From 7d9ff96091312885a5dc5471c55e3b8afdbdb632 Mon Sep 17 00:00:00 2001 From: Kittywhiskers Van Gogh <63189531+kwvg@users.noreply.github.com> Date: Sun, 13 Oct 2024 16:46:11 +0000 Subject: [PATCH] merge bitcoin#16981: Improve runtime performance of --reindex --- src/Makefile.bench.include | 1 + src/bench/load_external.cpp | 63 ++++++++++++++++++++++++++++ src/streams.h | 48 ++++++++++++++------- src/test/streams_tests.cpp | 67 ++++++++++++++++++++++++++++-- src/validation.cpp | 46 ++++++++++++-------- test/functional/feature_reindex.py | 50 ++++++++++++++++++++++ 6 files changed, 238 insertions(+), 37 deletions(-) create mode 100644 src/bench/load_external.cpp diff --git a/src/Makefile.bench.include b/src/Makefile.bench.include index 133bc282ef..6e13fd173b 100644 --- a/src/Makefile.bench.include +++ b/src/Makefile.bench.include @@ -35,6 +35,7 @@ bench_bench_dash_SOURCES = \ bench/ccoins_caching.cpp \ bench/gcs_filter.cpp \ bench/hashpadding.cpp \ + bench/load_external.cpp \ bench/merkle_root.cpp \ bench/mempool_eviction.cpp \ bench/mempool_stress.cpp \ diff --git a/src/bench/load_external.cpp b/src/bench/load_external.cpp new file mode 100644 index 0000000000..6bc359a408 --- /dev/null +++ b/src/bench/load_external.cpp @@ -0,0 +1,63 @@ +// Copyright (c) 2022 The Bitcoin Core developers +// Distributed under the MIT software license, see the accompanying +// file COPYING or https://www.opensource.org/licenses/mit-license.php. + +#include +#include +#include +#include +#include + +/** + * The LoadExternalBlockFile() function is used during -reindex and -loadblock. + * + * Create a test file that's similar to a datadir/blocks/blk?????.dat file, + * It contains around 134 copies of the same block (typical size of real block files). + * For each block in the file, LoadExternalBlockFile() won't find its parent, + * and so will skip the block. (In the real system, it will re-read the block + * from disk later when it encounters its parent.) + * + * This benchmark measures the performance of deserializing the block (or just + * its header, beginning with PR 16981). + */ +static void LoadExternalBlockFile(benchmark::Bench& bench) +{ + const auto testing_setup{MakeNoLogFileContext(CBaseChainParams::MAIN)}; + + // Create a single block as in the blocks files (magic bytes, block size, + // block data) as a stream object. + const fs::path blkfile{testing_setup.get()->m_path_root / "blk.dat"}; + CDataStream ss(SER_DISK, 0); + auto params{Params()}; + ss << params.MessageStart(); + ss << static_cast(benchmark::data::block813851.size()); + // We can't use the streaming serialization (ss << benchmark::data::block813851) + // because that first writes a compact size. + ss.write(MakeByteSpan(benchmark::data::block813851)); + + // Create the test file. + { + // "wb+" is "binary, O_RDWR | O_CREAT | O_TRUNC". + FILE* file{fsbridge::fopen(blkfile, "wb+")}; + // Make the test block file about 128 MB in length. + for (size_t i = 0; i < MAX_BLOCKFILE_SIZE / ss.size(); ++i) { + if (fwrite(ss.data(), 1, ss.size(), file) != ss.size()) { + throw std::runtime_error("write to test file failed\n"); + } + } + fclose(file); + } + + CChainState& chainstate{testing_setup->m_node.chainman->ActiveChainstate()}; + std::multimap blocks_with_unknown_parent; + FlatFilePos pos; + bench.run([&] { + // "rb" is "binary, O_RDONLY", positioned to the start of the file. + // The file will be closed by LoadExternalBlockFile(). + FILE* file{fsbridge::fopen(blkfile, "rb")}; + chainstate.LoadExternalBlockFile(file, &pos, &blocks_with_unknown_parent); + }); + fs::remove(blkfile); +} + +BENCHMARK(LoadExternalBlockFile); diff --git a/src/streams.h b/src/streams.h index ced2b1f76f..76feda569f 100644 --- a/src/streams.h +++ b/src/streams.h @@ -641,7 +641,6 @@ private: uint64_t nRewind; //!< how many bytes we guarantee to rewind std::vector vchBuf; //!