Compute transmutability from `rustc_target::abi::Layout`
In its first step of computing transmutability, `rustc_transmutability`
constructs a byte-level representation of type layout (`Tree`). Previously, this
representation was computed for ADTs by inspecting the ADT definition and
performing our own layout computations. This process was error-prone, verbose,
and limited our ability to analyze many types (particularly default-repr types).
In this PR, we instead construct `Tree`s from `rustc_target::abi::Layout`s. This
helps ensure that layout optimizations are reflected our analyses, and increases
the kinds of types we can now analyze, including:
- default repr ADTs
- transparent unions
- `UnsafeCell`-containing types
Overall, this PR expands the expressvity of `rustc_transmutability` to be much
closer to the transmutability analysis performed by miri. Future PRs will work
to close the remaining gaps (e.g., support for `Box`, raw pointers, `NonZero*`,
coroutines, etc.).
2024-03-19 14:49:13 +00:00
|
|
|
//@ check-pass
|
2021-07-03 16:18:13 +00:00
|
|
|
//! A struct must have a well-defined layout to participate in a transmutation.
|
|
|
|
|
|
|
|
#![crate_type = "lib"]
|
|
|
|
#![feature(transmutability)]
|
|
|
|
#![allow(dead_code, incomplete_features, non_camel_case_types)]
|
|
|
|
|
|
|
|
mod assert {
|
2024-08-27 14:05:54 +00:00
|
|
|
use std::mem::{Assume, TransmuteFrom};
|
2021-07-03 16:18:13 +00:00
|
|
|
|
|
|
|
pub fn is_maybe_transmutable<Src, Dst>()
|
|
|
|
where
|
2024-08-27 14:05:54 +00:00
|
|
|
Dst: TransmuteFrom<Src, {
|
2022-08-18 19:39:14 +00:00
|
|
|
Assume {
|
|
|
|
alignment: true,
|
|
|
|
lifetimes: true,
|
|
|
|
safety: true,
|
|
|
|
validity: true,
|
|
|
|
}
|
|
|
|
}>
|
2021-07-03 16:18:13 +00:00
|
|
|
{}
|
|
|
|
}
|
|
|
|
|
Compute transmutability from `rustc_target::abi::Layout`
In its first step of computing transmutability, `rustc_transmutability`
constructs a byte-level representation of type layout (`Tree`). Previously, this
representation was computed for ADTs by inspecting the ADT definition and
performing our own layout computations. This process was error-prone, verbose,
and limited our ability to analyze many types (particularly default-repr types).
In this PR, we instead construct `Tree`s from `rustc_target::abi::Layout`s. This
helps ensure that layout optimizations are reflected our analyses, and increases
the kinds of types we can now analyze, including:
- default repr ADTs
- transparent unions
- `UnsafeCell`-containing types
Overall, this PR expands the expressvity of `rustc_transmutability` to be much
closer to the transmutability analysis performed by miri. Future PRs will work
to close the remaining gaps (e.g., support for `Box`, raw pointers, `NonZero*`,
coroutines, etc.).
2024-03-19 14:49:13 +00:00
|
|
|
fn should_accept_repr_rust()
|
2021-07-03 16:18:13 +00:00
|
|
|
{
|
|
|
|
fn unit() {
|
|
|
|
struct repr_rust;
|
Compute transmutability from `rustc_target::abi::Layout`
In its first step of computing transmutability, `rustc_transmutability`
constructs a byte-level representation of type layout (`Tree`). Previously, this
representation was computed for ADTs by inspecting the ADT definition and
performing our own layout computations. This process was error-prone, verbose,
and limited our ability to analyze many types (particularly default-repr types).
In this PR, we instead construct `Tree`s from `rustc_target::abi::Layout`s. This
helps ensure that layout optimizations are reflected our analyses, and increases
the kinds of types we can now analyze, including:
- default repr ADTs
- transparent unions
- `UnsafeCell`-containing types
Overall, this PR expands the expressvity of `rustc_transmutability` to be much
closer to the transmutability analysis performed by miri. Future PRs will work
to close the remaining gaps (e.g., support for `Box`, raw pointers, `NonZero*`,
coroutines, etc.).
