Skip to content

Commit

Permalink
Rollup merge of rust-lang#115476 - RalfJung:abi-compat-docs, r=Mark-S…
Browse files Browse the repository at this point in the history
…imulacrum

document ABI compatibility

I don't think we have any central place where we document our ABI compatibility rules, so let's create one. The `fn()` pointer type seems like a good place since ABI questions can only become relevant when invoking a function through a function pointer.

This will likely need T-lang FCP.
  • Loading branch information
matthiaskrgr authored Nov 17, 2023
2 parents 4770d91 + 8f03a55 commit 1cabedc
Show file tree
Hide file tree
Showing 3 changed files with 117 additions and 8 deletions.
7 changes: 4 additions & 3 deletions library/core/src/option.rs
Original file line number Diff line number Diff line change
Expand Up @@ -119,27 +119,28 @@
//! # Representation
//!
//! Rust guarantees to optimize the following types `T` such that
//! [`Option<T>`] has the same size and alignment as `T`. In some
//! [`Option<T>`] has the same size, alignment, and [function call ABI] as `T`. In some
//! of these cases, Rust further guarantees that
//! `transmute::<_, Option<T>>([0u8; size_of::<T>()])` is sound and
//! produces `Option::<T>::None`. These cases are identified by the
//! second column:
//!
//! | `T` | `transmute::<_, Option<T>>([0u8; size_of::<T>()])` sound? |
//! |---------------------------------------------------------------------|----------------------------------------------------------------------|
//! | [`Box<U>`] | when `U: Sized` |
//! | [`Box<U>`] (specifically, only `Box<U, Global>`) | when `U: Sized` |
//! | `&U` | when `U: Sized` |
//! | `&mut U` | when `U: Sized` |
//! | `fn`, `extern "C" fn`[^extern_fn] | always |
//! | [`num::NonZero*`] | always |
//! | [`ptr::NonNull<U>`] | when `U: Sized` |
//! | `#[repr(transparent)]` struct around one of the types in this list. | when it holds for the inner type |
//!
//! [^extern_fn]: this remains true for any other ABI: `extern "abi" fn` (_e.g._, `extern "system" fn`)
//! [^extern_fn]: this remains true for any argument/return types and any other ABI: `extern "abi" fn` (_e.g._, `extern "system" fn`)
//!
//! [`Box<U>`]: ../../std/boxed/struct.Box.html
//! [`num::NonZero*`]: crate::num
//! [`ptr::NonNull<U>`]: crate::ptr::NonNull
//! [function call ABI]: ../primitive.fn.html#abi-compatibility
//!
//! This is called the "null pointer optimization" or NPO.
//!
Expand Down
110 changes: 109 additions & 1 deletion library/core/src/primitive_docs.rs
Original file line number Diff line number Diff line change
Expand Up @@ -1493,7 +1493,7 @@ mod prim_ref {}
///
/// ### Casting to and from integers
///
/// You cast function pointers directly to integers:
/// You can cast function pointers directly to integers:
///
/// ```rust
/// let fnptr: fn(i32) -> i32 = |x| x+2;
Expand All @@ -1519,6 +1519,114 @@ mod prim_ref {}
/// Note that all of this is not portable to platforms where function pointers and data pointers
/// have different sizes.
///
/// ### ABI compatibility
///
/// Generally, when a function is declared with one signature and called via a function pointer with
/// a different signature, the two signatures must be *ABI-compatible* or else calling the function
/// via that function pointer is Undefined Behavior. ABI compatibility is a lot stricter than merely
/// having the same memory layout; for example, even if `i32` and `f32` have the same size and
/// alignment, they might be passed in different registers and hence not be ABI-compatible.
///
/// ABI compatibility as a concern only arises in code that alters the type of function pointers,
/// code that imports functions via `extern` blocks, and in code that combines `#[target_feature]`
/// with `extern fn`. Altering the type of function pointers is wildly unsafe (as in, a lot more
/// unsafe than even [`transmute_copy`][mem::transmute_copy]), and should only occur in the most
/// exceptional circumstances. Most Rust code just imports functions via `use`. `#[target_feature]`
/// is also used rarely. So, most likely you do not have to worry about ABI compatibility.
///
/// But assuming such circumstances, what are the rules? For this section, we are only considering
/// the ABI of direct Rust-to-Rust calls, not linking in general -- once functions are imported via
/// `extern` blocks, there are more things to consider that we do not go into here.
///
/// For two signatures to be considered *ABI-compatible*, they must use a compatible ABI string,
/// must take the same number of arguments, the individual argument types and the return types must
/// be ABI-compatible, and the target feature requirements must be met (see the subsection below for
/// the last point). The ABI string is declared via `extern "ABI" fn(...) -> ...`; note that
/// `fn name(...) -> ...` implicitly uses the `"Rust"` ABI string and `extern fn name(...) -> ...`
/// implicitly uses the `"C"` ABI string.
///
/// The ABI strings are guaranteed to be compatible if they are the same, or if the caller ABI
/// string is `$X-unwind` and the callee ABI string is `$X`, where `$X` is one of the following:
/// "C", "aapcs", "fastcall", "stdcall", "system", "sysv64", "thiscall", "vectorcall", "win64".
///
/// The following types are guaranteed to be ABI-compatible:
///
/// - `*const T`, `*mut T`, `&T`, `&mut T`, `Box<T>` (specifically, only `Box<T, Global>`), and
/// `NonNull<T>` are all ABI-compatible with each other for all `T`. They are also ABI-compatible
