Merge bitcoin/bitcoin#20516: Well-defined CAddress disk serialization, and addrv2 anchors.dat

f8866e8c324be3322fa507c2ceb1de35d148d0f1 Add roundtrip fuzz tests for CAddress serialization (Pieter Wuille)
e2f0548b52a4b2ba3edf77e3f21365f1e8f270a4 Use addrv2 serialization in anchors.dat (Pieter Wuille)
8cd8f37dfe3ffb73a09f3ad773603d9d89452245 Introduce well-defined CAddress disk serialization (Pieter Wuille)

Pull request description:

  Alternative to #20509.

  This makes the `CAddress` disk serialization format well defined, and uses it to enable addrv2 support in anchors.dat (in a way that's compatible with older software). The new format is:
  - The first 4 bytes store a format version number. Its low 19 bits are ignored (as those historically stored the `CLIENT_VERSION`), but its high 13 bits specify the actual serialization:
    - 0x00000000: LE64 encoding for `nServices`, V1 encoding for `CService` (like pre-BIP155 network serialization).
    - 0x20000000: CompactSize encoding for `nServices`, V2 encoding for `CService` (like BIP155 network serialization).
    - Any other value triggers an unsupported format error on deserialization, and can be used for future format changes.
  - The `ADDRV2_FORMAT` flag in the stream's version does not determine the actual serialization format; it only sets whether or not V2 encoding is permitted.

ACKs for top commit:
  achow101:
    ACK f8866e8c324be3322fa507c2ceb1de35d148d0f1
  laanwj:
    Code review ACK f8866e8c324be3322fa507c2ceb1de35d148d0f1
  vasild:
    ACK f8866e8c324be3322fa507c2ceb1de35d148d0f1
  jonatack:
    ACK f8866e8c324be3322fa507c2ceb1de35d148d0f1 tested rebased to master and built/run/restarted with DEBUG_ADDRMAN, peers.dat and anchors ser/deser seems fine
  hebasto:
    ACK f8866e8c324be3322fa507c2ceb1de35d148d0f1, tested on Linux Mint 20.1 (x86_64).

Tree-SHA512: 3898f8a8c51783a46dd0aae03fa10060521f5dd6e79315fe95ba807689e78f202388ffa28c40bf156c6f7b1fc2ce806b155dcbe56027df73d039a55331723796

1 files changed, 36 insertions, 8 deletions

diff --git a/src/test/fuzz/deserialize.cpp b/src/test/fuzz/deserialize.cpp
index 1290c78712..decfc2610c 100644
--- a/src/test/fuzz/deserialize.cpp
+++ b/src/test/fuzz/deserialize.cpp
@@ -53,9 +53,9 @@ struct invalid_fuzzing_input_exception : public std::exception {
 };
 
 template <typename T>
-CDataStream Serialize(const T& obj, const int version = INIT_PROTO_VERSION)
+CDataStream Serialize(const T& obj, const int version = INIT_PROTO_VERSION, const int ser_type = SER_NETWORK)
 {
-    CDataStream ds(SER_NETWORK, version);
+    CDataStream ds(ser_type, version);
     ds << obj;
     return ds;
 }
@@ -69,9 +69,9 @@ T Deserialize(CDataStream ds)
 }
 
 template <typename T>
-void DeserializeFromFuzzingInput(FuzzBufferType buffer, T& obj, const std::optional<int> protocol_version = std::nullopt)
+void DeserializeFromFuzzingInput(FuzzBufferType buffer, T& obj, const std::optional<int> protocol_version = std::nullopt, const int ser_type = SER_NETWORK)
 {
-    CDataStream ds(buffer, SER_NETWORK, INIT_PROTO_VERSION);
+    CDataStream ds(buffer, ser_type, INIT_PROTO_VERSION);
     if (protocol_version) {
         ds.SetVersion(*protocol_version);
     } else {
@@ -92,9 +92,9 @@ void DeserializeFromFuzzingInput(FuzzBufferType buffer, T& obj, const std::optio
 }
 
 template <typename T>
-void AssertEqualAfterSerializeDeserialize(const T& obj, const int version = INIT_PROTO_VERSION)
+void AssertEqualAfterSerializeDeserialize(const T& obj, const int version = INIT_PROTO_VERSION, const int ser_type = SER_NETWORK)
 {
-    assert(Deserialize<T>(Serialize(obj, version)) == obj);
+    assert(Deserialize<T>(Serialize(obj, version, ser_type)) == obj);
 }
 
 } // namespace
@@ -251,9 +251,37 @@ FUZZ_TARGET_DESERIALIZE(messageheader_deserialize, {
         DeserializeFromFuzzingInput(buffer, mh);
         (void)mh.IsCommandValid();
 })
-FUZZ_TARGET_DESERIALIZE(address_deserialize, {
+FUZZ_TARGET_DESERIALIZE(address_deserialize_v1_notime, {
         CAddress a;
-        DeserializeFromFuzzingInput(buffer, a);
+        DeserializeFromFuzzingInput(buffer, a, INIT_PROTO_VERSION);
+        // A CAddress without nTime (as is expected under INIT_PROTO_VERSION) will roundtrip
+        // in all 5 formats (with/without nTime, v1/v2, network/disk)
+        AssertEqualAfterSerializeDeserialize(a, INIT_PROTO_VERSION);
+        AssertEqualAfterSerializeDeserialize(a, PROTOCOL_VERSION);
+        AssertEqualAfterSerializeDeserialize(a, 0, SER_DISK);
+        AssertEqualAfterSerializeDeserialize(a, PROTOCOL_VERSION | ADDRV2_FORMAT);
+        AssertEqualAfterSerializeDeserialize(a, ADDRV2_FORMAT, SER_DISK);
+})
+FUZZ_TARGET_DESERIALIZE(address_deserialize_v1_withtime, {
+        CAddress a;
+        DeserializeFromFuzzingInput(buffer, a, PROTOCOL_VERSION);
+        // A CAddress in V1 mode will roundtrip in all 4 formats that have nTime.
+        AssertEqualAfterSerializeDeserialize(a, PROTOCOL_VERSION);
+        AssertEqualAfterSerializeDeserialize(a, 0, SER_DISK);
+        AssertEqualAfterSerializeDeserialize(a, PROTOCOL_VERSION | ADDRV2_FORMAT);
+        AssertEqualAfterSerializeDeserialize(a, ADDRV2_FORMAT, SER_DISK);
+})
+FUZZ_TARGET_DESERIALIZE(address_deserialize_v2, {
+        CAddress a;
+        DeserializeFromFuzzingInput(buffer, a, PROTOCOL_VERSION | ADDRV2_FORMAT);
+        // A CAddress in V2 mode will roundtrip in both V2 formats, and also in the V1 formats
+        // with time if it's V1 compatible.
+        if (a.IsAddrV1Compatible()) {
+            AssertEqualAfterSerializeDeserialize(a, PROTOCOL_VERSION);
+            AssertEqualAfterSerializeDeserialize(a, 0, SER_DISK);
+        }
+        AssertEqualAfterSerializeDeserialize(a, PROTOCOL_VERSION | ADDRV2_FORMAT);
+        AssertEqualAfterSerializeDeserialize(a, ADDRV2_FORMAT, SER_DISK);
 })
 FUZZ_TARGET_DESERIALIZE(inv_deserialize, {
         CInv i;