User-Space thread balancing + refactoring

.dockerignore

@@ -19,7 +19,7 @@ Dockerfile
 /frontend/build/
 /frontend/build/**
 /frontend/node_modules/
-/backend/modules/cppqueue
+/backend/modules/cppregex
 /backend/modules/proxy
 docker-compose.yml
 

.gitignore

@@ -21,8 +21,10 @@
 /frontend/build/**
 /frontend/dist/
 /frontend/dist/**
-/backend/modules/cppqueue
+/backend/modules/cppregex
-/backend/binsrc/cppqueue
+/backend/binsrc/cppregex
+/backend/modules/cpproxy
+/backend/binsrc/cpproxy
 /backend/modules/proxy
 /docker-compose.yml
 /firegex-compose.yml

Dockerfile

@@ -15,9 +15,9 @@ RUN bun run build
 
 #Building main conteiner
 FROM --platform=$TARGETARCH registry.fedoraproject.org/fedora:latest
-RUN dnf -y update && dnf install -y python3-pip @development-tools gcc-c++ \
+RUN dnf -y update && dnf install -y python3-devel python3-pip @development-tools gcc-c++ \
-    libnetfilter_queue-devel libnfnetlink-devel libmnl-devel libcap-ng-utils \
+    libnetfilter_queue-devel libnfnetlink-devel libmnl-devel libcap-ng-utils nftables \
-    nftables vectorscan-devel libtins-devel python3-nftables libpcap-devel boost-devel
+    vectorscan-devel libtins-devel python3-nftables libpcap-devel boost-devel
 
 RUN mkdir -p /execute/modules
 WORKDIR /execute
@@ -26,8 +26,8 @@ ADD ./backend/requirements.txt /execute/requirements.txt
 RUN pip3 install --no-cache-dir --break-system-packages -r /execute/requirements.txt --no-warn-script-location
 
 COPY ./backend/binsrc /execute/binsrc
-RUN g++ binsrc/nfqueue.cpp -o modules/cppqueue -std=c++23 -O3 -lnetfilter_queue -pthread -lnfnetlink $(pkg-config --cflags --libs libtins libhs libmnl)
+RUN g++ binsrc/nfregex.cpp -o modules/cppregex -std=c++23 -O3 -lnetfilter_queue -pthread -lnfnetlink $(pkg-config --cflags --libs libtins libhs libmnl)
-#RUN g++ binsrc/nfproxy-tun.cpp -o modules/cpproxy -std=c++23 -O3 -lnetfilter_queue -pthread -lnfnetlink $(pkg-config --cflags --libs libtins libmnl)
+RUN g++ binsrc/nfproxy-tun.cpp -o modules/cpproxy -std=c++23 -O3 -lnetfilter_queue -pthread -lnfnetlink $(pkg-config --cflags --libs libtins libmnl python3)
 
 COPY ./backend/ /execute/
 COPY --from=frontend /app/dist/ ./frontend/

backend/app.py

@@ -42,14 +42,6 @@ app = FastAPI(
     title="Firegex API",
     version=API_VERSION,
 )
-utils.socketio = socketio.AsyncServer(
-    async_mode="asgi",
-    cors_allowed_origins=[],
-    transports=["websocket"]
-)
-
-sio_app = socketio.ASGIApp(utils.socketio, socketio_path="/sock/socket.io", other_asgi_app=app)
-app.mount("/sock", sio_app)
 
 if DEBUG:
     app.add_middleware(
@@ -61,6 +53,15 @@ if DEBUG:
     )
 
 
+utils.socketio = socketio.AsyncServer(
+    async_mode="asgi",
+    cors_allowed_origins=[],
+    transports=["websocket"]
+)
+
+sio_app = socketio.ASGIApp(utils.socketio, socketio_path="/sock/socket.io", other_asgi_app=app)
+app.mount("/sock", sio_app)
+
 def APP_STATUS(): return "init" if db.get("password") is None else "run"
 def JWT_SECRET(): return db.get("secret")
 
@@ -197,10 +198,11 @@ if __name__ == '__main__':
     os.chdir(os.path.dirname(os.path.realpath(__file__)))
     uvicorn.run(
         "app:app",
-        host=None, #"::" if DEBUG else None,
+        host="::" if DEBUG else None,
         port=FIREGEX_PORT,
-        reload=False,#DEBUG,
+        reload=DEBUG,
         access_log=True,
-        workers=1, # Multiple workers will cause a crash due to the creation
+        workers=1, # Firewall module can't be replicated in multiple workers
-                  # of multiple processes with separated memory
+                   # Later the firewall module will be moved to a separate process
+                   # The webserver will communicate using redis (redis is also needed for websockets)
     )

backend/binsrc/classes/netfilter.cpp

@@ -1,244 +1,101 @@
-#include <linux/netfilter/nfnetlink_queue.h>
+#include <vector>
-#include <libnetfilter_queue/libnetfilter_queue.h>
-#include <linux/netfilter/nfnetlink_conntrack.h>
-#include <libmnl/libmnl.h>
-#include <linux/netfilter.h>
-#include <linux/netfilter/nfnetlink.h>
-#include <linux/types.h>
 #include <thread>
-#include <iostream>
+#include <type_traits>
-#include <functional>
+#include "../utils.cpp"
-#include <netinet/in.h>
+#include "nfqueue.cpp"
-
-using namespace std;
 
 #ifndef NETFILTER_CLASS_CPP
 #define NETFILTER_CLASS_CPP
 
-typedef int QueueCallbackFunction(const nlmsghdr *, const mnl_socket*, void *);
+namespace Firegex {
+namespace NfQueue {
 
-struct nfqueue_execution_data_tmp{
+template <typename Derived>
-    mnl_socket* nl = nullptr;
+class ThreadNfQueue {
-    function<QueueCallbackFunction> queue_cb = nullptr;
+public:
-    void *data = nullptr;
+    ThreadNfQueue() = default;
+    virtual ~ThreadNfQueue() = default;
+
+    std::thread thr;
+    BlockingQueue<PktRequest<Derived>*> queue;
+
+    virtual void before_loop() {}
+	virtual void handle_next_packet(PktRequest<Derived>* pkt){}
+    
+    void loop() {
+        static_cast<Derived*>(this)->before_loop();
+        for(;;) {
+            PktRequest<Derived>* pkt;
+            queue.take(pkt);
+            static_cast<Derived*>(this)->handle_next_packet(pkt);
+            delete pkt;
+        }
+    }
+
+    void run_thread_loop() {
+        thr = std::thread([this]() { this->loop(); });
+    }
 };
 
-class NfQueueExecutor {
+template <typename Worker, typename = is_base_of<ThreadNfQueue<Worker>, Worker>>
-    private:
+void __real_handler(PktRequest<std::vector<Worker>>* pkt) {
-    size_t BUF_SIZE = 0xffff + (MNL_SOCKET_BUFFER_SIZE/2);
+    const size_t idx = hash_stream_id(pkt->sid) % pkt->ctx->size();
-    char *queue_msg_buffer = nullptr;
-   	QueueCallbackFunction * _queue_callback_hook = nullptr;
-    public:
 
-    unsigned int portid;
+    auto* converted_pkt = reinterpret_cast<PktRequest<Worker>*>(pkt);
-    u_int16_t queue_num;
+    converted_pkt->ctx = &((*pkt->ctx)[idx]);
-    mnl_socket* nl = nullptr;
     
-	NfQueueExecutor(u_int16_t queue_num, QueueCallbackFunction* queue_cb): queue_num(queue_num), _queue_callback_hook(queue_cb){
+    converted_pkt->ctx->queue.put(converted_pkt);
-		nl = mnl_socket_open(NETLINK_NETFILTER);
+}
-		
-		if (nl == nullptr) { throw runtime_error( "mnl_socket_open" );}
-
-		if (mnl_socket_bind(nl, 0, MNL_SOCKET_AUTOPID) < 0) {
-			mnl_socket_close(nl);
-			throw runtime_error( "mnl_socket_bind" );
-		}
-		portid = mnl_socket_get_portid(nl);
-
-		queue_msg_buffer = (char*) malloc(BUF_SIZE);
-
-		if (!queue_msg_buffer) {
-			mnl_socket_close(nl);
-			throw runtime_error( "allocate receive buffer" );
-		}
-
-		if (_send_config_cmd(NFQNL_CFG_CMD_BIND) < 0) {
-			_clear();
-			throw runtime_error( "mnl_socket_send" );
-		}
-		//TEST if BIND was successful
-		if (_send_config_cmd(NFQNL_CFG_CMD_NONE) < 0) { // SEND A NONE cmmand to generate an error meessage
-			_clear();
-			throw runtime_error( "mnl_socket_send" );
-		}
-		if (_recv_packet() == -1) { //RECV the error message
-			_clear();
-			throw runtime_error( "mnl_socket_recvfrom" );
-		}
-
-		struct nlmsghdr *nlh = (struct nlmsghdr *) queue_msg_buffer;
-		
-		if (nlh->nlmsg_type != NLMSG_ERROR) {
-			_clear();
-			throw runtime_error( "unexpected packet from kernel (expected NLMSG_ERROR packet)" );
-		}		
-		//nfqnl_msg_config_cmd
-		nlmsgerr* error_msg = (nlmsgerr *)mnl_nlmsg_get_payload(nlh);	
-
-		// error code taken from the linux kernel:
-		// https://elixir.bootlin.com/linux/v5.18.12/source/include/linux/errno.h#L27
-		#define ENOTSUPP	524	/* Operation is not supported */
-
-		if (error_msg->error != -ENOTSUPP) {
-			_clear();
-			throw invalid_argument( "queueid is already busy" );
-		}
-		
-		//END TESTING BIND
-		nlh = nfq_nlmsg_put(queue_msg_buffer, NFQNL_MSG_CONFIG, queue_num);
-		nfq_nlmsg_cfg_put_params(nlh, NFQNL_COPY_PACKET, 0xffff);
-
-		mnl_attr_put_u32(nlh, NFQA_CFG_FLAGS, htonl(NFQA_CFG_F_GSO));
-		mnl_attr_put_u32(nlh, NFQA_CFG_MASK, htonl(NFQA_CFG_F_GSO));
-
-		if (mnl_socket_sendto(nl, nlh, nlh->nlmsg_len) < 0) {
-			_clear();
-			throw runtime_error( "mnl_socket_send" );
-		}
-
-	}
-
-    NfQueueExecutor(u_int16_t queue_num): NfQueueExecutor(queue_num, nullptr) {}
-
-    // --- Functions to be implemented by the user
-
-    virtual void before_loop() {
-        // Do nothing by default
-    }
-
-    virtual void * callback_data_fetch(){
-        return nullptr;
-    }
-
-    // --- End of functions to be implemented by the user
-
-	void run(){
-		/*
-		* ENOBUFS is signalled to userspace when packets were lost
-		* on kernel side.  In most cases, userspace isn't interested
-		* in this information, so turn it off.
-		*/
-		int ret = 1;
-		mnl_socket_setsockopt(nl, NETLINK_NO_ENOBUFS, &ret, sizeof(int));
-		
-        before_loop();
-
-		for (;;) {
-			ret = _recv_packet();
-			if (ret == -1) {
-				throw runtime_error( "mnl_socket_recvfrom" );
-			}
-			nfqueue_execution_data_tmp data = {
-                nl: nl,
-                queue_cb: _queue_callback_hook,
-                data: callback_data_fetch()
-            };
-			ret = mnl_cb_run(queue_msg_buffer, ret, 0, portid, _real_queue_cb, &data);
-			if (ret < 0){
-				throw runtime_error( "mnl_cb_run" );
-			}
-		}
-	}
 
