validation: rename snapshot chainstate dir

This changes the snapshot's leveldb chainstate dir name from
`chainstate_[blockhash]` to `chainstate_snapshot`. This simplifies
later logic that loads snapshot data, and enforces the limitation
of a single snapshot at any given time.

Since we still need to persis the blockhash of the base block, we
write that out to a file (`chainstate_snapshot/base_blockhash`) for
later use during initialization, so that we can reinitialize the
snapshot chainstate.

Co-authored-by: Russell Yanofsky <russ@yanofsky.org>

src/Makefile.am

@@ -395,6 +395,7 @@ libbitcoin_node_a_SOURCES = \
   node/minisketchwrapper.cpp \
   node/psbt.cpp \
   node/transaction.cpp \
+  node/utxo_snapshot.cpp \
   node/validation_cache_args.cpp \
   noui.cpp \
   policy/fees.cpp \
@@ -900,6 +901,7 @@ libbitcoinkernel_la_SOURCES = \
   node/blockstorage.cpp \
   node/chainstate.cpp \
   node/interface_ui.cpp \
+  node/utxo_snapshot.cpp \
   policy/feerate.cpp \
   policy/fees.cpp \
   policy/packages.cpp \

src/node/utxo_snapshot.cpp

@@ -0,0 +1,79 @@
+// Copyright (c) 2022 The Bitcoin Core developers
+// Distributed under the MIT software license, see the accompanying
+// file COPYING or http://www.opensource.org/licenses/mit-license.php.
+
+#include <node/utxo_snapshot.h>
+
+#include <fs.h>
+#include <logging.h>
+#include <streams.h>
+#include <uint256.h>
+#include <validation.h>
+
+#include <cstdio>
+#include <optional>
+
+namespace node {
+
+bool WriteSnapshotBaseBlockhash(Chainstate& snapshot_chainstate)
+{
+    AssertLockHeld(::cs_main);
+    assert(snapshot_chainstate.m_from_snapshot_blockhash);
+
+    const std::optional<fs::path> chaindir = snapshot_chainstate.CoinsDB().StoragePath();
+    assert(chaindir); // Sanity check that chainstate isn't in-memory.
+    const fs::path write_to = *chaindir / node::SNAPSHOT_BLOCKHASH_FILENAME;
+
+    FILE* file{fsbridge::fopen(write_to, "wb")};
+    AutoFile afile{file};
+    if (afile.IsNull()) {
+        LogPrintf("[snapshot] failed to open base blockhash file for writing: %s\n",
+                  fs::PathToString(write_to));
+        return false;
+    }
+    afile << *snapshot_chainstate.m_from_snapshot_blockhash;
+
+    if (afile.fclose() != 0) {
+        LogPrintf("[snapshot] failed to close base blockhash file %s after writing\n",
+                  fs::PathToString(write_to));
+        return false;
+    }
+    return true;
+}
+
+std::optional<uint256> ReadSnapshotBaseBlockhash(fs::path chaindir)
+{
+    if (!fs::exists(chaindir)) {
+        LogPrintf("[snapshot] cannot read base blockhash: no chainstate dir " /* Continued */
+            "exists at path %s\n", fs::PathToString(chaindir));
+        return std::nullopt;
+    }
+    const fs::path read_from = chaindir / node::SNAPSHOT_BLOCKHASH_FILENAME;
+    const std::string read_from_str = fs::PathToString(read_from);
+
+    if (!fs::exists(read_from)) {
+        LogPrintf("[snapshot] snapshot chainstate dir is malformed! no base blockhash file " /* Continued */
+            "exists at path %s. Try deleting %s and calling loadtxoutset again?\n",
+            fs::PathToString(chaindir), read_from_str);
+        return std::nullopt;
+    }
+
+    uint256 base_blockhash;
+    FILE* file{fsbridge::fopen(read_from, "rb")};
+    AutoFile afile{file};
+    if (afile.IsNull()) {
+        LogPrintf("[snapshot] failed to open base blockhash file for reading: %s\n",
+            read_from_str);
+        return std::nullopt;
+    }
+    afile >> base_blockhash;
+
+    if (std::fgetc(afile.Get()) != EOF) {
+        LogPrintf("[snapshot] warning: unexpected trailing data in %s\n", read_from_str);
+    } else if (std::ferror(afile.Get())) {
+        LogPrintf("[snapshot] warning: i/o error reading %s\n", read_from_str);
+    }
+    return base_blockhash;
+}
+
+} // namespace node

src/node/utxo_snapshot.h

@@ -6,8 +6,14 @@
 #ifndef BITCOIN_NODE_UTXO_SNAPSHOT_H
 #define BITCOIN_NODE_UTXO_SNAPSHOT_H
 
+#include <fs.h>
 #include <uint256.h>
 #include <serialize.h>
+#include <validation.h>
+
+#include <optional>
+
+extern RecursiveMutex cs_main;
 
 namespace node {
 //! Metadata describing a serialized version of a UTXO set from which an
@@ -33,6 +39,27 @@ public:
 