< the buffer -protected: //! read data from the source to fill the buffer bool Fill() { unsigned int pos = nSrcPos % vchBuf.size(); @@ -659,6 +658,28 @@ protected: return true; } + //! Advance the stream's read pointer (m_read_pos) by up to 'length' bytes, + //! filling the buffer from the file so that at least one byte is available. + //! Return a pointer to the available buffer data and the number of bytes + //! (which may be less than the requested length) that may be accessed + //! beginning at that pointer. + std::pair AdvanceStream(size_t length) + { + assert(m_read_pos <= nSrcPos); + if (m_read_pos + length > nReadLimit) { + throw std::ios_base::failure("Attempt to position past buffer limit"); + } + // If there are no bytes available, read from the file. + if (m_read_pos == nSrcPos && length > 0) Fill(); + + size_t buffer_offset{static_cast(m_read_pos % vchBuf.size())}; + size_t buffer_available{static_cast(vchBuf.size() - buffer_offset)}; + size_t bytes_until_source_pos{static_cast(nSrcPos - m_read_pos)}; + size_t advance{std::min({length, buffer_available, bytes_until_source_pos})}; + m_read_pos += advance; + return std::make_pair(&vchBuf[buffer_offset], advance); + } + public: CBufferedFile(FILE* fileIn, uint64_t nBufSize, uint64_t nRewindIn, int nTypeIn, int nVersionIn) : nType(nTypeIn), nVersion(nVersionIn), nSrcPos(0), m_read_pos(0), nReadLimit(std::numeric_limits::max()), nRewind(nRewindIn), vchBuf(nBufSize, std::byte{0}) @@ -696,24 +717,21 @@ public: //! read a number of bytes void read(Span dst) { - if (dst.size() + m_read_pos > nReadLimit) { - throw std::ios_base::failure("Read attempted past buffer limit"); - } while (dst.size() > 0) { - if (m_read_pos == nSrcPos) - Fill(); - unsigned int pos = m_read_pos % vchBuf.size(); - size_t nNow = dst.size(); - if (nNow + pos > vchBuf.size()) - nNow = vchBuf.size() - pos; - if (nNow + m_read_pos > nSrcPos) - nNow = nSrcPos - m_read_pos; - memcpy(dst.data(), &vchBuf[pos], nNow); - m_read_pos += nNow; - dst = dst.subspan(nNow); + auto [buffer_pointer, length]{AdvanceStream(dst.size())}; + memcpy(dst.data(), buffer_pointer, length); + dst = dst.subspan(length); } } + //! Move the read position ahead in the stream to the given position. + //! Use SetPos() to back up in the stream, not SkipTo(). + void SkipTo(const uint64_t file_pos) + { + assert(file_pos >= m_read_pos); + while (m_read_pos < file_pos) AdvanceStream(file_pos - m_read_pos); + } + //! return the current reading position uint64_t GetPos() const { return m_read_pos; diff --git a/src/test/streams_tests.cpp b/src/test/streams_tests.cpp index 2eae80bf8b..2f8b76d592 100644 --- a/src/test/streams_tests.cpp +++ b/src/test/streams_tests.cpp @@ -251,7 +251,7 @@ BOOST_AUTO_TEST_CASE(streams_buffered_file) BOOST_CHECK(false); } catch (const std::exception& e) { BOOST_CHECK(strstr(e.what(), - "Read attempted past buffer limit") != nullptr); + "Attempt to position past buffer limit") != nullptr); } // The default argument removes the limit completely. BOOST_CHECK(bf.SetLimit()); @@ -320,7 +320,7 @@ BOOST_AUTO_TEST_CASE(streams_buffered_file) BOOST_CHECK(!bf.SetPos(0)); // But we should now be positioned at least as far back as allowed // by the rewind window (relative to our farthest read position, 40). - BOOST_CHECK(bf.GetPos() <= 30); + BOOST_CHECK(bf.GetPos() <= 30U); // We can explicitly close the file, or the destructor will do it. bf.fclose(); @@ -328,6 +328,55 @@ BOOST_AUTO_TEST_CASE(streams_buffered_file) fs::remove("streams_test_tmp"); } +BOOST_AUTO_TEST_CASE(streams_buffered_file_skip) +{ + fs::path streams_test_filename = m_args.GetDataDirBase() / "streams_test_tmp"; + FILE* file = fsbridge::fopen(streams_test_filename, "w+b"); + // The value at each offset is the byte offset (e.g. byte 1 in the file has the value 0x01). + for (uint8_t j = 0; j < 40; ++j) { + fwrite(&j, 1, 1, file); + } + rewind(file); + + // The buffer is 25 bytes, allow rewinding 10 bytes. + CBufferedFile bf(file, 25, 10, 222, 333); + + uint8_t i; + // This is like bf >> (7-byte-variable), in that it will cause data + // to be read from the file into memory, but it's not copied to