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         console_loglevel = 10;
 236         return 0;
 237 }
 238
 239 static int __init quiet_kernel(char *str)
 240 {
 241         console_loglevel = 4;
 242         return 0;
 243 }
 244
 245 early_param("debug", debug_kernel);
 246 early_param("quiet", quiet_kernel);
 247
 248 static int __init loglevel(char *str)
 249 {
 250         get_option(&str, &console_loglevel);
 251         return 0;
 252 }
 253
 254 early_param("loglevel", loglevel);
 255
 256 /*
 257  * Unknown boot options get handed to init, unless they look like
 258  * failed parameters
 259  */
 260 static int __init unknown_bootoption(char *param, char *val)
 261 {
 262         /* Change NUL term back to "=", to make "param" the whole string. */
 263         if (val) {
 264                 /* param=val or param="val"? */
 265                 if (val == param+strlen(param)+1)
 266                         val[-1] = '=';
 267                 else if (val == param+strlen(param)+2) {
 268                         val[-2] = '=';
 269                         memmove(val-1, val, strlen(val)+1);
 270                         val--;
 271                 } else
 272                         BUG();
 273         }
 274
 275         /* Handle obsolete-style parameters */
 276         if (obsolete_checksetup(param))
 277                 return 0;
 278
 279         /*
 280          * Preemptive maintenance for "why didn't my misspelled command
 281          * line work?"
 282          */
 283         if (strchr(param, '.') && (!val || strchr(param, '.') < val)) {
 284                 printk(KERN_ERR "Unknown boot option `%s': ignoring\n", param);
 285                 return 0;
 286         }
 287
 288         if (panic_later)
 289                 return 0;
 290
 291         if (val) {
 292                 /* Environment option */
 293                 unsigned int i;
 294                 for (i = 0; envp_init[i]; i++) {
 295                         if (i == MAX_INIT_ENVS) {
 296                                 panic_later = "Too many boot env vars at `%s'";
 297                                 panic_param = param;
 298                         }
 299                         if (!strncmp(param, envp_init[i], val - param))
 300                                 break;
 301                 }
 302                 envp_init[i] = param;
 303         } else {
 304                 /* Command line option */
 305                 unsigned int i;
 306                 for (i = 0; argv_init[i]; i++) {
 307                         if (i == MAX_INIT_ARGS) {
 308                                 panic_later = "Too many boot init vars at `%s'";
 309                                 panic_param = param;
 310                         }
 311                 }
 312                 argv_init[i] = param;
 313         }
 314         return 0;
 315 }
 316
 317 #ifdef CONFIG_DEBUG_PAGEALLOC
 318 int __read_mostly debug_pagealloc_enabled = 0;
 319 #endif
 320
 321 static 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
 338 static 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
 351
 352 #ifdef CONFIG_X86_LOCAL_APIC
 353 static void __init smp_init(void)
 354 {
 355         APIC_init_uniprocessor();
 356 }
 357 #else
 358 #define smp_init()      do { } while (0)
 359 #endif
 360
 361 static inline void setup_per_cpu_areas(void) { }
 362 static inline void setup_nr_cpu_ids(void) { }
 363 static inline void smp_prepare_cpus(unsigned int maxcpus) { }
 364
 365 #else
 366
 367 #if NR_CPUS > BITS_PER_LONG
 368 cpumask_t cpu_mask_all __read_mostly = CPU_MASK_ALL;
 369 EXPORT_SYMBOL(cpu_mask_all);
 370 #endif
 371
 372 /* Setup number of possible processor ids */
 373 int nr_cpu_ids __read_mostly = NR_CPUS;
 374 EXPORT_SYMBOL(nr_cpu_ids);
 375
 376 /* An arch may set nr_cpu_ids earlier if needed, so this would be redundant */
 377 static void __init setup_nr_cpu_ids(void)
 378 {
 379         int cpu, highest_cpu = 0;
 380
 381         for_each_possible_cpu(cpu)
 382                 highest_cpu = cpu;
 383
 384         nr_cpu_ids = highest_cpu + 1;
 385 }
 386
 387 #ifndef CONFIG_HAVE_SETUP_PER_CPU_AREA
 388 unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
 389
 390 EXPORT_SYMBOL(__per_cpu_offset);
 391
 392 static void __init setup_per_cpu_areas(void)
 393 {
 394         unsigned long size, i;
