💬 maflcko commented on issue "CI timeouts":
(https://github.com/bitcoin/bitcoin/issues/30851#issuecomment-2354875962)
I am seeing a 2h timeout on running `clang --version`, even before any CI-specific code is run: https://github.com/maflcko/bitcoin-core-with-ci/actions/runs/10897268811/job/30238399156

So I fail to see how this could be anything other than a GitHub hardware, software, or data corruption bug.

I don't think this can be fixed on our side, apart from removing the task.

💬 TheCharlatan commented on pull request "depends: sqlite 3.46.1":
(https://github.com/bitcoin/bitcoin/pull/29991#issuecomment-2354884459)
Guix build (aarch64):
```
773a4d0dbc05ceb91439802ce3374b252f45d2a4c69c3fa5930d9450d8c1f340 guix-build-def6dd0c597f/output/aarch64-linux-gnu/SHA256SUMS.part
c7719110515db339e6ca87f8255338eb21ed2df2fa89e2ae72a10860a6fff2a5 guix-build-def6dd0c597f/output/aarch64-linux-gnu/bitcoin-def6dd0c597f-aarch64-linux-gnu-debug.tar.gz
83a75cb1242f725f234e16799791890d0aebd2efab1632ac3db300b73ad664a9 guix-build-def6dd0c597f/output/aarch64-linux-gnu/bitcoin-def6dd0c597f-aarch64-linux-gnu.tar.gz
f6d09f3839
...

💬 maflcko commented on pull request "ci: Use macos-14 GHA image":
(https://github.com/bitcoin/bitcoin/pull/30913#discussion_r1762647527)
That sounds like a bug in the CI config. However, it should be harmless, because `HOST` is unused, because depends isn't used here.

I'll add an unrelated bugfix commit to remove the wrong config value.

💬 stratospher commented on pull request "addrman: change internal id counting to int64_t":
(https://github.com/bitcoin/bitcoin/pull/30568#discussion_r1762661233)
I'm fine with the version on the branch now because `Serialise()` and `Unserialise()` are symmetrical. `nIndex` is [written as int](https://github.com/bitcoin/bitcoin/blob/51f7668d31e2624e41c7ce77fe33162802808f3f/src/addrman.cpp#L220) into peers.dat in `Serialise()` and [read as int](https://github.com/bitcoin/bitcoin/blob/51f7668d31e2624e41c7ce77fe33162802808f3f/src/addrman.cpp#L325) from peers.dat in `Unserialise()`.

💬 stratospher commented on pull request "addrman: change internal id counting to int64_t":
(https://github.com/bitcoin/bitcoin/pull/30568#issuecomment-2354910571)
ACK 51f7668d31e2624e41c7ce77fe33162802808f3f. I think it's a good change to make the nId space large(64 bits) so that the nId values are distinct.

🤔 hebasto reviewed a pull request: "build: speed up by flattening the dependency graph"
(https://github.com/bitcoin/bitcoin/pull/30911#pullrequestreview-2308818404)
Concept ACK.

a39a915c226775396f6505efe77a4aba4d0ae3ad
Did you consider setting `CMAKE_ADD_CUSTOM_COMMAND_DEPENDS_EXPLICIT_ONLY` globally?

aaf30ef89465c18228b4e46fd1378b9087274cbe
> turns out this triggers an [upstream bug](https://gitlab.kitware.com/cmake/cmake/-/issues/24058), which I guess can be considered an edge-case until fixed in CMake. I've added 2 per-lib opt-outs as a result.

This bug may also affect the `crc32c` library after [switching](https://github.com/bitcoin/bitcoin/p
...

💬 Sjors commented on pull request "multiprocess: Add IPC wrapper for Mining interface":
(https://github.com/bitcoin/bitcoin/pull/30510#discussion_r1762668345)
02c6a1ed58867fa14b1e0c42f476ef45e64ea7ea `-> (result: BlockTemplate);`

🚀 fanquake merged a pull request: "guix: (explicitly) build Linux GCC with `--enable-cet`"
(https://github.com/bitcoin/bitcoin/pull/30438)

💬 maflcko commented on pull request "build: speed up by flattening the dependency graph":
(https://github.com/bitcoin/bitcoin/pull/30911#issuecomment-2354922939)
> ... though currently only Ninja benefits from this being set, and only CMake >= 3.27 understands it.

I wonder if one CI task should be using Ninja (and cmake >= 3.27), if it isn't too hard to implement. Otherwise this config will remain untested and errors may sneak in to the master branch, only being detected after merge.

💬 hebasto commented on pull request "build: speed up by flattening the dependency graph":
(https://github.com/bitcoin/bitcoin/pull/30911#discussion_r1762679548)
A note for reviewers. The Guix builds are using CMake 3.24.2 (the `cmake-minimal` package).

🤔 hebasto reviewed a pull request: "ci: Use macos-14 GHA image"
(https://github.com/bitcoin/bitcoin/pull/30913#pullrequestreview-2308880257)
Concept ACK.

Perhaps explicitly mention the switch from the `x86_64` architecture to `arm64`?

👋 fanquake's pull request is ready for review: "depends: Qt 5.15.15"
(https://github.com/bitcoin/bitcoin/pull/30774)

💬 fanquake commented on pull request "depends: Qt 5.15.15":
(https://github.com/bitcoin/bitcoin/pull/30774#issuecomment-2355041535)
> but wasn't the idea to use qt6 once cmake was done? IIRC there was a tracking/meta issue, but I can't find it?

Maybe #24798? It's not really clear what the status of any migration is.

💬 maflcko commented on issue "Increasing self-hosted runner raw performance":
(https://github.com/bitcoin/bitcoin/issues/30852#issuecomment-2355045583)
> So I think the best way forward would be to have all workers of the same type on the same machine. (I am still running some performance tests to get an estimate if this make sense at all)

Did a quick sanity check with a 100G shared ccache on a single machine `cpx51` and all tasks finished in under a hour. Some tasks were faster than the AX52 run above, some were slower:

![Screenshot 2024-09-17 at 11-11-05 Merge commit 'fa99e4521b6fc0e7f6636d40bc0d6a7325227374' into main-with-ci - Cirrus
...

💬 maflcko commented on pull request "ci: Use macos-14 GHA image (x86_64-apple-darwin22.6.0 -> arm64-apple-darwin23.6.0)":
(https://github.com/bitcoin/bitcoin/pull/30913#issuecomment-2355068096)
> Perhaps explicitly mention the switch from the `x86_64` architecture to `arm64`?

Sure, done

💬 dergoegge commented on pull request "doc: update Eclipser fuzzing documentation":
(https://github.com/bitcoin/bitcoin/pull/30908#issuecomment-2355075262)
I think we could even remove eclipser from the docs entirely. As you noted as well, it looks unmaintained and I don't think any of us are actively using it.

That being said it would be nice to have an example of a maintained concolic testing tool in the docs (e.g. https://github.com/eurecom-s3/symqemu) but that could be done as a follow up.

💬 laanwj commented on pull request "streams: cache file position within AutoFile":
(https://github.com/bitcoin/bitcoin/pull/30884#discussion_r1762896709)
> Is there any reason why AutoFile::ignore(nSize) couldn't be removed and AutoFile::seek(nSize, SEEK_CURR) used instead?

Keep in mind not all streams support seeking, so it would at the least be a conditional optimization. Agree it's not worth it as long as the amounts skipped are small (generally smaller or equal to the page size).

💬 laanwj commented on pull request "streams: cache file position within AutoFile":
(https://github.com/bitcoin/bitcoin/pull/30884#issuecomment-2355119437)
Code review ACK a240e150e837b5a95ed19765a2e8b7c5b6013f35

💬 maflcko commented on pull request "doc: update Eclipser fuzzing documentation":
(https://github.com/bitcoin/bitcoin/pull/30908#issuecomment-2355137275)
> That being said it would be nice to have an example of a maintained concolic testing tool in the docs (e.g. https://github.com/eurecom-s3/symqemu) but that could be done as a follow up.

Agree. Also, it would be nice to (at the same time) add a synthetic bug (or prior real bug) that shows how the tool works and shows that (in this specific instance at least) it provides real value over just afl+libFuzzer+honggfuzz, using a saturated or at least well-developed corpus based on the existing dev
...

💬 marcofleon commented on pull request "refactor: TxDownloadManager + fuzzing":
(https://github.com/bitcoin/bitcoin/pull/30110#issuecomment-2355149293)
Ran `txdownloadman_impl` and it crashed on this assertion:
```
operator(): Assertion `!txdownload_impl.RecentRejectsFilter().contains(package.front()->GetWitnessHash().ToUint256())' failed.
```
Base64 of the input: 4d3Myytlcf9a3czLK2UxoFrjzMtlMQBSWt3Myyv//w==

Input file if it's any easier:
[txdownload_impl_crash.txt](https://github.com/user-attachments/files/17026573/txdownload_impl_crash.txt)