start_kernel分析

start_kernel是linux内核的入口函数，定义在init/main.c里面。

start_kernel里面包含了一系列的初始化代码。

<span style="color:#666666;">asmlinkage void __init </span><strong style="color: rgb(102, 102, 102);">start_kernel</strong><span style="color:#666666;">(void)
{
char * command_line;
extern const struct kernel_param __start___param[], __stop___param[];

smp_setup_processor_id(); //确定SMP系统中每个CPU的id

/*
* Need to run as early as possible, to initialize the
* lockdep hash:
*/
lockdep_init(); //初始化互斥锁的dependency。
debug_objects_early_init(); //初始化debug kernel相关

/*
* Set up the the initial canary ASAP:
*/
boot_init_stack_canary(); //stack_canary的是带防止栈溢出攻击保护的堆栈。

cgroup_init_early();

local_irq_disable(); //这个太直白了。
early_boot_irqs_off(); //这个也很直白。

/*
* Interrupts are still disabled. Do necessary setups, then
* enable them
*/
tick_init(); //初始化time ticket，时钟
boot_cpu_init(); //用以启动的CPU进行初始化。也就是初始化CPU0
page_address_init();//初始化页面
printk(KERN_NOTICE "%s", linux_banner);
setup_arch(&command_line); //CPU架构相关的初始化
mm_init_owner(&init_mm, &init_task); //初始化内存管理
setup_command_line(command_line); //处理启动命令行
setup_nr_cpu_ids();
setup_per_cpu_areas();
smp_prepare_boot_cpu();//准备boot_cpu.   /* arch-specific boot-cpu hooks */

build_all_zonelists(NULL); //建立什么样的zone??
page_alloc_init(); //初始化page allocation相关结构

printk(KERN_NOTICE "Kernel command line: %s\n", boot_command_line);
parse_early_param();
parse_args("Booting kernel", static_command_line, __start___param,
__stop___param - __start___param,
&unknown_bootoption);//解析启动参数
/*
* These use large bootmem allocations and must precede
* kmem_cache_init()
*/
pidhash_init();//初始化process ID hash表
vfs_caches_init_early(); //文件系统caches预初始化
sort_main_extable(); //初始化exception table
trap_init(); //初始化trap，用以处理错误执行代码
mm_init(); //初始化内存管理
/*
* Set up the scheduler prior starting any interrupts (such as the
* timer interrupt). Full topology setup happens at smp_init()
* time - but meanwhile we still have a functioning scheduler.
*/
</span><strong style="color: rgb(102, 102, 102);">sched_init();</strong><span style="color:#666666;"> //进程调度初始化
/*
* Disable preemption - early bootup scheduling is extremely
* fragile until we cpu_idle() for the first time.
*/
preempt_disable();
if (!irqs_disabled()) {
printk(KERN_WARNING "start_kernel(): bug: interrupts were "
"enabled *very* early, fixing it\n");
local_irq_disable();
}
rcu_init(); //Read_Copy_Update机制初始
radix_tree_init();
/* init some links before init_ISA_irqs() */
early_irq_init();
init_IRQ(); //初始化中断
prio_tree_init(); //这是啥？以后要弄懂
init_timers(); //初始化时钟
hrtimers_init();//初始化高精时钟
softirq_init();//初始化软中断
timekeeping_init();//初始化时钟源
time_init();//初始化时间例程
profile_init(); //Profile?这是什么profile?
if (!irqs_disabled())
printk(KERN_CRIT "start_kernel(): bug: interrupts were "
"enabled early\n");
early_boot_irqs_on();
local_irq_enable();

/* Interrupts are enabled now so all GFP allocations are safe. */
gfp_allowed_mask = __GFP_BITS_MASK;

kmem_cache_init_late();//初始化CPU Cache

/*
* HACK ALERT! This is early. We're enabling the console before
* we've done PCI setups etc, and console_init() must be aware of
* this. But we do want output early, in case something goes wrong.
*/
console_init(); //初始化console
if (panic_later)
panic(panic_later, panic_param);

lockdep_info();

/*
* Need to run this when irqs are enabled, because it wants
* to self-test [hard/soft]-irqs on/off lock inversion bugs
* too:
*/
locking_selftest(); //自测试锁

#ifdef CONFIG_BLK_DEV_INITRD
if (initrd_start && !initrd_below_start_ok &&
page_to_pfn(virt_to_page((void *)initrd_start)) < min_low_pfn) {
printk(KERN_CRIT "initrd overwritten (0x%08lx < 0x%08lx) - "
"disabling it.\n",
page_to_pfn(virt_to_page((void *)initrd_start)),
min_low_pfn);
initrd_start = 0;
}
#endif
page_cgroup_init(); //页面初始
enable_debug_pagealloc(); //页面分配debug启用
kmemleak_init(); //memory leak 侦测初始化
debug_objects_mem_init();
idr_init_cache();
setup_per_cpu_pageset(); //设置每个CPU的页面集合
numa_policy_init();
if (late_time_init)
late_time_init();
sched_clock_init();//初始化调度时钟
calibrate_delay(); //协同不同CPU的时钟
pidmap_init();//pid是process id还是processor id?
anon_vma_init();//anonymous page?什么意思？
#ifdef CONFIG_X86
if (efi_enabled)
efi_enter_virtual_mode();
#endif
thread_info_cache_init(); //初始化thread info
cred_init(); //credential
fork_init(totalram_pages); //初始化fork
proc_caches_init(); //初始化/proc的cache?
buffer_init(); // buffer
key_init(); //key
security_init();//security
dbg_late_init();//debug
vfs_caches_init(totalram_pages);//文件系统cache初始化
signals_init();//signal
/* rootfs populating might need page-writeback */
page_writeback_init();page_writeback
#ifdef CONFIG_PROC_FS
proc_root_init();
#endif
cgroup_init();
cpuset_init(); //cpuset
taskstats_init_early(); //task
delayacct_init();