 395         char *ptr;
 396         unsigned long nr_possible_cpus = num_possible_cpus();
 397
 398         /* Copy section for each CPU (we discard the original) */
 399         size = ALIGN(PERCPU_ENOUGH_ROOM, PAGE_SIZE);
 400         ptr = alloc_bootmem_pages(size * nr_possible_cpus);
 401
 402         for_each_possible_cpu(i) {
 403                 __per_cpu_offset[i] = ptr - __per_cpu_start;
 404                 memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
 405                 ptr += size;
 406         }
 407 }
 408 #endif /* CONFIG_HAVE_SETUP_PER_CPU_AREA */
 409
 410 /* Called by boot processor to activate the rest. */
 411 static void __init smp_init(void)
 412 {
 413         unsigned int cpu;
 414
 415         /* FIXME: This should be done in userspace --RR */
 416         for_each_present_cpu(cpu) {
 417                 if (num_online_cpus() >= setup_max_cpus)
 418                         break;
 419                 if (!cpu_online(cpu))
 420                         cpu_up(cpu);
 421         }
 422
 423         /* Any cleanup work */
 424         printk(KERN_INFO "Brought up %ld CPUs\n", (long)num_online_cpus());
 425         smp_cpus_done(setup_max_cpus);
 426 }
 427
 428 #endif
 429
 430 /*
 431  * We need to store the untouched command line for future reference.
 432  * We also need to store the touched command line since the parameter
 433  * parsing is performed in place, and we should allow a component to
 434  * store reference of name/value for future reference.
 435  */
 436 static void __init setup_command_line(char *command_line)
 437 {
 438         saved_command_line = alloc_bootmem(strlen (boot_command_line)+1);
 439         static_command_line = alloc_bootmem(strlen (command_line)+1);
 440         strcpy (saved_command_line, boot_command_line);
 441         strcpy (static_command_line, command_line);
 442 }
 443
 444 /*
 445  * We need to finalize in a non-__init function or else race conditions
 446  * between the root thread and the init thread may cause start_kernel to
 447  * be reaped by free_initmem before the root thread has proceeded to
 448  * cpu_idle.
 449  *
 450  * gcc-3.4 accidentally inlines this function, so use noinline.
 451  */
 452
 453 static void noinline __init_refok rest_init(void)
 454         __releases(kernel_lock)
 455 {
 456         int pid;
 457
 458         kernel_thread(kernel_init, NULL, CLONE_FS | CLONE_SIGHAND);
 459         numa_default_policy();
 460         pid = kernel_thread(kthreadd, NULL, CLONE_FS | CLONE_FILES);
 461         kthreadd_task = find_task_by_pid(pid);
 462         unlock_kernel();
 463
 464         /*
 465          * The boot idle thread must execute schedule()
 466          * at least once to get things moving:
 467          */
 468         init_idle_bootup_task(current);
 469         preempt_enable_no_resched();
 470         schedule();
 471         preempt_disable();
 472
 473         /* Call into cpu_idle with preempt disabled */
 474         cpu_idle();
 475 }
 476
 477 /* Check for early params. */
 478 static int __init do_early_param(char *param, char *val)
 479 {
 480         struct obs_kernel_param *p;
 481
 482         for (p = __setup_start; p < __setup_end; p++) {
 483                 if ((p->early && strcmp(param, p->str) == 0) ||
 484                     (strcmp(param, "console") == 0 &&
 485                      strcmp(p->str, "earlycon") == 0)
 486                 ) {
 487                         if (p->setup_func(val) != 0)
 488                                 printk(KERN_WARNING
 489                                        "Malformed early option '%s'\n", param);
 490                 }
 491         }
 492         /* We accept everything at this stage. */
 493         return 0;
 494 }
 495
 496 /* Arch code calls this early on, or if not, just before other parsing. */
 497 void __init parse_early_param(void)
 498 {
 499         static __initdata int done = 0;
 500         static __initdata char tmp_cmdline[COMMAND_LINE_SIZE];
 501
 502         if (done)
 503                 return;
 504
 505         /* All fall through to do_early_param. */
 506         strlcpy(tmp_cmdline, boot_command_line, COMMAND_LINE_SIZE);
 507         parse_args("early options", tmp_cmdline, NULL, 0, do_early_param);
 508         done = 1;
 509 }
 510
 511 /*
 512  *      Activate the first processor.
