最新消息:20190717 VPS服务器:Vultr新加坡,WordPress主题:大前端D8,统一介绍入口:关于

Oops Messages中英文版【转】

工作和技术 crifan 536浏览 0评论

Oops Messages中英文版【转】

转者注:这个是别人翻译的。英文原文,可以在这下载到:

Debugging Techniques

也就是 Linux Device Drivers, Third Edition,简称LLD3中的第四章中的内容。


    An oops displays the processor status at the time of the fault, including the contents of the CPU registers and other seemingly incomprehensible information. The message is generated by printk statements in the fault handler (arch/*/kernel/traps.c) and is dispatched as described earlier in Section 4.2.1).

    一条oops消息能够显示发生故障时处理器的状态,以及CPU寄存器的内容和其他从表面难以理解的信息。该消息是由容错处理中的printk语句产生的(arch/*kernel/traps.c)并按照4.2.1小节中描述的方式进行分派。

    Let’s look at one such message. Here’s what results from dereferencing a NULL pointer on a PC running Version 2.6 of the kernel. The most relevant information here is the instruction pointer (EIP), the address of the faulty instruction.

    下面我们来看一条这样的消息。这是通过在一台运行着2.6内核的PC机上废弃一个NULL指针所引起的。其中最有关的信息是指令指针(EIP),就是故障指令的地址。

Unable to handle kernel NULL pointer dereference at virtual address 00000000
printing eip:
d083a064
Oops: 0002 [#1]
SMP
CPU:    0
EIP:    0060:[<d083a064>]    Not tainted
EFLAGS: 00010246   (2.6.6)
EIP is at faulty_write+0x4/0x10 [faulty]
eax: 00000000   ebx: 00000000   ecx: 00000000   edx: 00000000
esi: cf8b2460   edi: cf8b2480   ebp: 00000005   esp: c31c5f74
ds: 007b   es: 007b   ss: 0068
Process bash (pid: 2086, threadinfo=c31c4000 task=cfa0a6c0)
Stack: c0150558 cf8b2460 080e9408 00000005 cf8b2480 00000000 cf8b2460 cf8b2460
       fffffff7 080e9408 c31c4000 c0150682 cf8b2460 080e9408 00000005 cf8b2480
       00000000 00000001 00000005 c0103f8f 00000001 080e9408 00000005 00000005
Call Trace:
[<c0150558>] vfs_write+0xb8/0x130
[<c0150682>] sys_write+0x42/0x70
[<c0103f8f>] syscall_call+0x7/0xb
Code: 89 15 00 00 00 00 c3 90 8d 74 26 00 83 ec 0c b8 00 a6 83 d0

    This message was generated by writing to a device owned by the faulty module, a module built deliberately to demonstrate failures. The implementation of the write method of faulty.c is trivial:

    这条消息是由一个问题模块向其设备执行写操作时引起的,该模块是特意为示范故障而构建的。faulty.c中的write函数很普通:

ssize_t faulty_write (struct file *filp, const char _ _user *buf, size_t count, loff_t *pos)
{
    /* make a simple fault by dereferencing a NULL pointer */
    *(int *)0 = 0;
    return 0;
}

    As you can see, what we do here is dereference a NULL pointer. Since 0 is never a valid pointer value, a fault occurs, which the kernel turns into the oops message shown earlier. The calling process is then killed.

    如你所见,我们在这里做的就是废弃一个NULL指针。因为0从来都不是一个可用的指针值,所以会引发一个故障,内核会简单地将其转换为oops消息并显示。然后其调用进程会被杀死。

The faulty module has a different fault condition in its read implementation: faulty

该示例模块在其read函数中则有着不同的故障条件:

ssize_t faulty_read(struct file *filp, char _ _user *buf, size_t count, loff_t *pos)
{
    int ret;
    char stack_buf[4];
    /* Let’s try a buffer overflow */
    memset(stack_buf, 0xff, 20);
    if (count > 4)
        count = 4;
    /* copy 4 bytes to the user */
    ret = copy_to_user(buf, stack_buf, count);
    if (!ret)
        return count;
    return ret;
}

    This method copies a string into a local variable; unfortunately, the string is longer than the destination array. The resulting buffer overflow causes an oops when the function returns. Since the return instruction brings the instruction pointer to nowhere land, this kind of fault is much harder to trace, and you can get something such as the following:

    该函数将一个字符串赋给一个局部变量;不幸的是,字符串的长度超出了目标数组的范围。当函数返回时就会导致缓冲区溢出而引起一条oops消息。由于返回指令会带来指向空地址的指针,因此这类故障更加难以跟踪,你将会看到下面这样的信息:

EIP:    0010:[<00000000>]
Unable to handle kernel paging request at virtual address ffffffff
printing eip:
ffffffff
Oops: 0000 [#5]
SMP
CPU:    0
EIP:    0060:[<ffffffff>]    Not tainted
EFLAGS: 00010296   (2.6.6)
EIP is at 0xffffffff
eax: 0000000c   ebx: ffffffff   ecx: 00000000   edx: bfffda7c
esi: cf434f00   edi: ffffffff   ebp: 00002000   esp: c27fff78
ds: 007b   es: 007b   ss: 0068
Process head (pid: 2331, threadinfo=c27fe000 task=c3226150)
Stack: ffffffff bfffda70 00002000 cf434f20 00000001 00000286 cf434f00 fffffff7
       bfffda70 c27fe000 c0150612 cf434f00 bfffda70 00002000 cf434f20 00000000
       00000003 00002000 c0103f8f 00000003 bfffda70 00002000 00002000 bfffda70
Call Trace:
[<c0150612>] sys_read+0x42/0x70
[<c0103f8f>] syscall_call+0x7/0xb
Code: Bad EIP value.

