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/efi.h>
  43 #include <linux/tick.h>
  44 #include <linux/interrupt.h>
  45 #include <linux/taskstats_kern.h>
  46 #include <linux/delayacct.h>
  47 #include <linux/unistd.h>
  48 #include <linux/rmap.h>
  49 #include <linux/mempolicy.h>
  50 #include <linux/key.h>
  51 #include <linux/unwind.h>
  52 #include <linux/buffer_head.h>
  53 #include <linux/debug_locks.h>
  54 #include <linux/lockdep.h>
  55 #include <linux/pid_namespace.h>
  56 #include <linux/device.h>
  57 #include <linux/kthread.h>
  58
  59 #include <asm/io.h>
  60 #include <asm/bugs.h>
  61 #include <asm/setup.h>
  62 #include <asm/sections.h>
  63 #include <asm/cacheflush.h>
  64
  65 #ifdef CONFIG_X86_LOCAL_APIC
  66 #include <asm/smp.h>
  67 #endif
  68
  69 /*
  70  * This is one of the first .c files built. Error out early if we have compiler
  71  * trouble.
  72  *
  73  * Versions of gcc older than that listed below may actually compile and link
  74  * okay, but the end product can have subtle run time bugs.  To avoid associated
  75  * bogus bug reports, we flatly refuse to compile with a gcc that is known to be
  76  * too old from the very beginning.
  77  */
  78 #if (__GNUC__ < 3) || (__GNUC__ == 3 && __GNUC_MINOR__ < 2)
  79 #error Sorry, your GCC is too old. It builds incorrect kernels.
  80 #endif
  81
  82 #if __GNUC__ == 4 && __GNUC_MINOR__ == 1 && __GNUC_PATCHLEVEL__ == 0
  83 #warning gcc-4.1.0 is known to miscompile the kernel.  A different compiler version is recommended.
  84 #endif
  85
  86 static int kernel_init(void *);
  87
  88 extern void init_IRQ(void);
  89 extern void fork_init(unsigned long);
  90 extern void mca_init(void);
  91 extern void sbus_init(void);
  92 extern void signals_init(void);
  93 extern void pidhash_init(void);
  94 extern void pidmap_init(void);
  95 extern void prio_tree_init(void);
  96 extern void radix_tree_init(void);
  97 extern void free_initmem(void);
  98 #ifdef  CONFIG_ACPI
  99 extern void acpi_early_init(void);
 100 #else
 101 static inline void acpi_early_init(void) { }
 102 #endif
 103 #ifndef CONFIG_DEBUG_RODATA
 104 static inline void mark_rodata_ro(void) { }
 105 #endif
 106
 107 #ifdef CONFIG_TC
 108 extern void tc_init(void);
 109 #endif
 110
 111 enum system_states system_state;
 112 EXPORT_SYMBOL(system_state);
 113
 114 /*
 115  * Boot command-line arguments
 116  */
 117 #define MAX_INIT_ARGS CONFIG_INIT_ENV_ARG_LIMIT
 118 #define MAX_INIT_ENVS CONFIG_INIT_ENV_ARG_LIMIT
 119
 120 extern void time_init(void);
 121 /* Default late time init is NULL. archs can override this later. */
 122 void (*late_time_init)(void);
 123 extern void softirq_init(void);
 124
 125 /* Untouched command line saved by arch-specific code. */
 126 char __initdata boot_command_line[COMMAND_LINE_SIZE];
 127 /* Untouched saved command line (eg. for /proc) */
 128 char *saved_command_line;
 129 /* Command line for parameter parsing */
 130 static char *static_command_line;
 131
 132 static char *execute_command;
 133 static char *ramdisk_execute_command;
 134
 135 #ifdef CONFIG_SMP
 136 /* Setup configured maximum number of CPUs to activate */
 137 static unsigned int __initdata max_cpus = NR_CPUS;
 138
 139 /*
 140  * Setup routine for controlling SMP activation
 141  *
 142  * Command-line option of "nosmp" or "maxcpus=0" will disable SMP
 143  * activation entirely (the MPS table probe still happens, though).
 144  *
 145  * Command-line option of "maxcpus=<NUM>", where <NUM> is an integer
 146  * greater than 0, limits the maximum number of CPUs activated in
 147  * SMP mode to <NUM>.
