80
  81 static int kernel_init(void *);
  82
  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);
  92 #ifdef  CONFIG_ACPI
  93 extern void acpi_early_init(void);
  94 #else
  95 static inline void acpi_early_init(void) { }
  96 #endif
  97 #ifndef CONFIG_DEBUG_RODATA
  98 static inline void mark_rodata_ro(void) { }
  99 #endif
 100
 101 #ifdef CONFIG_TC
 102 extern void tc_init(void);
 103 #endif
 104
 105 enum system_states system_state;
 106 EXPORT_SYMBOL(system_state);
 107
 108 /*
 109  * Boot command-line arguments
 110  */
 111 #define MAX_INIT_ARGS CONFIG_INIT_ENV_ARG_LIMIT
 112 #define MAX_INIT_ENVS CONFIG_INIT_ENV_ARG_LIMIT
 113
 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);
 118
 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;
 125
 126 static char *execute_command;
 127 static char *ramdisk_execute_command;
 128
 129 #ifdef CONFIG_SMP
 130 /* Setup configured maximum number of CPUs to activate */
 131 unsigned int __initdata setup_max_cpus = NR_CPUS;
 132
 133 /*
 134  * Setup routine for controlling SMP activation
 135  *
 136  * Command-line option of "nosmp" or "maxcpus=0" will disable SMP
 137  * activation entirely (the MPS table probe still happens, though).
 138  *
 139  * Command-line option of "maxcpus=<NUM>", where <NUM> is an integer
 140  * greater than 0, limits the maximum number of CPUs activated in
 141  * SMP mode to <NUM>.
 142  */
 143 #ifndef CONFIG_X86_IO_APIC
 144 static inline void disable_ioapic_setup(void) {};
 145 #endif
 146
 147 static int __init nosmp(char *str)
 148 {
 149         setup_max_cpus = 0;
 150         disable_ioapic_setup();
 151         return 0;
 152 }
 153
 154 early_param("nosmp", nosmp);
 155
 156 static int __init maxcpus(char *str)
 157 {
 158         get_option(&str, &setup_max_cpus);
 159         if (setup_max_cpus == 0)
 160                 disable_ioapic_setup();
 161
 162         return 0;
 163 }
 164
 165 early_param("maxcpus", maxcpus);
 166 #else
 167 #define setup_max_cpus NR_CPUS
 168 #endif
 169
 170 /*
 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.
 174  *
 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.
 178  */
 179 unsigned int reset_devices;
 180 EXPORT_SYMBOL(reset_devices);
 181
 182 static int __init set_reset_devices(char *str)
 183 {
 184         reset_devices = 1;
 185         return 1;
 186 }
 187
 188 __setup("reset_devices", set_reset_devices);
 189
 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;
 193
 194 extern struct obs_kernel_param __setup_start[], __setup_end[];
 195
 196 static int __init obsolete_checksetup(char *line)
 197 {
 198         struct obs_kernel_param *p;
 199         int had_early_param = 0;
 200
 201         p = __setup_start;
 202         do {
 203                 int n = strlen(p->str);
 204                 if (!strncmp(line, p->str, n)) {
 205                         if (p->early) {
 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] == '=')
 211                                         had_early_param = 1;
 212                         } else if (!p->setup_func) {
 213                                 printk(KERN_WARNING "Parameter %s is obsolete,"
 214                                        " ignored\n", p->str);
 215                                 return 1;
 216                         } else if (p->setup_func(line + n))
 217                                 return 1;
 218                 }
 219                 p++;
 220         } while (p < __setup_end);
 221
 222         return had_early_param;
 223 }
 224
 225 /*
 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
 228  */
 229 unsigned long loops_per_jiffy = (1<<12);
 230
 231 EXPORT_SYMBOL(loops_per_jiffy);
 232
 233 static int __init debug_kernel(char *str)
 234 {
 235         if (*str)
 236                 return 0;
 237         console_loglevel = 10;
 238         return 1;
 239 }
 240
 241 static int __init quiet_kernel(char *str)
 242 {
 243         if (*str)
 244                 return 0;
 245         console_loglevel = 4;
 246         return 1;
 247 }
 248
 249 __setup("debug", debug_kernel);
 250 __setup("quiet", quiet_kernel);
 251
 252 static int __init loglevel(char *str)
 253 {
 254         get_option(&str, &console_loglevel);
 255         return 1;
 256 }
 257
 258 __setup("loglevel=", loglevel);
 259
 260 /*
 261  * Unknown boot options get handed to init, unless they look like
 262  * failed parameters
 263  */