us. + bf.SkipTo(7); + BOOST_CHECK_EQUAL(bf.GetPos(), 7U); + bf >> i; + BOOST_CHECK_EQUAL(i, 7); + + // The bytes in the buffer up to offset 7 are valid and can be read. + BOOST_CHECK(bf.SetPos(0)); + bf >> i; + BOOST_CHECK_EQUAL(i, 0); + bf >> i; + BOOST_CHECK_EQUAL(i, 1); + + bf.SkipTo(11); + bf >> i; + BOOST_CHECK_EQUAL(i, 11); + + // SkipTo() honors the transfer limit; we can't position beyond the limit. + bf.SetLimit(13); + try { + bf.SkipTo(14); + BOOST_CHECK(false); + } catch (const std::exception& e) { + BOOST_CHECK(strstr(e.what(), "Attempt to position past buffer limit") != nullptr); + } + + // We can position exactly to the transfer limit. + bf.SkipTo(13); + BOOST_CHECK_EQUAL(bf.GetPos(), 13U); + + bf.fclose(); + fs::remove(streams_test_filename); +} + BOOST_AUTO_TEST_CASE(streams_buffered_file_rand) { // Make this test deterministic. @@ -358,7 +407,7 @@ BOOST_AUTO_TEST_CASE(streams_buffered_file_rand) // sizes; the boundaries of the objects can interact arbitrarily // with the CBufferFile's internal buffer. These first three // cases simulate objects of various sizes (1, 2, 5 bytes). - switch (InsecureRandRange(5)) { + switch (InsecureRandRange(6)) { case 0: { uint8_t a[1]; if (currentPos + 1 > fileSize) @@ -396,6 +445,16 @@ BOOST_AUTO_TEST_CASE(streams_buffered_file_rand) break; } case 3: { + // SkipTo is similar to the "read" cases above, except + // we don't receive the data. + size_t skip_length{static_cast(InsecureRandRange(5))}; + if (currentPos + skip_length > fileSize) continue; + bf.SetLimit(currentPos + skip_length); + bf.SkipTo(currentPos + skip_length); + currentPos += skip_length; + break; + } + case 4: { // Find a byte value (that is at or ahead of the current position). size_t find = currentPos + InsecureRandRange(8); if (find >= fileSize) @@ -412,7 +471,7 @@ BOOST_AUTO_TEST_CASE(streams_buffered_file_rand) currentPos++; break; } - case 4: { + case 5: { size_t requestPos = InsecureRandRange(maxPos + 4); bool okay = bf.SetPos(requestPos); // The new position may differ from the requested position diff --git a/src/validation.cpp b/src/validation.cpp index 844ad7eb6e..0e9ca1b609 100644 --- a/src/validation.cpp +++ b/src/validation.cpp @@ -4603,6 +4603,8 @@ void CChainState::LoadExternalBlockFile( unsigned int nMaxBlockSize = MaxBlockSize(); // This takes over fileIn and calls fclose() on it in the CBufferedFile destructor CBufferedFile blkdat(fileIn, 2*nMaxBlockSize, nMaxBlockSize+8, SER_DISK, CLIENT_VERSION); + // nRewind indicates where to resume scanning in case something goes wrong, + // such as a block fails to deserialize. uint64_t nRewind = blkdat.GetPos(); while (!blkdat.eof()) { if (ShutdownRequested()) return; @@ -4626,28 +4628,30 @@ void CChainState::LoadExternalBlockFile( continue; } catch (const std::exception&) { // no valid block header found; don't complain + // (this happens at the end of every blk.dat file) break; } try { - // read block - uint64_t nBlockPos = blkdat.GetPos(); + // read block header + const uint64_t nBlockPos{blkdat.GetPos()}; if (dbp) dbp->nPos = nBlockPos; blkdat.SetLimit(nBlockPos + nSize); - std::shared_ptr pblock = std::make_shared(); - CBlock& block = *pblock; - blkdat >> block; - nRewind = blkdat.GetPos(); - - uint256 hash = block.GetHash(); + CBlockHeader header; + blkdat >> header; + const uint256 hash{header.GetHash()}; + // Skip the rest of this block (this may read from disk into memory); position to the marker before the + // next block, but it's still possible to rewind to the start of the current block (without a disk read). + nRewind = nBlockPos + nSize; + blkdat.SkipTo(nRewind); { LOCK(cs_main); // detect out of order blocks, and store them for later - if (hash != m_params.GetConsensus().hashGenesisBlock && !m_blockman.LookupBlockIndex(block.hashPrevBlock)) { + if (hash != m_params.GetConsensus().hashGenesisBlock && !m_blockman.LookupBlockIndex(header.hashPrevBlock)) { LogPrint(BCLog::REINDEX, "%s: Out of order block %s, parent %s not known\n", __func__, hash.ToString(), - block.hashPrevBlock.ToString()); + header.hashPrevBlock.ToString()); if (dbp && blocks_with_unknown_parent) { - blocks_with_unknown_parent->emplace(block.hashPrevBlock, *dbp); + blocks_with_unknown_parent->emplace(header.hashPrevBlock, *dbp); } continue; } @@ -4655,13 +4659,19 @@ void CChainState::LoadExternalBlockFile( // process in case the block isn't known yet const CBlockIndex* pindex = m_blockman.LookupBlockIndex(hash); if (!pindex || (pindex->nStatus & BLOCK_HAVE_DATA) == 0) { - BlockValidationState state; - if (AcceptBlock(pblock, state, nullptr, true, dbp, nullptr)) { - nLoaded++; - } - if (state.IsError()) { - break; - } + // This block can be processed immediately; rewind to its start, read and deserialize it. + blkdat.SetPos(nBlockPos); + std::shared_ptr pblock{std::make_shared()}; + blkdat >> *pblock; + nRewind = blkdat.GetPos(); + + BlockValidationState state; + if (AcceptBlock(pblock, state, nullptr, true, dbp, nullptr)) { + nLoaded++; + } + if (state.IsError()) { + break; + } } else if (hash != m_params.GetConsensus().hashGenesisBlock && pindex->nHeight % 1000 == 0) { LogPrint(BCLog::REINDEX, "Block Import: already had block %s at height %d\n", hash.ToString(), pindex->nHeight); } diff --git a/test/functional/feature_reindex.py b/test/functional/feature_reindex.py index 8c15a74e2f..588d8766ad 100755 --- a/test/functional/feature_reindex.py +++ b/test/functional/feature_reindex.py @@ -7,9 +7,12 @@ - Start a single node and generate 3 blocks. - Stop the node and restart it with -reindex. Verify that the node has reindexed up to block 3. - Stop the node and restart it with -reindex-chainstate. Verify that the node has reindexed up to block 3. +- Verify that out-of-order blocks are correctly processed, see LoadExternalBlockFile() """ +import os from test_framework.test_framework import BitcoinTestFramework +from test_framework.p2p import MAGIC_BYTES from test_framework.util import assert_equal @@ -27,6 +30,50 @@ class ReindexTest(BitcoinTestFramework): assert_equal(self.nodes[0].getblockcount(), blockcount) # start_node is blocking on reindex self.log.info("Success") + # Check that blocks can be processed out of order + def out_of_order(self): + # The previous test created 24 blocks + assert_equal(self.nodes[0].getblockcount(), 24) + self.stop_nodes() + + # In this test environment, blocks will always be in order (since + # we're generating them rather than getting them from peers), so to + # test out-of-order handling, swap blocks 1 and 2 on disk. + blk0 = os.path.join(self.nodes[0].datadir, self.nodes[0].chain, 'blocks', 'blk00000.dat') + with open(blk0, 'r+b') as bf: + # Read at least the first few blocks (including genesis) + b = bf.read(2000) + + # Find the offsets of blocks 2, 3, and 4 (the first 3 blocks beyond genesis) + # by searching for the regtest marker bytes (see pchMessageStart). + def find_block(b, start): + return b.find(MAGIC_BYTES["regtest"], start)+4 + + genesis_start = find_block(b, 0) + assert_equal(genesis_start, 4) + b2_start = find_block(b, genesis_start) + b3_start = find_block(b, b2_start) + b4_start = find_block(b, b3_start) + + # Blocks 2 and 3 should be the same size. + assert_equal(b3_start-b2_start, b4_start-b3_start) + + # Swap the second and third blocks (don't disturb the genesis block). + bf.seek(b2_start) + bf.write(b[b3_start:b4_start]) + bf.write(b[b2_start:b3_start]) + + # The reindexing code should detect and accommodate out of order blocks. + with self.nodes[0].assert_debug_log([ + 'LoadExternalBlockFile: Out of order block', + 'LoadExternalBlockFile: Processing out of order child', + ]): + extra_args = [["-reindex"]] + self.start_nodes(extra_args) + + # All blocks should be accepted and processed. + assert_equal(self.nodes[0].getblockcount(), 24) + def run_test(self): for txindex in [0, 1]: self.reindex(False, txindex) @@ -34,5 +81,8 @@ class ReindexTest(BitcoinTestFramework): self.reindex(False, txindex) self.reindex(True, txindex) + self.out_of_order() + + if __name__ == '__main__': ReindexTest().main()