Newer versions of the macOS SDK, have introduced code like:
```cpp
if defined(__has_feature) && __has_feature(modules)
define USE_CLANG_TYPES 1
else
define USE_CLANG_TYPES 0
endif
if USE_CLANG_TYPES
include <sys/_types/_ptrdiff_t.h>
include <sys/_types/_size_t.h>
include <sys/_types/_va_list.h>
include <sys/_types/_wchar_t.h>
endif
```
which is currently causing compile failures due to undeclared types,
which manifest when C++ modules are enabled. Note that the usage of
"modules" in LLVM is can be a bit ambiguous, see:
https://github.com/llvm/llvm-project/issues/55891.
For now, explcitly disable cxx modules using `-fno-cxx-modules`. This
resolves the include/compilation issues.
Related discussion:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116827https://github.com/llvm/llvm-project/pull/150349https://github.com/llvm/llvm-project/issues/57432
Add m_connect_block_view to ChainState's CoinsViews.
Call CreateResetGuard inside ConnectTip to ensure the view
is Reset after each block, avoiding repeated memory allocations.
Co-authored-by: l0rinc <pap.lorinc@gmail.com>
CCoinsViewCache::CreateResetGuard returns a guard that calls
Reset on the cache when the guard goes out of scope.
This RAII pattern ensures the cache is always properly reset
when it leaves current scope.
Co-authored-by: l0rinc <pap.lorinc@gmail.com>
Co-authored-by: sedited <seb.kung@gmail.com>
Add a Reset() method to CCoinsViewCache that clears cacheCoins,
cachedCoinsUsage, and hashBlock without flushing to the base view.
Co-authored-by: l0rinc <pap.lorinc@gmail.com>
Co-authored-by: sedited <seb.kung@gmail.com>
Use the AFL++ shared memory ID environment variable to create
a deterministic datadir path. This prevents accumulation of stale
directories after a fuzz iteration crashes or times out. During
long fuzz campaigns, this accumulation has occasionally resulted
in running out of disk space.
2845f10a2be0fee13b2772d24e948052243782b8 test: extend FreeBSD ephemeral port range fix to P2P listeners (node)
34bed0ed8c449a3834927cec3447dbe6c74edf3d test: use IP_PORTRANGE_HIGH on FreeBSD for dynamic port allocation (woltx)
Pull request description:
Reopening #34336. I’ve now tested it on FreeBSD and confirmed it works.
On FreeBSD, the default ephemeral port range (10000-65535) overlaps with the test framework's static port range (11000-26000), possibly causing intermittent "address already in use" failures when tests use dynamic port allocation (`port=0`).
This PR adds a helper that sets `IP_PORTRANGE_HIGH` via `setsockopt()` before binding, requesting ports from 49152-65535 instead, which avoids the overlap, as suggested in https://github.com/bitcoin/bitcoin/issues/34331#issuecomment-3767161843 by @maflcko .
From FreeBSD's [sys/netinet/in.h](https://cgit.freebsd.org/src/tree/sys/netinet/in.h):
```c
#define IP_PORTRANGE 19
#define IP_PORTRANGE_HIGH 1
#define IPPORT_EPHEMERALFIRST 10000 /* default range start */
#define IPPORT_HIFIRSTAUTO 49152 /* high range start */
```
See also: FreeBSD https://man.freebsd.org/cgi/man.cgi?query=ip&sektion=4 man page.
Fixes#34331
ACKs for top commit:
vasild:
ACK 2845f10a2be0fee13b2772d24e948052243782b8
hebasto:
ACK 2845f10a2be0fee13b2772d24e948052243782b8, I have reviewed the code and it looks OK.
Tree-SHA512: ce501ce3e8a4023e07bad572df2b85d6829becf133813e4529aebba83e4eba59fa8b48e9d2197ebbb226adaf3054fad720775a787244d6b38c0078ee086102f4
Add `AssertLockHeld(cs_main)` to `ChainstateManager::UpdateIBDStatus()` given `EXCLUSIVE_LOCKS_REQUIRED(cs_main)`.
Fix outdated comment about constness of `ChainstateManager::IsInitialBlockDownload()` (build passes without it).
And since we're touching it, we might as well mark `ChainstateManager::IsInitialBlockDownload()` as `noexcept` now.
Follow-up to #34253.
Co-authored-by: Martin Zumsande <mzumsande@gmail.com>
fad042235bd6054d99d3f5a07529276b0138b484 refactor: Remove remaining std::bind, check via clang-tidy (MarcoFalke)
Pull request description:
`std::bind` has many issues:
* It is verbose in a meaningless way
* Overriden args are silently accepted and dropped at runtime without a compile error. Same for accidental duplicates.
One could use `std::bind_front` similar to commit fa267551c4eaef577db92e248c4b6d31d0c8bc77. Though, I think the remaining cases are better off with lambdas.
So do that here, and enable the `modernize-avoid-bind` clang-tidy rule to avoid `std::bind` bugs in the future.
ACKs for top commit:
fjahr:
Code review ACK fad042235bd6054d99d3f5a07529276b0138b484
purpleKarrot:
Code review ACK fad042235bd6054d99d3f5a07529276b0138b484
Tree-SHA512: 38b17e26eda3ae47d84a8c34298309dc1eeb4ed434fda58b5803ef031c4c2edfb17222f5208f37af727bf340e32b37c7f81784f461d2b65fbc6227f3cd53eea4
Currently, if the node replies to any command with an error during
the test framework's setup(), we log the generic and not really useful
"Unexpected exception" from the BaseException catch, with no further
information.
This isn't helpful for diagnosing the issue. Fix it by explicitly handling
JSONRPCException and logging the response error message and http status
code.
Currently, if an exception is thrown at the top-level HTTP request
handler (prior to invoking the command), the program crashes.
Ideally, each handler should catch all exceptions internally and
be responsible for sanitizing them and crafting the client response.
This is because only the handler knows the correct response format,
which differs per server type. However, because this cannot always
be guaranteed, it is safer to also catch exceptions in the top-level
server code, log the unexpected error, and disconnect the socket.
This both guards against crashes caused by uncaught exceptions and
prevents the client from hanging indefinitely while waiting for a
response that will never arrive.
The following diff can be used to trigger the crash in master
(just run single node functional tests like feature_shutdown.py):
```
diff --git a/src/httprpc.cpp b/src/httprpc.cpp
--- a/src/httprpc.cpp
+++ b/src/httprpc.cpp
@@ -103,6 +103,9 @@
static bool HTTPReq_JSONRPC(const std::any& context, HTTPRequest* req)
{
+ static int i = 0; // skip initial requests as they are used in the RPC warmup phase.
+ if (i++ > 3) throw std::runtime_error("error from json rpc handler");
+
// JSONRPC handles only POST
if (req->GetRequestMethod() != HTTPRequest::POST) {
req->WriteReply(HTTP_BAD_METHOD, "JSONRPC server handles only POST requests");
```
Note:
This leaves a TODO in the code because error responses should eventually
be specialized per server type. REST clients expect plain text responses,
while JSON-RPC clients expect a JSON error object.
The TODO is there because this is not consistently enforced everywhere
in the current codebase, and we should tackle them all at once.
e770392084aa52e5568cf001da4d537fda1d71b3 test: addrman: test self-announcement time penalty handling (Bruno Garcia)
Pull request description:
This PR adds a test case for addrman that verifies that addresses announcing themselves (addr == source) are exempt from time penalties, while addresses announced by others receive the expected penalty.
