47f705b33fc1381d96c99038e2110e6fe2b2f883 tests: add fuzz tests for BitSet (Pieter Wuille)
59a6df6bd584701f820ad60a10d9d477bf0236b5 util: add BitSet (Pieter Wuille)
Pull request description:
Extracted from #30126.
This introduces the `BitSet` data structure, inspired by `std::bitset`, but with a few features that cannot be implemented on top without efficiency loss:
* Finding the first set bit (`First`)
* Finding the last set bit (`Last`)
* Iterating over all set bits (`begin` and `end`).
And a few other operators/member functions that help readability for #30126:
* `operator-` for set subtraction
* `Overlaps()` for testing whether intersection is non-empty
* `IsSupersetOf()` for testing (non-strict) supersetness
* `IsSubsetOf()` for testing (non-strict) subsetness
* `Fill()` to construct a set with all numbers from 0 to n-1, inclusive
* `Singleton()` to construct a set with one specific element.
Everything is tested through a simulation-based fuzz test that compares the behavior with normal `std::bitset` equivalent operations.
ACKs for top commit:
instagibbs:
ACK 47f705b33f
achow101:
ACK 47f705b33fc1381d96c99038e2110e6fe2b2f883
cbergqvist:
re-ACK 47f705b33fc1381d96c99038e2110e6fe2b2f883
theStack:
Code-review ACK 47f705b33fc1381d96c99038e2110e6fe2b2f883
Tree-SHA512: e451bf4b801f193239ee434b6b614f5a2ac7bb49c70af5aba24c2ac0c54acbef4672556800e4ac799ae835632bdba716209c5ca8c37433a6883dab4eb7cd67c1
5bc2077e8f592442b089affdf0b5795fbc053bb8 validation: allow to specify frequency for -checkblockindex (Martin Zumsande)
d5a631b9597e5029a5048d9b8ad84ea4536bbac0 validation: improve performance of CheckBlockIndex (Martin Zumsande)
32c80413fdb063199f3bee719c4651bd63f05fce bench: add benchmark for checkblockindex (Martin Zumsande)
Pull request description:
`CheckBlockIndex() ` are consistency checks that are currently enabled by default on regtest.
The function is rather slow, which is annoying if you
* attempt to run it on other networks, especially if not fully synced
* want to generate a long chain on regtest and see block generation slow down because you forgot to disable `-checkblockindex` or don't know it existed.
One reason why it's slow is that in order to be able to traverse the block tree depth-first from genesis, it inserts pointers to all block indices into a `std::multimap` - for which inserts and lookups become slow once there are hundred thousands of entries.
However, typically the block index is mostly chain-like with just a few forks so a multimap isn't really needed for the most part. This PR suggests to store the block indices of the chain ending in the best header in a vector instead, and store only the rest of the indices in a multimap. This does not change the actual consistency checks that are being performed for each index, just the way the block index tree is stored and traversed.
This adds a bit of complication to make sure each block is visited (note that there are asserts that check it), making sure that the two containers are traversed correctly, but it speeds up the function considerably:
On master, a single invocation of `CheckBlockIndex` takes ~1.4s on mainnet for me (4.9s on testnet which has >2.4 million blocks).
With this branch, the runtime goes down to ~0.27s (0.85s on testnet).This is a speedup by a factor ~5.
ACKs for top commit:
achow101:
ACK 5bc2077e8f592442b089affdf0b5795fbc053bb8
furszy:
ACK 5bc2077e8f592442b089affdf0b5795fbc053bb8
ryanofsky:
Code review ACK 5bc2077e8f592442b089affdf0b5795fbc053bb8. Just added suggested assert and simplification since last review
Tree-SHA512: 6b9c3e3e5069d6152b45a09040f962380d114851ff0f9ff1771cf8cad7bb4fa0ba25cd787ceaa3dfa5241fb249748e2ee6987af0ccb24b786a5301b2836f8487
24bc46c83b39149f4845a575a82337eb46d91bdb cli: Add warning for duplicate port definition (tdb3)
e208fb5d3bea4c1fb750cb0028819635ecdeb415 cli: Sanitize ports in rpcconnect and rpcport (tdb3)
Pull request description:
Adds invalid port detection to bitcoin-cli for -rpcconnect and -rpcport.
In addition to detecting malformed/invalid ports (e.g. those outside of the 16-bit port range, not numbers, etc.), bitcoin-cli also now considers usage of port 0 to be invalid. bitcoin-cli previously considered port 0 to be valid and attempted to use it to reach bitcoind.
Functional tests were added for invalid port detection as well as port prioritization.
Additionally, a warning is provided when a port is specified in both -rpcconnect and -rpcport.
This PR is an alternate approach to PR #27820 (e.g. SplitHostPort is unmodified).
Considered an alternative to 127.0.0.1 being specified in functional tests, but at first glance, this might need an update to test_framework/util.py (e.g. rpc_url), which might be left to a future PR.
ACKs for top commit:
S3RK:
light code review ACK 24bc46c83b39149f4845a575a82337eb46d91bdb
achow101:
ACK 24bc46c83b39149f4845a575a82337eb46d91bdb
cbergqvist:
re ACK 24bc46c83b39149f4845a575a82337eb46d91bdb
Tree-SHA512: c83ab6a30a08dd1ac8b368a7dcc2b4f23170f0b61dd67ffcad7bcda05096d333bcb9821fba11018151f55b2929c0a333bfec15b8bb863d83f41fc1974c6efca5
09ef322acc0a88a9e119f74923399598984c68f6 [[refactor]] Check CTxMemPool options in constructor (TheCharlatan)
Pull request description:
The tests should run the same checks on the mempool options that the init code also applies. The downside to this patch is that the log line may now be printed more than once in the for loop.
This was originally noticed here https://github.com/bitcoin/bitcoin/pull/25290#discussion_r900272797.
ACKs for top commit:
stickies-v:
re-ACK 09ef322acc0a88a9e119f74923399598984c68f6 . Fixed unreachable assert and updated docstring, and also added an exception for "-maxmempool must be at least " in the `tx_pool` fuzz test, which makes sense when looking at how the mempool options are constructed in `SetMempoolConstraints`.
achow101:
ACK 09ef322acc0a88a9e119f74923399598984c68f6
ryanofsky:
Code review ACK 09ef322acc0a88a9e119f74923399598984c68f6. Just fuzz test error checking fix and updated comment since last review
Tree-SHA512: eb3361411c2db70be17f912e3b14d9cb9c60fb0697a1eded952c3b7e8675b7d783780d45c52e091931d1d80fe0f0280cee98dd57a3100def13af20259d9d1b9e
7cbfd7a7ce45ac68d6041f42f468862f5c193d8c refactor: rename (macho) ld64 to lld (fanquake)
d851451705aa67175cbafb06f455e6969131f90f ci: update deps for macOS cross build (fanquake)
9ebdd5e9e04be3355309d3096da00db37b4d540c depends: update install docs for macOS cross compilation (fanquake)
fb74fd66cb2b016f07e47dc58d81b33034ce851b depends: remove no-longer used llvm_* vars from macOS build (fanquake)
9043f12425d869d8bf7f2be1bb12368e41d09c5d depends: no-longer pass -B to clang in macOS cross-compile (fanquake)
f9994b025e1d72b4c0f5267de54caa2bfae136a3 depends: remove native LLVM package (fanquake)
e9a44faf14854c5b05a297df182d10706f438a83 depends: remove FORCE_USE_SYSTEM_CLANG (fanquake)
9946618f6174422d9ba93bf96c278269bebacf39 guix: use clang-toolchain-18 for macOS build (fanquake)
Pull request description:
Remove `FORCE_USE_SYSTEM_CLANG` in favour of always using the system Clang and lld for macOS cross-compilation; rather than downloading precompiled blobs.
For example, anyone using Ubuntu 24.04 should be able to `apt install clang llvm lld .. etc`, and then cross-compile for macOS using:
```bash
# clang --version
Ubuntu clang version 18.1.3 (1)
make -C depends HOST=arm64-apple-darwin FORCE_USE_SYSTEM_CLANG=1
./autogen.sh
CONFIG_SITE=/path/to/depends/arm64-apple-darwin/share/config.site ./configure
make
# file src/qt/bitcoin-qt
src/qt/bitcoin-qt: Mach-O 64-bit arm64 executable, flags:<NOUNDEFS|DYLDLINK|TWOLEVEL|WEAK_DEFINES|BINDS_TO_WEAK|PIE|HAS_TLV_DESCRIPTORS>
```
Note that the minimum supported version of Clang we will support for macOS cross-compilation will likely be more recent than our other minimum supported version of Clang, due to compiler/linker option usage.
ACKs for top commit:
Sjors:
tACK 7cbfd7a7ce45ac68d6041f42f468862f5c193d8c
theuni:
ACK 7cbfd7a7ce45ac68d6041f42f468862f5c193d8c
TheCharlatan:
Nice, ACK 7cbfd7a7ce45ac68d6041f42f468862f5c193d8c
Tree-SHA512: 1499e29b3b238c5c85278c38e8fb6bb5e7883db3443f62b6bf397c5d761bedbc054962be645a9defce15266f0a969bb4b3ccd28b6e4dd874472857b928f185d1
ab98e6fd03970d6b5a593674c84e762a47b90ea6 test: add coverage for errors for `combinerawtransaction` RPC (brunoerg)
Pull request description:
This PR adds test coverage for the following errors for the `combinerawtransaction` RPC:
* Tx decode failed
* Missing transactions
* Input not found or already spent
For reference: https://maflcko.github.io/b-c-cov/total.coverage/src/rpc/rawtransaction.cpp.gcov.html
ACKs for top commit:
maflcko:
lgtm ACK ab98e6fd03970d6b5a593674c84e762a47b90ea6
tdb3:
ACK ab98e6fd03970d6b5a593674c84e762a47b90ea6
Tree-SHA512: 8a133c25dad2e1b236e0278a88796f60f763e3fd6fbbc080f926bb23f9dcc55599aa242d6e0c4ec15a179d9ded10a1f17ee5b6063719107ea84e6099f10416b2
0000276b31cea5e443a59d94a98c569293ada951 test: Remove redundant verack check (MarcoFalke)
Pull request description:
Currently the sync in `connect_nodes` mentions the `version` and `verack` message types, but only checks the `verack`. Neither check is required, as the `pong` check implies both. In case of failure, the debug log will have to be consulted anyway, so the redundant check doesn't add value.
Also clarify in the comments that the goal is to check the flag `fSuccessfullyConnected` indirectly.
ACKs for top commit:
furszy:
utACK 0000276b31ce
brunoerg:
ACK 0000276b31cea5e443a59d94a98c569293ada951
tdb3:
ACK 0000276b31cea5e443a59d94a98c569293ada951
Tree-SHA512: f9ddcb1436d2f70da462a8dd470ecfc90a534dd6507c23877ef7626e7c02326c077001a42ad0171a87fba5c5275d1970d8c5e5d82c56c8412de944856fdfd6db
39d135e79f3f0c40dfd8fad2c53723d533cd19b4 test: MiniWallet: respect fee_rate for target_weight, use in mempool_limit.py (Sebastian Falbesoner)
b2f0a9f8b0776d49ef1639310311ca50435a2a0a test: add framework functional test for MiniWallet's tx padding (Sebastian Falbesoner)
c17550bc3a68faa1eb82d1bdb767b41d8cd85a6b test: MiniWallet: fix tx padding (`target_weight`) for large sizes, improve accuracy (Sebastian Falbesoner)
Pull request description:
MiniWallet allows to create padded transactions that are equal or slightly above a certain `target_weight` (first introduced in PR #25379, commit 1d6b438ef0ccd05e1522ac38b44f847c1d93e72f), which can be useful especially for mempool-related tests, e.g. for policy limit checks or scenarios to trigger mempool eviction. Currently the `target_weight` parameter doesn't play together with `fee_rate` though, as the fee calculation is incorrectly based on the tx vsize before the padding output is added, so the fee-rate is consequently far off. This means users are forced to pass an absolute fee, which can be quite inconvenient and leads to lots of duplicated "calculate absolute fee from fee-rate and vsize" code with the pattern `fee = (feerate / 1000) * (weight // 4)` on the call-sites.
This PR first improves the tx padding itself to be more accurate, adds a functional test for it, and fixes the `fee_rate` treatment for the `{create,send}_self_transfer` methods. (Next step would be to enable this also for the `_self_transfer_multi` methods, but those currently don't even offer a `fee_rate` parameter). Finally, the ability to pass both `target_weight` and `fee_rate` is used in the `mempool_limit.py` functional test. There might be more use-cases in other tests, that could be done in a follow-up.
ACKs for top commit:
rkrux:
tACK [39d135e](39d135e79f)
ismaelsadeeq:
Code Review ACK 39d135e79f3f0c40dfd8fad2c53723d533cd19b4 🚀
glozow:
light review ACK 39d135e79f3f0c40dfd8fad2c53723d533cd19b4
Tree-SHA512: 6bf8e853a921576d463291d619cdfd6a7e74cf92f61933a563800ac0b3c023a06569b581243166906f56b3c5e8858fec2d8a6910d55899e904221f847eb0953d
d1581c6048478cf70c5fb9ec5ebc178f16b376b8 test: doc: fix units in tx size standardness test (s/vbytes/weight units) (Sebastian Falbesoner)
Pull request description:
This small fixup PR is a late follow-up for #17947 (commit 4537ba5f21ad8afb705325cd8e15dd43877eb28f), where the wrong units has been used in the comments for the large tx composition.
ACKs for top commit:
tdb3:
ACK d1581c6048478cf70c5fb9ec5ebc178f16b376b8
ismaelsadeeq:
ACK d1581c6048478cf70c5fb9ec5ebc178f16b376b8
glozow:
ACK d1581c6048478cf70c5fb9ec5ebc178f16b376b8
Tree-SHA512: ea2de42174f9dca0608275ea377c852ebddc5a04a2b32248ce808aea33d7e00cdee3a225b24c0cf426c69646cccbbc31273c62f7bc1647bb3443a61de3b15670
f68cba29b3be0dec7877022b18a193a3b78c1099 blockman: Replace m_reindexing with m_blockfiles_indexed (Ryan Ofsky)
1b1c6dcca0cc891bd35d29b61628c39098cd94ce test: Add functional test for continuing a reindex (TheCharlatan)
201c1a92824c71ae646d5bba9963871b1d704cc1 indexes: Don't wipe indexes again when already reindexing (TheCharlatan)
804f09dfa116300914e2aeef05ed9710dd504e8c kernel: Add less confusing reindex options (Ryan Ofsky)
e17255322378076edce3ef6f06cd36ca58d2e236 validation: Remove needs_init from LoadBlockIndex (TheCharlatan)
533eab7d67d78f217f74909662133086b79ea808 bugfix: Streamline setting reindex option (TheCharlatan)
Pull request description:
When restarting `bitcoind` during an ongoing reindex without setting the `-reindex` flag again, the block and coins db is left intact, but any data from the optional indexes is discarded. While not a bug per se, wiping the data again is
wasteful, both in terms of having to write it again, as well as potentially leading to longer startup times. So keep the index data instead when continuing a prior reindex.
Also includes a bugfix and smaller code cleanups around the reindexing code. The bug was introduced in b47bd959207e82555f07e028cc2246943d32d4c3: "kernel: De-globalize fReindex".
ACKs for top commit:
stickies-v:
ACK f68cba29b3be0dec7877022b18a193a3b78c1099
fjahr:
Code review ACK f68cba29b3be0dec7877022b18a193a3b78c1099
furszy:
Code review ACK f68cba29b3be0dec7877022b18a193a3b78c1099
ryanofsky:
Code review ACK f68cba29b3be0dec7877022b18a193a3b78c1099. Only changes since last review were cherry-picking suggested commits that rename variables, improving comments, and making some tweaks to test code.
