d79249d2799e7f450d5708ca3366461f5f2069bc ci: add chimera Linux LTO CI job (fanquake)
Pull request description:
Adds a CI config based on using [Chimera Linux](https://chimera-linux.org/). This might be interesting for any of the following:
* Chimera is based on LLVM & musl libc - we test both of these in isolation, but not together.
* No GNU components. I don't think we have an existing Linux CI job that doesn't have a gcc/stdlibc++ install. This exercises the depends logic for a fully LLVM/Clang/lld only build, including building the native tools (related to #33902).
* We don't currently have a job with LTO enabled (here using CMakes `CMAKE_INTERPROCEDURAL_OPTIMIZATION`, which is `-flto=thin` for LLVM/Clang). I think this is worth having generally (we do use LTO in some other places, like oss-fuzz). If runtime is too much of an issue, then it could also be dropped. (Chimera itself is also compiled with LTO).
QT in depends doesn't build (#32744), so is excluded for now.
Chimera has pointed out at least a few quirks, i.e #34390, #34408 and https://github.com/bitcoin/bitcoin/pull/29963#discussion_r2707922298.
ACKs for top commit:
maflcko:
lgtm ACK d79249d2799e7f450d5708ca3366461f5f2069bc
hebasto:
ACK d79249d2799e7f450d5708ca3366461f5f2069bc.
Tree-SHA512: 1174a7462bf2e7433a2c27a6cf398e94b05db42bb414629c71cf9f9a297ca269e173ae1b7517b30510b494b4397f918eef706d3c75c4286767c5557aeb6db4c7
b65a3d80093b992a2574d5762a8cd03ce2eb4412 iwyu: Fix patch to prefer `<cstdint>` (Hennadii Stepanov)
Pull request description:
The goal of the [patch](https://github.com/bitcoin/bitcoin/blob/master/ci/test/01_iwyu.patch) is to suggest C++ headers rather than their C counterparts. However, for fixed width integer types, the patched IWYU currently suggests `<cinttypes>` where `<cstdint>` is sufficient.
This PR fixes this behavior.
ACKs for top commit:
maflcko:
lgtm ACK b65a3d80093b992a2574d5762a8cd03ce2eb4412
furszy:
utACK b65a3d80093b992a2574d5762a8cd03ce2eb4412
willcl-ark:
utACK b65a3d80093b992a2574d5762a8cd03ce2eb4412
Tree-SHA512: 695efdd44b92a642401738572e49c8b6591aa4463d387107fdf3d2f7c9c4b39f4097cb82413752caf9e8890dcca7246a894e562a1dd17023b05a7e455705beac
PreSelectedInputs is confusing to use. it's `total_amount`
might store total amount or effective amount based on SFFO.
ex: we might accidentally sum preselected inputs effective
amount (named `total_amount`) with automatically selected
inputs actual total amount.
CoinsResult has a cleaner interface with separate fields
for both these amounts.
2 behavioural changes:
1. no more default assert error if effective value is unset
- previously PreSelectedInputs::Insert() called
COutput::GetEffectiveValue() which assert failed
if the optional was unset.
- now we don't default assert anymore.
* in GUI/getAvailableBalance better not to assert.
* SelectCoins's preselected inputs always contain a
feerate, so effective amount should be set.
explicitly added an assertion to ensure this.
2. FetchSelectedInputs uses OutputType::UNKNOWN as key to
populate CoinsResult's coins map. it's discarded later.
faa4ab113cc9e300b3b8dce0c774d0a33a555883 ci: Drop valgrind fuzz from GHA matrix (MarcoFalke)
Pull request description:
The valgrind fuzz task is problematic, because:
* It is redundant with the msan fuzz task, which has std lib hardening enabled, so often UB is diagnosed before it even happens in the valgrind task.
* All issues so far found by the valgrind fuzz task were also found by the hardened msan fuzz task.
* All other issues were false-positives, which are hard to debug, and confusing and tedious to work around.
I don't think there is any value in asking pull request authors to debug valgrind false-positives that they triggered by accident. So remove the task for now.
I know that there are some devs, who like to keep the task, but if the task is kept, it should come with clear instructions on how to deal with false-postives in pull requests.
I am not proposing to remove the config itself, and I am happy to continue maintaining it, like it was done before. However, as of now, running it in the GHA matrix is of negative or questionable benefit.
ACKs for top commit:
l0rinc:
ACK faa4ab113cc9e300b3b8dce0c774d0a33a555883
fanquake:
ACK faa4ab113cc9e300b3b8dce0c774d0a33a555883 - hopefully we can revisit re-adding soon. To be clear, I don't agree with the rationale from #34304, or the initial changes there. The case here, and the fact that it is causing disruption in this repo, is more pressing.
Tree-SHA512: 59272f4b5b01c3b8ee6078ea635441f17776d4d8923f1adacdabdbb00bd2eb0234b30dc5b27938e29f8e79b3c3bebed5f339ae36c2c8fb17ea9d3a2884bee986
This makes it easier to:
* Run the exact command of any CI type and step locally
* Re-Run older CI tasks on GHA and using the latest merged config.
(.github/ci-windows.py is merged with master on re-runs, but
.github/workflows/ci.yml is NOT)
Also, writing it in Python has benefits:
* Any developer (even non-Windows ones) can read and modify the script.
* Python is already required for tests, so no new dependency is needed.
This refactor does not change behavior. However, it avoids a vector
copy, which can lead to a minimal speed-up of 1%-5%, depending on the
call-site. This is mostly relevant for the fuzz tests and utils that
read large blobs of data (like a full block).
returns the total amount (if SFFO), otherwise the effective amount.
previously, this was the logic in calculating
PreSelectedInputs::total_amount when PreSelectedInputs::Insert()
was called.
return optional to force callers to explicitly handle the case
when effective amount optional is not set.
coins.size() would be the number of the OutputType keys in the map.
whereas Size() would return total number of COutput objects when
flattening the map.
before #25806, set<COutput> was used and would not
contain same COutputs in the set.
now we use set<shared_ptr<COutput>> and it might be
possible for 2 distinct shared_ptr (different pointer
address but same COutputs) to be added into the set.
so preserve previous behaviour by making sure values
in the set are also distinct
e67a676df9af5ece5307438ae1b4ddb0730e3482 fix: uptime RPC returns 0 on first call (Lőrinc)
Pull request description:
### Problem
#34328 switched uptime to use monotonic time, but `g_startup_time` was a function-local static in `GetUptime()`, meaning it was initialized on first call rather than at program start.
This caused the first uptime RPC to always return 0.
### Fix
Move `g_startup_time` to namespace scope so it initializes at program start, ensuring the first `uptime()` call returns actual elapsed time.
### Reproducer
Revert the fix and run the test or alternatively:
```bash
cmake -B build && cmake --build build --target bitcoind bitcoin-cli -j$(nproc)
./build/bin/bitcoind -regtest -daemon
sleep 10
./build/bin/bitcoin-cli -regtest uptime
./build/bin/bitcoin-cli -regtest stop
```
<details>
<summary>Before (uptime is initialized on first call)</summary>
```bash
Bitcoin Core starting
0
Bitcoin Core stopping
```
</details>
<details>
<summary>After (first uptime call is in-line with sleep)</summary>
```bash
Bitcoin Core starting
10
Bitcoin Core stopping
```
</details>
----
Fixes#34423, added reporter as coauthor.
ACKs for top commit:
maflcko:
lgtm ACK e67a676df9af5ece5307438ae1b4ddb0730e3482
optout21:
crACK e67a676df9af5ece5307438ae1b4ddb0730e3482
carloantinarella:
Tested ACK e67a676df9af5ece5307438ae1b4ddb0730e3482
achow101:
ACK e67a676df9af5ece5307438ae1b4ddb0730e3482
musaHaruna:
Tested ACK [e67a676](e67a676df9)
Tree-SHA512: b156f7ed15c3fbb50e8a15f6c9a0d4e2ffb956d0c6ef949e0f8a661a564a20c0d3ed2149fae75ce7e2baa9326788d5037e726e7a7ac2c6ef4e70e4862cd5763f
54d039305823f67688ec9116757d8244f84badc6 FUZZ: Test that BnB finds best solution (Murch)
Pull request description:
BnB’s solution is the input set with the lowest waste score, excluding any supersets of other solution candidates.
This fuzz test compares a brute force search with the BnB result to ensure that BnB succeeds.
ACKs for top commit:
achow101:
ACK 54d039305823f67688ec9116757d8244f84badc6
brunoerg:
ACK 54d039305823f67688ec9116757d8244f84badc6
Tree-SHA512: 96b6a822f53311d9a76abe8c217794e0a2dd5bd713db0a15dc70e065099b8245c430e1174e24133e0a802218ff0f2943dfcc3d638c3716485d5607c452854e7d
322c4ec4422a2e03fc81e61b1315b1245ba47d2f build: replace WERROR with CMAKE_COMPILE_WARNING_AS_ERROR (fanquake)
Pull request description:
`CMAKE_COMPILE_WARNING_AS_ERROR` was added to CMake in 3.24.
`--compile-no-warning-as-error` can be used, if needed, in future, to
suppress the `CMAKE_COMPILE_WARNING_AS_ERROR` behaviour from a CI
config.
Potential alternative to #33297. Closes#33284.
See https://cmake.org/cmake/help/latest/prop_tgt/COMPILE_WARNING_AS_ERROR.html.
ACKs for top commit:
purpleKarrot:
re-ACK 322c4ec4422a2e03fc81e61b1315b1245ba47d2f
maflcko:
review ACK 322c4ec4422a2e03fc81e61b1315b1245ba47d2f 📩
hebasto:
re-ACK 322c4ec4422a2e03fc81e61b1315b1245ba47d2f.
Tree-SHA512: 1e9b41e87568c1021aecf9414cfc70a3215d1663a13adb1f32493f2dba18978b53888520775d1aa2945224b0404ef83bda0837dddf5f8a089cafb940b856598c
f2b8acc0edb6f1acf58cd3d7423d6ad60e8d90c8 remove glozow from trusted keys (glozow)
Pull request description:
As planned!
Also bump trusted-git-root to 88a7294356e75bbaa136c9427c64e239f7c6fd40, which is the last merge signed with this key.
ACKs for top commit:
achow101:
ACK f2b8acc0edb6f1acf58cd3d7423d6ad60e8d90c8
stickies-v:
ACK f2b8acc0edb6f1acf58cd3d7423d6ad60e8d90c8
Tree-SHA512: e68025c519ab003ad464783325502c1a6a3f560c0708a9383332cb0971be2039f462fc95f85373bb075191c165ca8e3904312a66e41b52d1b99dd2a8a837be9a
-Werror is added to the previous releases job, given it runs on Ubuntu
22.04, which uses an older CMake.
`--compile-no-warning-as-error` can be used, if needed, in future, to
suppress the `CMAKE_COMPILE_WARNING_AS_ERROR` behaviour from a CI
config.
CMAKE_COMPILE_WARNING_AS_ERROR was added to CMake in 3.24.
See https://cmake.org/cmake/help/latest/prop_tgt/COMPILE_WARNING_AS_ERROR.html.
Co-authored-by: willcl-ark <will8clark@gmail.com>
a50d0b6720f300987d2b3d82f4fb3a2336259887 build: don't pass on boost dependency to kernel consumers (Cory Fields)
Pull request description:
This is unnecessary now that the kernel now exports a (boost-less) API.
Noticed while slimming down boost dependencies in #34495.
ACKs for top commit:
stickies-v:
ACK a50d0b6720f300987d2b3d82f4fb3a2336259887
hebasto:
ACK a50d0b6720f300987d2b3d82f4fb3a2336259887, I have reviewed the code and it looks OK. I tested it by applying the Boost-specifc commits from https://github.com/bitcoin/bitcoin/pull/34143 and building with depends.
Tree-SHA512: e2d12356f41dd51dd729362121a33bd4f395821d53dd9a0bb0d5d6a53aba2ca2064e0709d9799dc6751b3d61ea576d2efc0e28296fdba26f2809dbcb0feabe44
1f8f7d477ae0d33bd96f7936889c17bd40805fb9 Change BlockRequestAllowed() to take ref (optout)
Pull request description:
As [suggested here](https://github.com/bitcoin/bitcoin/pull/34416#discussion_r2745302958), a minor refactor of `PeerManagerImpl::BlockRequestAllowed()` to take reference parameter (instead of pointer). The motivation is to make the code safer, by minimizing the risk of null-dereference, and to be more consistent.
The change is local to the `PeerManagerImpl::BlockRequestAllowed()` class.
Related to #34440.
ACKs for top commit:
maflcko:
review ACK 1f8f7d477ae0d33bd96f7936889c17bd40805fb9 🎐
l0rinc:
tested ACK 1f8f7d477ae0d33bd96f7936889c17bd40805fb9
stickies-v:
ACK 1f8f7d477ae0d33bd96f7936889c17bd40805fb9
Tree-SHA512: 9c2de2d3e7d067e018db7ec54c79f512ccc1da54574d4fb362f6697ee6e235783779d7094cf20856cd34e08a1dbc74609d8351fe7b2287cd8ec0c836ef07be19
Member variables of `CConnman::m_private_broadcast` (introduced in
https://github.com/bitcoin/bitcoin/pull/29415) could influence the tests
which creates non-determinism if the same instance of `CConnman` is used
for repeated test iterations.
So, reset the state of `CConnman::m_private_broadcast` from
`ConnmanTestMsg::Reset()`. Currently this affects the fuzz tests
`process_message` and `process_messages`.
Reported in
https://github.com/bitcoin/bitcoin/issues/34476#issuecomment-3849088794
42ee31e80c99bdb4d6affdc9dc22a0f3d5da7b59 doc: fix broken bpftrace installation link (jayvaliya)
Pull request description:
The bpftrace project has moved from the `iovisor` organization to its own
`bpftrace` organization on GitHub. The old installation documentation link
(`https://github.com/iovisor/bpftrace/blob/master/INSTALL.md`) is now broken (404).
The project also restructured their documentation - installation instructions
are now in the README.md under the "Quick Start" section rather than a
separate INSTALL.md file.
ACKs for top commit:
maflcko:
lgtm ACK 42ee31e80c99bdb4d6affdc9dc22a0f3d5da7b59
Tree-SHA512: 5320e747b77fe9bb347a98396364f3942214b6d3ee69707f62452df0aaaa2f6cfd58fc3b6155f64c4acb00f1a4dba8a751618805e75b823454c73ac93877ad4f
The bpftrace project moved from iovisor/bpftrace to bpftrace/bpftraceand
removed the separate INSTALL.md file. Installation instructionsare now
in the README.md Quick Start section.
BnB’s solution is the input set with the lowest waste score, excluding
any supersets of other solution candidates.
This fuzz test compares a brute force search with the BnB result to
ensure that BnB succeeds.
Pass `--failfast` to the functional test runner in `.github/ci-test-each-commit-exec.py`.
Stop after the first failure to surface the root cause sooner and keep logs readable when testing ancestor commits.
Co-authored-by: MarcoFalke <*~=`'#}+{/-|&$^_@721217.xyz>
Extend `test-each-commit` to run on every non-head pull request commit.
The PR tip is excluded because it is already covered by other CI jobs.
Runner was changed to a performant cirrus runner.
Co-authored-by: MarcoFalke <*~=`'#}+{/-|&$^_@721217.xyz>
db2effaca4cf82bf806596d16f9797d3692e2da7 scripted-diff: refactor: CWallet::Create() -> CreateNew() (David Gumberg)
27e021ebc0dd3517a71f3ddb38ed265a19693d4c wallet: Correctly log stats for encrypted messages. (David Gumberg)
d8bec61be233b9cb6d5db886e8f1c1f058288fb5 wallet: remove loading logic from CWallet::Create (David Gumberg)
f35acc893fb3378b2ad39608fe254d33af6cce9f refactor: wallet: Factor out `WriteVersion()` from `PopulateWalletFromDB()` (David Gumberg)
e12ff8aca049ec7b054cb3047a167c7ce8dbd421 test: wallet: Split create and load (David Gumberg)
70dbc79b09acf7b1515532ee20c7533c938ffb70 wallet: Use CWallet::LoadExisting() for loading existing wallets. (David Gumberg)
ae66e011646266abb67b31027bc29e0ce1d08ad4 wallet: Create separate function for wallet load (David Gumberg)
bc69070416c62a88d8f4029280ec10d6f9ec8d20 refactor: Wallet stats logging in its own function (David Gumberg)
a9d64cd49c69dafd6496ccb5aef4cd6d8898966b wallet: Remove redundant birth time update (David Gumberg)
b4a49cc7275efc16d4a4179ed34b50de5bb7367e wallet: Move argument parsing to before DB load (David Gumberg)
b15a94a618c53041e97ccfface3045a0642777e1 refactor: Split out wallet argument loading (David Gumberg)
a02c4a82d88a3e9a24ec2aa0b828b8cc533dde58 refactor: Move -walletbroadcast setting init (David Gumberg)
411caf72815bdf2e176e790a4c63f745517c4bb4 wallet: refactor: PopulateWalletFromDB use switch statement. (David Gumberg)
a48e23f566ccaf9b81fe0684885972d9ee34afd3 refactor: wallet: move error handling to PopulateWalletFromDB() (David Gumberg)
0972785fd723b9b3c84844bf999d6e08e163ef9d wallet: Delete unnecessary PopulateWalletFromDB() calls (David Gumberg)
f0a046094e4c4b5f3af0e453492077f4911e0132 scripted-diff: refactor: CWallet::LoadWallet->PopulateWalletFromDB (David Gumberg)
Pull request description:
This PR is mostly a refactor which splits out logic used for creating wallets and for loading wallets, both of which are presently contained in `CWallet::Create()` into `CWallet::CreateNew()` and `CWallet::LoadExisting()`
The real win of this PR is that `CWallet::Create()` uses a very bad heuristic for trying to guess whether or not it is supposed to be creating a new wallet or loading an existing wallet:
370c592612/src/wallet/wallet.cpp (L2882-L2885)
This heuristic assumes that wallets with no `ScriptPubKeyMans` are being created, which sounds reasonable, but as demonstrated in #32112 and #32111, this can happen when the user tries to load a wallet file that is corrupted, both issues are fixed by this PR and any other misbehavior for wallet files which succeeded the broken heuristic's sniff test for new wallets.
It was already the case that every caller of `CWallet::Create()` knows whether it is creating a wallet or loading one, so we can avoid replacing this bad heuristic with another one, and just shift the burden to the caller.
ACKs for top commit:
achow101:
ACK db2effaca4cf82bf806596d16f9797d3692e2da7
polespinasa:
approach ACK db2effaca4cf82bf806596d16f9797d3692e2da7
w0xlt:
reACK db2effaca4cf82bf806596d16f9797d3692e2da7
murchandamus:
ACK db2effaca4cf82bf806596d16f9797d3692e2da7
rkrux:
ACK db2effaca4cf82bf806596d16f9797d3692e2da7
Tree-SHA512: c28d60e0a3001058da3fd2bdbe0726c7ebe742a4b900a1dee2e5132eccc22e49619cb747a99b4032b000eafd4aa2fdd4ec244c32be2012aba809fdc94b5f6ecd
fad9dd1a8891770846f3f98c60bebf2c2bf72e05 test: kernel test fixups (MarcoFalke)
fabb58d42dc203b91f6ec6261f4bac94ee8df0a2 test: Use clang-tidy named args for create_chainman (MarcoFalke)
fa51594c5c0fe27e55d580dfab046e1226c6d83b refactor: Small style fixups in src/kernel/bitcoinkernel.cpp (MarcoFalke)
Pull request description:
Just some small style and test fixups after https://github.com/bitcoin/bitcoin/pull/30595#pullrequestreview-3420542946
ACKs for top commit:
stickies-v:
re-ACK fad9dd1a8891770846f3f98c60bebf2c2bf72e05
frankomosh:
Code Review ACK fad9dd1a8891770846f3f98c60bebf2c2bf72e05. All changes are sound refactoring with no functional issues. Nice improvements to readability (named args in create_chainman, span.data(), range checks now properly require non-empty).
Tree-SHA512: 0a92e871b4db75a590acad39672594625e402895bc0d36635d36ec2fe8dce7cc2c5cb6ebf2a92bc14617d94648b84bffb95ff783cea71bd91ac4a9871ef5dbef
fa33acec89f08b747449dbcd5248ee6d448a6648 Revert "valgrind: add suppression for bug 472219" (MarcoFalke)
Pull request description:
Now that several Alpine releases are out, it seems fine to remove this suppression.
This should work since Alpine 3.19: https://pkgs.alpinelinux.org/packages?name=valgrind&branch=v3.19&repo=&arch=aarch64, which just dropped out of `main` support (https://alpinelinux.org/releases/)
ACKs for top commit:
fanquake:
ACK fa33acec89f08b747449dbcd5248ee6d448a6648
Tree-SHA512: 2a58475d20f8214c012b77efb7fe145e810a1506736341f8debea74dabd31499899a968bc2d4a88008b2a3878b4fe5e48fa14c83a2e943a91943bf0e1e159b6f
4dfb6eef70d719a79904cabc4519d7a725de130a test: Add DERSIG tests to script_tests (billymcbip)
884978f3894ac7d96f113a00bbcce45c9785d44a test: Fix a STRICTENC test in script_tests (billymcbip)
527e8ca7b54515e129484824e4df66b5dafdb45f test: Remove outdated comment in script_tests (billymcbip)
Pull request description:
1. Remove a comment referencing a file that no longer exists in the codebase: `script_invalid.json`.
2. Fix a test that isn't implemented as intended. The idea is to test execution order by providing a signature that would cause script failure when parsed. An empty signature does not cause script failure in `CHECKMULTISIG`. Use `OP_1` for the second signature instead of `OP_0`.
3. Copy existing `STRICTENC` tests and change the flag to `DERSIG`. `DERSIG` is a consensus flag (unlike `STRICTENC`), so it'd be good to have dedicated test cases.
`script_tests` pass on my end.
ACKs for top commit:
darosior:
reACK 4dfb6eef70d719a79904cabc4519d7a725de130a
sipa:
ACK 4dfb6eef70d719a79904cabc4519d7a725de130a
Tree-SHA512: aea4aa5199530804561e9f597f69d6cffd7a40d4830919f9371fe97e4d04d8f10e8ed29b91e65e007a3f6e3a38e2881f88dff25831e741ad7592a12ed02b801a
The goal of the patch is to suggest C++ headers rather than their C
counterparts. However, for fixed width integer types, the patched IWYU
currently suggests `<cinttypes>` where `<cstdint>` is sufficient.
This change fixes this behavior.
02b5f6078d65c3a2f9ba8b30474d8201516c5c4b fees: make flushes log debug only (ismaelsadeeq)
Pull request description:
This is a simple PR that updates the flushing log to use debug-level logging under the estimatefee category. It also ensures the log consistently includes only the full file path.
The motivation behind this is that the "Flushed fee estimates to fee_estimates.dat." logs can become noisy; it's done after one hour, so hiding it in the debug estimatefee category seems reasonable.
---
I left the logs when the file is not found as info because that should only occur when you start a fresh node, change datadir, or explicitly delete the file
ACKs for top commit:
achow101:
ACK 02b5f6078d65c3a2f9ba8b30474d8201516c5c4b
furszy:
utACK 02b5f6078d65c3a2f9ba8b30474d8201516c5c4b
l0rinc:
Lightly tested ACK 02b5f6078d65c3a2f9ba8b30474d8201516c5c4b
sipa:
utACK 02b5f6078d65c3a2f9ba8b30474d8201516c5c4b
Tree-SHA512: 3e4b822caa23db9b30f1bb8990b36f35dcfcd82dbfb27c319463447da1df988eded84c47d5319ad637c822bdd04f0a727a176da3632c40d786332b6a5aaa6d89
