Fix excessive memory use when flushing chainstate and EvoDB (#3008)

* Specialize Serialize(CSizeComputer&) for BLS objects

Speeds up size computation for BLS objects.

* Track memory usage in CDBTransaction and CEvoDB

* Take memory used by CEvoDB/CDBTransaction into account when flushing

* Implement specialized SerReadWrite for immer maps

This allows to use READWRITE and fixes compilation errors with CSizeComputer

* Log EvoDB memory usage independently

Co-Authored-By: UdjinM6 <UdjinM6@users.noreply.github.com>

src/bls/bls.h

@@ -170,6 +170,11 @@ public:
     }
 
 public:
+    inline void Serialize(CSizeComputer& s) const
+    {
+        s.seek(SerSize);
+    }
+
     template <typename Stream>
     inline void Serialize(Stream& s) const
     {
@@ -356,6 +361,11 @@ public:
         return *this;
     }
 
+    inline void Serialize(CSizeComputer& s) const
+    {
+        s.seek(BLSObject::SerSize);
+    }
+
     template<typename Stream>
     inline void Serialize(Stream& s) const
     {

src/dbwrapper.h

@@ -550,6 +550,7 @@ class CDBTransaction {
 protected:
     Parent &parent;
     CommitTarget &commitTarget;
+    ssize_t memoryUsage{0}; // signed, just in case we made an error in the calculations so that we don't get an overflow
 
     struct DataStreamCmp {
         static bool less(const CDataStream& a, const CDataStream& b) {
@@ -563,6 +564,8 @@ protected:
     };
 
     struct ValueHolder {
+        size_t memoryUsage;
+        ValueHolder(size_t _memoryUsage) : memoryUsage(_memoryUsage) {}
         virtual ~ValueHolder() = default;
         virtual void Write(const CDataStream& ssKey, CommitTarget &parent) = 0;
     };
@@ -570,7 +573,7 @@ protected:
 
     template <typename V>
     struct ValueHolderImpl : ValueHolder {
-        ValueHolderImpl(const V &_value) : value(_value) { }
+        ValueHolderImpl(const V &_value, size_t _memoryUsage) : ValueHolder(_memoryUsage), value(_value) {}
 
         virtual void Write(const CDataStream& ssKey, CommitTarget &commitTarget) {
             // we're moving the value instead of copying it. This means that Write() can only be called once per
@@ -604,9 +607,18 @@ public:
 
     template <typename V>
     void Write(const CDataStream& ssKey, const V& v) {
-        deletes.erase(ssKey);
+        auto valueMemoryUsage = ::GetSerializeSize(v, SER_DISK, CLIENT_VERSION);
+
+        if (deletes.erase(ssKey)) {
+            memoryUsage -= ssKey.size();
+        }
         auto it = writes.emplace(ssKey, nullptr).first;
-        it->second = std::make_unique<ValueHolderImpl<V>>(v);
+        if (it->second) {
+            memoryUsage -= ssKey.size() + it->second->memoryUsage;
+        }
+        it->second = std::make_unique<ValueHolderImpl<V>>(v, valueMemoryUsage);
+
+        memoryUsage += ssKey.size() + valueMemoryUsage;
     }
 
     template <typename K, typename V>
@@ -656,13 +668,20 @@ public:
     }
 
     void Erase(const CDataStream& ssKey) {
-        writes.erase(ssKey);
-        deletes.emplace(ssKey);
+        auto it = writes.find(ssKey);
+        if (it != writes.end()) {
+            memoryUsage -= ssKey.size() + it->second->memoryUsage;
+            writes.erase(it);
+        }
+        if (deletes.emplace(ssKey).second) {
+            memoryUsage += ssKey.size();
+        }
     }
 
     void Clear() {
         writes.clear();
         deletes.clear();
+        memoryUsage = 0;
     }
 
     void Commit() {
@@ -679,6 +698,19 @@ public:
         return writes.empty() && deletes.empty();
     }
 
+    size_t GetMemoryUsage() const {
+        if (memoryUsage < 0) {
+            // something went wrong when we accounted/calculated used memory...
+            static volatile bool didPrint = false;
+            if (!didPrint) {
+                LogPrintf("CDBTransaction::%s -- negative memoryUsage (%d)", __func__, memoryUsage);
+                didPrint = true;
+            }
+            return 0;
+        }
+        return (size_t)memoryUsage;
+    }
+
     CDBTransactionIterator<CDBTransaction>* NewIterator() {
         return new CDBTransactionIterator<CDBTransaction>(*this);
     }

src/evo/deterministicmns.h

@@ -195,6 +195,21 @@ void UnserializeImmerMap(Stream& is, immer::map<K, T, Hash, Equal>& m)
     }
 }
 
+// For some reason the compiler is not able to choose the correct Serialize/Deserialize methods without a specialized
+// version of SerReadWrite. It otherwise always chooses the version that calls a.Serialize()
+template<typename Stream, typename K, typename T, typename Hash, typename Equal>
+inline void SerReadWrite(Stream& s, const immer::map<K, T, Hash, Equal>& m, CSerActionSerialize ser_action)
+{
+    ::SerializeImmerMap(s, m);
+}
+
+template<typename Stream, typename K, typename T, typename Hash, typename Equal>
+inline void SerReadWrite(Stream& s, immer::map<K, T, Hash, Equal>& obj, CSerActionUnserialize ser_action)
+{
+    ::UnserializeImmerMap(s, obj);
+}
+
+
 class CDeterministicMNList
 {
 public:
@@ -226,13 +241,8 @@ public:
     {
         READWRITE(blockHash);
         READWRITE(nHeight);
-        if (ser_action.ForRead()) {
-            UnserializeImmerMap(s, mnMap);
-            UnserializeImmerMap(s, mnUniquePropertyMap);
-        } else {
-            SerializeImmerMap(s, mnMap);
-            SerializeImmerMap(s, mnUniquePropertyMap);
-        }
+        READWRITE(mnMap);
+        READWRITE(mnUniquePropertyMap);
     }
 
 public:

src/evo/evodb.h

@@ -73,6 +73,11 @@ public:
         return db;
     }
 
+    size_t GetMemoryUsage()
+    {
+        return rootDBTransaction.GetMemoryUsage();
+    }
+
     bool CommitRootTransaction();
 
     bool VerifyBestBlock(const uint256& hash);

src/validation.cpp

@@ -2423,6 +2423,7 @@ bool static FlushStateToDisk(const CChainParams& chainparams, CValidationState &
     }
     int64_t nMempoolSizeMax = gArgs.GetArg("-maxmempool", DEFAULT_MAX_MEMPOOL_SIZE) * 1000000;
     int64_t cacheSize = pcoinsTip->DynamicMemoryUsage() * DB_PEAK_USAGE_FACTOR;
+    cacheSize += evoDb->GetMemoryUsage() * DB_PEAK_USAGE_FACTOR;
     int64_t nTotalSpace = nCoinCacheUsage + std::max<int64_t>(nMempoolSizeMax - nMempoolUsage, 0);
     // The cache is large and we're within 10% and 10 MiB of the limit, but we have time now (not in the middle of a block processing).
     bool fCacheLarge = mode == FLUSH_STATE_PERIODIC && cacheSize > std::max((9 * nTotalSpace) / 10, nTotalSpace - MAX_BLOCK_COINSDB_USAGE * 1024 * 1024);
@@ -2563,6 +2564,7 @@ void static UpdateTip(CBlockIndex *pindexNew, const CChainParams& chainParams) {
       log(chainActive.Tip()->nChainWork.getdouble())/log(2.0), (unsigned long)chainActive.Tip()->nChainTx,
       DateTimeStrFormat("%Y-%m-%d %H:%M:%S", chainActive.Tip()->GetBlockTime()),
       GuessVerificationProgress(chainParams.TxData(), chainActive.Tip()), pcoinsTip->DynamicMemoryUsage() * (1.0 / (1<<20)), pcoinsTip->GetCacheSize());
+    strMessage += strprintf(" evodb_cache=%.1fMiB", evoDb->GetMemoryUsage() * (1.0 / (1<<20)));
     if (!warningMessages.empty())
         strMessage += strprintf(" warning='%s'", boost::algorithm::join(warningMessages, ", "));
     LogPrintf("%s\n", strMessage);