 
-	~NfQueueExecutor() {
+template <typename Worker, typename = is_base_of<ThreadNfQueue<Worker>, Worker>>
-		_send_config_cmd(NFQNL_CFG_CMD_UNBIND);
+class MultiThreadQueue {
-		_clear();
+    static_assert(std::is_base_of_v<ThreadNfQueue<Worker>, Worker>,
-	}
+        "Worker must inherit from ThreadNfQueue<Worker>");
 
-    private:
+private:
-
+    std::vector<Worker> workers;
-    static int _real_queue_cb(const nlmsghdr *nlh, void *data_ptr) {
+    NfQueue<std::vector<Worker>, __real_handler<Worker>> * nfq;
-        nfqueue_execution_data_tmp* info = (nfqueue_execution_data_tmp*) data_ptr;
+    uint16_t queue_num_;
-        if (info->queue_cb == nullptr) return MNL_CB_OK;
-        return info->queue_cb(nlh, info->nl, info->data);
-    }
-
-	inline void _clear(){
-		if (queue_msg_buffer != nullptr) {
-			free(queue_msg_buffer);
-			queue_msg_buffer = nullptr;
-		}
-		mnl_socket_close(nl);
-		nl = nullptr;
-	}
-
-    inline ssize_t _send_config_cmd(nfqnl_msg_config_cmds cmd){
-		struct nlmsghdr *nlh = nfq_nlmsg_put(queue_msg_buffer, NFQNL_MSG_CONFIG, queue_num);
-		nfq_nlmsg_cfg_put_cmd(nlh, AF_INET, cmd);
-		return mnl_socket_sendto(nl, nlh, nlh->nlmsg_len);
-	}
-
-	inline ssize_t _recv_packet(){
-		return mnl_socket_recvfrom(nl, queue_msg_buffer, BUF_SIZE);
-	}	
-
-};
 	
     
-template <typename Executor, typename = enable_if_t<is_base_of_v<NfQueueExecutor, Executor>>>
+public:
-class NFQueueSequence{
+    const size_t n_threads;
+    static constexpr int QUEUE_BASE_NUM = 1000;
 
-	private:
+    explicit MultiThreadQueue(size_t n_threads) 
-		vector<Executor *> nfq;
+        : n_threads(n_threads), workers(n_threads) 
-		uint16_t _init;
+    {
-		uint16_t _end;
+        if(n_threads == 0) throw std::invalid_argument("At least 1 thread required");
-		vector<thread> threads;
-	public:
-		static const int QUEUE_BASE_NUM = 1000;
         
-		NFQueueSequence(uint16_t seq_len){
+        for(uint16_t qnum = QUEUE_BASE_NUM; ; qnum++) {
-			if (seq_len <= 0) throw invalid_argument("seq_len <= 0");
+            try {
-			nfq = vector<Executor*>(seq_len);
+                nfq = new NfQueue<std::vector<Worker>, __real_handler<Worker>>(qnum);
-			_init = QUEUE_BASE_NUM;
+                queue_num_ = qnum;
-			while(nfq[0] == nullptr){
-				if (_init+seq_len-1 >= 65536){
-					throw runtime_error("NFQueueSequence: too many queues!");
-				}
-				for (int i=0;i<seq_len;i++){
-					try{
-						nfq[i] = new Executor(_init+i);
-					}catch(const invalid_argument e){
-						for(int j = 0; j < i; j++) {
-							delete nfq[j];
-							nfq[j] = nullptr;
-						}
-						_init += seq_len - i;
                 break;
             }
+            catch(const std::invalid_argument&) {
+                if(qnum == std::numeric_limits<uint16_t>::max())
+                    throw std::runtime_error("No available queue numbers");
             }
         }
-			_end = _init + seq_len - 1;
-		}
-		
-		void start(){
-			if (threads.size() != 0) throw runtime_error("NFQueueSequence: already started!");
-			for (int i=0;i<nfq.size();i++){
-				threads.push_back(thread([executor = nfq[i]](){
-                    executor->run();
-                }));
-			}
     }
 
-		void join(){
+    ~MultiThreadQueue() {
-			for (int i=0;i<nfq.size();i++){
+        delete nfq;
-				threads[i].join();
-			}
-			threads.clear();
     }
 
-		uint16_t init(){
+    void start() {
-			return _init;
+        for(auto& worker : workers) {
+            worker.run_thread_loop();
+        }
+		for (;;){
+        	nfq->handle_next_packet(&workers);
 		}
-		uint16_t end(){
-			return _end;
     }
 
-		~NFQueueSequence(){
+    uint16_t queue_num() const { return queue_num_; }
-			for (int i=0;i<nfq.size();i++){
-				delete nfq[i];
-			}
-		}
 };
 