 148  */
 149 static int __init nosmp(char *str)
 150 {
 151         max_cpus = 0;
 152         return 0;
 153 }
 154
 155 early_param("nosmp", nosmp);
 156
 157 static int __init maxcpus(char *str)
 158 {
 159         get_option(&str, &max_cpus);
 160         return 1;
 161 }
 162
 163 __setup("maxcpus=", maxcpus);
 164 #else
 165 #define max_cpus NR_CPUS
 166 #endif
 167
 168 /*
 169  * If set, this is an indication to the drivers that reset the underlying
 170  * device before going ahead with the initialization otherwise driver might
 171  * rely on the BIOS and skip the reset operation.
 172  *
 173  * This is useful if kernel is booting in an unreliable environment.
 174  * For ex. kdump situaiton where previous kernel has crashed, BIOS has been
 175  * skipped and devices will be in unknown state.
 176  */
 177 unsigned int reset_devices;
 178 EXPORT_SYMBOL(reset_devices);
 179
 180 static int __init set_reset_devices(char *str)
 181 {
 182         reset_devices = 1;
 183         return 1;
 184 }
 185
 186 __setup("reset_devices", set_reset_devices);
 187
 188 static char * argv_init[MAX_INIT_ARGS+2] = { "init", NULL, };
 189 char * envp_init[MAX_INIT_ENVS+2] = { "HOME=/", "TERM=linux", NULL, };
 190 static const char *panic_later, *panic_param;
 191
 192 extern struct obs_kernel_param __setup_start[], __setup_end[];
 193
 194 static int __init obsolete_checksetup(char *line)
 195 {
 196         struct obs_kernel_param *p;
 197         int had_early_param = 0;
 198
 199         p = __setup_start;
 200         do {
 201                 int n = strlen(p->str);
 202                 if (!strncmp(line, p->str, n)) {
 203                         if (p->early) {
 204                                 /* Already done in parse_early_param?
 205                                  * (Needs exact match on param part).
 206                                  * Keep iterating, as we can have early
 207                                  * params and __setups of same names 8( */
 208                                 if (line[n] == '\0' || line[n] == '=')
 209                                         had_early_param = 1;
 210                         } else if (!p->setup_func) {
 211                                 printk(KERN_WARNING "Parameter %s is obsolete,"
 212                                        " ignored\n", p->str);
 213                                 return 1;
 214                         } else if (p->setup_func(line + n))
 215                                 return 1;
 216                 }
 217                 p++;
 218         } while (p < __setup_end);
 219
 220         return had_early_param;
 221 }
 222
 223 /*
 224  * This should be approx 2 Bo*oMips to start (note initial shift), and will
 225  * still work even if initially too large, it will just take slightly longer
 226  */
 227 unsigned long loops_per_jiffy = (1<<12);
 228
 229 EXPORT_SYMBOL(loops_per_jiffy);
 230
 231 static int __init debug_kernel(char *str)
 232 {
 233         if (*str)
 234                 return 0;
 235         console_loglevel = 10;
 236         return 1;
 237 }
 238
 239 static int __init quiet_kernel(char *str)
 240 {
 241         if (*str)
 242                 return 0;
 243         console_loglevel = 4;
 244         return 1;
 245 }
 246
 247 __setup("debug", debug_kernel);
 248 __setup("quiet", quiet_kernel);
 249
 250 static int __init loglevel(char *str)
 251 {
 252         get_option(&str, &console_loglevel);
 253         return 1;
 254 }
 255
 256 __setup("loglevel=", loglevel);
 257
 258 /*
 259  * Unknown boot options get handed to init, unless they look like
 260  * failed parameters
 261  */
 262 static int __init unknown_bootoption(char *param, char *val)
 263 {
 264         /* Change NUL term back to "=", to make "param" the whole string. */
 265         if (val) {
 266                 /* param=val or param="val"? */
 267                 if (val == param+strlen(param)+1)
 268                         val[-1] = '=';
 269                 else if (val == param+strlen(param)+2) {
 270                         val[-2] = '=';
 271                         memmove(val-1, val, strlen(val)+1);
 272                         val--;
 273                 } else
 274                         BUG();
 275         }
 276
 277         /* Handle obsolete-style parameters */
 278         if (obsolete_checksetup(param))
 279                 return 0;
 280
 281         /*
 282          * Preemptive maintenance for "why didn't my mispelled command
 283          * line work?"
