If the miner code is faulty and does not include any transactions in a block,
the code segfaults when it tries to access block transactions. Instead, add a
check that safely aborts the process.
1afc0e4aa1 doc: remove potentially confusing ChainstateManager comment (James O'Beirne)
769a1ef9fd test: Add tests with maleated snapshot data (Fabian Jahr)
4d8de04f32 tests: add snapshot activation test (James O'Beirne)
31d225274f tests: add deterministic chain generation unittest fixture (James O'Beirne)
6606a4f8c6 move-onlyish: break out CreateUTXOSnapshot from dumptxoutset (James O'Beirne)
ad949ba449 txdb: don't reset during in-memory cache resize (James O'Beirne)
f6e2da5fb7 simplify ChainstateManager::SnapshotBlockhash() return semantics (James O'Beirne)
7a6c46b37e chainparams: add allowed assumeutxo values (James O'Beirne)
Pull request description:
This is part of the [assumeutxo project](https://github.com/bitcoin/bitcoin/projects/11):
Parent PR: #15606
Issue: #15605
Specification: https://github.com/jamesob/assumeutxo-docs/tree/master/proposal
---
This change proposes logic for activating UTXO snapshots, which is unused at the moment aside from an included unittest. There are a few moveonyish/refactoring commits to allow for halfway decent unittests.
Basic structure is included for specifying and checking the assumeutxo hash values used to validate activated snapshots. Initially I had specified a few height/hash pairs for mainnet in this change, but because of the security-critical nature of those parameters, I figured it was better to leave their inclusion to a future PR that includes only that change - my intent being that reviewers will be more likely to verify those parameters firsthand in a dedicated PR.
Aside from that and the snapshot activation logic, there are a few related changes:
- ~~allow caching the `nChainTx` value in the CCoinsViewDB; this is set during snapshot activation. Because we don't necessarily have access to the full chain at the time of snapshot load, this value is communicated through the snapshot metadata and must be cached within the chainstate to survive restarts.~~
- break out `CreateUTXOSnapshot()` from dumptxoutset. This is essentially a move-only change to allow the reuse of snapshot creation logic from within unittests.
- ...and a few other misc. changes that are solely related to unittests.
The move-onlyish commit is most easily reviewed with `--color-moved=zebra`.
ACKs for top commit:
fjahr:
Code review ACK 1afc0e4aa1
laanwj:
Code review ACK 1afc0e4aa1
Tree-SHA512: a4e4f0698f00a53ec298b5e8b7ef1c9fdf0185f95139d1b1f63cfdf6cbbd6d17b8c6e51bbf1de2e5f1a946bf49f8466232698ef55acce5a012c80b067da366ea
ef21fb7313 zmq test: speedup test by whitelisting peers (immediate tx relay) (Sebastian Falbesoner)
5c6546362d zmq test: fix flakiness by using more robust sync method (Sebastian Falbesoner)
8666033630 zmq test: accept arbitrary sequence start number in ZMQSubscriber (Sebastian Falbesoner)
6014d6e1b5 zmq test: dedup message reception handling in ZMQSubscriber (Sebastian Falbesoner)
Pull request description:
Fixes#20934 by using the "sync up" method described in https://github.com/bitcoin/bitcoin/issues/20538#issuecomment-738791868.
After improving robustness with this approach (commits 1-3), it turned out that there were still some fails, but those were unrelated to zmq: Out of 500 runs, 3 times `sync_mempool()` or `sync_blocks()` timed out, which can happen because the trickle relay time has no upper bound -- hence in rare cases, it takes longer than 60s. This is fixed by enabling immediate tx relay on node1 (commit 4), which as a nice side-effect also gives us a rough 2x speedup for the test.
For further details, also see the explanations in the commit messages.
There is no guarantee that the test is still not flaky, but it would help if potential reviewers would run the following script locally and report how many runs failed (feel free to do less than 1000 runs, as this takes quite a long if ran with `--valgrind`):
```
#!/bin/sh
OUTPUT_FILE=./zmq_results
echo ===== repeated zmq test ===== > $OUTPUT_FILE
for i in `seq 1000`; do
echo ------------------------
echo ----- test run $i -----
echo ------------------------
echo --- $i --- >> $OUTPUT_FILE
./test/functional/interface_zmq.py --valgrind
if [ $? -ne 0 ]; then
echo "FAILED. /o\\" >> $OUTPUT_FILE
else
echo "PASSED. \\o/" >> $OUTPUT_FILE
fi
done
echo Failed test runs:
grep FAILED $OUTPUT_FILE | wc -l
```
ACKs for top commit:
jonatack:
Light ACK ef21fb7313 with the caveat that I was unable to make the test fail with valgrind both here and on master, so I can't vouch that it actually fixes the CI flakiness. The test does run ~2x faster with this.
Tree-SHA512: 7a1e7592fbbd98e69e1e1294486b91253e589c72b3c6bbb7f587028ec07cca59b7d984e4ebf256c4bc3e8a529ec77d31842f3dd874038aea0b684abfea50306a
a5e15ae45c scripted-diff: rename ping members (John Newbery)
45dcf22661 [net processing] Move ping data fields to net processing (John Newbery)
dd2646d12c [net processing] Move ping timeout logic to net processing (John Newbery)
0b43b81f69 [net processing] Move send ping message logic into function (John Newbery)
1a07600b4b [net] Add RunInactivityChecks() (John Newbery)
f8b3058992 [net processing] Add Peer& arg to MaybeDiscourageAndDisconnect() (John Newbery)
Pull request description:
This continues the work of moving application layer data into net_processing, by moving all ping data into the new Peer object added in #19607.
For motivation, see #19398.
ACKs for top commit:
glozow:
reACK a5e15ae45c
MarcoFalke:
review ACK a5e15ae45c🥉
amitiuttarwar:
ACK a5e15ae45c
Tree-SHA512: fb84241613d6a6e1f2832fa5378030b5877a02e8308188f57ab545a6eaf2ab731a93abb7dcd3a7f7285bb66700f938096378a8e90cd6a3e6f3309f81d85a344e
ffff84a9cb fuzz: Remove expensive and redundant muhash from crypto fuzz target (MarcoFalke)
Pull request description:
Remove because it is redundant with `src/test/fuzz/muhash.cpp` and incredibly expensive
ACKs for top commit:
practicalswift:
Tested ACK ffff84a9cb
Tree-SHA512: c91ea2406db857127c789b9cdeb714a719d88b54132e9cef74fffd229532d874b6c043353793ec687504b5784afc74995f8982243d41f976b63d57454a5ed339
Moves the logic to prevent running inactivity checks until
the peer has been connected for -peertimeout time into its
own function. This will be reused by net_processing later.
96635e6177 init: use GetNetworkNames() in -onlynet help (Jon Atack)
0dbde700a6 rpc: use GetNetworkNames() in getnetworkinfo and getpeerinfo helps (Jon Atack)
1c3af37881 net: create GetNetworkNames() (Jon Atack)
b45eae4d53 net: update NET_UNROUTABLE to not_publicly_routable in GetNetworkName() (Jon Atack)
Pull request description:
per the IRC discussion today at http://www.erisian.com.au/bitcoin-core-dev/log-2021-01-19.html#l-87
- return a more helpful string name for `Network::NET_UNROUTABLE`: "not_publicly_routable" instead of "unroutable"
- update the RPC getpeerinfo "network" help, and automate it and the getnetworkinfo "network#name" and the -onlynet help doc generation
ACKs for top commit:
theStack:
re-ACK 96635e6177🌳
MarcoFalke:
review ACK 96635e6177🐗
Tree-SHA512: 511a7d987126b48a7a090739aa7c4964b6186a3ff8f5f7eec9233dd816c6b7a6dc91b3ea6b824aa68f218a8a3ebdc6ffd214e9a88af38f2bf23f3257c4284c3a
2ee4a7a9ec net: remove CNode::m_inbound_onion defaults for explicitness (Jon Atack)
24bda56c29 net: make CNode::m_inbound_onion public, drop getter, update tests (Jon Atack)
Pull request description:
Refactoring only, no change in behavior. This is a quick follow-up to #20210 to address these review comments:
- https://github.com/bitcoin/bitcoin/pull/20210#discussion_r528835313
- https://github.com/bitcoin/bitcoin/pull/20210#discussion_r550860416
- https://github.com/bitcoin/bitcoin/pull/20210#issuecomment-766093925
Changes:
- make the `CNode::m_inbound_onion class` member public, update the Doxygen comment, drop the getter, and update the tests
- remove the `CNode::m_inbound_onion` default value initialization in the ctor declaration and the member initializer in favor of always passing it explicitly to the ctor where we initialize it dynamically, to both clarify the caller code and to allow the compiler to warn if it is uninitialized in the ctor or omitted in the caller
ACKs for top commit:
MarcoFalke:
review ACK 2ee4a7a9ec🏀
vasild:
ACK 2ee4a7a9ec
Tree-SHA512: 72961c91168885a9d881756b10bad9d587f5ce297d5a6493c23e267b7178ff22b697bc6868e7761d6304e372d2781453d30011a020afd506b1e308b4ffa5feee
3f8776a139 Re-add dead code detection (flack)
Pull request description:
This re-adds unreachable code detection for Python based on `vulture`.
Effectively, this reverts f4beb4996d. The difference to the previous version is that this runs with the `--min-confidence 100` setting. From https://pypi.org/project/vulture/:
> Use `--min-confidence 100` to only report code that is guaranteed to be unused within the analyzed files.
So this should avoid the previous issues where static analysis had wrong positives due to the dynamic nature of Python code by only reporting things that are unambiguous (such as code after a `return` statement). As such, there is not suppressions list.
My motivation was mainly #21081 which would have been caught by this (as can be seen by the CI run failing). This is still marked as draft because #21081 is needed to get the linter to pass. Also, there is a second problem that this found (see https://github.com/bitcoin/bitcoin/pull/19509/files#r571454691). From what I can tell, this is a spurious type comment that could just be removed (or if that line has no side effects it could also be deleted altogether?). I could add a commit here to fix it, but I wanted to see if there is interest in having this linter again in the first place
ACKs for top commit:
practicalswift:
ACK 3f8776a139
Tree-SHA512: 52314ad4f627d969de1eb15375ca677ed86a2e816fe773756a1ce22421214ba407b5a09a4bf701a3aab1a10c7b336f548e4cef3327edf154acba55e987db21f6
Can be reviewed with --word-diff-regex=.
-BEGIN VERIFY SCRIPT-
sed -i --regexp-extended 's/(PeerManagerImpl::.*\)).*LOCKS_.*\)/\1/g' ./src/net_processing.cpp
-END VERIFY SCRIPT-
f64adc1eed test: remove unused function xor_bytes (Sebastian Falbesoner)
Pull request description:
The function `xor_bytes` was introduced in commit 3c226639eb (#19953, BIP340-342 validation), even [code-reviewed](https://github.com/bitcoin/bitcoin/pull/19953/files#r509383731), but actually never used. The [default signing algorithm in BIP340](https://github.com/bitcoin/bips/blob/master/bip-0340.mediawiki#Default_Signing) needs a xor operation, but this step is currently done by a single xor operation on large integer operands:
```
t = (sec ^ int.from_bytes(TaggedHash("BIP0340/aux", aux), 'big')).to_bytes(32, 'big')
```
Alternatively, we could keep the function and as well use it:
```diff
--- a/test/functional/test_framework/key.py
+++ b/test/functional/test_framework/key.py
@@ -492,7 +492,7 @@ def sign_schnorr(key, msg, aux=None, flip_p=False, flip_r=False):
P = SECP256K1.affine(SECP256K1.mul([(SECP256K1_G, sec)]))
if SECP256K1.has_even_y(P) == flip_p:
sec = SECP256K1_ORDER - sec
- t = (sec ^ int.from_bytes(TaggedHash("BIP0340/aux", aux), 'big')).to_bytes(32, 'big')
+ t = xor_bytes(sec.to_bytes(32, 'big'), TaggedHash("BIP0340/aux", aux))
kp = int.from_bytes(TaggedHash("BIP0340/nonce", t + P[0].to_bytes(32, 'big') + msg), 'big') % SECP256K1_ORDER
assert kp != 0
R = SECP256K1.affine(SECP256K1.mul([(SECP256K1_G, kp)]))
```
ACKs for top commit:
practicalswift:
cr ACK f64adc1eed: untested unused code should be removed
Tree-SHA512: e9afae303488f19c6f6f44874d3537ed1c8164a197490e2b4e34853db886b858825b719648fa1a30b95177dcee9cf241f94ee9b835f0a2cae07024ce38a8c090
6452190841 net_processing: simplify MaybeSetPeerAsAnnouncingHeaderAndIDs args (Anthony Towns)
39c2a69bc2 net_processing: move MaybeSetPeerAsAnnouncingHeadersAndIDs into PeerManagerImpl (Anthony Towns)
7b7117efd0 net_processing: simplify ProcessGetData and FindTxForGetData args (Anthony Towns)
34207b9004 net_processing: move FindTxForGetData and ProcessGetData to PeerManagerImpl (Anthony Towns)
d44084883a net_processing: simplify PeerManageImpl method args (Anthony Towns)
a490f0a056 net_processing: move MarkBlockAs*, TipMayBeStale, FindNextBlocksToDL to PeerManagerImpl (Anthony Towns)
052d9bc7e5 net_processing: simplify AlreadyHaveTx args (Anthony Towns)
eeac506250 net_processing: move AlreadyHaveTx into PeerManageImpl (Anthony Towns)
9781c08a33 net_processing: move some globals into PeerManagerImpl (Anthony Towns)
Pull request description:
Turns some globals into member variables, and simplifies the parameter list for some of net_processing's internal functions. Mostly just serves as a code cleanup at this point.
ACKs for top commit:
jnewbery:
Code review ACK 6452190841
ariard:
Code Review ACK 6452190, changes are pretty straightforward.
MarcoFalke:
Concept ACK 6452190841 I have not reviewed this, but I left a comment 🐡
Tree-SHA512: 381361f9dbfeb851a5522ead3165ce1447a0f212ddea4b483aa38975559ee5ed03a4ba69c24fd69f36847a1eddfef05785f5cbb2fcec5fe50f8b336e8047c3b1
5200929bfe depends: Include GUIX_ENVIRONMENT in id string (Carl Dong)
4c7d418588 depends: Improve id string robustness (Carl Dong)
b3bdff42b5 build: Proper quoting for var printing targets (Carl Dong)
Pull request description:
```
Environment variables and search paths can drastically effect the
operation of build tools.
Include these in our id string to mitigate against false cache hits.
```
Note to builders: This will invalidate all depends output caches in `BASE_CACHE`
ACKs for top commit:
laanwj:
re-ACK 5200929bfe
Tree-SHA512: e70c98da89cde90dc54bc3be89b925787cf94bbf246e27cc9345816b312073d78a02215448f731f21d8cf033c455234a2377ff1d66c00e1f3db69c9c9687d027
aa929abf8d [docs] Update developer notes to discourage very long lines (John Newbery)
Pull request description:
Mandatory rules on line lengths are bad - there will always be cases where a longer line is more readable than the alternative.
However, very long lines for no good reason _do_ hurt readability. For example, this declaration in validation.h is 274 chars:
```c++
bool ConnectTip(BlockValidationState& state, const CChainParams& chainparams, CBlockIndex* pindexNew, const std::shared_ptr<const CBlock>& pblock, ConnectTrace& connectTrace, DisconnectedBlockTransactions& disconnectpool) EXCLUSIVE_LOCKS_REQUIRED(cs_main, m_mempool.cs);
```
That won't fit on one line without wrapping on my 27" monitor with a comfortable font size. Much easier to read is something like:
```c++
bool ConnectTip(BlockValidationState& state, const CChainParams& chainparams,
CBlockIndex* pindexNew, const std::shared_ptr<const CBlock>& pblock,
ConnectTrace& connectTrace, DisconnectedBlockTransactions& disconnectpool)
EXCLUSIVE_LOCKS_REQUIRED(cs_main, m_mempool.cs);
```
Therefore, _discourage_ (don't forbid) line lengths greater than 100 characters in our developer style guide.
100 chars is somewhat arbitrary. The old standard was 80, but that seems very limiting with modern displays.
ACKs for top commit:
fanquake:
ACK aa929abf8d - this is basically just something to point too when a PR has unreasonably long lines for no particularly reason.
practicalswift:
ACK aa929abf8d
amitiuttarwar:
ACK aa929abf8d
theStack:
ACK aa929abf8d
glozow:
ACK aa929abf8d
Tree-SHA512: 17f1b11f811137497ede8851ede93fa612dc622922b5ad7ac8f065ea026d9a718db5b92325754b74d24012b4d45c4e2cd5cd439a6a8d34bbabf5da927d783970
fa051c2386 doc: Guix is shipped in Debian and Ubuntu (MarcoFalke)
Pull request description:
ACKs for top commit:
practicalswift:
ACK fa051c2386🚀
Tree-SHA512: f72f546cfc20cf1cc0c26c2306ac06416ada87661596fe811b497cce646aa286dc4aee832145bf838b13fbd3c5f064519eb8c0b4525eb562f2f04f20e2876ffc
9305862f71 wallet: load flags before everything else (Sjors Provoost)
Pull request description:
Load and set wallet flags before processing other records. That way we can take them into account while processing those other records.
Suggested here:
https://github.com/bitcoin/bitcoin/pull/16546#discussion_r572334983
ACKs for top commit:
laanwj:
Code review ACK 9305862f71
gruve-p:
ACK 9305862f71
achow101:
ACK 9305862f71
Tree-SHA512: 7104523e369ce3c670571fe5e8b52c67b9ca92b8e36a2da5eb6f9f8bf8ed0544897007257204b68f6f371d682b3ef0d0635d36e6e8416ac74af1999d9fbc869c
c943326d3c doc/bips: Add BIPs 43, 44, 49, and 84 (Luke Dashjr)
Pull request description:
If you don't like what they say, please suggest alternatives ;)
ACKs for top commit:
prusnak:
ACK c943326
Tree-SHA512: 7db93f8491289657ec45df30e557eb8572b35201eb29aed1b11bf3949924fce56b4e2d71e1f0acf5d24a01278c0dec99790d632f04c15117010c4ac564368d6b
The benchmark BlockToJsonVerbose only tests generating (and destroying)
the JSON data structure, but serializing into a string is also a
performance critical aspect of the RPC calls.
Also, use ankerl::nanobench::doNotOptimizeAway to make sure the compiler
can't optimize the result of the calls away.
d4187e4619 [test] Use mocktime in test_seed_peers() (Dhruv Mehta)
015637dd44 [refactor] Correct log message in net.cpp (Dhruv Mehta)
Pull request description:
The test now takes less than 5 seconds instead of more than 2 minutes
Further context: https://github.com/bitcoin/bitcoin/pull/19884/files#r575336503
Before:
```
2021-02-12T17:22:25.980000Z TestFramework (INFO): Test seed peers, this will take about 2 minutes
2021-02-12T17:24:30.472000Z TestFramework (INFO): Test -networkactive option
```
After:
```
2021-02-12T17:33:39.224000Z TestFramework (INFO): Test seed peers
2021-02-12T17:33:43.139000Z TestFramework (INFO): Test -networkactive option
```
Top commit has no ACKs.
Tree-SHA512: 6d8df7d4433c96268694577e4c10a346785e076d45fa220091875e55def200100e7b827fac2a1f7853a2c2c39e9661e06288dca8c645da9e13d4318a4ff2172e
