👍 laanwj approved a pull request: "build: Drop option to disable hardening."
(https://github.com/bitcoin/bitcoin/pull/32071#pullrequestreview-2689416322)
Concept and code review ACK ecf2046d4b5c43ddf64f62f09cd3ed70dd5caafb
rationale: https://github.com/bitcoin/bitcoin/pull/32038#issuecomment-2717727013 https://github.com/bitcoin/bitcoin/pull/32038#issuecomment-2720530698

💬 maflcko commented on pull request "ci: [lint] Use Cirrus dockerfile cache":
(https://github.com/bitcoin/bitcoin/pull/31948#issuecomment-2728489736)
cc @willcl-ark

💬 Sjors commented on pull request "qt: doc: adapt outdated binary paths to CMake changes":
(https://github.com/bitcoin-core/gui/pull/858#issuecomment-2728501875)
utACK 7ebc458a8cb994bc3c0c129da61353968d955bc2

💬 Sjors commented on pull request "[29.x] backports and rc2":
(https://github.com/bitcoin/bitcoin/pull/32062#issuecomment-2728536081)
ACK f86997eb03d825b7c963adb1463ba83105628a74

I cherry-picked the pull requests, ran `gen-bitcoin-conf.sh` (built with ` -DWITH_ZMQ=ON`) and compared to this PR (minus the version bump commits).

I do get a slightly difference in `share/examples/bitcoin.conf`, maybe due to a build option?

```
-# estimatefee, http, i2p, ipc, leveldb, libevent, mempool,
+# estimatefee, http, i2p, ipc, leveldb, libevent, lock, mempool,
```

📝 fanquake opened a pull request: "depends: patch Qt rounding bugs"
(https://github.com/bitcoin/bitcoin/pull/32081)
When compiled under -O0, this code was causing runtime failures in the 32-bit Clang gui wallet tests. Work around that by importing a commit from upstream: https://github.com/qt/qtbase/commit/8c8b9a4173f4add522ec13de85107deba7c82da0.