 513  */
 514
 515 static void __init boot_cpu_init(void)
 516 {
 517         int cpu = smp_processor_id();
 518         /* Mark the boot cpu "present", "online" etc for SMP and UP case */
 519         cpu_set(cpu, cpu_online_map);
 520         cpu_set(cpu, cpu_present_map);
 521         cpu_set(cpu, cpu_possible_map);
 522 }
 523
 524 void __init __weak smp_setup_processor_id(void)
 525 {
 526 }
 527
 528 void __init __weak thread_info_cache_init(void)
 529 {
 530 }
 531
 532 asmlinkage void __init start_kernel(void)
 533 {
 534         char * command_line;
 535         extern struct kernel_param __start___param[], __stop___param[];
 536
 537         smp_setup_processor_id();
 538
 539         /*
 540          * Need to run as early as possible, to initialize the
 541          * lockdep hash:
 542          */
 543         unwind_init();
 544         lockdep_init();
 545         cgroup_init_early();
 546
 547         local_irq_disable();
 548         early_boot_irqs_off();
 549         early_init_irq_lock_class();
 550
 551 /*
 552  * Interrupts are still disabled. Do necessary setups, then
 553  * enable them
 554  */
 555         lock_kernel();
 556         tick_init();
 557         boot_cpu_init();
 558         page_address_init();
 559         printk(KERN_NOTICE);
 560         printk(linux_banner);
 561         setup_arch(&command_line);
 562         setup_command_line(command_line);
 563         unwind_setup();
 564         setup_per_cpu_areas();
 565         setup_nr_cpu_ids();
 566         smp_prepare_boot_cpu(); /* arch-specific boot-cpu hooks */
 567
 568         /*
 569          * Set up the scheduler prior starting any interrupts (such as the
 570          * timer interrupt). Full topology setup happens at smp_init()
 571          * time - but meanwhile we still have a functioning scheduler.
 572          */
 573         sched_init();
 574         /*
 575          * Disable preemption - early bootup scheduling is extremely
 576          * fragile until we cpu_idle() for the first time.
 577          */
 578         preempt_disable();
 579         build_all_zonelists();
 580         page_alloc_init();
 581         printk(KERN_NOTICE "Kernel command line: %s\n", boot_command_line);
 582         parse_early_param();
 583         parse_args("Booting kernel", static_command_line, __start___param,
 584                    __stop___param - __start___param,
 585                    &unknown_bootoption);
 586         if (!irqs_disabled()) {
 587                 printk(KERN_WARNING "start_kernel(): bug: interrupts were "
 588                                 "enabled *very* early, fixing it\n");
 589                 local_irq_disable();
 590         }
 591         sort_main_extable();
 592         trap_init();
 593         rcu_init();
 594         init_IRQ();
 595         pidhash_init();
 596         init_timers();
 597         hrtimers_init();
 598         softirq_init();
 599         timekeeping_init();
 600         time_init();
 601         profile_init();
 602         if (!irqs_disabled())
 603                 printk("start_kernel(): bug: interrupts were enabled early\n");
 604         early_boot_irqs_on();
 605         local_irq_enable();
 606
 607         /*
 608          * HACK ALERT! This is early. We're enabling the console before
 609          * we've done PCI setups etc, and console_init() must be aware of
 610          * this. But we do want output early, in case something goes wrong.