+}} // namespace Firegex::NfQueue
 #endif // NETFILTER_CLASS_CPP

backend/binsrc/classes/nfqueue.cpp

@@ -0,0 +1,370 @@
+
+#ifndef NFQUEUE_CLASS_CPP
+#define NFQUEUE_CLASS_CPP
+
+#include <libnetfilter_queue/libnetfilter_queue.h>
+#include <linux/netfilter/nfnetlink_queue.h>
+#include <tins/tcp_ip/stream_identifier.h>
+#include <libmnl/libmnl.h>
+#include <tins/tins.h>
+
+using namespace std;
+
+namespace Firegex{
+namespace NfQueue{
+
+enum class FilterAction{ DROP, ACCEPT, MANGLE, NOACTION };
+enum class L4Proto { TCP, UDP, RAW };
+typedef Tins::TCPIP::StreamIdentifier stream_id;
+
+
+template<typename T>
+class PktRequest {
+	private:
+	FilterAction action = FilterAction::NOACTION;
+	mnl_socket* nl = nullptr;
+	nfgenmsg * nfg = nullptr;
+	nfqnl_msg_packet_hdr *ph;
+	shared_ptr<char[]> packet_buffer; // Will be deallocated here
+	size_t data_size = 0;
+	public:
+	const bool is_ipv6;
+	Tins::IP* ipv4 = nullptr;
+	Tins::IPv6* ipv6 = nullptr;
+	Tins::TCP* tcp = nullptr;
+	Tins::UDP* udp = nullptr;
+	const L4Proto l4_proto;
+	const bool is_input;
+	
+	const string packet;
+	const string data;
+	const stream_id sid;
+
+	T* ctx;
+
+	private:
+
+	inline void fetch_data_size(Tins::PDU* pdu){
+		auto inner = pdu->inner_pdu();
+		if (inner == nullptr){
+			data_size = 0;
+		}else{
+			data_size = inner->size();
+		}
+	}
+
+	L4Proto fill_l4_info(){
+		if (is_ipv6){
+			tcp = ipv6->find_pdu<Tins::TCP>();
+			if (tcp == nullptr){
+				udp = ipv6->find_pdu<Tins::UDP>();
+				if (udp == nullptr){
+					fetch_data_size(ipv6);
+					return L4Proto::RAW;
+				}else{
+					fetch_data_size(udp);
+					return L4Proto::UDP;
+				}
+			}else{
+				fetch_data_size(tcp);
+				return L4Proto::TCP;
+			}
+		}else{
+			tcp = ipv4->find_pdu<Tins::TCP>();
+			if (tcp == nullptr){
+				udp = ipv4->find_pdu<Tins::UDP>();
+				if (udp == nullptr){
+					fetch_data_size(ipv4);
+					return L4Proto::RAW;
+				}else{
+					fetch_data_size(udp);
+					return L4Proto::UDP;
+				}
+			}else{
+				fetch_data_size(tcp);
+				return L4Proto::TCP;
+			}
+		}
+	}
+
+	public:
+
+	PktRequest(shared_ptr<char[]> buf, Tins::IP* ipv4, const char* payload, size_t plen, stream_id sid, T* ctx, mnl_socket* nl, nfgenmsg *nfg, nfqnl_msg_packet_hdr *ph, bool is_input):
+		is_ipv6(false), ipv4(ipv4), packet(string(payload, plen)), sid(sid), ctx(ctx), nl(nl), nfg(nfg), ph(ph),
+		is_input(is_input), packet_buffer(buf), l4_proto(fill_l4_info()), data(string(payload+(plen-data_size), data_size)) {}
+
+	PktRequest(shared_ptr<char[]> buf, Tins::IPv6* ipv6, const char* payload, size_t plen, stream_id sid, T* ctx, mnl_socket* nl, nfgenmsg *nfg, nfqnl_msg_packet_hdr *ph, bool is_input):
+		is_ipv6(true), ipv6(ipv6), packet(string(payload, plen)), sid(sid), ctx(ctx), nl(nl), nfg(nfg), ph(ph),
+		is_input(is_input), packet_buffer(buf), l4_proto(fill_l4_info()), data(string(payload+(plen-data_size), data_size)) {}
+
+	void drop(){
+		if (action == FilterAction::NOACTION){
+			action = FilterAction::DROP;
+			perfrom_action();
+		}else{
+			throw invalid_argument("Cannot drop a packet that has already been dropped or accepted");
+		}
+	}
+
+	void accept(){
+		if (action == FilterAction::NOACTION){
+			action = FilterAction::ACCEPT;
+			perfrom_action();
+		}else{
+			throw invalid_argument("Cannot accept a packet that has already been dropped or accepted");
+		}
+	}
+
+	void mangle(){
+		if (action == FilterAction::NOACTION){
+			action = FilterAction::MANGLE;
+			perfrom_action();
+		}else{
+			throw invalid_argument("Cannot mangle a packet that has already been accepted or dropped");
+		}
+	}
+
+	FilterAction get_action(){
+		return action;
+	}
+
+	~PktRequest(){
+		if (ipv4 != nullptr){
+            delete ipv4;
+        }
+        if (ipv6 != nullptr){
+            delete ipv6;
+        }
+	}
+
+	private:
+	void perfrom_action(){
+		char buf[MNL_SOCKET_BUFFER_SIZE];
+		struct nlmsghdr *nlh_verdict;
+		nlh_verdict = nfq_nlmsg_put(buf, NFQNL_MSG_VERDICT, ntohs(nfg->res_id));
+		switch (action)
+		{
+			case FilterAction::ACCEPT:
+				nfq_nlmsg_verdict_put(nlh_verdict, ntohl(ph->packet_id), NF_ACCEPT );
+				break;
+			case FilterAction::DROP:
+				nfq_nlmsg_verdict_put(nlh_verdict, ntohl(ph->packet_id), NF_DROP );
+				break;
+			case FilterAction::MANGLE:{
+				if (is_ipv6){
+					nfq_nlmsg_verdict_put_pkt(nlh_verdict, ipv6->serialize().data(), ipv6->size());
+				}else{
+					nfq_nlmsg_verdict_put_pkt(nlh_verdict, ipv4->serialize().data(), ipv4->size());
+				}
+				nfq_nlmsg_verdict_put(nlh_verdict, ntohl(ph->packet_id), NF_ACCEPT );
+				break;
+			}
+			default:
+				throw invalid_argument("Invalid action");
+		}
+		if (mnl_socket_sendto(nl, nlh_verdict, nlh_verdict->nlmsg_len) < 0) {
+			throw runtime_error( "mnl_socket_send" );
+		}
+	}
+
+};
+
+struct internal_nfqueue_execution_data_tmp{
+    mnl_socket* nl = nullptr;
+    void *data = nullptr;
+	shared_ptr<char[]> packet_buffer;
+};
+
+const size_t NFQUEUE_BUFFER_SIZE = 0xffff + (MNL_SOCKET_BUFFER_SIZE/2);
+/*  NfQueue wrapper class to handle nfqueue packets
+    this class is made to be possible enqueue multiple packets to multiple threads
+    --> handle function is responsable to delete the PktRequest object */
+template <typename D, void handle_func(PktRequest<D>*)>
+class NfQueue {
+    private:
+	mnl_socket* nl = nullptr;
+	unsigned int portid;
+    public:
+
+	const uint16_t queue_num;
+
+	NfQueue(u_int16_t queue_num): queue_num(queue_num) {
+		nl = mnl_socket_open(NETLINK_NETFILTER);
+		
+		if (nl == nullptr) { throw runtime_error( "mnl_socket_open" );}
+
+		if (mnl_socket_bind(nl, 0, MNL_SOCKET_AUTOPID) < 0) {
+			mnl_socket_close(nl);
+			throw runtime_error( "mnl_socket_bind" );
+		}
+		portid = mnl_socket_get_portid(nl);
+
+		char queue_msg_buffer[NFQUEUE_BUFFER_SIZE];
+
+		if (_send_config_cmd(NFQNL_CFG_CMD_BIND, queue_msg_buffer) < 0) {
+			_clear();
+			throw runtime_error( "mnl_socket_send" );
+		}
+		//TEST if BIND was successful
+		if (_send_config_cmd(NFQNL_CFG_CMD_NONE, queue_msg_buffer) < 0) { // SEND A NONE command to generate an error meessage
+			_clear();
+			throw runtime_error( "mnl_socket_send" );
+		}
+		if (_recv_packet(queue_msg_buffer) == -1) { //RECV the error message
+			_clear();
+			throw runtime_error( "mnl_socket_recvfrom" );
+		}
+
+		struct nlmsghdr *nlh = (struct nlmsghdr *) queue_msg_buffer;
+		
+		if (nlh->nlmsg_type != NLMSG_ERROR) {
+			_clear();
+			throw runtime_error( "unexpected packet from kernel (expected NLMSG_ERROR packet)" );
+		}		
+		//nfqnl_msg_config_cmd
+		nlmsgerr* error_msg = (nlmsgerr *)mnl_nlmsg_get_payload(nlh);	
+
+		// error code taken from the linux kernel:
+		// https://elixir.bootlin.com/linux/v5.18.12/source/include/linux/errno.h#L27
+		#define ENOTSUPP	524	/* Operation is not supported */
+
+		if (error_msg->error != -ENOTSUPP) {
+			_clear();
+			throw invalid_argument( "queueid is already busy" );
+		}
+		
+		//END TESTING BIND
+		nlh = nfq_nlmsg_put(queue_msg_buffer, NFQNL_MSG_CONFIG, queue_num);
+		nfq_nlmsg_cfg_put_params(nlh, NFQNL_COPY_PACKET, 0xffff);
+
+		mnl_attr_put_u32(nlh, NFQA_CFG_FLAGS, htonl(NFQA_CFG_F_GSO));
+		mnl_attr_put_u32(nlh, NFQA_CFG_MASK, htonl(NFQA_CFG_F_GSO));
+
+		if (mnl_socket_sendto(nl, nlh, nlh->nlmsg_len) < 0) {
+			_clear();
+			throw runtime_error( "mnl_socket_send" );
+		}
+
+		/*
+		* ENOBUFS is signalled to userspace when packets were lost
+		* on kernel side.  In most cases, userspace isn't interested
+		* in this information, so turn it off.
+		*/
+		int tmp = 1;
+		mnl_socket_setsockopt(nl, NETLINK_NO_ENOBUFS, &tmp, sizeof(int));
+
+	}
+
+	void handle_next_packet(D* data){
+		auto queue_msg_buffer = make_shared<char[]>(NFQUEUE_BUFFER_SIZE);
+		int ret = _recv_packet(queue_msg_buffer.get(), NFQUEUE_BUFFER_SIZE);
+		if (ret == -1) {
+			throw runtime_error( "mnl_socket_recvfrom" );
+		}
+		internal_nfqueue_execution_data_tmp raw_ptr = {
+			nl: nl,
+			data: data,
+			packet_buffer: queue_msg_buffer
+		};
+
+		ret = mnl_cb_run(queue_msg_buffer.get(), ret, 0, portid, _real_queue_cb, &raw_ptr);
+		if (ret <= 0){
+			cerr << "[error] [NfQueue.handle_next_packet] mnl_cb_run error with: " << ret << endl;
+			throw runtime_error( "mnl_cb_run error!" );
+		}
+	}
+	
+	~NfQueue() {
+		char queue_msg_buffer[NFQUEUE_BUFFER_SIZE];
+		_send_config_cmd(NFQNL_CFG_CMD_UNBIND, queue_msg_buffer);
+		_clear();
+	}
+    
+    private:
+
+	template<typename T, typename = enable_if_t<is_base_of_v<Tins::PDU, T>>>
+	static void inline _send_verdict(shared_ptr<char[]> raw_buf, T* packet, char *payload, uint16_t plen, nfgenmsg *nfg, nfqnl_msg_packet_hdr *ph, internal_nfqueue_execution_data_tmp* ctx, bool is_input){
+		handle_func(new PktRequest<D>(
+			raw_buf, packet, payload, plen,
+			stream_id::make_identifier(*packet),
+			(D*)ctx->data, ctx->nl, nfg, ph, is_input
+		));
+	}
+
+    static int _real_queue_cb(const nlmsghdr *nlh, void *data_ptr) {
+		
+        internal_nfqueue_execution_data_tmp* info = (internal_nfqueue_execution_data_tmp*) data_ptr;
+
+		//Extract attributes from the nlmsghdr
+		nlattr *attr[NFQA_MAX+1] = {};
+		
+		if (nfq_nlmsg_parse(nlh, attr) < 0) {
+			cerr << "[error] [NfQueue._real_queue_cb] problems parsing" << endl;
+			return MNL_CB_ERROR;
+		}
+		if (attr[NFQA_PACKET_HDR] == nullptr) {
+			cerr << "[error] [NfQueue._real_queue_cb] packet header not set" << endl;
+			return MNL_CB_ERROR;
+		}
+		if (attr[NFQA_MARK] == nullptr) {
+			cerr << "[error] [NfQueue._real_queue_cb] mark not set" << endl;
+			return MNL_CB_ERROR;
+		}
+		
+		//Get Payload
+		uint16_t plen = mnl_attr_get_payload_len(attr[NFQA_PAYLOAD]);
+		char *payload = (char *)mnl_attr_get_payload(attr[NFQA_PAYLOAD]);
+		
+		//Return result to the kernel
+		struct nfqnl_msg_packet_hdr *ph = (nfqnl_msg_packet_hdr*) mnl_attr_get_payload(attr[NFQA_PACKET_HDR]);
+		struct nfgenmsg *nfg = (nfgenmsg *)mnl_nlmsg_get_payload(nlh);
+
+		bool is_input = ntohl(mnl_attr_get_u32(attr[NFQA_MARK])) & 0x1; // == 0x1337 that is odd
+		// Check IP protocol version
+		if ( (payload[0] & 0xf0) == 0x40 ){
+			_send_verdict(info->packet_buffer, new Tins::IP((uint8_t*)payload, plen), payload, plen, nfg, ph, info, is_input);
+		}else{
+			_send_verdict(info->packet_buffer, new Tins::IPv6((uint8_t*)payload, plen), payload, plen, nfg, ph, info, is_input);
+		}
+		return MNL_CB_OK;
+    }
+
+	inline void _clear(){
+		if (nl != nullptr) {
+			mnl_socket_close(nl);
+			nl = nullptr;
+		}
+	}
+
+    inline ssize_t _send_config_cmd(nfqnl_msg_config_cmds cmd, char* buf){
+		struct nlmsghdr *nlh = nfq_nlmsg_put(buf, NFQNL_MSG_CONFIG, queue_num);
+		nfq_nlmsg_cfg_put_cmd(nlh, AF_INET, cmd);
+		return mnl_socket_sendto(nl, nlh, nlh->nlmsg_len);
+	}
+
+	inline ssize_t _recv_packet(char* buf, size_t buf_size = NFQUEUE_BUFFER_SIZE){
+		return mnl_socket_recvfrom(nl, buf, buf_size);
+	}	
+
+};
+
+
+
+uint32_t hash_stream_id(const stream_id &sid) {
+    uint32_t addr_hash = 0;
+    const uint32_t* min_addr = reinterpret_cast<const uint32_t*>(sid.min_address.data());
+    const uint32_t* max_addr = reinterpret_cast<const uint32_t*>(sid.max_address.data());
+    addr_hash ^= min_addr[0] ^ min_addr[1] ^ min_addr[2] ^ min_addr[3];
+    addr_hash ^= max_addr[0] ^ max_addr[1] ^ max_addr[2] ^ max_addr[3];
+
+    uint32_t ports = (static_cast<uint32_t>(sid.min_address_port) << 16) | sid.max_address_port;
+    
+    uint32_t hash = addr_hash ^ ports;
+    
+    hash *= 0x9e3779b9;
+    
+    return hash;
+}
+
+}}
+#endif // NFQUEUE_CLASS_CPP

backend/binsrc/nfproxy-tun.cpp

@@ -1,26 +1,40 @@
+#define PY_SSIZE_T_CLEAN
+#include <Python.h>
+
 #include "proxytun/settings.cpp"
 #include "proxytun/proxytun.cpp"
 #include "classes/netfilter.cpp"
 #include <syncstream>
 #include <iostream>
+#include <stdexcept>
+#include <cstdlib>
 
 using namespace std;
+using namespace Firegex::PyProxy;
+using Firegex::NfQueue::MultiThreadQueue;
+
+ssize_t read_check(int __fd, void *__buf, size_t __nbytes){
+	ssize_t bytes = read(__fd, __buf, __nbytes);
+	if (bytes == 0){
+		cerr << "[fatal] [updater] read() returned EOF" << endl;
+		throw invalid_argument("read() returned EOF");
+	}
+	if (bytes < 0){
+		cerr << "[fatal] [updater] read() returned an error" << bytes << endl;
+		throw invalid_argument("read() returned an error");
+	}
+	return bytes;
+}
 
 void config_updater (){
 	while (true){
-		//TODO read config getline(cin, line);
+		uint32_t code_size;
-		if (cin.eof()){
+		read_check(STDIN_FILENO, &code_size, 4);
-			cerr << "[fatal] [updater] cin.eof()" << endl;
+		vector<uint8_t> code(code_size);
-			exit(EXIT_FAILURE);
+		read_check(STDIN_FILENO, code.data(), code_size);
-		}
-		if (cin.bad()){
-			cerr << "[fatal] [updater] cin.bad()" << endl;
-			exit(EXIT_FAILURE);
-		}
 		cerr << "[info] [updater] Updating configuration" << endl;
-		
 		try{
-			//TODO add data config.reset(new PyCodeConfig(""));
+			config.reset(new PyCodeConfig(code));
 			cerr << "[info] [updater] Config update done" << endl;
 			osyncstream(cout) << "ACK OK" << endl;
 		}catch(const std::exception& e){
@@ -31,18 +45,28 @@ void config_updater (){
 }
 
 int main(int argc, char *argv[]){
+
+	Py_Initialize();
+	atexit(Py_Finalize);
+
+	if (freopen(nullptr, "rb", stdin) == nullptr){ // We need to read from stdin binary data
+		cerr << "[fatal] [main] Failed to reopen stdin in binary mode" << endl;
+		return 1;
+	}
 	int n_of_threads = 1;
    	char * n_threads_str = getenv("NTHREADS");
    	if (n_threads_str != nullptr) n_of_threads = ::atoi(n_threads_str);
 	if(n_of_threads <= 0) n_of_threads = 1;
 
-	config.reset(new PyCodeConfig(""));
+	config.reset(new PyCodeConfig());
+	MultiThreadQueue<PyProxyQueue> queue(n_of_threads);
 
-	NFQueueSequence<PyProxyQueue> queues(n_of_threads);
+	osyncstream(cout) << "QUEUE " << queue.queue_num() << endl;
-	queues.start();
+	cerr << "[info] [main] Queue: " << queue.queue_num() << " threads assigned: " << n_of_threads << endl;
-
-	osyncstream(cout) << "QUEUES " << queues.init() << " " << queues.end() << endl;
-	cerr << "[info] [main] Queues: " << queues.init() << ":" << queues.end() << " threads assigned: " << n_of_threads << endl;
 
+	thread qthr([&](){
+		queue.start();
+	});
 	config_updater();
+	qthr.join();
 }

backend/binsrc/nfregex.cpp

@@ -5,6 +5,8 @@
 #include <iostream>
 
 using namespace std;
+using namespace Firegex::Regex;
+using Firegex::NfQueue::MultiThreadQueue;
 
 void config_updater (){
 	string line;
@@ -55,11 +57,16 @@ int main(int argc, char *argv[]){
 
 	regex_config.reset(new RegexRules(stream_mode));
 
-	NFQueueSequence<RegexQueue> queues(n_of_threads);
-	queues.start();
 	
-	osyncstream(cout) << "QUEUES " << queues.init() << " " << queues.end() << endl;
+	MultiThreadQueue<RegexNfQueue> queue_manager(n_of_threads);
-	cerr << "[info] [main] Queues: " << queues.init() << ":" << queues.end() << " threads assigned: " << n_of_threads << " stream mode: " << stream_mode << endl;
 
+	osyncstream(cout) << "QUEUE " << queue_manager.queue_num() << endl;
+	cerr << "[info] [main] Queue: " << queue_manager.queue_num() << " threads assigned: " << n_of_threads << " stream mode: " << stream_mode << endl;
+
+	thread qthr([&](){
+		queue_manager.start();
+	});
 	config_updater();
+	qthr.join();
+
 }

backend/binsrc/proxytun/proxytun.cpp

@@ -13,7 +13,6 @@
 #include <linux/types.h>
 #include <stdexcept>
 #include <thread>
-#include <hs.h>
 #include <syncstream>
 #include <iostream>
 #include "../classes/netfilter.cpp"
@@ -24,24 +23,32 @@ using Tins::TCPIP::Stream;
 using Tins::TCPIP::StreamFollower;
 using namespace std;
 
-class PyProxyQueue: public NfQueueExecutor {
+namespace Firegex {
+namespace PyProxy {
+
+class PyProxyQueue: public NfQueue::ThreadNfQueue<PyProxyQueue> {
 	public:
 	stream_ctx sctx;
+	StreamFollower follower;
+
+	struct {
+		bool matching_has_been_called = false;
+		bool already_closed = false;
+		bool result;
+		NfQueue::PktRequest<PyProxyQueue>* pkt;
+	} match_ctx;
 
     void before_loop() override {
-		sctx.follower.new_stream_callback(bind(on_new_stream, placeholders::_1, &sctx));
+		follower.new_stream_callback(bind(on_new_stream, placeholders::_1, this));
-		sctx.follower.stream_termination_callback(bind(on_stream_close, placeholders::_1, &sctx));
+		follower.stream_termination_callback(bind(on_stream_close, placeholders::_1, this));
     }
 
-    void * callback_data_fetch() override{
+	bool filter_action(NfQueue::PktRequest<PyProxyQueue>* pkt){
-        return &sctx;
-    }
-
-	static bool filter_action(packet_info& info){
 		shared_ptr<PyCodeConfig> conf = config;
-		auto stream_search = info.sctx->streams_ctx.find(info.sid);
+
-		pyfilter_ctx stream_match;
+		auto stream_search = sctx.streams_ctx.find(pkt->sid);
-		if (stream_search == info.sctx->streams_ctx.end()){
+		pyfilter_ctx* stream_match;
+		if (stream_search == sctx.streams_ctx.end()){
 			// TODO: New pyfilter_ctx
 		}else{
 			stream_match = stream_search->second;
@@ -59,86 +66,76 @@ class PyProxyQueue: public NfQueueExecutor {
 	}
 
 	//If the stream has already been matched, drop all data, and try to close the connection
-	static void keep_fin_packet(stream_ctx* sctx){
+	static void keep_fin_packet(PyProxyQueue* pkt){
-		sctx->match_info.matching_has_been_called = true;
+		pkt->match_ctx.matching_has_been_called = true;
-		sctx->match_info.already_closed = true;
+		pkt->match_ctx.already_closed = true;
 	}
 	
-	static void on_data_recv(Stream& stream, stream_ctx* sctx, string data) {
+	static void on_data_recv(Stream& stream, PyProxyQueue* pkt, string data) {
-		sctx->match_info.matching_has_been_called = true;
+		pkt->match_ctx.matching_has_been_called = true;
-		sctx->match_info.already_closed = false;
+		pkt->match_ctx.already_closed = false;
-		bool result = filter_action(*sctx->match_info.pkt_info);
+		bool result = pkt->filter_action(pkt->match_ctx.pkt);
 		if (!result){
-			sctx->clean_stream_by_id(sctx->match_info.pkt_info->sid);
+			pkt->sctx.clean_stream_by_id(pkt->match_ctx.pkt->sid);
-			stream.client_data_callback(bind(keep_fin_packet, sctx));
+			stream.client_data_callback(bind(keep_fin_packet, pkt));
-			stream.server_data_callback(bind(keep_fin_packet, sctx));
+			stream.server_data_callback(bind(keep_fin_packet, pkt));
 		}
-		sctx->match_info.result = result;
+		pkt->match_ctx.result = result;
 	}
 	
 	//Input data filtering
-	static void on_client_data(Stream& stream, stream_ctx* sctx) {
+	static void on_client_data(Stream& stream, PyProxyQueue* pkt) {
-		sctx->match_info.pkt_info->is_input = true;
+		on_data_recv(stream, pkt, string(stream.client_payload().begin(), stream.client_payload().end()));
-		on_data_recv(stream, sctx, string(stream.client_payload().begin(), stream.client_payload().end()));
 	}
 	
 	//Server data filtering
-	static void on_server_data(Stream& stream, stream_ctx* sctx) {
+	static void on_server_data(Stream& stream, PyProxyQueue* pkt) {
-		sctx->match_info.pkt_info->is_input = false;
+		on_data_recv(stream, pkt, string(stream.server_payload().begin(), stream.server_payload().end()));
-		on_data_recv(stream, sctx, string(stream.server_payload().begin(), stream.server_payload().end()));
 	}
 	
 	// A stream was terminated. The second argument is the reason why it was terminated
-	static void on_stream_close(Stream& stream, stream_ctx* sctx) {
+	static void on_stream_close(Stream& stream, PyProxyQueue* pkt) {
 		stream_id stream_id = stream_id::make_identifier(stream);
-		sctx->clean_stream_by_id(stream_id);
+		pkt->sctx.clean_stream_by_id(stream_id);
 	}
 	
-	static void on_new_stream(Stream& stream, stream_ctx* sctx) {
+	static void on_new_stream(Stream& stream, PyProxyQueue* pkt) {
 		stream.auto_cleanup_payloads(true);
 		if (stream.is_partial_stream()) {
 			//TODO take a decision about this...
 			stream.enable_recovery_mode(10 * 1024);
 		}
-		stream.client_data_callback(bind(on_client_data, placeholders::_1, sctx));
+		stream.client_data_callback(bind(on_client_data, placeholders::_1, pkt));
-		stream.server_data_callback(bind(on_server_data, placeholders::_1, sctx));
+		stream.server_data_callback(bind(on_server_data, placeholders::_1, pkt));
-		stream.stream_closed_callback(bind(on_stream_close, placeholders::_1, sctx));
+		stream.stream_closed_callback(bind(on_stream_close, placeholders::_1, pkt));
 	}
 
-	template<typename T>
+
-	static void build_verdict(T packet, uint8_t *payload, uint16_t plen, nlmsghdr *nlh_verdict, nfqnl_msg_packet_hdr *ph, stream_ctx* sctx, bool is_ipv6){
+	void handle_next_packet(NfQueue::PktRequest<PyProxyQueue>* pkt) override{
-		Tins::TCP* tcp = packet.template find_pdu<Tins::TCP>();
+		if (pkt->l4_proto != NfQueue::L4Proto::TCP){
-		if (!tcp){
 			throw invalid_argument("Only TCP and UDP are supported");
 		}
-		Tins::PDU* application_layer = tcp->inner_pdu();
+		Tins::PDU* application_layer = pkt->tcp->inner_pdu();
 		u_int16_t payload_size = 0;
 		if (application_layer != nullptr){
 			payload_size = application_layer->size();
 		}
-		packet_info pktinfo{
+		match_ctx.matching_has_been_called = false;
-			payload: string(payload+plen - payload_size, payload+plen),
+		match_ctx.pkt = pkt;
-			sid: stream_id::make_identifier(packet),
+		if (pkt->is_ipv6){
-			is_ipv6: is_ipv6,
+			follower.process_packet(*pkt->ipv6);
-			sctx: sctx,
+		}else{
-			packet_pdu: &packet,
+			follower.process_packet(*pkt->ipv4);
-			tcp: tcp,
+		}
-		};
-		sctx->match_info.matching_has_been_called = false;
-		sctx->match_info.pkt_info = &pktinfo;
-		sctx->follower.process_packet(packet);
 		// Do an action only is an ordered packet has been received
-		if (sctx->match_info.matching_has_been_called){
+		if (match_ctx.matching_has_been_called){
-			bool empty_payload = pktinfo.payload.empty();
+			bool empty_payload = payload_size == 0;
 			//In this 2 cases we have to remove all data about the stream
-			if (!sctx->match_info.result || sctx->match_info.already_closed){
+			if (!match_ctx.result || match_ctx.already_closed){
-				#ifdef DEBUG
+				sctx.clean_stream_by_id(pkt->sid);
-					cerr << "[DEBUG] [NetfilterQueue.build_verdict] Stream matched, removing all data about it" << endl;
-				#endif
-				sctx->clean_stream_by_id(pktinfo.sid);
 				//If the packet has data, we have to remove it
 				if (!empty_payload){
-					Tins::PDU* data_layer = tcp->release_inner_pdu();
+					Tins::PDU* data_layer = pkt->tcp->release_inner_pdu();
 					if (data_layer != nullptr){
 						delete data_layer;
 					}
@@ -146,64 +143,23 @@ class PyProxyQueue: public NfQueueExecutor {
 				//For the first matched data or only for data packets, we set FIN bit
 				//This only for client packets, because this will trigger server to close the connection
 				//Packets will be filtered anyway also if client don't send packets
-				if ((!sctx->match_info.result || !empty_payload) && is_input){
+				if ((!match_ctx.result || !empty_payload) && pkt->is_input){
-					tcp->set_flag(Tins::TCP::FIN,1);
+					pkt->tcp->set_flag(Tins::TCP::FIN,1);
-					tcp->set_flag(Tins::TCP::ACK,1);
+					pkt->tcp->set_flag(Tins::TCP::ACK,1);
-					tcp->set_flag(Tins::TCP::SYN,0);
+					pkt->tcp->set_flag(Tins::TCP::SYN,0);
 				}
 				//Send the edited packet to the kernel
-				nfq_nlmsg_verdict_put_pkt(nlh_verdict, packet.serialize().data(), packet.size());
+				return pkt->mangle();
 			}
 		}
-		nfq_nlmsg_verdict_put(nlh_verdict, ntohl(ph->packet_id), NF_ACCEPT );
+		return pkt->accept();
-
 	}
 
-	static int queue_cb(const nlmsghdr *nlh, const mnl_socket* nl, void *data_ptr) {
-	
-		stream_ctx* sctx = (stream_ctx*)data_ptr;
-	
-		//Extract attributes from the nlmsghdr
-		nlattr *attr[NFQA_MAX+1] = {};
-		
-		if (nfq_nlmsg_parse(nlh, attr) < 0) {
-			perror("problems parsing");
-			return MNL_CB_ERROR;
-		}
-		if (attr[NFQA_PACKET_HDR] == nullptr) {
-			fputs("metaheader not set\n", stderr);
-			return MNL_CB_ERROR;
-		}
-		//Get Payload
-		uint16_t plen = mnl_attr_get_payload_len(attr[NFQA_PAYLOAD]);
-		uint8_t *payload = (uint8_t *)mnl_attr_get_payload(attr[NFQA_PAYLOAD]);
-		
-		//Return result to the kernel
-		struct nfqnl_msg_packet_hdr *ph = (nfqnl_msg_packet_hdr*) mnl_attr_get_payload(attr[NFQA_PACKET_HDR]);
-		struct nfgenmsg *nfg = (nfgenmsg *)mnl_nlmsg_get_payload(nlh);
-		char buf[MNL_SOCKET_BUFFER_SIZE];
-		struct nlmsghdr *nlh_verdict;
-	
-		nlh_verdict = nfq_nlmsg_put(buf, NFQNL_MSG_VERDICT, ntohs(nfg->res_id));
-		// Check IP protocol version
-		if ( (payload[0] & 0xf0) == 0x40 ){
-			build_verdict(Tins::IP(payload, plen), payload, plen, nlh_verdict, ph, sctx, false);
-		}else{
-			build_verdict(Tins::IPv6(payload, plen), payload, plen, nlh_verdict, ph, sctx, true);
-		}
-	
-		if (mnl_socket_sendto(nl, nlh_verdict, nlh_verdict->nlmsg_len) < 0) {
-			throw runtime_error( "mnl_socket_send" );
-		}
-		return MNL_CB_OK;
-	}
-
-	PyProxyQueue(int queue) : NfQueueExecutor(queue, &queue_cb) {}
-
 	~PyProxyQueue() {
 		sctx.clean();
 	}
 
 };
 
+}}
 #endif // PROXY_TUNNEL_CLASS_CPP

backend/binsrc/proxytun/settings.cpp

@@ -1,21 +1,17 @@
 #ifndef PROXY_TUNNEL_SETTINGS_CPP
 #define PROXY_TUNNEL_SETTINGS_CPP
 
-#include <iostream>
-#include <cstring>
-#include <sstream>
-#include "../utils.hpp"
 #include <vector>
-#include <hs.h>
 #include <memory>
 
 using namespace std;
 
 class PyCodeConfig{
 	public:
-		const string code;
+		const vector<uint8_t> code;
 	public:
-		PyCodeConfig(string pycode): code(pycode){}
+		PyCodeConfig(vector<uint8_t> pycode): code(pycode){}
+		PyCodeConfig(): code(vector<uint8_t>()){}
 		
 		~PyCodeConfig(){}
 };

backend/binsrc/proxytun/stream_ctx.cpp

@@ -3,11 +3,8 @@
 #define STREAM_CTX_CPP
 
 #include <iostream>
-#include <tins/tcp_ip/stream_follower.h>
 #include <tins/tcp_ip/stream_identifier.h>
 
-using Tins::TCPIP::Stream;
-using Tins::TCPIP::StreamFollower;
 using namespace std;
 
 typedef Tins::TCPIP::StreamIdentifier stream_id;
@@ -17,44 +14,25 @@ struct pyfilter_ctx {
 	string pycode;
 };
 
-typedef map<stream_id, pyfilter_ctx> matching_map;
+typedef map<stream_id, pyfilter_ctx*> matching_map;
-
-struct packet_info;
-
-struct tcp_stream_tmp {
-	bool matching_has_been_called = false;
-	bool already_closed = false;
-	bool result;
-	packet_info *pkt_info;
-};
 
 struct stream_ctx {
 	matching_map streams_ctx;
-	StreamFollower follower;
+
-	tcp_stream_tmp match_info;
 	void clean_stream_by_id(stream_id sid){
 		auto stream_search = streams_ctx.find(sid);
 		if (stream_search != streams_ctx.end()){
 			auto stream_match = stream_search->second;
 			//DEALLOC PY GLOB TODO
+			delete stream_match;
 		}
 	}
 	void clean(){
 		for (auto ele: streams_ctx){
 			//TODO dealloc ele.second.pyglob
+			delete ele.second;
 		}
 	}
 };
 
-struct packet_info {
-	string payload;
-	stream_id sid;
-	bool is_input;
-	bool is_ipv6;
-	stream_ctx* sctx;
-	Tins::PDU* packet_pdu;
-	Tins::TCP* tcp;
-};
-
-
 #endif // STREAM_CTX_CPP

backend/binsrc/regex/regex_rules.cpp

@@ -4,13 +4,16 @@
 #include <iostream>
 #include <cstring>
 #include <sstream>
-#include "../utils.hpp"
+#include "../utils.cpp"
 #include <vector>
 #include <hs.h>
 #include <memory>
 
 using namespace std;
 