/// with each other for _different_ `T` if they have the same metadata type (`<T as
/// Pointee>::Metadata`).
/// - `usize` is ABI-compatible with the `uN` integer type of the same size, and likewise `isize` is
/// ABI-compatible with the `iN` integer type of the same size.
/// - Any two `fn` (function pointer) types are ABI-compatible with each other if they have the same
/// ABI string or the ABI string only differs in a trailing `-unwind`, independent of the rest of
/// their signature. (This means you can pass `fn()` to a function expecting `fn(i32)`, and the
/// call will be valid ABI-wise. The callee receives the result of transmuting the function pointer
/// from `fn()` to `fn(i32)`; that transmutation is itself a well-defined operation, it's just
/// almost certainly UB to later call that function pointer.)
/// - Any two types with size 0 and alignment 1 are ABI-compatible.
/// - A `repr(transparent)` type `T` is ABI-compatible with its unique non-trivial field, i.e., the
/// unique field that doesn't have size 0 and alignment 1 (if there is such a field).
/// - `i32` is ABI-compatible with `NonZeroI32`, and similar for all other integer types with their
/// matching `NonZero*` type.
/// - If `T` is guaranteed to be subject to the [null pointer
/// optimization](option/index.html#representation), then `T` and `Option<T>` are ABI-compatible.
///
/// Furthermore, ABI compatibility satisfies the following general properties:
///
/// - Every type is ABI-compatible with itself.
/// - If `T1` and `T2` are ABI-compatible, then two `repr(C)` types that only differ because one
/// field type was changed from `T1` to `T2` are ABI-compatible.
/// - If `T1` and `T2` are ABI-compatible and `T2` and `T3` are ABI-compatible, then so are `T1` and
/// `T3` (i.e., ABI-compatibility is transitive).
/// - If `T1` and `T2` are ABI-compatible, then so are `T2` and `T1` (i.e., ABI-compatibility is
/// symmetric).
///
/// More signatures can be ABI-compatible on specific targets, but that should not be relied upon
/// since it is not portable and not a stable guarantee.
///
/// Noteworthy cases of types *not* being ABI-compatible in general are:
/// * `bool` vs `u8`, and `i32` vs `u32`: on some targets, the calling conventions for these types
/// differ in terms of what they guarantee for the remaining bits in the register that are not
/// used by the value.
/// * `i32` vs `f32` are not compatible either, as has already been mentioned above.
/// * `struct Foo(u32)` and `u32` are not compatible (without `repr(transparent)`) since structs are
/// aggregate types and often passed in a different way than primitives like `i32`.
///
/// Note that these rules describe when two completely known types are ABI-compatible. When
/// considering ABI compatibility of a type declared in another crate (including the standard
/// library), consider that any type that has a private field or the `#[non_exhaustive]` attribute
/// may change its layout as a non-breaking update unless documented otherwise -- so for instance,
/// even if such a type is a 1-ZST or `repr(transparent)` right now, this might change with any
/// library version bump.
///
/// If the declared signature and the signature of the function pointer are ABI-compatible, then the
/// function call behaves as if every argument was [`transmute`d][mem::transmute] from the
/// type in the function pointer to the type at the function declaration, and the return value is
/// [`transmute`d][mem::transmute] from the type in the declaration to the type in the
/// pointer. All the usual caveats and concerns around transmutation apply; for instance, if the
/// function expects a `NonNullI32` and the function pointer uses the ABI-compatible type
/// `Option<NonNullI32>`, and the value used for the argument is `None`, then this call is Undefined
/// Behavior since transmuting `None::<NonNullI32>` to `NonNullI32` violates the non-null
/// requirement.
///
/// #### Requirements concerning target features
///
/// Under some conditions, the signature used by the caller and the callee can be ABI-incompatible
/// even if the exact same ABI string and types are being used. As an example, the
/// `std::arch::x86_64::__m256` type has a different `extern "C"` ABI when the `avx` feature is
/// enabled vs when it is not enabled.
///
/// Therefore, to ensure ABI compatibility when code using different target features is combined
/// (such as via `#[target_feature]`), we further require that one of the following conditions is
/// met:
///
/// - The function uses the `"Rust"` ABI string (which is the default without `extern`).
/// - Caller and callee are using the exact same set of target features. For the callee we consider
/// the features enabled (via `#[target_feature]` and `-C target-feature`/`-C target-cpu`) at the
/// declaration site; for the caller we consider the features enabled at the call site.
/// - Neither any argument nor the return value involves a SIMD type (`#[repr(simd)]`) that is not
/// behind a pointer indirection (i.e., `*mut __m256` is fine, but `(i32, __m256)` is not).
///
/// ### Trait implementations
///
/// In this documentation the shorthand `fn (T₁, T₂, …, Tₙ)` is used to represent non-variadic
Expand Down
8 changes: 4 additions & 4 deletions tests/ui/abi/compatibility.rs
Original file line number Diff line number Diff line change
Expand Up @@ -231,8 +231,7 @@ macro_rules! test_abi_compatible {
};
}

