Merge #10148: Use non-atomic flushing with block replay

176c021 [qa] Test non-atomic chainstate writes (Suhas Daftuar)
d6af06d Dont create pcoinsTip until after ReplayBlocks. (Matt Corallo)
eaca1b7 Random db flush crash simulator (Pieter Wuille)
0580ee0 Adapt memory usage estimation for flushing (Pieter Wuille)
013a56a Non-atomic flushing using the blockchain as replay journal (Pieter Wuille)
b3a279c [MOVEONLY] Move LastCommonAncestor to chain (Pieter Wuille)

Tree-SHA512: 47ccc62303f9075c44d2a914be75bd6969ff881a857a2ff1227f05ec7def6f4c71c46680c5a28cb150c814999526797dc05cf2701fde1369c06169f46eccddee

1 files changed, 105 insertions, 12 deletions

diff --git a/src/validation.cpp b/src/validation.cpp
index 8d47b52ebe..216ba3d4a5 100644
--- a/src/validation.cpp
+++ b/src/validation.cpp
@@ -96,7 +96,7 @@ namespace {
 
     struct CBlockIndexWorkComparator
     {
-        bool operator()(CBlockIndex *pa, CBlockIndex *pb) const {
+        bool operator()(const CBlockIndex *pa, const CBlockIndex *pb) const {
             // First sort by most total work, ...
             if (pa->nChainWork > pb->nChainWork) return false;
             if (pa->nChainWork < pb->nChainWork) return true;
@@ -1331,17 +1331,19 @@ int ApplyTxInUndo(Coin&& undo, CCoinsViewCache& view, const COutPoint& out)
             return DISCONNECT_FAILED; // adding output for transaction without known metadata
         }
     }
-    view.AddCoin(out, std::move(undo), undo.fCoinBase);
+    // The potential_overwrite parameter to AddCoin is only allowed to be false if we know for
+    // sure that the coin did not already exist in the cache. As we have queried for that above
+    // using HaveCoin, we don't need to guess. When fClean is false, a coin already existed and
+    // it is an overwrite.
+    view.AddCoin(out, std::move(undo), !fClean);
 
     return fClean ? DISCONNECT_OK : DISCONNECT_UNCLEAN;
 }
 
 /** Undo the effects of this block (with given index) on the UTXO set represented by coins.
- *  When UNCLEAN or FAILED is returned, view is left in an indeterminate state. */
+ *  When FAILED is returned, view is left in an indeterminate state. */
 static DisconnectResult DisconnectBlock(const CBlock& block, const CBlockIndex* pindex, CCoinsViewCache& view)
 {
-    assert(pindex->GetBlockHash() == view.GetBestBlock());
-
     bool fClean = true;
 
     CBlockUndo blockUndo;
@@ -1779,7 +1781,7 @@ bool static FlushStateToDisk(const CChainParams& chainparams, CValidationState &
         nLastSetChain = nNow;
     }
     int64_t nMempoolSizeMax = GetArg("-maxmempool", DEFAULT_MAX_MEMPOOL_SIZE) * 1000000;
-    int64_t cacheSize = pcoinsTip->DynamicMemoryUsage() * DB_PEAK_USAGE_FACTOR;
+    int64_t cacheSize = pcoinsTip->DynamicMemoryUsage();
     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);
@@ -1946,6 +1948,7 @@ bool static DisconnectTip(CValidationState& state, const CChainParams& chainpara
     int64_t nStart = GetTimeMicros();
     {
         CCoinsViewCache view(pcoinsTip);
+        assert(view.GetBestBlock() == pindexDelete->GetBlockHash());
         if (DisconnectBlock(block, pindexDelete, view) != DISCONNECT_OK)
             return error("DisconnectTip(): DisconnectBlock %s failed", pindexDelete->GetBlockHash().ToString());
         bool flushed = view.Flush();
@@ -3417,20 +3420,25 @@ bool static LoadBlockIndexDB(const CChainParams& chainparams)
     pblocktree->ReadFlag("txindex", fTxIndex);
     LogPrintf("%s: transaction index %s\n", __func__, fTxIndex ? "enabled" : "disabled");
 
+    return true;
+}
+
+void LoadChainTip(const CChainParams& chainparams)
+{
+    if (chainActive.Tip() && chainActive.Tip()->GetBlockHash() == pcoinsTip->GetBestBlock()) return;
+
     // Load pointer to end of best chain
     BlockMap::iterator it = mapBlockIndex.find(pcoinsTip->GetBestBlock());
     if (it == mapBlockIndex.end())
-        return true;
+        return;
     chainActive.SetTip(it->second);
 
     PruneBlockIndexCandidates();
 
-    LogPrintf("%s: hashBestChain=%s height=%d date=%s progress=%f\n", __func__,
+    LogPrintf("Loaded best chain: hashBestChain=%s height=%d date=%s progress=%f\n",
         chainActive.Tip()->GetBlockHash().ToString(), chainActive.Height(),
         DateTimeStrFormat("%Y-%m-%d %H:%M:%S", chainActive.Tip()->GetBlockTime()),
         GuessVerificationProgress(chainparams.TxData(), chainActive.Tip()));
-
-    return true;
 }
 
 CVerifyDB::CVerifyDB()
@@ -3499,6 +3507,7 @@ bool CVerifyDB::VerifyDB(const CChainParams& chainparams, CCoinsView *coinsview,
         }
         // check level 3: check for inconsistencies during memory-only disconnect of tip blocks
         if (nCheckLevel >= 3 && pindex == pindexState && (coins.DynamicMemoryUsage() + pcoinsTip->DynamicMemoryUsage()) <= nCoinCacheUsage) {
+            assert(coins.GetBestBlock() == pindex->GetBlockHash());
             DisconnectResult res = DisconnectBlock(block, pindex, coins);
             if (res == DISCONNECT_FAILED) {
                 return error("VerifyDB(): *** irrecoverable inconsistency in block data at %d, hash=%s", pindex->nHeight, pindex->GetBlockHash().ToString());
@@ -3538,6 +3547,92 @@ bool CVerifyDB::VerifyDB(const CChainParams& chainparams, CCoinsView *coinsview,
     return true;
 }
 
