252d1a70fb ci: use Debian Bullseye in ARM CI (fanquake)
Pull request description:
This works around an issue when trying to use `std::filesystem::remove_all` with the ARM GCC on Buster. Has been split out of #20744.
See commentary starting here: https://github.com/bitcoin/bitcoin/pull/20744#issuecomment-810279549.
Also: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93201.
ACKs for top commit:
MarcoFalke:
cr ACK 252d1a70fb
hebasto:
ACK 252d1a70fb, I have reviewed the code and it looks OK, I agree it can be merged.
Tree-SHA512: ca71f5cb07fe06c1c7f0160935e667ffeb62bd6a1a89b54124b5633c5c176347a2207aaa5eca68938ed89db9778d357e42b677115d4ed386fa2d7d2ffa5025ad
3174425255 Cleanup headers after #20788 (Hennadii Stepanov)
Pull request description:
This is a header cleanup after #20788.
ACKs for top commit:
vasild:
ACK 3174425255
Tree-SHA512: 1c21b1ba43841880625289174f10e5b333f6eb857f448e1e4114b1ecdf32a6044ec91c5987c1d66806c1d408a4e3d46569eb41d69a0acb8296601d7c203d9f1d
865ee1af20 Fix Qt test broken by #22219 (Russell Yanofsky)
Pull request description:
It looks like this should have been caught by CI but there might have been a conflict with a recently merged PR like #19101. Failure was reported by fanquake https://github.com/bitcoin/bitcoin/pull/22219#issuecomment-920496509
Fix avoids null WalletClient pointer dereference in address book test by adding MakeWalletClient call and making address book test initialization more consistent with wallet test initialization:
865ee1af20/src/qt/test/addressbooktests.cpp (L63-L66)865ee1af20/src/qt/test/wallettests.cpp (L141-L144)
ACKs for top commit:
fanquake:
ACK 865ee1af20 - I'm merging this now to unbreak the build.
Tree-SHA512: 1f32b7fc79fa79fcf8600d23063896cbc7f8bbcff39d95747ecd546e754581f0f36ece3098ddecded175afccbb3709b4232da39a400dda23b7e550f361b515fb
fad86061e5 doc: Update snap release process for new versioning scheme (MarcoFalke)
Pull request description:
ACKs for top commit:
jarolrod:
ACK fad86061e5
fanquake:
ACK fad86061e5 - moving these instructions to the packaging repo.
Tree-SHA512: 80a9d1484a6544e44a7e57967a62d6e3084efe946de40ac5f0ad0aeb79399d578a7b690186b33d394402d383715741203ebd09ff3f376a5f657045ed96f273e7
e4709c7b56 Start using init makeNode, makeChain, etc methods (Russell Yanofsky)
Pull request description:
Use `interfaces::Init::make*` methods instead of `interfaces::Make*` functions, so interfaces can be constructed differently in different executable without having to change any code. (So for example `bitcoin-gui` can make an `interfaces::Node` pointer that communicates with a `bitcoin-node` subprocess, while `bitcoin-qt` can make an `interfaces::Node` pointer that controls node code in the same process.)
---
This PR is part of the [process separation project](https://github.com/bitcoin/bitcoin/projects/10). The commit was first part of larger PR #10102.
ACKs for top commit:
jamesob:
reACK e4709c7b56
achow101:
ACK e4709c7b56
benthecarman:
utACK e4709c7b56
Tree-SHA512: 580c1979dbb2ef444157c8e53041e70d15ddeee77e5cbdb34f70b6d228cc2d2fe3843825f172da84e506200c58f7e0932f7cd4c006bb5058c1f4e43259394834
This is done by removing an unnecessary loop in Good_() and looping
through the new tables in MakeTried() more efficiently, choosing a
starting value that allow us to stop early in typical cases.
Co-authored-by: John Newbery <john@johnnewbery.com>
Adjust CheckBlockIndex to account for
- assumed-valid block indexes lacking transaction data, and
- setBlockIndexCandidates for the background chainstate not containing certain entries
which rely on assumed-valid ancestors.
Mark the block index entries that are beneath the snapshot base block as
assumed-valid. Subsequent commits will make use of this flag in other
parts of the system.
Instead of (ab)using the existing BLOCK_VALID_* flags to mark CBlockIndex entries which
we haven't yet fully validated (but assume validity for use with UTXO snapshot
loading), introduce a status flag that specifically marks an assumed-valid state.
This state is then removed in RaiseValidity() when the block has actually been
validated.
This distinction will allow us to make the necessary changes to various parts of the
system to facilitate assumeutxo/background chainstate validation but without leaking
details like snapshot height, as we had done previously.