 611          */
 612         console_init();
 613         if (panic_later)
 614                 panic(panic_later, panic_param);
 615
 616         lockdep_info();
 617
 618         /*
 619          * Need to run this when irqs are enabled, because it wants
 620          * to self-test [hard/soft]-irqs on/off lock inversion bugs
 621          * too:
 622          */
 623         locking_selftest();
 624
 625 #ifdef CONFIG_BLK_DEV_INITRD
 626         if (initrd_start && !initrd_below_start_ok &&
 627                         initrd_start < min_low_pfn << PAGE_SHIFT) {
 628                 printk(KERN_CRIT "initrd overwritten (0x%08lx < 0x%08lx) - "
 629                     "disabling it.\n",initrd_start,min_low_pfn << PAGE_SHIFT);
 630                 initrd_start = 0;
 631         }
 632 #endif
 633         vfs_caches_init_early();
 634         cpuset_init_early();
 635         mem_init();
 636         enable_debug_pagealloc();
 637         cpu_hotplug_init();
 638         kmem_cache_init();
 639         setup_per_cpu_pageset();
 640         numa_policy_init();
 641         if (late_time_init)
 642                 late_time_init();
 643         calibrate_delay();
 644         pidmap_init();
 645         pgtable_cache_init();
 646         prio_tree_init();
 647         anon_vma_init();
 648 #ifdef CONFIG_X86
 649         if (efi_enabled)
 650                 efi_enter_virtual_mode();
 651 #endif
 652         thread_info_cache_init();
 653         fork_init(num_physpages);
 654         proc_caches_init();
 655         buffer_init();
 656         unnamed_dev_init();
 657         key_init();
 658         security_init();
 659         vfs_caches_init(num_physpages);
 660         radix_tree_init();
 661         signals_init();
 662         /* rootfs populating might need page-writeback */
 663         page_writeback_init();
 664 #ifdef CONFIG_PROC_FS
 665         proc_root_init();
 666 #endif
 667         cgroup_init();
 668         cpuset_init();
 669         taskstats_init_early();
 670         delayacct_init();
 671
 672         check_bugs();
 673
 674         acpi_early_init(); /* before LAPIC and SMP init */
 675
 676         /* Do the rest non-__init'ed, we're now alive */
 677         rest_init();
 678 }
 679
 680 static int __initdata initcall_debug;
 681
 682 static int __init initcall_debug_setup(char *str)
 683 {
 684         initcall_debug = 1;
 685         return 1;
 686 }
 687 __setup("initcall_debug", initcall_debug_setup);
 688
 689 extern initcall_t __initcall_start[], __initcall_end[];
 690
 691 static void __init do_initcalls(void)
 692 {
 693         initcall_t *call;
 694         int count = preempt_count();
 695
 696         for (call = __initcall_start; call < __initcall_end; call++) {
 697                 ktime_t t0, t1, delta;
 698                 char *msg = NULL;
 699                 char msgbuf[40];
 700                 int result;
 701
 702                 if (initcall_debug) {
 703                         print_fn_descriptor_symbol("calling  %s()\n",
 704                                         (unsigned long) *call);
 705                         t0 = ktime_get();
 706                 }
 707
 708                 result = (*call)();
 709
 710                 if (initcall_debug) {
 711                         t1 = ktime_get();
 712                         delta = ktime_sub(t1, t0);
 713
 714                         print_fn_descriptor_symbol("initcall %s()",
 715                                         (unsigned long) *call);
 716                         printk(" returned %d after %Ld msecs\n", result,
 717                                 (unsigned long long) delta.tv64 >> 20);
 718                 }
 719
 720                 if (result && result != -ENODEV && initcall_debug) {
 721                         sprintf(msgbuf, "error code %d", result);
 722                         msg = msgbuf;
 723                 }
 724                 if (preempt_count() != count) {
 725                         msg = "preemption imbalance";
 726                         preempt_count() = count;
 727                 }
 728                 if (irqs_disabled()) {
 729                         msg = "disabled interrupts";
 730                         local_irq_enable();
 731                 }
 732                 if (msg) {
 733                         print_fn_descriptor_symbol(KERN_WARNING "initcall %s()",
 734                                         (unsigned long) *call);
 735                         printk(" returned with %s\n", msg);
 736                 }
 737         }
 738
 739         /* Make sure there is no pending stuff from the initcall sequence */
 740         flush_scheduled_work();
 741 }
 742
 743 /*
 744  * Ok, the machine is now initialized. None of the devices
 745  * have been touched yet, but the CPU subsystem is up and
 746  * running, and memory and process management works.
