OK6410A 开发板 (八) 63 linux-5.11 OK6410A linux内核空间常见的异常情景及分析 kernel exception

2022-08-22 来源：csdn

异常情景的level1

异常情景的level1 是最底层 show 当前状况的 api

其他具体的异常情景会根据自身情景调用这些api

oops

打印

Unable to ... 到 ---[ end trace

行为

如果调用panic,就会做panic的行为,否则只是打印

是什么时候会调用panic

1. panic_on_oops 为 1

2. in_interrupt

2.1 硬中断

2.2 软中断

2.3 NMI

panic

打印

Kernel panic ...

行为

通知感兴趣模块,死机或重启

被谁调用的

只有oops会调用panic

BUG

!CONFIG_BUG #define BUG() do {} while (1)

CONFIG_BUG && !HAVE_ARCH_BUG #define BUG() do {

printk('BUG: failure at %s:%d/%s()!n', __FILE__, __LINE__, __func__);

barrier_before_unreachable();

panic('BUG!');

} while (0)

CONFIG_BUG && HAVE_ARCH_BUG #define BUG() _BUG(__FILE__, __LINE__, BUG_INSTR_VALUE)

#define BUG() _BUG(__FILE__, __LINE__, BUG_INSTR_VALUE) // #define BUG_INSTR_VALUE 0xe7f001f2 // 以这个值为arm的二进制代码执行,会产生未定义指令异常

#define _BUG(file, line, value) __BUG(file, line, value)

#define __BUG(__file, __line, __value)

do {

asm volatile('1:t' BUG_INSTR(__value) 'n'

'.pushsection .rodata.str, 'aMS', %progbits, 1n'

'2:t.asciz ' #__file 'n'

'.popsectionn'

'.pushsection __bug_table,'aw'n'

'.align 2n'

'3:t.word 1b, 2bn'

't.hword ' #__line ', 0n'

'.popsection');

unreachable();

} while (0)

BUG_ON

#define BUG_ON(condition) do { if (unlikely(condition)) BUG(); } while (0)

level2

level2 的话有很多,level 2 对应了具体的异常情景

比如除0 的函数路径

比如缺页的函数路径

1. 踩内存、硬件Bit位翻转导致的数据异常

一般需要同时编译出来的符号表（vmlinux）进行反汇编分析

Unable to handle kernel paging request at virtual address XXXXXXXXX

2. 调用BUG // arm64 BUG 直接调用 panic

代码中出现异常导致调用到BUG()触发的主动重启

Kernel BUG at ffffff8008311ac8 [verbose debug info unavailable]

3. oom

Out of memory意味着已无可用的memory，这样的问题必然存在一些耗费资源的进程耗尽了memory的资源触发的KE

Kernel panic - not syncing: Out of memory and no killable processes...

4. 未定义指令异常

一般是CPU/DRAM不稳定导致的问题

Internal error: Oops - undefined instruction

5. 卡死导致硬件看门狗超时

多见于底层频繁irq/bus卡死，导致kicker无法被schedule，从而引起watch dog触发中断

引导系统进入FIQ处理流程，最终call到BUG触发重启。 // wathchdog 是 fiq ???

（表现为HWT和Hardware Reboot）

6. 软件看门狗超时

MTK平台有hang_detect的机制，system_server进程每30S回去软件喂狗

如果系统卡在超过10min/11min/12min/14min时,会dump出不同的信息用于debug。

level2 实例

clcdfb_probe

825 printk('xxxxxxxxxxxxxxxxxxxxxxxxxxxx1n');

826 int *p = 0;

827 int a = 0;

828 a = *p;

829 printk('xxxxxxxxxxxxxxxxxxxxxxxxxxxx2n');

#0 __loop_delay () at arch/arm/lib/delay-loop.S:47

#1 0xc0045544 in panic (fmt=0xc0b1a010 'Attempted to kill init! exitcode=0x%08xn') at kernel/panic.c:200

#2 0xc00486cc in find_child_reaper (father=0xee8a8000) at kernel/exit.c:471

#3 0xc0048998 in forget_original_parent (father=0xee8a8000, dead=0xee8977a0) at kernel/exit.c:560

