Bitcoin Core Github
44 subscribers
120K links
Download Telegram
💬 instagibbs commented on pull request "Ephemeral Dust":
(https://github.com/bitcoin/bitcoin/pull/30239#issuecomment-2334086913)
CI failures seem like spurious timeouts of random tests
💬 sdaftuar commented on pull request "cluster mempool: optimized candidate search":
(https://github.com/bitcoin/bitcoin/pull/30286#issuecomment-2334102610)
> @sdaftuar Is it possible that some of these clusters are "related" (as in are really modifications of one another with a few transactions added/removed)? That could explain the strong correlations between consecutive ones.

I do believe those last 4 examples are likely to be related, as all were found within minutes of each other on the same day.
💬 ryanofsky commented on pull request "interfaces: #30697 follow ups":
(https://github.com/bitcoin/bitcoin/pull/30828#discussion_r1747172543)
In commit "chain: disallow overwriting settings with null values" (4627174d80838e1c6848f9dff47a2ca4a9a35091)

Would be good to just delete this line entirely and make overwriteRwSetting just call updateRwSetting without making a special case for null. Setting a null setting should just delete a setting consistently and not have different behavior for different functions.

(Sorry I didn't notice this line before and mistakenly thought overwriteRwSetting would allow writing null values to the
...
📝 maflcko opened a pull request: " test: Work around boost compilation error "
(https://github.com/bitcoin/bitcoin/pull/30834)
💬 Sjors commented on pull request "test: add mocked Sock that can read/write custom data and/or CNetMessages":
(https://github.com/bitcoin/bitcoin/pull/30205#issuecomment-2334121626)
re-ACK e7cf8e8fc658e4559fcc64279c7bd854773ad6c2

This makes my dimmed zebra very happy.
💬 maflcko commented on pull request "build: work around issue with Boost <= 1.80 and Clang >= 18":
(https://github.com/bitcoin/bitcoin/pull/30821#issuecomment-2334122738)
I still haven't tested this, but the draft at https://github.com/bitcoin/bitcoin/pull/30834 may or may not work around the test compilation error.
💬 fanquake commented on pull request "test: Work around boost compilation error":
(https://github.com/bitcoin/bitcoin/pull/30834#issuecomment-2334129103)
It looks like this works using clang 20 (`Debian clang version 20.0.0 (++20240903115437+0748f4227cd6-1~exp1~20240903115620.935)`) & Boost 1.74.0.
👋 maflcko's pull request is ready for review: "test: Work around boost compilation error"
(https://github.com/bitcoin/bitcoin/pull/30834)
💬 maflcko commented on pull request "test: Work around boost compilation error":
(https://github.com/bitcoin/bitcoin/pull/30834#issuecomment-2334133682)
Thanks for testing. It would be good to also check with Clang 18 (and any other known-to-fail combination).
💬 fanquake commented on pull request "test: Work around boost compilation error":
(https://github.com/bitcoin/bitcoin/pull/30834#issuecomment-2334134054)
Note also, I somehow discovered only now that there is a separate Boost 1.81.0 package available for Bookworm, and the same package is in the backports for Bullseye. So we can also direct any users to those packages.
💬 fanquake commented on pull request "[28.x] Further backports and rc2":
(https://github.com/bitcoin/bitcoin/pull/30827#discussion_r1747184890)
This can now be swapped out for #30834.
💬 sipa commented on issue "28.0rc1 synchronizes much slower on Windows":
(https://github.com/bitcoin/bitcoin/issues/30833#issuecomment-2334157710)
@andrewtoth I cannot imagine that causing a 5x slowdown, though?
💬 hebasto commented on pull request "test: Work around boost compilation error":
(https://github.com/bitcoin/bitcoin/pull/30834#issuecomment-2334163228)
> It would be good to also check with Clang 18 (and any other known-to-fail combination).

Tested on Fedora 40 using the downloaded [Boost 1.74](https://archives.boost.io/release/1.74.0/source/) and commands as follows:
```
$ cmake -B build -DCMAKE_C_COMPILER=clang-18 -DCMAKE_CXX_COMPILER=clang++-18 -DBoost_INCLUDE_DIR=/home/hebasto/boost_1_74_0
$ cmake --build build -t test_bitcoin
```

No build errors.
👍 hebasto approved a pull request: "test: Work around boost compilation error"
(https://github.com/bitcoin/bitcoin/pull/30834#pullrequestreview-2286400951)
ACK fa9d7d5d205ada8915cbbc29599ab8e7bf1fffe0.
💬 jamesob commented on pull request "rpc: add getdescriptoractivity":
(https://github.com/bitcoin/bitcoin/pull/30708#issuecomment-2334181480)
I think this is ready for "actual" review, in case anyone was wondering.
💬 instagibbs commented on pull request "cluster mempool: optimized candidate search":
(https://github.com/bitcoin/bitcoin/pull/30286#issuecomment-2334213839)
Confirmed that at least for example 19, it simply can't find the optimal solution until the commit it's included, even if total iterations are cranked up a couple orders of magnitude.
💬 sipa commented on pull request "cluster mempool: optimized candidate search":
(https://github.com/bitcoin/bitcoin/pull/30286#issuecomment-2334217170)
Example cluster 18:
![example18](https://github.com/user-attachments/assets/7859f0dd-986c-4907-8953-d950bd104f61)
💬 fanquake commented on pull request "cmake: add `USE_SOURCE_PERMISSIONS` to all `configure_file()` usage":
(https://github.com/bitcoin/bitcoin/pull/30823#issuecomment-2334225403)
Guix Build:
```bash
f2e883f15be3d30cc27f1e5ba96ccaa18a9201a242f6dfa7bc65595192249f47 guix-build-1f054eca4e77/output/arm64-apple-darwin/SHA256SUMS.part
53379e0ac143a7c150563ac10973ee7af5dc827e53f96a8b2519cdf1992ece63 guix-build-1f054eca4e77/output/arm64-apple-darwin/bitcoin-1f054eca4e77-arm64-apple-darwin-unsigned.tar.gz
8446ce177d54bac5b185faaecf7a7f34590b512b8ffc690f43d461418196f99c guix-build-1f054eca4e77/output/arm64-apple-darwin/bitcoin-1f054eca4e77-arm64-apple-darwin-unsigned.zip
88
...
💬 zaidmstrr commented on pull request "fix: handle invalid `-rpcbind` port earlier":
(https://github.com/bitcoin/bitcoin/pull/30679#issuecomment-2334241361)
Hey, I'm reviewing this PR, and I saw that the test `run_invalid_bind_test` you implemented in `rpc_bind.py` is not invoking without explicitly providing the `--ipv4` or `--ipv6` flags, so is there any particular reason for that?
📝 hebasto opened a pull request: "build: Introduce "Kernel" installation component"
(https://github.com/bitcoin/bitcoin/pull/30835)
This PR enables building and installing only `libbitcoinkernel`, without the need to disable other targets during the project build system generation:

```
$ rm -rf build && cmake -B build -DBUILD_KERNEL_LIB=ON
$ cmake --build build --target bitcoinkernel
$ cmake --install build --component Kernel --prefix /home/hebasto/INSTALL
-- Install configuration: "RelWithDebInfo"
-- Installing: /home/hebasto/INSTALL/lib/libbitcoinkernel.so
```

Please note, that only the `bitcoinkernel` target i
...