undefined reference to `void multiprocess(int, std::vector >, void (*)(Photon*))'

Question

I am new to c++, and trying to implement multithreading in my code. However, I get this weird error during compilation:

photongen.cc:(.text+0x90a): undefined reference to `void multiprocess(int, std::vector<Photon, std::allocator >, void ()(Photon))'

I've declared multithreading function in my header file and defined it in another cpp file.

My code snippets: Photongen:

void photongen() {

// Iterating over Muon list
for (auto m = muonList.begin(); m != muonList.end(); ++m) {
    
    nP = 0; // Initialise the number of photons to 0

    vector<Photon> tempP;   // Temporary vector to store generated photons
    
    // Muon propagation
    bool stopLoop = false;
    while (!stopLoop) {
        
        // Photon generation
        if (nPhotons()) {
            for (int i = 0; i < nPhotons(); i++){

                // Create photon instance
                Photon *P = new Photon(m->x, m->y, m->z, m->run, m->event, i);
                tempP.push_back(*P);
                delete P;   // Free memory by photon object
            }
        }

        // Check if muon has left the detector and stop loop...
        ...
    }

    multiprocess<Photon>(1, tempP, &propagation);

    tempP.clear();
}}

propagation

void propagation(Photon *P) {
    if (photonProp(P)) {
        pLock.lock();
        photonList.push_back(*P);
        nP++;
        pLock.unlock();
    }}

multiprocess

template <typename T>
void multiprocess(int nThreads, vector<T> arg, void (*func)(T*)) {

    int i;

    for (i=0; i < arg.size() - nThreads; i += nThreads) {
        
        thread t[nThreads];

        for (int j = 0; j < nThreads; j++){
            // cout << i+j << endl;
           t[j] = thread(func, arg[i+j]);
        }

        for (int j = 0; j < nThreads; j++){
            t[j].join();
        }
    }

    

    thread t[arg.size()%nThreads];

    for (int j = 0; j < arg.size()%nThreads; j++){
        t[j] = thread(func, arg[i + j]);
    }

    for (int j = 0; j < arg.size()%nThreads; j++){
        t[j].join();
    }
    
}