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/bootmem.h>
  26 #include <linux/tty.h>
  27 #include <linux/gfp.h>
  28 #include <linux/percpu.h>
  29 #include <linux/kmod.h>
  30 #include <linux/kernel_stat.h>
  31 #include <linux/start_kernel.h>
  32 #include <linux/security.h>
  33 #include <linux/smp.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/debugobjects.h>
  56 #include <linux/lockdep.h>
  57 #include <linux/pid_namespace.h>
  58 #include <linux/device.h>
  59 #include <linux/kthread.h>
  60 #include <linux/sched.h>
  61 #include <linux/signal.h>
  62 #include <linux/idr.h>
  63
  64 #include <asm/io.h>
  65 #include <asm/bugs.h>
  66 #include <asm/setup.h>
  67 #include <asm/sections.h>
  68 #include <asm/cacheflush.h>
  69
  70 #ifdef CONFIG_X86_LOCAL_APIC
  71 #include <asm/smp.h>
  72 #endif
  73
  74 /*
  75  * This is one of the first .c files built. Error out early if we have compiler
  76  * trouble.
  77  */
  78
  79 #if __GNUC__ == 4 && __GNUC_MINOR__ == 1 && __GNUC_PATCHLEVEL__ == 0
  80 #warning gcc-4.1.0 is known to miscompile the kernel.  A different compiler version is recommended.
  81 #endif
  82
  83 static int kernel_init(void *);
  84
  85 extern void init_IRQ(void);
  86 extern void fork_init(unsigned long);
  87 extern void mca_init(void);
  88 extern void sbus_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         /*
 416          * Set up the current CPU as possible to migrate to.
 417          * The other ones will be done by cpu_up/cpu_down()
 418          */
 419         cpu = smp_processor_id();
 420         cpu_set(cpu, cpu_active_map);
 421
 422         /* FIXME: This should be done in userspace --RR */
 423         for_each_present_cpu(cpu) {
 424                 if (num_online_cpus() >= setup_max_cpus)
 425                         break;
 426                 if (!cpu_online(cpu))
 427                         cpu_up(cpu);
 428         }
 429
 430         /* Any cleanup work */
 431         printk(KERN_INFO "Brought up %ld CPUs\n", (long)num_online_cpus());
 432         smp_cpus_done(setup_max_cpus);
 433 }
 434
 435 #endif
 436
 437 /*
 438  * We need to store the untouched command line for future reference.
 439  * We also need to store the touched command line since the parameter
 440  * parsing is performed in place, and we should allow a component to
 441  * store reference of name/value for future reference.
 442  */
 443 static void __init setup_command_line(char *command_line)
 444 {
 445         saved_command_line = alloc_bootmem(strlen (boot_command_line)+1);
 446         static_command_line = alloc_bootmem(strlen (command_line)+1);
 447         strcpy (saved_command_line, boot_command_line);
 448         strcpy (static_command_line, command_line);
 449 }
 450
 451 /*
 452  * We need to finalize in a non-__init function or else race conditions
 453  * between the root thread and the init thread may cause start_kernel to
 454  * be reaped by free_initmem before the root thread has proceeded to
 455  * cpu_idle.
 456  *
 457  * gcc-3.4 accidentally inlines this function, so use noinline.
 458  */
 459
 460 static void noinline __init_refok rest_init(void)
 461         __releases(kernel_lock)
 462 {
 463         int pid;
 464
 465         kernel_thread(kernel_init, NULL, CLONE_FS | CLONE_SIGHAND);
 466         numa_default_policy();
 467         pid = kernel_thread(kthreadd, NULL, CLONE_FS | CLONE_FILES);
 468         kthreadd_task = find_task_by_pid_ns(pid, &init_pid_ns);
 469         unlock_kernel();
 470
 471         /*
 472          * The boot idle thread must execute schedule()
 473          * at least once to get things moving:
 474          */
 475         init_idle_bootup_task(current);
 476         preempt_enable_no_resched();
 477         schedule();
 478         preempt_disable();
 479
 480         /* Call into cpu_idle with preempt disabled */
 481         cpu_idle();
 482 }
 483
 484 /* Check for early params. */
 485 static int __init do_early_param(char *param, char *val)
 486 {
 487         struct obs_kernel_param *p;
 488
 489         for (p = __setup_start; p < __setup_end; p++) {
 490                 if ((p->early && strcmp(param, p->str) == 0) ||
 491                     (strcmp(param, "console") == 0 &&
 492                      strcmp(p->str, "earlycon") == 0)
 493                 ) {
 494                         if (p->setup_func(val) != 0)
 495                                 printk(KERN_WARNING
 496                                        "Malformed early option '%s'\n", param);
 497                 }
 498         }
 499         /* We accept everything at this stage. */
 500         return 0;
 501 }
 502
 503 /* Arch code calls this early on, or if not, just before other parsing. */
 504 void __init parse_early_param(void)
 505 {
 506         static __initdata int done = 0;
 507         static __initdata char tmp_cmdline[COMMAND_LINE_SIZE];
 508
 509         if (done)
 510                 return;
 511
 512         /* All fall through to do_early_param. */
 513         strlcpy(tmp_cmdline, boot_command_line, COMMAND_LINE_SIZE);
 514         parse_args("early options", tmp_cmdline, NULL, 0, do_early_param);
 515         done = 1;
 516 }
 517
 518 /*
 519  *      Activate the first processor.
