Every captured commit receives deterministic security triage and a separate communication-quality score. Security candidates and broader second-pass signals receive full-patch Ollama analysis.
Message quality measures whether a commit identifies its scope, purpose, rationale, testing, and supporting references. It does not change the security-severity score.
This commit only changes documentation comments and enables a stricter compiler warning for documentation style. It does not alter any executable code, so it cannot introduce or fix a security vulnerability on its own.
This change fixes a privacy leak in Bitcoin Core's wallet. When users used the 'bumpfee' feature to replace a transaction with a higher fee, the replacement could sometimes end up with an older 'locktime' than the original transaction. Thi…
Privacy/fingerprinting issue in wallet transaction creationReplacement transaction locktime can be lower than original transaction locktimeAnti-fee-sniping random backdating can produce identifiable on-chain pattern
This patch fixes a bug in Bitcoin Core's wallet transaction creation where extremely large output amounts could cause an integer overflow or even crash the node. Before the fix, a user could craft a transaction whose outputs added up to mo…
Integer overflow in recipient amount summationNode abort via failed assertion in coin selectionPossible creation of a transaction with wrapped/invalid total output value
This change fixes how Bitcoin Core sends your Tor control password to the Tor daemon. Previously, the password was wrapped in quotes and only double-quote characters were escaped. If a password contained a backslash, Tor would misinterpret…
Improper escaping of protocol string (QuotedString)Potential Tor control command boundary confusionAuthentication bypass/failure risk for Tor control
This commit marks the existing 'recortar el video' feature as deprecated. It doesn't remove it, but it now shows a warning and requires a special flag to use. The change is meant to protect users from accidentally losing access to their bi…
Deprecation of risky wallet-encryption workflowKnown data-retention risk: unencrypted key material may remain on disk after late encryptionKnown backup/restore risk: post-encryption keys are not covered by pre-encryption backups
This is a code-cleanup change in Bitcoin Core. It replaces many uses of `assert(false)` and similar 'this should never happen' markers with a new macro called `AssertUnreachable()`. The goal is to make the code clearer for developers and a…
No security-relevant functional change: process still aborts on unreachable pathsRefactor only: replaces one abort mechanism with another equivalent oneAdds a new lint rule to enforce the preferred macro going forward
This is a minor cleanup inside a single test file. It changes how a fake transaction output is built in a unit test so the empty script is written explicitly rather than created implicitly by resizing a list. The actual test behavior and t…
This is a routine subtree update for the secp256k1 cryptographic library inside Bitcoin Core. It pulls in several upstream fixes, most of which are build-system cleanups, test improvements, and API hardening. The most user-visible changes …
Silent Payments API no longer aborts on empty input/output arrays; returns 0 per documented behaviorSilent Payments label-tweak overflow now rejected with ARG_CHECK instead of being silently handledecmult_const gains a _ge variant that clears intermediate Jacobian coordinates to reduce secret leakage
This commit only cleans up Bitcoin Core's own functional test suite. It removes unused multisig test code, fixes outdated docstrings, and drops a no-longer-needed command-line flag in test scripts. None of these changes affect the actual B…
This commit only cleans up two internal test scripts for the Bitcoin Core project. It replaces a manual network disconnect/reconnect pattern with a simpler way to create an unconnected block header, and removes a hard-coded number in favor…
This is a tiny code cleanup in a Bitcoin Core test file. It replaces the hard-coded number 3 with the actual number of test nodes, making the test slightly easier to maintain. It does not change any real Bitcoin Core software behavior and …
This change hides an old, always-false field called 'iswatchonly' from the getaddressinfo API response unless the user explicitly asks for deprecated features. It is a routine cleanup, not a security fix, and does not change how wallets or…
This commit only cleans up Bitcoin Core's own automated test suite. It removes unused test nodes and an unnecessary double restart in a few wallet-related test scripts. There is no change to the actual Bitcoin node software that users run,…
This change fixes a bug where Bitcoin Core would disconnect every peer sharing the same IP address when one peer misbehaved. The problem is especially bad for users connecting through Tor, because all Tor traffic exits through a small set …
Denial of service via shared IP disconnectTor exit-node IP collision exploited to disconnect honest peersPeer-to-peer network isolation of inbound connections
This commit fixes a bug where Bitcoin Core would disconnect every peer sharing the same IP address when one of them misbehaved. Because Tor exit nodes share a single IP address among many users, a malicious user could deliberately misbehav…
Denial-of-service via shared IP disconnectTor exit-node IP collision exploited to disconnect honest peersPeer-disconnection logic changed from subnet match to single-node flag
This change is purely a code-cleanup and linting improvement. It removes repetitive 'IWYU pragma: keep' comments from build-config header includes and instead marks the generated config header itself with a single 'always_keep' pragma. The…
This is a code cleanup change that makes the type system express 'this pointer can never be null' for several internal objects and for block templates. It removes redundant null checks and adds helper functions to create such non-null poin…
No security-relevant bug is patchedChange is purely a type-system refactorRemoves dead null-check error paths rather than adding safety checks
This commit only changes a test file to make a timing check less strict. It removes an upper-bound timing assertion that was causing false test failures on slow CI machines. There is no change to Bitcoin Core's actual network or wallet cod…
This commit is a routine cleanup, not a security fix. It removes support for an old, confusingly-named option ('confTarget') in two Bitcoin wallet commands (bumpfee and psbtbumpfee), leaving only the clearer 'conf_target' name unless the u…
This is a tiny change to a Bitcoin Core test helper that makes fee-checking tests slightly more tolerant. It is not a security fix and does not affect production wallet code or consensus rules. It simply reduces how often an extremely rare…
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Links an issue, advisory, or supporting reference
Why it was queued
access controlmerge-commit duplicate discount
AI analysis · Informational 15/100
This is a tiny code cleanup in Bitcoin Core's network handling. It removes an unnecessary second check that 'peer' exists, because the surrounding code already confirmed it exists. There is no functional change and no security risk.
