💬 HowHsu commented on pull request "index: handle case where pindex_prev equals chain tip in NextSyncBlock()":
(https://github.com/bitcoin/bitcoin/pull/32875#discussion_r2362131224)
> I still think this should need a test as well, otherwise it just seems like a random change - especially in a PR that claims to change a comment only

Hi l0rinc,

Just like what I said, revert this commit doesn't cause anything wrong, the work this PR does is just to handle a case explicitly rather than leaving it in `return chain.Next(chain.FindFork(pindex_prev));` which is not for that case, and may cause confusion. So this PR is code clean work.
I've updated the PR title, it isn't abou
...

👍 TheCharlatan approved a pull request: "Update libmultiprocess subtree to fix intermittent mptest hang"
(https://github.com/bitcoin/bitcoin/pull/33412#pullrequestreview-3243996515)
ACK c49a43591f88dc199cc04e76f3cfcb7ba136f1a6

💬 vasild commented on pull request "test: don't throw from the destructor of DebugLogHelper":
(https://github.com/bitcoin/bitcoin/pull/33388#issuecomment-3311443431)
`2dcf0c69b8...2427939935`: take https://github.com/bitcoin/bitcoin/pull/33388#discussion_r2359730526

💬 vasild commented on pull request "test: don't throw from the destructor of DebugLogHelper":
(https://github.com/bitcoin/bitcoin/pull/33388#discussion_r2362310458)
Done, thanks!

🤔 hodlinator reviewed a pull request: "log: always print initial signature verification state"
(https://github.com/bitcoin/bitcoin/pull/33336#pullrequestreview-3243827867)
Reviewed 38025edd1978670bfa3c32cf518acd0b01644167

💬 hodlinator commented on pull request "log: always print initial signature verification state":
(https://github.com/bitcoin/bitcoin/pull/33336#discussion_r2362289291)
nit:
```suggestion
- node5 starts with no -assumevalid parameter. Reindex to deterministically hit
"assumevalid hash not in headers" and "below minimum chainwork".
```

💬 hodlinator commented on pull request "log: always print initial signature verification state":
(https://github.com/bitcoin/bitcoin/pull/33336#discussion_r2362319419)
nit: Had a back & forth out-of-band about increasing the `with`-block scopes. While it makes the code layout nicer, it makes for slightly sloppier tests. But initializing a p2p connection and then doing some asserts at the end of the block doesn't seem too bad, so this is nit-level.

💬 hodlinator commented on pull request "log: always print initial signature verification state":
(https://github.com/bitcoin/bitcoin/pull/33336#discussion_r2362210661)
Thanks!

💬 hodlinator commented on pull request "log: always print initial signature verification state":
(https://github.com/bitcoin/bitcoin/pull/33336#discussion_r2362200979)
Don't see much of a point to change `m_prev_script_checks_logged` types twice within the PR (`atomic_bool` -> `optional<bool>` -> `optional<string>`)?
Would suggest swapping the order of 00194b5c and 0ab8939c and then squashing 0ab8939c together with ed9d573d.

💬 hodlinator commented on pull request "log: always print initial signature verification state":
(https://github.com/bitcoin/bitcoin/pull/33336#discussion_r2362231753)
Thanks for combining the variables!

Could you please lay down your aversion to using `const char*` here? Benefits I see over `std::string`:
* Allocation-free.
* Quick comparisons (pointer values rather than fetching the two cache lines where the strings are allocated on the heap and doing string comparisons).

I know it's not performance-critical, but don't see any major drawbacks to it. `std::string` has never been claimed to be a zero-cost abstraction alternative to `const char*`.

💬 hodlinator commented on pull request "log: always print initial signature verification state":
(https://github.com/bitcoin/bitcoin/pull/33336#discussion_r2362266224)
(Thanks for changing to `self.wait_until(lambda: self.nodes[3].getblockcount() == 1)`).

💬 fanquake commented on pull request "p2p: Increase tx relay rate":
(https://github.com/bitcoin/bitcoin/pull/28592#issuecomment-3311507080)
cc @mzumsande

🚀 fanquake merged a pull request: "Update libmultiprocess subtree to fix intermittent mptest hang"
(https://github.com/bitcoin/bitcoin/pull/33412)

✅ fanquake closed an issue: "intermittent timeout in mptest unit test"
(https://github.com/bitcoin/bitcoin/issues/33244)

💬 fanquake commented on issue "intermittent timeout in mptest unit test":
(https://github.com/bitcoin/bitcoin/issues/33244#issuecomment-3311521428)
Closing this, given #33412.

✅ fanquake closed an issue: "intermittent Segmentation fault in mptest IPC path (callFnAsyncParams)"
(https://github.com/bitcoin/bitcoin/issues/33277)

💬 fanquake commented on issue "intermittent Segmentation fault in mptest IPC path (callFnAsyncParams)":
(https://github.com/bitcoin/bitcoin/issues/33277#issuecomment-3311522326)
Closing this, given #33412.

🤔 optout21 reviewed a pull request: "test: don't throw from the destructor of DebugLogHelper"
(https://github.com/bitcoin/bitcoin/pull/33388#pullrequestreview-3244102600)
crACK 2427939935f3e6669be6bf553be89639e0afabaa

💬 fanquake commented on issue "v30rc1 Weird GUI windowing behavior":
(https://github.com/bitcoin/bitcoin/issues/33432#issuecomment-3311531020)
> Doing sudo apt install libxcb-cursor0

Related to #32097.

💬 fanquake commented on issue "Linux download needs installation instructions":
(https://github.com/bitcoin/bitcoin/issues/32097#issuecomment-3311535425)
See #33432.
> Doing sudo apt install libxcb-cursor0 seems to have fixed the issue

💬 hebasto commented on issue "v30rc1 Weird GUI windowing behavior":
(https://github.com/bitcoin/bitcoin/issues/33432#issuecomment-3311581711)
@benthecarman
> Doing `sudo apt install libxcb-cursor0` seems to have fixed the issue but also removed dark mode for me!

What are the first 10 lines of your debug log file in both cases?