+/** Apply the effects of a block on the utxo cache, ignoring that it may already have been applied. */
+static bool RollforwardBlock(const CBlockIndex* pindex, CCoinsViewCache& inputs, const CChainParams& params)
+{
+    // TODO: merge with ConnectBlock
+    CBlock block;
+    if (!ReadBlockFromDisk(block, pindex, params.GetConsensus())) {
+        return error("ReplayBlock(): ReadBlockFromDisk failed at %d, hash=%s", pindex->nHeight, pindex->GetBlockHash().ToString());
+    }
+
+    for (const CTransactionRef& tx : block.vtx) {
+        if (!tx->IsCoinBase()) {
+            for (const CTxIn &txin : tx->vin) {
+                inputs.SpendCoin(txin.prevout);
+            }
+        }
+        // Pass check = true as every addition may be an overwrite.
+        AddCoins(inputs, *tx, pindex->nHeight, true);
+    }
+    return true;
+}
+
+bool ReplayBlocks(const CChainParams& params, CCoinsView* view)
+{
+    LOCK(cs_main);
+
+    CCoinsViewCache cache(view);
+
+    std::vector<uint256> hashHeads = view->GetHeadBlocks();
+    if (hashHeads.empty()) return true; // We're already in a consistent state.
+    if (hashHeads.size() != 2) return error("ReplayBlocks(): unknown inconsistent state");
+
+    uiInterface.ShowProgress(_("Replaying blocks..."), 0);
+    LogPrintf("Replaying blocks\n");
+
+    const CBlockIndex* pindexOld = nullptr;  // Old tip during the interrupted flush.
+    const CBlockIndex* pindexNew;            // New tip during the interrupted flush.
+    const CBlockIndex* pindexFork = nullptr; // Latest block common to both the old and the new tip.
+
+    if (mapBlockIndex.count(hashHeads[0]) == 0) {
+        return error("ReplayBlocks(): reorganization to unknown block requested");
+    }
+    pindexNew = mapBlockIndex[hashHeads[0]];
+
+    if (!hashHeads[1].IsNull()) { // The old tip is allowed to be 0, indicating it's the first flush.
+        if (mapBlockIndex.count(hashHeads[1]) == 0) {
+            return error("ReplayBlocks(): reorganization from unknown block requested");
+        }
+        pindexOld = mapBlockIndex[hashHeads[1]];
+        pindexFork = LastCommonAncestor(pindexOld, pindexNew);
+        assert(pindexFork != nullptr);
+    }
+
+    // Rollback along the old branch.
+    while (pindexOld != pindexFork) {
+        if (pindexOld->nHeight > 0) { // Never disconnect the genesis block.
+            CBlock block;
+            if (!ReadBlockFromDisk(block, pindexOld, params.GetConsensus())) {
+                return error("RollbackBlock(): ReadBlockFromDisk() failed at %d, hash=%s", pindexOld->nHeight, pindexOld->GetBlockHash().ToString());
+            }
+            LogPrintf("Rolling back %s (%i)\n", pindexOld->GetBlockHash().ToString(), pindexOld->nHeight);
+            DisconnectResult res = DisconnectBlock(block, pindexOld, cache);
+            if (res == DISCONNECT_FAILED) {
+                return error("RollbackBlock(): DisconnectBlock failed at %d, hash=%s", pindexOld->nHeight, pindexOld->GetBlockHash().ToString());
+            }
+            // If DISCONNECT_UNCLEAN is returned, it means a non-existing UTXO was deleted, or an existing UTXO was
+            // overwritten. It corresponds to cases where the block-to-be-disconnect never had all its operations
+            // applied to the UTXO set. However, as both writing a UTXO and deleting a UTXO are idempotent operations,
+            // the result is still a version of the UTXO set with the effects of that block undone.
+        }
+        pindexOld = pindexOld->pprev;
+    }
+
+    // Roll forward from the forking point to the new tip.
+    int nForkHeight = pindexFork ? pindexFork->nHeight : 0;
+    for (int nHeight = nForkHeight + 1; nHeight <= pindexNew->nHeight; ++nHeight) {
+        const CBlockIndex* pindex = pindexNew->GetAncestor(nHeight);
+        LogPrintf("Rolling forward %s (%i)\n", pindex->GetBlockHash().ToString(), nHeight);
+        if (!RollforwardBlock(pindex, cache, params)) return false;
+    }
+
+    cache.SetBestBlock(pindexNew->GetBlockHash());
+    cache.Flush();
+    uiInterface.ShowProgress("", 100);
+    return true;
+}
+
 bool RewindBlockIndex(const CChainParams& params)
 {
     LOCK(cs_main);
@@ -3687,8 +3782,6 @@ bool InitBlockIndex(const CChainParams& chainparams)
             CBlockIndex *pindex = AddToBlockIndex(block);
             if (!ReceivedBlockTransactions(block, state, pindex, blockPos, chainparams.GetConsensus()))
                 return error("LoadBlockIndex(): genesis block not accepted");
-            // Force a chainstate write so that when we VerifyDB in a moment, it doesn't check stale data
-            return FlushStateToDisk(chainparams, state, FLUSH_STATE_ALWAYS);
         } catch (const std::runtime_error& e) {
             return error("LoadBlockIndex(): failed to initialize block database: %s", e.what());
         }