⚠️ alexpkv opened an issue: "bitcoin ?"
(https://github.com/bitcoin/bitcoin/issues/31695)

✅ fanquake closed an issue: "bitcoin ?"
(https://github.com/bitcoin/bitcoin/issues/31695)

:lock: fanquake locked an issue: "bitcoin ?"
(https://github.com/bitcoin/bitcoin/issues/31695)

💬 achow101 commented on pull request "psbt: MuSig2 Fields":
(https://github.com/bitcoin/bitcoin/pull/31247#discussion_r1922921473)
Done

💬 achow101 commented on pull request "psbt: MuSig2 Fields":
(https://github.com/bitcoin/bitcoin/pull/31247#discussion_r1922921513)
Fixed

💬 achow101 commented on pull request "psbt: MuSig2 Fields":
(https://github.com/bitcoin/bitcoin/pull/31247#discussion_r1922921546)
Fixed

💬 achow101 commented on pull request "psbt: MuSig2 Fields":
(https://github.com/bitcoin/bitcoin/pull/31247#discussion_r1922921575)
Fixed

💬 kevkevinpal commented on pull request "lint: Call more checks from test_runner":
(https://github.com/bitcoin/bitcoin/pull/31653#discussion_r1922924258)
I think we can delete this line and then do the following on line 12
`export COMMIT_RANGE="HEAD~..HEAD"`

since now this export sometimes is empty

💬 EthanHeilman commented on pull request "fuzz: Expand script verification flag testing to segwit v0 and tapscript":
(https://github.com/bitcoin/bitcoin/pull/31460#issuecomment-2603480439)
@dergoegge This PR encouraged me to break the unit testing improvements I made to the Tapscript unit tests out the OP_CAT PR and into their own PR here: https://github.com/bitcoin/bitcoin/pull/31640

💬 l0rinc commented on pull request "doc: Amend notes on benchmarking":
(https://github.com/bitcoin/bitcoin/pull/31690#discussion_r1923155638)
It's not obvious to everyone. This is a useful change.
Some of the overly harsh/opinionated parts could indeed be dropped/reformulated, but it's important that newcomers know the basic directions.

💬 maflcko commented on pull request "ci: Skip read-write of default env vars":
(https://github.com/bitcoin/bitcoin/pull/31678#issuecomment-2603828725)
rebased

💬 maflcko commented on pull request "ci: Enable DEBUG=1 for one GCC-12+ build to catch 117966 regressions":
(https://github.com/bitcoin/bitcoin/pull/31522#issuecomment-2603830583)
rebased on centos 10 and silenced the gcc warning

💬 maflcko commented on pull request "doc: Amend notes on benchmarking":
(https://github.com/bitcoin/bitcoin/pull/31690#discussion_r1923198442)
> `Supporting the case for a performance improvement with benchmark results is encouraged`.

The benchmark also needs to be put into context in what real end-to-end scenario it would result in a measurable improvement. Just because a benchmark exists or can be written, doesn't mean that it is a goal to improve its performance. I think noting that the considerations are not one-dimensional is useful.

💬 NISTDoD commented on something "":
(https://github.com/bitcoin/bitcoin/commit/7629d36a530087806e80ed3c5866e8ec64319828#r151568735)
[5.tar.gz](https://github.com/user-attachments/files/18486354/5.tar.gz)

💬 maflcko commented on pull request "doc: Amend notes on benchmarking":
(https://github.com/bitcoin/bitcoin/pull/31690#discussion_r1923199247)
I am happy to push any change (ideally a diff) on the wording, which conveys the essence of the note better.

💬 maflcko commented on pull request "lint: Call more checks from test_runner":
(https://github.com/bitcoin/bitcoin/pull/31653#discussion_r1923223310)
The two should be equivalent, right? Happy to push, if I have to touch again.

💬 maflcko commented on pull request "lint: Call more checks from test_runner":
(https://github.com/bitcoin/bitcoin/pull/31653#discussion_r1923223396)
> looks like this adds a new log added to the CI (which I think is fine)

It shouldn't add a new log to the CI. The commits were logged in the CI before this pull request (See the `echo`).

The goal of this pull request is to move the logic into the test_runner, so that it is easier to run the test_runner locally and replicate the CI.

💬 maflcko commented on pull request "lint: Call more checks from test_runner":
(https://github.com/bitcoin/bitcoin/pull/31653#discussion_r1923223434)
I think it is unlikely for `git log` to fail, but since `RUST_BACKTRACE=1` is set, the error should be self-explanatory anyway. Possibly `expect` could be replaced by `unwrap`?

🤔 maflcko reviewed a pull request: "bench: fix used file that is not opened"
(https://github.com/bitcoin/bitcoin/pull/31693#pullrequestreview-2563757732)
What are the exact steps to reproduce or test this problem and fix?

💬 maflcko commented on pull request "bench: fix used file that is not opened":
(https://github.com/bitcoin/bitcoin/pull/31693#discussion_r1923241566)
Why would it not exist? If this can happen, I don't think the right fix will be to add this code to every unit/bench/fuzz test.