+namespace Firegex {
+namespace Regex {
+
 enum FilterDirection{ CTOS, STOC };
 
 struct decoded_regex {
@@ -181,10 +184,6 @@ void inline scratch_setup(regex_ruleset &conf, hs_scratch_t* & scratch){
 	}
 }
 
-struct matched_data{
+}}
-	unsigned int matched = 0;
-	bool has_matched = false;
-};
-
 #endif // REGEX_FILTER_CPP
 

backend/binsrc/regex/regexfilter.cpp

@@ -15,53 +15,59 @@
 #include <thread>
 #include <hs.h>
 #include <syncstream>
+#include <functional>
 #include <iostream>
 #include "../classes/netfilter.cpp"
 #include "stream_ctx.cpp"
 #include "regex_rules.cpp"
 
-using Tins::TCPIP::Stream;
-using Tins::TCPIP::StreamFollower;
 using namespace std;
 
-class RegexQueue: public NfQueueExecutor {
+
-	public:
+namespace Firegex {
+namespace Regex {
+
+using Tins::TCPIP::Stream;
+using Tins::TCPIP::StreamFollower;
+
+
+
+class RegexNfQueue : public NfQueue::ThreadNfQueue<RegexNfQueue> {
+public:
 	stream_ctx sctx;
+	u_int16_t latest_config_ver = 0;
+	StreamFollower follower;
+	struct {
+		bool matching_has_been_called = false;
+		bool already_closed = false;
+		bool result;
+		NfQueue::PktRequest<RegexNfQueue>* pkt;
+	} match_ctx;
 	
-    void before_loop() override {
-		sctx.follower.new_stream_callback(bind(on_new_stream, placeholders::_1, &sctx));
-		sctx.follower.stream_termination_callback(bind(on_stream_close, placeholders::_1, &sctx));
-    }
 
-    void * callback_data_fetch() override{
+	bool filter_action(NfQueue::PktRequest<RegexNfQueue>* pkt){
-        return &sctx;
-    }
-
-	static bool filter_action(packet_info& info){
 		shared_ptr<RegexRules> conf = regex_config;
-		auto current_version = conf->ver();
-		if (current_version != info.sctx->latest_config_ver){
-			#ifdef DEBUG
-			cerr << "[DEBUG] [filter_callback] Configuration has changed (" << current_version << "!=" << info.sctx->latest_config_ver << "), cleaning scratch spaces" << endl;
-			#endif
-			info.sctx->clean();
-			info.sctx->latest_config_ver = current_version;
-		}
-		scratch_setup(conf->input_ruleset, info.sctx->in_scratch);
-		scratch_setup(conf->output_ruleset, info.sctx->out_scratch);
 
-		hs_database_t* regex_matcher = info.is_input ? conf->input_ruleset.hs_db : conf->output_ruleset.hs_db;
+		auto current_version = conf->ver();
+		if (current_version != latest_config_ver){
+			sctx.clean();
+			latest_config_ver = current_version;
+		}
+		scratch_setup(conf->input_ruleset, sctx.in_scratch);
+		scratch_setup(conf->output_ruleset, sctx.out_scratch);
+	
+		hs_database_t* regex_matcher = pkt->is_input ? conf->input_ruleset.hs_db : conf->output_ruleset.hs_db;
 		if (regex_matcher == nullptr){
 			return true;
 		}
 		
-		#ifdef DEBUG
+		struct matched_data{
-		cerr << "[DEBUG] [filter_callback] Matching packet with " << (info.is_input ? "input" : "output") << " ruleset" << endl;
+			unsigned int matched = 0;
-		#endif
+			bool has_matched = false;
+		} match_res;
 
-		matched_data match_res;
 		hs_error_t err;
-		hs_scratch_t* scratch_space = info.is_input ? info.sctx->in_scratch: info.sctx->out_scratch;
+		hs_scratch_t* scratch_space = pkt->is_input ? sctx.in_scratch: sctx.out_scratch;
 		auto match_func = [](unsigned int id, auto from, auto to, auto flags, auto ctx){
 			auto res = (matched_data*)ctx;
 			res->has_matched = true;
@@ -70,49 +76,32 @@ class RegexQueue: public NfQueueExecutor {
 		};
 		hs_stream_t* stream_match;
 		if (conf->stream_mode()){
-			matching_map* match_map = info.is_input ? &info.sctx->in_hs_streams : &info.sctx->out_hs_streams;
+			matching_map* match_map = pkt->is_input ? &sctx.in_hs_streams : &sctx.out_hs_streams;
-			#ifdef DEBUG
+			auto stream_search = match_map->find(pkt->sid);
-			cerr << "[DEBUG] [filter_callback] Dumping match_map " << match_map << endl;
-			for (auto ele: *match_map){
-				cerr << "[DEBUG] [filter_callback] " << ele.first << " -> " << ele.second << endl;
-			}
-			cerr << "[DEBUG] [filter_callback] End of match_map" << endl;
-			#endif
-			auto stream_search = match_map->find(info.sid);
 			
 			if (stream_search == match_map->end()){
-				
-				#ifdef DEBUG
-				cerr << "[DEBUG] [filter_callback] Creating new stream matcher for " << info.sid << endl;
-				#endif
 				if (hs_open_stream(regex_matcher, 0, &stream_match) != HS_SUCCESS) {
 					cerr << "[error] [filter_callback] Error opening the stream matcher (hs)" << endl;
 					throw invalid_argument("Cannot open stream match on hyperscan");
 				}
-				if (info.is_tcp){
+				if (pkt->l4_proto == NfQueue::L4Proto::TCP){
-					match_map->insert_or_assign(info.sid, stream_match);
+					match_map->insert_or_assign(pkt->sid, stream_match);
 				}
 			}else{
 				stream_match = stream_search->second;
 			}
-			#ifdef DEBUG
-			cerr << "[DEBUG] [filter_callback] Matching as a stream" << endl;
-			#endif
 			err = hs_scan_stream(
-				stream_match,info.payload.c_str(), info.payload.length(),
+				stream_match,pkt->data.c_str(), pkt->data.size(),
 				0, scratch_space, match_func, &match_res
 			);
 		}else{
-			#ifdef DEBUG
-			cerr << "[DEBUG] [filter_callback] Matching as a block" << endl;
-			#endif
 			err = hs_scan(
-				regex_matcher,info.payload.c_str(), info.payload.length(),
+				regex_matcher,pkt->data.c_str(), pkt->data.size(),
 				0, scratch_space, match_func, &match_res
 			);
 		}
 		if (
-			!info.is_tcp && conf->stream_mode() && 
+			pkt->l4_proto != NfQueue::L4Proto::TCP && conf->stream_mode() && 
 			hs_close_stream(stream_match, scratch_space, nullptr, nullptr) != HS_SUCCESS
 		){
 			cerr << "[error] [filter_callback] Error closing the stream matcher (hs)" << endl;
@@ -123,117 +112,34 @@ class RegexQueue: public NfQueueExecutor {
 			throw invalid_argument("Error while matching the stream with hyperscan");
 		}
 		if (match_res.has_matched){
-			auto rules_vector = info.is_input ? conf->input_ruleset.regexes : conf->output_ruleset.regexes;
+			auto& rules_vector = pkt->is_input ? conf->input_ruleset.regexes : conf->output_ruleset.regexes;
 			osyncstream(cout) << "BLOCKED " << rules_vector[match_res.matched] << endl;
 			return false;
 		}
 		return true;
 	}
 
-	//If the stream has already been matched, drop all data, and try to close the connection
+	void handle_next_packet(NfQueue::PktRequest<RegexNfQueue>* pkt) override{
-	static void keep_fin_packet(stream_ctx* sctx){
+		bool empty_payload = pkt->data.size() == 0;
-		sctx->match_info.matching_has_been_called = true;
+		if (pkt->tcp){
-		sctx->match_info.already_closed = true;
+			match_ctx.matching_has_been_called = false;
-	}
+			match_ctx.pkt = pkt;
 