 264 static int __init unknown_bootoption(char *param, char *val)
 265 {
 266         /* Change NUL term back to "=", to make "param" the whole string. */
 267         if (val) {
 268                 /* param=val or param="val"? */
 269                 if (val == param+strlen(param)+1)
 270                         val[-1] = '=';
 271                 else if (val == param+strlen(param)+2) {
 272                         val[-2] = '=';
 273                         memmove(val-1, val, strlen(val)+1);
 274                         val--;
 275                 } else
 276                         BUG();
 277         }
 278
 279         /* Handle obsolete-style parameters */
 280         if (obsolete_checksetup(param))
 281                 return 0;
 282
 283         /*
 284          * Preemptive maintenance for "why didn't my misspelled command
 285          * line work?"
 286          */
 287         if (strchr(param, '.') && (!val || strchr(param, '.') < val)) {
 288                 printk(KERN_ERR "Unknown boot option `%s': ignoring\n", param);
 289                 return 0;
 290         }
 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 = "Too many boot env vars at `%s'";
 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 = "Too many boot init vars at `%s'";
 313                                 panic_param = param;
 314                         }
 315                 }
 316                 argv_init[i] = param;
 317         }
 318         return 0;
 319 }
 320
 321 #ifdef CONFIG_DEBUG_PAGEALLOC
 322 int __read_mostly debug_pagealloc_enabled = 0;
 323 #endif
 324
 325 static int __init init_setup(char *str)
 326 {
 327         unsigned int i;
 328
 329         execute_command = str;
 330         /*
 331          * In case LILO is going to boot us with default command line,
 332          * it prepends "auto" before the whole cmdline which makes
 333          * the shell think it should execute a script with such name.
 334          * So we ignore all arguments entered _before_ init=... [MJ]
 335          */
 336         for (i = 1; i < MAX_INIT_ARGS; i++)
 337                 argv_init[i] = NULL;
 338         return 1;
 339 }
 340 __setup("init=", init_setup);
 341
 342 static int __init rdinit_setup(char *str)
 343 {
 344         unsigned int i;
 345
 346         ramdisk_execute_command = str;
 347         /* See "auto" comment in init_setup */
 348         for (i = 1; i < MAX_INIT_ARGS; i++)
 349                 argv_init[i] = NULL;
 350         return 1;
 351 }
 352 __setup("rdinit=", rdinit_setup);
 353
 354 #ifndef CONFIG_SMP
 355
 356 #ifdef CONFIG_X86_LOCAL_APIC
 357 static void __init smp_init(void)
 358 {
 359         APIC_init_uniprocessor();
 360 }
 361 #else
 362 #define smp_init()      do { } while (0)
 363 #endif
 364
 365 static inline void setup_per_cpu_areas(void) { }
 366 static inline void smp_prepare_cpus(unsigned int maxcpus) { }
 367
 368 #else
 369
 370 #ifndef CONFIG_HAVE_SETUP_PER_CPU_AREA
 371 unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
 372
 373 EXPORT_SYMBOL(__per_cpu_offset);
 374
 375 static void __init setup_per_cpu_areas(void)
 376 {
 377         unsigned long size, i;
 378         char *ptr;
 379         unsigned long nr_possible_cpus = num_possible_cpus();
 380
 381         /* Copy section for each CPU (we discard the original) */
 382         size = ALIGN(PERCPU_ENOUGH_ROOM, PAGE_SIZE);
 383         ptr = alloc_bootmem_pages(size * nr_possible_cpus);
 384
 385         for_each_possible_cpu(i) {
 386                 __per_cpu_offset[i] = ptr - __per_cpu_start;
 387                 memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
 388                 ptr += size;
 389         }
 390 }
 391 #endif /* CONFIG_HAVE_SETUP_PER_CPU_AREA */
 392
 393 /* Called by boot processor to activate the rest. */
 394 static void __init smp_init(void)
 395 {
 396         unsigned int cpu;
 397
 398         /* FIXME: This should be done in userspace --RR */
 399         for_each_present_cpu(cpu) {
 400                 if (num_online_cpus() >= setup_max_cpus)
 401                         break;
 402                 if (!cpu_online(cpu))
 403                         cpu_up(cpu);
 404         }
 405
 406         /* Any cleanup work */
 407         printk(KERN_INFO "Brought up %ld CPUs\n", (long)num_online_cpus());
 408         smp_cpus_done(setup_max_cpus);
 409 }
 410
 411 #endif
 412
 413 /*
 414  * We need to store the untouched command line for future reference.
 415  * We also need to store the touched command line since the parameter
 416  * parsing is performed in place, and we should allow a component to
 417  * store reference of name/value for future reference.
 418  */
 419 static void __init setup_command_line(char *command_line)
 420 {
 421         saved_command_line = alloc_bootmem(strlen (boot_command_line)+1);
 422         static_command_line = alloc_bootmem(strlen (command_line)+1);
 423         strcpy (saved_command_line, boot_command_line);
 424         strcpy (static_command_line, command_line);
 425 }
 426
 427 /*
 428  * We need to finalize in a non-__init function or else race conditions
 429  * between the root thread and the init thread may cause start_kernel to
 430  * be reaped by free_initmem before the root thread has proceeded to
 431  * cpu_idle.
 432  *
 433  * gcc-3.4 accidentally inlines this function, so use noinline.
 434  */
 435
 436 static void noinline __init_refok rest_init(void)
 437         __releases(kernel_lock)
 438 {
 439         int pid;
 440
 441         kernel_thread(kernel_init, NULL, CLONE_FS | CLONE_SIGHAND);
 442         numa_default_policy();
 443         pid = kernel_thread(kthreadd, NULL, CLONE_FS | CLONE_FILES);
 444         kthreadd_task = find_task_by_pid(pid);
 445         unlock_kernel();
 446
 447         /*
 448          * The boot idle thread must execute schedule()
 449          * at least once to get things moving:
 450          */
 451         init_idle_bootup_task(current);
 452         preempt_enable_no_resched();
 453         schedule();
 454         preempt_disable();
 455
 456         /* Call into cpu_idle with preempt disabled */
 457         cpu_idle();
 458 }
 459
 460 /* Check for early params. */
 461 static int __init do_early_param(char *param, char *val)
 462 {
 463         struct obs_kernel_param *p;
 464
 465         for (p = __setup_start; p < __setup_end; p++) {
 466                 if ((p->early && strcmp(param, p->str) == 0) ||
 467                     (strcmp(param, "console") == 0 &&
 468                      strcmp(p->str, "earlycon") == 0)
 469                 ) {
 470                         if (p->setup_func(val) != 0)
 471                                 printk(KERN_WARNING
 472                                        "Malformed early option '%s'\n", param);
 473                 }
 474         }
 475         /* We accept everything at this stage. */
 476         return 0;
 477 }
 478
 479 /* Arch code calls this early on, or if not, just before other parsing. */
 480 void __init parse_early_param(void)
 481 {
 482         static __initdata int done = 0;
 483         static __initdata char tmp_cmdline[COMMAND_LINE_SIZE];
 484
 485         if (done)
 486                 return;
 487
 488         /* All fall through to do_early_param. */
 489         strlcpy(tmp_cmdline, boot_command_line, COMMAND_LINE_SIZE);
 490         parse_args("early options", tmp_cmdline, NULL, 0, do_early_param);
 491         done = 1;
 492 }
 493
 494 /*
 495  *      Activate the first processor.