check_bugs();

acpi_early_init();//acpi /* before LAPIC and SMP init */
sfi_init_late(); //simple firmware interface

ftrace_init();

/* Do the rest non-__init'ed, we're now alive */
</span><strong><span style="color:#ff0000;"> rest_init();</span></strong><span style="color:#666666;">
}</span>

将这些函数归归类，以后就按照类型来读：

CPU初始化

smp_setup_processor_id()

boot_cpu_init()

setup_arch(&command_line);

setup_nr_cpu_ids()

setup_per_cpu_areas()

smp_prepare_boot_cpu()

setup_per_cpu_pageset();

calibrate_delay();

cpuset_init();

内存管理初始化

boot_init_stack_canary()

page_address_init();

mm_init_owner();

page_alloc_init();

mm_init();

rcu_init();

kmem_cache_init_late();

page_cgroup_init();

kmemleak_init();

numa_policy_init();

anon_vma_init();

page_writeback_init();

进程管理

pidhash_init();

sched_init();

sched_clock_init()

pidmap_init();

fork_init(totalram_pages);

taskstats_init_early();

文件系统

vfs_caches_init_early();

thread_info_cache_init();

vfs_caches_init(totalram_pages);

中断

early_irq_init();

init_IRQ();

softirq_init();

同步互斥

lockdep_init();

lockdep_info();

locking_selftest();

时钟

tick_init();

init_timers();

hrtimers_init();

timekeeping_init();

time_init();

调试

debug_objects_early_init();

console_init();

enable_debug_pagealloc();

debug_objects_mem_init();

dbg_late_init();

其他

sort_main_extable();

trap_init();

efi_enter_virtual_mode();

cred_init();

proc_caches_init();

buffer_init();

key_init();

security_init();

signals_init();

proc_root_init();

delayacct_init();

check_bugs();

acpi_early_init();

sfi_init_late();

未知

cgroup_init_early();

build_all_zonelists(NULL);

preempt_disable();

radix_tree_init();

prio_tree_init();

profile_init();

idr_init_cache();

cgroup_init();

ftrace_init();

<span style="color:#666666;">static noinline void __init_refok </span><strong><span style="color:#ff0000;">rest_init</span></strong><span style="color:#666666;">(void)
{
int pid;

rcu_scheduler_starting();
/*
* We need to spawn init first so that it obtains pid 1, however
* the init task will end up wanting to create kthreads, which, if
* we schedule it before we create kthreadd, will OOPS.
*/
kernel_thread(</span><strong><span style="color:#ff0000;">kernel_init</span></strong><span style="color:#666666;">, NULL, CLONE_FS);//创建1号进程
numa_default_policy();
pid = kernel_thread(kthreadd, NULL, CLONE_FS | CLONE_FILES);
rcu_read_lock();
kthreadd_task = find_task_by_pid_ns(pid, &init_pid_ns);
rcu_read_unlock();
complete(&kthreadd_done);

/*
* The boot idle thread must execute schedule()
* at least once to get things moving:
*/
init_idle_bootup_task(current);
schedule_preempt_disabled();
/* Call into cpu_idle with preempt disabled */
cpu_startup_entry(CPUHP_ONLINE);
}</span>

static int __ref <strong><span style="color:#ff0000;">kernel_init</span></strong>(void *unused)
{
int ret;

kernel_init_freeable();
/* need to finish all async __init code before freeing the memory */
async_synchronize_full();
free_initmem();
mark_rodata_ro();
system_state = SYSTEM_RUNNING;
numa_default_policy();

flush_delayed_fput();

if (ramdisk_execute_command) {
ret = run_init_process(ramdisk_execute_command);
if (!ret)
return 0;
pr_err("Failed to execute %s (error %d)\n",
ramdisk_execute_command, ret);
}

/*
* We try each of these until one succeeds.
*
* The Bourne shell can be used instead of init if we are
* trying to recover a really broken machine.
*/
if (execute_command) {
ret = run_init_process(execute_command);
if (!ret)
return 0;
pr_err("Failed to execute %s (error %d).  Attempting defaults...\n",
execute_command, ret);
}
if (!try_to_run_init_process("/sbin/init") ||
!try_to_run_init_process("/etc/init") ||
!try_to_run_init_process("/bin/init") ||
!try_to_run_init_process("/bin/sh"))
return 0;

panic("No working init found.  Try passing init= option to kernel. "
"See Linux Documentation/init.txt for guidance.");
}