解决海思平台下C++启动线程报pure virtual method called, terminate called without an active exception问题

背景

最近在移植已有C++11的项目到海思平台中，工程师反馈遇到了下面的报错

pure virtual method called
terminate called without an active exception

pure virtual method called

terminate called without an active exception

而且卡了好几天都没解决。CSDN上也有人反馈说没有解决，到底该如何解决呢?

排查

首先看看环境。使用的工具链是arm-hisiv600-linux，其自身是gcc 4.9.4，板子是Hi3531D，CPU是A9核心。这里首先做了一个最简的例子编译并测试。

#include <iostream>
#include <thread>

int main() {
    std::thread thread_object([](){
        std::cout << "print in thread" << std::endl;
    });
    thread_object.join();
    std::cout << "thread joined" << std::endl;
    return 0;
}

#include <iostream>

#include <thread>

int main() {

std::thread thread_object([](){

std::cout << "print in thread" << std::endl;

});

thread_object.join();

std::cout << "thread joined" << std::endl;

return 0;

}

编译命令如下

arm-hisiv600-linux-gnueabi-g++ -o hello_cpp hello.cpp -std=c++11 -mcpu=cortex-a9 -mfpu=neon-fp16

1 2	arm-hisiv600-linux-gnueabi-g++ -o hello_cpp hello.cpp -std=c++11 -mcpu=cortex-a9 -mfpu=neon-fp16

编译出来一切正常，通过readelf确认下是否架构信息正确

$ readelf -a hello_cpp | grep Tag
  Tag_CPU_name: "Cortex-A9"
  Tag_CPU_arch: v7
  Tag_CPU_arch_profile: Application
  Tag_ARM_ISA_use: Yes
  Tag_THUMB_ISA_use: Thumb-2
  Tag_ABI_PCS_wchar_t: 4
  Tag_ABI_FP_rounding: Needed
  Tag_ABI_FP_denormal: Needed
  Tag_ABI_FP_exceptions: Needed
  Tag_ABI_FP_number_model: IEEE 754
  Tag_ABI_align_needed: 8-byte
  Tag_ABI_align_preserved: 8-byte, except leaf SP
  Tag_ABI_enum_size: int
  Tag_CPU_unaligned_access: v6
  Tag_MPextension_use: Allowed
  Tag_Virtualization_use: TrustZone

$ readelf -a hello_cpp | grep Tag

Tag_CPU_name: "Cortex-A9"

Tag_CPU_arch: v7

Tag_CPU_arch_profile: Application

Tag_ARM_ISA_use: Yes

Tag_THUMB_ISA_use: Thumb-2

Tag_ABI_PCS_wchar_t: 4

Tag_ABI_FP_rounding: Needed

Tag_ABI_FP_denormal: Needed

Tag_ABI_FP_exceptions: Needed

Tag_ABI_FP_number_model: IEEE 754

Tag_ABI_align_needed: 8-byte

Tag_ABI_align_preserved: 8-byte, except leaf SP

Tag_ABI_enum_size: int

Tag_CPU_unaligned_access: v6

Tag_MPextension_use: Allowed

Tag_Virtualization_use: TrustZone

看出结果是没问题的，Cortex-A9 + armv7 + Thumb2

结果放到板子上跑却遇到了另一个报错

# ./hello_cpp
terminate called after throwing an instance
'std::system_error'
 what(): Enable multithreading to use std::thread: Operation not permitted
Aborted

# ./hello_cpp

terminate called after throwing an instance

'std::system_error'

what(): Enable multithreading to use std::thread: Operation not permitted

Aborted

至少好像信息有些变化，我们再针对这个gcc4.9和std::thread的operation not permitted再搜索下看，结果直接就查到了这个链接，文中提到了加入”-thread”编译选项可以解决问题。

接下来我尝试了下新的编译命令

arm-hisiv600-linux-gnueabi-g++ -o hello_cpp hello.cpp -std=c++11 -thread -mcpu=cortex-a9 -mfpu=neon-fp16

1 2	arm-hisiv600-linux-gnueabi-g++ -o hello_cpp hello.cpp -std=c++11 -thread -mcpu=cortex-a9 -mfpu=neon-fp16

编译完成后再次运行，就一切正常了。

分析

那为什么一开始gcc 4.9在编译带std::thread的时候不能正常启用pthread呢? 这可能是个bug，并且可以通过以下手段进行验证。对未加-pthread编译出来的文件使用readelf查看其链接到的动态链接库，

$ readelf -d hello_cpp | grep "共享库"
 0x00000001 (NEEDED)                     共享库：[libstdc++.so.6]
 0x00000001 (NEEDED)                     共享库：[libm.so.6]
 0x00000001 (NEEDED)                     共享库：[libgcc_s.so.1]
 0x00000001 (NEEDED)                     共享库：[libc.so.6]

$ readelf -d hello_cpp | grep "共享库"

0x00000001 (NEEDED) 共享库：[libstdc++.so.6]

0x00000001 (NEEDED) 共享库：[libm.so.6]

0x00000001 (NEEDED) 共享库：[libgcc_s.so.1]

0x00000001 (NEEDED) 共享库：[libc.so.6]

发现其实不包含libpthread.so.0，其实这就是原因。从编译到链接都不包含线程库支持。

唉~真是坑啊