# LLVM: implement call_async_alloc · gitcafe/zig

[View on GitCafe](https://git.cafe/gitcafe/zig/commit/f1b7c76ae9c015eeadeeff6ffcdf1523afce43f5)

Repository: [gitcafe/zig](https://git.cafe/gitcafe/zig)

Visibility: public

Requested revision: f1b7c76ae9c015eeadeeff6ffcdf1523afce43f5

Requested commit: f1b7c76ae9c015eeadeeff6ffcdf1523afce43f5

Commit: f1b7c76ae9c015eeadeeff6ffcdf1523afce43f5

Tree: 9f18b9d6edfaf3f550236f3503b6720736888fbd

Author: Andrew Kelley

Committer: Andrew Kelley

## Message

```
LLVM: implement call_async_alloc

It's based on a runtime alloca based on the frame size. This allows us
to lower async function calls before lowering the callee, and allows
updating callees without updating the callers.

Sema: add `std.debug.Trace` integration for `unneeded`. This helps us
debug when `LazySrcLoc.unneeded` was incorrectly used.

```

## Parents

- [a6eb83bd1b7739156ab01e3fc55588f67d924daa](https://git.cafe/gitcafe/zig/commit/a6eb83bd1b7739156ab01e3fc55588f67d924daa?format=markdown)

[Source at this commit](https://git.cafe/gitcafe/zig/tree/f1b7c76ae9c015eeadeeff6ffcdf1523afce43f5?format=markdown)