如何使用堆栈跟踪来确定崩溃的原因？

Question

我了解堆栈跟踪背后的“大创意”，但实际上如何使用堆栈跟踪来确定崩溃的原因？例如，我正在编写一个 c++ 宏，而我的解释器会输出以下堆栈跟踪：

>  *** Break *** illegal instruction
> 
> 
> 
> =========================================================== There was a crash. This is the entire stack trace of all threads:
> ===========================================================
> #0  0x00007fb64ef29bbc in waitpid () from /lib64/libc.so.6
> #1  0x00007fb64eea7ea2 in do_system () from /lib64/libc.so.6
> #2  0x00007fb64ffb3d84 in TUnixSystem::StackTrace (this=0x98e960) at /cvmfs/myusername/trunk/centos7/source/root-6.16.00/core/unix/src/TUnixSystem.cxx:2413
> #3  0x00007fb64ffb64bc in TUnixSystem::DispatchSignals (this=0x98e960, sig=kSigIllegalInstruction) at
> /cvmfs/myusername/trunk/centos7/source/root-6.16.00/core/unix/src/TUnixSystem.cxx:3644
> #4  <signal handler called>
> #5  0x00007fb65057f54a in ?? ()
> #6  0x000000000146e4ec in ?? ()
> #7  0x000000000028dc9d in ?? ()
> #8  0x0028dcf80146e51c in ?? ()
> #9  0x00007ffd08a37890 in ?? ()
> #10 0x0000000000051b93 in ?? ()
> #11 0x000000000000a372 in ?? ()
> #12 0x0000a3e900051c01 in ?? ()
> #13 0x000000000000146e in ?? ()
> #14 0x00007ffd08a3788f in ?? ()
> #15 0x0000000000000000 in ?? ()
> ===========================================================
> 
> 
> The lines below might hint at the cause of the crash. You may get help
> by asking at the ROOT forum http://root.cern.ch/forum Only if you are
> really convinced it is a bug in ROOT then please submit a report at
> http://root.cern.ch/bugs Please post the ENTIRE stack trace from above
> as an attachment in addition to anything else that might help us
> fixing this issue.
> ===========================================================
> #5  0x00007fb65057f54a in ?? ()
> #6  0x000000000146e4ec in ?? ()
> #7  0x000000000028dc9d in ?? ()
> #8  0x0028dcf80146e51c in ?? ()
> #9  0x00007ffd08a37890 in ?? ()
> #10 0x0000000000051b93 in ?? ()
> #11 0x000000000000a372 in ?? ()
> #12 0x0000a3e900051c01 in ?? ()
> #13 0x000000000000146e in ?? ()
> #14 0x00007ffd08a3788f in ?? ()
> #15 0x0000000000000000 in ?? ()
> ===========================================================

Answer 1

堆栈跟踪背后的想法是查看崩溃时谁在调用谁。因此，您可以检查产生故障的函数并查看它是如何被调用的。
您在示例中看到的是函数的地址，当然，如果堆栈以某种方式损坏，您也可以找到无效地址。
要查看更多有用的信息，您应该使用调试信息编译代码，这样您不仅会看到地址，还会看到函数的名称。 例如，在 gcc 中，您可以使用“-g”启用调试信息，即：

gcc -g test.c -o test

以下是“test.c”错误代码的示例：

void foo()
{
    int a=1/0;
}
void foo1()
{
    foo();
}
void foo2()
{
    foo1();
}
void foo3()
{
    foo2();
}

int main()
{
    foo3();
};

这是它产生的回溯（在 gdb 中运行，信息可用，因为我使用 -g 标志进行编译）

#0  0x0000000000401101 in foo () at test.c:3
#1  0x0000000000401117 in foo1 () at test.c:7
#2  0x0000000000401128 in foo2 () at test.c:11
#3  0x0000000000401139 in foo3 () at test.c:15
#4  0x000000000040114a in main () at test.c:20