Tree-SHA512: b252228cc76e9f1eaac56d5bd9e4eac23408e0fc04aeffd97a85417f046229364673ee1ca7410b9b6e7b692b03f13ece17c42a10176da0d7e975a8915deb98ca
This adds a bitset module that implements a BitSet<N> class, a variant
of std::bitset with a few additional features that cannot be implemented
in a wrapper without performance loss (specifically, finding first and
last bit set, or iterating over all set bits).
fa780e1c25e8e98253e32d93db65f78a0092433f build: Remove --enable-gprof (MarcoFalke)
Pull request description:
It is unclear what benefit this option has, given that:
* `gprof` requires re-compilation (`perf` and other tools can instead be used on existing executables)
* `gprof` requires hardening to be disabled
* `gprof` doesn't work with `clang`
* `perf` is documented in the dev-notes, and test notes, and embedded into the functional test framework; `gprof` isn't
* Anyone really wanting to use it could pass the required flags to `./configure`
* I couldn't find any mention of the use of `gprof` in the discussions in this repo, apart from the initial pull request adding it (cfaac2a60f3ac63ae8deccb03d88bd559449b78c)
* Keeping it means that it needs to be maintained and ported to CMake
Fix all issues by removing it.
ACKs for top commit:
TheCharlatan:
ACK fa780e1c25e8e98253e32d93db65f78a0092433f
hebasto:
ACK fa780e1c25e8e98253e32d93db65f78a0092433f, I have reviewed the code and it looks OK.
willcl-ark:
crACK fa780e1c25e8e98253e32d93db65f78a0092433f
Tree-SHA512: 0a9ff363ac2bec8b743878a4e3147f18bc16823d00c5007568432c36320bd0199b13b6d0ce828a9a83c2cc434c058afaa64eb2eccfbd93ed85b81ce10c41760c
0d3ef83433805d3f367130fd5bd227a8ed5a7ccd ci: Use relative paths in `win64-native` CI job consistently (Hennadii Stepanov)
501aceefcf7536cbdead5bcb53b13f2fec7ab6be ci: Remove no longer needed workaround for GHA Windows images (Hennadii Stepanov)
Pull request description:
This PR:
1. Removes no longer needed workaround for GHA Windows images.
GHA Windows images previously had multiple VC Build Tools installed, which required specifying the `VCPKG_PLATFORM_TOOLSET_VERSION` explicitly to avoid linker errors. This issue has been resolved as per
https://github.com/actions/runner-images/issues/9701.
2. Switches all references to temporary files to relative ones for consistency and readability.
ACKs for top commit:
sipsorcery:
ACK 0d3ef83433805d3f367130fd5bd227a8ed5a7ccd.
maflcko:
ACK 0d3ef83433805d3f367130fd5bd227a8ed5a7ccd
Tree-SHA512: e832b87fc6dee1e9d1eb452797f16b732e776c2ecdbe3dc9e64cc48ce9b5b89c569d5b96b999423ae1261ff4bf684b7003af84d8024ef5260682f531c4e8ff5e
15796d4b61342f75548b20a18c670ed21d102ba8 build: warn on self-assignment (Cory Fields)
53372f21767be449bb452fc3f5fe7f16286ae371 refactor: disable self-assign warning for tests (Cory Fields)
Pull request description:
Belt-and suspenders after #30234. Self-assignment should be safe _and_ discouraged.
We used to opt out of this warning because something deep in our serialization/byteswapping code could self-assign, but that doesn't appear to be the case anymore.
ACKs for top commit:
maflcko:
ACK 15796d4b61342f75548b20a18c670ed21d102ba8
fanquake:
ACK 15796d4b61342f75548b20a18c670ed21d102ba8 - not a huge fan of inline pragma usage, but this seems fine, given it's to work around an already-fixed compiler bug, and we'll only be carrying it for a shortish time in any case.
Tree-SHA512: 1f95f7c730b974ad1da55ebd381040bac312f2f380fff9d569ebab91d7c1963592a84d1613d81d96238c6f5a66aa40deebba68a76f6b24b02150d0a77c769654
fab01b5220c28a334b451ed9625bd3914c48e6af refactor: performance-for-range-copy in psbt.h (MarcoFalke)
Pull request description:
A copy of the partial signatures is not required before serializing them.
For reference, this was found by switching the codebase to `std::span` (not sure why it wasn't found with `Span` though):
```
./psbt.h:240:23: error: loop variable is copied but only used as const reference; consider making it a const reference [performance-for-range-copy,-warnings-as-errors]
240 | for (auto sig_pair : partial_sigs) {
| ^
| const &
ACKs for top commit:
tdb3:
ACK fab01b5220c28a334b451ed9625bd3914c48e6af
theStack:
utACK fab01b5220c28a334b451ed9625bd3914c48e6af
Tree-SHA512: b55513d8191118499716684190ee568d171b50ae9171d246ca6e047f0cfd3ec14c9453d721e88af55e47bb41fa66cbafdbfb47bc5f9b8d82789e0a9b634b371b
1f6ab1215bbb1f8a5f1743c3c413b95ad08090df minor: remove unnecessary semicolons from RPC content type examples (Matthew Zipkin)
b22529529823c0cb5916ac318c8536e9107b7e78 test: use json-rpc 2.0 in all functional tests by default (Matthew Zipkin)
391843b0297db03d71a8d88ab77609e2ad230bf2 bitcoin-cli: use json-rpc 2.0 (Matthew Zipkin)
d39bdf339772166a5545ae811e58b7764af093a8 test: remove unused variable in interface_rpc.py (Matthew Zipkin)
0ead71df8c83a2f9eae1220544ec84dcf38a0326 doc: update and link for JSON-RPC 2.0 (Matthew Zipkin)
Pull request description:
This is a follow-up to #27101.
- Addresses [post-merge comments ](https://github.com/bitcoin/bitcoin/pull/27101#discussion_r1606723428)
- bitcoin-cli uses JSON-RPC 2.0
- functional tests use JSON-RPC 2.0 by default (exceptions are in the regression tests added by #27101)
ACKs for top commit:
tdb3:
ACK 1f6ab1215bbb1f8a5f1743c3c413b95ad08090df
cbergqvist:
ACK 1f6ab1215bbb1f8a5f1743c3c413b95ad08090df
Tree-SHA512: 49bf14c70464081280216ece538a2f5ec810bac80a86a83ad3284f0f1b017edf755a1a74a45be279effe00218170cafde7c2de58aed07097a95c2c6b837a6b6c
2599655c1fb8e7d0b8407d2be249977372cb73ff guix: bump time-machine to f0bb724211872cd6158fce6162e0b8c73efed126 (fanquake)
Pull request description:
Includes:
LLVM 18.1.x (#30201)
GCC 13.x (#29881)
git-minimal 2.41.0 -> 2.45.1
Kernel Headers 6.1.80 -> 6.1.92
moreutils 0.68 -> 0.69
Commits like https://git.savannah.gnu.org/cgit/guix.git/commit/?id=7b0f145802f0c2c785014293d748721678fef824, which should improve the bootstrap situation (#30042). This can somewhat be visualised by comparing the (simplified) dependencies of guix itself, between the two time-machines.
Master:
PR:
Note that in the case of this PR, we are better off, no-longer having to build a number of tex packages, ruby, cairo, graphics libs, openssl 1.x etc.
ACKs for top commit:
TheCharlatan:
ACK 2599655c1fb8e7d0b8407d2be249977372cb73ff
Tree-SHA512: 9c5675a5d563c17744c89c8a392bc7865aa1c9e71a5e3044c23f31e51458dac28c0c238d2055c86dc732df595dae60bcbc8b85266b23b7991c4c770d56f7d23a
This is a just a mechanical change, renaming and inverting the meaning
of the indexing variable.
"m_blockfiles_indexed" is a more straightforward name for this variable
because this variable just indicates whether or not
<datadir>/blocks/blk?????.dat files have been indexed in the
<datadir>/blocks/index LevelDB database. The name "m_reindexing" was
more confusing, it could be true even if -reindex was not specified, and
false when it was specified. Also, the previous name unnecessarily
required thinking about the whole reindexing process just to understand
simple checks in validation code about whether blocks were indexed.
The motivation for this change is to follow up on previous commits,
moving away from having multiple variables called "reindex" internally,
and instead naming variables individually after what they do and
represent.
Before this change continuing a reindex without the -reindex flag set
would leave the block and coins db intact, but discard the data of the
optional indexes. While not a bug per se, wiping the data again is
wasteful, both in terms of having to write it again, and potentially
leading to longer startup times.
When initially running a reindex, both the block index and any further
activated indexes are wiped. On an index's Init(), both the best block
stored by the index and the chain's tip are null. An index's m_synced
member is therefore true. This means that it will process blocks through
validation events while the reindex is running.
Currently, if the reindex is continued without the user re-specifying
the reindex flag, the block index is preserved but further index data is
wiped. This leads to the stored best block being null, but the chain tip
existing. The m_synced member will be set to false. The index will not
process blocks through the validation interface, but instead use the
background sync once the reindex is completed.
If the index is preserved (this change) after a restart its best block
may potentially match the chain tip. The m_synced member will be set to
true and the index can process validation events during the rest of the
reindex.
Drop confusing kernel options:
BlockManagerOpts::reindex
ChainstateLoadOptions::reindex
ChainstateLoadOptions::reindex_chainstate
Replacing them with more straightforward options:
ChainstateLoadOptions::wipe_block_tree_db
ChainstateLoadOptions::wipe_chainstate_db
Having two options called "reindex" which did slightly different things
was needlessly confusing (one option wiped the block tree database, and
the other caused block files to be rescanned). Also the previous set of
options did not allow rebuilding the block database without also
rebuilding the chainstate database, when it should be possible to do
those independently.
30a01134cdec37e7467fcd6eee8b0ae3890a131c [doc] update bips.md for 431 (glozow)
9dbe6a03f0d6e70ccdf8e8715f888c0c17216bee [test] wallet uses CURRENT_VERSION which is 2 (glozow)
539404fe0fc0346b3aa77c330b38a5a0ad6565b2 [policy] make v3 transactions standard (glozow)
052ede75aff5c9f3a0a422ef413852eabeecc665 [refactor] use TRUC_VERSION in place of 3 (glozow)
Pull request description:
Make `nVersion=3` (which is currently nonstandard on mainnet) standard.
Note that we will treat these transactions as Topologically Restricted Until Confirmation (TRUC). Spec is in BIP 431 and implementation is in #28948, #29306, and #29873
See #27463 for overall project tracking, and #29319 for information about relevance to cluster mempool.
ACKs for top commit:
sdaftuar:
utACK 30a01134c
achow101:
ACK 30a01134cdec37e7467fcd6eee8b0ae3890a131c
instagibbs:
utACK 30a01134cdec37e7467fcd6eee8b0ae3890a131c
murchandamus:
ACK 30a01134cdec37e7467fcd6eee8b0ae3890a131c
ismaelsadeeq:
ACK 30a01134cdec37e7467fcd6eee8b0ae3890a131c 🛰️
Tree-SHA512: 2a4aec0442c860e792a061d83e36483c1f1b426f946efbdf664c8db97a596e498b535707e1d3a900218429486ea69fd4552e3d476526a6883cbd5556c6534b48
7b8eea067f188c0b0e52ef21b01aedd37667a237 tests: add fuzz tests for VecDeque (Pieter Wuille)
62fd24af6a3fe1569662c2802f59bb68a0172087 util: add VecDeque (Pieter Wuille)
Pull request description:
Extracted from #30126.
This adds a `VecDeque` data type, inspired by `std::deque`, but backed by a single allocated memory region used as a ring buffer instead of a linked list of arrays. This gives better memory locality and less allocation overhead, plus better guarantees (some C++ standard library implementations, though not libstdc++ and libc++, use a separate allocation per element in a deque).
It is intended for the candidate set search queue in #30126, but may be useful as a replacement for `std::deque` in other places too. It's not a full drop-in replacement, as I did not add iteration support which is unnecessary for the intended use case, but nothing prevents adding that if needed.
Everything is tested through a simulation-based fuzz test that compares the behavior with normal `std::deque` equivalent operations, both for trivially-copyable/destructible types and others.
ACKs for top commit:
instagibbs:
reACK 7b8eea067f
cbergqvist:
re-ACK 7b8eea067f188c0b0e52ef21b01aedd37667a237
hebasto:
re-ACK 7b8eea067f188c0b0e52ef21b01aedd37667a237, I've verified changes since my recent [review](https://github.com/bitcoin/bitcoin/pull/30161#pullrequestreview-2103018546) with
glozow:
ACK 7b8eea067f
Tree-SHA512: 1b62f3ba1a43a1293d8c9de047e2399442e74c46de2df81406151fe27538716ce265f35fb6779ee56d77a39cddf8fb4b4e15bda8f04ebf3b149e2f05fa55cb21
It does not control any actual logic and the log message as well as the
comment are obsolete, since no database initialization takes place there
anymore. Log messages indicating when indexes and chainstate databases
are loaded exist in other places.
Reverts a bug introduced in b47bd959207e82555f07e028cc2246943d32d4c3
"kernel: De-globalize fReindex". The change leads to a GUI user being
prompted to re-index on a chainstate loading failure more than once as
well as the node actually not reindexing if the user chooses to. Fix
this by setting the reindexing option instead of the atomic, which can
be safely re-used to indicate that a reindex should be attempted.
The bug specifically is caused by the chainman, and thus the blockman
and its m_reindexing atomic being destroyed on every iteration of
the for loop.
The reindex option for ChainstateLoadOptions is currently also set in a
confusing way. By using the reindex atomic, it is not obvious in which
scenario it is true or false.
The atomic is controlled by both the user passing the -reindex option,
the user chosing to reindex if something went wrong during chainstate
loading when running the gui, and by reading the reindexing flag from
the block tree database in LoadBlockIndexDB. In practice this read is
done through the chainstate module's CompleteChainstateInitialization's
call to LoadBlockIndex. Since this is only done after the reindex option
is set already, it does not have an effect on it.
Make this clear by using the reindex option from the blockman opts which
is only controlled by the user.
GHA Windows images previously had multiple VC Build Tools installed,
which required specifying the `VCPKG_PLATFORM_TOOLSET_VERSION`
explicitly to avoid linker errors. This issue has been resolved as per
https://github.com/actions/runner-images/issues/9701.
fa52e13ee81fcc7543890dbd6986fcb55168583f test: Remove struct.pack from almost all places (MarcoFalke)
fa826db477a981b48bff53021f9695a5f6682dc0 scripted-diff: test: Use int.to_bytes over struct packing (MarcoFalke)
faf2a975ad46799d075e3a70c699da0d8182aab9 test: Use int.to_bytes over struct packing (MarcoFalke)
faf3cd659a72473a1aa73c4367a145f4ec64f146 test: Normalize struct.pack format (MarcoFalke)
Pull request description:
`struct.pack` has many issues:
* The format string consists of characters that may be confusing and may need to be looked up in the documentation, as opposed to using easy to understand self-documenting code.
This lead to many test bugs, which weren't hit, which is fine, but still confusing. Ref: https://github.com/bitcoin/bitcoin/pull/29400, https://github.com/bitcoin/bitcoin/pull/29399, https://github.com/bitcoin/bitcoin/pull/29363, fa3886b7c69cbbe564478f30bb2c35e9e6b1cffa, ...
Fix all issues by using the built-in `int` helpers `to_bytes` via a scripted diff.
Review notes:
* For `struct.pack` and `int.to_bytes` the error behavior is the same, although the error messages are not identical.
* Two `struct.pack` remain. One for float serialization in a C++ code comment, and one for native serialization.
ACKs for top commit:
achow101:
ACK fa52e13ee81fcc7543890dbd6986fcb55168583f
rkrux:
tACK [fa52e13](fa52e13ee8)
theStack:
Code-review ACK fa52e13ee81fcc7543890dbd6986fcb55168583f
Tree-SHA512: ee80d935b68ae43d1654b047e84ceb80abbd20306df35cca848b3f7874634b518560ddcbc7e714e2e7a19241e153dee64556dc4701287ae811e26e4f8c57fe3e
This is an STL-like container that interface-wise looks like std::deque, but
is backed by a (fixed size, with vector-like capacity/reserve) circular buffer.
