Reputation: 1021
I am a novice in the field of C++ multithread program. I want to read block data from my disk and do some computation based on these data. In order to simplify the code, I write the following demo to test my idea. However, I found some questions.
#include <iostream>
#include <thread>
#include <string>
#include <mutex>
#include <vector>
#include <condition_variable>
std::condition_variable cv;
std::mutex cv_m;
bool is_print;
void read_value(std::vector<int> &data)
{
for (int j = 1; j < 5; j++)
{
std::cout << "read data iteration is " << j << std::endl;
std::lock_guard<std::mutex> lk(cv_m);
for (int i = 0; i < 5; i++)
data.at(i) = i * j;
is_print = true;
cv.notify_one();
}
}
void print_value(const std::vector<int> &data)
{
std::cout << "output data" << std::endl;
std::unique_lock<std::mutex> lk(cv_m);
cv.wait(lk, []{return is_print;});
for (int i = 0; i < data.size(); i++)
std::cout << data[i] << std::endl;
is_print = false;
}
int main()
{
is_print = false;
std::vector<int> data = {0, 0, 0, 0, 0};
std::thread thread_read(read_value, data);
std::thread thread_print(print_value, data);
thread_print.join();
thread_read.join();
return 0;
}
In this demo, I use "read_value" to simulate read data from disk and I want to read data from disk several times. So, I add an outer loop in the function "read_value". After that, I use "print_value" function to output the data. However, in such case, I meet the error as the following, which tells me that I use the deleted function. I am not sure what the problem is and how to read data several times. Thank you so much!
In file included from
/Users/zsk/Downloads/programming/C++/multi_thread/multi_thread/main.cpp:10:
/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/thread:342:5: error: attempt to use a deleted function
__invoke(_VSTD::move(_VSTD::get<1>(__t)), _VSTD::move(_VSTD::get<_Indices>(__t))...);
^
/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/thread:352:5: note: in instantiation of function template specialization 'std::__1::__thread_execute<std::__1::unique_ptr<std::__1::__thread_struct, std::__1::default_delete<std::__1::__thread_struct> >, void (*)(std::__1::vector<int, std::__1::allocator<int> > &), std::__1::vector<int, std::__1::allocator<int> > , 2>' requested here
__thread_execute(*__p, _Index());
^
/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/v1/thread:368:47: note: in instantiation of function template specialization 'std::__1::__thread_proxy<std::__1::tuple<std::__1::unique_ptr<std::__1::__thread_struct, std::__1::default_delete<std::__1::__thread_struct> >, void (*)(std::__1::vector<int, std::__1::allocator<int> > &), std::__1::vector<int, std::__1::allocator<int> > > >' requested here
int __ec = __libcpp_thread_create(&__t_, &__thread_proxy<_Gp>, __p.get());
^
/Users/zsk/Downloads/programming/C++/multi_thread/multi_thread/main.cpp:51:17: note: in instantiation of function template specialization 'std::__1::thread::thread<void (&)(std::__1::vector<int, std::__1::allocator<int> > &), std::__1::vector<int, std::__1::allocator<int> > &, void>' requested here
std::thread thread_read(read_value, data);
Upvotes: 0
Views: 929
Reputation: 11317
Working with threads is surprising if you try to use references. The other answer is completely right, use std::ref
to pass the arguments.
Personally, I prefer to use lambdas instead:
std::thread thread_read([&data](){ read_value(data); });
std::thread thread_print([&data = std::as_const(data)](){ print_value(data); });
It might not be as elegant as the std::ref
though it uses the concepts you already know from all other places instead of requiring you to remember the unexpected copying.
Upvotes: 0
Reputation: 409136
The problem is that the std::thread
object needs to be able to copy the arguments you pass to the threads. And references can't be copied.
To solve this you need to use std::ref
or std::cref
to wrap the arguments:
std::thread thread_read(read_value, std::ref(data));
std::thread thread_print(print_value, std::cref(data));
Upvotes: 5