Lower-priorityMerge bitcoin/bitcoin#36554: cmake: Reject warnings in `check_cxx_source_compiles_with_flags()`, use `-msse4.1` for SHA-NIby merge-script · ddea9196 · Nov 2, 2026 · 6 filesMessage 91 · StrongTriage 0Details
Commit message · merge-script
Merge bitcoin/bitcoin#36554: cmake: Reject warnings in `check_cxx_source_compiles_with_flags()`, use `-msse4.1` for SHA-NI
5bb7c5646a7f11e910c620107ba873bb4ecb3f3f cmake, crypto: Use `-msse4.1` instead of `-msse4` for x86 SHA-NI (Hennadii Stepanov) 9ab42235c1567442c73c3c0316a309c7d7a10283 cmake: Treat warnings as errors in `CheckSourceCompilesWithFlags` (Hennadii Stepanov) c4763b716db175fa106c47a1cd68abf07e762e6b cmake: Make sanitizer and fuzzer checks warning-free (Hennadii Stepanov) 8ba6cb08dfafaa0dc5ee5f95f20ee4124bb518f2 cmake, minisketch: Make CLMUL check warning-free (Hennadii Stepanov) 89bc05fe832d597c72f175e8a473555c963d5a7d cmake, crc32c: Make SSE4.2 check warning-free (Hennadii Stepanov) 53865a0addc240cb6db3ad03b2d578420990a243 cmake: Initialize variables in ARMv8 SHA-NI check (Hennadii Stepanov) cd3353c6892e3c85bf6dc52b0e6bd98c493532b3 cmake: Add `[[maybe_unused]]` annotation to atomic library check (Hennadii Stepanov)
Pull request description:
This is a prerequisite for #31507.
1. The first commit fixes a bug in the C++ atomic library check: ```console $ git rev-parse HEAD 01b51c0f1fa8af242fc37891d666c3d940045d80 $ cmake -B build -DAPPEND_CXXFLAGS="-Wall -Werror" 2>&1 | grep -A 8 STD_ATOMIC -- Performing Test STD_ATOMIC_LINKS_WITHOUT_LIBATOMIC -- Performing Test STD_ATOMIC_LINKS_WITHOUT_LIBATOMIC - Failed -- Performing Test STD_ATOMIC_NEEDS_LINK_TO_LIBATOMIC -- Performing Test STD_ATOMIC_NEEDS_LINK_TO_LIBATOMIC - Failed CMake Error at cmake/module/TestAppendRequiredLibraries.cmake:91 (message): Cannot figure out how to use std::atomic. Call Stack (most recent call first): cmake/introspection.cmake:23 (test_append_atomic_library) CMakeLists.txt:453 (include)
Run Build Command(s): /usr/bin/ninja-build -v cmTC_88170 [1/2] /usr/lib64/ccache/c++ -DHAVE_SSE42 -msse4.2 -O2 -g -std=c++20 -fPIE -o CMakeFiles/cmTC_88170.dir/src.cxx.o -c /home/hebasto/dev/bitcoin/build/CMakeFiles/CMakeScratch/TryCompile-fn8wHI/src.cxx -Wconversion -Werror FAILED: [code=1] CMakeFiles/cmTC_88170.dir/src.cxx.o /usr/lib64/ccache/c++ -DHAVE_SSE42 -msse4.2 -O2 -g -std=c++20 -fPIE -o CMakeFiles/cmTC_88170.dir/src.cxx.o -c /home/hebasto/dev/bitcoin/build/CMakeFiles/CMakeScratch/TryCompile-fn8wHI/src.cxx -Wconversion -Werror /home/hebasto/dev/bitcoin/build/CMakeFiles/CMakeScratch/TryCompile-fn8wHI/src.cxx: In function ‘int main()’: /home/hebasto/dev/bitcoin/build/CMakeFiles/CMakeScratch/TryCompile-fn8wHI/src.cxx:11:22: error: conversion from ‘uint64_t’ {aka ‘long unsigned int’} to ‘unsigned int’ may change value [-Werror=conversion] 11 | l = _mm_crc32_u8(l, 0); | ^ /home/hebasto/dev/bitcoin/build/CMakeFiles/CMakeScratch/TryCompile-fn8wHI/src.cxx:12:23: error: conversion from ‘uint64_t’ {aka ‘long unsigned int’} to ‘unsigned int’ may change value [-Werror=conversion] 12 | l = _mm_crc32_u32(l, 0); | ^ /home/hebasto/dev/bitcoin/build/CMakeFiles/CMakeScratch/TryCompile-fn8wHI/src.cxx:14:12: error: conversion from ‘uint64_t’ {aka ‘long unsigned int’} to ‘int’ may change value [-Werror=conversion] 14 | return l; | ^ cc1plus: all warnings being treated as errors ninja: build stopped: subcommand failed.
exitCode: 1 ```
4. Similarly, the CLMUL check in minisketch was made warning-free.
5. The sanitizer and fuzzer checks was made warning-free.
6. We expect that neither the compiler flags nor the test source triggers any warnings, such as `-Wunknown-argument`. Therefore, append `working_compiler_werror_flag`, as done in [`try_append_cxx_flags()`](https://github.com/bitcoin/bitcoin/blob/master/cmake/module/TryAppendCXXFlags.cmake).
