< bitcoin-git>
bitcoin/master 34b04ee Hennadii Stepanov: refactor: Add TSA annotations to the WorkQueue class members
< bitcoin-git>
bitcoin/master d2f6d29 MarcoFalke: Merge bitcoin/bitcoin#21895: refactor: Add TSA annotations to the WorkQueu...
< bitcoin-git>
[bitcoin] MarcoFalke merged pull request #21895: refactor: Add TSA annotations to the WorkQueue class members (master...210509-tsa) https://github.com/bitcoin/bitcoin/pull/21895
< bitcoin-git>
[bitcoin] MarcoFalke opened pull request #21909: fuzz: Limit max insertions in timedata fuzz test (master...2105-fuzzMedian) https://github.com/bitcoin/bitcoin/pull/21909
< bitcoin-git>
[bitcoin] jonatack closed pull request #21506: p2p, refactor: make NetPermissionFlags an enum class (master...NetPermissionFlags-enum-class) https://github.com/bitcoin/bitcoin/pull/21506
< bitcoin-git>
[bitcoin] jonatack closed pull request #21690: test: use higher value and per-platform assert in cnetaddr link-local test (master...cnetaddr-link-local-test) https://github.com/bitcoin/bitcoin/pull/21690
< jonatack>
note that the bugfix PR was the second one opened for this issue, had 3 acks, was a blocker for two other PRs (one to make NetPermissions an enum class and one to remove the multi-bit flags), and i thought it should be backported to 0.21.x
< wumpus>
was there anything blocking #21644 from being merged?
< jonatack>
wumpus: imo no but this is a situation i still haven't figured out, e.g. is there silent opposition, lack of time, or it got lost in the stack
< jonatack>
was just cleaning things up before starting five weeks of seminars
< jonatack>
kallewoof: (thanks)
< wumpus>
jonatack: most likely the latter, i mean, it seems impossible to keep on top of the whole stack of PRs, it's good to remind maintainers if something seems like it should be merged but isn't
< hebasto>
was ever "silent opposition" a thing in this project?
< wumpus>
not seeing any opposition so let's reopen 21644 and merge i'd say
< wumpus>
hebasto: if itis, definitely not for non-controversial bugfixes
< wumpus>
i could imagine it in one case: say, a PR for something that has been discussed to death many times since 2011, and everyone is just tired arguing about it again
< wumpus>
but anyhow not the case here, or often
< jonatack>
i could imagine ppl looking at it and thinking, "hm yes but maybe better first to remove the multiflag instead but i don't have time right now, will get back to it" etc
< jonatack>
e.g. order-of-changes or something
< jonatack>
+ everyone is busy + deep stack, yes
< kallewoof>
feel free to ping me when you get stuck on PRs lacking reviews... Work keeps me busy enough to not really be a big contributor at the moment, but I enjoy taking a break and reviewing stuff sometimes. (unless it goes over my head, then sorry, find someone smarter :P)
< bitcoin-git>
[bitcoin] jonatack reopened pull request #21644: p2p, bugfix: use NetPermissions::HasFlag() in CConnman::Bind() (master...NetPermissionFlags-noban-bugfix) https://github.com/bitcoin/bitcoin/pull/21644
< wumpus>
as for #21690, i think having the test depend on local networking configuration is a drawback, though not necessarily prohibitive; some tests necessarily involve interfacing with the operating system, might want to keep them separated from pure utility function tests
< gribble>
https://github.com/bitcoin/bitcoin/issues/21690 | test: use higher value and per-platform assert in cnetaddr link-local test by jonatack · Pull Request #21690 · bitcoin/bitcoin · GitHub
< jonatack>
yes, my understanding is the 32 suffix was just set a bit too low to avoid some local configs when the CI was updated (cirrus), higher value worth giving a try