 496  */
 497
 498 static void __init boot_cpu_init(void)
 499 {
 500         int cpu = smp_processor_id();
 501         /* Mark the boot cpu "present", "online" etc for SMP and UP case */
 502         cpu_set(cpu, cpu_online_map);
 503         cpu_set(cpu, cpu_present_map);
 504         cpu_set(cpu, cpu_possible_map);
 505 }
 506
 507 void __init __attribute__((weak)) smp_setup_processor_id(void)
 508 {
 509 }
 510
 511 asmlinkage void __init start_kernel(void)
 512 {
 513         char * command_line;
 514         extern struct kernel_param __start___param[], __stop___param[];
 515
 516         smp_setup_processor_id();
 517
 518         /*
 519          * Need to run as early as possible, to initialize the
 520          * lockdep hash:
 521          */
 522         unwind_init();
 523         lockdep_init();
 524         cgroup_init_early();
 525
 526         local_irq_disable();
 527         early_boot_irqs_off();
 528         early_init_irq_lock_class();
 529
 530 /*
 531  * Interrupts are still disabled. Do necessary setups, then
 532  * enable them
 533  */
 534         lock_kernel();
 535         tick_init();
 536         boot_cpu_init();
 537         page_address_init();
 538         printk(KERN_NOTICE);
 539         printk(linux_banner);
 540         setup_arch(&command_line);
 541         setup_command_line(command_line);
 542         unwind_setup();
 543         setup_per_cpu_areas();
 544         smp_prepare_boot_cpu(); /* arch-specific boot-cpu hooks */
 545
 546         /*
 547          * Set up the scheduler prior starting any interrupts (such as the
 548          * timer interrupt). Full topology setup happens at smp_init()
 549          * time - but meanwhile we still have a functioning scheduler.
 550          */
 551         sched_init();
 552         /*
 553          * Disable preemption - early bootup scheduling is extremely
 554          * fragile until we cpu_idle() for the first time.
 555          */
 556         preempt_disable();
 557         build_all_zonelists();
 558         page_alloc_init();
 559         enable_debug_pagealloc();
 560         printk(KERN_NOTICE "Kernel command line: %s\n", boot_command_line);
 561         parse_early_param();
 562         parse_args("Booting kernel", static_command_line, __start___param,
 563                    __stop___param - __start___param,
 564                    &unknown_bootoption);
 565         if (!irqs_disabled()) {
 566                 printk(KERN_WARNING "start_kernel(): bug: interrupts were "
 567                                 "enabled *very* early, fixing it\n");
 568                 local_irq_disable();
 569         }
 570         sort_main_extable();
 571         trap_init();
 572         rcu_init();
 573         init_IRQ();
 574         pidhash_init();
 575         init_timers();
 576         hrtimers_init();
 577         softirq_init();
 578         timekeeping_init();
 579         time_init();
 580         profile_init();
 581         if (!irqs_disabled())
 582                 printk("start_kernel(): bug: interrupts were enabled early\n");
 583         early_boot_irqs_on();
 584         local_irq_enable();
 585
 586         /*
 587          * HACK ALERT! This is early. We're enabling the console before
 588          * we've done PCI setups etc, and console_init() must be aware of
 589          * this. But we do want output early, in case something goes wrong.