Changes that actually make use of this flag follow in future commits.
Only perform certain behavior (namely that related to servicing
the getblocktemplate RPC call) for the active chainstate when
calling UpdateTip.
Co-authored-by: Jon Atack <jon@atack.com>
PR #22697 introduced a reproducible issue in commit 181a1207 that causes the
addrman tried table to fail consistency checks and significantly lose peer
entries when the `-asmap` configuration option is used.
The issue occurs on bitcoind restart due to an initialization order change
in `src/init.cpp` in that commit whereby CAddrman asmap is set after
deserializing `peers.dat`, rather than before.
Issue reported on the `#bitcoin-core-dev` IRC channel starting at
https://www.erisian.com.au/bitcoin-core-dev/log-2021-08-23.html#l-263.
```
addrman lost 22813 new and 2 tried addresses due to collisions or invalid addresses
ADDRMAN CONSISTENCY CHECK FAILED!!! err=-17
bitcoind: ./addrman.h:707: void CAddrMan::Check() const: Assertion `false' failed. Aborted
```
How to reproduce:
- `git checkout 181a1207` and recompile
- launch bitcoind with `-asmap` and `-checkaddrman=1` config options
- restart bitcoind
- bitcoind aborts on second call to `CAddrMan::Check()`
This commit adds a regression test to reproduce the case; it passes or fails
with the same error.
Co-authored-by: John Newbery <john@johnnewbery.com>
Co-authored-by: Martin Zumsande <mzumsande@gmail.com>
5cc783f5f3 qt: ensure translator comments end in full stop (Jarol Rodriguez)
Pull request description:
This is a follow-up to #318 which addresses this [nit](https://github.com/bitcoin-core/gui/pull/318#discussion_r706856893) by addressing it globally.
This ensures that all GUI translator comments end in a full stop. If a comment does not end in a full stop, a translator may think that the rest of the comment is being cut off.
While here, add a colon to the word "see" for any comments touched which point to look at a link.
ACKs for top commit:
hebasto:
ACK 5cc783f5f3, I have reviewed the code and it looks OK, I agree it can be merged.
shaavan:
Code Review ACK 5cc783f5f3
Tree-SHA512: 67a1d56175c974e0af9b460fa44163f7ce139a7b81cfaf8ed2c0e7fb6d5120957c3135d96010aeb6229689468e36673fe9571b5a8c3e1c07e047aba1bd563444
c88f43f1ac doc: Add historical release notes for 22.0 (W. J. van der Laan)
Pull request description:
.
Top commit has no ACKs.
Tree-SHA512: c074301e7a1fdbd0f0cf6ee662db52cb98b244376baf9fa1ca43692cd5967c963ab417e96736f701ea6cf8215ab9fdd3fe8fe258c93e09ca783931e4f2eb611d
3765c486ef qt: fix bitcoin-qt app categorization on apple silicon (Jarol Rodriguez)
Pull request description:
`System Information` contains many insights into various aspects of your macOS system; the 'Applications' tab contains info on apps. Starting with macOS 11, the 'kind' column under the 'Applications' tab started displaying the CPU architecture of the application. The options are; apple silicon, intel, universal. Previously, the `kind` column indicated where the application originated. The change was made to conveniently determine if the app installed was built to run natively on the new M1 CPU or an intel app that will run under rosetta. Of course, there are several other tools to confirm this; the 'kind' column provides a user-friendly way.
We expect that Bitcoin Core compiled, built, and deployed on an intel CPU will be classified as `Intel`. Similarly, we expect that if this is done on an M1 mac, the resulting app is classified as `Apple Silicon`. In reality, Bitcoin-qt built and deployed on an M1 mac will be classified as `IOS`. This behavior is incorrect and should be fixed.
We fix this by setting the `CFBundleSupportedPlatforms` in our info.plist to the value of `MacOSX`. In doing this, we are telling macOS, "We do not support IOS; stop it!".
Tested and confirmed that this is a no-op on macOS < 11.
