Kernel 启动流程梳理
Posted Li-Yongjun
tags:
篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了Kernel 启动流程梳理相关的知识，希望对你有一定的参考价值。
内核生命周期
uboot 打印完 Starting kernel . . .，就完成了自己的使命，控制权便交给了 kernel 的第一条指令，也就是下面这个函数
init/main.c
asmlinkage __visible void __init start_kernel(void)
...
rest_init();
start_kernel 相当于内核的 main 函数，内核的生命周期就是从执行这个函数的第一条语句开始的，直到最后一个函数 reset_init()，内核将不再从这个函数中返回，而是陷入这个函数里面的一个 while(1) 死循环，这个死循环被作为 idle 进程，也就是 0 号进程。
所以，内核的生命周期，就是一个完整的 start_kernel 函数。始于 start_kernel 函数的第一条语句，停留在最后的死循环。
init 进程
kernel 会创建众多内核线程，来持续致力于内存、磁盘、CPU 的管理，其中有两个内核线程比较重要，需要我们重点讲解，那就是 1 号内核线程 kernel_init 和 2 号内核线程 kthreadd。1 号内核线程最终会被用户的第一个进程 init 代替，也就成了 1 号进程。如下：
# ps
PID USER COMMAND
1 root init
2 root [kthreadd]
3 root [rcu_gp]
4 root [rcu_par_gp]
7 root [kworker/u4:0-ev]
8 root [mm_percpu_wq]
9 root [ksoftirqd/0]
...
COMMAND 这一列，带中括号的是内核线程，不带中括号的是用户进程。从 PID 统一编址就可以看出，它俩地位是一样的。
下面我们深入分析一下从 start_kernel 到最终运行 init 进程，kernel 都经历了什么
打印
添加打印，是分析流程的好方法。
asmlinkage __visible void __init start_kernel(void)
char *command_line;
char *after_dashes;
set_task_stack_end_magic(&init_task);
smp_setup_processor_id();
debug_objects_early_init();
cgroup_init_early();
local_irq_disable();
early_boot_irqs_disabled = true;
/*
* Interrupts are still disabled. Do necessary setups, then
* enable them.
*/
boot_cpu_init();
page_address_init();
pr_notice("%s", linux_banner);
setup_arch(&command_line);
/*
* Set up the the initial canary and entropy after arch
* and after adding latent and command line entropy.
*/
add_latent_entropy();
add_device_randomness(command_line, strlen(command_line));
boot_init_stack_canary();
mm_init_cpumask(&init_mm);
setup_command_line(command_line);
setup_nr_cpu_ids();
setup_per_cpu_areas();
smp_prepare_boot_cpu(); /* arch-specific boot-cpu hooks */
boot_cpu_hotplug_init();
build_all_zonelists(NULL);
page_alloc_init();
pr_notice("Kernel command line: %s\\n", boot_command_line);
parse_early_param();
after_dashes = parse_args("Booting kernel",
static_command_line, __start___param,
__stop___param - __start___param,
-1, -1, NULL, &unknown_bootoption);
if (!IS_ERR_OR_NULL(after_dashes))
parse_args("Setting init args", after_dashes, NULL, 0, -1, -1,
NULL, set_init_arg);
jump_label_init();
/*
* These use large bootmem allocations and must precede
* kmem_cache_init()
*/
setup_log_buf(0);
vfs_caches_init_early();
sort_main_extable();
trap_init();
mm_init();
ftrace_init();
/* trace_printk can be enabled here */
early_trace_init();
/*
* 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.
*/
sched_init();
/*
* Disable preemption - early bootup scheduling is extremely
* fragile until we cpu_idle() for the first time.
*/
preempt_disable();
if (WARN(!irqs_disabled(),
"Interrupts were enabled *very* early, fixing it\\n"))
local_irq_disable();
radix_tree_init();
/*
* Set up housekeeping before setting up workqueues to allow the unbound
* workqueue to take non-housekeeping into account.
*/
housekeeping_init();
/*
* Allow workqueue creation and work item queueing/cancelling
* early. Work item execution depends on kthreads and starts after
* workqueue_init().
*/
workqueue_init_early();
rcu_init();
/* Trace events are available after this */
trace_init();
if (initcall_debug)
initcall_debug_enable();
context_tracking_init();
/* init some links before init_ISA_irqs() */
early_irq_init();
init_IRQ();
tick_init();
rcu_init_nohz();
init_timers();
hrtimers_init();
softirq_init();
timekeeping_init();
time_init();
sched_clock_postinit();
printk_safe_init();
perf_event_init();
profile_init();
call_function_init();
WARN(!irqs_disabled(), "Interrupts were enabled early\\n");
early_boot_irqs_disabled = false;
local_irq_enable();
kmem_cache_init_late();
/*
* 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();
printk("## start_kernel() --> console_init()\\n");
if (panic_later)
panic("Too many boot %s vars at `%s'", 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();
/*
* This needs to be called before any devices perform DMA
* operations that might use the SWIOTLB bounce buffers. It will
* mark the bounce buffers as decrypted so that their usage will
* not cause "plain-text" data to be decrypted when accessed.
*/
mem_encrypt_init();
#ifdef CONFIG_BLK_DEV_INITRD
if (initrd_start && !initrd_below_start_ok &&
page_to_pfn(virt_to_page((void *)initrd_start)) < min_low_pfn)
pr_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_ext_init();
kmemleak_init();
debug_objects_mem_init();
setup_per_cpu_pageset();
numa_policy_init();
acpi_early_init();
if (late_time_init)
late_time_init();
calibrate_delay();
pid_idr_init();
anon_vma_init();
#ifdef CONFIG_X86
if (efi_enabled(EFI_RUNTIME_SERVICES))
efi_enter_virtual_mode();
#endif
thread_stack_cache_init();
cred_init();
fork_init();
proc_caches_init();
uts_ns_init();
buffer_init();
key_init();
security_init();
dbg_late_init();
vfs_caches_init();
pagecache_init();
signals_init();
seq_file_init();
proc_root_init();
nsfs_init();
cpuset_init();
cgroup_init();
taskstats_init_early();
delayacct_init();
check_bugs();
acpi_subsystem_init();
arch_post_acpi_subsys_init();
sfi_init_late();
if (efi_enabled(EFI_RUNTIME_SERVICES))
efi_free_boot_services();
printk("## run rest_init()\\n");
/* Do the rest non-__init'ed, we're now alive */
rest_init();
printk("## after rest_init()\\n");
一开始尝试在函数刚开始就添加 printk 打印，结果发现添加完 printk 后内核起不来，最后保守起见，在 131 行 console_init(); 后才开始添加打印。
Starting kernel ...
[ 0.000000] Booting Linux on physical CPU 0x0
[ 0.000000] Linux version 4.18.12 (liyongjun@Box) (gcc version 9.3.0 (Buildroot 2021.05)) #14 SMP Thu Nov 25 00:37:30 CST 2021
[ 0.000000] CPU: ARMv7 Processor [410fc074] revision 4 (ARMv7), cr=10c5387d
[ 0.000000] CPU: div instructions available: patching division code
[ 0.000000] CPU: PIPT / VIPT nonaliasing data cache, VIPT aliasing instruction cache
[ 0.000000] OF: fdt: Machine model: LeMaker Banana Pi
[ 0.000000] Memory policy: Data cache writealloc
[ 0.000000] cma: Reserved 16 MiB at 0x7ec00000
[ 0.000000] psci: probing for conduit method from DT.
[ 0.000000] psci: Using PSCI v0.1 Function IDs from DT
[ 0.000000] random: get_random_bytes called from start_kernel+0xa0/0x430 with crng_init=0
[ 0.000000] percpu: Embedded 16 pages/cpu @(ptrval) s34444 r8192 d22900 u65536
[ 0.000000] Built 1 zonelists, mobility grouping on. Total pages: 260202
[ 0.000000] Kernel command line: console=ttyS0,57600 earlyprintk root=/dev/mmcblk0p2 rootwait
[ 0.000000] Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
[ 0.000000] Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
[ 0.000000] Memory: 1011460K/1046952K available (6144K kernel code, 418K rwdata, 1524K rodata, 1024K init, 240K bss, 19108K reserved, 16384K cma-reserved, 244136K highmem)
[ 0.000000] Virtual kernel memory layout:
[ 0.000000] vector : 0xffff0000 - 0xffff1000 ( 4 kB)
[ 0.000000] fixmap : 0xffc00000 - 0xfff00000 (3072 kB)
[ 0.000000] vmalloc : 0xf0800000 - 0xff800000 ( 240 MB)
[ 0.000000] lowmem : 0xc0000000 - 0xf0000000 ( 768 MB)
[ 0.000000] pkmap : 0xbfe00000 - 0xc0000000 ( 2 MB)
[ 0.000000] modules : 0xbf000000 - 0xbfe00000 ( 14 MB)
[ 0.000000] .text : 0x(ptrval) - 0x(ptrval) (7136 kB)
[ 0.000000] .init : 0x(ptrval) - 0x(ptrval) (1024 kB)
[ 0.000000] .data : 0x(ptrval) - 0x(ptrval) ( 419 kB)
[ 0.000000] .bss : 0x(ptrval) - 0x(ptrval) ( 241 kB)
[ 0.000000] SLUB: HWalign=64, Order=0-3, MinObjects=0, CPUs=2, Nodes=1
[ 0.000000] Hierarchical RCU implementation.
[ 0.000000] RCU restricting CPUs from NR_CPUS=8 to nr_cpu_ids=2.
[ 0.000000] RCU: Adjusting geometry for rcu_fanout_leaf=16, nr_cpu_ids=2
[ 0.000000] NR_IRQS: 16, nr_irqs: 16, preallocated irqs: 16
[ 0.000000] GIC: Using split EOI/Deactivate mode
[ 0.000000] arch_timer: cp15 timer(s) running at 24.00MHz (phys).
