Files
bitcoin/test/functional
Ryan Ofsky c0e91efdb3 Merge bitcoin/bitcoin#35295: validation: fetch block input prevouts in parallel during ConnectBlock
dc1c17c085 doc: add release notes (Andrew Toth)
0e10937184 fuzz: add coins_view_stacked fuzz harness to test concurrent leveldb reads (Andrew Toth)
ce610a6ff4 fuzz: update harnesses to cover CoinsViewOverlay::StartFetching (Andrew Toth)
760fb22dc3 test: add unit tests for CoinsViewOverlay::StartFetching (Andrew Toth)
d69a3b20de doc: update CoinsViewOverlay docstring to describe parallel fetching (Andrew Toth)
ab2a379237 coins: fetch inputs in parallel (Andrew Toth)
fdf283036a coins: add ready flag to InputToFetch (Andrew Toth)
ede11b8314 validation: collect block inputs in CoinsViewOverlay before ConnectBlock (Andrew Toth)
f82043af50 coins: introduce thread pool in CoinsViewOverlay (Andrew Toth)
5bf1c32008 validation: add -prevoutfetchthreads configuration option (Andrew Toth)

Pull request description:

  This PR is a continuation of https://github.com/bitcoin/bitcoin/pull/31132. All outstanding issues raised there have been resolved, but the volume of stale comments can make that change difficult to review.

  Currently, when connecting a block, each input prevout is looked up one at a time. For every input we first check the in-memory coins cache, and on a miss we make a synchronous round-trip to the chainstate LevelDB to read the coin from disk. Because these lookups happen serially as the block is being validated, the disk read latency stacks up and dominates the time spent in `ConnectBlock` whenever many inputs are not already in the cache.

  This PR moves those disk reads onto a pool of worker threads that run in parallel with block connection. Before entering `ConnectBlock` the block is handed to a `CoinsViewOverlay`, which kicks off the workers to begin fetching all of the block's prevouts from disk and warming the cache. The main validation thread continues to do exactly the same work it does today, hitting the cache for each input in order. The only difference is that by the time it asks, the coin is much more likely to already be there. There are no validation logic or consensus behavior changes. This is purely a parallelization of an existing read pattern.

  The number of fetcher threads is configurable via `-prevoutfetchthreads=<n>`, defaulting to 8 and capped at 16. Setting it to 0 disables input fetching entirely and reverts to the previous serial behavior.

  We have measured large performance gains for IBD and `-reindex-chainstate`, as well as worst-case steady-state block connection at the tip. l0rinc ran many thorough benchmarking passes on the original PR across multiple machines, storage types, dbcache sizes[^1], operating systems[^2], and fetcher thread counts[^3]. Many other contributors also posted their benchmark results in the original PR. IBD speedups range from 1.18× to over 3× faster[^4]. Worst-case block connection time for network-attached storage was over 2× faster[^5]. Flamegraph comparisons before and after this change are available[^6].

  On safety: `ConnectBlock` runs while holding `cs_main`, so nothing else in the node can mutate the chainstate while the fetchers are reading it.

  On LevelDB: [concurrent reads are fully supported](https://github.com/bitcoin/bitcoin/blob/master/src/leveldb/include/leveldb/db.h#L44) and [documented as such](https://github.com/bitcoin/bitcoin/blob/master/src/leveldb/doc/index.md#concurrency). We already rely on this in production today against our other LevelDB-backed databases. The `txindex` DB is read by multiple simultaneous HTTP RPC worker threads via the `getrawtransaction` RPC. The `blockfilterindex` DB is called concurrently from both the P2P `cfilters` / `cfheaders` / `cfcheckpt` message handlers on the `msghand` thread, and from the `getblockfilter` RPC on the HTTP RPC worker threads. We have not yet been issuing concurrent reads against the chainstate DB, but there is no LevelDB-side reason we can't. In fact, the chainstate DB is already being touched by more than one thread on master, because LevelDB schedules its own background compaction work.

  For reviewers:

  The main change is `CoinsViewOverlay` gets 1 new public and 2 new private methods.