| On [#22546](https://github.com/bitcoin/bitcoin/pull/22546) Branch | [#22546](https://github.com/bitcoin/bitcoin/pull/22546) + PR Branch |
| ------------------------------------------------------------------- | ----------- |
|  |  |
**To Test:**
For testing, our base branch will be [#22546](https://github.com/bitcoin/bitcoin/pull/22546). Please perform the following steps on the base branch and then the base branch with the commit from this PR cherry-picked onto it:
- Have an M1 mac
- Compile and deploy bitcoin
- Open up the deployed *.dmg, installed the bundled app
- Eject the bitcoin dmg that should currently be mounted
- Navigate to System Information -> Applications
- Click on the top-left Apple icon
- Click about this mac
- Click on the 'Storage' tab
- Click on the 'Manage...' button
- On the left, click on 'Applications'
- Sort by Name
- Look for the Bitcoin Core application
- Base Branch: The kind column should state that this application is of type `IOS`
- PR Branch: The kind column should state that this application is of type `Apple Silicon`
Note: Intel users on at least macOS 11 can help test by confirming that the application still shows up as kind=`Intel`
ACKs for top commit:
hebasto:
ACK 3765c486ef
Tree-SHA512: 666672025e81e59fe1803859a7f9a4fd3b93a3aba05a163ce223c36081dd579b866d071455608011a19d9ba0c3e9f564cca0c4cb941452f2b51f4ef0dfead1fa
08634e82c6 fix typos in logging messages (ShubhamPalriwala)
d447ded6ba replace: self.nodes[0] with node (ShubhamPalriwala)
dddca3899c test: use MiniWallet in mempool_limit.py (ShubhamPalriwala)
Pull request description:
This is a PR proposed in #20078
This PR enables running another non-wallet functional test even when the wallet is disabled thanks to the MiniWallet, i.e. it can be run even when bitcoin-core is compiled with --disable-wallet.
It also includes changes in wallet.py in the form of a new method, `create_large_transactions()` for the MiniWallet to create large transactions.
Efforts for this feature started in #20874 but were not continued and that PR was closed hence I picked this up.
To test this PR locally, compile and build bitcoin-core without the wallet and run:
```
$ test/functional/mempool_limit.py
```
ACKs for top commit:
amitiuttarwar:
ACK 08634e8, only git changes since last push (and one new line).
Zero-1729:
ACK 08634e82c6🧉
Tree-SHA512: 0f744ad26bf7a5a784aac1ed5077b59c95a36d1ff3ad0087ffd10ac8d5979f7362c63c20c2ce2bfa650fda02dfbcd60b1fceee049a2465c8d221cce51c20369f
9bd168bf54 qt: add missing tooltips to options menu settings (Jarol Rodriguez)
Pull request description:
This adds missing tooltips to the text of the `Size of database cache` and the `Number of script verification threads` settings.
All settings in the Options window will now have appropriate tooltip texts.
ACKs for top commit:
jonatack:
ACK 9bd168bf54 tested on Debian 5.10.46-4 (2021-08-03)
hebasto:
ACK 9bd168bf54, tested on Linux Mint 20.2 (Qt 5.12.8).
Tree-SHA512: d71946bfee33c624a8b79eafe514d2c902090a40bc25097be4c7da4a80270f53305002af1b27d5fd082a0f45f838e22036632f9445918c4b8898073b33c09c08
On master, the deployed bitcoin-qt application is categorized as an IOS application.
This is obviously incorrect, and the built executable is not an IOS executable.
To fix this, we set the CFBundleSupportedPlatforms key in our info.plist
This avoids having to remember to set it whenever mocktime is used with
peer connections. Also, it might help avoiding disconnects when
attaching a debugger to a running test.
This ensures that all gui translator comments end in a full stop.
If a comment does not end in a full stop, a translator may think
that the rest of the comment is being cut off.
While here, add a colon to the word "see" for any comments
touched which point to look at a link.
0b869df1c9 qt: Add cancel button to configuration options popup (Shashwat)
Pull request description:
This PR renames the **OK** button to **Continue** and adds a **Cancel** button to the configuration options pop-up.
This feature will give the user an option to abort opening the configuration file if they want to. This is an essential helpful feature that was missing in the master branch.
In some windows managers such as Windows I3. The exit button at the top right corner is missing. So this feature becomes crucial there. And even when the exit button is there, it doesn't prevent the opening of the configuration file even when pressed.
Additionally, it will always be possible to close using Keyboard Shortcut. This PR helps accessibility for those who need to use a mouse.
<table>
<tr>
<td>Master
</td>
<td>PR
</td>
</tr>
<tr>
<td>
</td>
<td>
</td>
</tr>
</table>
ACKs for top commit:
hebasto:
ACK 0b869df1c9, tested on Linux Mint 20.2 (Qt 5.12.8):
prayank23:
tACK 0b869df1c9
Tree-SHA512: c314e8b84064134f028f66f5015eb0f6ba33d5d4174c9ff49dcb5d2b577dce6019f59f9c7913393a415a323ea98c26febf5ca26e3e2102e7a1d31171e01937f1
The new name describes better what the bool does and also limits the confusion of the three different concepts:
* fBlocksOnly (This bool to skip tx invs)
* -blocksonly (A setting to ignore incoming txs)
* block-relay-only (A connection type in the block-relay-only P2P graph)
