36.Linux驱动调试-根据oops定位错误代码行

1.当驱动有误时,比如,访问的内存地址是非法的,便会打印一大串的oops出来

1.1以LED驱动为例

将open()函数里的ioremap()屏蔽掉,直接使用物理地址的GPIOF,如下图所示:

 

1.2然后编译装载26th_segmentfault并执行测试程序后,内核便打印了oops出来,如下图所示:

 

 

2.接下来,我们便来分析oops:

Unable to handle kernel paging request at virtual address 56000050
      //无法处理内核页面请求的虚拟地址56000050

pgd = c3850000
[56000050] *pgd=00000000

Internal error: Oops: 5 [#1]
        //内部错误oops

Modules linked in: 26th_segmentfault
        //表示内部错误发生在26th_segmentfault.ko驱动模块里

CPU: 0    Not tainted  (2.6.22.6 #2)
PC is at first_drv_open+0x78/0x12c [26th_segmentfault]
        //PC值:程序运行成功的最后一次地址,位于first_drv_open()函数里,偏移值0x78,该函数总大小0x12c

LR is at 0xc0365ed8             //LR值

/*发生错误时的各个寄存器值*/
pc : [<bf000078>]    lr : [<c0365ed8>]    psr: 80000013
sp : c3fcbe80  ip : c0365ed8  fp : c3fcbe94
r10: 00000000  r9 : c3fca000  r8 : c04df960
r7 : 00000000  r6 : 00000000  r5 : bf000de4  r4 : 00000000
r3 : 00000000  r2 : 56000050  r1 : 00000001  r0 : 00000052

Flags: Nzcv  IRQs on  FIQs on  Mode SVC_32  Segment user
Control: c000717f  Table: 33850000  DAC: 00000015
Process 26th_segmentfau (pid: 813, stack limit = 0xc3fca258)
            //发生错误时,进程名称为26th_segmentfault

Stack: (0xc3fcbe80 to 0xc3fcc000)                     //栈信息
be80: c06d7660 c3e880c0 c3fcbebc c3fcbe98 c008d888 bf000010 00000000 c04df960
bea0: c3e880c0 c008d73c c0474e20 c3fb9534 c3fcbee4 c3fcbec0 c0089e48 c008d74c
bec0: c04df960 c3fcbf04 00000003 ffffff9c c002c044 c380a000 c3fcbefc c3fcbee8
bee0: c0089f64 c0089d58 00000000 00000002 c3fcbf68 c3fcbf00 c0089fb8 c0089f40
bf00: c3fcbf04 c3fb9534 c0474e20 00000000 00000000 c3851000 00000101 00000001
bf20: 00000000 c3fca000 c04c90a8 c04c90a0 ffffffe8 c380a000 c3fcbf68 c3fcbf48
bf40: c008a16c c009fc70 00000003 00000000 c04df960 00000002 be84ce38 c3fcbf94
bf60: c3fcbf6c c008a2f4 c0089f88 00008588 be84ce84 00008718 0000877c 00000005
bf80: c002c044 4013365c c3fcbfa4 c3fcbf98 c008a3a8 c008a2b0 00000000 c3fcbfa8
bfa0: c002bea0 c008a394 be84ce84 00008718 be84ce30 00000002 be84ce38 be84ce30
bfc0: be84ce84 00008718 0000877c 00000003 00008588 00000000 4013365c be84ce58
bfe0: 00000000 be84ce28 0000266c 400c98e0 60000010 be84ce30 30002031 30002431

Backtrace:                                        //回溯信息
[<bf000000>] (first_drv_open+0x0/0x12c [26th_segmentfault]) from [<c008d888>] (chrdev_open+0x14c/0x164)
 r5:c3e880c0 r4:c06d7660
[<c008d73c>] (chrdev_open+0x0/0x164) from [<c0089e48>] (__dentry_open+0x100/0x1e8)
 r8:c3fb9534 r7:c0474e20 r6:c008d73c r5:c3e880c0 r4:c04df960
[<c0089d48>] (__dentry_open+0x0/0x1e8) from [<c0089f64>] (nameidata_to_filp+0x34/0x48)
[<c0089f30>] (nameidata_to_filp+0x0/0x48) from [<c0089fb8>] (do_filp_open+0x40/0x48)
 r4:00000002
[<c0089f78>] (do_filp_open+0x0/0x48) from [<c008a2f4>] (do_sys_open+0x54/0xe4)
 r5:be84ce38 r4:00000002
[<c008a2a0>] (do_sys_open+0x0/0xe4) from [<c008a3a8>] (sys_open+0x24/0x28)
[<c008a384>] (sys_open+0x0/0x28) from [<c002bea0>] (ret_fast_syscall+0x0/0x2c)
Code: bf000094 bf0000b4 bf0000d4 e5952000 (e5923000)
Segmentation fault

