💬 Sjors commented on pull request "Add multiprocess binaries to release build (except Windows)":
(https://github.com/bitcoin/bitcoin/pull/30975#discussion_r1919757571)
I look the latter, and removed the `_linux` and `_darwin` specific variables.

cc @fanquake

💬 Sjors commented on pull request "Add multiprocess binaries to release build (except Windows)":
(https://github.com/bitcoin/bitcoin/pull/30975#issuecomment-2597719684)
Testing https://github.com/bitcoin/bitcoin/pull/30975#discussion_r1919757571 with cherry-picked #31675.

💬 TheCharlatan commented on pull request "guix: use GCC 13 to build releases":
(https://github.com/bitcoin/bitcoin/pull/29881#issuecomment-2597722897)
Guix builds (aarch64):
```
463ef2208f9c5ce3cf884fd2cf8fba8f2aabded19ba43620178aea41e8c8fd1d guix-build-a1add80c80bc/output/aarch64-linux-gnu/SHA256SUMS.part
3a87fe313164fed822437e35511febdbdf0d8e9feda099dfef4fd9031735916f guix-build-a1add80c80bc/output/aarch64-linux-gnu/bitcoin-a1add80c80bc-aarch64-linux-gnu-debug.tar.gz
a20f31417cbb9994eed0e88700c3e1f8c714282425cbe5fb33a5c7ac7b7c68c2 guix-build-a1add80c80bc/output/aarch64-linux-gnu/bitcoin-a1add80c80bc-aarch64-linux-gnu.tar.gz
51fbc8b8a
...

💬 hebasto commented on pull request "Add multiprocess binaries to release build (except Windows)":
(https://github.com/bitcoin/bitcoin/pull/30975#discussion_r1919762346)
FWIW, building the `capnp` package still [fails](https://github.com/bitcoin/bitcoin/pull/31484#issuecomment-2543029821) on OpenBSD.

💬 Sjors commented on pull request "Implement BIP 370 PSBTv2":
(https://github.com/bitcoin/bitcoin/pull/21283#issuecomment-2597726644)
@whitslack @instagibbs thanks for mentioning CLN. Is it possible to turn off its round tripping so I can test this PR against it? And are there any good nodes on signet or testnet4?

👍 maflcko approved a pull request: "ci: Supply `--platform` argument to `docker` commands."
(https://github.com/bitcoin/bitcoin/pull/31657#pullrequestreview-2558295404)
lgtm ACK e1c7c29727faf2d0a5c6fa01505418aac26958fe

I think some code can be removed to be more streamlined, but not a blocker.

💬 maflcko commented on pull request "ci: Supply `--platform` argument to `docker` commands.":
(https://github.com/bitcoin/bitcoin/pull/31657#discussion_r1919767906)
Why the code above? Seems easier to just force "linux" (the fallback/default), which documents the requirements. It seems to work with podman, and I presume also works for the others?



```suggestion
export CI_IMAGE_PLATFORM=${CI_IMAGE_PLATFORM:-"linux"} # native arch by default
```

💬 maflcko commented on pull request "ci: Supply `--platform` argument to `docker` commands.":
(https://github.com/bitcoin/bitcoin/pull/31657#discussion_r1919775445)
> Don't fully grasp the context of #28138 but `native_nowallet_libbitcoinkernel`

If it is no longer happening, I guess the workaround can be removed.

🤔 hebasto reviewed a pull request: "build: Make config warnings fatal if -DWERROR=ON"
(https://github.com/bitcoin/bitcoin/pull/31665#pullrequestreview-2558311790)
I don't like the idea of overloading the `WERROR` build option with additional functionality, as this is not what users would expect.

However, if this PR proceeds in its current direction, the new functionality must be clearly documented in the `WERROR` build option's help string.

> Although one thing that could be improved is making the error more clear, as in the current output, it's hidden far back in the scrollback.

There is no need to continue after encountering an error:
```diff
...

💬 hebasto commented on pull request "build: Make config warnings fatal if -DWERROR=ON":
(https://github.com/bitcoin/bitcoin/pull/31665#discussion_r1919777981)
style nit: 2 space indentation.

💬 Sjors commented on pull request "Add multiprocess binaries to release build (except Windows)":
(https://github.com/bitcoin/bitcoin/pull/30975#discussion_r1919786698)
I'll add a check for `freebsd`.

💬 hebasto commented on pull request "Add multiprocess binaries to release build (except Windows)":
(https://github.com/bitcoin/bitcoin/pull/30975#discussion_r1919791861)
7ab858765d90470f941851d8cf55d616014ee2df:

Why are native packages being removed?

💬 hebasto commented on pull request "init: Lock blocksdir in addition to datadir":
(https://github.com/bitcoin/bitcoin/pull/31674#issuecomment-2597778189)
Concept ACK.

💬 Sjors commented on pull request "Add multiprocess binaries to release build (except Windows)":
(https://github.com/bitcoin/bitcoin/pull/30975#discussion_r1919798868)
Ah that's probably why CI broke.

💬 TheCharlatan commented on pull request "kernel: Introduce initial C header API":
(https://github.com/bitcoin/bitcoin/pull/30595#issuecomment-2597803878)
Updated 4dde75858a3b08f84d71176c7be14bae62020b1f -> 3eefbda3710aad3adc4aac3a6a36e381f9e9f555 ([kernelApi_16](https://github.com/TheCharlatan/bitcoin/tree/kernelApi_16) -> [kernelApi_17](https://github.com/TheCharlatan/bitcoin/tree/kernelApi_17), [compare](https://github.com/TheCharlatan/bitcoin/compare/kernelApi_16..kernelApi_17))

* Get new method for computing cache sizes from https://github.com/bitcoin/bitcoin/pull/31483

💬 fanquake commented on pull request "build: Make config warnings fatal if -DWERROR=ON":
(https://github.com/bitcoin/bitcoin/pull/31665#issuecomment-2597811603)
> I don't like the idea of overloading the WERROR build option with additional functionality, as this is not what users would expect.

Can you elaborate on why this would be unexpected? This is an option to turn warnings into errors, and it'd just be turning more warnings into errors. I'm assuming any user who want to use this locally is exactly the kind of user who'd rather see more warnings than less?

🚀 fanquake merged a pull request: "[test] fix p2p_orphan_handling.py empty orphanage check"
(https://github.com/bitcoin/bitcoin/pull/31675)

💬 fanquake commented on pull request "Add multiprocess binaries to release build (except Windows)":
(https://github.com/bitcoin/bitcoin/pull/30975#issuecomment-2597826643)
https://github.com/bitcoin/bitcoin/actions/runs/12826202676/job/35765674846?pr=30975#step:7:3078:
```bash
node0 2025-01-17T09:33:34.664692Z [capnp-loop] [ipc/capnp/protocol.cpp:35] [IpcLogFn] [ipc] {bitcoin-node-25673/b-capnp-loop-71206} EventLoop::loop done, cancelling event listeners.
node0 2025-01-17T09:33:34.664706Z [capnp-loop] [ipc/capnp/protocol.cpp:35] [IpcLogFn] [ipc] {bitcoin-node-25673/b-capnp-loop-71206} EventLoop::loop bye.
test 2025-01-17T09:33:34.712000Z TestFramework (E
...

💬 Sjors commented on pull request "Add multiprocess binaries to release build (except Windows)":
(https://github.com/bitcoin/bitcoin/pull/30975#issuecomment-2597838860)
@fanquake I'm unable to reproduce that locally, will file an issue on the multiprocess repo to track.

Rebased after #31675 and fix the confusion between freebsd and openbsd.

💬 Sjors commented on pull request "Implement BIP 370 PSBTv2":
(https://github.com/bitcoin/bitcoin/pull/21283#issuecomment-2597892874)
@bigspider thanks for your [reply](https://groups.google.com/g/bitcoindev/c/OPx9qvmr5Nc/m/8mp8VHjQBQAJ) on the mailinglist. So I guess another way to test this PR is to patch HWI to not do the conversion between PSBT 0 and 2, and then make (any?) transaction with a Ledger that runs at least 2.0.0.