Can be recreated with:
```bash
make -C depends/ DEBUG=1 CFLAGS="-O0" CXXFLAGS="-O0" HOST=i686-pc-linux-gnu
cmake -B build --toolchain /root/ci_scratch/depends/i686-pc-linux-gnu/toolchain.cmake -DCMAKE_BUILD_TYPE=Debug -DCMAKE_C_COMPILER='clang;-m32' -DCMAKE_CX
...

✅ fanquake closed a pull request: "build: align debugging flags to `-O0`"
(https://github.com/bitcoin/bitcoin/pull/29796)

💬 fanquake commented on pull request "build: align debugging flags to `-O0`":
(https://github.com/bitcoin/bitcoin/pull/29796#issuecomment-2728537087)
> Can you describe why you think it's important that they align?

Mostly for more intuitive behaviour. I think this was more interesting when it was going to be -Og or -O1. Maybe the better approach is to discuss this (and the passing through of flags) further as part of #31920. I'll split out the Qt cherry-pick and close this PR. At least this has surfaced a couple bugs.

🤔 stratospher reviewed a pull request: "net: Only attempt v2->v1 transport downgrade if online"
(https://github.com/bitcoin/bitcoin/pull/32073#pullrequestreview-2689560351)
`fNetworkActive` doesn't indicate whether our node is online/offline - it's just a node specific option configured using RPC/GUI to enable/disable P2P network activity. So even with this patch, we'd still attempt for a v1 reconnection if we're trying to establish a v2 connection during the time period when our internet goes off :)

(Could you update the online/offline terminology in the PR?)

If we're attempting a v2 connection during the time period when the `fNetworkActive` option is toggl
...

👍 hebasto approved a pull request: "qt: doc: adapt outdated binary paths to CMake changes"
(https://github.com/bitcoin-core/gui/pull/858#pullrequestreview-2689645732)
ACK 7ebc458a8cb994bc3c0c129da61353968d955bc2.

Thank you!

🚀 hebasto merged a pull request: "qt: doc: adapt outdated binary paths to CMake changes"
(https://github.com/bitcoin-core/gui/pull/858)

💬 hebasto commented on pull request "[29.x] backports and rc2":
(https://github.com/bitcoin/bitcoin/pull/32062#issuecomment-2728699985)
> Can also pull in [bitcoin-core/gui#858](https://github.com/bitcoin-core/gui/pull/858) (if not this rc than next).

+1

💬 jsarenik commented on issue "Release Schedule for 29.0":
(https://github.com/bitcoin/bitcoin/issues/31029#issuecomment-2728735213)
Are there any release candidate [builds](https://bitcoincore.org/bin/bitcoin-core-29.0) available yet? All I want is a Linux x86_64 build. What could I do to help? Thanks.

💬 TheCharlatan commented on pull request "build, ci: Fix linking `bitcoin-chainstate.exe` to `bitcoinkernel.dll` on Windows":
(https://github.com/bitcoin/bitcoin/pull/31158#issuecomment-2728786844)
I'm not sure about the current approach of this change. I think we could go the way of manually exporting symbols in our c++ code, but then I would call into question the need for this project to ship a C header in the first place. Having two symbol-exporting headers, one for our internal code, and one for a potential future C header seems very confusing.

Also, as maflcko already said, we could just not directly use these globals from code that links to the shared library. If I understand th
...

💬 maflcko commented on issue "Release Schedule for 29.0":
(https://github.com/bitcoin/bitcoin/issues/31029#issuecomment-2728893884)
@jsarenik If you just want to build a tag or commit for yourself, you can use the `doc/build-*.md` instructions, or the guix build (`contrib/guix/README.md`). With the guix build you can even contribute attestations, see https://github.com/bitcoin/bitcoin/blob/master/doc/release-process.md#building. If you are waiting on the pre-compiled rc binary, I suspect it will be available some time this week.

💬 glozow commented on pull request "[29.x] backports and rc2":
(https://github.com/bitcoin/bitcoin/pull/32062#issuecomment-2728946957)
> Can also pull in [bitcoin-core/gui#858](https://github.com/bitcoin-core/gui/pull/858) (if not this rc than next).

added

💬 suiyuan1314 commented on pull request "docs: fix typos":
(https://github.com/bitcoin/bitcoin/pull/32006#discussion_r1998453937)
Thanks for your reply, I have removed this.

💬 vasild commented on pull request "net: Only attempt v2->v1 transport downgrade if online":
(https://github.com/bitcoin/bitcoin/pull/32073#discussion_r1998469228)
`fNetworkActive` will be read 2 times in `CConnman::DisconnectNodes()`. Given that it can change any time, it could happen that is has one value in the first read and another value here in the second read (this is actually inside a loop so it is many second reads, one for each node). This makes it a bit involved to reason all possible outcomes and could produce unexpected results.

I think it would be better if the variable is read just once in the `CConnman::DisconnectNodes()` method. This wi
...

🤔 vasild reviewed a pull request: "net: Only attempt v2->v1 transport downgrade if online"
(https://github.com/bitcoin/bitcoin/pull/32073#pullrequestreview-2690008496)
Approach ACK 0f4e20728d231935f0140845370090a760ae93af

💬 maflcko commented on pull request "docs: fix typos":
(https://github.com/bitcoin/bitcoin/pull/32006#discussion_r1998471816)
You'll have to adjust the pull title, description and commit message as well to include the rationale. Removing a section is a bit different from just a typo fix.

💬 TheCharlatan commented on pull request "multiprocess: Add libmultiprocess git subtree":
(https://github.com/bitcoin/bitcoin/pull/31741#discussion_r1998479545)
@hebasto should we unset `CMAKE_RUNTIME_OUTPUT_DIRECTORY` here to avoid `mpgen` and `mpgen` landing in the `bin/` directory? Maybe it would be better to move the `libmultiprocess` subtree up one level?