b22901dfa9 Avoid explicitly computing diagram; compare based on chunks (Pieter Wuille)
Pull request description:
This merges the `BuildDiagramFromChunks` and `CompareFeeRateDiagram` introduced in #29242 into a single `CompareChunks` function, which operates on sorted chunk data rather than diagrams, instead computing the diagram on the fly.
This avoids the need for the construction of an intermediary diagram object, and removes the slightly arbitrary "all diagrams must start at (0, 0)" requirement.
Not a big deal, but I think the result is a bit cleaner and not really more complicated.
ACKs for top commit:
glozow:
reACK b22901d
instagibbs:
reACK b22901dfa9
Tree-SHA512: ca37bdf61d9a9cb5435f4da73e97ead33bf65828ad9af49b87336b1ece70db8ced1c21f517fc6eb6d616311c91f3da75ecae6b9bd42547133e3a3c5320b7816d
13adbf733f remove unneeded environment option from cpp-subprocess (Sebastian Falbesoner)
2088777ba0 remove unneeded cwd option from cpp-subprocess (Sebastian Falbesoner)
03ffb09c31 remove unneeded bufsize option from cpp-subprocess (Sebastian Falbesoner)
79c3036373 remove unneeded close_fds option from cpp-subprocess (Sebastian Falbesoner)
62db8f8e5a remove unneeded session_leader option from cpp-subprocess (Sebastian Falbesoner)
80d008c66d remove unneeded defer_spawn option from cpp-subprocess (Sebastian Falbesoner)
cececad7b2 remove unneeded preexec function option from cpp-subprocess (Sebastian Falbesoner)
633e45b2e2 remove unneeded shell option from cpp-subprocess (Sebastian Falbesoner)
Pull request description:
The newly introduced cpp-subprocess library provides a good number of options for the `Popen` class:
0de63b8b46/src/util/subprocess.hpp (L1009-L1020)
Some of them are either not fully implemented (`shell`, missing an implementation on Windows), implemented in an ugly way (e.g. using "Impoverished, meager, needy, truly needy version of type erasure" for `preexec_func` according to the author's own words) or simply unlikely to be ever needed for our external signer use-case (`defer_spawn`). Instead of maintaining incomplete and/or unneeded code, I'd suggest to get rid of it and only keep support for options if there is a strong reason for it.
ACKs for top commit:
achow101:
ACK 13adbf733f
hebasto:
re-ACK 13adbf733f.
Tree-SHA512: 8270da27891cb659da2ef6062a23f4b86331859b15ac27b79ae7433b14f5bd7efaba621f2b3ba1953708d0f38377a8bd23ef1cc0f28b9c152ac8958dd9eec6b0
d5a715536e build: remove boost::process dependency for building external signer support (Sebastian Falbesoner)
70434b1c44 external_signer: replace boost::process with cpp-subprocess (Sebastian Falbesoner)
cc8b9875b1 Add `cpp-subprocess` header-only library (Hennadii Stepanov)
Pull request description:
Closes https://github.com/bitcoin/bitcoin/issues/24907.
This PR is based on **theStack**'s [work](https://github.com/bitcoin/bitcoin/issues/24907#issuecomment-1466087049).
The `subprocess.hpp` header has been sourced from the [upstream repo](https://github.com/arun11299/cpp-subprocess) with the only modification being the removal of convenience functions, which are not utilized in our codebase.
Windows-related changes will be addressed in subsequent follow-ups.
ACKs for top commit:
achow101:
reACK d5a715536e
Sjors:
re-tACK d5a715536e
theStack:
Light re-ACK d5a715536e
fanquake:
ACK d5a715536e - with the expectation that this code is going to be maintained as our own. Next PRs should:
Tree-SHA512: d7fb6fecc3f5792496204190afb7d85b3e207b858fb1a75efe483c05260843b81b27d14b299323bb667c990e87a07197059afea3796cf218ed8b614086bd3611
fa9f36baba build: Remove HAVE_GMTIME_R (MarcoFalke)
fa72dcbfa5 refactor: FormatISO8601* without gmtime* (MarcoFalke)
fa2c486afc Revert "time: add runtime sanity check" (MarcoFalke)
Pull request description:
Now that the `ChronoSanityCheck` has passed for everyone with C++17 and is guaranteed by C++20 to always pass, remove it.
Also, remove `gmtime_r` and `gmtime_s` and replace them with `year_month_day`+`hh_mm_ss` from C++20.
ACKs for top commit:
sipa:
utACK fa9f36baba
fanquake:
ACK fa9f36baba - more std lib & even less stuff to port.