 520  */
 521
 522 static void __init boot_cpu_init(void)
 523 {
 524         int cpu = smp_processor_id();
 525         /* Mark the boot cpu "present", "online" etc for SMP and UP case */
 526         cpu_set(cpu, cpu_online_map);
 527         cpu_set(cpu, cpu_present_map);
 528         cpu_set(cpu, cpu_possible_map);
 529 }
 530
 531 void __init __weak smp_setup_processor_id(void)
 532 {
 533 }
 534
 535 void __init __weak thread_info_cache_init(void)
 536 {
 537 }
 538
 539 asmlinkage void __init start_kernel(void)
 540 {
 541         char * command_line;
 542         extern struct kernel_param __start___param[], __stop___param[];
 543
 544         smp_setup_processor_id();
 545
 546         /*
 547          * Need to run as early as possible, to initialize the
 548          * lockdep hash:
 549          */
 550         unwind_init();
 551         lockdep_init();
 552         debug_objects_early_init();
 553         cgroup_init_early();
 554
 555         local_irq_disable();
 556         early_boot_irqs_off();
 557         early_init_irq_lock_class();
 558
 559 /*
 560  * Interrupts are still disabled. Do necessary setups, then
 561  * enable them
 562  */
 563         lock_kernel();
 564         tick_init();
 565         boot_cpu_init();
 566         page_address_init();
 567         printk(KERN_NOTICE);
 568         printk(linux_banner);
 569         setup_arch(&command_line);
 570         mm_init_owner(&init_mm, &init_task);
 571         setup_command_line(command_line);
 572         unwind_setup();
 573         setup_per_cpu_areas();
 574         setup_nr_cpu_ids();
 575         smp_prepare_boot_cpu(); /* arch-specific boot-cpu hooks */
 576
 577         /*
 578          * Set up the scheduler prior starting any interrupts (such as the
 579          * timer interrupt). Full topology setup happens at smp_init()
 580          * time - but meanwhile we still have a functioning scheduler.
 581          */
 582         sched_init();
 583         /*
 584          * Disable preemption - early bootup scheduling is extremely
 585          * fragile until we cpu_idle() for the first time.
 586          */
 587         preempt_disable();
 588         build_all_zonelists();
 589         page_alloc_init();
 590         printk(KERN_NOTICE "Kernel command line: %s\n", boot_command_line);
 591         parse_early_param();
 592         parse_args("Booting kernel", static_command_line, __start___param,
 593                    __stop___param - __start___param,
 594                    &unknown_bootoption);
 595         if (!irqs_disabled()) {
 596                 printk(KERN_WARNING "start_kernel(): bug: interrupts were "
 597                                 "enabled *very* early, fixing it\n");
 598                 local_irq_disable();
 599         }
 600         sort_main_extable();
 601         trap_init();
 602         rcu_init();
 603         init_IRQ();
 604         pidhash_init();
 605         init_timers();
 606         hrtimers_init();
 607         softirq_init();
 608         timekeeping_init();
 609         time_init();
 610         sched_clock_init();
 611         profile_init();
 612         if (!irqs_disabled())
 613                 printk("start_kernel(): bug: interrupts were enabled early\n");
 614         early_boot_irqs_on();
 615         local_irq_enable();
 616
 617         /*
 618          * HACK ALERT! This is early. We're enabling the console before
 619          * we've done PCI setups etc, and console_init() must be aware of
 620          * this. But we do want output early, in case something goes wrong.
 621          */
 622         console_init();
 623         if (panic_later)
 624                 panic(panic_later, panic_param);
 625
 626         lockdep_info();
 627
 628         /*
 629          * Need to run this when irqs are enabled, because it wants
 630          * to self-test [hard/soft]-irqs on/off lock inversion bugs
 631          * too:
 632          */
 633         locking_selftest();
 634
 635 #ifdef CONFIG_BLK_DEV_INITRD
 636         if (initrd_start && !initrd_below_start_ok &&
 637             page_to_pfn(virt_to_page((void *)initrd_start)) < min_low_pfn) {
 638                 printk(KERN_CRIT "initrd overwritten (0x%08lx < 0x%08lx) - "
 639                     "disabling it.\n",
 640                     page_to_pfn(virt_to_page((void *)initrd_start)),
 641                     min_low_pfn);
 642                 initrd_start = 0;
 643         }
 644 #endif
 645         vfs_caches_init_early();
 646         cpuset_init_early();
 647         mem_init();
 648         enable_debug_pagealloc();
 649         cpu_hotplug_init();
 650         kmem_cache_init();
 651         debug_objects_mem_init();
 652         idr_init_cache();
 653         setup_per_cpu_pageset();
 654         numa_policy_init();
 655         if (late_time_init)
 656                 late_time_init();
 657         calibrate_delay();
 658         pidmap_init();
 659         pgtable_cache_init();
 660         prio_tree_init();
 661         anon_vma_init();
 662 #ifdef CONFIG_X86
 663         if (efi_enabled)
 664                 efi_enter_virtual_mode();
 665 #endif
 666         thread_info_cache_init();
 667         fork_init(num_physpages);
 668         proc_caches_init();
 669         buffer_init();
 670         unnamed_dev_init();
 671         key_init();
 672         security_init();
 673         vfs_caches_init(num_physpages);
 674         radix_tree_init();
 675         signals_init();
 676         /* rootfs populating might need page-writeback */
 677         page_writeback_init();
 678 #ifdef CONFIG_PROC_FS
 679         proc_root_init();
 680 #endif
 681         cgroup_init();
 682         cpuset_init();
 683         taskstats_init_early();
 684         delayacct_init();
 685
 686         check_bugs();
 687
 688         acpi_early_init(); /* before LAPIC and SMP init */
 689
 690         /* Do the rest non-__init'ed, we're now alive */
 691         rest_init();
 692 }
 693
 694 static int initcall_debug;
 695
 696 static int __init initcall_debug_setup(char *str)
 697 {
 698         initcall_debug = 1;
 699         return 1;
 700 }
 701 __setup("initcall_debug", initcall_debug_setup);
 702
 703 int do_one_initcall(initcall_t fn)
 704 {
 705         int count = preempt_count();
 706         ktime_t t0, t1, delta;
 707         char msgbuf[64];
 708         int result;
 709
 710         if (initcall_debug) {
 711                 printk("calling  %pF @ %i\n", fn, task_pid_nr(current));
 712                 t0 = ktime_get();
 713         }
 714
 715         result = fn();
 716
 717         if (initcall_debug) {
 718                 t1 = ktime_get();
 719                 delta = ktime_sub(t1, t0);
 720
 721                 printk("initcall %pF returned %d after %Ld msecs\n",
 722                         fn, result,
 723                         (unsigned long long) delta.tv64 >> 20);
 724         }
 725
 726         msgbuf[0] = 0;
 727
 728         if (result && result != -ENODEV && initcall_debug)
 729                 sprintf(msgbuf, "error code %d ", result);
 730
 731         if (preempt_count() != count) {
 732                 strlcat(msgbuf, "preemption imbalance ", sizeof(msgbuf));
 733                 preempt_count() = count;
 734         }
 735         if (irqs_disabled()) {
 736                 strlcat(msgbuf, "disabled interrupts ", sizeof(msgbuf));
 737                 local_irq_enable();
 738         }
 739         if (msgbuf[0]) {
 740                 printk("initcall %pF returned with %s\n", fn, msgbuf);
 741         }
 742
 743         return result;
 744 }
 745
 746
 747 extern initcall_t __initcall_start[], __initcall_end[], __early_initcall_end[];
 748
 749 static void __init do_initcalls(void)
 750 {
 751         initcall_t *call;
 752
 753         for (call = __early_initcall_end; call < __initcall_end; call++)
 754                 do_one_initcall(*call);
 755
 756         /* Make sure there is no pending stuff from the initcall sequence */
 757         flush_scheduled_work();
 758 }
 759
 760 /*
 761  * Ok, the machine is now initialized. None of the devices
 762  * have been touched yet, but the CPU subsystem is up and
 763  * running, and memory and process management works.
 764  *
 765  * Now we can finally start doing some real work..
 766  */
 767 static void __init do_basic_setup(void)
 768 {
 769         rcu_init_sched(); /* needed by module_init stage. */
 770         /* drivers will send hotplug events */
 771         init_workqueues();
 772         usermodehelper_init();
 773         driver_init();
 774         init_irq_proc();
 775         do_initcalls();
 776 }
 777
 778 static void __init do_pre_smp_initcalls(void)
 779 {
 780         initcall_t *call;
 781
 782         for (call = __initcall_start; call < __early_initcall_end; call++)
 783                 do_one_initcall(*call);
 784 }
 785
 786 static void run_init_process(char *init_filename)
 787 {
 788         argv_init[0] = init_filename;
 789         kernel_execve(init_filename, argv_init, envp_init);
 790 }
 791
 792 /* This is a non __init function. Force it to be noinline otherwise gcc
 793  * makes it inline to init() and it becomes part of init.text section
 794  */
 795 static int noinline init_post(void)
 796 {
 797         free_initmem();
 798         unlock_kernel();
 799         mark_rodata_ro();
 800         system_state = SYSTEM_RUNNING;
 801         numa_default_policy();
 802
 803         if (sys_open((const char __user *) "/dev/console", O_RDWR, 0) < 0)
 804                 printk(KERN_WARNING "Warning: unable to open an initial console.\n");
 805
 806         (void) sys_dup(0);
 807         (void) sys_dup(0);
 808
 809         current->signal->flags |= SIGNAL_UNKILLABLE;
 810
 811         if (ramdisk_execute_command) {
 812                 run_init_process(ramdisk_execute_command);
 813                 printk(KERN_WARNING "Failed to execute %s\n",
 814                                 ramdisk_execute_command);
 815         }
 816
 817         /*
 818          * We try each of these until one succeeds.
 819          *
 820          * The Bourne shell can be used instead of init if we are
 821          * trying to recover a really broken machine.
 822          */
 823         if (execute_command) {
 824                 run_init_process(execute_command);
 825                 printk(KERN_WARNING "Failed to execute %s.  Attempting "
 826                                         "defaults...\n", execute_command);
 827         }
 828         run_init_process("/sbin/init");
 829         run_init_process("/etc/init");
 830         run_init_process("/bin/init");
 831         run_init_process("/bin/sh");
 832
 833         panic("No init found.  Try passing init= option to kernel.");
 834 }
 835
 836 static int __init kernel_init(void * unused)
 837 {
 838         lock_kernel();
 839         /*
 840          * init can run on any cpu.
 841          */
 842         set_cpus_allowed_ptr(current, CPU_MASK_ALL_PTR);
 843         /*
 844          * Tell the world that we're going to be the grim
 845          * reaper of innocent orphaned children.
 846          *
 847          * We don't want people to have to make incorrect
 848          * assumptions about where in the task array this
 849          * can be found.
 850          */
 851         init_pid_ns.child_reaper = current;
 852
 853         cad_pid = task_pid(current);
 854
 855         smp_prepare_cpus(setup_max_cpus);
 856
 857         do_pre_smp_initcalls();
 858
 859         smp_init();
 860         sched_init_smp();
 861
 862         cpuset_init_smp();
 863
 864         do_basic_setup();
 865
 866         /*
 867          * check if there is an early userspace init.  If yes, let it do all
 868          * the work
 869          */
 870
 871         if (!ramdisk_execute_command)
 872                 ramdisk_execute_command = "/init";
 873
 874         if (sys_access((const char __user *) ramdisk_execute_command, 0) != 0) {
 875                 ramdisk_execute_command = NULL;
 876                 prepare_namespace();
 877         }
 878
 879         /*
 880          * Ok, we have completed the initial bootup, and
 881          * we're essentially up and running. Get rid of the
 882          * initmem segments and start the user-mode stuff..
 883          */
 884         init_post();
 885         return 0;
 886 }