Rémy Rakic

@willcrichton The `.rustfmt.toml` change and new `log` file, seem unintentional.

Notable people to cc: - @nagisa and @nikic, for their linker & LLVM expertise - @1000teslas and @hkratz, since they've worked on/introduced the `-Zgcc-ld` flag - @bstrie and @gankra, since...

@bjorn3 it's still a meeting proposal, zulip threads are created only for MCPs I think ?

I just cloned machete as of 3 days ago, it was not cargo installed. The error occurred on windows and ubuntu under WSL, so I suspect it's some windows-related weirdness.

Another interesting datapoint in the same spirit, would be to also track the state of rust perf between releases in addition to nightlies. It would be interesting to show possible...

Which reminds me: you mentioned `for await in it` [here](https://github.com/estebank/iterator_item/blob/e125b300d50cbe0e74a2de8d3bc732162433a10b/iterator_item_macros/src/lib.rs#L196-L197) but IIUC from this test above, `syn` won't be able to parse that and visit that node at this location...

The max-rss hit looks noticeable, is that expected ?

keccak has been kinda noisy recently, right ?  The significance thresholding didn't find this [5% regression](https://perf.rust-lang.org/compare.html?start=211637d0802a1c17d41b414e091e9a8691b26068&end=8f68c43ca6a6381a4d73f887f112e9fb95769905&stat=instructions:u) relevant either.

Potentially interesting: having them in integration tests would mean we wouldn't have to update the test code when migrating away from `#[rustc_args_required_const]` as described in #1022. They'd be in another...

The first task to update the help page will be done by https://github.com/rust-lang/rustc-perf/pull/1434