-	static void on_data_recv(Stream& stream, stream_ctx* sctx, string data) {
+			if (pkt->ipv4){
-		sctx->match_info.matching_has_been_called = true;
+				follower.process_packet(*pkt->ipv4);
-		sctx->match_info.already_closed = false;
-		bool result = filter_action(*sctx->match_info.pkt_info);
-		#ifdef DEBUG
-			cerr << "[DEBUG] [NetfilterQueue.on_data_recv] result: " << result << endl;
-		#endif
-		if (!result){
-			#ifdef DEBUG
-				cerr << "[DEBUG] [NetfilterQueue.on_data_recv] Stream matched, removing all data about it" << endl;
-			#endif
-			sctx->clean_stream_by_id(sctx->match_info.pkt_info->sid);
-			stream.client_data_callback(bind(keep_fin_packet, sctx));
-			stream.server_data_callback(bind(keep_fin_packet, sctx));
-		}
-		sctx->match_info.result = result;
-	}
-	
-	//Input data filtering
-	static void on_client_data(Stream& stream, stream_ctx* sctx) {
-		on_data_recv(stream, sctx, string(stream.client_payload().begin(), stream.client_payload().end()));
-	}
-	
-	//Server data filtering
-	static void on_server_data(Stream& stream, stream_ctx* sctx) {
-		on_data_recv(stream, sctx, string(stream.server_payload().begin(), stream.server_payload().end()));
-	}
-	
-	// A stream was terminated. The second argument is the reason why it was terminated
-	static void on_stream_close(Stream& stream, stream_ctx* sctx) {
-		stream_id stream_id = stream_id::make_identifier(stream);
-		#ifdef DEBUG
-			cerr << "[DEBUG] [NetfilterQueue.on_stream_close] Stream terminated, deleting all data" << endl;
-		#endif
-		sctx->clean_stream_by_id(stream_id);
-	}
-	
-	static void on_new_stream(Stream& stream, stream_ctx* sctx) {
-		#ifdef DEBUG
-			cerr << "[DEBUG] [NetfilterQueue.on_new_stream] New stream detected" << endl;
-		#endif
-		stream.auto_cleanup_payloads(true);
-		if (stream.is_partial_stream()) {
-			#ifdef DEBUG
-				cerr << "[DEBUG] [NetfilterQueue.on_new_stream] Partial stream detected" << endl;
-			#endif
-			stream.enable_recovery_mode(10 * 1024);
-		}
-		stream.client_data_callback(bind(on_client_data, placeholders::_1, sctx));
-		stream.server_data_callback(bind(on_server_data, placeholders::_1, sctx));
-		stream.stream_closed_callback(bind(on_stream_close, placeholders::_1, sctx));
-	}
-
-	template<typename T>
-	static void build_verdict(T packet, uint8_t *payload, uint16_t plen, nlmsghdr *nlh_verdict, nfqnl_msg_packet_hdr *ph, stream_ctx* sctx, bool is_input, bool is_ipv6){
-		Tins::TCP* tcp = packet.template find_pdu<Tins::TCP>();
-	
-		if (tcp){
-			Tins::PDU* application_layer = tcp->inner_pdu();
-			u_int16_t payload_size = 0;
-			if (application_layer != nullptr){
-				payload_size = application_layer->size();
-			}
-			packet_info pktinfo{
-				payload: string(payload+plen - payload_size, payload+plen),
-				sid: stream_id::make_identifier(packet),
-				is_input: is_input,
-				is_tcp: true,
-				is_ipv6: is_ipv6,
-				sctx: sctx,
-				packet_pdu: &packet,
-				layer4_pdu: tcp,
-			};
-			sctx->match_info.matching_has_been_called = false;
-			sctx->match_info.pkt_info = &pktinfo;
-			#ifdef DEBUG
-				cerr << "[DEBUG] [NetfilterQueue.build_verdict] TCP Packet received " << packet.src_addr() << ":" << tcp->sport() << " -> " << packet.dst_addr() << ":" << tcp->dport() << " thr: " << this_thread::get_id() <<  ", sending to libtins StreamFollower" << endl;
-			#endif
-			sctx->follower.process_packet(packet);
-			#ifdef DEBUG
-			if (sctx->tcp_match_util.matching_has_been_called){
-				cerr << "[DEBUG] [NetfilterQueue.build_verdict] StreamFollower has called matching functions" << endl;
 			}else{
-				cerr << "[DEBUG] [NetfilterQueue.build_verdict] StreamFollower has NOT called matching functions" << endl;
+				follower.process_packet(*pkt->ipv6);
 			}
-			#endif
+	
 			// Do an action only is an ordered packet has been received
-			if (sctx->match_info.matching_has_been_called){
+			if (match_ctx.matching_has_been_called){
-				bool empty_payload = pktinfo.payload.empty();
+	
 				//In this 2 cases we have to remove all data about the stream
-				if (!sctx->match_info.result || sctx->match_info.already_closed){
+				if (!match_ctx.result || match_ctx.already_closed){
-					#ifdef DEBUG
+					sctx.clean_stream_by_id(pkt->sid);
-						cerr << "[DEBUG] [NetfilterQueue.build_verdict] Stream matched, removing all data about it" << endl;
-					#endif
-					sctx->clean_stream_by_id(pktinfo.sid);
 					//If the packet has data, we have to remove it
 					if (!empty_payload){
-						Tins::PDU* data_layer = tcp->release_inner_pdu();
+						Tins::PDU* data_layer = pkt->tcp->release_inner_pdu();
 						if (data_layer != nullptr){
 							delete data_layer;
 						}
@@ -241,106 +147,83 @@ class RegexQueue: public NfQueueExecutor {
 					//For the first matched data or only for data packets, we set FIN bit
 					//This only for client packets, because this will trigger server to close the connection
 					//Packets will be filtered anyway also if client don't send packets
-					if ((!sctx->match_info.result || !empty_payload) && is_input){
+					if ((!match_ctx.result || !empty_payload) && pkt->is_input){
-						tcp->set_flag(Tins::TCP::FIN,1);
+						pkt->tcp->set_flag(Tins::TCP::FIN,1);
-						tcp->set_flag(Tins::TCP::ACK,1);
+						pkt->tcp->set_flag(Tins::TCP::ACK,1);
-						tcp->set_flag(Tins::TCP::SYN,0);
+						pkt->tcp->set_flag(Tins::TCP::SYN,0);
 					}
 					//Send the edited packet to the kernel
-					nfq_nlmsg_verdict_put_pkt(nlh_verdict, packet.serialize().data(), packet.size());
+					return pkt->mangle();
 				}
 			}
-			nfq_nlmsg_verdict_put(nlh_verdict, ntohl(ph->packet_id), NF_ACCEPT );
+			return pkt->accept();
 		}else{
-			Tins::UDP* udp = packet.template find_pdu<Tins::UDP>();
+			if (!pkt->udp){
-			if (!udp){
 				throw invalid_argument("Only TCP and UDP are supported");
 			}
-			Tins::PDU* application_layer = udp->inner_pdu();
+			if(empty_payload){
-			u_int16_t payload_size = 0;
+				return pkt->accept();
-			if (application_layer != nullptr){
+			}else if (filter_action(pkt)){
-				payload_size = application_layer->size();
+				return pkt->accept();
-			}
-			if((udp->inner_pdu() == nullptr)){
-				nfq_nlmsg_verdict_put(nlh_verdict, ntohl(ph->packet_id), NF_ACCEPT );
-			}
-			packet_info pktinfo{
-				payload: string(payload+plen - payload_size, payload+plen),
-				sid: stream_id::make_identifier(packet),
-				is_input: is_input,
-				is_tcp: false,
-				is_ipv6: is_ipv6,
-				sctx: sctx,
-				packet_pdu: &packet,
-				layer4_pdu: udp,
-			};
-			if (filter_action(pktinfo)){
-				nfq_nlmsg_verdict_put(nlh_verdict, ntohl(ph->packet_id), NF_ACCEPT );
 			}else{
-				nfq_nlmsg_verdict_put(nlh_verdict, ntohl(ph->packet_id), NF_DROP );
+				return pkt->drop();
 			}
 		}
 	}
-	
+	//If the stream has already been matched, drop all data, and try to close the connection
-	static int queue_cb(const nlmsghdr *nlh, const mnl_socket* nl, void *data_ptr) {
+	static void keep_fin_packet(RegexNfQueue* nfq){
-	
+		nfq->match_ctx.matching_has_been_called = true;
-		stream_ctx* sctx = (stream_ctx*)data_ptr;
+		nfq->match_ctx.already_closed = true;
-	
-		//Extract attributes from the nlmsghdr
-		nlattr *attr[NFQA_MAX+1] = {};
-		
-		if (nfq_nlmsg_parse(nlh, attr) < 0) {
-			perror("problems parsing");
-			return MNL_CB_ERROR;
-		}
-		if (attr[NFQA_PACKET_HDR] == nullptr) {
-			fputs("metaheader not set\n", stderr);
-			return MNL_CB_ERROR;
-		}
-		if (attr[NFQA_MARK] == nullptr) {
-			fputs("mark not set\n", stderr);
-			return MNL_CB_ERROR;
-		}
-		//Get Payload
-		uint16_t plen = mnl_attr_get_payload_len(attr[NFQA_PAYLOAD]);
-		uint8_t *payload = (uint8_t *)mnl_attr_get_payload(attr[NFQA_PAYLOAD]);
-		
-		//Return result to the kernel
-		struct nfqnl_msg_packet_hdr *ph = (nfqnl_msg_packet_hdr*) mnl_attr_get_payload(attr[NFQA_PACKET_HDR]);
-		struct nfgenmsg *nfg = (nfgenmsg *)mnl_nlmsg_get_payload(nlh);
-		char buf[MNL_SOCKET_BUFFER_SIZE];
-		struct nlmsghdr *nlh_verdict;
-	
-		nlh_verdict = nfq_nlmsg_put(buf, NFQNL_MSG_VERDICT, ntohs(nfg->res_id));
-	
-		bool is_input = ntohl(mnl_attr_get_u32(attr[NFQA_MARK])) & 0x1; // == 0x1337 that is odd
-		#ifdef DEBUG
-			cerr << "[DEBUG] [NetfilterQueue.queue_cb] Packet received" << endl;
-			cerr << "[DEBUG] [NetfilterQueue.queue_cb] Packet ID: " << ntohl(ph->packet_id) << endl;
-			cerr << "[DEBUG] [NetfilterQueue.queue_cb] Payload size: " << plen << endl;
-			cerr << "[DEBUG] [NetfilterQueue.queue_cb] Is input: " << is_input << endl;
-		#endif
-		
-		// Check IP protocol version
-		if ( (payload[0] & 0xf0) == 0x40 ){
-			build_verdict(Tins::IP(payload, plen), payload, plen, nlh_verdict, ph, sctx, is_input, false);
-		}else{
-			build_verdict(Tins::IPv6(payload, plen), payload, plen, nlh_verdict, ph, sctx, is_input, true);
 	}
 
-		if (mnl_socket_sendto(nl, nlh_verdict, nlh_verdict->nlmsg_len) < 0) {
+	static void on_data_recv(Stream& stream, RegexNfQueue* nfq, string data) {
-			throw runtime_error( "mnl_socket_send" );
+		nfq->match_ctx.matching_has_been_called = true;
+		nfq->match_ctx.already_closed = false;
+		bool result = nfq->filter_action(nfq->match_ctx.pkt);
+		if (!result){
+			nfq->sctx.clean_stream_by_id(nfq->match_ctx.pkt->sid);
+			stream.client_data_callback(bind(keep_fin_packet, nfq));
+			stream.server_data_callback(bind(keep_fin_packet, nfq));
+		}
+		nfq->match_ctx.result = result;
 	}
 
-		return MNL_CB_OK;
+	//Input data filtering
+	static void on_client_data(Stream& stream, RegexNfQueue* nfq) {
+		on_data_recv(stream, nfq, string(stream.client_payload().begin(), stream.client_payload().end()));
 	}
 
-	RegexQueue(int queue) : NfQueueExecutor(queue, &queue_cb) {}
+	//Server data filtering
+	static void on_server_data(Stream& stream, RegexNfQueue* nfq) {
+		on_data_recv(stream, nfq, string(stream.server_payload().begin(), stream.server_payload().end()));
+	}
 