    In this case, we see only part of the call stack (vfs_read and faulty_read are missing), and the kernel complains about a "bad EIP value." That complaint, and the offending address (ffffffff) listed at the beginning are both hints that the kernel stack has been corrupted.

    这种情况下,你只能看到部分函数调用的堆栈情况(vfs_readfaulty_read丢失了),而且内核会为了一个“不可用的EIP值”而抱怨。这种抱怨以及开始部分列出的讨厌地址(ffffffff)都暗示了内核堆栈已经坍塌。

    In general, when you are confronted with an oops, the first thing to do is to look at the location where the problem happened, which is usually listed separately from the call stack. In the first oops shown above, the relevant line is:

    通常,当你面临一个oops时,首要问题就是查看故障的发生位置,它通常会与函数调用的堆栈信息分开列出。对于上面第一个oops,与此相关的信息是:

EIP is at faulty_write+0x4/0x10 [faulty]

    Here we see that we were in the function faulty_write , which is located in the faulty module (which is listed in square brackets). The hex numbers indicate that the instruction pointer was 4 bytes into the function, which appears to be 10 (hex) bytes long. Often that is enough to figure out what the problem is.

     这里可以看出我们正位于faulty模块(方括号中的是模块名称)faulty_write函数中。十六进制的数字指明了该函数中的指令指针长度为4字节,而现在看起来则有10(十六进制)字节长。通常这足以查明问题的所在了。

     If you need more information, the call stack shows you how you got to where things fell apart. The stack itself is printed in hex form; with a bit of work, you can often determine the values of local variables and function parameters from the stack listing. Experienced kernel developers can benefit from a certain amount of pattern recognition here; for example, if we look at the stack listing from the faulty_read oops:

    如果你需要更多信息,函数调用的堆栈信息将会告诉你怎样找到已崩溃的东西。堆栈信息会以十六进制列出;稍加分析,你就能从中辨别出局部变量以及函数参数。有经验的内核开发者会从中获得很大的帮助;例如,faulty_read函数的堆栈信息如下所示:

Stack: ffffffff bfffda70 00002000 cf434f20 00000001 00000286 cf434f00 fffffff7
            bfffda70 c27fe000 c0150612 cf434f00 bfffda70 00002000 cf434f20 00000000
            00000003 00002000 c0103f8f 00000003 bfffda70 00002000 00002000 bfffda70

   The ffffffff at the top of the stack is part of our string that broke things. On the x86 architecture, by default, the user-space stack starts just below 0xc0000000; thus, the recurring value 0xbfffda70 is probably a user-space stack address; it is, in fact, the address of the buffer passed to the read system call, replicated each time it is passed down the kernel call chain. On the x86 (again, by default), kernel space starts at 0xc0000000, so values above that are almost certainly kernel-space addresses, and so on.

   位于堆栈顶部的ffffffff是引发故障的字符串的一部分。在x86体系中,默认用户空间中的堆栈地址是小于0xc00000000的;因此,其中0xbfffda70很有可能是一个用户空间的堆栈地址;实际上它就是传递给read系统调用的缓冲区的地址,它在内核调用链中每次被下传时都会被复制。在x86(再次说明,缺省的),内核空间地址起始自0xc00000000,所以可以基本确定凡是大于该值的地址都是属于内核空间的地址。

   Finally, when looking at oops listings, always be on the lookout for the "slab poisoning" values discussed at the beginning of this chapter. Thus, for example, if you get a kernel oops where the offending address is 0xa5a5a5a5, you are almost certainly forgetting to initialize dynamic memory somewhere.

   最后要注意的一点是,当你查看oops信息时,始终要留意本章开始时讨论的“slab poisoning”的值。因此,如果一条内核oops中出现了讨厌的地址值0xa5a5a5a5,那么你肯定是在什么地方忘记初始化动态分配的内存了。

   Please note that you see a symbolic call stack (as shown above) only if your kernel is built with the CONFIG_KALLSYMS option turned on. Otherwise, you see a bare, hexadecimal listing, which is far less useful until you have decoded it in other ways.

   请注意要想看到一条可读的调用堆栈信息(如上所示),你必须要在构建内核时启用CONFIG_KALLSYMS选项。否则你将会看到一个原始的十六进制列表,在你使用其他方法译解它之前它几乎没什么用处。

转载请注明:在路上 » Oops Messages中英文版【转】

发表我的评论
取消评论

表情

Hi,您需要填写昵称和邮箱!

  • 昵称 (必填)
  • 邮箱 (必填)
  • 网址
53 queries in 0.115 seconds, using 18.86MB memory