#4 0xc0048cd4 in exit_notify (tsk=0xee8a8000, group_dead=1) at kernel/exit.c:596

#5 0xc0049878 in do_exit (code=11) at kernel/exit.c:767

#6 0xc001dacc in oops_end (flags=1610613011, regs=0xee897b88, signr=11) at arch/arm/kernel/traps.c:313

#7 0xc001db7c in die (str=0xc0b17148 'Oops', regs=0xee897b88, err=23) at arch/arm/kernel/traps.c:333

#8 0xc002f960 in __do_kernel_fault (mm=0x0 <__vectors_start>, addr=0, fsr=23, regs=0xee897b88) at arch/arm/mm/fault.c:150

#9 0xc0b0dc00 in do_page_fault (addr=0, fsr=23, regs=0xee897b88) at arch/arm/mm/fault.c:392

#10 0xc0008798 in do_DataAbort (addr=0, fsr=23, regs=0xee897b88) at arch/arm/mm/fault.c:550

#11 0xc0b0c9d8 in __dabt_svc () at arch/arm/kernel/entry-armv.S:197

#12 0xc0b0c9d8 in __dabt_svc () at arch/arm/kernel/entry-armv.S:197

#13 0xc0b0c9d8 in __dabt_svc () at arch/arm/kernel/entry-armv.S:197

xxxxxxxxxxxxxxxxxxxxxxxxxxxx1

Unable to handle kernel NULL pointer dereference at virtual address 00000000

pgd = c0004000

[00000000] *pgd=00000000

Internal error: Oops: 17 [#1] SMP ARM

Modules linked in:

CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.0.0 #10

Hardware name: ARM-Versatile Express

task: ee8a8000 ti: ee896000 task.ti: ee896000

PC is at clcdfb_probe+0x44/0x2f0

LR is at clcdfb_probe+0x30/0x2f0

pc : [] lr : [] psr: 60000013

sp : ee897bd0 ip : 00000004 fp : 00000000

r10: 00000000 r9 : 00000000 r8 : 00000000

r7 : 00000000 r6 : 00000000 r5 : c0affb44 r4 : 00000000

r3 : 00000000 r2 : c10b37a8 r1 : 60000013 r0 : 0000001d

Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment kernel

Control: 10c5387d Table: 8ebcc059 DAC: 00000015

Process swapper/0 (pid: 1, stack limit = 0xee896210)

Stack: (0xee897bd0 to 0xee898000)