-	~RegexQueue() {
+	// A stream was terminated. The second argument is the reason why it was terminated
+	static void on_stream_close(Stream& stream, RegexNfQueue* nfq) {
+		stream_id stream_id = stream_id::make_identifier(stream);
+		nfq->sctx.clean_stream_by_id(stream_id);
+	}
+
+	static void on_new_stream(Stream& stream, RegexNfQueue* nfq) {
+		stream.auto_cleanup_payloads(true);
+		if (stream.is_partial_stream()) {
+			stream.enable_recovery_mode(10 * 1024);
+		}
+		stream.client_data_callback(bind(on_client_data, placeholders::_1, nfq));
+		stream.server_data_callback(bind(on_server_data, placeholders::_1, nfq));
+		stream.stream_closed_callback(bind(on_stream_close, placeholders::_1, nfq));
+	}
+
+	void before_loop() override{
+		follower.new_stream_callback(bind(on_new_stream, placeholders::_1, this));
+		follower.stream_termination_callback(bind(on_stream_close, placeholders::_1, this));
+	}
+
+	~RegexNfQueue(){
 		sctx.clean();
 	}
 
 };
 
+}}
 #endif // REGEX_FILTER_CLASS_CPP

backend/binsrc/regex/stream_ctx.cpp

@@ -4,30 +4,19 @@
 
 #include <iostream>
 #include <hs.h>
-#include <tins/tcp_ip/stream_follower.h>
 #include <tins/tcp_ip/stream_identifier.h>
+#include <functional>
+#include <map>
+#include "regexfilter.cpp"
 
-using Tins::TCPIP::Stream;
-using Tins::TCPIP::StreamFollower;
 using namespace std;
 
+namespace Firegex {
+namespace Regex {
+
 typedef Tins::TCPIP::StreamIdentifier stream_id;
 typedef map<stream_id, hs_stream_t*> matching_map;
 
-/* Considering to use unorder_map using this hash of stream_id 
-
-namespace std {
-	template<>
-	struct hash<stream_id> {
-		size_t operator()(const stream_id& sid) const
-		{
-			return std::hash<std::uint32_t>()(sid.max_address[0] + sid.max_address[1] + sid.max_address[2] + sid.max_address[3] + sid.max_address_port + sid.min_address[0] + sid.min_address[1] + sid.min_address[2] + sid.min_address[3] + sid.min_address_port);
-		}
-	};
-}
-
-*/
-
 #ifdef DEBUG
 ostream& operator<<(ostream& os, const Tins::TCPIP::StreamIdentifier::address_type &sid){
 	bool first_print = false;
@@ -46,24 +35,11 @@ ostream& operator<<(ostream& os, const stream_id &sid){
 }
 #endif
 
-
-struct packet_info;
-
-struct tcp_stream_tmp {
-	bool matching_has_been_called = false;
-	bool already_closed = false;
-	bool result;
-	packet_info *pkt_info;
-};
-
 struct stream_ctx {
 	matching_map in_hs_streams;
 	matching_map out_hs_streams;
 	hs_scratch_t* in_scratch = nullptr;
 	hs_scratch_t* out_scratch = nullptr;
-	u_int16_t latest_config_ver = 0;
-	StreamFollower follower;
-	tcp_stream_tmp match_info;
 
 	void clean_scratches(){
 		if (out_scratch != nullptr){
@@ -77,9 +53,6 @@ struct stream_ctx {
 	}
 
 	void clean_stream_by_id(stream_id sid){
-		#ifdef DEBUG
-		cerr << "[DEBUG] [NetfilterQueue.clean_stream_by_id] Cleaning stream context of " << sid << endl;
-		#endif
 		auto stream_search = in_hs_streams.find(sid);
 		hs_stream_t* stream_match;
 		if (stream_search != in_hs_streams.end()){
@@ -103,11 +76,6 @@ struct stream_ctx {
 	}
 
 	void clean(){
-
-		#ifdef DEBUG
-		cerr << "[DEBUG] [NetfilterQueue.clean] Cleaning stream context" << endl;
-		#endif
-
 		if (in_scratch){
 			for(auto ele: in_hs_streams){
 				if (hs_close_stream(ele.second, in_scratch, nullptr, nullptr) != HS_SUCCESS) {
@@ -131,16 +99,5 @@ struct stream_ctx {
 	}
 };
 
-struct packet_info {
+}}
-	string payload;
-	stream_id sid;
-	bool is_input;
-	bool is_tcp;
-	bool is_ipv6;
-	stream_ctx* sctx;
-	Tins::PDU* packet_pdu;
-	Tins::PDU* layer4_pdu;
-};
-
-
 #endif // STREAM_CTX_CPP

backend/binsrc/utils.cpp

@@ -0,0 +1,64 @@
+#include <string>
+#include <unistd.h>
+#include <queue>
+#include <condition_variable>
+
+#ifndef UTILS_CPP
+#define UTILS_CPP
+
+bool unhexlify(std::string const &hex, std::string &newString) {
+   try{
+      int len = hex.length();
+      for(int i=0; i< len; i+=2)
+      {
+         std::string byte = hex.substr(i,2);
+         char chr = (char) (int)strtol(byte.c_str(), nullptr, 16);
+         newString.push_back(chr);
+      }
+      return true;
+   }
+   catch (...){
+      return false;
+   }
+}
+
+template<typename T, int MAX = 1024> //same of kernel nfqueue max
+class BlockingQueue
+{
+private:
+    std::mutex mut;
+    std::queue<T> private_std_queue;
+    std::condition_variable condNotEmpty;
+    std::condition_variable condNotFull;
+    size_t count; // Guard with Mutex
+public:
+    void put(T new_value)
+    {
+      
+        std::unique_lock<std::mutex> lk(mut);
+        //Condition takes a unique_lock and waits given the false condition
+        condNotFull.wait(lk,[this]{
+            if (count == MAX) {
+               return false;
+            }else{
+               return true;
+            }
+            
+        });
+        private_std_queue.push(new_value);
+        count++;
+        condNotEmpty.notify_one();
+    }
+    void take(T& value)
+    {
+        std::unique_lock<std::mutex> lk(mut);
+         //Condition takes a unique_lock and waits given the false condition
+        condNotEmpty.wait(lk,[this]{return !private_std_queue.empty();});
+        value=private_std_queue.front();
+        private_std_queue.pop();
+        count--;
+        condNotFull.notify_one();
+    }
+};
+
+#endif // UTILS_CPP

backend/binsrc/utils.hpp

@@ -1,23 +0,0 @@
-#include <string>
-#include <unistd.h>
-
-#ifndef UTILS_HPP
-#define UTILS_HPP
-
-bool unhexlify(std::string const &hex, std::string &newString) {
-   try{
-      int len = hex.length();
-      for(int i=0; i< len; i+=2)
-      {
-         std::string byte = hex.substr(i,2);
-         char chr = (char) (int)strtol(byte.c_str(), nullptr, 16);
-         newString.push_back(chr);
-      }
-      return true;
-   }
-   catch (...){
-      return false;
-   }
-}
-
-#endif

backend/modules/nfproxy/firegex.py

@@ -57,9 +57,9 @@ class FiregexInterceptor:
             self.process.kill()
             raise Exception("Invalid binary output")
         line = line_fut.decode()
-        if line.startswith("QUEUES "):
+        if line.startswith("QUEUE "):
             params = line.split()
-            return (int(params[1]), int(params[2]))
+            return (int(params[1]), int(params[1]))
         else:
             self.process.kill()
             raise Exception("Invalid binary output")

backend/modules/nfproxy/nftables.py

@@ -62,6 +62,7 @@ class FiregexTables(NFTableManager):
                 "expr": [
                         {'match': {'left': {'payload': {'protocol': ip_family(srv.ip_int), 'field': 'saddr'}}, 'op': '==', 'right': nftables_int_to_json(srv.ip_int)}},
                         {'match': {"left": { "payload": {"protocol": str(srv.proto), "field": "sport"}}, "op": "==", "right": int(srv.port)}},
+                        {"mangle": {"key": {"meta": {"key": "mark"}},"value": 0x1338}},
                         {"queue": {"num": str(init) if init == end else {"range":[init, end] }, "flags": ["bypass"]}}
                 ]
             }}},
@@ -72,6 +73,7 @@ class FiregexTables(NFTableManager):
                 "expr": [
                         {'match': {'left': {'payload': {'protocol': ip_family(srv.ip_int), 'field': 'daddr'}}, 'op': '==', 'right': nftables_int_to_json(srv.ip_int)}},
                         {'match': {"left": { "payload": {"protocol": str(srv.proto), "field": "dport"}}, "op": "==", "right": int(srv.port)}},
+                        {"mangle": {"key": {"meta": {"key": "mark"}},"value": 0x1337}},
                         {"queue": {"num": str(init) if init == end else {"range":[init, end] }, "flags": ["bypass"]}}
                     ]
             }}}

backend/modules/nfregex/firegex.py

@@ -84,7 +84,7 @@ class FiregexInterceptor:
         return self
     
     async def _start_binary(self):
-        proxy_binary_path = os.path.join(os.path.dirname(os.path.abspath(__file__)),"../cppqueue")
+        proxy_binary_path = os.path.join(os.path.dirname(os.path.abspath(__file__)),"../cppregex")
         self.process = await asyncio.create_subprocess_exec(
             proxy_binary_path,
             stdout=asyncio.subprocess.PIPE, stdin=asyncio.subprocess.PIPE,
@@ -97,9 +97,9 @@ class FiregexInterceptor:
             self.process.kill()
             raise Exception("Invalid binary output")
         line = line_fut.decode()
-        if line.startswith("QUEUES "):
+        if line.startswith("QUEUE "):
             params = line.split()
-            return (int(params[1]), int(params[2]))
+            return (int(params[1]), int(params[1]))
         else:
             self.process.kill()
             raise Exception("Invalid binary output")

backend/modules/nfregex/nftables.py

@@ -48,10 +48,12 @@ class FiregexTables(NFTableManager):
     def add(self, srv:Service, queue_range):
         
         for ele in self.get():
-            if ele.__eq__(srv): return
+            if ele.__eq__(srv):
+                return
                 
         init, end = queue_range
-        if init > end: init, end = end, init
+        if init > end:
+            init, end = end, init
         self.cmd(
             { "insert":{ "rule": {
                 "family": "inet",

tests/api_test.py

@@ -15,7 +15,7 @@ puts(f"{args.address}", color=colors.yellow)
 firegex = FiregexAPI(args.address)
 
 #Connect to Firegex
-if firegex.status()["status"] =="init":
+if firegex.status()["status"] == "init":
     if (firegex.set_password(args.password)):
         puts(f"Sucessfully set password to {args.password} ✔", color=colors.green)
     else: