👋 hebasto's pull request is ready for review: "build: Enhance Ccache performance across worktrees and build trees"
(https://github.com/bitcoin/bitcoin/pull/30861)

💬 glozow commented on pull request "test: p2p: fix sending of manual INVs in tx download test":
(https://github.com/bitcoin/bitcoin/pull/31658#issuecomment-2596595387)
> > Good find! I think that `p2p_orphan_handling.py` has a similar [spot](https://github.com/bitcoin/bitcoin/blob/e7c479495509c068215b73f6df070af2d406ae15/test/functional/p2p_orphan_handling.py#L431).
>
> Ah right, missed that one. If I interpret the `test_orphan_inherit_rejection` sub-test correctly, we want to send the INV with MSG_TX there to "announce again with potentially different witness", so the solution in this case to avoid ignoring the INV would be to add peer3 with wtxidrelay=Fal
...

💬 hebasto commented on pull request "build: Enhance Ccache performance across worktrees and build trees":
(https://github.com/bitcoin/bitcoin/pull/30861#issuecomment-2596595610)
> What is the status of this?

Reworked. The PR description has been updated.

💬 hebasto commented on pull request "build: Enhance Ccache performance across worktrees and build trees":
(https://github.com/bitcoin/bitcoin/pull/30861#discussion_r1919073119)
I did not test this branch for code coverage reports.

💬 instagibbs commented on pull request "[test] fix p2p_orphan_handling.py empty orphanage check":
(https://github.com/bitcoin/bitcoin/pull/31675#issuecomment-2596606081)
ACK 2e75ebb6169da08b04c4769555c4c84d6b5ca0ec

don't see any other problematic orphanage checks

💬 hebasto commented on pull request "build: Enhance Ccache performance across worktrees and build trees":
(https://github.com/bitcoin/bitcoin/pull/30861#discussion_r1919076576)
If it happens, the project might be re-built with `-DWITH_CCACHE=OFF`.

💬 glozow commented on pull request "p2p: track and use all potential peers for orphan resolution":
(https://github.com/bitcoin/bitcoin/pull/31397#discussion_r1919078592)
Good point. I seem to recall drafting a version that randomly selected 1 peer. That maybe saves us from a peer purposefully disconnecting (since they don't know whether we assigned them), but doesn't have this redundancy problem? What do you think?

💬 glozow commented on pull request "p2p: track and use all potential peers for orphan resolution":
(https://github.com/bitcoin/bitcoin/pull/31397#discussion_r1919083797)
Added in #31666

💬 glozow commented on pull request "p2p: track and use all potential peers for orphan resolution":
(https://github.com/bitcoin/bitcoin/pull/31397#discussion_r1919085298)
Ah, makes sense - I thought you meant a different bit of code. Added in #31666

💬 maflcko commented on pull request "build: Enhance Ccache performance across worktrees and build trees":
(https://github.com/bitcoin/bitcoin/pull/30861#discussion_r1919085335)
This should be harmless to set in the CI, because dir should be the same for all tasks anyway. If it isn't then the compiler should differ as well.

💬 glozow commented on pull request "p2p: track and use all potential peers for orphan resolution":
(https://github.com/bitcoin/bitcoin/pull/31397#discussion_r1919085616)
I feel like they're about the same, so going to leave as is

💬 glozow commented on pull request "p2p: track and use all potential peers for orphan resolution":
(https://github.com/bitcoin/bitcoin/pull/31397#discussion_r1919085794)
Added in #31666

💬 glozow commented on pull request "p2p: track and use all potential peers for orphan resolution":
(https://github.com/bitcoin/bitcoin/pull/31397#discussion_r1919085877)
Added in #31666

💬 glozow commented on pull request "p2p: track and use all potential peers for orphan resolution":
(https://github.com/bitcoin/bitcoin/pull/31397#discussion_r1919085955)
Added in #31666

💬 glozow commented on pull request "p2p: track and use all potential peers for orphan resolution":
(https://github.com/bitcoin/bitcoin/pull/31397#discussion_r1919086045)
Added in #31666

💬 hebasto commented on pull request "build: Enhance Ccache performance across worktrees and build trees":
(https://github.com/bitcoin/bitcoin/pull/30861#issuecomment-2596652055)
The CI failure is [unrelated](https://github.com/bitcoin/bitcoin/pull/31675).

💬 hebasto commented on pull request "build: Enhance Ccache performance across worktrees and build trees":
(https://github.com/bitcoin/bitcoin/pull/30861#discussion_r1919101037)
This is no longer relevant for the recent [push](https://github.com/bitcoin/bitcoin/pull/30861#issuecomment-2596595610).

🤔 pablomartin4btc reviewed a pull request: "gui, psbt: Use SIGHASH_DEFAULT when signing PSBTs"
(https://github.com/bitcoin-core/gui/pull/850#pullrequestreview-2557125142)
utACK 3e97ff9c5eaa3160426ba112930b047404c54c9e

I agree with @Sjors but tha'ts not covered by doc at the `/** Signature hash types/flags */` `enum` definition (`src/script/interpreter.h`)?

> https://github.com/bitcoin/bitcoin/commit/e94bb4e5fc6dea67844a8495b948b679649b7b65 makes this problem go away though.

is that part of a PR?

💬 maflcko commented on pull request "init: Lock blocksdir in addition to datadir":
(https://github.com/bitcoin/bitcoin/pull/31674#discussion_r1919104106)
I guess it is fine to ignore any "hidden" files (starting with `.`) as an alternative.

💬 hebasto commented on pull request "build: Enhance Ccache performance across worktrees and build trees":
(https://github.com/bitcoin/bitcoin/pull/30861#discussion_r1919104517)
In the recent [push](https://github.com/bitcoin/bitcoin/pull/30861#issuecomment-2596595610), this branch relies on setting the `base_dir` option by the user because the required value depends on actual user's build environment.