7bc0: 60000013 c10c8afc c10c37ac eea0ec00

7be0: 60000013 eea0ec64 c10c8b60 c10c8b60 60000013 00030003 60000013 00000000

7c00: 00000000 60000013 eea0ec64 60000013 eea0ec64 eea0ec64 60000013 eea0ec64

7c20: eea0ec64 eea0ec00 00000001 60000013 c10fb134 00000000 00000000 eea0ec64

7c40: 60000013 00000000 c0affb44 c0615614 c10fb134 eea0ec00 eebe7144 80000001

7c60: eebe71e0 eebe71e0 c10fb134 eebe7140 00000004 c0314138 00000001 c0b95370

7c80: 00000000 00000001 eea0ecc4 eea0ecc4 00000001 eea0ec00 00000000 c0b95370

7ca0: eea0ec00 eea0ec08 c10c37ac c10c37c4 c10c37c4 eea0ec00 eea0ec00 00000000

7cc0: 00000000 00000000 c0affb44 c068a554 c10c37c4 eea0ec00 eea0ec64 eea0ec64

7ce0: eea0ec64 c069d468 ffffffff eea0ec00 00000000 eea0ec64 eea0ec64 00000000

7d00: 00000001 c1107f6c c1107f6c eea0ec08 eea0ed3c ffffffff eea0ecc4 00000000

7d20: ffffffff eea0ecc4 00000000 00000001 eea0ecc4 eea0ecc4 00000000 c0b95378

7d40: eea0ec00 00000001 eea0ec00 00000000 00000000 00000000 c0affb44 c068a90c

7d60: eea0ec00 c10c37c4 eea0ec08 eea0ec00 c0b95488 eea0ec00 eea0ec34 00000000

7d80: ee95f9c0 c068ab90 c10c37c4 eea0ec00 00000000 eea0ec00 c10c37c4 ee897dc8

7da0: ee95f980 eea0ec00 c10c8a98 c10c37c4 ee95e774 c0687aa8 c068aac0 c10c37c4

7dc0: 00000000 c10c3aa8 ee95e75c ee95f9b4 eea0ec00 00000000 c10c8994 c068ac0c

7de0: ee95e770 c10c37c4 c0688858 c06888b0 c0b8e7f0 ee93ef80 00000000 c10c37c4

7e00: 000080d0 00000050 000000d0 00000050 eea7f180 c0b951c4 c10c3aa8 00000000

7e20: ef5d0620 c068bec4 00000000 c10c37c4 00000000 00000000 00000006 c0615a04

7e40: 00000000 c10c37c4 eeb31380 c0d57280 00000000 c0008dc8 ee891130 c007dc78

7e60: 000000a2 c0d5724c 00000006 c0d0eb30 000000a2 c0cd69f8 eefeb32a c0b12200

7e80: 000000d0 c10fa0ec 60000153 c10b1eec c10fa0ec c10fa0ec 00000000 00000001

7ea0: eefeb329 eefeb320 c0b1bb10 60000153 c10b1ef0 60000153 c10b1eec 00000000

7ec0: 60000013 60000013 60000013 a0000113 00000000 00000000 00000000 c0d0f40c

7ee0: 00000006 00000006 c0d0eb30 c10b1dc4 c10b1dc4 00000006 c10b1dc4 c0d901bc

7f00: ee8af6c0 c0d0f460 00000000 00000006 00000000 c0d0f4ac 00000000 c0d0f724

7f20: 2df17000 ee8a84c0 00000000 c10ac7e4 c10ac7e4 ee8a8000 eefc05c0 eefc05c0

7f40: 00000001 ee8a83f4 ee8a83f4 ee8a83f4 ee8a83f4 ee8a8000 60000153 c10b8e98

7f60: 60000153 ee8a842c ee8a842c ee8a842c 00000000 ee8a842c ee8a842c 60000153

7f80: 60000153 ee8a83f4 ee8a8000 ee8a8000 00000000 c0affb54 00000000 00000000

7fa0: 00000000 eefc05c0 00000000 c0014de0 00000000 00000000 00000000 00000000

7fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000

7fe0: 00000000 00000000 00000000 00000000 00000013 00000000 00000000 00000000

[] (clcdfb_probe) from [] (amba_probe+0x11c/0x238)

[] (amba_probe) from [] (really_probe+0x210/0x424)

[] (really_probe) from [] (driver_probe_device+0x84/0xac)

[] (driver_probe_device) from [] (__driver_attach+0xd0/0x124)

[] (__driver_attach) from [] (bus_for_each_dev+0x90/0xd8)

[] (bus_for_each_dev) from [] (driver_attach+0x28/0x38)

[] (driver_attach) from [] (bus_add_driver+0x15c/0x2a8)

[] (bus_add_driver) from [] (driver_register+0x10c/0x17c)

[] (driver_register) from [] (amba_driver_register+0x88/0x98)

[] (amba_driver_register) from [] (amba_clcdfb_init+0x34/0x40)

[] (amba_clcdfb_init) from [] (do_one_initcall+0x68/0x190)

[] (do_one_initcall) from [] (do_initcall_level+0xc8/0x100)

[] (do_initcall_level) from [] (do_initcalls+0x1c/0x40)

[] (do_initcalls) from [] (do_basic_setup+0x28/0x34)

[] (do_basic_setup) from [] (kernel_init_freeable+0x218/0x2c4)

[] (kernel_init_freeable) from [] (kernel_init+0x10/0x16c)

[] (kernel_init) from [] (ret_from_fork+0x14/0x34)

Code: e58d3068 e3a03000 e58d3064 e59d3068 (e5933000)

---[ end trace 3d127599db201418 ]---

Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b

---[ end Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b