Robin Freyler

Partially this is indeed done in the `for_each_operator!` macro, however, this issue was meant more widespread as a means of putting documentation hints onto any Rust definition that are specific...

Having https://github.com/paritytech/wasmi/issues/924 implemented makes this optimization simpler to implement since we no longer have to differentiate between `i32.add r c` and `i32.sub r c` variants of the above instruction sequences.

I appreciate the overall direction. I am the maintainer of the Wasmi crate and am using `BTreeMap` instead of `HashMap` pervasively there. This is also true for its dependencies such...

> BTreeMap has a non-trivial impact on compile time performance and it too can have worse than average performance under adversarial circumstances (inssrting elements in ascending order is supposedly slower...

> > Are you referring to this forum post from 2 days ago? > > Yes > > > Some users run Wasmi within Wasm (wasm32-unknown-unknown) executed via Wasmtime. >...

> I was talking with @tschneidereit yesterday about this issue and one point he brought up is that it would be reasonable to have a compile-time Cargo feature on the...

> My thinking was that the type aliases in wasmparser would become type definitions which internally #[cfg] for hash maps or tree-based maps. For the tree-based-indexmap-version I was thinking we'd...

> @Robbepop sounds good! And also to make sure I clarify I don't plan on personally pursuing the BTreeMap approach of wasmparser, but I wanted to outline how I think...

The Wasm `tail-calls` proposal has just been voted to move to Phase 4 (WG Standardization) in the CG meeting today. (https://github.com/WebAssembly/proposals/pull/157) Since Wasm is one of the most important targets...

Given the [long history of this feature request](https://github.com/rust-lang/rust/issues/102952#issuecomment-1455096639) my feeling is that the next step is to find a few people: namely, people from the lang and compiler team to...