// Compatibility of pointers is probably de-facto guaranteed,
// but that does not seem to be documented.
// Compatibility of pointers.
test_abi_compatible!(ptr_mut, *const i32, *mut i32);
test_abi_compatible!(ptr_pointee, *const i32, *const Vec<i32>);
test_abi_compatible!(ref_mut, &i32, &mut i32);
Expand All @@ -241,14 +240,15 @@ test_abi_compatible!(box_ptr, Box<i32>, *const i32);
test_abi_compatible!(nonnull_ptr, NonNull<i32>, *const i32);
test_abi_compatible!(fn_fn, fn(), fn(i32) -> i32);

// Some further guarantees we will likely (have to) make.
// Compatibility of 1-ZST.
test_abi_compatible!(zst_unit, Zst, ());
#[cfg(not(any(target_arch = "sparc64")))]
test_abi_compatible!(zst_array, Zst, [u8; 0]);
test_abi_compatible!(nonzero_int, NonZeroI32, i32);

// `DispatchFromDyn` relies on ABI compatibility.
// This is interesting since these types are not `repr(transparent)`.
// This is interesting since these types are not `repr(transparent)`. So this is not part of our
// public ABI guarantees, but is relied on by the compiler.
test_abi_compatible!(rc, Rc<i32>, *mut i32);
test_abi_compatible!(arc, Arc<i32>, *mut i32);

Expand Down

0 comments on commit 1cabedc

Please sign in to comment.