  - `StartFetching`: public method called in lieu of `CreateResetGuard` before we enter `ConnectBlock`. It still returns a `ResetGuard` so the view is `Reset` before the block it is working on leaves scope. This kicks off worker threads who each just run `while (ProcessInput()) {}` and then return.
  - `StopFetching`: private method called on `Reset` whenever the guard leaves scope or `Flush`. Stops all threads and clears multi threaded state.
  - `ProcessInput`: private method that fetches a single input prevout. Returns `true` if an input was fetched and `false` otherwise. This is the only method on `CoinsViewOverlay` that is called concurrently by multiple threads. Every other method on the overlay is still called synchronously on the main thread.

  The `CoinsViewOverlay::FetchCoinFromBase` method is also extended to lookup the coins fetched from `ProcessInput` first before falling back to `base->PeekCoin`.

  Mutating methods `Reset` and `Flush` are overridden in `CoinsViewOverlay` to call `StopFetching` first.

  [^1]: https://github.com/bitcoin/bitcoin/pull/31132#pullrequestreview-3515011880
  [^2]: https://github.com/bitcoin/bitcoin/pull/31132#issuecomment-3767758819
  [^3]: https://github.com/bitcoin/bitcoin/pull/31132#issuecomment-3617721711
  [^4]: https://github.com/bitcoin/bitcoin/pull/31132#issuecomment-3678847806
  [^5]: https://github.com/bitcoin/bitcoin/pull/31132#issuecomment-4071032270
  [^6]: https://github.com/bitcoin/bitcoin/pull/31132#issuecomment-3617315125

ACKs for top commit:
  l0rinc:
    reACK dc1c17c085
  willcl-ark:
    ACK dc1c17c085
  theStack:
    re-ACK dc1c17c085
  ryanofsky:
    Code review ACK dc1c17c085 with changes to StopFetching and AllInputsConsumed checking behavior since last review.

Tree-SHA512: 89c1c2890f65aac5cd546edc44504956c47b6fada256d3b86ced47e6dd8c72f633a4357753b3b9805b9ba6ed02790822090d70578aba2964baf50d7eb956864c
2026-07-08 20:49:48 -04:00
..
2026-06-30 00:33:00 +02:00
2026-06-04 21:58:56 +10:00

Functional tests

Writing Functional Tests

Example test

The file test/functional/example_test.py is a heavily commented example of a test case that uses both the RPC and P2P interfaces. If you are writing your first test, copy that file and modify to fit your needs.

Coverage

Assuming the build directory is build, running build/test/functional/test_runner.py with the --coverage argument tracks which RPCs are called by the tests and prints a report of uncovered RPCs in the summary. This can be used (along with the --extended argument) to find out which RPCs we don't have test cases for.

Style guidelines

  • Where possible, try to adhere to PEP-8 guidelines
  • Use a python linter like flake8 before submitting PRs to catch common style nits (eg trailing whitespace, unused imports, etc)
  • The oldest supported Python version is specified in doc/dependencies.md. Consider using pyenv, which checks .python-version, to prevent accidentally introducing modern syntax from an unsupported Python version. The CI linter job also checks this, but possibly not in all cases.
  • See the python lint script that checks for violations that could lead to bugs and issues in the test code.
  • Use type hints in your code to improve code readability and to detect possible bugs earlier.
  • Avoid wildcard imports.
  • If more than one name from a module is needed, use lexicographically sorted multi-line imports in order to reduce the possibility of potential merge conflicts.
  • Use a module-level docstring to describe what the test is testing, and how it is testing it.
  • When subclassing the BitcoinTestFramework, place overrides for the set_test_params(), add_options() and setup_xxxx() methods at the top of the subclass, then locally-defined helper methods, then the run_test() method.
  • Use f'{x}' for string formatting in preference to '{}'.format(x) or '%s' % x.
  • Use platform.system() for detecting the running operating system and os.name to check whether it's a POSIX system (see also the skip_if_platform_not_{linux,posix} methods in the BitcoinTestFramework class, which can be used to skip a whole test depending on the platform).

