Skip to content

Conversation

svix-jplatte
Copy link
Contributor

Kind of a follow-up to #3285.

The last commit in particular improves rebuild times of our internal codebase by about 5%. Seems like the compiler has an easier time due to one layer of genericism being sidestepped. I did this because I saw tower::util::MapResponse show up a bunch in a compiler self-profile trace.

Route::new applies a `map_response(into_response)` already, so the
service being given to it doesn't have to involve an into_response call
itself.
@jplatte jplatte merged commit a9638f7 into tokio-rs:main Mar 28, 2025
18 checks passed
@svix-jplatte svix-jplatte deleted the cleanup branch March 28, 2025 10:07
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants