Merge #16202: p2p: Refactor network message deserialization

ed2dc5e48abed1cde6ab98025dc8212917d47d21 Add override/final modifiers to V1TransportDeserializer (Pieter Wuille)
f342a5e61a73e1edf389b662d265d20cf26a1d51 Make resetting implicit in TransportDeserializer::Read() (Pieter Wuille)
6a91499496d76c2b3e84489e9723b60514fb08db Remove oversized message detection from log and interface (Pieter Wuille)
b0e10ff4df3d4c70fb172ea8c3128c82e6e368bb Force CNetMessage::m_recv to use std::move (Jonas Schnelli)
efecb74677222f6c70adf7f860c315f430d39ec4 Use adapter pattern for the network deserializer (Jonas Schnelli)
1a5c656c3169ba525f84145d19ce8c64f2cf1efb Remove transport protocol knowhow from CNetMessage / net processing (Jonas Schnelli)
6294ecdb8bb4eb7049a18c721ee8cb4a53d80a06 Refactor: split network transport deserializing from message container (Jonas Schnelli)

Pull request description:

  **This refactors the network message deserialization.**

  * It transforms the `CNetMessage` into a transport protocol agnostic message container.
  * A new class `TransportDeserializer` (unique pointer of `CNode`)  is introduced, handling the network buffer reading and the decomposing to a `CNetMessage`
  * **No behavioral changes** (in terms of disconnecting, punishing)
  * Moves the checksum finalizing into the `SocketHandler` thread (finalizing was in `ProcessMessages` before)

  The **optional last commit** makes the `TransportDeserializer` following an adapter pattern (polymorphic interface) to make it easier to later add a V2 transport protocol deserializer.

  Intentionally not touching the sending part.

  Pre-Requirement for BIP324 (v2 message transport protocol).
  Replacement for #14046 and inspired by a [comment](https://github.com/bitcoin/bitcoin/pull/14046#issuecomment-431528330) from sipa

ACKs for top commit:
  promag:
    Code review ACK ed2dc5e48abed1cde6ab98025dc8212917d47d21.
  marcinja:
    Code review ACK ed2dc5e48abed1cde6ab98025dc8212917d47d21
  ryanofsky:
    Code review ACK ed2dc5e48abed1cde6ab98025dc8212917d47d21. 4 cleanup commits added since last review. Unaddressed comments:
  ariard:
    Code review and tested ACK ed2dc5e.

Tree-SHA512: bab8d87464e2e8742529e488ddcdc8650f0c2025c9130913df00a0b17ecdb9a525061cbbbd0de0251b76bf75a8edb72e3ad0dbf5b79e26f2ad05d61b4e4ded6d

3 files changed, 133 insertions, 70 deletions

diff --git a/src/net.cpp b/src/net.cpp
index c0b39925c6..7ae88b47a0 100644
--- a/src/net.cpp
+++ b/src/net.cpp
@@ -567,42 +567,28 @@ bool CNode::ReceiveMsgBytes(const char *pch, unsigned int nBytes, bool& complete
     nLastRecv = nTimeMicros / 1000000;
     nRecvBytes += nBytes;
     while (nBytes > 0) {
-
-        // get current incomplete message, or create a new one
-        if (vRecvMsg.empty() ||
-            vRecvMsg.back().complete())
-            vRecvMsg.push_back(CNetMessage(Params().MessageStart(), SER_NETWORK, INIT_PROTO_VERSION));
-
-        CNetMessage& msg = vRecvMsg.back();
-
         // absorb network data
-        int handled;
-        if (!msg.in_data)
-            handled = msg.readHeader(pch, nBytes);
-        else
-            handled = msg.readData(pch, nBytes);
-
-        if (handled < 0)
-            return false;
-
-        if (msg.in_data && msg.hdr.nMessageSize > MAX_PROTOCOL_MESSAGE_LENGTH) {
-            LogPrint(BCLog::NET, "Oversized message from peer=%i, disconnecting\n", GetId());
-            return false;
-        }
+        int handled = m_deserializer->Read(pch, nBytes);
+        if (handled < 0) return false;
 
         pch += handled;
         nBytes -= handled;
 
-        if (msg.complete()) {
+        if (m_deserializer->Complete()) {
+            // decompose a transport agnostic CNetMessage from the deserializer
+            CNetMessage msg = m_deserializer->GetMessage(Params().MessageStart(), nTimeMicros);
+
             //store received bytes per message command
             //to prevent a memory DOS, only allow valid commands
-            mapMsgCmdSize::iterator i = mapRecvBytesPerMsgCmd.find(msg.hdr.pchCommand);
+            mapMsgCmdSize::iterator i = mapRecvBytesPerMsgCmd.find(msg.m_command);
             if (i == mapRecvBytesPerMsgCmd.end())
                 i = mapRecvBytesPerMsgCmd.find(NET_MESSAGE_COMMAND_OTHER);
             assert(i != mapRecvBytesPerMsgCmd.end());
-            i->second += msg.hdr.nMessageSize + CMessageHeader::HEADER_SIZE;
+            i->second += msg.m_raw_message_size;
+
+            // push the message to the process queue,
+            vRecvMsg.push_back(std::move(msg));
 