2.1上面的回溯信息,表示了函数的整个调用过程

比如上面的回溯信息表示:

  • sys_open()->do_sys_open()->do_filp_open()->nameidata_to_filp()->chrdev_open()->first_drv_open();

最终错误出在了first_drv_open();

 若内核没有配置回溯信息显示,则就不会打印函数调用过程,可以修改内核的.config文件,添加:

 //CONFIG_FRAME_POINTER,表示帧指针,用fp寄存器表示

 内核里,就会通过fp寄存器记录函数的运行位置,并存到栈里,然后当出问题时,从栈里调出fp寄存器,查看函数的调用关系,就可以看到回溯信息.

(PS:若不配置,也可以直接通过栈来分析函数调用过程,在下章会分析到:http://www.cnblogs.com/lifexy/p/8011966.html)

2.2而有些内核的环境不同,opps也可能不会打印出上面的:

Modules linked in: 26th_segmentfault
PC is at first_drv_open+0x78/0x12c [26th_segmentfault]

这些相关信息, 只打印PC值,就根本无法知道,到底是驱动模块出的问题,还是内核自带的函数出的问题? 

所以oops里的最重要内容还是这一段: pc : [<bf000078>]

 

2.3那么如何来确定,该PC值地址位于内核的函数,还是我们装载的驱动模块?

答:

可以在内核源码的根目录下通过的“vi System.map”来查看,该文件保存了内核里所有(符号、函数)的虚拟地址映射,比如下图的内核函数root_dev_setup():

 

通过vi命令的:0和:$命令行,可以看到内核的虚拟地址是c0004000~c03cebf4

所以,pc值bf000078为的驱动模块的地址值 

 

2.4当有多个驱动装载时,又如何区分PC值是哪个驱动的函数的地址值?

答:通过/proc/kallsyms来查看:

#cat /proc/kallsyms  //(kernel all symbols)查看所有的内核标号(包括内核函数,装载的驱动函数,变量符号等)的地址值

或者:

#cat  /proc/kallsyms> /kallsyms.txt    //将地址值放入kallsyms.txt中

如下图所示,在kallsyms.txt里,找到pc值bf000078位于26th_segmentfault驱动里first_drv_open()函数下的bf000000+0x78

 

2.5然后将驱动生成反汇编:

arm-linux-objdump -D 26th_segmentfault.ko >26th_segmentfault.dis //反汇编

 

2.6打开反汇编:

如下图所示,左边是kallsyms.txt,右边是26th_segmentfault.dis反汇编

 

显然pc值bf000078,就位于反汇编的78地址处:

Disassembly of section .text:         //.text段起始地址为0x00
00000000 <first_drv_open>:

38: e59fc0e8   ldr   ip, [pc, #232]; 128 <.text+0x128> //ip=.text段+0x128里的内容
... ...

50: e585c000   str   ip, [r5]       //r5=.text段+0x128里的内容
... ...

74: e5952000   ldr   r2, [r5]           //r2=.text段+0x128里的内容
78: e5923000   ldr   r3, [r2]                  // r3=.text段+0x128里的内容
7c: e3c33c3f   bic   r3,  r3,   #16128    ;0x3f00  //清除0x56000050的bit8~13
... ...

128:  56000050  undefined                //.text段+0x128里的内容=0x56000050

从上面看到,78地址处,主要是将0x56000050(r2)地址里的内容放入r3中.

0x56000050是个物理地址,在linux眼中便是个非法地址,所以出错

并找到出错地方位于first_drv_open ()函数下:

 

3.若发生错误的驱动位于内核的地址值时

3.1还是以26th_segmentfault.c为例,首先加入内核:

#cp 26th_segmentfault.c   /linux-2.6.22.6/drivers/char/  //将有问题的驱动复制到字符驱动目录下

 

#vi Makefile

添加:

obj-y    += 26th_segmentfault.o                            //y:将该驱动放入内核中

 

3.2然后make uImage装载新内核后,再运行测试程序,便会打印出opps信息

3.3在内核源码的根目录下通过:

# arm-none-linux-gnueabi-objdump -D vmlinux > vmlinux.dis

将整个内核反汇编, vmlinux:未压缩的内核

3.4 vi vmlinux.dis,然后通过oops信息的PC值直接来查找地址即可

 

接下来下章便通过信息来分析函数调用过程:http://www.cnblogs.com/lifexy/p/8011966.html

 

posted @ 2017-12-08 18:49  诺谦  阅读(5915)  评论(0编辑  收藏  举报