💬 andrewtoth commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485086567)
Done.

💬 andrewtoth commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485086761)
I couldn't get `clang-format` to work, but I made this a one-liner and moved under `GetCacheSize`.

💬 andrewtoth commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485086827)
Done.

💬 andrewtoth commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485087021)
Did it anyways.

💬 andrewtoth commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485087120)
Done.

💬 andrewtoth commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485087167)
Done.

💬 andrewtoth commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485087197)
Done.

💬 andrewtoth commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485087251)
Removed it.

💬 andrewtoth commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485087341)
Done.

💬 andrewtoth commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485087547)
Done.

💬 andrewtoth commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485087694)
Fixed. But, it wouldn't affect the correctness of the test.

💬 andrewtoth commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485087978)
Rewrote this part to make it more clear. I was trying to be clever by avoiding some extra checks, but they are probably meaningless and clarity is better here.

💬 andrewtoth commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485157101)
I wrote benchmarks to check how fast it was to construct, since this is the work that is done not in parallel:

```C++
static void SortedVectorBenchmark(benchmark::Bench& bench)
{
CBlock block;
DataStream{benchmark::data::block413567} >> TX_WITH_WITNESS(block);
std::vector<Txid> v{};
v.reserve(block.vtx.size());

bench.run([&] {
for (const auto& tx : block.vtx) {
v.emplace_back(tx->GetHash());
}
std::sort(v.begin(), v.end());
...

🤔 danielabrozzoni reviewed a pull request: "doc: document fingerprinting risk when operating node on multiple networks"
(https://github.com/bitcoin/bitcoin/pull/33750#pullrequestreview-3409743891)
ACK e346ecae830e10310979e5f64de63e043a383ff1

I really like how it's written, it's really fair and balanced, without sounding scary. Thanks for working on this! :)

Just a small nit, you included the suggestions from https://github.com/bitcoin/bitcoin/pull/33750#discussion_r2481496048 and https://github.com/bitcoin/bitcoin/pull/33750#discussion_r2482405261 in the second commit (e346ecae830e10310979e5f64de63e043a383ff1). This makes the first commit modify only tor.md, while the second one cha
...

💬 l0rinc commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485569647)
I don't think we should abort, it's not the fetcher's job to validate. If we didn't, we could even use short ids for the intra-block spends (64 bit likely won't even result in a single duplicate, and when they do xollide we would just do a db check)

💬 l0rinc commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485572792)
atomics work via CAS, they need to retry in case of high contention. the previous solution didn't have contention, the threads all knew beforehand what to work on.

💬 l0rinc commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485575807)
that's my point, I don't think we should validate at all, it would simplify the code if we didn't

✅ maflcko closed an issue: "Decouple datacarrier size and count limits (Draft PR)"
(https://github.com/bitcoin/bitcoin/issues/33595)

💬 maflcko commented on issue "Decouple datacarrier size and count limits (Draft PR)":
(https://github.com/bitcoin/bitcoin/issues/33595#issuecomment-3479289768)
The feature request didn't seem to attract much positive attention in the past. Thus, closing due to lack of interest, progress and direction. It is possible to re-open this issue or create a new issue referencing it, if there is fresh interest.

💬 l0rinc commented on pull request "validation: fetch block inputs on parallel threads >20% faster IBD":
(https://github.com/bitcoin/bitcoin/pull/31132#discussion_r2485579371)
how so? Can we assert the behavior of the main cache as well so that the previous version doesn't pass?