Commit Graph
4 Commits
Author SHA1 Message Date
aa69a484eb Add support for running LLVM optimizer. (#6225)
Adds a flag `--optimize=<mode>` that specifies what to optimize for:

* `--optimize=none` turns off the optimizer as much as possible, but
still respects always_inline.
* `--optimize=debug` aims to be the equivalent of `-Og` / `-O1`, and
provides optimizations that don't affect the ability to debug the
program. This is the default.
* `--optimize=size` optimizes for the size of the produced program, and
aims to be the equivalent of `-Oz`.
* `--optimize=speed` optimizes for the execution time of the produced
program, and aims to be the equivalent of `-O3`.

Following the approach taken by Clang, the optimization level feeds into
both the configuration of the LLVM pass pipeline and the attributes
added to function definitions generated by the frontend.

Optimization is performed in a new phase, `optimize`, which runs between
`lower` and `codegen`.

---------

Co-authored-by: Dana Jansens <danakj@orodu.net>
Co-authored-by: Chandler Carruth <chandlerc@gmail.com>
2025-11-05 00:15:14 +00:00
Richard Smith a1a35c207e Unify "needs thunk" logic. (#6277)
Remove duplication between determining whether a parameter needs custom
thunk mapping and whether a function needs a thunk. Now a function needs
a thunk if any parameter or the return type does.

This fixes some inconsistencies; previously:
- We would not require a thunk when passing an `unsigned int`, but if we
  had a thunk we'd pass `unsigned int` indirectly.
- We would always require a thunk for an enum parameter, even though
  we'd actually pass it directly if its underlying type is a 32- or
  64-bit integer.
- We would require a thunk for a nullable pointer, even though
  we arrange for all pointer types to have the same ABI in Carbon and
  C++, including nullable pointers / Optional(T*).

This also causes us to use a thunk for rvalue reference return types,
which we used to miscompile.

Depends on #6276.
2025-10-27 22:49:29 +00:00
Richard SmithandJon Ross-Perkins 1e7b7e53ae C++ interop: support for default arguments. (#6108)
The general strategy here is to force use of a thunk when we want to use
default arguments, and have Clang generate uses of the default arguments
on its side of the thunk.

To support this, change the key type used in `clang_decls` from being
just a `Decl*` to being a pair of `Decl*` and number of parameters in
the case of function decls. Import distinct `SemIR::Function`s for each
number of parameters that's used, and corresponding distinct thunks.

---------

Co-authored-by: Jon Ross-Perkins <jperkins@google.com>
2025-09-24 01:07:59 +00:00
Richard Smith d37f1ae6b5 Add return value support to C++ thunks. (#5976)
Based on #5948. A couple of tricky parts:

* When generating the C++ side of the thunk, we are given a pointer to
the location to emplace the return value. The only mechanism C++
provides to perform this emplacement is using placement `operator new`,
which requires a library function in the `<new>` header. We handle this
by declaring that library function ourselves, and rely on Clang not
actually needing a definition for it (which the standard library owns).

* On the Carbon side of the thunk, we want to form an initializing
expression as the result of the call. We don't have a way of expressing
in SemIR that an initializing expression performs its initialization by
storing through a pointer, so this PR adds a new initializing
instruction, `InPlaceInit`, to model an initialization that's performed
opaquely in-place.
2025-08-26 02:23:38 +00:00