Previously, if the builder exported $VERSION in their environment (as
past Gitian-building docs told them to), but their HEAD does not
actually point to v$VERSION, their build outputs will differ from those
of other builders.
This is because the contrib/guix/guix-* scripts only ever act on the
current git worktree, and does not try to check out $VERSION if $VERSION
is set in the environment.
Setting $VERSION only makes the scripts pretend like the current
worktree is $VERSION.
This problem was seen in jonatack's attestation for all.SHA256SUMS,
where only his bitcoin-22.0rc3-osx-signed.dmg differed from everyone
else's.
Here is my deduced sequence of events:
1. Aug 27th: He guix-builds 22.0rc3 and uploads his attestations up to
guix.sigs
2. Aug 30th, sometime after POSIX time 1630310848: he pulls the latest
changes from master in the same worktree where he guix-built 22.0rc3
and ends up at 7be143a960
3. Aug 30th, sometime before POSIX time 1630315907: With his worktree
still on 7be143a960, he guix-codesigns. Normally, this would result
in outputs going in guix-build-7be143a960e2, but he had
VERSION=22.0rc3 in his environment, so the guix-* scripts pretended
like he was building 22.0rc3, and used 22.0rc3's guix-build directory
to locate un-codesigned outputs and dump codesigned ones.
However, our SOURCE_DATE_EPOCH defaults to the POSIX time of HEAD
(7be143a960), which made all timestamps in the resulting codesigned
DMG 1630310848, 7be143a960e2's POSIX timestamp. This differs from the
POSIX timestamp of 22.0rc3, which is 1630348517. Note that the
windows codesigning procedure does not consider SOURCE_DATE_EPOCH.
We resolve this by only allowing VERSION overrides via the FORCE_VERSION
environment variable.
f293c68be0 MOVEONLY: getting mempool conflicts to policy/rbf (glozow)
8d71796335 [validation] quit RBF logic earlier and separate loops (glozow)
badb9b11a6 call SignalsOptInRBF instead of checking all inputs (glozow)
e0df41d7d5 [validation] default conflicting fees and size to 0 (glozow)
b001b9f6de MOVEONLY: BIP125 max conflicts limit to policy/rbf.h (glozow)
Pull request description:
See #22675 for motivation, this is one chunk of it. It extracts some BIP125 logic into policy/rbf:
- Defines a constant for specifying the maximum number of mempool entries we'd consider replacing by RBF
- Calls the available `SignalsOptInRBF` function instead of manually iterating through inputs
- Moves the logic for getting the list of conflicting mempool entries to a helper function
- Also does a bit of preparation for future moves - moving declarations around, etc
Also see #22677 for addressing the circular dependency.
ACKs for top commit:
jnewbery:
Code review ACK f293c68be0
theStack:
Code-review ACK f293c68be0📔
ariard:
ACK f293c68b
Tree-SHA512: a60370994569cfc91d4b2ad5e94542d4855a48927ae8b174880216074e4fa50d4523dd4ee36efdd6edf2bf7adb87a8beff9c3aaaf6dd323b286b287233e63790
fa001602cd ci: Re-enable verify-commits.py check (MarcoFalke)
fa880b10d6 ci: Unconditionally set the global git author name in cirrys.yml (MarcoFalke)
Pull request description:
Might be useful to detect bugs in the script itself or an accidentally missed signature.
ACKs for top commit:
josibake:
ACK fa001602cd
Zero-1729:
tACK fa001602cd
fanquake:
untested ACK fa001602cd
Tree-SHA512: 8a13a67d325f2477f4088d1034f0d5e4e04937a01ee3c738435fe66394c02b9f33225529952ad331b0ba19b63ca4b2f26911cb5d264890159840cf3e09085969
We need to append-to rather than set CXXFLAGS, otherwise we loose -O2 &
-pipe. Currently this results in zeromq being built without optimizations
at all (or whatever the compiler would default too, essentially always -O0).
Bdb is the same, for the CXX portion of its code. C code has been built
with -O2.
Boost has actually been uneffected because it receives -O3 from it's own
build flags.
0bd882b740 refactor: remove RecursiveMutex cs_nBlockSequenceId (Sebastian Falbesoner)
Pull request description:
The RecursiveMutex `cs_nBlockSequenceId` is only used at one place in `CChainState::ReceivedBlockTransactions()` to atomically read-and-increment the nBlockSequenceId member:
83daf47898/src/validation.cpp (L2973-L2976)
~~For this simple use-case, we can make the member `std::atomic` instead to achieve the same result (see https://en.cppreference.com/w/cpp/atomic/atomic/operator_arith).~~
~~This is related to #19303. As suggested in the issue, I first planned to change the `RecursiveMutex` to `Mutex` (still possible if the change doesn't get Concept ACKs), but using a Mutex for this simple operation seems to be overkill. Note that at the time when this mutex was introduced (PR #3370, commit 75f51f2a63) `std::atomic` were not used in the codebase yet -- according to `git log -S std::atomic` they have first appeared in 2016 (commit 7e908c7b82), probably also because the compilers didn't support them properly earlier.~~
At this point, the cs_main lock is set, hence we can use a plain int for the member and mark it as guarded by cs_main.