It fixes the following mutant (https://corecheck.dev/mutation/src/addrman.cpp#L561):
```diff
diff --git a/src/addrman.cpp b/src/addrman.cpp
index 206b54118e..c6a045fd8d 100644
--- a/src/addrman.cpp
+++ b/src/addrman.cpp
@@ -558,7 +558,7 @@ bool AddrManImpl::AddSingle(const CAddress& addr, const CNetAddr& source, std::c
AddrInfo* pinfo = Find(addr, &nId);
// Do not set a penalty for a source's self-announcement
- if (addr == source) {
+ if (addr != source) {
time_penalty = 0s;
}
```
ACKs for top commit:
maflcko:
review ACK e770392084aa52e5568cf001da4d537fda1d71b3 🐤
achow101:
ACK e770392084aa52e5568cf001da4d537fda1d71b3
fjahr:
Code review ACK e770392084aa52e5568cf001da4d537fda1d71b3
naiyoma:
tACK e770392084aa52e5568cf001da4d537fda1d71b3
Tree-SHA512: ec029d1e1e979f91840af944984cad530a1ce9a0eceb123230817f0ef3b9ad47253eebc4c953d350de2d904b59496fcd4757123c8bd63cf0e09c3581da48fff8
2ee7f9b259059d59e127852ea898b58183604b46 coins: assume `GetCoin` only returns unspent coins (Andrew Toth)
eec551aaf1dff4cccc15e486d5618a8a44d8314c fuzz: keep `coinscache_sim` backend free of spent coins (Andrew Toth)
3e4155fcefe0aafcc9cb84640e303e05477605a3 test: do not return spent coins from `CCoinsViewTest::GetCoin` (Andrew Toth)
ee1e40f58000921e95f08bcb199a452eb5c4d9b2 txdb: assert `CCoinsViewDB::GetCoin` only returns unspent coins (Lőrinc)
Pull request description:
This PR is split out from #33018 to keep that PR focused on removing the `FRESH-but-not-DIRTY` cache state.
### Problem
`::GetCoin()` is an interface for querying the UTXO set, so production implementations should only ever return unspent coins. Tests should mimic this to provide useful feedback.
### Fix:
* Add a fail-fast assertion that `CCoinsViewDB::GetCoin()` never returns a spent coin.
* Align unit tests and fuzz simulations with the production `GetCoin()` contract by never returning spent coins.
* Replace the unreachable “spent coin returned by parent” handling in `CCoinsViewCache::FetchCoin()` with `Assert(!coin.IsSpent())`, drop outdated `spent+FRESH` docs, and tighten `SanityCheck()` invariants.
Behavior is unchanged, it just aligns our tests to exercise valid states.
ACKs for top commit:
andrewtoth:
re-ACK 2ee7f9b259059d59e127852ea898b58183604b46
optout21:
crACK 2ee7f9b259059d59e127852ea898b58183604b46
achow101:
ACK 2ee7f9b259059d59e127852ea898b58183604b46
w0xlt:
reACK 2ee7f9b259059d59e127852ea898b58183604b46
Tree-SHA512: be21cc09690410fc04ca25e1ba47aae6186bc037e413b3bb1e6e9a04e6364cbfac5a2fcdc49b638fec848cd29243fab0cc0581b9923f34fafe8366828f690ed4
and_v() must never be 'd'. This is not a bug fix since this was
unreachable in valid Miniscripts: the first sub of an and_v() must be of
type V, which conflicts with (i.e. never has) property 'd'.
It was introduced in ab9edbd6b6eb3efbca11f16fa467c3c0ef905708 and never
used since. It seems it might have been intended to be exposed as part
of a public library interface, which has since been superseded.
The only call site uses SCRIPT_ERR_ERROR_COUNT directly.
1f60ca360eb83fa7982b1aac402eaaf477294197 wallet: fix removeprunedfunds bug with conflicting transactions (Martin Zumsande)
Pull request description:
`removeprunedfunds` removes all entries from `mapTxSpends` for the inputs of the pruned tx. However, this is incorrect, because there could be multiple entries from conflicting transactions (that shouldn't be removed as well). This could lead to the wallet creating invalid transactions, trying to double spend utxos.
The bug persists when the conflicting tx was mined, because the wallet trusts its internal accounting instead of calling `AddToSpends` again.
The added test should fail on master.
ACKs for top commit:
achow101:
ACK 1f60ca360eb83fa7982b1aac402eaaf477294197
fjahr:
tACK 1f60ca360eb83fa7982b1aac402eaaf477294197
furszy:
utACK 1f60ca360eb83fa7982b1aac402eaaf477294197
vasild:
ACK 1f60ca360eb83fa7982b1aac402eaaf477294197
Tree-SHA512: 3cc9ed547530fd53e25721177b76ab2e1eae16ce2c0e63fc01b20fdbf8bd02655dae51167ad56f9dec748d34c61ce65d38f993370820601f8257c73b876a3347
905dfdee86d679f8ea31d841bceb77a5724a6b1b test: use ModuleNotFoundError in interface_ipc.py (fanquake)
Pull request description:
Change this so we catch the case where the capnp shared libs have been updated, and can no-longer be loaded by the Python module, resulting in a skipped test, even though pycapnp is installed. i.e:
```bash
stderr:
Traceback (most recent call last):
File "/root/ci_scratch/build/test/functional/interface_ipc.py", line 20, in <module>
import capnp # type: ignore[import] # noqa: F401
^^^^^^^^^^^^
File "/usr/local/lib64/python3.14/site-packages/capnp/__init__.py", line 36, in <module>
from .version import version as __version__
File "/usr/local/lib64/python3.14/site-packages/capnp/version.py", line 1, in <module>
from .lib.capnp import _CAPNP_VERSION_MAJOR as LIBCAPNP_VERSION_MAJOR # noqa: F401
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ImportError: libcapnpc.so.1.0.1: cannot open shared object file: No such file or directory
```
Failing in this way should make it clear that `pycapnp` needs to be reinstalled/rebuilt.
If `pycapnp` is not installed, the test still skips as expected:
```bash
Remaining jobs: [interface_ipc.py]
1/1 - interface_ipc.py skipped (capnp module not available.)
TEST | STATUS | DURATION
interface_ipc.py | ○ Skipped | 0 s
```
Fixes: #34016.
ACKs for top commit:
maflcko:
lgtm ACK 905dfdee86d679f8ea31d841bceb77a5724a6b1b
hebasto:
ACK 905dfdee86d679f8ea31d841bceb77a5724a6b1b, I have reviewed the code and it looks OK. However, I'm [not able](https://github.com/bitcoin/bitcoin/issues/34016#issuecomment-3799532047) to reproduce https://github.com/bitcoin/bitcoin/issues/34016.
sedited:
ACK 905dfdee86d679f8ea31d841bceb77a5724a6b1b
Tree-SHA512: 3cedbe8fc51cc18f1c993f7747d20905f3bf94c736db99a9c4090f5823bf8c09dfbc19ef03c573d504dcdfba6ea0f7d088a7f4563b220742c9a441167c04cfd6
fa9c92d7b6394194696e014e48346a8b8562e16a log: Print warning about privacy-sensitive log info unconditionally (MarcoFalke)
Pull request description:
There is a warning about logs containing privacy-sensitive information. However, it is only printed when at least one debug log category is enabled.
This is confusing, because:
* Setting (let's say) `-debug=reindex` enables this warning, but it is hard to see what sensitive logs could be contained in reindex debug logs.
* Dropping `-debug=reindex` again disabled this warning, but the wallet continues to log txids (and other sensitive stuff) at info level.
So instead of implying the wrong thing, it would be better to remove this log line (because it should be common sense), or log it unconditionally.