 284          */
 285         if (strchr(param, '.') && (!val || strchr(param, '.') < val)) {
 286                 printk(KERN_ERR "Unknown boot option `%s': ignoring\n", param);
 287                 return 0;
 288         }
 289
 290         if (panic_later)
 291                 return 0;
 292
 293         if (val) {
 294                 /* Environment option */
 295                 unsigned int i;
 296                 for (i = 0; envp_init[i]; i++) {
 297                         if (i == MAX_INIT_ENVS) {
 298                                 panic_later = "Too many boot env vars at `%s'";
 299                                 panic_param = param;
 300                         }
 301                         if (!strncmp(param, envp_init[i], val - param))
 302                                 break;
 303                 }
 304                 envp_init[i] = param;
 305         } else {
 306                 /* Command line option */
 307                 unsigned int i;
 308                 for (i = 0; argv_init[i]; i++) {
 309                         if (i == MAX_INIT_ARGS) {
 310                                 panic_later = "Too many boot init vars at `%s'";
 311                                 panic_param = param;
 312                         }
 313                 }
 314                 argv_init[i] = param;
 315         }
 316         return 0;
 317 }
 318
 319 static int __init init_setup(char *str)
 320 {
 321         unsigned int i;
 322
 323         execute_command = str;
 324         /*
 325          * In case LILO is going to boot us with default command line,
 326          * it prepends "auto" before the whole cmdline which makes
 327          * the shell think it should execute a script with such name.
 328          * So we ignore all arguments entered _before_ init=... [MJ]
 329          */
 330         for (i = 1; i < MAX_INIT_ARGS; i++)
 331                 argv_init[i] = NULL;
 332         return 1;
 333 }
 334 __setup("init=", init_setup);
 335
 336 static int __init rdinit_setup(char *str)
 337 {
 338         unsigned int i;
 339
 340         ramdisk_execute_command = str;
 341         /* See "auto" comment in init_setup */
 342         for (i = 1; i < MAX_INIT_ARGS; i++)
 343                 argv_init[i] = NULL;
 344         return 1;
 345 }
 346 __setup("rdinit=", rdinit_setup);
 347
 348 #ifndef CONFIG_SMP
 349
 350 #ifdef CONFIG_X86_LOCAL_APIC
 351 static void __init smp_init(void)
 352 {
 353         APIC_init_uniprocessor();
 354 }
 355 #else
 356 #define smp_init()      do { } while (0)
 357 #endif
 358
 359 static inline void setup_per_cpu_areas(void) { }
 360 static inline void smp_prepare_cpus(unsigned int maxcpus) { }
 361
 362 #else
 363
 364 #ifdef __GENERIC_PER_CPU
 365 unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
 366
 367 EXPORT_SYMBOL(__per_cpu_offset);
 368
 369 static void __init setup_per_cpu_areas(void)
 370 {
 371         unsigned long size, i;
 372         char *ptr;
 373         unsigned long nr_possible_cpus = num_possible_cpus();
 374
 375         /* Copy section for each CPU (we discard the original) */
 376         size = ALIGN(PERCPU_ENOUGH_ROOM, PAGE_SIZE);
 377         ptr = alloc_bootmem_pages(size * nr_possible_cpus);
 378
 379         for_each_possible_cpu(i) {
 380                 __per_cpu_offset[i] = ptr - __per_cpu_start;
 381                 memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
 382                 ptr += size;
 383         }
 384 }
 385 #endif /* !__GENERIC_PER_CPU */
 386
 387 /* Called by boot processor to activate the rest. */
 388 static void __init smp_init(void)
 389 {
 390         unsigned int cpu;
 391
 392 #ifndef CONFIG_HOTPLUG_CPU
 393         cpu_possible_map = cpu_present_map;
 394 #endif
 395
 396         /* FIXME: This should be done in userspace --RR */
 397         for_each_present_cpu(cpu) {
 398                 if (num_online_cpus() >= max_cpus)
 399                         break;
 400                 if (!cpu_online(cpu))
 401                         cpu_up(cpu);
 402         }
 403
 404         /* Any cleanup work */
 405         printk(KERN_INFO "Brought up %ld CPUs\n", (long)num_online_cpus());
 406         smp_cpus_done(max_cpus);
 407 }
 408
 409 #endif
 410
 411 /*
 412  * We need to store the untouched command line for future reference.
 413  * We also need to store the touched command line since the parameter
 414  * parsing is performed in place, and we should allow a component to
 415  * store reference of name/value for future reference.
 416  */
 417 static void __init setup_command_line(char *command_line)
 418 {
 419         saved_command_line = alloc_bootmem(strlen (boot_command_line)+1);
 420         static_command_line = alloc_bootmem(strlen (command_line)+1);
 421         strcpy (saved_command_line, boot_command_line);
 422         strcpy (static_command_line, command_line);
 423 }
 424
 425 /*
 426  * We need to finalize in a non-__init function or else race conditions
 427  * between the root thread and the init thread may cause start_kernel to
 428  * be reaped by free_initmem before the root thread has proceeded to
 429  * cpu_idle.
 430  *
 431  * gcc-3.4 accidentally inlines this function, so use noinline.
 432  */
 433
 434 static void noinline __init_refok rest_init(void)
 435         __releases(kernel_lock)
 436 {
 437         int pid;
 438
 439         kernel_thread(kernel_init, NULL, CLONE_FS | CLONE_SIGHAND);
 440         numa_default_policy();
 441         pid = kernel_thread(kthreadd, NULL, CLONE_FS | CLONE_FILES);
 442         kthreadd_task = find_task_by_pid(pid);
 443         unlock_kernel();
 444
 445         /*
 446          * The boot idle thread must execute schedule()
 447          * at least once to get things moving:
 448          */
 449         init_idle_bootup_task(current);
 450         preempt_enable_no_resched();
 451         schedule();
 452         preempt_disable();
 453
 454         /* Call into cpu_idle with preempt disabled */
 455         cpu_idle();
 456 }
 457
 458 /* Check for early params. */
 459 static int __init do_early_param(char *param, char *val)
 460 {
 461         struct obs_kernel_param *p;
 462
 463         for (p = __setup_start; p < __setup_end; p++) {
 464                 if ((p->early && strcmp(param, p->str) == 0) ||
 465                     (strcmp(param, "console") == 0 &&
 466                      strcmp(p->str, "earlycon") == 0)
 467                 ) {
 468                         if (p->setup_func(val) != 0)
 469                                 printk(KERN_WARNING
 470                                        "Malformed early option '%s'\n", param);
 471                 }
 472         }
 473         /* We accept everything at this stage. */
 474         return 0;
 475 }
 476
 477 /* Arch code calls this early on, or if not, just before other parsing. */
 478 void __init parse_early_param(void)
 479 {
 480         static __initdata int done = 0;
 481         static __initdata char tmp_cmdline[COMMAND_LINE_SIZE];
 482
 483         if (done)
 484                 return;
 485
 486         /* All fall through to do_early_param. */
 487         strlcpy(tmp_cmdline, boot_command_line, COMMAND_LINE_SIZE);
 488         parse_args("early options", tmp_cmdline, NULL, 0, do_early_param);
 489         done = 1;
 490 }
 491
 492 /*
 493  *      Activate the first processor.
 494  */
 495
 496 static void __init boot_cpu_init(void)
 497 {
 498         int cpu = smp_processor_id();
 499         /* Mark the boot cpu "present", "online" etc for SMP and UP case */
 500         cpu_set(cpu, cpu_online_map);
 501         cpu_set(cpu, cpu_present_map);
 502         cpu_set(cpu, cpu_possible_map);
 503 }
 504
 505 void __init __attribute__((weak)) smp_setup_processor_id(void)
 506 {
 507 }
 508
 509 asmlinkage void __init start_kernel(void)
 510 {
 511         char * command_line;
 512         extern struct kernel_param __start___param[], __stop___param[];
 513
 514         smp_setup_processor_id();
 515
 516         /*
 517          * Need to run as early as possible, to initialize the
 518          * lockdep hash:
 519          */
 520         unwind_init();
 521         lockdep_init();
 522
 523         local_irq_disable();
 524         early_boot_irqs_off();
 525         early_init_irq_lock_class();
 526
 527 /*
 528  * Interrupts are still disabled. Do necessary setups, then
 529  * enable them
 530  */
 531         lock_kernel();
 532         tick_init();
 533         boot_cpu_init();
 534         page_address_init();
 535         printk(KERN_NOTICE);
 536         printk(linux_banner);
 537         setup_arch(&command_line);
 538         setup_command_line(command_line);
 539         unwind_setup();
 540 #ifndef CONFIG_HOTPLUG_CPU
 541         if (max_cpus < 2)
 542                 cpu_possible_map = cpu_online_map;
 543 #endif
 544         setup_per_cpu_areas();
 545         smp_prepare_boot_cpu(); /* arch-specific boot-cpu hooks */
 546
 547         /*
 548          * Set up the scheduler prior starting any interrupts (such as the
 549          * timer interrupt). Full topology setup happens at smp_init()
 550          * time - but meanwhile we still have a functioning scheduler.
 551          */
 552         sched_init();
 553         /*
 554          * Disable preemption - early bootup scheduling is extremely
 555          * fragile until we cpu_idle() for the first time.
 556          */
 557         preempt_disable();
 558         build_all_zonelists();
 559         page_alloc_init();
 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         kmem_cache_init();
 616         setup_per_cpu_pageset();
 617         numa_policy_init();
 618         if (late_time_init)
 619                 late_time_init();
 620         calibrate_delay();
 621         pidmap_init();
 622         pgtable_cache_init();
 623         prio_tree_init();
 624         anon_vma_init();
 625 #ifdef CONFIG_X86
 626         if (efi_enabled)
 627                 efi_enter_virtual_mode();
 628 #endif
 629         fork_init(num_physpages);
 630         proc_caches_init();
 631         buffer_init();
 632         unnamed_dev_init();
 633         key_init();
 634         security_init();
 635         vfs_caches_init(num_physpages);
 636         radix_tree_init();
 637         signals_init();
 638         /* rootfs populating might need page-writeback */
 639         page_writeback_init();
 640 #ifdef CONFIG_PROC_FS
 641         proc_root_init();
 642 #endif
 643         cpuset_init();
 644         taskstats_init_early();
 645         delayacct_init();
 646
 647         check_bugs();
 648
 649         acpi_early_init(); /* before LAPIC and SMP init */
 650
 651         /* Do the rest non-__init'ed, we're now alive */
 652         rest_init();
 653 }
 654
 655 static int __initdata initcall_debug;
 656
 657 static int __init initcall_debug_setup(char *str)
 658 {
 659         initcall_debug = 1;
 660         return 1;
 661 }
 662 __setup("initcall_debug", initcall_debug_setup);
 663
 664 extern initcall_t __initcall_start[], __initcall_end[];
 665
 666 static void __init do_initcalls(void)
 667 {
 668         initcall_t *call;
 669         int count = preempt_count();
 670
 671         for (call = __initcall_start; call < __initcall_end; call++) {
 672                 ktime_t t0, t1, delta;
 673                 char *msg = NULL;
 674                 char msgbuf[40];
 675                 int result;
 676
 677                 if (initcall_debug) {
 678                         printk("Calling initcall 0x%p", *call);
 679                         print_fn_descriptor_symbol(": %s()",
 680                                         (unsigned long) *call);
 681                         printk("\n");
 682                         t0 = ktime_get();
 683                 }
 684
 685                 result = (*call)();
 686
 687                 if (initcall_debug) {
 688                         t1 = ktime_get();
 689                         delta = ktime_sub(t1, t0);
 690
 691                         printk("initcall 0x%p", *call);
 692                         print_fn_descriptor_symbol(": %s()",
 693                                         (unsigned long) *call);
 694                         printk(" returned %d.\n", result);
 695
 696                         printk("initcall 0x%p ran for %Ld msecs: ",
 697                                 *call, (unsigned long long)delta.tv64 >> 20);
 698                         print_fn_descriptor_symbol("%s()\n",
 699                                 (unsigned long) *call);
 700                 }
 701
 702                 if (result && result != -ENODEV && initcall_debug) {
 703                         sprintf(msgbuf, "error code %d", result);
 704                         msg = msgbuf;
 705                 }
 706                 if (preempt_count() != count) {
 707                         msg = "preemption imbalance";
 708                         preempt_count() = count;
 709                 }
 710                 if (irqs_disabled()) {
 711                         msg = "disabled interrupts";
 712                         local_irq_enable();
 713                 }
 714                 if (msg) {
 715                         printk(KERN_WARNING "initcall at 0x%p", *call);
 716                         print_fn_descriptor_symbol(": %s()",
 717                                         (unsigned long) *call);
 718                         printk(": returned with %s\n", msg);
 719                 }
 720         }
 721
 722         /* Make sure there is no pending stuff from the initcall sequence */
 723         flush_scheduled_work();
 724 }
 725
 726 /*
 727  * Ok, the machine is now initialized. None of the devices
 728  * have been touched yet, but the CPU subsystem is up and
 729  * running, and memory and process management works.
