Having `@par title` followed by an empty line renders improperly in
Doxygen - it results in a paragraph with a title but without a body.
https://www.doxygen.nl/manual/commands.html#cmdpar
This also results in a compiler warning (or error) with Clang 19:
```
./txmempool.h:368:34: error: empty paragraph passed to '@par' command [-Werror,-Wdocumentation]
368 | * @par Consistency guarantees
| ~~~~~~~~~~~~~~~~~~~~~~~~~~^
1 error generated.
```
fa33a63bd9 fuzz: Speed up PickValue in txorphan (MarcoFalke)
Pull request description:
`PickValue` will advance a begin iterator on the `outpoints` set, which is expensive, because it only has a `++` operator. As it is called in a loop of `num_in` (~`outpoints.size()`), the runtime is `O(outpoints.size() ^ 2)`.
Fix it by making the runtime linear.
ACKs for top commit:
glozow:
ACK fa33a63bd9, thanks for taking the suggestion
dergoegge:
utACK fa33a63bd9
Tree-SHA512: 33f440d97c6834d907d43a8d29e4fb2c995f0d244460bd079af100f13d3607a53e44a0db52f4eb5c487d98df0ff4f2f6d987bf94b922ae9f4506f1295ad6214c
8c935e625e depends: Fix CMake-generated `libevent*.pc` files (Hennadii Stepanov)
Pull request description:
Broken out of #30454. This is a backport of the merged upstream PR: https://github.com/libevent/libevent/pull/1622.
Note that after #29835 we might end up dropping pkg-config and using the installed CMake files directly, but that depends on whether or not enough distros actually ship those files.
Either way, having fixed up .pc files won't hurt.
ACKs for top commit:
hebasto:
ACK 8c935e625e.
fanquake:
ACK 8c935e625e
Tree-SHA512: 259c2ad78fb9e90370a7205dc71c40acda1a872f6509435133bc1c4c2c3de57366e80679aa083e13ed85e7966883dc470c0147ee171a2ed0171a18cd5ffc99b3
fa7bee13bf lint: Use git clone --depth=1 (MarcoFalke)
fadb7c2a91 lint: Add missing docker.io prefix to ci/lint_imagefile (MarcoFalke)
Pull request description:
Currently, the `ci/lint_imagefile` may pick the wrong (non-native) architecture due to the missing prefix.
For example, assuming the user has previously pulled an s390x image:
```
$ podman run --rm 'docker.io/s390x/debian:bookworm' dpkg --print-architecture
exec /usr/bin/dpkg: exec format error
```
Now, `debian:bookworm` will refer to the same image:
```
$ podman run --rm 'debian:bookworm' dpkg --print-architecture
exec /usr/bin/dpkg: exec format error
```
However, `docker.io/debian:bookworm` works fine:
```
$ podman run --rm 'docker.io/debian:bookworm' dpkg --print-architecture
arm64
```
(Also includes a nit-fix from https://github.com/bitcoin/bitcoin/pull/30499#discussion_r1686470495)
ACKs for top commit:
paplorinc:
utACK fa7bee13bf
hebasto:
ACK fa7bee13bf.
Tree-SHA512: 4b6d562c14c67bef984ad25f6a3a1ef7f1059dc2859c603c45083b36bcacafa3248fc74176e2e4626fdc39507e9353f458ddbc4077f805c03e970df46af02224
0388ad0d65 depends: switch zmq to CMake (Cory Fields)
fefb3bbe5b depends: add zeromq no librt patch (fanquake)
a522ef1542 depends: add zeromq cmake minimum patch (fanquake)
cbbc229adf depends: add zeromq windows usage patch (fanquake)
2de68d6d38 depends: add zeromq builtin sha1 patch (fanquake)
0c8605253a depends: add zeromq mktemp macos patch (fanquake)
Pull request description:
This picks up a change, which is a switch to building zeromq with CMake. It includes a number of patches, some which have already been upstreamed (see each patch for details).
ACKs for top commit:
hebasto:
ACK 0388ad0d65.
Tree-SHA512: 5567e432b4e4e0446c41d502bd61810a80b329dea2399b5d9d9f6e79acc450d1c6ba861c8238ba895de98338cfc5dc44ad2bf86ee8c222ecb3fbf47d6eb60da4
fa8d73e86e lint: Use consistent out-of-tree build for python and test_runner (MarcoFalke)
fa0f859885 doc: Clarify intent of ./ci/lint_run_all.sh (MarcoFalke)
fa9ad59f87 lint: Use $CI_RETRY_EXE when building ./ci/lint_imagefile (MarcoFalke)
Pull request description:
Fixes https://github.com/bitcoin/bitcoin/issues/30496
Seems odd to sometimes do an out-of-tree build (via `./ci/lint_imagefile`, see `test/lint/README.md`) and sometimes not (via Cirrus CI, see `./ci/lint_run_all.sh`).
Fix it by doing an out-of-tree build consistently in the same location.
Also, fix `$CI_RETRY_EXE`, while touching this.
ACKs for top commit:
josibake:
utACK fa8d73e86e
willcl-ark:
utACK fa8d73e86e
paplorinc:
utACK fa8d73e86e
Tree-SHA512: 4181ca14299a798850f5e05f180f3305a3378081ca8dabf6ab2da6115997cc17f6ef0f10db9b2b31618e59231083e5c4a971432d27b4d77903e655be21155abb
Previous code was confusing and brittle. For example, the full import
"source ./ci/test/00_setup_env.sh" and $PATH overwrite was not needed.
Fix it by simply copying the exe to /ci_retry and use that in
$CI_RETRY_EXE.
This is also a fix, because previously ci/lint_imagefile did use an
empty $CI_RETRY_EXE.
a8e3af1a82 qa: Do not assume running `feature_asmap.py` from source directory (Hennadii Stepanov)
9bf7ca6cad qa: Consider `cache` and `config.ini` relative to invocation directory (Hennadii Stepanov)
a0473442d1 scripted-diff: Add `__file__` argument to `BitcoinTestFramework.init()` (Hennadii Stepanov)
Pull request description:
This PR includes changes split from https://github.com/bitcoin/bitcoin/pull/30454. They improve the functional test framework, allowing users to [run individual functional tests](https://github.com/hebasto/bitcoin/issues/146) from the build directory in the new CMake-based build system.
This functionality is not available for out-of-source builds using the current Autotools-based build system, which always requires write permissions for the source directory. Nevertheless, this PR can be tested as suggested in https://github.com/bitcoin/bitcoin/pull/30463#issuecomment-2232618421:
1. Make an out-of-source build:
```
$ ./autogen.sh
$ mkdir ../build && cd ../build
$ ../bitcoin/configure
$ make
```
2. Create a symlink in the build directory to a functional test:
```
$ ln --symbolic ../../../bitcoin/test/functional/wallet_disable.py ./test/functional/
```
3. Run this symlink:
```
$ ./test/functional/wallet_disable.py
```
The last command fails on the master branch:
```
Traceback (most recent call last):
File "/home/hebasto/git/build/./test/functional/wallet_disable.py", line 31, in <module>
DisableWalletTest().main()
^^^^^^^^^^^^^^^^^^^
File "/home/hebasto/git/bitcoin/test/functional/test_framework/test_framework.py", line 106, in __init__
self.parse_args()
File "/home/hebasto/git/bitcoin/test/functional/test_framework/test_framework.py", line 210, in parse_args
config.read_file(open(self.options.configfile))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
FileNotFoundError: [Errno 2] No such file or directory: '/home/hebasto/git/bitcoin/test/config.ini'
```
and succeeds with this PR.
ACKs for top commit:
maflcko:
tested ACK a8e3af1a82🎨
glozow:
ACK a8e3af1a82, tested with the steps in op
stickies-v:
ACK a8e3af1a82
Tree-SHA512: 899e4efc09edec13ea3f5b47825d03173fb21d3569c360deda7fa6a56b99b4d24e09ad4f0883bad1ee926b1c706e47ba07c6a6160c63c07c82b3cf4ae5816e91
7703884ab1 Fix MSVC warning C4273 "inconsistent dll linkage" (Hennadii Stepanov)
Pull request description:
Broken out of https://github.com/bitcoin/bitcoin/pull/30454.
When using CMake, the user can select the MSVC runtime library to be:
1) Statically-linked (with the corresponding `x64-windows-static` vcpkg triplet) or
2) Dynamically-linked (with the corresponding `x64-windows` vcpkg triplet)
In the latter case, the compiler emits the [C4273](https://learn.microsoft.com/en-us/cpp/error-messages/compiler-warnings/compiler-warning-level-1-c4273) warning.
As the "Necessary on some platforms" comment does not apply to MSVC, skip the declaration for MSVC.
The MSVC build system in the master branch supports the statically-linked runtime only: ed739d14b5/build_msvc/common.init.vcxproj.in (L65)
ACKs for top commit:
sipa:
utACK 7703884ab1
sipsorcery:
utACK 7703884ab1.
theuni:
utACK 7703884ab1
Tree-SHA512: a42e1a0d48973217462e703c418f3e9ef9cb5236267c1bf32912aacaf68976cdd2b9229168523f7c2a99ee3f2fb1bf8add4f342796bdb1e4063ca026b761db51
d318c4ef56 depends: bump libmultiprocess for CMake fixes (Cory Fields)
Pull request description:
Broken out of #30454 . Bumped [even further](4883197abc (r1684802528)) after https://github.com/chaincodelabs/libmultiprocess/pull/98 was merged upstream.
hebasto Presumably this approach works now with the CMake branch?
ACKs for top commit:
ryanofsky:
Code review ACK d318c4ef56.
Tree-SHA512: 4b5491f73c0063d09e8339829cd831b1f4c441dd7b55a22037c9337c80cce19bb00a3e5cf925efa77d6d4e89ab45482f40f5799bc14948f8cabcbad3c3549430
When using CMake, the user can select the MSVC runtime library to be:
1) Statically-linked (with the corresponding `x64-windows-static` vcpkg
triplet) or
2) Dynamically-linked (with the corresponding `x64-windows` vcpkg
triplet)
In the latter case, the compiler emits the C4273 warning.
As the "Necessary on some platforms" comment does not apply to MSVC,
skip the declaration for MSVC.
The CMake WIN32_WINNT autodetection is broken, and must be set
manually. We may want to set is explicitly in any case, but the
brokenness should also be fixed upstream.
Also patch out depends paths, that would cause non-determinism.
Co-authored-by: fanquake <fanquake@gmail.com>
Adds more testing options for creating an environment without networking
and a validation interface. This is useful for improving the performance
of the utxo snapshot fuzz test, which constructs a new TestingSetup on
each iteration.
ff4f3deb7b depends: use CMake to build FreeType (fanquake)
Pull request description:
Switches Freetype to be built with CMake.
ACKs for top commit:
theuni:
ACK ff4f3deb7b
hebasto:
ACK ff4f3deb7b, I've verified the actual compile options, they look sane.
Tree-SHA512: e9e4348975998539fde88a84d110d53dbac50ae9cc3fa692d15e09313d6fdb6acb3bb23533786a645fc836091075b4487d6de42ef78ba3a44de46d06360aef4f
Also pulls out the guarding assert and calls it explicitly before the test function is called. This is already done before the existing call of the test function so it was not needed there.
- "Valid snapshot file, but referencing a snapshot block that turns out
to be invalid, or has an invalid parent" has been addressed in #30267
- "An ancestor of snapshot block" - If chain tip refers to blocks in this context then any successful load is addressing this because if we have synced past the snapshot base block we fail because we don't need assumeutxo anymore. And if this is about headers then this is the `test_headers_not_synced()` case.
- "A descendant of the snapshot block" - If this refers to blocks the
`test_snapshot_with_less_work()` addressed this and if it is just headers in this case again it would be represented in all of the successful loads in the test.
Co-authored-by: Alfonso Roman Zubeldia <alfonsoromanz24@gmail.com>
23333b7ed2 net: Allow DNS lookups on nodes with IPV6 lo only (Max Edwards)
Pull request description:
This is similar to (but does not fix) https://github.com/bitcoin/bitcoin/issues/13155 which I believe is the same issue but in libevent.
The issue is on a host that has IPV6 enabled but only a loopback IP address `-proxy=[::1]` will fail as `[::1]` is not considered valid by `getaddrinfo` with `AI_ADDRCONFIG` flag. I think the loopback interface should be considered valid and we have a functional test that will try to test this: `feature_proxy.py`.
To replicate the issue, run `feature_proxy.py` inside a docker container that has IPV6 loopback ::1 address without specifically giving that container an external IPV6 address. This should be the default with recent versions of docker. IPV6 on loopback interface was enabled in docker engine 26 and later ([https://docs.docker.com/engine/release-notes/26.0/#bug-fixes-and-enhancements-2](https://docs.docker.com/engine/release-notes/26.0/#bug-fixes-and-enhancements-2)).
`AI_ADDRCONFIG` was introduced to prevent slow DNS lookups on systems that were IPV4 only.
References:
Man section on `AI_ADDRCONFIG`:
```
If hints.ai_flags includes the AI_ADDRCONFIG flag, then IPv4 addresses are returned in the list pointed to by res only if the local system has at least one IPv4 address configured, and IPv6 addresses
are returned only if the local system has at least one IPv6 address configured. The loopback address is not considered for this case as valid as a configured address. This flag is useful on, for ex‐
ample, IPv4-only systems, to ensure that getaddrinfo() does not return IPv6 socket addresses that would always fail in connect(2) or bind(2).
```
[AI_ADDRCONFIG considered harmful Wiki entry by Fedora](https://fedoraproject.org/wiki/QA/Networking/NameResolution/ADDRCONFIG)
[Mozilla discussing slow DNS without AI_ADDRCONFIG and also localhost issues with it](https://bugzilla.mozilla.org/show_bug.cgi?id=467497)
ACKs for top commit:
achow101:
ACK 23333b7ed2
tdb3:
ACK 23333b7ed2
pinheadmz:
ACK 23333b7ed2
Tree-SHA512: 5ecd8c72d1e1c28e3ebff07346381d74eaddef98dca830f6d3dbf098380562fa68847d053c0d84cc8ed19a45148ceb5fb244e4820cf63dccb10ab3db53175020
55b6d7be68 validation: Don't load a snapshot if it's not in the best header chain. (Martin Zumsande)
Pull request description:
This was suggested by me in the discussion of #30288, which has more context.
If the snapshot is not an ancestor of the most-work header (`m_best_header`), syncing from that alternative chain leading to `m_best_header` should be prioritised. Therefore it's not useful loading the snapshot in this situation.
If the other chain turns out to be invalid or the chain with the snapshot retrieves additional headers so that it's the most-work one again (see functional test), `m_best_header` will change and loading the snapshot will be possible again.
Because of the work required to generate a conflicting headers chain, a situation with two conflicting chains should only be possible under extreme circumstances, such as major forks.
ACKs for top commit:
fjahr:
re-ACK 55b6d7be68
achow101:
ACK 55b6d7be68
alfonsoromanz:
Re ACK 55b6d7be68
Tree-SHA512: 4fbea5ab1038ae353fc949a186041cf9b397e7ce4ac59ff36f881c9437b4f22ada922490ead5b2661389eb1ca0f3d1e7e7e6a4261057678643e71594a691ac36
fac932bf93 refactor: Use util::Split to avoid a harmless unsigned-integer-overflow (MarcoFalke)
fab54db9f1 rest: Reject negative outpoint index in getutxos parsing (MarcoFalke)
Pull request description:
In `rest_getutxos` outpoint indexes such as `+N` or `-N` are accepted. This should be harmless, because any index out of range should be treated as a non-existent utxo. However, a negative index can't exist ever, so it seems better to reject all signs, whether `+` or `-`.
ACKs for top commit:
achow101:
ACK fac932bf93
hodlinator:
ut-ACK fac932bf93
tdb3:
re ACK fac932bf93
danielabrozzoni:
ACK fac932bf93
brunoerg:
reACK fac932bf93
Tree-SHA512: 8f1a75248cb61e1c4beceded6ed170db83b07f30fbcf93a26acfffc00ec4546572366eff87907a7e1423d7d3a2a9e57a0a7a9bacb787c86463f842d7161c16bc
faed5d3870 test: Non-Shy version sender (MarcoFalke)
Pull request description:
After `add_outbound_p2p_connection`, the test framework normally sends a version message only in reply to a received version. This is fine, but the protocol does not require this and tolerates a version to be sent earlier.
However, this is untested, and the missing test coverage leads to bugs being missed. For example https://github.com/bitcoin/bitcoin/pull/30394#pullrequestreview-2166824948
Fix it by adding a test.
ACKs for top commit:
brunoerg:
ACK faed5d3870
tdb3:
ACK faed5d3870
theStack:
tACK faed5d3870
glozow:
ACK faed5d3870
Tree-SHA512: dbf527a39c932e994a1e8248ba78058000811a4bf69275278f1fd1e545716ac4d2d3be5dcf362976bbafa2a49f91d13e3601daf71d29e9c556179b01af62c03c
c504b6997b refactor: add coinbase constraints to BlockCreateOptions (Sjors Provoost)
6b4c817d4b refactor: pass BlockCreateOptions to createNewBlock (Sjors Provoost)
323cfed595 refactor: use CHECK_NONFATAL to avoid single-use symbol (Sjors Provoost)
Pull request description:
When generating a block template through e.g. getblocktemplate RPC, we reserve 4000 weight units and 400 sigops. Pools use this space for their coinbase outputs.
At least one pool patched their Bitcoin Core node to adjust these hardcoded values. They eventually [produced an invalid block](https://bitcoin.stackexchange.com/questions/117837/how-many-sigops-are-in-the-invalid-block-783426) which exceeded the sigops limit.
The existince of such patches suggests it may be useful to make this value configurable. This PR would make such a change easier. However, the main motivation is that in the Stratum v2 spec requires the pool to communicate the maximum bytes they intend
to add to the coinbase outputs.
Specifically the `CoinbaseOutputDataSize` message which is part of the [Template Distribution Protocol](https://github.com/stratum-mining/sv2-spec/blob/main/07-Template-Distribution-Protocol.md#71-coinbaseoutputdatasize-client---server) has a field `coinbase_output_max_additional_size`.
A proposed change to the spec adds the max additional sigops as well: https://github.com/stratum-mining/sv2-spec/pull/86. Whether that change makes it into the spec is not important though, as adding both to `BlockAssembler::Options` makes sense.
The first commit is a test refactor followup for #30335, related to the code that's changed here, but not required.
The second commit introduces BlockCreateOptions, with just `use_mempool`.
The thirds commit adds `coinbase_max_additional_weight` and `coinbase_output_max_additional_sigops` to `BlockCreateOptions`. They use the originally hardcoded values, and no existing caller overrides these defaults. This changes in #29432.
ACKs for top commit:
itornaza:
tested ACK c504b6997b
ryanofsky:
Code review ACK c504b6997b
ismaelsadeeq:
Code review ACK c504b6997b
Tree-SHA512: de2fa085f47048c91d95524e03f909f6f27f175c1fefa3d6106445e7eb5cf5b710eda6ea5b641cf3b4704a4e4e0181a0c829003b9fd35465f2a46167e5d64487
If the binaries don't exist, the Guix build has failed for some other
reason.
There's no need to check for unknown architectures, or executable
formats, as the only ones that could be built are those that we've
configured toolchains for in Guix.
We've also been doing this inconsistently across the two scripts.
44f08786f4 test: Fix MSVC warning C4101 "unreferenced local variable" (Hennadii Stepanov)
5d25a82b9a univalue, refactor: Convert indentation tabs to spaces (Hennadii Stepanov)
Pull request description:
This PR is split from https://github.com/bitcoin/bitcoin/pull/30454 and addresses MSVC warning [C4101](https://learn.microsoft.com/en-us/cpp/error-messages/compiler-warnings/compiler-warning-level-3-c4101) "unreferenced local variable". The current MSVC build system in the master branch skips building univalue tests, so it is not affected.
No behaviour changes.
ACKs for top commit:
kevkevinpal:
utACK [44f0878](44f08786f4)
maflcko:
ACK 44f08786f4
theuni:
trivial ACK 44f08786f4.
Tree-SHA512: 661d3b40ddb4f7915de7a65ccb27a24da88ae499ce03c036099007260b0597e83738f1a3a420985b51f798ee309ade32988c6d78f4ffed401099b175a0b2025b
734076c6de [wallet, rpc]: add `max_tx_weight` to tx funding options (ismaelsadeeq)
b6fc5043c1 [wallet]: update the data type of `change_output_size`, `change_spend_size` and `tx_noinputs_size` to `int` (ismaelsadeeq)
baab0d2d43 [doc]: update reason for deducting change output weight (ismaelsadeeq)
7f61d31a5c [refactor]: update coin selection algorithms input parameter `max_weight` name (ismaelsadeeq)
Pull request description:
This PR taken over from #29264
The PR added an option `max_tx_weight` to transaction funding RPC's that ensures the resulting transaction weight does not exceed the specified `max_tx_weight` limit.
If `max_tx_weight` is not given `MAX_STANDARD_TX_WEIGHT` is used as the max threshold.
This PR addressed outstanding review comments in #29264
For more context and rationale behind this PR see https://delvingbitcoin.org/t/lightning-transactions-with-v3-and-ephemeral-anchors/418/11?u=instagibbs
ACKs for top commit:
achow101:
ACK 734076c6de
furszy:
utACK 734076c6de
rkrux:
reACK [734076c](734076c6de)
Tree-SHA512: 013501aa443d239ee2ac01bccfc5296490c27b4edebe5cfca6b96c842375e895e5cfeb5424e82e359be581460f8be92095855763a62779a18ccd5bdfdd7ddce7
When generating a block template through e.g. getblocktemplate RPC,
we reserve 4000 weight units and 400 sigops. Pools use this space
for their coinbase outputs.
At least one pool patched their Bitcoin Core node to adjust
these hardcoded values. They eventually produced an invalid
block which exceeded the sigops limit.
https://bitcoin.stackexchange.com/questions/117837/how-many-sigops-are-in-the-invalid-block-783426
The existince of such patches suggests it may be useful to
make this value configurable. This commit would make such a
change easier.
The main motivation however is that the Stratum v2 spec
requires the pool to communicate the maximum bytes they intend
to add to the coinbase outputs. A proposed change to the spec
would also require them to communicate the maximum number of sigops.
This commit also documents what happens when
-blockmaxweight is lower than the coinbase
reserved value.
Co-authored-by: Ryan Ofsky <ryan@ofsky.org>
