💬 hebasto commented on pull request "build: Introduce CMake-based build system":
(https://github.com/bitcoin/bitcoin/pull/30454#discussion_r1749271553)
The unused dead code has been deleted in https://github.com/bitcoin/bitcoin/pull/30803.

💬 fanquake commented on pull request "build: Introduce CMake-based build system":
(https://github.com/bitcoin/bitcoin/pull/30454#discussion_r1749271611)
I was under the impression that happened automatically on wine installation. Will also re-test.

💬 hebasto commented on pull request "build: Minor build system fixes and amendments":
(https://github.com/bitcoin/bitcoin/pull/30803#issuecomment-2336730196)
> lgtm ACK [cc9d65d](https://github.com/bitcoin/bitcoin/commit/cc9d65dc05f802aaaa108a2edcf30c3758c3f459)
>
> Let's wait for the outcome of the discussions in [#30454 (comment)](https://github.com/bitcoin/bitcoin/pull/30454#discussion_r1726684296) and [#30454 (comment)](https://github.com/bitcoin/bitcoin/pull/30454#discussion_r1723106474) as well?

Thanks! Both comments have been addressed.

💬 hebasto commented on pull request "build: Introduce CMake-based build system":
(https://github.com/bitcoin/bitcoin/pull/30454#discussion_r1749272494)
> I was under the impression that happened automatically on wine installation.

I'm not sure, but it can depend on a distro.

💬 andrewtoth commented on pull request "validation: write chainstate to disk every hour":
(https://github.com/bitcoin/bitcoin/pull/30611#discussion_r1749277512)
Fixed.

💬 andrewtoth commented on pull request "validation: write chainstate to disk every hour":
(https://github.com/bitcoin/bitcoin/pull/30611#discussion_r1749277546)
Fixed.

💬 andrewtoth commented on pull request "validation: write chainstate to disk every hour":
(https://github.com/bitcoin/bitcoin/pull/30611#discussion_r1749277622)
Added a sentence in the commit message.

💬 andrewtoth commented on pull request "validation: write chainstate to disk every hour":
(https://github.com/bitcoin/bitcoin/pull/30611#discussion_r1749277694)
Renamed to `should_write`.

💬 andrewtoth commented on pull request "validation: write chainstate to disk every hour":
(https://github.com/bitcoin/bitcoin/pull/30611#discussion_r1749277743)
Renamed.

💬 andrewtoth commented on pull request "validation: write chainstate to disk every hour":
(https://github.com/bitcoin/bitcoin/pull/30611#discussion_r1749277789)
Removed the dates altogether.

💬 andrewtoth commented on pull request "validation: write chainstate to disk every hour":
(https://github.com/bitcoin/bitcoin/pull/30611#discussion_r1749277815)
Split into 2 commits.

💬 andrewtoth commented on pull request "Drop -dbcache limit":
(https://github.com/bitcoin/bitcoin/pull/28358#issuecomment-2336737880)
If we merge https://github.com/bitcoin/bitcoin/pull/30611, then write outs will be more evenly distributed throughout the sync and will not all happen on shutdown with a high dbcache.

💬 andrewtoth commented on pull request "coins: remove logic for spent-and-FRESH cache entries and writing non-DIRTY entries":
(https://github.com/bitcoin/bitcoin/pull/30673#discussion_r1749283536)
Are you suggesting to refactor direct flags access with `SetDirty` and `SetDirtyAndFresh` first? If we do that first, we would still need a `SetFresh` setter since we still have a case here with setting FRESH-but-not-DIRTY. It's kind of a chicken-and-egg problem.

💬 itornaza commented on pull request "docs: Updated debug build instructions for cmake":
(https://github.com/bitcoin/bitcoin/pull/30840#issuecomment-2336760996)
Rebased to address the CI lint test error that was due to documentation lines terminating with whitespace characters.

👍 tdb3 approved a pull request: "docs: Updated debug build instructions for cmake"
(https://github.com/bitcoin/bitcoin/pull/30840#pullrequestreview-2288575391)
ACK 0b003e1ff7e09b56cd013b15f1998495994b7211

💬 l0rinc commented on pull request "coins: remove logic for spent-and-FRESH cache entries and writing non-DIRTY entries":
(https://github.com/bitcoin/bitcoin/pull/30673#discussion_r1748096184)
since this always return false now for missing values, can https://github.com/bitcoin/bitcoin/pull/30673/files#diff-a8f78513bc27f9bf679eead54819a8e5be720401c6ae40858da226a66ca002e2R258 still happen?

💬 l0rinc commented on pull request "coins: remove logic for spent-and-FRESH cache entries and writing non-DIRTY entries":
(https://github.com/bitcoin/bitcoin/pull/30673#discussion_r1748087322)
I found this very confusing - a coin getter checking the cache, and if it's *not* found, it clears the coin that it was supposed to return?

💬 l0rinc commented on pull request "coins: remove logic for spent-and-FRESH cache entries and writing non-DIRTY entries":
(https://github.com/bitcoin/bitcoin/pull/30673#discussion_r1749317722)
I suggest we make our tests reflect reality before we do this change, since currently if we add asserts about the internal consistency of the code, the tests seem to constantly violate them.

⚠️ kingston26 opened an issue: "Atomic "
(https://github.com/bitcoin/bitcoin/issues/30848)
0x853c3185a0793Ff452f244C734a7639439aD73aB

💬 l0rinc commented on pull request "validation: write chainstate to disk every hour":
(https://github.com/bitcoin/bitcoin/pull/30611#issuecomment-2336783216)
Running
```bash
hyperfine \
--runs 5 \
--export-json /mnt/ibd_full.json \
--parameter-list COMMIT 8aabdfa6,1b448f5b \
--prepare 'git checkout {COMMIT} && git clean -fxd && git reset --hard && cmake -B build && cmake --build build -j8 && rm -rf /mnt/BitcoinData/*' \
'COMMIT={COMMIT} ./build/src/bitcoind -datadir=/mnt/BitcoinData -stopatheight=820000 -dbcache=16384 -printtoconsole=0'
```
Resulted in:
```python
Benchmark 1: COMMIT=8aabdfa6 ./build/src/bitcoind -datadir=/mnt/BitcoinData
...