 590          */
 591         console_init();
 592         if (panic_later)
 593                 panic(panic_later, panic_param);
 594
 595         lockdep_info();
 596
 597         /*
 598          * Need to run this when irqs are enabled, because it wants
 599          * to self-test [hard/soft]-irqs on/off lock inversion bugs
 600          * too:
 601          */
 602         locking_selftest();
 603
 604 #ifdef CONFIG_BLK_DEV_INITRD
 605         if (initrd_start && !initrd_below_start_ok &&
 606                         initrd_start < min_low_pfn << PAGE_SHIFT) {
 607                 printk(KERN_CRIT "initrd overwritten (0x%08lx < 0x%08lx) - "
 608                     "disabling it.\n",initrd_start,min_low_pfn << PAGE_SHIFT);
 609                 initrd_start = 0;
 610         }
 611 #endif
 612         vfs_caches_init_early();
 613         cpuset_init_early();
 614         mem_init();
 615         cpu_hotplug_init();
 616         kmem_cache_init();
 617         setup_per_cpu_pageset();
 618         numa_policy_init();
 619         if (late_time_init)
 620                 late_time_init();
 621         calibrate_delay();
 622         pidmap_init();
 623         pgtable_cache_init();
 624         prio_tree_init();
 625         anon_vma_init();
 626 #ifdef CONFIG_X86
 627         if (efi_enabled)
 628                 efi_enter_virtual_mode();
 629 #endif
 630         fork_init(num_physpages);
 631         proc_caches_init();
 632         buffer_init();
 633         unnamed_dev_init();
 634         key_init();
 635         security_init();
 636         vfs_caches_init(num_physpages);
 637         radix_tree_init();
 638         signals_init();
 639         /* rootfs populating might need page-writeback */
 640         page_writeback_init();
 641 #ifdef CONFIG_PROC_FS
 642         proc_root_init();
 643 #endif
 644         cgroup_init();
 645         cpuset_init();
 646         taskstats_init_early();
 647         delayacct_init();
 648
 649         check_bugs();
 650
 651         acpi_early_init(); /* before LAPIC and SMP init */
 652
 653         /* Do the rest non-__init'ed, we're now alive */
 654         rest_init();
 655 }
 656
 657 static int __initdata initcall_debug;
 658
 659 static int __init initcall_debug_setup(char *str)
 660 {
 661         initcall_debug = 1;
 662         return 1;
 663 }
 664 __setup("initcall_debug", initcall_debug_setup);
 665
 666 extern initcall_t __initcall_start[], __initcall_end[];
 667
 668 static void __init do_initcalls(void)
 669 {
 670         initcall_t *call;
 671         int count = preempt_count();
 672
 673         for (call = __initcall_start; call < __initcall_end; call++) {
 674                 ktime_t t0, t1, delta;
 675                 char *msg = NULL;
 676                 char msgbuf[40];
 677                 int result;
 678
 679                 if (initcall_debug) {
 680                         printk("Calling initcall 0x%p", *call);
 681                         print_fn_descriptor_symbol(": %s()",
 682                                         (unsigned long) *call);
 683                         printk("\n");
 684                         t0 = ktime_get();
 685                 }
 686
 687                 result = (*call)();
 688
 689                 if (initcall_debug) {
 690                         t1 = ktime_get();
 691                         delta = ktime_sub(t1, t0);
 692
 693                         printk("initcall 0x%p", *call);
 694                         print_fn_descriptor_symbol(": %s()",
 695                                         (unsigned long) *call);
 696                         printk(" returned %d.\n", result);
 697
 698                         printk("initcall 0x%p ran for %Ld msecs: ",
 699                                 *call, (unsigned long long)delta.tv64 >> 20);
 700                         print_fn_descriptor_symbol("%s()\n",
 701                                 (unsigned long) *call);
 702                 }
 703
 704                 if (result && result != -ENODEV && initcall_debug) {
 705                         sprintf(msgbuf, "error code %d", result);
 706                         msg = msgbuf;
 707                 }
 708                 if (preempt_count() != count) {
 709                         msg = "preemption imbalance";
 710                         preempt_count() = count;
 711                 }
 712                 if (irqs_disabled()) {
 713                         msg = "disabled interrupts";
 714                         local_irq_enable();
 715                 }
 716                 if (msg) {
 717                         printk(KERN_WARNING "initcall at 0x%p", *call);
 718                         print_fn_descriptor_symbol(": %s()",
 719                                         (unsigned long) *call);
 720                         printk(": returned with %s\n", msg);
 721                 }
 722         }
 723
 724         /* Make sure there is no pending stuff from the initcall sequence */
 725         flush_scheduled_work();
 726 }
 727
 728 /*
 729  * Ok, the machine is now initialized. None of the devices
 730  * have been touched yet, but the CPU subsystem is up and
 731  * running, and memory and process management works.
 732  *
 733  * Now we can finally start doing some real work..
