💬 olegrok commented on issue "ci: don't pass GIT_EXEC_PATH inside test container":
(https://github.com/bitcoin/bitcoin/issues/32935#issuecomment-3057060249)
> Did you use?

No. I hoped that everything will works fine out-of-box. So here I just highlight possible issue. It was really hard to debug it.

> I don't think a list with excludes is maintainable.

That's true. So feel free to close this issue if you don't think anything actually can't be improved here.

💬 maflcko commented on pull request "rpc, test: Fix JSON parsing errors in unloadwallet and getdescriptoractivity RPCs":
(https://github.com/bitcoin/bitcoin/pull/32845#discussion_r2197460325)
Yeah, thanks @stickies-v. Sorry for missing this. Obviously commit 529d2ce24570aae426ce19e87cb22f124dd09470 is not quite right. It should either be dropped, because there isn't really any problem, or the already existing RPCArg::Optional logic should be used.

👍 stickies-v approved a pull request: "log: Properly log warnings with warn loglevel in addrdb"
(https://github.com/bitcoin/bitcoin/pull/32933#pullrequestreview-3005268716)
ACK fa894b0f3e13dcc55fd42cec2c56d4aa2115194d

⚠️ fanquake opened an issue: "cmake: searching across directories for config files"
(https://github.com/bitcoin/bitcoin/issues/32938)
This can be recreated on at least macOS, due to #32707.

If there are two checkouts of the source, the build in one directory, can fail to configure, because of files in the other directory:
```bash
git clone https://github.com/bitcoin/bitcoin/ bitcoin_a
git clone https://github.com/bitcoin/bitcoin/ bitcoin_b

cd bitcoin_a
make -C depends HOST=x86_64-w64-mingw32
# fails to build libevent (#32707)
cd ../bitcoin_b
# fails to configure, because of the libevent config files (for a different target)
...

💬 maflcko commented on issue "ci: don't pass GIT_EXEC_PATH inside test container":
(https://github.com/bitcoin/bitcoin/issues/32935#issuecomment-3057091584)
I think the possible solutions here are:

* Keep everything as-is and just let people figure out that they need to follow the docs exactly.
* Rename all env vars used internally in the ci to have some prefix like `CI_...` and then filter for those and only pass them. (easy-ish change, but still a breaking change)
* Wrap the ci invocation into an arg-parser (larger re-write, also a breaking change)
* something else?

💬 ryanofsky commented on pull request "ipc: add bitcoin-mine test program":
(https://github.com/bitcoin/bitcoin/pull/30437#issuecomment-3057186819)
Updated 1a87ed000e09b10c173ba87d4bf2f79bf63abfa5 -> 2910315d1d8a3879f4e0bd50bf90870b642bfede ([`pr/mine.26`](https://github.com/ryanofsky/bitcoin/commits/pr/mine.26) -> [`pr/mine.27`](https://github.com/ryanofsky/bitcoin/commits/pr/mine.27), [compare](https://github.com/ryanofsky/bitcoin/compare/pr/mine.26..pr/mine.27)) fixing unused python import lint error https://cirrus-ci.com/task/5421333066022912

💬 Sjors commented on pull request "ipc: add bitcoin-mine test program":
(https://github.com/bitcoin/bitcoin/pull/30437#issuecomment-3057188134)
ACK 2910315d1d8a3879f4e0bd50bf90870b642bfede

💬 l0rinc commented on pull request "[IBD] multi-byte block obfuscation":
(https://github.com/bitcoin/bitcoin/pull/31144#issuecomment-3057258422)
[Latest push](https://github.com/bitcoin/bitcoin/compare/e036ba4ddaeb24a2361c356f7f270ee60bfff984..64fbca986cbaf3990d15d26bf6336fa8116f1d56) contains (as usual, the second push is just a changeless rebase):
* `Obfuscation` constructor only accepts fixed-size span now - and is explicit
* fuzz test changes were moved earlier to fix per-commit tests and we're using `ConsumeFixedLengthByteVector` now instead of either a uint64_t or random length vector
* `Obfuscation::SIZE_BYTES` was originally n
...

🚀 fanquake merged a pull request: "bench: Avoid tmp files in pwd"
(https://github.com/bitcoin/bitcoin/pull/32890)

👍 Sjors approved a pull request: "ipc: Handle unclean shutdowns better"
(https://github.com/bitcoin/bitcoin/pull/32345#pullrequestreview-3005455760)
utACK 0a9a1940782fad8dc3494c4aacee0a06c030eb5c

💬 Sjors commented on pull request "ipc: Handle unclean shutdowns better":
(https://github.com/bitcoin/bitcoin/pull/32345#discussion_r2197581718)
In f8fd3959d51f6f80b428828c4fd965e06a8fa19e _ipc: Use EventLoopRef instead of addClient/removeClient_: if you have to retouch it would be good to document `m_loop` and `m_loop_ref`.

📝 l0rinc opened a pull request: "refactor: avoid double hashing in `SourceLocationHasher`"
(https://github.com/bitcoin/bitcoin/pull/32939)
Feeds the file name string directly into the hasher instead of pre-hashing it first. The resulting hash changes of course, but that doesn't affect us since it's only stored in-memory.

Fixes: https://github.com/bitcoin/bitcoin/pull/32604#discussion_r2188644325

A few other related nits were also addressed here:
* Comment was redundant, `CSipHasher` is used in other hashers all over the codebase
* `static_cast<size_t>` skewed the whole line, changed to shorter functional-style cast
* struc
...

💬 stickies-v commented on pull request "rpc, test: Fix JSON parsing errors in unloadwallet and getdescriptoractivity RPCs":
(https://github.com/bitcoin/bitcoin/pull/32845#discussion_r2197642119)
> "descriptors" documentation could be improved, just quickly wrote something down. E.g. not sure if order matters here, or if there's anything else to look out for.

We should probably just leave the name as "scanobjects" (instead of my change to "descriptors"), to avoid introducing backwards incompatible API changes. The description should be change-able without issue.

💬 l0rinc commented on pull request "log: Mitigate disk filling attacks by rate limiting LogPrintf, LogInfo, LogWarning, LogError":
(https://github.com/bitcoin/bitcoin/pull/32604#discussion_r2197642575)
Fixed it in https://github.com/bitcoin/bitcoin/pull/32939

🤔 stickies-v reviewed a pull request: "refactor: avoid double hashing in `SourceLocationHasher`"
(https://github.com/bitcoin/bitcoin/pull/32939#pullrequestreview-3005632806)
Since there are quite a few follow-ups in #32604, it might make sense to do them all in one (or at least larger, related PRs) instead of carving it out into dozens of small ones?

💬 stickies-v commented on pull request "refactor: avoid double hashing in `SourceLocationHasher`":
(https://github.com/bitcoin/bitcoin/pull/32939#discussion_r2197694485)
As we've already discussed [here](https://github.com/bitcoin/bitcoin/pull/32604#discussion_r2163614334), I don't think it would be an improvement to remove this comment. Using `CSipHasher` over `std::hash` is non-trivial and helped me understand the code.

💬 hebasto commented on issue "cmake: searching across directories for config files":
(https://github.com/bitcoin/bitcoin/issues/32938#issuecomment-3057410237)
Could you please confirm whether the following patch helps:
```diff
--- a/depends/patches/libevent/cmake_fixups.patch
+++ b/depends/patches/libevent/cmake_fixups.patch
@@ -7,7 +7,7 @@ cmake: set minimum version to 3.5
#