ACKs for top commit:
Zero-1729:
ACK 0bd882b
promag:
Code review ACK 0bd882b740.
hebasto:
ACK 0bd882b740
Tree-SHA512: 435271ac8f877074099ddb31436665b500e555f7cab899e5c8414af299b154d1249996be500e8fdeff64e4639bcaf7386e12510b738ec6f20e415e7e35afaea9
The RecursiveMutex cs_nBlockSequenceId is only used at one place in
CChainState::ReceivedBlockTransactions() to atomically read-and-increment the
nBlockSequenceId member. At this point, the cs_main lock is set, hence we can
use a plain int for the member and mark it as guarded by cs_main.
e251726aff build, qt: Fix typo in QtInputSupport check (Hennadii Stepanov)
Pull request description:
This typo slipped in fecb3723b6 (#21565).
I apologize for that as a reviewer of #21565.
ACKs for top commit:
fanquake:
ACK e251726aff
Tree-SHA512: c14d603ed5a9b7e4b6503bfb93e66b800d5a107f21a8375c6f98465ef937e3e55b9c4fc160407c7e3ee402b495c5820ebb35ec6cb92956c1ee4092415635b915
In order to change the KnapsackSolver tests to call KnapsackSolver, we
need KnapsackGroupOutputs to create the OutputGroups filtered with the
filter criteria.
The long term feerate is really the highest feerate that the user is
comfortable with making consolidatory transactions. This is should thus
be something that can be configured by the user via a new startup option
-consolidatefeerate. The default value is 10 sat/vbyte, chosen
arbitrarily (it seems like a reasonable number).
The m_ prefix indicates that a variable is a data member. Using it as
a parameter name is misleading.
Also update the name of the function from copyStats to CopyStats to
comply with our style guide.
b367745cfe ci: Make Cirrus CI Windows build with --enable-werror (Hennadii Stepanov)
c713bb2b24 Fix Windows build with --enable-werror on Ubuntu Focal (Hennadii Stepanov)
Pull request description:
This PR makes possible to cross-compile Windows build with `--enable-werror --enable-suppress-external-warnings`.
Some problems are fixed, others are silenced.
Also `--enable-werror` is enabled for Cirrus CI Windows build (the last one on Cirrus CI without `--enable-werror`).
ACKs for top commit:
practicalswift:
cr ACK b367745cfe: patch looks correct
laanwj:
Code review ACK b367745cfe
vasild:
ACK b367745cfe
jarolrod:
ACK b367745cfe
Tree-SHA512: 64f5c99b7dad4c0efce80cd45d7074f275bd8411235dc9e0841287bdab64b812c6f8f9d632c35531d0b8210148531f53aaaac77be7699b29d2d6aaae304dbee0
faf7e485e9 Set regtest.BIP65Height = 111 to speed up tests (MarcoFalke)
Pull request description:
No need to waste time by forcing creation of more than 1000 blocks to get the benefits of being able to test BIP 65. Also, reducing the height makes it more likely that (third-party) tests are conforming to BIP 65, which is enforced on mainnet for all new blocks.
ACKs for top commit:
theStack:
re-ACK faf7e485e9📍
Zero-1729:
re-ACK faf7e485e9
kristapsk:
ACK faf7e485e9
Tree-SHA512: 79a8263e7233838666b9b636b496a8b9eb12398c779f9434677e1d62816732c0a7c7b3e73965be1fb0038d35e05e5a90e665bd74e9610104127dfc4ea38169bf
ab1461d5d3 qt: Add copy IP/Netmask action for banned peer (Shashwat)
Pull request description:
This PR adds a Copy IP/Netmask context menu action to the Banned Peers Table.
This feature is helpful if a node using GUI might want to alert its peer about a particular malicious user. So it can copy that user’s IP/Netmask and broadcast it to its peers so they can ban it instantly using the setban command in the console.
| Master | PR |
| ----------- | ----------- |
|  |  |
ACKs for top commit:
jarolrod:
re-ACK ab1461d
hebasto:
re-ACK ab1461d5d3, tested on Linux Mint 20.2 (Qt 5.12.8).
Tree-SHA512: a528f089bd4cb5b51fec987550d21c2587459ad80f854b55850bc62c776c21f3fa31052a17e2b0e9e9d0b3468799c8070ed306543730fb7b324f283847151e17
b8aa84b1a1 qt, refactor: Replace `if` check with `assert` (Hennadii Stepanov)
fcdc8b0fcb qt, refactor: Drop redundant signalling in WalletView::setWalletModel (Hennadii Stepanov)
37dcf161d3 qt, refactor: Emit WalletView::encryptionStatusChanged signal directly (Hennadii Stepanov)
7d0d4c0490 qt: Add WalletFrame::currentWalletSet signal (Hennadii Stepanov)
Pull request description:
This PR makes signal-slot paths to reach `setHDStatus` and `setEncryptionStatus` functions shorter and easier to reason about them.
Required to simplify #398 (see https://github.com/bitcoin-core/gui/pull/398#discussion_r686094883).
---
**Note for reviewers.** Please verify that "Encrypt Wallet..." menu item, and the following icons
and updated properly in each and every possible scenario.
ACKs for top commit:
jarolrod:
tACK b8aa84b1a1
Talkless:
Code review ACK b8aa84b1a1. Did build on Debian Sid with Qt 5.15.2 but no actual testing performed.
ryanofsky:
Code review ACK b8aa84b1a1. Only change since last review was rebase
Tree-SHA512: 275737cdba02baff71049df41bc24089e916f96326dd2dea26ec607c7949cb3aae368eeabbe3ad5a0a27651503a1d65536873726de854c5f6af259bcc29727e7
017597767b Add I2P network SetReachable/IsReachable unit test assertions (Jon Atack)
b87a9c4d13 Improve doc/i2p.md regarding I2P router options/versions (Jon Atack)
bebcf785c0 Update i2p.md and tor.md regarding -onlynet config option (Jon Atack)
Pull request description:
This pull addresses https://github.com/bitcoin/bitcoin/issues/22634#issuecomment-894104681 and various user feedback/questions, updates the -onlynet documentation in doc/i2p.md and doc/tor.md per #22651 (src/init.cpp is already fine) and fills in some missing I2P unit test coverage.
Note: this PR depends in part on whether #22651 is merged in order to propose the correct -onlynet documentation (it is currently aligned with the change in #22651), so that PR should be decided or merged first.
ACKs for top commit:
Rspigler:
Re-ACK 017597767b
prayank23:
reACK 017597767b
vasild:
ACK 017597767b
Tree-SHA512: ae606437522bfccdfb7508108cddc7dfede2385e30a0561dbd007b784ed2639962c28552eb0e9336412faa323637fe964c26b8d8fc6dcf9fc63734ac00d05736
7720d4f650 test: fix failure in feature_nulldummy.py on single-core machines (Sebastian Falbesoner)
646b3885f7 test: refactor: use named args for block_submit in feature_nulldummy.py (Sebastian Falbesoner)
Pull request description:
On single-core machines, executing the test `feature_nulldummy.py` results in the following assertion error:
```
...
2021-08-18T15:37:58.805000Z TestFramework (INFO): Test 4: Non-NULLDUMMY base multisig transaction is invalid after activation
2021-08-18T15:37:58.814000Z TestFramework (ERROR): Assertion failed
Traceback (most recent call last):
File "[...]/test/functional/test_framework/test_framework.py", line 131, in main
self.run_test()
File "[...]/test/functional/feature_nulldummy.py", line 107, in run_test
self.block_submit(self.nodes[0], [test4tx], accept=False)
File "[...]/test/functional/feature_nulldummy.py", line 134, in block_submit
assert_equal(None if accept else 'block-validation-failed', node.submitblock(block.serialize().hex()))
File "[...]/test/functional/test_framework/util.py", line 49, in assert_equal
raise AssertionError("not(%s)" % " == ".join(str(arg) for arg in (thing1, thing2) + args))
AssertionError: not(block-validation-failed == non-mandatory-script-verify-flag (Dummy CHECKMULTISIG argument must be zero))
2021-08-18T15:37:58.866000Z TestFramework (INFO): Stopping nodes
...
```
There are hardly any single-core machines around anymore, but the behaviour can be reproduced on a multi-core machine by patching the function `GetNumCores()` to return 1 on the master branch and running `feature_nulldummy.py`:
```diff
diff --git a/src/util/system.cpp b/src/util/system.cpp
index 30d410381..149b512fc 100644
--- a/src/util/system.cpp
+++ b/src/util/system.cpp
@@ -1338,7 +1338,7 @@ bool SetupNetworking()
int GetNumCores()
{
- return std:🧵:hardware_concurrency();
+ return 1;
}
```
As solution, parallel script verification is disabled (`-par=1`) and the exact reject reason is checked, which also increases the precision of the test (the possibility that the block is rejected because of another unintended reason is ruled out). See also related PR #22711 which applies the same approach for the p2p segwit test. The PR also includes a refactoring commit which changes the calls to `self.block_submit()` to use named arguments and removes the default value for parameter `accept` (i.e. explicitely passing `accept=...` is mandatory), with the aim to increase the test readability.
ACKs for top commit:
josibake:
ACK 7720d4f650
Saviour1001:
Tested ACK <code>[7720d4f](7720d4f650)</code>
Tree-SHA512: 8a31ebab3e2ab38e555d7a23139b3324a134a0dedc5b879a2419348ae858323882dbbfcbbf88b68e4f8d7eea8cfe43ee19da1d0d2a36c93ae7878c4980cac31d