7. `-msse4` is merely a synonym for `-msse4.2` in both GCC and Clang, but it is not accepted by the clang-cl frontend. Furthermore, the x86 SHA-NI code uses only SSE4.1 intrinsics and no SSE4.2 ones.
Replace `-msse4` with `-msse4.1` to resolve both issues.
The absence of SSE4.2 usage can be verified by building with `-DAPPEND_CXXFLAGS=-mno-sse4.2`.
ACKs for top commit: fanquake: ACK 5bb7c5646a7f11e910c620107ba873bb4ecb3f3f hodlinator: re-ACK 5bb7c5646a7f11e910c620107ba873bb4ecb3f3f
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
defensive validationfuzzing or regression evidencemerge-commit duplicate discount
Security candidateMerge bitcoin/bitcoin#36712: fees: prevent divide by zero in CFeeRateby merge-script · 38ece2a9 · Nov 2, 2026 · 3 filesMessage 91 · StrongModerate 60Details
Commit message · merge-script
Merge bitcoin/bitcoin#36712: fees: prevent divide by zero in CFeeRate
f442fa20d437f02518c327736206185bd6f6a1a1 fees: prevent divide by zero in CFeeRate (ismaelsadeeq)
Pull request description:
A CFeeRate with size 0 `GetFeePerK()` and `CFeeRate+=` evaluated `EvaluateFeeDown(1000)` unconditionally, dividing by that zero size and crashing.
Return 0 for an empty fee rate, matching `GetFee()` behaviour, and add minimal unit test coverage.
This was reported by Jacob F Cudney (aragondragon)
ACKs for top commit: pablomartin4btc: re-ACK f442fa20d437f02518c327736206185bd6f6a1a1 polespinasa: re-ACK f442fa20d437f02518c327736206185bd6f6a1a1 musaHaruna: Code review ACK [f442fa2](https://github.com/bitcoin/bitcoin/pull/36712/changes/f442fa20d437f02518c327736206185bd6f6a1a1)
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
access controlmerge-commit duplicate discount
AI analysis · Moderate 60/100
This commit fixes a divide-by-zero crash in Bitcoin Core's fee-rate calculation. When a transaction fee rate was created with a size of zero, calling certain methods would try to divide by zero and crash the program. The fix makes these methods safely return zero instead, matching the behavior of related code, and adds tests to prevent the crash from coming back.
Security candidateMerge bitcoin/bitcoin#36305: net: Use exclusive P2P listen sockets on Windowsby Hennadii Stepanov · 3581d56b · Nov 2, 2026 · 2 filesMessage 100 · StrongModerate 62Details
Commit message · Hennadii Stepanov
Merge bitcoin/bitcoin#36305: net: Use exclusive P2P listen sockets on Windows
db31a3ad1f75538d07707d373adbfbe90ba8e328 test: check P2P listen port exclusivity (winterrdog) 0dbd271eb4a3263b8b81ccf22876429076bae574 net: use exclusive P2P listen sockets on Windows (winterrdog)
Pull request description:
on Windows, `SO_REUSEADDR` does not carry over the same semantics as on POSIX systems. a second socket can set `SO_REUSEADDR` and bind to a port that is _already in use_ by another process running under the _same_ user account. this means another local process could bind to the P2P listen port and receive incoming connections intended for `bitcoind`.
this matters because the P2P listener is expected to have exclusive ownership of its address and port. allowing another process to bind to it could let that process _intercept_ incoming P2P connections. this is in a similar spirit to #36169, which made the RPC listener exclusive on Windows.
to fix this, we use `SO_EXCLUSIVEADDRUSE` for the P2P listen socket on Windows instead of `SO_REUSEADDR`, while retaining `SO_REUSEADDR` on other platforms. the added test verifies the expected behaviour by attempting to bind a second socket to the same port. on Windows, this fails with `WSAEACCES` (`10013`), which Python exposes as `PermissionError` (an `OSError` subclass).
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
memory safetycredential or privilege statemerge-commit duplicate discount
AI analysis · Moderate 62/100
This change fixes a Windows-specific networking issue in Bitcoin Core. Previously, another program running on the same Windows computer under the same user account could hijack Bitcoin's peer-to-peer network port, potentially intercepting incoming connections meant for the Bitcoin node. The fix makes the listening socket exclusive on Windows, similar to a change already made for the RPC port. Other operating systems are unaffected because their behavior was already safe.
AI review queuedMerge bitcoin/bitcoin#36618: test: Wait for index sync in all testsby merge-script · e310efb2 · Nov 2, 2026 · 18 filesMessage 91 · StrongInformational 15Details
Commit message · merge-script
Merge bitcoin/bitcoin#36618: test: Wait for index sync in all tests
fa74dc72d942e1628cea3c143bb56eed1457bbce test: refactor: Wait for index sync by default (MarcoFalke) fa8009ba2997a0eb3e6e1a657db507e5a8b8b529 test: Wait for index sync in wallet_fast_rescan.py (MarcoFalke)
Pull request description:
Fixes https://github.com/bitcoin/bitcoin/issues/36564 by adding the missing wait on the index sync.