-            msg.nTime = nTimeMicros;
             complete = true;
         }
     }
@@ -636,8 +622,7 @@ int CNode::GetSendVersion() const
     return nSendVersion;
 }
 
-
-int CNetMessage::readHeader(const char *pch, unsigned int nBytes)
+int V1TransportDeserializer::readHeader(const char *pch, unsigned int nBytes)
 {
     // copy data to temporary parsing buffer
     unsigned int nRemaining = 24 - nHdrPos;
@@ -658,9 +643,10 @@ int CNetMessage::readHeader(const char *pch, unsigned int nBytes)
         return -1;
     }
 
-    // reject messages larger than MAX_SIZE
-    if (hdr.nMessageSize > MAX_SIZE)
+    // reject messages larger than MAX_SIZE or MAX_PROTOCOL_MESSAGE_LENGTH
+    if (hdr.nMessageSize > MAX_SIZE || hdr.nMessageSize > MAX_PROTOCOL_MESSAGE_LENGTH) {
         return -1;
+    }
 
     // switch state to reading message data
     in_data = true;
@@ -668,7 +654,7 @@ int CNetMessage::readHeader(const char *pch, unsigned int nBytes)
     return nCopy;
 }
 
-int CNetMessage::readData(const char *pch, unsigned int nBytes)
+int V1TransportDeserializer::readData(const char *pch, unsigned int nBytes)
 {
     unsigned int nRemaining = hdr.nMessageSize - nDataPos;
     unsigned int nCopy = std::min(nRemaining, nBytes);
@@ -685,14 +671,44 @@ int CNetMessage::readData(const char *pch, unsigned int nBytes)
     return nCopy;
 }
 
-const uint256& CNetMessage::GetMessageHash() const
+const uint256& V1TransportDeserializer::GetMessageHash() const
 {
-    assert(complete());
+    assert(Complete());
     if (data_hash.IsNull())
         hasher.Finalize(data_hash.begin());
     return data_hash;
 }
 
+CNetMessage V1TransportDeserializer::GetMessage(const CMessageHeader::MessageStartChars& message_start, int64_t time) {
+    // decompose a single CNetMessage from the TransportDeserializer
+    CNetMessage msg(std::move(vRecv));
+
+    // store state about valid header, netmagic and checksum
+    msg.m_valid_header = hdr.IsValid(message_start);
+    msg.m_valid_netmagic = (memcmp(hdr.pchMessageStart, message_start, CMessageHeader::MESSAGE_START_SIZE) == 0);
+    uint256 hash = GetMessageHash();
+
+    // store command string, payload size
+    msg.m_command = hdr.GetCommand();
+    msg.m_message_size = hdr.nMessageSize;
+    msg.m_raw_message_size = hdr.nMessageSize + CMessageHeader::HEADER_SIZE;
+
+    msg.m_valid_checksum = (memcmp(hash.begin(), hdr.pchChecksum, CMessageHeader::CHECKSUM_SIZE) == 0);
+    if (!msg.m_valid_checksum) {
+        LogPrint(BCLog::NET, "CHECKSUM ERROR (%s, %u bytes), expected %s was %s\n",
+                 SanitizeString(msg.m_command), msg.m_message_size,
+                 HexStr(hash.begin(), hash.begin()+CMessageHeader::CHECKSUM_SIZE),
+                 HexStr(hdr.pchChecksum, hdr.pchChecksum+CMessageHeader::CHECKSUM_SIZE));
+    }
+
+    // store receive time
+    msg.m_time = time;
+
+    // reset the network deserializer (prepare for the next message)
+    Reset();
+    return msg;
+}
+
 size_t CConnman::SocketSendData(CNode *pnode) const EXCLUSIVE_LOCKS_REQUIRED(pnode->cs_vSend)
 {
     auto it = pnode->vSendMsg.begin();
@@ -1344,9 +1360,9 @@ void CConnman::SocketHandler()
                     size_t nSizeAdded = 0;
                     auto it(pnode->vRecvMsg.begin());
                     for (; it != pnode->vRecvMsg.end(); ++it) {
-                        if (!it->complete())
-                            break;
-                        nSizeAdded += it->vRecv.size() + CMessageHeader::HEADER_SIZE;
+                        // vRecvMsg contains only completed CNetMessage
+                        // the single possible partially deserialized message are held by TransportDeserializer
+                        nSizeAdded += it->m_raw_message_size;
                     }
                     {
                         LOCK(pnode->cs_vProcessMsg);
@@ -2676,6 +2692,8 @@ CNode::CNode(NodeId idIn, ServiceFlags nLocalServicesIn, int nMyStartingHeightIn
     } else {
         LogPrint(BCLog::NET, "Added connection peer=%d\n", id);
     }
+
+    m_deserializer = MakeUnique<V1TransportDeserializer>(V1TransportDeserializer(Params().MessageStart(), SER_NETWORK, INIT_PROTO_VERSION));
 }
 
 CNode::~CNode()
diff --git a/src/net.h b/src/net.h
index 44655abf80..af73361c08 100644
--- a/src/net.h
+++ b/src/net.h
@@ -609,56 +609,105 @@ public:
 
 
 
-
+/** Transport protocol agnostic message container.
+ * Ideally it should only contain receive time, payload,
+ * command and size.
+ */
 class CNetMessage {
+public:
+    CDataStream m_recv;                  // received message data
+    int64_t m_time = 0;                  // time (in microseconds) of message receipt.
+    bool m_valid_netmagic = false;
+    bool m_valid_header = false;
+    bool m_valid_checksum = false;
+    uint32_t m_message_size = 0;         // size of the payload
+    uint32_t m_raw_message_size = 0;     // used wire size of the message (including header/checksum)
+    std::string m_command;
+
+    CNetMessage(CDataStream&& recv_in) : m_recv(std::move(recv_in)) {}
+
+    void SetVersion(int nVersionIn)
+    {
+        m_recv.SetVersion(nVersionIn);
+    }
+};
+
+/** The TransportDeserializer takes care of holding and deserializing the
+ * network receive buffer. It can deserialize the network buffer into a
+ * transport protocol agnostic CNetMessage (command & payload)
+ */
+class TransportDeserializer {
+public:
+    // returns true if the current deserialization is complete
+    virtual bool Complete() const = 0;
+    // set the serialization context version
+    virtual void SetVersion(int version) = 0;
+    // read and deserialize data
+    virtual int Read(const char *data, unsigned int bytes) = 0;
+    // decomposes a message from the context
+    virtual CNetMessage GetMessage(const CMessageHeader::MessageStartChars& message_start, int64_t time) = 0;
+    virtual ~TransportDeserializer() {}
+};
+
+class V1TransportDeserializer final : public TransportDeserializer
+{
 private:
     mutable CHash256 hasher;
     mutable uint256 data_hash;
-public:
     bool in_data;                   // parsing header (false) or data (true)
-
     CDataStream hdrbuf;             // partially received header
     CMessageHeader hdr;             // complete header
-    unsigned int nHdrPos;
-
     CDataStream vRecv;              // received message data
+    unsigned int nHdrPos;
     unsigned int nDataPos;
 
-    int64_t nTime;                  // time (in microseconds) of message receipt.
+    const uint256& GetMessageHash() const;
+    int readHeader(const char *pch, unsigned int nBytes);
+    int readData(const char *pch, unsigned int nBytes);
 
-    CNetMessage(const CMessageHeader::MessageStartChars& pchMessageStartIn, int nTypeIn, int nVersionIn) : hdrbuf(nTypeIn, nVersionIn), hdr(pchMessageStartIn), vRecv(nTypeIn, nVersionIn) {
+    void Reset() {
+        vRecv.clear();
+        hdrbuf.clear();
         hdrbuf.resize(24);
         in_data = false;
         nHdrPos = 0;
         nDataPos = 0;
-        nTime = 0;
+        data_hash.SetNull();
+        hasher.Reset();
     }
 
-    bool complete() const
+public:
+
+    V1TransportDeserializer(const CMessageHeader::MessageStartChars& pchMessageStartIn, int nTypeIn, int nVersionIn) : hdrbuf(nTypeIn, nVersionIn), hdr(pchMessageStartIn), vRecv(nTypeIn, nVersionIn) {
+        Reset();
+    }
+
+    bool Complete() const override
     {
         if (!in_data)
             return false;
         return (hdr.nMessageSize == nDataPos);
     }
-
-    const uint256& GetMessageHash() const;
-
-    void SetVersion(int nVersionIn)
+    void SetVersion(int nVersionIn) override
     {
         hdrbuf.SetVersion(nVersionIn);
         vRecv.SetVersion(nVersionIn);
     }
-
-    int readHeader(const char *pch, unsigned int nBytes);
-    int readData(const char *pch, unsigned int nBytes);
+    int Read(const char *pch, unsigned int nBytes) override {
+        int ret = in_data ? readData(pch, nBytes) : readHeader(pch, nBytes);
+        if (ret < 0) Reset();
+        return ret;
+    }
+    CNetMessage GetMessage(const CMessageHeader::MessageStartChars& message_start, int64_t time) override;
 };
 