ACKs for top commit:
l0rinc:
ACK fa9c92d7b6394194696e014e48346a8b8562e16a
sedited:
ACK fa9c92d7b6394194696e014e48346a8b8562e16a
Tree-SHA512: 42f71b030e7722203f225f04e979143e829dae3556f64e322a791361a3b9c16150d53bb7bb9a99839c975d9052115770b9473138acc58baeee457253526fd892
ab649ce459450533ffdf068b00472259b72cf07a guix: documented shasum gathering command (janb84)
Pull request description:
When a PR requires proof of Guix builds (sha256sums), the PR author or reviewer uses a not well documented command to collect the sha256sums of build outputs or manually gathers them from files.
This pull request introduces a new section in the documentation, providing some documentation on the command's functionality and usage.
ACKs for top commit:
willcl-ark:
ACK ab649ce459450533ffdf068b00472259b72cf07a
sedited:
ACK ab649ce459450533ffdf068b00472259b72cf07a
Tree-SHA512: 0188663ad117b636c7d32a1b655db97610f558cfcffe4abd6f0fb097b3990db0dc6d23ab972926fefd2531b21f429742dcbea6b0fa579d22d5da7a7d6a4c753e
fa578d9434fdb090d27c7b5598dcd7f0ff0965cc lint: [move-only] Move python related lints to lint_py.rs (MarcoFalke)
fa392c31e7b9a487b61c881b06bb1e3787dcf4d9 lint: [move-only] Move repo related lints to lint_repo_hygiene.rs (MarcoFalke)
fab0cfa987c916a17c9cdd57636798866a56d74d lint: [move-only] Move cpp related lints to lint_cpp.rs (MarcoFalke)
fa3e48e3fd4d01dd8453994d214c24297aa83f62 lint: [move-only] Move docs related lints to lint_docs.rs (MarcoFalke)
fad09e77dbe521eb35206cd72536619225b18550 lint: [move-only] Move text related lints to text_format.rs (MarcoFalke)
faf40c2f848d7b5542862a4fd10b41cc6d418a90 lint: [move-only] Move util functions to util.rs (MarcoFalke)
Pull request description:
The single, large `main.rs` file is fine, but at some point it becomes harder to read.
So reduce the size by pulling functions out into modules.
This can be reviewed with the git option: `--color-moved=dimmed-zebra`
ACKs for top commit:
l0rinc:
Lightly tested code review ACK fa578d9434fdb090d27c7b5598dcd7f0ff0965cc
sedited:
ACK fa578d9434fdb090d27c7b5598dcd7f0ff0965cc
Tree-SHA512: f1e29fd3cf695fb6634d0b9f9e55508992b4b9885afee9dbe4d5d9e99cad3061e7141f39acbfe69d698422888169128cd7658a6dc991fd904b8520328b51586d
fdc9fe2da6a8640b11a2871f8b653764652f8c1f ci, iwyu: Fix warnings in `src/primitives` and treat them as errors (Hennadii Stepanov)
Pull request description:
This PR [continues](https://github.com/bitcoin/bitcoin/pull/33725#issuecomment-3466897433) the ongoing effort to enforce IWYU warnings.
See [Developer Notes](https://github.com/bitcoin/bitcoin/blob/master/doc/developer-notes.md#using-iwyu).
ACKs for top commit:
maflcko:
review ACK fdc9fe2da6a8640b11a2871f8b653764652f8c1f 📀
janb84:
ACK fdc9fe2da6a8640b11a2871f8b653764652f8c1f
sedited:
ACK fdc9fe2da6a8640b11a2871f8b653764652f8c1f
Tree-SHA512: d290545c7aab477b4a5bf121b694899a78e0526be72efa31fa4205b0fd840e6e8240d32f9134a18c9dc58c5f91e7847d7f20ca34f8d2edc4d541ac858ec0dccc
Otherwise, this may pick the wrong arch like s390x, if a such a podman
container was most recently used.