Tree-SHA512: a9e7e805b757b7dade0bcc3f95273a7dc4f68622630d74838339789dd203ad7542d36b2e090a93b2bc5a7ecc383207dd7ec82c68147108bdac7ce44f088c8c9a
Upstream repo: https://github.com/arun11299/cpp-subprocess
Commit: 4025693decacaceb9420efedbf4967a04cb028e7
The "Convenience Functions" section is unused in our codebase, so it has
been removed.
7295986778 Unit tests for CalculateFeerateDiagramsForRBF (Greg Sanders)
b767e6bd47 test: unit test for ImprovesFeerateDiagram (Greg Sanders)
7e89b659e1 Add fuzz test for FeeFrac (Greg Sanders)
4d6528a3d6 fuzz: fuzz diagram creation and comparison (Greg Sanders)
e9c5aeb11d test: Add tests for CompareFeerateDiagram and CheckConflictTopology (Greg Sanders)
588a98dccc fuzz: Add fuzz target for ImprovesFeerateDiagram (Greg Sanders)
2079b80854 Implement ImprovesFeerateDiagram (Greg Sanders)
66d966dcfa Add FeeFrac unit tests (Greg Sanders)
ce8e22542e Add FeeFrac utils (Greg Sanders)
Pull request description:
This is a smaller piece of https://github.com/bitcoin/bitcoin/pull/28984 broken off for easier review.
Up to date explanation of diagram checks are here: https://delvingbitcoin.org/t/mempool-incentive-compatibility/553
This infrastructure has two near term applications prior to cluster mempool:
1) Limited Package RBF(https://github.com/bitcoin/bitcoin/pull/28984): We want to allow package RBF only when we know it improves the mempool. This narrowly scoped functionality allows use with v3-like topologies, and will be expanded at some point post-cluster mempool when diagram checks can be done efficiently against bounded cluster sizes.
2) Replacement for single tx RBF(in a cluster size of up to two) against conflicts of up to cluster size two. `ImprovesFeerateDiagram` interface will have to change for this use-case, which is a future direction to solve certain pins and improve mempool incentive compatibility: https://delvingbitcoin.org/t/ephemeral-anchors-and-mev/383#diagram-checks-fix-this-3
And longer-term, this would be the proposed way we would compute incentive compatibility for all conflicts, post-cluster mempool.
ACKs for top commit:
sipa:
utACK 7295986778
glozow:
code review ACK 7295986778
murchandamus:
utACK 7295986778
ismaelsadeeq:
Re-ACK 7295986778
willcl-ark:
crACK 7295986778
sdaftuar:
ACK 7295986778
Tree-SHA512: 79593e5a087801c06f06cc8b73aa3e7b96ab938d3b90f5d229c4e4bfca887a77b447605c49aa5eb7ddcead85706c534ac5eb6146ae2396af678f4beaaa5bea8e
fa39151394 refactor: Remove unused error() (MarcoFalke)
fad0335517 scripted-diff: Replace error() with LogError() (MarcoFalke)
fa808fb749 refactor: Make error() return type void (MarcoFalke)
fa1d624348 scripted-diff: return error(...); ==> error(...); return false; (MarcoFalke)
fa9a5e80ab refactor: Add missing {} around error() calls (MarcoFalke)
Pull request description:
`error(...)` has many issues:
* It is often used in the context of `return error(...)`, implying that it has a "fancy" type, creating confusion with `util::Result/Error`
* `-logsourcelocations` does not work with it, because it will pretend the error happened inside of `logging.h`
* The log line contains `ERROR: `, as opposed to `[error]`, like for other errors logged with `LogError`.
Fix all issues by removing it.
ACKs for top commit:
fjahr:
re-utACK fa39151394
stickies-v:
re-ACK fa39151394, no changes since 4a903741b0
ryanofsky:
Code review ACK fa39151394. Just rebase since last review
Tree-SHA512: ec5bb502ab0d3733fdb14a8a00762638fce0417afd8dd6294ae0d485ce2b7ca5b1efeb50fc2cd7467f6c652e4ed3e99b0f283b08aeca04bbfb7ea4f2c95d283a
This fixes the log output when -logsourcelocations is used.
Also, instead of 'ERROR:', the log will now say '[error]', like other
errors logged with LogError.
-BEGIN VERIFY SCRIPT-
sed -i --regexp-extended 's! error\("([^"]+)"! LogError("\1\\n"!g' $( git grep -l ' error(' ./src/ )
-END VERIFY SCRIPT-
d5228efb53 kernel: Remove dependency on CScheduler (TheCharlatan)
06069b3913 scripted-diff: Rename MainSignals to ValidationSignals (TheCharlatan)
0d6d2b650d scripted-diff: Rename SingleThreadedSchedulerClient to SerialTaskRunner (TheCharlatan)
4abde2c4e3 [refactor] Make MainSignals RAII styled (TheCharlatan)
84f5c135b8 refactor: De-globalize g_signals (TheCharlatan)
473dd4b97a [refactor] Prepare for g_signals de-globalization (TheCharlatan)
3fba3d5dee [refactor] Make signals optional in mempool and chainman (TheCharlatan)
Pull request description:
By defining a virtual interface class for the scheduler client, users of the kernel can now define their own event consuming infrastructure, without having to spawn threads or rely on the scheduler design.
