The context of this commit is an effort to decouple the dbwrapper header
file from leveldb includes. To this end, the includes are moved to the
dbwrapper implementation file. This is done as part of the kernel
project to reduce the number of required includes for users of the
kernel.
Keep the generic serialization in the header, while moving
leveldb-specifics to the implementation file.
Since CharCast is no longer needed in the header, move it to the
implementation file.
The context of this commit is an effort to decouple the dbwrapper header
file from leveldb includes. To this end, the includes are moved to the
dbwrapper implementation file. This is done as part of the kernel
project to reduce the number of required includes for users of the
kernel.
Make it a static function in dbwrapper.cpp, since it is not used
elsewhere and when left in the header, would expose a leveldb type.
The context of this commit is an effort to decouple the dbwrapper header
file from leveldb includes. To this end, the includes are moved to the
dbwrapper implementation file. This is done as part of the kernel
project to reduce the number of required includes for users of the
kernel.
Keep the generic serialization in the header, while moving
leveldb-specifics to the implementation file.
The context of this commit is an effort to decouple the dbwrapper header
file from leveldb includes. To this end, the includes are moved to the
dbwrapper implementation file. This is done as part of the kernel
project to reduce the number of required includes for users of the
kernel.
Keep the generic serialization in the header, while moving
leveldb-specifics to the implementation file.
The context of this commit is an effort to decouple the dbwrapper header
file from leveldb includes. To this end, the includes are moved to the
dbwrapper implementation file. This is done as part of the kernel
project to reduce the number of required includes for users of the
kernel.
Hide the leveldb::Iterator member variable with a pimpl in order not to
expose it directly in the header.
Also, move CDBWrapper::NewIterator to the dbwrapper implementation to
use the pimpl for CDBIterator initialziation.
The context of this commit is an effort to decouple the dbwrapper header
file from leveldb includes. To this end, the includes are moved to the
dbwrapper implementation file. This is done as part of the kernel
project to reduce the number of required includes for users of the
kernel.
Keep the generic serialization in the header, while moving
leveldb-specifics to the implementation file.
The context of this commit is an effort to decouple the dbwrapper header
file from leveldb includes. To this end, the includes are moved to the
dbwrapper implementation file. This is done as part of the kernel
project to reduce the number of required includes for users of the
kernel.
Keep the generic serialization in the header, while moving
leveldb-specifics to the implementation file.
The context of this commit is an effort to decouple the dbwrapper header
file from leveldb includes. To this end, the includes are moved to the
dbwrapper implementation file. This is done as part of the kernel
project to reduce the number of required includes for users of the
kernel.
Keep the generic serialization in the header, while moving
leveldb-specifics to the implementation file.
The context of this commit is an effort to decouple the dbwrapper header
file from leveldb includes. To this end, the includes are moved to the
dbwrapper implementation file. This is done as part of the kernel
project to reduce the number of required includes for users of the
kernel.
Hide the leveldb::WriteBatch member variable with a pimpl in order not
to expose it directly in the header.
Also move CDBBatch::Clear to the dbwrapper implementation to use the new
impl_batch.
The context of this commit is an effort to decouple the dbwrapper header
file from leveldb includes. To this end, the includes are moved to the
dbwrapper implementation file. This is done as part of the kernel
project to reduce the number of required includes for users of the
kernel.
Keep the generic serialization in the header, while moving
leveldb-specifics to the implementation file.
The context of this commit is an effort to decouple the dbwrapper header
file from leveldb includes. To this end, the includes are moved to the
dbwrapper implementation file. This is done as part of the kernel
project to reduce the number of required includes for users of the
kernel.
Keep the generic serialization in the header, while moving
leveldb-specifics to the implementation file.
The context of this commit is an effort to decouple the dbwrapper header
file from leveldb includes. To this end, the includes are moved to the
dbwrapper implementation file. This is done as part of the kernel
project to reduce the number of required includes for users of the
kernel.
Wrap leveldb::DestroyDB in a helper function without exposing
leveldb-specifics.
Also, add missing optional include.
The context of this commit is an effort to decouple the dbwrapper header
file from leveldb includes. To this end, the includes are moved to the
dbwrapper implementation file. This is done as part of the kernel
project to reduce the number of required includes for users of the
kernel.
06199a995f refactor: Revert addition of univalue sighash string check (TheCharlatan)
0b47c16215 doc: Correct release-notes for sighashtype exceptions (TheCharlatan)
Pull request description:
This is a follow up for #28113.
The string type check is already done by the rpc parser / RPCHelpMan. Re-doing it is adding dead code. Instead, throwing an exception when the assumption does not hold is the already correct behavior. Pointed out in this [comment](https://github.com/bitcoin/bitcoin/pull/28113/files#r1274568557).
Also correct the release note for the correct sighashtype exception change. There is no change in the handling of non-string sighashtype arugments. Pointed out in this [comment](https://github.com/bitcoin/bitcoin/pull/28113/files#r1274567555).
ACKs for top commit:
MarcoFalke:
lgtm ACK 06199a995f
jonatack:
Tested ACK 06199a995f
stickies-v:
ACK 06199a995f
Tree-SHA512: 3faa6b3d2247624c0973df8d79c09fbf1f90ffb99f1be484e359b528f485c31affea45976759bd206e4c81cbb54ebba5ad0ef4127d1deacbfe2a58153fcc94ee
fa940f41ea Remove unused raw-pointer read helper from univalue (MarcoFalke)
Pull request description:
The helpers are unused outside of tests and redundant with the existing `bool read(std::string_view raw);`.
Fix both issues by removing them.
Also, simplify the tests code by removing a `std::string` constructor where possible.
ACKs for top commit:
stickies-v:
utACK fa940f41ea
TheCharlatan:
tACK fa940f41ea
Tree-SHA512: 60c154c1046f01551335af79bf820a6104844f63e89977271b4336b3cd06ac3bab1379e18b7bc61d12bef7446029e91c16541ddecf9e88bc8bc897fc1f6ee2c8
131314b62e fuzz: increase coverage of the descriptor targets (Antoine Poinsot)
90a24741e7 fuzz: add a new, more efficient, descriptor parsing target (Antoine Poinsot)
d60229ede5 fuzz: make the parsed descriptor testing into a function (Antoine Poinsot)
Pull request description:
The current descriptor parsing fuzz target requires valid public or private keys to be provided. This is unnecessary as we are only interested in fuzzing the descriptor parsing logic here (other targets are focused on fuzzing keys serializations). And it's pretty inefficient, especially for formats that need a checksum (`xpub`, `xprv`, WIF).
This introduces a new target that mocks the keys as an index in a list of precomputed keys. Keys are represented as 2 hex characters in the descriptor. The key type (private, public, extended, ..) is deterministically based on this one-byte value. Keys are deterministically generated at target initialization. This is much more efficient and also largely reduces the size of the seeds.
TL;DR: for instance instead of requiring the fuzzer to generate a `pk(xpub6DdBu7pBoyf7RjnUVhg8y6LFCfca2QAGJ39FcsgXM52Pg7eejUHLBJn4gNMey5dacyt4AjvKzdTQiuLfRdK8rSzyqZPJmNAcYZ9kVVEz4kj)` to parse a valid descriptor, it just needs to generate a `pk(03)`.
Note we only mock the keys themselves, not the entire descriptor key expression. As we want to fuzz the real code that parses the rest of the key expression (origin, derivation paths, ..).
This is a target i used for reviewing #17190 and #27255, and figured it was worth PR'ing on its own since the added complexity for mocking the keys is minimal and it could help prevent introducing bugs to the descriptor parsing logic much more efficiently.
ACKs for top commit:
MarcoFalke:
re-ACK 131314b62e🐓
achow101:
ACK 131314b62e
Tree-SHA512: 485a8d6a0f31a3a132df94dc57f97bdd81583d63507510debaac6a41dbbb42fa83c704ff3f2bd0b78c8673c583157c9a3efd79410e5e79511859e1470e629118
108c6255bc test: remove unused `totalOut` code (brunoerg)
0fc3deee9a test: remove unecessary `decoderawtransaction` calls (brunoerg)
Pull request description:
This PR removes in `wallet_fundrawtransaction`:
- unecessary variables/calls to `decoderawtransaction`
- unused `totalOut` variable and its related code (`totalOut` is used in some functions to test change, in other ones its value is not used)
ACKs for top commit:
kevkevinpal:
utACK [108c625](108c6255bc)
MarcoFalke:
lgtm ACK 108c6255bc
Tree-SHA512: c352524f3633146117534c79bd1a24523a7068f13a17d0b8a425cc3c85d62cb769a79ea60db8b075b137da2a0cc43142c43a23ca5af89246ff86cd824e37cf17
08eb5f1b67 ci: document that -Wreturn-type has been fixed upstream (Windows) (fanquake)
Pull request description:
`noreturn` attributes have been added to the mingw-w64 headers, 1690994f51, meaning that [from 11.0.0 onwards](https://www.mingw-w64.org/changelog/), you'll no-longer see `-Wreturn-type` warnings when using `assert(false)`.
Add -Wno-return-type to the Windows CI, where is should have been all
along, and document why it's required. This can be dropped when we are
using the fixed version of the mingw-w64 headers there.
Drop the -Werror -Wno-return-type special case from our build system.
-Wreturn-type is on by default in Clang and GCC.
The new mingw-w64 header behaviour can be checked on Ubuntu mantic, [which ships with 11.0.0](https://packages.ubuntu.com/mantic/mingw-w64), using:
```cpp
#include <cassert>
int f(){ assert(false); }
int main() {
return 0;
}
```
On Mantic (with 11.0.0):
```bash
x86_64-w64-mingw32-g++ test.cpp -Wreturn-type
# nada
```
On Lunar ([with 10.0.0](https://packages.ubuntu.com/lunar/mingw-w64)):
```bash
x86_64-w64-mingw32-g++ test.cpp -Wreturn-type
test.cpp: In function 'int f()':
test.cpp:3:25: warning: no return statement in function returning non-void [-Wreturn-type]
3 | int f(){ assert(false); }
| ^
```
ACKs for top commit:
TheCharlatan:
ACK 08eb5f1b67
Tree-SHA512: 9cd4310a96abd87bf8ceb37949ad0259fe4adee3367c604f4c4ad521a0cf09bdcc5dd305db19a0f45ce74c85178b0d739e2fca5ad0fc841ac935523a23b28a7f
This check is already done by the rpc parser. Re-doing it is adding dead
code. Instead, throwing an exception when the assumption does not hold
is the already correct behavior.
To make the fuzz test more accurate and not swallow all runtime errors,
add a check that the passed in UniValue sighash argument is either a
string or null.
Co-authored-by: stickies-v <stickies-v@protonmail.com>
In `wallet_fundrawtransaction`, `totalOut` is used in
some functions to check if the change is correct. In
other ones, it has been created but never used.
fa9108f85a refactor: Use reinterpret_cast where appropriate (MarcoFalke)
3333f950d4 refactor: Avoid casting away constness (MarcoFalke)
fa6394dd10 refactor: Remove unused C-style casts (MarcoFalke)
Pull request description:
Using a C-style cast to convert pointer types to a byte-like pointer type has many issues:
* It may accidentally and silently throw away `const`.
* It forces reviewers to check that it doesn't accidentally throw away `const`.
For example, on current master a `const char*` is cast to `unsigned char*` (without `const`), see d23fda0584/src/span.h (L273) . This can lead to UB, and the only reason why it didn't lead to UB is because the return type added back the `const`. (Obviously this would break if the return type was deduced via `auto`)
Fix all issues by adding back the `const` and using `reinterpret_cast` where appropriate.
ACKs for top commit:
darosior:
re-utACK fa9108f85a
hebasto:
re-ACK fa9108f85a.
john-moffett:
ACK fa9108f85a
Tree-SHA512: 87f6e4b574f9bd96d4e0f2a0631fd0a9dc6096e5d4f1b95042fe9f197afc2fe9a24e333aeb34fed11feefcdb184a238fe1ea5aff10d580bb18d76bfe48b76a10
c648bdbda2 test: create wallet specific for test_locked_wallet case (furszy)
Pull request description:
Coming from https://github.com/bitcoin/bitcoin/pull/28089#discussion_r1265478128.
Several test cases are relying on the node1 default wallet, which thanks to 'test_locked_wallet' is encrypted.
And can be only accessed within a specific timeframe (100ms), a duration internally set by the same test.
This situation introduces a potential race condition, where other tests must complete their operations within
the specified 100ms window to pass (otherwise the wallet gets re-locked and they fail).
This can be seen running the test in valgrind (https://github.com/bitcoin/bitcoin/pull/28089), where other test cases fail due the wallet re-locking
itself after the 100ms.
ACKs for top commit:
MarcoFalke:
lgtm ACK c648bdbda2
ishaanam:
utACK c648bdbda2
Tree-SHA512: 01cde5a4a0cb3405adb9ea3c1f73841f3fa237d1162268ed06f0d49ca38541006b423a029e0b5e5955e1aa7e018c4600d894e555a68cf17ff60a4b8be58f4aa9
faca9a3d5a test: Avoid intermittent issues due to async events in validationinterface_tests (MarcoFalke)
Pull request description:
Currently the tests have many issues:
* They setup the genesis block, even though it is not needed
* They queue an async `UpdatedBlockTip` even, which causes intermittent issues: https://github.com/bitcoin/bitcoin/issues/28146#issuecomment-1650064645
Fix all issues by trimming down the setup to just `ChainTestingSetup`.
ACKs for top commit:
Crypt-iQ:
tACK faca9a3d5a
Tree-SHA512: 4449040330f89bbaf5ce5b2052417c160b451c373987fdf1069596c07834ed81f0aea1506d53c7d2cd21062b27332d30679285dae194b272fd0cb9ce5ded32cf
d0c6cc4abe suppressions: note that 'type:ClassName::MethodName' should be used (fanquake)
Pull request description:
Now that the symbolizer is back in play, suppressions can once-again be targeted to functions, rather than file-wide.
ACKs for top commit:
MarcoFalke:
lgtm ACK d0c6cc4abe
hebasto:
ACK d0c6cc4abe
Tree-SHA512: fb65398eae18a6ebc5f8414275c568cf2664ab5357c2b3160f3bf285b67bc3af788225c5dba3c824c0e098627789450bec775375f52529d71c6ef700a9632d65
53c990ad34 test: fix `feature_addrman.py` on big-endian systems (Sebastian Falbesoner)
Pull request description:
The test `feature_addrman.py` currently serializes the addrdb without specifying endianness for `int`s, so the machine's native byte order is used (see https://docs.python.org/3/library/struct.html#byte-order-size-and-alignment) and the generated `peers.dat` would be invalid on big-endian systems (our internal (de)serializers always use little-endian, see `ser_{read,write}data32`). Fix this by explicitly specifying little-endian serialization via the `<` character in `struct.pack(...)`.
This is not detected by CI as we unfortunately don't run functional tests on big-endian systems there (I think we definitely should!).
ACKs for top commit:
MarcoFalke:
lgtm ACK 53c990ad34🔚
Tree-SHA512: 513af6f1f785a713e7a8ef3a57fcd3fe2520a7d537f63a9c8e1f4bdea4c2f605fd4c35001623d6b13458883dbc256f24943684ab8f224055c22bf8d8eeee5fe2
07c59eda00 Don't derive secure_allocator from std::allocator (Casey Carter)
Pull request description:
Giving the C++ Standard Committee control of the public interface of your type means they will break it. C++23 adds a new `allocate_at_least` member to `std::allocator`. Very bad things happen when, say, `std::vector` uses `allocate_at_least` from `secure_allocator`'s base to allocate memory which it then tries to free with `secure_allocator::deallocate`.
(Discovered by microsoft/STL#3712, which will be reverted by microsoft/STL#3819 before it ships.)
ACKs for top commit:
jonatack:
re-ACK 07c59eda00 no change since my previous ACK apart from squashing the commits
achow101:
ACK 07c59eda00
john-moffett:
ACK 07c59eda00 Reviewed and tested. Performance appears unaffected in my environment.
Tree-SHA512: 23606c40414d325f5605a9244d4dd50907fdf5f2fbf70f336accb3a2cb98baa8acd2972f46eab1b7fdec1d28a843a96b06083cd2d09791cda7c90ee218e5bbd5
6960c81cbf kernel: Remove Univalue from kernel library (TheCharlatan)
10eb3a9faa kernel: Split ParseSighashString (TheCharlatan)
Pull request description:
Besides the build system changes, this is a mostly move-only change for moving the few UniValue-related functions out of kernel files.
UniValue is not required by any of the kernel components and a JSON library should not need to be part of a consensus library.
ACKs for top commit:
achow101:
ACK 6960c81cbf
theuni:
Re-ACK 6960c81cbf
stickies-v:
re-ACK 6960c81cbf
Tree-SHA512: d92e4cb4e12134c94b517751bd746d39f9b8da528ec3a1c94aaedcce93274a3bae9277832e8a7c0243c13df0397ca70ae7bbb24ede200018c569f8d81103c1da
faa8c1be26 fuzz: Re-enable symbolize=1 in ASAN_OPTIONS (MarcoFalke)
Pull request description:
Looks like this fixed itself somehow and is no longer reproducible?
ACKs for top commit:
fanquake:
ACK faa8c1be26
Tree-SHA512: 67d2d6349cc7485f32bebabc18869ab101ae66a778a40ff9ddb037980997e600d7c6d1e0a17a011fa2a4ba07c73594b087dd781248cb8351f2688bc4cf6e587d
Affects both secure_allocator and zero_after_free_allocator.
Giving the C++ Standard Committee control of the public interface of your type means they will break it. C++23 adds a new `allocate_at_least` member to `std::allocator`. Very bad things happen when, say, `std::vector` uses `allocate_at_least` from `secure_allocator`'s base to allocate memory which it then tries to free with `secure_allocator::deallocate`.
Drive-by: Aggressively remove facilities unnecessary since C++11 from both allocators to keep things simple.