     SERIALIZE_METHODS(SnapshotMetadata, obj) { READWRITE(obj.m_base_blockhash, obj.m_coins_count); }
 };
+
+//! The file in the snapshot chainstate dir which stores the base blockhash. This is
+//! needed to reconstruct snapshot chainstates on init.
+//!
+//! Because we only allow loading a single snapshot at a time, there will only be one
+//! chainstate directory with this filename present within it.
+const fs::path SNAPSHOT_BLOCKHASH_FILENAME{"base_blockhash"};
+
+//! Write out the blockhash of the snapshot base block that was used to construct
+//! this chainstate. This value is read in during subsequent initializations and
+//! used to reconstruct snapshot-based chainstates.
+bool WriteSnapshotBaseBlockhash(Chainstate& snapshot_chainstate)
+    EXCLUSIVE_LOCKS_REQUIRED(::cs_main);
+
+//! Read the blockhash of the snapshot base block that was used to construct the
+//! chainstate.
+std::optional<uint256> ReadSnapshotBaseBlockhash(fs::path chaindir)
+    EXCLUSIVE_LOCKS_REQUIRED(::cs_main);
+
+constexpr std::string_view SNAPSHOT_CHAINSTATE_SUFFIX = "_snapshot";
+
 } // namespace node
 
 #endif // BITCOIN_NODE_UTXO_SNAPSHOT_H

src/streams.h

@@ -488,12 +488,14 @@ public:
     AutoFile(const AutoFile&) = delete;
     AutoFile& operator=(const AutoFile&) = delete;
 
-    void fclose()
+    int fclose()
     {
+        int retval{0};
         if (file) {
-            ::fclose(file);
+            retval = ::fclose(file);
             file = nullptr;
         }
+        return retval;
     }
 
     /** Get wrapped FILE* with transfer of ownership.

src/test/validation_chainstatemanager_tests.cpp

@@ -232,8 +232,17 @@ BOOST_FIXTURE_TEST_CASE(chainstatemanager_activate_snapshot, TestChain100Setup)
         *chainman.ActiveChainstate().m_from_snapshot_blockhash,
         *chainman.SnapshotBlockhash());
 
+    {
+        LOCK(::cs_main);
+
+        // Note: WriteSnapshotBaseBlockhash() is implicitly tested above.
+        BOOST_CHECK_EQUAL(
+            *node::ReadSnapshotBaseBlockhash(m_args.GetDataDirNet() / "chainstate_snapshot"),
+            *chainman.SnapshotBlockhash());
+
         // Ensure that the genesis block was not marked assumed-valid.
-    BOOST_CHECK(WITH_LOCK(::cs_main, return !chainman.ActiveChain().Genesis()->IsAssumedValid()));
+        BOOST_CHECK(!chainman.ActiveChain().Genesis()->IsAssumedValid());
+    }
 
     const AssumeutxoData& au_data = *ExpectedAssumeutxo(snapshot_height, ::Params());
     const CBlockIndex* tip = WITH_LOCK(chainman.GetMutex(), return chainman.ActiveTip());

src/validation.cpp

@@ -1515,7 +1515,7 @@ void Chainstate::InitCoinsDB(
     fs::path leveldb_name)
 {
     if (m_from_snapshot_blockhash) {
-        leveldb_name += "_" + m_from_snapshot_blockhash->ToString();
+        leveldb_name += node::SNAPSHOT_CHAINSTATE_SUFFIX;
     }
 
     m_coins_views = std::make_unique<CoinsViews>(
@@ -4837,9 +4837,17 @@ bool ChainstateManager::ActivateSnapshot(
             static_cast<size_t>(current_coinstip_cache_size * SNAPSHOT_CACHE_PERC));
     }
 
-    const bool snapshot_ok = this->PopulateAndValidateSnapshot(
+    bool snapshot_ok = this->PopulateAndValidateSnapshot(
         *snapshot_chainstate, coins_file, metadata);
 
+    // If not in-memory, persist the base blockhash for use during subsequent
+    // initialization.
+    if (!in_memory) {
+        LOCK(::cs_main);
+        if (!node::WriteSnapshotBaseBlockhash(*snapshot_chainstate)) {
+            snapshot_ok = false;
+        }
+    }
     if (!snapshot_ok) {
         WITH_LOCK(::cs_main, this->MaybeRebalanceCaches());
         return false;

test/lint/lint-circular-dependencies.py

@@ -14,6 +14,7 @@ import sys
 EXPECTED_CIRCULAR_DEPENDENCIES = (
     "chainparamsbase -> util/system -> chainparamsbase",
     "node/blockstorage -> validation -> node/blockstorage",
+    "node/utxo_snapshot -> validation -> node/utxo_snapshot",
     "policy/fees -> txmempool -> policy/fees",
     "qt/addresstablemodel -> qt/walletmodel -> qt/addresstablemodel",
     "qt/recentrequeststablemodel -> qt/walletmodel -> qt/recentrequeststablemodel",