💬 pinheadmz commented on issue "getbestblockhash is sometimes taking a very long time":
(https://github.com/bitcoin/bitcoin/issues/32733#issuecomment-3109585516)
I am running Bitcoin Core v28.0.0 with `txindex=1` on an RPi 4 with 8 GB RAM and 1 TB USB-SSD and I have no performance issues at all. I also run LND and some other bitcoin monitoring software that makes RPCs every 1 second.

Just for a sanity check, can you ensure that your bitcoin data directory is being written to external SSD and not the MicroSD card?

💬 furszy commented on pull request "index: initial sync speedup, parallelize process":
(https://github.com/bitcoin/bitcoin/pull/26966#discussion_r2226285266)
Absolutely

💬 glozow commented on pull request "p2p: TxOrphanage revamp cleanups":
(https://github.com/bitcoin/bitcoin/pull/32941#discussion_r2226282529)
done

💬 glozow commented on pull request "p2p: TxOrphanage revamp cleanups":
(https://github.com/bitcoin/bitcoin/pull/32941#discussion_r2226282789)
done

💬 glozow commented on pull request "p2p: TxOrphanage revamp cleanups":
(https://github.com/bitcoin/bitcoin/pull/32941#discussion_r2226282376)
Fixed the typos. I didn't mention the `getorphantxs` fields since we haven't mentioned them in release notes before

💬 glozow commented on pull request "p2p: TxOrphanage revamp cleanups":
(https://github.com/bitcoin/bitcoin/pull/32941#discussion_r2226286848)
Added the option again, a warning, and a more explicit recommendation in the release notes.

💬 glozow commented on pull request "p2p: TxOrphanage revamp cleanups":
(https://github.com/bitcoin/bitcoin/pull/32941#discussion_r2226284019)
Thanks, removed that commit and leaving as is

💬 furszy commented on pull request "index: initial sync speedup, parallelize process":
(https://github.com/bitcoin/bitcoin/pull/26966#discussion_r2226304853)
> In " init: provide thread pool to indexes " [82fa9d2](https://github.com/bitcoin/bitcoin/commit/82fa9d29653b445118fc2d03e2ced520a5d4c7dc)
>
> Define the max and then use it here and in checking the bounds

Since the maximum number of threads for indexes should depend on how many threads Core has at runtime and the number of available processors, I don't think hardcoding it here is the best approach. I agree with adding it on the error side: https://github.com/bitcoin/bitcoin/pull/26966#di
...

💬 furszy commented on pull request "index: initial sync speedup, parallelize process":
(https://github.com/bitcoin/bitcoin/pull/26966#discussion_r2226307172)
> In "util: introduce general purpose thread pool" [7d984c3](https://github.com/bitcoin/bitcoin/commit/7d984c3b2dca085ef7f49d21568b06c7b89c7807)
>
> Also verify that no work queue size is 0.

sure. Done.

💬 pinheadmz commented on pull request "fees: rpc: `estimatesmartfee` now returns a fee rate estimate during low network activity":
(https://github.com/bitcoin/bitcoin/pull/32395#issuecomment-3109644770)
concept ACK -- I think like https://github.com/bitcoin/bitcoin/pull/32528 a little fudge factor against what is "most correct" is worth the tradeoff for a more predictable user experience, especially on test networks

💬 furszy commented on pull request "index: initial sync speedup, parallelize process":
(https://github.com/bitcoin/bitcoin/pull/26966#discussion_r2226319906)
Done

🤔 glozow reviewed a pull request: "package validation: relax the package-not-child-with-unconfirmed-parents rule"
(https://github.com/bitcoin/bitcoin/pull/31385#pullrequestreview-3048518467)
Thanks @ishaanam! Took all suggestions

💬 glozow commented on pull request "package validation: relax the package-not-child-with-unconfirmed-parents rule":
(https://github.com/bitcoin/bitcoin/pull/31385#discussion_r2226318223)
changed them to `parent_relative, child_relative`

💬 glozow commented on pull request "package validation: relax the package-not-child-with-unconfirmed-parents rule":
(https://github.com/bitcoin/bitcoin/pull/31385#discussion_r2226317096)
Good point, added a definition

💬 glozow commented on pull request "package validation: relax the package-not-child-with-unconfirmed-parents rule":
(https://github.com/bitcoin/bitcoin/pull/31385#discussion_r2226317312)
done

💬 glozow commented on pull request "package validation: relax the package-not-child-with-unconfirmed-parents rule":
(https://github.com/bitcoin/bitcoin/pull/31385#discussion_r2226318676)
done, thanks

💬 furszy commented on pull request "index: initial sync speedup, parallelize process":
(https://github.com/bitcoin/bitcoin/pull/26966#discussion_r2226322535)
done as suggested

💬 achow101 commented on pull request "wallet, rpc: add v3 transaction creation and wallet support":
(https://github.com/bitcoin/bitcoin/pull/32896#discussion_r2226356487)
In 2bcdf53a58ee6cd7c70f2afd97a16fce5dcc21c6 "rpc: Support v3 raw transactions creation"

If `createrawtransaction` is going to support setting the version, then `createpsbt`, `walletcreatefundedpsbt`, `send`, and `sendall` need to as well.

💬 achow101 commented on pull request "wallet, rpc: add v3 transaction creation and wallet support":
(https://github.com/bitcoin/bitcoin/pull/32896#discussion_r2226343248)
In 40c6d135f37a1d8dd64cbea3ca30229c2e012fa2 "wallet: throw error at conflicting tx versions in pre-selected inputs"

We want to avoid querying `mapWallet` as much as possible for performance. `txo.GetWalletTx` returns a reference to the `CWalletTx` that can be used here instead of a `mapWallet` lookup.

👍 l0rinc approved a pull request: "test: check proper OP_2ROT behavior"
(https://github.com/bitcoin/bitcoin/pull/33047#pullrequestreview-3048598782)
ACK b94c6356a29b03def6337c91caabb3b8642187e8

I can confirm that the whole test suite passes when the stack erase is removed from the script interpreter code.

<details>
<summary>Details</summary>

```bash
rm -rf build && cmake -B build -G Ninja -DCMAKE_BUILD_TYPE=Debug && ninja -C build && build/bin/test_bitcoin --run_test=script_tests
```

```diff
diff --git a/src/script/interpreter.cpp b/src/script/interpreter.cpp
index 61ea7f4503..efb1af8137 100644
--- a/src/script/interpreter.
...