fa7238300c fuzz: Limit fuzzed time to years 2000-2100 (MarcoFalke)
Pull request description:
It doesn't make sense to fuzz times in the past, as Bitcoin Core will refuse to start in the past.
Fix that and also remove a sanitizer suppression, which would be hit in net_processing in `ProcessMessage`:
```cpp
if (addr.nTime <= 100000000 || addr.nTime > nNow + 10 * 60)
addr.nTime = nNow - 5 * 24 * 60 * 60; // <-- Here
```
This changes the format of fuzz inputs. Previously a time value was (de)serialized as 40 bytes, now it is 32 bytes.
ACKs for top commit:
mzumsande:
Code Review ACK fa7238300c
Tree-SHA512: ca6e7233beec2d9ef9fd481d8f1331942a4d2c8fe518b857629bebcc53a4f42ae123b994cf5d359384a0a8022098ff5a9c146600bc2593c6d88734e25bc240ad
In each of the critical sections, only the the guarded variable is
accessed, without any chance that within one section another one is
called. Hence, we can use an ordinary Mutex instead of RecursiveMutex.
-BEGIN VERIFY SCRIPT-
sed -i 's/std::string m_command;/std::string m_type;/g' ./src/net.h
sed -i 's/* command and size./* type and size./g' ./src/net.h
sed -i 's/msg.m_command/msg.m_type/g' ./src/net.cpp ./src/net_processing.cpp ./src/test/fuzz/p2p_transport_serialization.cpp
-END VERIFY SCRIPT-
On a period boundary, getdeploymentinfo (and previously getblockchaininfo)
would report the status and statistics for the next block rather than
the current block. Change this to always report the status/statistics
of the current block, but add status-next to report the status for the
next block.
In each of the critical sections, only the the guarded variables are
accessed, without any chance that within one section another one is
called. Hence, we can use an ordinary Mutex instead of RecursiveMutex.
aa8a65e4a8 test: use MiniWallet for mempool_accept.py (Sebastian Falbesoner)
b24f6c6855 test: MiniWallet: support default `from_node` for creating txs (Sebastian Falbesoner)
f30041c914 test: create txs with current `nVersion` (2) by default (Sebastian Falbesoner)
2f79786822 test: refactor: add constant for sequence number `SEQUENCE_FINAL` (Sebastian Falbesoner)
Pull request description:
This PR enables one more of the non-wallet functional tests (mempool_accept.py) to be run even with the Bitcoin Core wallet disabled by using the MiniWallet instead, as proposed in #20078.
It also includes some other minor changes that came up while working on the replacement:
* [commit 1/4] replace magic number 0xffffffff for a tx's nSequence with a new constant `SEQUENCE_FINAL`
* [commit 2/4] create `CTransaction` instances with the current nVersion=2 by default, in order to use BIP68 for tests
* [commit 3/4] support default `from_node` parameter for creating txs (this is a stripped down version of PR #24025)
ACKs for top commit:
MarcoFalke:
re-ACK aa8a65e4a8📊
Tree-SHA512: 34cd085ea4147ad5bd3f3321c84204064ceb95f382664c7fe29062c1bbc79d9d9465c5e46d35e11c416f2f3cd46030c90a09b518c829c73ae40d060be5e4c9cb
c62d763fc3 Necessary improvements to make configure work without libevent installed (Perlover)
091ccc38c2 The evhttp_connection_get_peer function from libevent changes the type of the second parameter. Fixing the problem. (Perlover)
Pull request description:
The second parameter of evhttp_connection_get_peer in libevent already has type as `const char **`
The compilation of bitcoind with the fresh libevent occurs errors
Details: https://github.com/bitcoin/bitcoin/issues/23606
ACKs for top commit:
laanwj:
Code review ACK c62d763fc3
luke-jr:
tACK c62d763fc3
Tree-SHA512: d1c8062d90bd0d55c582dae2c3a7e5ee1b6c7ca872bf4aa7fe6f45a52ac4a8f59464215759d961f8efde0efbeeade31b08daf9387d7d50d7622baa1c06992d83
This distribution is used for more than just the next inv send, so make
the name more generic.
Also rename to "exponential" to avoid the confusion that this is a
poisson distribution.
-BEGIN VERIFY SCRIPT-
ren() { sed -i "s/\<$1\>/$2/g" $(git grep -l "$1" ./src) ; }
ren PoissonNextSend GetExponentialRand
ren "a poisson timer" "an exponential timer"
-END VERIFY SCRIPT-
b5c9bb5cb9 util: Restore GetIntArg saturating behavior (James O'Beirne)
Pull request description:
The new locale-independent atoi64 method introduced in #20452 parses large integer values higher than maximum representable value as 0 instead of the maximum value, which breaks backwards compatibility. This commit restores compatibility and adds test coverage for this case in terms of the related GetIntArg and strtoll functions.
Specifically, command line or bitcoin.conf integer values greater than `9223372036854775807` (`2**63-1`) used to be parsed as `9223372036854775807` before #20452. Then #20452 caused them to be parsed as `0`. And after this PR they will be parsed as `9223372036854775807` again.
This change is a stripped-down alternative version of #23841 by jamesob
ACKs for top commit:
jamesob:
Github ACK b5c9bb5cb9
vincenzopalazzo:
ACK b5c9bb5cb9
MarcoFalke:
review ACK b5c9bb5cb9🌘
Tree-SHA512: 4e8abdbabf3cf4713cf5a7c5169539159f359ab4109a4e7e644cc2e9b2b0c3c532fad9f6b772daf015e1c5340ce59280cd9a41f2730afda6099cbf636b7d23ae
Shave off an extra 100 or so instructions from the
validation:block_connected tracepoint by reusing a nearby
GetTimeMicros(). This brings the tracepoint down to 54 instructions.
Still high, but much better than the previous ~154 and 8000 instructions
which it was originally.
Signed-off-by: William Casarin <jb55@jb55.com>
In the validation:block_connected tracepoint, we call block->GetHash(),
which ends up calling CBlockHeader::GetHash(), executing around 8000
serialization instructions. We don't need to do this extra work, because
block->GetHash() is already called further up in the function. Let's
save that value as a local variable and re-use it in our tracepoint so
there is no unnecessary tracepoint overhead.
Signed-off-by: William Casarin <jb55@jb55.com>
3c4f5d2a20 scripted-diff: Rename functions to drop mention of boost (Hennadii Stepanov)
Pull request description:
Nothing specific to Boost is in `GUIUtil::qstringToBoostPath` and `GUIUtil::boostPathToQString` functions.
Related to bitcoin/bitcoin#20744.
ACKs for top commit:
laanwj:
Concept and code review ACK 3c4f5d2a20
promag:
Code review ACK 3c4f5d2a20👋
Tree-SHA512: 0c8790979783a067811f7699b4ce4c204f6e0818d5f32986ec24b2c71583b4496d7a0e0c0361dd77c7e641a75d983fee35cd51ef722bbca9a5f13194efb3b4c0
a229451590 build: Point Guix to the current top of the "version-1.4.0" branch (Hennadii Stepanov)
Pull request description:
On master (c561f2f06e) the commit in Guix repo from bitcoin/bitcoin#23778 seems unavailable:
```
$ git checkout fa17abf1af09570708daa28dd40ffbc932ebe25c
fatal: reference is not a tree: fa17abf1af09570708daa28dd40ffbc932ebe25c
```
This PR points Guix to the current top of the "version-1.4.0" branch.
Fixes#24040.
ACKs for top commit:
MarcoFalke:
Approach ACK a229451590
fanquake:
ACK a229451590 - from what I've seen on the mailing list there shouldn't be any more force pushing.
Tree-SHA512: c58f846fb0afd51b5c2dd33034e9d593aec5d5b49e9f2232af70ae1224da8408ad4e05aa314609350d92a6400e354a816b988226e3572198c3f839ab33913164
ce95fb36af Remove cs_main lock annotation from ChainstateManager.m_blockman (Ryan Ofsky)
Pull request description:
`BlockManager` is a large data structure, and `cs_main` is not required to take its address or access every part of it. Individual `BlockManager` fields and methods which do require `cs_main` like `m_block_index` and `LookupBlockIndex` are already annotated separately, and these other annotations describe locking requirements more accurately and do a better job enforcing thread safety.
Since `cs_main` is not needed to access the address of the m_block object, this commit drops `cs_main` LOCK calls which were added pointlessly to satisfy this annotation in the past.
Code changes were made by dongcarl, I just wrote the commit description
ACKs for top commit:
MarcoFalke:
cr ACK ce95fb36af
dongcarl:
crACK ce95fb36af
jonatack:
ACK ce95fb36af per `git range-diff db1f04f 5a1c413 ce95fb3` change since last push is rebase and dropping braces on the member initialization
Tree-SHA512: b18a6ebcc70bea750485f04d4feb7bb28450fea2176e513be9cc242e9f63b24254c5659e74eb6d6045c706a3aaeb94688937b25b7ca7653f8aa3cf8c18845d5a
9d3e95d77c [bugfix] prevent UnicodeDecodeError errors when opening log file in feature_init.py (sogoagain)
Pull request description:
Should fix#23989
To fix a bug, I modified `feature_init.py` to open the log file as a byte stream when opening it.
thank you.
ACKs for top commit:
MarcoFalke:
review ACK 9d3e95d77c
Tree-SHA512: 6e3e57cac5f4865b3894ee4e9fcd9eb2690e824af20e34b4595722bd7043b0c3fe417cc1bfcff25fbab95c66418d3fce13434bd63d0244875a867d08853a5644
The new locale-independent atoi64 method introduced in #20452 parses
large integer values higher than maximum representable value as 0
instead of the maximum value, which breaks backwards compatibility.
This commit restores compatibility and adds test coverage for this case
in terms of the related GetIntArg and strtoll functions.
Co-authored-by: Ryan Ofsky <ryan@ofsky.org>
If no `from_node` parameter is passed explicitely to the
`create_self_transfer` method, the test node passed in the course
of creating the MiniWallet instance is used. This seems to
be the main use-case in most of the current functional
tests, i.e. in many instances the calls can be shortened.
Make it possible to override from the command line (without recompiling)
the addrman check ratio in non-addrman fuzz tests (connman and
deserialize) instead of hardcoding it to 0:
```
FUZZ=connman ./src/test/fuzz/fuzz --checkaddrman=5
```
Make it possible to override from the command line (without recompiling)
the addrman check ratio in addrman fuzz tests instead of hardcoding it
to 0:
```
FUZZ=addrman ./src/test/fuzz/fuzz --checkaddrman=5
```