-
 /** Information about a peer */
 class CNode
 {
     friend class CConnman;
 public:
+    std::unique_ptr<TransportDeserializer> m_deserializer;
+
     // socket
     std::atomic<ServiceFlags> nServices{NODE_NONE};
     SOCKET hSocket GUARDED_BY(cs_hSocket);
diff --git a/src/net_processing.cpp b/src/net_processing.cpp
index 4a3076e3c7..fd31c962c2 100644
--- a/src/net_processing.cpp
+++ b/src/net_processing.cpp
@@ -3272,41 +3272,37 @@ bool PeerLogicValidation::ProcessMessages(CNode* pfrom, std::atomic<bool>& inter
             return false;
         // Just take one message
         msgs.splice(msgs.begin(), pfrom->vProcessMsg, pfrom->vProcessMsg.begin());
-        pfrom->nProcessQueueSize -= msgs.front().vRecv.size() + CMessageHeader::HEADER_SIZE;
+        pfrom->nProcessQueueSize -= msgs.front().m_raw_message_size;
         pfrom->fPauseRecv = pfrom->nProcessQueueSize > connman->GetReceiveFloodSize();
         fMoreWork = !pfrom->vProcessMsg.empty();
     }
     CNetMessage& msg(msgs.front());
 
     msg.SetVersion(pfrom->GetRecvVersion());
-    // Scan for message start
-    if (memcmp(msg.hdr.pchMessageStart, chainparams.MessageStart(), CMessageHeader::MESSAGE_START_SIZE) != 0) {
-        LogPrint(BCLog::NET, "PROCESSMESSAGE: INVALID MESSAGESTART %s peer=%d\n", SanitizeString(msg.hdr.GetCommand()), pfrom->GetId());
+    // Check network magic
+    if (!msg.m_valid_netmagic) {
+        LogPrint(BCLog::NET, "PROCESSMESSAGE: INVALID MESSAGESTART %s peer=%d\n", SanitizeString(msg.m_command), pfrom->GetId());
         pfrom->fDisconnect = true;
         return false;
     }
 
-    // Read header
-    CMessageHeader& hdr = msg.hdr;
-    if (!hdr.IsValid(chainparams.MessageStart()))
+    // Check header
+    if (!msg.m_valid_header)
     {
-        LogPrint(BCLog::NET, "PROCESSMESSAGE: ERRORS IN HEADER %s peer=%d\n", SanitizeString(hdr.GetCommand()), pfrom->GetId());
+        LogPrint(BCLog::NET, "PROCESSMESSAGE: ERRORS IN HEADER %s peer=%d\n", SanitizeString(msg.m_command), pfrom->GetId());
         return fMoreWork;
     }
-    std::string strCommand = hdr.GetCommand();
+    const std::string& strCommand = msg.m_command;
 
     // Message size
-    unsigned int nMessageSize = hdr.nMessageSize;
+    unsigned int nMessageSize = msg.m_message_size;
 
     // Checksum
-    CDataStream& vRecv = msg.vRecv;
-    const uint256& hash = msg.GetMessageHash();
-    if (memcmp(hash.begin(), hdr.pchChecksum, CMessageHeader::CHECKSUM_SIZE) != 0)
+    CDataStream& vRecv = msg.m_recv;
+    if (!msg.m_valid_checksum)
     {
-        LogPrint(BCLog::NET, "%s(%s, %u bytes): CHECKSUM ERROR expected %s was %s\n", __func__,
-           SanitizeString(strCommand), nMessageSize,
-           HexStr(hash.begin(), hash.begin()+CMessageHeader::CHECKSUM_SIZE),
-           HexStr(hdr.pchChecksum, hdr.pchChecksum+CMessageHeader::CHECKSUM_SIZE));
+        LogPrint(BCLog::NET, "%s(%s, %u bytes): CHECKSUM ERROR peer=%d\n", __func__,
+           SanitizeString(strCommand), nMessageSize, pfrom->GetId());
         return fMoreWork;
     }
 
@@ -3314,7 +3310,7 @@ bool PeerLogicValidation::ProcessMessages(CNode* pfrom, std::atomic<bool>& inter
     bool fRet = false;
     try
     {
-        fRet = ProcessMessage(pfrom, strCommand, vRecv, msg.nTime, chainparams, connman, interruptMsgProc);
+        fRet = ProcessMessage(pfrom, strCommand, vRecv, msg.m_time, chainparams, connman, interruptMsgProc);
         if (interruptMsgProc)
             return false;
         if (!pfrom->vRecvGetData.empty())