See also the CI_IMAGE_PLATFORM setting in the "other" CI, which does the
same.
c8abac994122b67de8145579489576db5df2b3d3 ci: mount .git dir rw (ci)
Pull request description:
On merges to master we set LINT_CI_SANITY_CHECK_COMMIT_SIG (when "GITHUB_REPOSITORY == bitcoin/bitcoin") which runs verify-commits.py.
This requires write access to the .git directory.
Make the mounted .git directory writable.
This is currently not run on PR branches or locally which caused a miss during review.
Ideally we can have the same checks running in PRs as on merges to master to avoid future discrepancies like this.
ACKs for top commit:
maflcko:
lgtm ACK c8abac994122b67de8145579489576db5df2b3d3
l0rinc:
untested code review ACK c8abac994122b67de8145579489576db5df2b3d3
Tree-SHA512: 7ae4f63227ecffe1dc9003454a7473d6d592550af2e1c899457f34a947e5604b04c13319fb8979f36789ae7787bed62066be60697d163ad5ebedde3fbe8ce45f
On merges to master we set LINT_CI_SANITY_CHECK_COMMIT_SIG (when
"GITHUB_REPOSITORY == bitcoin/bitcoin") which runs verify-commits.py.
This requires write access to the .git directory.
Make the mounted .git directory writable.
This is currently not run on PR branches or locally which caused a miss
during review.
3400db80401d65ba16b52e5055486c75cd1412ff doc: add missing param description to SRD (yancy)
Pull request description:
The params documentation is missing `change_fee` and the description is lacking recent changes.
ACKs for top commit:
murchandamus:
ACK 3400db80401d65ba16b52e5055486c75cd1412ff
brunoerg:
code review ACK 3400db80401d65ba16b52e5055486c75cd1412ff
Tree-SHA512: 8f6fac0d92873c5c9f77b19fbc0c6ecfb425b2a6b3d5f5ad69c82ed706b21cf4627e68c71acbc43661000e6063e8f8dbcd3b8ff60e3c727bdcba497d13ee1383
ddae1b4efa568b6099d5bece941f8f2733faf29e ci: remove gnu-getopt usage (fanquake)
Pull request description:
This is used for argument parsing in the `retry` script, however we don't use the script with any arguments. So remove the unused code, and the dependency on `gnu-getopt`.
This came up in the context of adding new CI jobs, where gnu-getopt might not be available, or working properly. It seemed easier to just remove the unused code, than look for more workarounds.
ACKs for top commit:
maflcko:
review ACK ddae1b4efa568b6099d5bece941f8f2733faf29e 🔀
sedited:
ACK ddae1b4efa568b6099d5bece941f8f2733faf29e
Tree-SHA512: a73cf61fe0965127f87f1725b3a25a305ebfd354c318f5f44ecfa20da02ba72fef42dca656dae07f6e1ece956b9d7c58e99edb124d968a4bffb2ce6ac8fc018b
fad7bd9ba3eef03fcdd7cb17011ea0c6e483c767 noui: Remove always empty caption while formatting (MarcoFalke)
fa8ebeb332325604e8ca6080262543e10de4e46c refactor: [gui] Document that the title is always empty for node message (MarcoFalke)
fafe71b743a0637d16812d26430d99464cab0cee refactor: Remove empty caption from ThreadSafeMessageBox (MarcoFalke)
fa8d0088e76d4def59dff92bfb2ebbfc6cd4c195 refactor: Remove empty caption from ThreadSafeQuestion (MarcoFalke)
fa0195499ca611b513d9d1986d79c5e3a58cd0f2 refactor: [gui] Use lambdas over std::bind (MarcoFalke)
eeee1e341fa59b5b0b05f974105104fb2a0df9c3 refactor: Remove trailing semicolon after ADD_SIGNALS_DECL_WRAPPER (MarcoFalke)
Pull request description:
Currently, the user interface (noui, gui) has a caption for each message. However, the caption has many issues:
* It is always hard-coded to the empty string.
* This is confusing and tedious when reading or maintaining the code.
* It is redundant, because `noui` will ignore the caption and set the logging prefix (error, warning, info) based on the `style`.
* The gui does prefer to set the title based on the caption, but since it the caption is always empty, the fallback will always be used.
Fix all issues by removing it.
ACKs for top commit:
hebasto:
ACK fad7bd9ba3eef03fcdd7cb17011ea0c6e483c767, I have reviewed the code and it looks OK. Tested on Ubuntu 25.10.
sedited:
ACK fad7bd9ba3eef03fcdd7cb17011ea0c6e483c767
Tree-SHA512: 58ef538b9b3e1cfdcf2955f6de9b8cee335edbf6339723cb693cb4d584817904c962dac5199ee44d7e2860a5332dec1a6abf47e621eb5cf919aa1cdae271b55f
477c5504e05f9031449cdbf62bf329eac427cb0c coins: replace `std::distance` with unambiguous pointer subtraction (Lőrinc)
Pull request description:
### Problem
Calling `std::distance(nullptr, nullptr)` has ambiguous status in the C++ standard [iterator.requirements.general](https://eel.is/c++draft/iterator.requirements.general#7):
> Iterators can also have singular values that are not associated with any sequence. Results of most expressions are undefined for singular values.
It seems to work correctly in every implementation we use, but [LWG 1213](https://cplusplus.github.io/LWG/issue1213) ("Meaning of valid and singular iterator underspecified") has been Open since 2009, acknowledging that the standard's wording on this topic is unclear.
<details>
<summary>Details</summary>
The [iterator.requirements.general](https://eel.is/c++draft/iterator.requirements.general#7) states:
> Iterators can also have singular values that are not associated with any sequence. Results of most expressions are undefined for singular values.
And [LWG 208](https://cplusplus.github.io/LWG/issue208)'s rationale explicitly confirms:
> Null pointers are singular.
Therefore they cannot form a valid range required by [std::distance](https://eel.is/c++draft/iterator.operations#4):
> Preconditions: last is reachable from first, or InputIterator meets the Cpp17RandomAccessIterator requirements and first is reachable from last.
</details>
### Fix
A previous version of this PR checked both values for `nullptr`, the current one uses unambiguously well-defined pointer subtraction instead, which is per [expr.add](https://eel.is/c++draft/expr.add#5):
> If P and Q both evaluate to null pointer values, the value is 0.
This applies on the first call before any memory is allocated, when both pointers are `nullptr`.
Using `operator-` directly is simpler and avoids the ambiguity entirely.
ACKs for top commit:
maflcko:
review ACK 477c5504e05f9031449cdbf62bf329eac427cb0c 🍶
optout21:
ACK 477c5504e05f9031449cdbf62bf329eac427cb0c
sedited:
ACK 477c5504e05f9031449cdbf62bf329eac427cb0c
Tree-SHA512: 5edfb19ab4820e2003928f60f20d4a5893bcd3c316afdfe91c9c06e9b465352769b2cddb0d0e2419ea083a906d35f4aada74149e81f4ea0315f8173ac538789f
5aeaa71c77ac31a1b05f8361356d2810cfc5bc28 lint: pass args from lint.py to cargo run in container (will)
c17a2adb8dc0e9706319ddecc42737be4268ac6a lint: upgrade lint scripts for worktrees (will)
Pull request description:
Fixes#29972
Use a single script to run the linter locally or in CI.
Works from inside a worktree.
ACKs for top commit:
maflcko:
review ACK 5aeaa71c77ac31a1b05f8361356d2810cfc5bc28 🔒
davidgumberg:
code review and lightly tested reACK 5aeaa71
l0rinc:
Tested (+ lightly reviewed) ACK 5aeaa71c77ac31a1b05f8361356d2810cfc5bc28
Tree-SHA512: 7c11f649b4752739d31c4f9e6306a98bd2e615b27a0819bbb5e7d9284b9e28bd9f424e145f16361f672f1a63441a1ae2f901c4f99759e997b72a4bf2d56d8d39
fa2e1b85dd6b263a77294566bd361d7873731045 build: Remove outdated comment about -ffile-prefix-map (MarcoFalke)
fa06cd4ba7303bd3036467cf00d12f36c94bcaff doc: Remove outdated -fdebug-prefix-map section in dev notes (MarcoFalke)
Pull request description:
This removes some docs. See the commit messages for an explanation.
ACKs for top commit:
l0rinc:
ACK fa2e1b85dd6b263a77294566bd361d7873731045
sedited:
ACK fa2e1b85dd6b263a77294566bd361d7873731045
Tree-SHA512: 6be33bdf9365be5fb75d39a48fd1295b193649775a00e8344123dc0f588da22f7efe80b1490dde2c74aea3d7fec6a3fa75785791296f3fb248ddf45e40b95eb7
e71c4df1685131f5ab48aac6ccb07ac944e91e9f refactor: replace manual promise with SyncWithValidationInterfaceQueue (ANtutov)
Pull request description:
`BroadcastTransaction()` now waits for validation callbacks using the built-in `validation_signals>SyncWithValidationInterfaceQueue()` instead of creating a local `std::promise` and scheduling a lambda. This removes an unnecessary allocation and uses the canonical API.
ACKs for top commit:
maflcko:
review ACK e71c4df1685131f5ab48aac6ccb07ac944e91e9f 🌃
rkrux:
lgtm ACK e71c4df1685131f5ab48aac6ccb07ac944e91e9f
sedited:
ACK e71c4df1685131f5ab48aac6ccb07ac944e91e9f
Tree-SHA512: 602994ba3c2ac91996068aee6eac7e788c3832d7ab949519a9420d2b59e2a67d2d4e67c3c9191ba60e9caa75f1524a95b0851fcd40b6732f6a9956a011b4a120
This is used for argument parsing in the retry script, however we don't
use the script with any arguments. So remove the unused code, and the
dependency on gnu-getopt.
This came up in the context of adding new CI jobs, where gnu-getopt
might not be available, or working properly. It seemed easier to just
remove the unused code, than look for more workarounds.
This allows to skip nullptr checks later in the code, both mentally and literally.
This can be reviewed via the git option:
--word-diff-regex=.
-BEGIN VERIFY SCRIPT-
sed --regexp-extended --in-place '
/^bool PeerManagerImpl::SendMessages\(/,/^}$/ {
s/auto& peer\{maybe_peer\}; .. alias cleaned up .*/Peer\& peer{*maybe_peer};/;
s/peer->/peer./g;
s/\*peer\>/peer/g;
/CNode\* pto\{&node\}; .. alias removed .*/d;
s/pto->/node./g;
s/\*pto\>/node/g;
}
' src/net_processing.cpp
sed --regexp-extended --in-place '
/^void PeerManagerImpl::ProcessMessage\(/,/^}$/ {
/Peer\* peer\{&peer_alias_removed_in_later_commit};/d;
s/peer_alias_removed_in_later_commit/peer/;
s/peer->/peer./g;
s/\*peer\>/peer/g;
}
' src/net_processing.cpp
sed --regexp-extended --in-place '
/^bool PeerManagerImpl::ProcessMessages\(/,/^}$/ {
s/auto& peer\{maybe_peer\}; .. alias cleaned up .*/Peer\& peer{*maybe_peer};/;
s/peer->/peer./g;
s/\*peer\>/peer/g;
/CNode\* pfrom\{&node\}; .. alias removed .*/d;
s/pfrom->/node./g;
s/\*pfrom\>/node/g;
}
' src/net_processing.cpp
-END VERIFY SCRIPT-
Calling this low-level function from tests is confusing, and also makes
it harder to change the peer manager implementation.
So juse use the pre-existing test helpers to achieve the same.