``` - 2026-10-18T05:06:44.451973Z [http.01] [wallet/scan.cpp:142] [ShouldFetchBlock] [scan] Fast rescan: inspect block 200 [4c063a93914ec0cb1604297f2f601dab9b9342bf073e6b80e1c11efd71714e2e] (WARNING: block filter not found!)
ACKs for top commit: ViniciusCestarii: ACK fa74dc72d942e1628cea3c143bb56eed1457bbce hebasto: ACK fa74dc72d942e1628cea3c143bb56eed1457bbce, the changes are correct.
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
signing or wallet pathmerge-commit duplicate discountsecond-pass: security-sensitive path
AI analysis · Informational 15/100
This commit only changes Bitcoin Core's internal functional test framework. It makes the test harness automatically wait for optional indexes (like transaction index, block filter index, and coin stats index) to finish syncing before continuing tests. It does not change the production Bitcoin node software that users run, so it cannot directly affect real wallets, funds, or network security.
Security candidateMerge bitcoin/bitcoin#36091: test: Add debug output to common tested typesby merge-script · d54844ea · Nov 2, 2026 · 18 filesMessage 91 · StrongInformational 15Details
Commit message · merge-script
Merge bitcoin/bitcoin#36091: test: Add debug output to common tested types
722d08e03980c359bae4e3b65551d6b8ba615a7a test: Convert `BOOST_CHECK` macros with known debug string (rustaceanrob) d735b94803a6d6e1ae8a3716a2ca7b1a63e1ca1e test: Add display `operator<<` overloads to `common.{h,cpp}` (rustaceanrob)
Pull request description:
If the `<<` operator exists, boost can use this as a debug output when a check fails (besides `BOOST_CHECK`). This adds `<<` for types commonly used across the unit tests that do not yet have a display output. Each callsite that uses these types are then updated to use macros that will print the debug output on failure. All of these display strings would be used as part of #35713, but we can also use them with boost today.
What this PR does: - Adds debug output for types used across multiple files, or types that have a clear debug string, and migrates the `BOOST_CHECK` calls. This covers the majority of custom types that did not yet have a debug string.
What this PR does _not_ do: - Migrate `BOOST_CHECK` callsites for types that already _do_ have a debug output (all primitive type comparisons, `CFeeRate`, etc). This would be a much larger refactor and could perhaps be done after. - Add debug output for `std::vector<T>` where `T` is streamable. These can be migrated to `BOOST_CHECK_EQUAL_COLLECTIONS`, but is a similar refactor to the one above. - Add debug output for `std::vector<unsigned char>` and related. Also a large refactor, and could use discussion. - Add debug output for types that would require a `friend` method in the source code (`ByRatio`)
ACKs for top commit: maflcko: review ACK 722d08e03980c359bae4e3b65551d6b8ba615a7a 🍟 ismaelsadeeq: reACK 722d08e03980c359bae4e3b65551d6b8ba615a7a
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
cryptography-sensitive pathsigning or wallet pathmerge-commit duplicate discount
AI analysis · Informational 15/100
This commit only changes Bitcoin Core's test code. It adds pretty-printing helpers for common data types and updates test assertions to use macros that print more helpful failure messages. There is no change to the actual Bitcoin node software that users run, so it cannot affect live wallets, transactions, or network behavior.
This can help to prevent out-of-bounds reads such as the one removed in 43a33e513c4044dc01d92e77da76f2e98faaedb5, or prevent other bugs.
There is some code that correctly uses signed chars, but this can be trivially made "tidy" by using C++11 narrowing init to `int` (or whatever type you want).
ACKs for top commit: jpk68: ACK fa6efa5da55c4b1acda336fb875e94b020c023a6
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode✓ Links an issue, advisory, or supporting reference
Why it was queued
merge-commit duplicate discount
Security candidateMerge bitcoin/bitcoin#36687: test: wait for the node to finish the last block in p2p_ping_ibdby merge-script · 6c01bd0b · Oct 31, 2026 · 1 fileMessage 91 · StrongInformational 15Details
Commit message · merge-script
Merge bitcoin/bitcoin#36687: test: wait for the node to finish the last block in p2p_ping_ibd
9af82f9872ca1a9e8d4645c537b23af4bc861ed2 test: wait for the node to finish the last block in p2p_ping_ibd (Greg Sanders)
Pull request description:
ProcessBlock() stamps m_last_block_time only after ProcessNewBlock() returns, so getblockcount can show the new height, with nothing in flight, before the stamp is taken. A bumpmocktime in that gap stamps the block with the bumped time, and the final bump then leaves the ping timeout exactly at its boundary, where it never fires. Sync with a ping after the last block, so the node has finished processing it before the clock moves.
The race has been present since the test was added in #36080.
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
This is a fix for a flaky automated test, not a security fix for the Bitcoin Core software itself. The test sometimes failed because it moved the simulated clock forward before the test node had fully finished recording the time of the last downloaded block. The patch adds a small synchronization step so the test waits for the node to finish processing before advancing the clock. It does not change production code or fix any vulnerability that could affect real users.
**Problem:** The four-argument `CBloomFilter` constructor is used by unit tests and the fuzz target to exercise Bloom filter sizing directly. Zero elements make the hash-function calculation divide by zero. The size calculation converts its floating-point result to `unsigned int` before applying the protocol limit, which is undefined for zero or invalid false-positive rates and for valid parameters whose calculated size exceeds the integer range.
**Fix:** Require false-positive rates in `[0, 1]`, construct the existing empty match-all state for zero elements, and use the largest permitted filter for a zero rate with positive elements. Clamp the calculated size before its integer conversion so large valid inputs remain within the protocol limit.
ACKs for top commit: rustaceanrob: Code review ACK c4302ddf806094fcc2b1e21593e25b2038191a07 sedited: ACK c4302ddf806094fcc2b1e21593e25b2038191a07 instagibbs: ACK c4302ddf806094fcc2b1e21593e25b2038191a07
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
access controlfuzzing or regression evidencemerge-commit duplicate discount
AI analysis · Low 35/100
This commit fixes a bug in Bitcoin Core's Bloom filter constructor that could perform undefined math when given unusual inputs, such as zero elements or a zero/invalid false-positive rate. The fix adds input checks and special-case handling so the constructor no longer divides by zero, takes the logarithm of zero, or converts an out-of-range floating-point size to an integer. The affected constructor is used only in tests and fuzzing, not in normal peer-to-peer network code, so the practical risk to live Bitcoin nodes is low.
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Links an issue, advisory, or supporting reference
Why it was queued
merge-commit duplicate discount
AI review queuedMerge bitcoin/bitcoin#36011: btcsignals: destroy callbacks on disconnectby merge-script · 862f6038 · Oct 31, 2026 · 2 filesMessage 100 · StrongLow 29Details
Commit message · merge-script
Merge bitcoin/bitcoin#36011: btcsignals: destroy callbacks on disconnect
a318b23131e6df2c64e8fe79cebc8e5babdf8b7f btcsignals: fall back to atomic_load/store without __cpp_lib_atomic_shared_ptr (Ryan Ofsky) f1d786c1ef5d813f61e405370a67df4cc6323885 btcsignals: destroy callbacks eagerly on disconnect (Ryan Ofsky) 7c5571adc46857cfcf77d81b6fc135dee5369749 test: characterize btcsignals callback destruction timing (Ryan Ofsky)
Pull request description:
Currently callback functions connected to signals do not get freed when their `connection::disconnect` method is called or when their associated `scoped_connection` is destroyed. They live as long as the signal does, even if they will never be called, and can only get [garbage collected](https://github.com/bitcoin/bitcoin/blob/0fd515bbb7070a502dd2fe6d16f5f3cbac591283/src/util/btcsignals.h#L240-L241) if new connections are added.
This behavior is different that previous boost signals behavior and causes a [deadlock](https://github.com/bitcoin/bitcoin/actions/runs/31848571315/job/94919834743) on GUI shutdown in #10102 because when the node registers signal callbacks (`handleInitMessage`, etc) that get forwarded to the GUI, the callbacks own `mp::EventLoopRef` references, and if the callbacks aren't freed, the node process can't shut down.
This change restores boost signals behavior freeing callbacks when they are disconnected or when `scoped_connection` objects are destroyed to avoid this problem and make callback lifetimes more predictable.
ACKs for top commit: hodlinator: re-ACK a318b23131e6df2c64e8fe79cebc8e5babdf8b7f sedited: Re-ACK a318b23131e6df2c64e8fe79cebc8e5babdf8b7f
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
This change fixes a bug in Bitcoin Core's internal signal/callback system where callback functions and the resources they own were not released when disconnected. Previously, disconnected callbacks lingered until the signal itself was destroyed or new connections triggered garbage collection. This caused a real-world deadlock during GUI shutdown because node-to-GUI callbacks held references that prevented the node process from exiting. The patch makes callbacks get destroyed immediately on disconnect (or after any in-flight call finishes), matching the behavior of the older Boost signals library.
AI review queuedMerge bitcoin/bitcoin#36611: test: consolidate mempool constantsby Ryan Ofsky · ab7e668c · Oct 31, 2026 · 10 filesMessage 91 · StrongInformational 15Details
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
signing or wallet pathmerge-commit duplicate discountsecond-pass: security-sensitive path
AI analysis · Informational 15/100
This commit is a test-only cleanup. It moves several hard-coded numbers used in Bitcoin Core's functional tests into a shared helper file so they are defined in one place. No production code, consensus rules, or network behavior were changed.
Lower-priorityMerge bitcoin/bitcoin#36701: ci: Disable tests on ARM32 in GHA yamlby merge-script · b048a3e8 · Oct 30, 2026 · 1 fileMessage 91 · StrongTriage 0Details
Commit message · merge-script
Merge bitcoin/bitcoin#36701: ci: Disable tests on ARM32 in GHA yaml
fa01c3c533b8d645a400d23e785d4467b07e8bb5 ci: Disable tests on ARM32 in GHA yaml (MarcoFalke)
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
fuzzing or regression evidencesigning or wallet pathmerge-commit duplicate discountsecond-pass: security-sensitive path
AI analysis · Informational 19/100
This commit removes an old feature in the Bitcoin Core wallet's transaction details window that displayed a merchant name for BIP70 payment requests. BIP70 is an outdated payment protocol that is no longer used. The removed code parsed certificate data in a simple, untested way. The change is described as a cleanup, but it also eliminates a small bit of code that handled untrusted data without proper validation.
Lower-priorityMerge bitcoin/bitcoin#36703: test: Sync blocks in test_witness_block in p2p_segwit.pyby merge-script · 07efc592 · Oct 30, 2026 · 1 fileMessage 91 · StrongTriage 0Details
Commit message · merge-script
Merge bitcoin/bitcoin#36703: test: Sync blocks in test_witness_block in p2p_segwit.py
fab8fd026cfb24c9f246fb60fbff155f3414bacd test: Sync blocks in test_witness_block in p2p_segwit.py (MarcoFalke)
Pull request description:
`test_witness_block` does not sync the blocks across nodes by default, so the test may intermittently fail.
Fixes https://github.com/bitcoin/bitcoin/issues/36700 by adding the sync.
ACKs for top commit: Crypt-iQ: crACK fab8fd026cfb24c9f246fb60fbff155f3414bacd
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
merge-commit duplicate discount
Lower-priorityMerge bitcoin/bitcoin#36594: net: make orphan reconsideration more granularby merge-script · 3c51322d · Oct 30, 2026 · 2 filesMessage 91 · StrongTriage 0Details
Commit message · merge-script
Merge bitcoin/bitcoin#36594: net: make orphan reconsideration more granular
83003cafaa81bb116526419246608658ede824b6 qa: test a single orphan is processed per peer turn, regardless of its validation result (Antoine Poinsot) 28a41e61ab5d993d344dc5c8180482eb146c6bd5 net: reconsider at most one orphan per call, whatever the validation result (Antoine Poinsot)
Pull request description:
We currently concentrate the validation of all the assigned peer’s reconsiderable orphans that are still missing inputs in a single message-processing turn. This ties orphan processing to validation result, and adds latency to processing other peers' messages.
Always process a single reconsiderable orphan per turn instead. This removes the need for an immediate validation result, and shifts the latency from other peers to the assigned peer.
ACKs for top commit: fametrano: tACK 83003cafaa81bb116526419246608658ede824b6 instagibbs: ACK 83003cafaa81bb116526419246608658ede824b6 ViniciusCestarii: ACK 83003cafaa81bb116526419246608658ede824b6 furszy: ACK 83003cafaa81bb116526419246608658ede824b6
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
merge-commit duplicate discount
AI review queueddoc: use doxygen comments for blocks containing @paramby fanquake · 0e9ac2cb · Oct 30, 2026 · 3 filesMessage 73 · AdequateInformational 15Details
Commit message · fanquake
doc: use doxygen comments for blocks containing @param
Plain /* comments are ignored by Doxygen and by clang's -Wdocumentation, so mistakes in them go unnoticed. Also wrap the CoinGrinder ASCII trees in @verbatim so Doxygen keeps their layout.
73/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context
Why it was queued
signing or wallet pathsecond-pass: security-sensitive path
AI analysis · Informational 15/100
This commit only changes code comments. It converts plain block comments into Doxygen-style documentation comments (starting with /** instead of /*) so that developer documentation tools can read them. It also wraps ASCII art diagrams in special tags so they keep their formatting. There are no changes to actual program behavior, and no security impact.
✓ Descriptive subject✓ Names a concrete action or component! No meaningful explanatory body
Why it was queued
signing or wallet pathsecond-pass: security-sensitive path
AI analysis · Informational 15/100
This commit only fixes documentation comments that describe whether function parameters are inputs, outputs, or both. No actual code logic was changed, so it cannot affect program behavior or introduce a security vulnerability.
AI review queuedMerge bitcoin/bitcoin#36071: refactor: fuzz: remove cost_of_change from coin-grinder fuzz targetby merge-script · d1fb1ba3 · Oct 30, 2026 · 1 fileMessage 91 · StrongInformational 15Details
Commit message · merge-script
Merge bitcoin/bitcoin#36071: refactor: fuzz: remove cost_of_change from coin-grinder fuzz target
5ec6446a1b83e673a51d9e86fa4c0c0e9930fccc refactor: fuzz: remove cost_of_change from coin-grinder fuzz target (yancy)
Pull request description:
coin-grinder selection algorithm does not take cost_of_change as a parameter.
> CoinGrinder doesn’t use coin_params.m_cost_of_change and the comparison of selection results from CoinGrinder, SRD, and Knapsack does not need it either as only the weight of the selections is relevant. So, [line 96](https://github.com/bitcoin/bitcoin/blob/2990bd773551c835ab0b6b59d85730adab2fd428/src/wallet/test/fuzz/coinselection.cpp#L96) in that fuzz test is both incorrect and dead code. It should be removed.
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
fuzzing or regression evidencesigning or wallet pathmerge-commit duplicate discountsecond-pass: security-sensitive path
AI analysis · Informational 15/100
This commit removes one unused line from a fuzz test (a type of automated test that throws random data at code to find bugs). The removed line set a 'cost of change' value that the CoinGrinder coin-selection algorithm never actually reads. The change only affects test code, not the real Bitcoin wallet, and does not fix or introduce any security issue.
Lower-priorityMerge bitcoin/bitcoin#36663: test: Use single node in mempool_reorg.pyby merge-script · d14b94dd · Oct 30, 2026 · 1 fileMessage 91 · StrongTriage 0Details
Commit message · merge-script
Merge bitcoin/bitcoin#36663: test: Use single node in mempool_reorg.py
fa3aeb51cfb1f664627730ccfded040a16ad2e91 test: Extract first test case in mempool_reorg.py (MarcoFalke) fa329d62c5194dd4e974bcb3cd314593d17b1a68 test: Use single node in mempool_reorg.py (MarcoFalke) fabf5c52ccba619060610f3da181802c3b193478 test: refactor: Use trigger_reorg helper in mempool_reorg.py (MarcoFalke)
Pull request description:
`mempool_reorg.py` is a bit confusing:
* There is a `trigger_reorg` helper, but it is not used consistently. * It tests tx trickle delay on node1, but the global args have immediate tx relay configured for node0. * Using two nodes is not needed, in light of Python helpers to create forks and P2P connections.
So fix all issues by using a single node and the Python helpers.
ACKs for top commit: jpk68: ACK fa3aeb51cfb1f664627730ccfded040a16ad2e91 mercie-ux: ACK fa3aeb51cfb1f664627730ccfded040a16ad2e91 Waridi318: ACK fa3aeb5 darosior: utACK fa3aeb51cfb1f664627730ccfded040a16ad2e91
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
merge-commit duplicate discount
Lower-priorityMerge bitcoin/bitcoin#36124: http: Make `HTTPRequest` update state internallyby merge-script · 6ed5aad0 · Oct 29, 2026 · 4 filesMessage 93 · StrongTriage 0Details
Commit message · merge-script
Merge bitcoin/bitcoin#36124: http: Make `HTTPRequest` update state internally
6a5212fb97c2c332fc3cb901f8e9a110622ef841 doc: Add missing names to activate grouping syntax (Hodlinator) 148210fe99bb54583fd47f2c39bf0b9390161185 refactor: Flatten HTTPRemoteClient::ReadRequest() into HTTPRemoteClient::TryReadRequestInternal() (Hodlinator) 9692e6f7d00f3fa5b1d57465730c8c50e090a06f refactor: Extract HTTPRemoteClient::TryReadRequestInternal() from static HTTPRemoteClient::TryReadRequest() (Hodlinator) 05897e53b2f92bcd28d2ced7ffd76d82b682b400 http: Extract HTTPRequest::Load() to remove SetState() (Hodlinator)
Pull request description:
### Issues in order of importance
#35829 made the fields of `HTTPRequest` & `HTTPRemoteClient` private, but still allows anyone with a reference to a request to update the internal state `enum`. https://github.com/bitcoin/bitcoin/pull/35829#discussion_r3860447518
`HTTPRemoteClient::TryReadRequest()` only needs to be a `static` method in order to forward a `shared_ptr` of a client to a new request.
### Commits solving these issues in same order
* http: Remove `HTTPRequest::SetState()` and introduce `HTTPRequest::Load()` - Removes the ability for other classes to directly modify the request state. * 2 refactors: Extract `HTTPRemoteClient::TryReadRequestInternal()` from a combination of `HTTPRemoteClient::TryReadRequest()` and `HTTPRemoteClient::ReadRequest()` (the latter is removed)
ACKs for top commit: pinheadmz: ACK 6a5212fb97c2c332fc3cb901f8e9a110622ef841
Part of https://github.com/bitcoin/bitcoin/issues/36692 (previously discussed in https://github.com/bitcoin/bitcoin/pull/35994): replace procedural, field-by-field construction of `COutPoint`, `CTxIn` and `CTxOut` in test code with constructor calls.
This PR covers cases that are context-dependent and need manual changes, independent of a scripted-diff for the trivial bulk: - `test/util/transaction_utils`: in `BuildSpendingTransaction`, the `scriptWitness` assignment has to be moved after `vin` is replaced, otherwise the witness is dropped. - `test/util/mining`: in `CreateBlockChain`, the coinbase input's fields were split by the output assignments, with comments that are now kept next to the constructor arguments. - `sigopcount_tests`: `BuildTxs` has the same witness ordering issue as `BuildSpendingTransaction`. - `bench/ccoins_caching`: The `scriptSig`s are built with `<<` appends, and the second output was created by `vout.resize(2)` but never assigned, so it is kept as a null `CTxOut{}`. - `transaction_tests`: `test_Get` builds the same transaction as `bench/ccoins_caching`, with the same `<<` appends and null second output.
**This change is test- and bench-only and does not change the constructed transactions.**
Notes for reviewers: - `CTxIn` has no witness constructor argument, so `scriptWitness` is assigned after `vin` is replaced. - `BuildTxs` is called repeatedly on the same transaction objects. Previously, `resize(1)` kept the existing inputs and only overwrote some of their fields; now they are replaced entirely. The fields that were left untouched by `BuildTxs` are never modified by the test, so the transactions are unchanged. - The null second output in `ccoins_caching` is copied from `test_Get` in `transaction_tests` and looks like an oversight. It is kept here to avoid behaviour changes.
ACKs for top commit: maflcko: review ACK bdf58dd89cc2fe2f0be5ac63755e821661022be6 🍠 purpleKarrot: But even when the approach does not follow the steps 1 and 2, it is a step in the right direction, hence ACK bdf58dd89cc2fe2f0be5ac63755e821661022be6
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
signing or wallet pathmerge-commit duplicate discountsecond-pass: security-sensitive path
AI analysis · Informational 15/100
This commit is a code cleanup inside Bitcoin Core's tests and benchmarks. It replaces a manual, field-by-field way of building fake transactions with a more direct constructor style. The author explicitly states that the actual transaction data being built is unchanged, and the diff confirms the values and structure stay the same. There is no security issue here.
1176a20d5263d6ae7ff80398928a2a5b1a0ebd6a doc: add release note for -acceptnonstdtxn changes (Anthony Towns) 1df5910720495eb2e8e453fb40da222dd794424a validation: Add compile-time checks for consensus flags (Anthony Towns) 425115713c5e37e78c9051b1b0a997ad62067ec8 validation: Check only (potential) block script flags when -acceptnonstdtxn is set (Anthony Towns) 703ebdf9de87ffb6010d89e3f0bd49e3c6cdb030 validation: Be explicit about whether CheckInputScripts is for block/mempool (Anthony Towns) 81ef18236fc74913344bd9e28b1e27a50a4e3ae7 tests: test tx rejection for both -acceptnonstdtxn=0 and 1 (Anthony Towns) 9b6342e5d295e337791984af2d42681d9c8e08bb tests: in p2p_segwit, check non-mandatory errors with -acceptnonstdtxn=0 node (Anthony Towns)
Pull request description:
This PR changes `-acceptnonstdtxn=1` so that it also skips the non-mandatory script checks, allowing txs that (eg) use OP_NOPx or OP_SUCCESS into the mempool. This remains only available on test nets.
ACKs for top commit: fjahr: reACK 1176a20d5263d6ae7ff80398928a2a5b1a0ebd6a instagibbs: reACK 1176a20d5263d6ae7ff80398928a2a5b1a0ebd6a
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
consensus or confidential-proof validationmerge-commit duplicate discount
AI analysis · Informational 22/100
This change loosens transaction acceptance rules only on Bitcoin test networks when a special command-line option (-acceptnonstdtxn=1) is used. It lets developers test unusual or experimental scripts in the mempool, while still enforcing all consensus rules that protect real money. Main network behavior is unchanged, and the option is not available there.
AI review queuedMerge bitcoin/bitcoin#36682: refactor: Use std::erase and std::erase_ifby merge-script · 363d9e9d · Oct 29, 2026 · 13 filesMessage 93 · StrongInformational 15Details
Commit message · merge-script
Merge bitcoin/bitcoin#36682: refactor: Use std::erase and std::erase_if
fa87bfe31dfda2d102c4e0cb5d39813cf84c7d1e refactor: Use std::erase and std::erase_if over iterator-based erase (MarcoFalke) fa93a02563145aca3f96badaabe302b2dfa331d5 refactor: Use std::erase and std::erase_if over list removal or vector erase-remove (MarcoFalke)
Pull request description:
Convert `remove` and `remove_if` to the C++20 `std::erase` and `std::erase_if` overloads.
The legacy ones can be confusing, because:
* Array-based containers (like vector) must use the erase-remove idiom. Forgetting the erase, or forgetting to pass the end iterator (`bugprone-inaccurate-erase`) can be a bug, see also https://github.com/bitcoin/bitcoin/pull/36588#discussion_r4345799022. So a call only to "remove" will not erase. * Node-based containers (like list or set) must not use the idiom. Using it will be a pessimization or won't compile. And correctly not using the idiom reads confusing, because a call to "remove" will actually erase. * They are more verbose.
Also convert simple iterator-erase (which is verbose, but fine) to the std overload.
ACKs for top commit: jpk68: ACK fa87bfe31dfda2d102c4e0cb5d39813cf84c7d1e hodlinator: ACK fa87bfe31dfda2d102c4e0cb5d39813cf84c7d1e w0xlt: ACK fa87bfe31dfda2d102c4e0cb5d39813cf84c7d1e
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode✓ Links an issue, advisory, or supporting reference
Why it was queued
signing or wallet pathparser or protocol pathmerge-commit duplicate discountsecond-pass: security-sensitive path
AI analysis · Informational 15/100
This commit is a code cleanup that replaces older, more verbose ways of removing items from lists and vectors with newer C++20 standard helpers called std::erase and std::erase_if. It does not change what the program does, only how the source code expresses it. There is no security bug being fixed here.
**Problem:** When flushing block or undo data fails, `FlushStateToDisk()` still writes block-index metadata and coins data, then advances `m_last_flushed_block`. With a separate `-blocksdir`, persistent failures there can leave metadata referring to block data that is not durable even when chainstate writes succeed.
**Fix:** Return the flush error before those writes, leaving the last-flushed marker and on-disk coins tip unchanged in the failing call. Later flushes can retry and write if they succeed, including during shutdown. `ConnectTip()` and `DisconnectTip()` propagate the error before updating the chain tip, after applying the coins changes in memory.
**History:** * [#27866](https://github.com/bitcoin/bitcoin/pull/27866#discussion_r1297313148) raised whether saving other state after a flush failure was preferable. This change preserves later retries while skipping writes in the failing call. * [#34897](https://github.com/bitcoin/bitcoin/pull/34897) introduced `m_last_flushed_block` to prevent persistent indexes from advancing past flushed chainstate. This fix keeps it at the last successful flush.
ACKs for top commit: optout21: reACK 12f691cf286ea9a020583d267b94efca102a4056 arejula27: reACK 12f691cf286ea9a020583d267b94efca102a4056 pzafonte: reACK 12f691cf286ea9a020583d267b94efca102a4056 maflcko: review ACK 12f691cf286ea9a020583d267b94efca102a4056 👟 ryanofsky: Code review ACK 12f691cf286ea9a020583d267b94efca102a4056. Thanks for putting the test first, it makes the behavior change easy to verify.
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference