Max Inden

> There are a bunch of `TODO`s (beyond those I marked) that probably need addressing. Sorry for the noise. Before solving some of the trickier TODOs, I wanted to make...

### Naming > We call the individual payloads "segments", which as a TCP person I find confusing. While it can be easily confused with the TCP usage of the same...

This pull request is ready for a full review. All TODOs are addressed.

Thank you for the reviews @larseggert and @martinthomson. > It seems to me like you could do this in pieces, starting with the encoder/decoder changes. Sounds good. I will address...

**Receive path** optimization has been merged through https://github.com/mozilla/neqo/pull/2184. While this pull request can serve as guidance for a future **send path** patch, I assume it is better to start from...

https://github.com/mozilla/neqo/pull/2636 disables the test on ARM. https://github.com/mozilla/neqo/actions/runs/15073341987/job/42374916088?pr=2639 is a recent failure on x86-64.

Reopening again, given recent failure: https://github.com/mozilla/neqo/actions/runs/15418677161/job/43387638309?pr=2691

> I've checked the usage in our server code, which is fine because attackers don't get to control memory allocations (we use pointer values for the hash). Still, that makes...

> a **lot** of variation run-to-run for some reason That is counterintuitive for me, given that it uses `test-fixtures` and thus does no IO via the OS. Let me know...

For what it is worth, here is #2444 on my machine: ``` ➜ neqo-http3 git:(test-streams-bench) ✗ cargo bench --features bench Benchmarking 1 streams of 1 bytes/multistream: Collecting 100 samples in...