-cmake_minimum_required(VERSION 3.1 FATAL_ERROR)
-+cmake_minimum_required(VERSION 3.5 FATAL_ERROR)
++cmake_minimum_required(VERSION 3.15 FATAL_ERROR)

if (POLICY CMP0054)
cmake_policy(SET CMP0054 NEW)
```
?

🤔 marcofleon reviewed a pull request: "p2p: improve TxOrphanage denial of service bounds"
(https://github.com/bitcoin/bitcoin/pull/31829#pullrequestreview-3005543794)
The `protected_count == 0` assertion failed in `txorphan_protected`.
<details>
<summary>crash</summary>

```bash
../../../../src/test/fuzz/txorphan.cpp:395 operator(): Assertion `protected_count == 0' failed.
==2872074== ERROR: libFuzzer: deadly signal
#0 0x558b2f142c41 in __sanitizer_print_stack_trace (/root/bitcoin/txorphanfuzzbuild/bin/fuzz+0x1c92c41) (BuildId: 2ed4a047818a3d1ea39c53799e090813bde04dba)
#1 0x558b2f097c08 in fuzzer::PrintStackTrace() (/root/bitcoin/txorphanfuzzb
...

💬 marcofleon commented on pull request "p2p: improve TxOrphanage denial of service bounds":
(https://github.com/bitcoin/bitcoin/pull/31829#discussion_r2197640228)
What's the point of `std::floor` here? Also I noticed this doesn't match the calculation in `GetLatencyScore()`. Is it meant to be an overestimate?

💬 marcofleon commented on pull request "p2p: improve TxOrphanage denial of service bounds":
(https://github.com/bitcoin/bitcoin/pull/31829#discussion_r2197660618)
We increment by the latency score above but then decrement here by announcement count. These aren't guaranteed to be the same afaict, so this might be what's causing the failure of the `protected_count == 0` assertion.