2024-03-19 14:49:13 +00:00
|
|
|
assert::is_maybe_transmutable::<repr_rust, ()>();
|
|
|
|
assert::is_maybe_transmutable::<u128, repr_rust>();
|
2021-07-03 16:18:13 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
fn tuple() {
|
|
|
|
struct repr_rust();
|
Compute transmutability from `rustc_target::abi::Layout`
In its first step of computing transmutability, `rustc_transmutability`
constructs a byte-level representation of type layout (`Tree`). Previously, this
representation was computed for ADTs by inspecting the ADT definition and
performing our own layout computations. This process was error-prone, verbose,
and limited our ability to analyze many types (particularly default-repr types).
In this PR, we instead construct `Tree`s from `rustc_target::abi::Layout`s. This
helps ensure that layout optimizations are reflected our analyses, and increases
the kinds of types we can now analyze, including:
- default repr ADTs
- transparent unions
- `UnsafeCell`-containing types
Overall, this PR expands the expressvity of `rustc_transmutability` to be much
closer to the transmutability analysis performed by miri. Future PRs will work
to close the remaining gaps (e.g., support for `Box`, raw pointers, `NonZero*`,
coroutines, etc.).
2024-03-19 14:49:13 +00:00
|
|
|
assert::is_maybe_transmutable::<repr_rust, ()>();
|
|
|
|
assert::is_maybe_transmutable::<u128, repr_rust>();
|
2021-07-03 16:18:13 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
fn braces() {
|
|
|
|
struct repr_rust{}
|
Compute transmutability from `rustc_target::abi::Layout`
In its first step of computing transmutability, `rustc_transmutability`
constructs a byte-level representation of type layout (`Tree`). Previously, this
representation was computed for ADTs by inspecting the ADT definition and
performing our own layout computations. This process was error-prone, verbose,
and limited our ability to analyze many types (particularly default-repr types).
In this PR, we instead construct `Tree`s from `rustc_target::abi::Layout`s. This
helps ensure that layout optimizations are reflected our analyses, and increases
the kinds of types we can now analyze, including:
- default repr ADTs
- transparent unions
- `UnsafeCell`-containing types
Overall, this PR expands the expressvity of `rustc_transmutability` to be much
closer to the transmutability analysis performed by miri. Future PRs will work
to close the remaining gaps (e.g., support for `Box`, raw pointers, `NonZero*`,
coroutines, etc.).
2024-03-19 14:49:13 +00:00
|
|
|
assert::is_maybe_transmutable::<repr_rust, ()>();
|
|
|
|
assert::is_maybe_transmutable::<u128, repr_rust>();
|
2021-07-03 16:18:13 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
fn aligned() {
|
|
|
|
#[repr(align(1))] struct repr_rust{}
|
Compute transmutability from `rustc_target::abi::Layout`
In its first step of computing transmutability, `rustc_transmutability`
constructs a byte-level representation of type layout (`Tree`). Previously, this
representation was computed for ADTs by inspecting the ADT definition and
performing our own layout computations. This process was error-prone, verbose,
and limited our ability to analyze many types (particularly default-repr types).
In this PR, we instead construct `Tree`s from `rustc_target::abi::Layout`s. This
helps ensure that layout optimizations are reflected our analyses, and increases
the kinds of types we can now analyze, including:
- default repr ADTs
- transparent unions
- `UnsafeCell`-containing types
Overall, this PR expands the expressvity of `rustc_transmutability` to be much
closer to the transmutability analysis performed by miri. Future PRs will work
to close the remaining gaps (e.g., support for `Box`, raw pointers, `NonZero*`,
coroutines, etc.).
2024-03-19 14:49:13 +00:00
|
|
|
assert::is_maybe_transmutable::<repr_rust, ()>();
|
|
|
|
assert::is_maybe_transmutable::<u128, repr_rust>();
|
2021-07-03 16:18:13 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
fn packed() {
|
|
|
|
#[repr(packed)] struct repr_rust{}
|
Compute transmutability from `rustc_target::abi::Layout`
In its first step of computing transmutability, `rustc_transmutability`
constructs a byte-level representation of type layout (`Tree`). Previously, this
representation was computed for ADTs by inspecting the ADT definition and
performing our own layout computations. This process was error-prone, verbose,
and limited our ability to analyze many types (particularly default-repr types).
In this PR, we instead construct `Tree`s from `rustc_target::abi::Layout`s. This
helps ensure that layout optimizations are reflected our analyses, and increases
the kinds of types we can now analyze, including:
- default repr ADTs
- transparent unions
- `UnsafeCell`-containing types
Overall, this PR expands the expressvity of `rustc_transmutability` to be much
closer to the transmutability analysis performed by miri. Future PRs will work
to close the remaining gaps (e.g., support for `Box`, raw pointers, `NonZero*`,
coroutines, etc.).
2024-03-19 14:49:13 +00:00
|
|
|
assert::is_maybe_transmutable::<repr_rust, ()>();
|
|
|
|
assert::is_maybe_transmutable::<u128, repr_rust>();
|
2021-07-03 16:18:13 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
fn nested() {
|
|
|
|
struct repr_rust;
|
|
|
|
#[repr(C)] struct repr_c(repr_rust);
|
Compute transmutability from `rustc_target::abi::Layout`
In its first step of computing transmutability, `rustc_transmutability`
constructs a byte-level representation of type layout (`Tree`). Previously, this
representation was computed for ADTs by inspecting the ADT definition and
performing our own layout computations. This process was error-prone, verbose,
and limited our ability to analyze many types (particularly default-repr types).
In this PR, we instead construct `Tree`s from `rustc_target::abi::Layout`s. This
helps ensure that layout optimizations are reflected our analyses, and increases
the kinds of types we can now analyze, including:
- default repr ADTs
- transparent unions
- `UnsafeCell`-containing types
Overall, this PR expands the expressvity of `rustc_transmutability` to be much
closer to the transmutability analysis performed by miri. Future PRs will work
to close the remaining gaps (e.g., support for `Box`, raw pointers, `NonZero*`,
coroutines, etc.).
2024-03-19 14:49:13 +00:00
|
|
|
assert::is_maybe_transmutable::<repr_c, ()>();
|
|
|
|
assert::is_maybe_transmutable::<u128, repr_c>();
|
2021-07-03 16:18:13 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
Compute transmutability from `rustc_target::abi::Layout`
In its first step of computing transmutability, `rustc_transmutability`
constructs a byte-level representation of type layout (`Tree`). Previously, this
representation was computed for ADTs by inspecting the ADT definition and
performing our own layout computations. This process was error-prone, verbose,
and limited our ability to analyze many types (particularly default-repr types).
In this PR, we instead construct `Tree`s from `rustc_target::abi::Layout`s. This
helps ensure that layout optimizations are reflected our analyses, and increases
the kinds of types we can now analyze, including:
- default repr ADTs
- transparent unions
- `UnsafeCell`-containing types
Overall, this PR expands the expressvity of `rustc_transmutability` to be much
closer to the transmutability analysis performed by miri. Future PRs will work
to close the remaining gaps (e.g., support for `Box`, raw pointers, `NonZero*`,
coroutines, etc.).
2024-03-19 14:49:13 +00:00
|
|
|
fn should_accept_repr_c()
|
2021-07-03 16:18:13 +00:00
|
|
|
{
|
|
|
|
fn unit() {
|
|
|
|
#[repr(C)] struct repr_c;
|
|
|
|
assert::is_maybe_transmutable::<repr_c, ()>();
|
|
|
|
assert::is_maybe_transmutable::<i128, repr_c>();
|
|
|
|
}
|
|
|
|
|
|
|
|
fn tuple() {
|
|
|
|
#[repr(C)] struct repr_c();
|
|
|
|
assert::is_maybe_transmutable::<repr_c, ()>();
|
|
|
|
assert::is_maybe_transmutable::<i128, repr_c>();
|
|
|
|
}
|
|
|
|
|
|
|
|
fn braces() {
|
|
|
|
#[repr(C)] struct repr_c{}
|
|
|
|
assert::is_maybe_transmutable::<repr_c, ()>();
|
|
|
|
assert::is_maybe_transmutable::<i128, repr_c>();
|
|
|
|
}
|
|
|
|
}
|