[ 0.000000] clocksource: arch_sys_counter: mask: 0xffffffffffffff max_cycles: 0x588fe9dc0, max_idle_ns: 440795202592 ns
[ 0.000007] sched_clock: 56 bits at 24MHz, resolution 41ns, wraps every 4398046511097ns
[ 0.000021] Switching to timer-based delay loop, resolution 41ns
[ 0.000334] clocksource: timer: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 79635851949 ns
[ 0.000586] clocksource: hstimer: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 6370868154 ns
[ 0.000822] Console: colour dummy device 80x30
[ 0.000844] ## start_kernel() --> console_init() // 132 行的打印
可是，发现在 console_init() 之前就有不少打印了，这个地方还是有些不解。猜测要么是在 console_init() 之前就已经可以打印了；要么是在 console_init() 之前先将打印缓存着，等初始化之后再打印。这点以后再研究吧。先插个眼👁。
接着看打印
[ 0.000844] ## start_kernel() --> console_init()
[ 0.000872] Calibrating delay loop (skipped), value calculated using timer frequency.. 48.00 BogoMIPS (lpj=240000)
[ 0.000886] pid_max: default: 32768 minimum: 301
[ 0.001054] Mount-cache hash table entries: 2048 (order: 1, 8192 bytes)
[ 0.001070] Mountpoint-cache hash table entries: 2048 (order: 1, 8192 bytes)
[ 0.001727] CPU: Testing write buffer coherency: ok
[ 0.001779] ## run rest_init()
从 133 行到 206 行，代码不少，打印却只有寥寥 5 行。少就说明不重要，那就不分析了，哈哈😁。
看下面的代码，乖乖，就剩一句了：rest_init();，并且第 210 行的 printk 等到系统完全起来都没有打印，说明 rest_init() 就没返回。看来是个扛把子。
static noinline void __ref rest_init(void)
struct task_struct *tsk;
int pid;
printk("## rcu_scheduler_starting()\\n");
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.
*/
pid = kernel_thread(kernel_init, NULL, CLONE_FS);
/*
* Pin init on the boot CPU. Task migration is not properly working
* until sched_init_smp() has been run. It will set the allowed
* CPUs for init to the non isolated CPUs.
*/
rcu_read_lock();
tsk = find_task_by_pid_ns(pid, &init_pid_ns);
set_cpus_allowed_ptr(tsk, cpumask_of(smp_processor_id()));
rcu_read_unlock();
printk("## bb\\n");
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();
printk("## cc\\n");
/*
* Enable might_sleep() and smp_processor_id() checks.
* They cannot be enabled earlier because with CONFIG_PREEMPT=y
* kernel_thread() would trigger might_sleep() splats. With
* CONFIG_PREEMPT_VOLUNTARY=y the init task might have scheduled
* already, but it's stuck on the kthreadd_done completion.
*/
system_state = SYSTEM_SCHEDULING;
printk("## dd\\n");
complete(&kthreadd_done);
/*
* The boot idle thread must execute schedule()
* at least once to get things moving:
*/
printk("## ee\\n");
schedule_preempt_disabled();
printk("## ff\\n");
/* Call into cpu_idle with preempt disabled */
cpu_startup_entry(CPUHP_ONLINE);
// printk("## ff\\n");
对应打印如下：
[ 0.001779] ## run rest_init()
[ 0.001786] ## rcu_scheduler_starting()
[ 0.002030] ## bb
[ 0.002089] ## cc
[ 0.002097] ## dd
[ 0.002104] ## ee
[ 0.002150] ## kernel_init()
[ 0.002199] /cpus/cpu@0 missing clock-frequency property
[ 0.002214] /cpus/cpu@1 missing clock-frequency property
[ 0.002228] CPU0: thread -1, cpu 0, socket 0, mpidr 80000000
[ 0.002799] Setting up static identity map for 0x40100000 - 0x40100060
[ 0.002972] Hierarchical SRCU implementation.
[ 0.003776] smp: Bringing up secondary CPUs ...
[ 0.014335] ## ff
打印了 kernel_init()，是因为调用了 schedule_preempt_disabled()，其内部调用了 schedule() 执行进程切换，进入睡眠。当然，在当前进程被唤醒时，程序也是从这个断点开始运行。
void __sched schedule_preempt_disabled(void)
sched_preempt_enable_no_resched();
schedule();
preempt_disable();
执行进程切换后，kernel_init() 才有机会运行
static int __ref kernel_init(void *unused)
int ret;
printk("## kernel_init()\\n");
kernel_init_freeable();
/* need to finish all async __init code before freeing the memory */
async_synchronize_full();
ftrace_free_init_mem();
jump_label_invalidate_initmem();
free_initmem();
printk("## 2.\\n");
mark_readonly();
printk("## 3.\\n");
system_state = SYSTEM_RUNNING;
numa_default_policy();
rcu_end_inkernel_boot();
printk("## 4.\\n");
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;
panic("Requested init %s failed (error %d).",
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")) 以上是关于Kernel 启动流程梳理的主要内容，如果未能解决你的问题，请参考以下文章