💬 ismaelsadeeq commented on pull request "test: use address_to_scriptpubkey instead of RPC call":
(https://github.com/bitcoin/bitcoin/pull/27349#discussion_r1150370139)
Thanks.

👍 fanquake approved a pull request: "ci: Use TSan new runtime (llvm-16, take 3)"
(https://github.com/bitcoin/bitcoin/pull/27298)
ACK faf4aca15a319a26aaec7127455f6db97c7039cc - I still see [this](https://github.com/bitcoin/bitcoin/pull/27298#issuecomment-1480041089) failure on aarch64, but that isn't really a regression, as running this tests was already broken for me. I'll open a separate issue, and we can follow up.

💬 glozow commented on pull request "test: refactor: dedup mempool_package_limits.py subtests via decorator":
(https://github.com/bitcoin/bitcoin/pull/27350#discussion_r1150385623)
Ah nice. But might as well just make it this?
```suggestion
high_fee = target_weight * 10
```

💬 glozow commented on pull request "test: refactor: dedup mempool_package_limits.py subtests via decorator":
(https://github.com/bitcoin/bitcoin/pull/27350#discussion_r1150382620)
(1) I think it could be useful to clarify that the subtest modifies mempool and creates the txns
(2) I prefer to reserve the term "submit" for sendraw and submitpackage

```suggestion
# 2) run the subtest, which may submit some transaction(s) to the mempool and create a list of hex transactions
# 3) testmempoolaccept the package hex and check that it fails with the error "package-mempool-limits" for each tx
# 4) after mining a block, clearing the pre-submitted transactions from mempool, su
...

💬 fanquake commented on pull request "refactor (tidy): Fixes after enable-debug configure":
(https://github.com/bitcoin/bitcoin/pull/27353#issuecomment-1486632813)
The failure here is #27316.

💬 fanquake commented on issue "wallet_create_tx.py "Not solvable pre-selected input" exception":
(https://github.com/bitcoin/bitcoin/issues/27316#issuecomment-1486633439)
Also seen here https://cirrus-ci.com/task/4836577990410240

🚀 fanquake merged a pull request: "net: #27257 follow-ups"
(https://github.com/bitcoin/bitcoin/pull/27324)

🚀 fanquake merged a pull request: "fuzz: Remove legacy int parse fuzz tests"
(https://github.com/bitcoin/bitcoin/pull/27344)

💬 fanquake commented on issue "-Wmaybe-uninitialized warnings under LTO":
(https://github.com/bitcoin/bitcoin/issues/27343#issuecomment-1486680561)
> Fuzz ones removed in https://github.com/bitcoin/bitcoin/pull/27344

Thanks. Retested at 8d31d769b7b504104f162a03c94bbc95dec7d849 and updated the op to drop those (other than in float).

💬 MarcoFalke commented on issue "-Wmaybe-uninitialized warnings under LTO":
(https://github.com/bitcoin/bitcoin/issues/27343#issuecomment-1486706789)
Yeah, that is a false positive. Not sure if they are going to fix it, but if you want to work around, you can try something like:


```diff
diff --git a/src/test/fuzz/float.cpp b/src/test/fuzz/float.cpp
index 8714ab8a04..57f8d4cec8 100644
--- a/src/test/fuzz/float.cpp
+++ b/src/test/fuzz/float.cpp
@@ -7,11 +7,13 @@
#include <test/fuzz/fuzz.h>
#include <test/fuzz/util.h>
#include <util/serfloat.h>
+#include <util/check.h>
#include <version.h>

#include <cassert>
#include <c
...

⚠️ MarcoFalke opened an issue: "wallet: Data race in GetOrCreateLegacyScriptPubKeyMan vs IsMine"
(https://github.com/bitcoin/bitcoin/issues/27354)
### Is there an existing issue for this?

- [X] I have searched the existing issues

### Current behaviour

tsan error

### Expected behaviour

no tsan error

### Steps to reproduce

I presume the error is intermittent, but the basic idea is that the validation interface thread (scheduler) may connect blocks and sync wallet transactions, thus call `IsMine` and access `m_spk_managers`. At the same time, the main thread may call `GetOrCreateLegacyScriptPubKeyMan`, also accessing `m_spk_managers`,
...

👍 willcl-ark approved a pull request: "init: Error if ignored bitcoin.conf file is found"
(https://github.com/bitcoin/bitcoin/pull/27302)
ACK ac9fee615

@ryanofsky Thanks for the extra clarification. I agree (and have now tested) that this PR does achieve that correctly.

I left a few tiny nits that don't need to be adressed unless you retouch (and want to address them).

💬 willcl-ark commented on pull request "init: Error if ignored bitcoin.conf file is found":
(https://github.com/bitcoin/bitcoin/pull/27302#discussion_r1150313881)
On a re-touch similar re-wording could be done here?

💬 willcl-ark commented on pull request "init: Error if ignored bitcoin.conf file is found":
(https://github.com/bitcoin/bitcoin/pull/27302#discussion_r1150313450)
If you re-touch, perhaps this could read:

'Test error is triggered when the datadir in use contains a bitcoin.conf file that would be ignored '

💬 willcl-ark commented on pull request "init: Error if ignored bitcoin.conf file is found":
(https://github.com/bitcoin/bitcoin/pull/27302#discussion_r1150550636)
There is `get_datadir_path()` in _test_framework/util.py_:

https://github.com/bitcoin/bitcoin/blob/b759cefe936ed3991633acff215ea1dcec5ece28/test/functional/test_framework/util.py#L417-L418

So perhaps could go next to that in a future PR?

💬 willcl-ark commented on pull request "init: Error if ignored bitcoin.conf file is found":
(https://github.com/bitcoin/bitcoin/pull/27302#discussion_r1150571443)
^ this can be resolved.

💬 TheCharlatan commented on pull request "init: Error if ignored bitcoin.conf file is found":
(https://github.com/bitcoin/bitcoin/pull/27302#issuecomment-1486858153)
Should this get its own release note? There are bound to be some node operators that will be caught off-guard by this change.

💬 ismaelsadeeq commented on pull request "test: use address_to_scriptpubkey instead of RPC call":
(https://github.com/bitcoin/bitcoin/pull/27349#issuecomment-1486864385)
> @ismaelsadeeq seems ready for review to me
Ready for review now.

💬 furszy commented on pull request "refactor: Remove CAddressBookData::destdata":
(https://github.com/bitcoin/bitcoin/pull/27224#discussion_r1150513279)
nit: could use structured bindings.
```c++
for (const auto& [dest, entry] : m_address_book) {
for (const auto& [id, request] : entry.receive_requests) {
values.emplace_back(request);
}
}
```

💬 furszy commented on pull request "refactor: Remove CAddressBookData::destdata":
(https://github.com/bitcoin/bitcoin/pull/27224#discussion_r1150541159)
Would be good to keep the encapsulation and don't access `m_address_book` from outside of the wallet class (otherwise we are reverting #25337 and, future, #26836).