Naming guidelines

  • Name the test <area>_test.py, where area can be one of the following:
    • feature for tests for full features that aren't wallet/mining/mempool, eg feature_rbf.py
    • interface for tests for other interfaces (REST, ZMQ, etc), eg interface_rest.py
    • mempool for tests for mempool behaviour, eg mempool_reorg.py
    • mining for tests for mining features, eg mining_prioritisetransaction.py
    • p2p for tests that explicitly test the p2p interface, eg p2p_disconnect_ban.py
    • rpc for tests for individual RPC methods or features, eg rpc_listtransactions.py
    • tool for tests for tools, eg tool_wallet.py
    • wallet for tests for wallet features, eg wallet_keypool.py
  • Use an underscore to separate words
    • exception: for tests for specific RPCs or command line options which don't include underscores, name the test after the exact RPC or argument name, eg rpc_decodescript.py, not rpc_decode_script.py
  • Don't use the redundant word test in the name, eg interface_zmq.py, not interface_zmq_test.py

General test-writing advice

  • Instead of inline comments or no test documentation at all, log the comments to the test log, e.g. self.log.info('Create enough transactions to fill a block'). Logs make the test code easier to read and the test logic easier to debug.
  • Set self.num_nodes to the minimum number of nodes necessary for the test. Having additional unrequired nodes adds to the execution time of the test as well as memory/CPU/disk requirements (which is important when running tests in parallel).
  • Avoid stop-starting the nodes multiple times during the test if possible. A stop-start takes several seconds, so doing it several times blows up the runtime of the test.
  • Set the self.setup_clean_chain variable in set_test_params() to True to initialize an empty blockchain and start from the Genesis block, rather than load a premined blockchain from cache with the default value of False. The cached data directories contain a 200-block pre-mined blockchain with the spendable mining rewards being split between four nodes. Each node has 25 mature block subsidies (25x50=1250 BTC) in its wallet. Using them is much more efficient than mining blocks in your test.
  • When calling RPCs with lots of arguments, consider using named keyword arguments instead of positional arguments to make the intent of the call clear to readers.
  • Many of the core test framework classes such as CBlock and CTransaction don't allow new attributes to be added to their objects at runtime like typical Python objects allow. This helps prevent unpredictable side effects from typographical errors or usage of the objects outside of their intended purpose.

RPC and P2P definitions

Test writers may find it helpful to refer to the definitions for the RPC and P2P messages. These can be found in the following source files:

  • /src/rpc/* for RPCs
  • /src/wallet/rpc* for wallet RPCs
  • ProcessMessage() in /src/net_processing.cpp for parsing P2P messages

Using the P2P interface

  • P2Ps can be used to test specific P2P protocol behavior. p2p.py contains test framework p2p objects and messages.py contains all the definitions for objects passed over the network (CBlock, CTransaction, etc, along with the network-level wrappers for them, msg_block, msg_tx, etc).

  • P2P tests have two threads. One thread handles all network communication with the bitcoind(s) being tested in a callback-based event loop; the other implements the test logic.

  • P2PConnection is the class used to connect to a bitcoind. P2PInterface contains the higher level logic for processing P2P payloads and connecting to the Bitcoin Core node application logic. For custom behaviour, subclass the P2PInterface object and override the callback methods.

P2PConnections can be used as such:

p2p_conn = node.add_p2p_connection(P2PInterface())
p2p_conn.send_and_ping(msg)

They can also be referenced by indexing into a TestNode's p2ps list, which contains the list of test framework p2p objects connected to itself (it does not include any TestNodes):

node.p2ps[0].sync_with_ping()

More examples can be found in p2p_unrequested_blocks.py, p2p_compactblocks.py.

Prototyping tests

The TestShell class exposes the BitcoinTestFramework functionality to interactive Python3 environments and can be used to prototype tests. This may be especially useful in a REPL environment with session logging utilities, such as IPython. The logs of such interactive sessions can later be adapted into permanent test cases.

Test framework modules

The following are useful modules for test developers. They are located in test/functional/test_framework/.

authproxy.py

Taken from the python-bitcoinrpc repository.

test_framework.py

Base class for functional tests.

util.py

Generally useful functions.

p2p.py

Test objects for interacting with a bitcoind node over the p2p interface.

script.py

Utilities for manipulating transaction scripts (originally from python-bitcoinlib)

key.py

Test-only secp256k1 elliptic curve implementation

blocktools.py

Helper functions for creating blocks and transactions.