 730  *
 731  * Now we can finally start doing some real work..
 732  */
 733 static void __init do_basic_setup(void)
 734 {
 735         /* drivers will send hotplug events */
 736         init_workqueues();
 737         usermodehelper_init();
 738         driver_init();
 739         init_irq_proc();
 740         do_initcalls();
 741 }
 742
 743 static int __initdata nosoftlockup;
 744
 745 static int __init nosoftlockup_setup(char *str)
 746 {
 747         nosoftlockup = 1;
 748         return 1;
 749 }
 750 __setup("nosoftlockup", nosoftlockup_setup);
 751
 752 static void __init do_pre_smp_initcalls(void)
 753 {
 754         extern int spawn_ksoftirqd(void);
 755 #ifdef CONFIG_SMP
 756         extern int migration_init(void);
 757
 758         migration_init();
 759 #endif
 760         spawn_ksoftirqd();
 761         if (!nosoftlockup)
 762                 spawn_softlockup_task();
 763 }
 764
 765 static void run_init_process(char *init_filename)
 766 {
 767         argv_init[0] = init_filename;
 768         kernel_execve(init_filename, argv_init, envp_init);
 769 }
 770
 771 /* This is a non __init function. Force it to be noinline otherwise gcc
 772  * makes it inline to init() and it becomes part of init.text section
 773  */
 774 static int noinline init_post(void)
 775 {
 776         free_initmem();
 777         unlock_kernel();
 778         mark_rodata_ro();
 779         system_state = SYSTEM_RUNNING;
 780         numa_default_policy();
 781
 782         if (sys_open((const char __user *) "/dev/console", O_RDWR, 0) < 0)
 783                 printk(KERN_WARNING "Warning: unable to open an initial console.\n");
 784
 785         (void) sys_dup(0);
 786         (void) sys_dup(0);
 787
 788         if (ramdisk_execute_command) {
 789                 run_init_process(ramdisk_execute_command);
 790                 printk(KERN_WARNING "Failed to execute %s\n",
 791                                 ramdisk_execute_command);
 792         }
 793
 794         /*
 795          * We try each of these until one succeeds.
 796          *
 797          * The Bourne shell can be used instead of init if we are
 798          * trying to recover a really broken machine.
 799          */
 800         if (execute_command) {
 801                 run_init_process(execute_command);
 802                 printk(KERN_WARNING "Failed to execute %s.  Attempting "
 803                                         "defaults...\n", execute_command);
 804         }
 805         run_init_process("/sbin/init");
 806         run_init_process("/etc/init");
 807         run_init_process("/bin/init");
 808         run_init_process("/bin/sh");
 809
 810         panic("No init found.  Try passing init= option to kernel.");
 811 }
 812
 813 static int __init kernel_init(void * unused)
 814 {
 815         lock_kernel();
 816         /*
 817          * init can run on any cpu.
 818          */
 819         set_cpus_allowed(current, CPU_MASK_ALL);
 820         /*
 821          * Tell the world that we're going to be the grim
 822          * reaper of innocent orphaned children.
 823          *
 824          * We don't want people to have to make incorrect
 825          * assumptions about where in the task array this
 826          * can be found.
 827          */
 828         init_pid_ns.child_reaper = current;
 829
 830         __set_special_pids(1, 1);
 831         cad_pid = task_pid(current);
 832
 833         smp_prepare_cpus(max_cpus);
 834
 835         do_pre_smp_initcalls();
 836
 837         smp_init();
 838         sched_init_smp();
 839
 840         cpuset_init_smp();
 841
 842         do_basic_setup();
 843
 844         /*
 845          * check if there is an early userspace init.  If yes, let it do all
 846          * the work
 847          */
 848
 849         if (!ramdisk_execute_command)
 850                 ramdisk_execute_command = "/init";
 851
 852         if (sys_access((const char __user *) ramdisk_execute_command, 0) != 0) {
 853                 ramdisk_execute_command = NULL;
 854                 prepare_namespace();
 855         }
 856
 857         /*
 858          * Ok, we have completed the initial bootup, and
 859          * we're essentially up and running. Get rid of the
 860          * initmem segments and start the user-mode stuff..
 861          */
 862         init_post();
 863         return 0;
 864 }