 734  */
 735 static void __init do_basic_setup(void)
 736 {
 737         /* drivers will send hotplug events */
 738         init_workqueues();
 739         usermodehelper_init();
 740         driver_init();
 741         init_irq_proc();
 742         do_initcalls();
 743 }
 744
 745 static int __initdata nosoftlockup;
 746
 747 static int __init nosoftlockup_setup(char *str)
 748 {
 749         nosoftlockup = 1;
 750         return 1;
 751 }
 752 __setup("nosoftlockup", nosoftlockup_setup);
 753
 754 static void __init do_pre_smp_initcalls(void)
 755 {
 756         extern int spawn_ksoftirqd(void);
 757
 758         migration_init();
 759         spawn_ksoftirqd();
 760         if (!nosoftlockup)
 761                 spawn_softlockup_task();
 762 }
 763
 764 static void run_init_process(char *init_filename)
 765 {
 766         argv_init[0] = init_filename;
 767         kernel_execve(init_filename, argv_init, envp_init);
 768 }
 769
 770 /* This is a non __init function. Force it to be noinline otherwise gcc
 771  * makes it inline to init() and it becomes part of init.text section
 772  */
 773 static int noinline init_post(void)
 774 {
 775         free_initmem();
 776         unlock_kernel();
 777         mark_rodata_ro();
 778         system_state = SYSTEM_RUNNING;
 779         numa_default_policy();
 780
 781         if (sys_open((const char __user *) "/dev/console", O_RDWR, 0) < 0)
 782                 printk(KERN_WARNING "Warning: unable to open an initial console.\n");
 783
 784         (void) sys_dup(0);
 785         (void) sys_dup(0);
 786
 787         if (ramdisk_execute_command) {
 788                 run_init_process(ramdisk_execute_command);
 789                 printk(KERN_WARNING "Failed to execute %s\n",
 790                                 ramdisk_execute_command);
 791         }
 792
 793         /*
 794          * We try each of these until one succeeds.
 795          *
 796          * The Bourne shell can be used instead of init if we are
 797          * trying to recover a really broken machine.
 798          */
 799         if (execute_command) {
 800                 run_init_process(execute_command);
 801                 printk(KERN_WARNING "Failed to execute %s.  Attempting "
 802                                         "defaults...\n", execute_command);
 803         }
 804         run_init_process("/sbin/init");
 805         run_init_process("/etc/init");
 806         run_init_process("/bin/init");
 807         run_init_process("/bin/sh");
 808
 809         panic("No init found.  Try passing init= option to kernel.");
 810 }
 811
 812 static int __init kernel_init(void * unused)
 813 {
 814         lock_kernel();
 815         /*
 816          * init can run on any cpu.
 817          */
 818         set_cpus_allowed(current, CPU_MASK_ALL);
 819         /*
 820          * Tell the world that we're going to be the grim
 821          * reaper of innocent orphaned children.
 822          *
 823          * We don't want people to have to make incorrect
 824          * assumptions about where in the task array this
 825          * can be found.
 826          */
 827         init_pid_ns.child_reaper = current;
 828
 829         __set_special_pids(1, 1);
 830         cad_pid = task_pid(current);
 831
 832         smp_prepare_cpus(setup_max_cpus);
 833
 834         do_pre_smp_initcalls();
 835
 836         smp_init();
 837         sched_init_smp();
 838
 839         cpuset_init_smp();
 840
 841         do_basic_setup();
 842
 843         /*
 844          * check if there is an early userspace init.  If yes, let it do all
 845          * the work
 846          */
 847
 848         if (!ramdisk_execute_command)
 849                 ramdisk_execute_command = "/init";
 850
 851         if (sys_access((const char __user *) ramdisk_execute_command, 0) != 0) {
 852                 ramdisk_execute_command = NULL;
 853                 prepare_namespace();
 854         }
 855
 856         /*
 857          * Ok, we have completed the initial bootup, and
 858          * we're essentially up and running. Get rid of the
 859          * initmem segments and start the user-mode stuff..
 860          */
 861         init_post();
 862         return 0;
 863 }