 747  *
 748  * Now we can finally start doing some real work..
 749  */
 750 static void __init do_basic_setup(void)
 751 {
 752         /* drivers will send hotplug events */
 753         init_workqueues();
 754         usermodehelper_init();
 755         driver_init();
 756         init_irq_proc();
 757         do_initcalls();
 758 }
 759
 760 static int __initdata nosoftlockup;
 761
 762 static int __init nosoftlockup_setup(char *str)
 763 {
 764         nosoftlockup = 1;
 765         return 1;
 766 }
 767 __setup("nosoftlockup", nosoftlockup_setup);
 768
 769 static void __init do_pre_smp_initcalls(void)
 770 {
 771         extern int spawn_ksoftirqd(void);
 772
 773         migration_init();
 774         spawn_ksoftirqd();
 775         if (!nosoftlockup)
 776                 spawn_softlockup_task();
 777 }
 778
 779 static void run_init_process(char *init_filename)
 780 {
 781         argv_init[0] = init_filename;
 782         kernel_execve(init_filename, argv_init, envp_init);
 783 }
 784
 785 /* This is a non __init function. Force it to be noinline otherwise gcc
 786  * makes it inline to init() and it becomes part of init.text section
 787  */
 788 static int noinline init_post(void)
 789 {
 790         free_initmem();
 791         unlock_kernel();
 792         mark_rodata_ro();
 793         system_state = SYSTEM_RUNNING;
 794         numa_default_policy();
 795
 796         if (sys_open((const char __user *) "/dev/console", O_RDWR, 0) < 0)
 797                 printk(KERN_WARNING "Warning: unable to open an initial console.\n");
 798
 799         (void) sys_dup(0);
 800         (void) sys_dup(0);
 801
 802         if (ramdisk_execute_command) {
 803                 run_init_process(ramdisk_execute_command);
 804                 printk(KERN_WARNING "Failed to execute %s\n",
 805                                 ramdisk_execute_command);
 806         }
 807
 808         /*
 809          * We try each of these until one succeeds.
 810          *
 811          * The Bourne shell can be used instead of init if we are
 812          * trying to recover a really broken machine.
 813          */
 814         if (execute_command) {
 815                 run_init_process(execute_command);
 816                 printk(KERN_WARNING "Failed to execute %s.  Attempting "
 817                                         "defaults...\n", execute_command);
 818         }
 819         run_init_process("/sbin/init");
 820         run_init_process("/etc/init");
 821         run_init_process("/bin/init");
 822         run_init_process("/bin/sh");
 823
 824         panic("No init found.  Try passing init= option to kernel.");
 825 }
 826
 827 static int __init kernel_init(void * unused)
 828 {
 829         lock_kernel();
 830         /*
 831          * init can run on any cpu.
 832          */
 833         set_cpus_allowed_ptr(current, CPU_MASK_ALL_PTR);
 834         /*
 835          * Tell the world that we're going to be the grim
 836          * reaper of innocent orphaned children.
 837          *
 838          * We don't want people to have to make incorrect
 839          * assumptions about where in the task array this
 840          * can be found.
 841          */
 842         init_pid_ns.child_reaper = current;
 843
 844         cad_pid = task_pid(current);
 845
 846         smp_prepare_cpus(setup_max_cpus);
 847
 848         do_pre_smp_initcalls();
 849
 850         smp_init();
 851         sched_init_smp();
 852
 853         cpuset_init_smp();
 854
 855         do_basic_setup();
 856
 857         /*
 858          * check if there is an early userspace init.  If yes, let it do all
 859          * the work
 860          */
 861
 862         if (!ramdisk_execute_command)
 863                 ramdisk_execute_command = "/init";
 864
 865         if (sys_access((const char __user *) ramdisk_execute_command, 0) != 0) {
 866                 ramdisk_execute_command = NULL;
 867                 prepare_namespace();
 868         }
 869
 870         /*
 871          * Ok, we have completed the initial bootup, and
 872          * we're essentially up and running. Get rid of the
 873          * initmem segments and start the user-mode stuff..
 874          */
 875         init_post();
 876         return 0;
 877 }