Removing `CScheduler` also allows removing the thread and exception modules from the kernel library.
To make the `CMainSignals` class easier to use from a kernel library perspective, remove its global instantiation and adopt RAII practices.
Renames `CMainSignals` to `ValidationSignals`, which more accurately describes its purpose and scope.
Also make the `ValidationSignals` in the `ChainstateManager` and CTxMemPool` optional. This could be useful in the future for using or testing these classes without having to instantiate any form of signal handling.
---
This PR is part of the [libbitcoinkernel project](https://github.com/bitcoin/bitcoin/issues/27587). It improves the kernel API and removes two modules from the kernel library.
ACKs for top commit:
maflcko:
re-ACK d5228efb53🌄
ryanofsky:
Code review ACK d5228efb53. Just comment change since last review.
vasild:
ACK d5228efb53
furszy:
diff ACK d5228ef
Tree-SHA512: e93a5f10eb6182effb84bb981859a7ce750e466efd8171045d8d9e7fe46e4065631d9f6f533c5967c4d34c9bb7d7a67e9f4593bd4c5b30cd7b3bbad7be7b331b
86b7f28d6c serialization: use internal endian conversion functions (Cory Fields)
432b18ca8d serialization: detect byteswap builtins without autoconf tests (Cory Fields)
297367b3bb crypto: replace CountBits with std::bit_width (Cory Fields)
52f9bba889 crypto: replace non-standard CLZ builtins with c++20's bit_width (Cory Fields)
Pull request description:
This replaces #28674, #29036, and #29057. Now ready for testing and review.
Replaces platform-specific endian and byteswap functions. This is especially useful for kernel, as it means that our deep serialization code no longer requires bitcoin-config.h.
I apologize for the size of the last commit, but it's hard to avoid making those changes at once.
All platforms now use our internal functions rather than libc or platform-specific ones, with the exception of MSVC.
Sadly, benchmarking showed that not all compilers are capable of detecting and optimizing byteswap functions, so compiler builtins are instead used where possible. However, they're now detected via macros rather than autoconf checks.
This[ matches how libc++ implements std::byteswap for c++23](https://github.com/llvm/llvm-project/blob/main/libcxx/include/__bit/byteswap.h#L26).
I suggest we move/rename `compat/endian.h`, but I left that out of this PR to avoid bikeshedding.
#29057 pointed out some irregularities in benchmarks. After messing with various compilers and configs for a few weeks with these changes, I'm of the opinion that we can't win on every platform every time, so we should take the code that makes sense going forward. That said, if any real-world slowdowns are caused here, we should obviously investigate.
ACKs for top commit:
maflcko:
ACK 86b7f28d6c📘
fanquake:
ACK 86b7f28d6c - we can finish pruning out the __builtin_clz* checks/usage once the minisketch code has been updated. This is more good cleanup pre-CMake & for the kernal.
Tree-SHA512: 715a32ec190c70505ffbce70bfe81fc7b6aa33e376b60292e801f60cf17025aabfcab4e8c53ebb2e28ffc5cf4c20b74fe3dd8548371ad772085c13aec8b7970e
By defining a virtual interface class for the scheduler client, users of
the kernel can now define their own event consuming infrastructure,
without having to spawn threads or rely on the scheduler design.
Removing CScheduler also allows removing the thread and
exception modules from the kernel library.
6666713041 refactor: Rename fs::path::u8string() to fs::path::utf8string() (MarcoFalke)
856c88776f ArgsManager: return path by value from GetBlocksDirPath() (Vasil Dimov)
fa3d9304e8 refactor: Remove pre-C++20 fs code (MarcoFalke)
fa00098e1a Add tests for C++20 std::u8string (MarcoFalke)
fa2bac08c2 refactor: Avoid copy/move in fs.h (MarcoFalke)
faea30227b refactor: Use C++20 std::chrono::days (MarcoFalke)
Pull request description:
This:
* Removes dead code.
* Avoids unused copies in some places.
* Adds copies in other places for safety.
ACKs for top commit:
achow101:
ACK 6666713041
ryanofsky:
Code review ACK 6666713041. Just documentation change since last review.
stickies-v:
re-ACK 6666713041
Tree-SHA512: 6176e44f30b310d51632ec2d3827c3819905d0ddc6a4b57acfcb6cfa1f9735176da75ee8ed4a4abd1296cb0b83bee9374cc6f91ffac87c19b63c435eeadf3f46
6db04be102 Get rid of shutdown.cpp/shutdown.h, use SignalInterrupt directly (Ryan Ofsky)
213542b625 refactor: Add InitContext function to initialize NodeContext with global pointers (Ryan Ofsky)
feeb7b816a refactor: Remove calls to StartShutdown from KernelNotifications (Ryan Ofsky)
6824eecaf1 refactor: Remove call to StartShutdown from stop RPC (Ryan Ofsky)
1d92d89edb util: Get rid of uncaught exceptions thrown by SignalInterrupt class (Ryan Ofsky)
ba93966368 refactor: Remove call to ShutdownRequested from IndexWaitSynced (Ryan Ofsky)
42e5829d97 refactor: Remove call to ShutdownRequested from HTTPRequest (Ryan Ofsky)
73133c36aa refactor: Add NodeContext::shutdown member (Ryan Ofsky)
f4a8bd6e2f refactor: Remove call to StartShutdown from qt (Ryan Ofsky)
f0c73c1336 refactor: Remove call to ShutdownRequested from rpc/mining (Ryan Ofsky)
263b23f008 refactor: Remove call to ShutdownRequested from chainstate init (Ryan Ofsky)
Pull request description:
This change drops `shutdown.h` and `shutdown.cpp` files, replacing them with a `NodeContext::shutdown` member which is used to trigger shutdowns directly. This gets rid of an unnecessary layer of indirection, and allows getting rid of the `kernel::g_context` global.
Additionally, this PR tries to improve error handling of `SignalInterrupt` code by marking relevant methods `[[nodiscard]]` to avoid the possibility of uncaught exceptions mentioned https://github.com/bitcoin/bitcoin/pull/27861#discussion_r1255496707.
Behavior is changing In a few cases which are noted in individual commit messages. Particularly: GUI code more consistently interrupts RPCs when it is shutting down, shutdown state no longer persists between unit tests, the stop RPC now returns an RPC error if requesting shutdown fails instead of aborting, and other failed shutdown calls now log errors instead of aborting.
This PR is a net reduction in lines of code, but in some cases the explicit error handling and lack of global shutdown functions do make it more verbose. The verbosity can be seen as good thing if it discourages more code from directly triggering shutdowns, and instead encourages code to return errors or send notifications that could be translated into shutdowns. Probably a number of existing shutdown calls could just be replaced by better error handling.
ACKs for top commit:
achow101:
ACK 6db04be102
TheCharlatan:
Re-ACK 6db04be102
maflcko:
ACK 6db04be102👗
stickies-v:
re-ACK 6db04be102
Tree-SHA512: 7a34cb69085f37e813c43bdaded1a0cbf6c53bd95fdde96f0cb45346127fc934604c43bccd3328231ca2f1faf712a7418d047ceabd22ef2dca3c32ebb659e634
fa3da629a1 Remove DirIsWritable, GetUniquePath (MarcoFalke)
fad3a9793b Return LockResult::ErrorWrite in LockDirectory (MarcoFalke)
fa0afe7408 refactor: Return enum in LockDirectory (MarcoFalke)
Pull request description:
`GetUniquePath` is only used in tests and in `DirIsWritable`. The check by `DirIsWritable` is redundant with the check done in `LockDirectory`.
Fix the redundancy by removing everything, except `LockDirectory`.
ACKs for top commit:
TheCharlatan:
Re-ACK fa3da629a1
hebasto:
ACK fa3da629a1, I have reviewed the code and it looks OK.
Tree-SHA512: e95f18cd586de7582e9c08ac7ddb860bfcfcbc8963804f45c5784c5e4c0598dc59ae7e45dd4daf30a5020dbf8433f5db2ad06e46a8676371982003790043c6c9
Treating std::string as UTF-8 is deprecated in std::filesystem::path
since C++20.
However, it makes this codebase easier to read and maintain to retain
the ability for std::string to hold UTF-8.
Replace exceptions thrown by signal and wait methods with [[nodiscard]] return
values.
This is mostly a refactoring, but there is a slight change of behavior if
AbortShutdown function fails. The original behavior which was unintentionally
changed in #27861 is restored, so it now triggers an assert failure again
instead of throwing an exception. (The AbortShutdown function is only ever
called in the the GUI version of Bitcoin Core when corruption is detected on
loading and the user tries to reindex.)
Problems with using exceptions were pointed out by MarcoFalke in
https://github.com/bitcoin/bitcoin/pull/27861#discussion_r1255496707.
