The deriving code has some complex parts involving iterations over
selflike args and also fields within structs and enum variants.
The return types for a few functions demonstrate this:
- `TraitDef::create_{struct_pattern,enum_variant_pattern}` returns a
`(P<ast::Pat>, Vec<(Span, Option<Ident>, P<Expr>)>)`
- `TraitDef::create_struct_field_accesses` returns a `Vec<(Span,
Option<Ident>, P<Expr>)>`.
This results in per-field data stored within per-selflike-arg data, with
lots of repetition within the per-field data elements. This then has to
be "transposed" in two places (`expand_struct_method_body` and
`expand_enum_method_body`) into per-self-like-arg data stored within
per-field data. It's all quite clumsy and confusing.
This commit rearranges things greatly. Data is obtained in the needed
form up-front, avoiding the need for transposition. Also, various
functions are split, removed, and added, to make things clearer and
avoid tuple return values.
The diff is hard to read, which reflects the messiness of the original
code -- there wasn't an easy way to break these changes into small
pieces. (Sorry!) It's a net reduction of 35 lines and a readability
improvement. The generated code is unchanged.
The deriving code has inconsistent terminology to describe args.
In some places it distinguishes between:
- the `&self` arg (if present), versus
- all other args.
In other places it distinguishes between:
- the `&self` arg (if present) and any other arguments with the same
type (in practice there is at most one, e.g. in `PartialEq::eq`),
versus
- all other args.
The terms "self_args" and "nonself_args" are sometimes used for the
former distinction, and sometimes for the latter. "args" is also
sometimes used for "all other args".
This commit makes the code consistently uses "self_args"/"nonself_args"
for the former and "selflike_args"/"nonselflike_args" for the latter.
This change makes the code easier to read.
The commit also adds a panic on an impossible path (the `Self_` case) in
`extract_arg_details`.
We currently do a match on the comparison of every field in a struct or
enum variant. But the last field has a degenerate match like this:
```
match ::core::cmp::Ord::cmp(&self.y, &other.y) {
::core::cmp::Ordering::Equal =>
::core::cmp::Ordering::Equal,
cmp => cmp,
},
```
This commit changes it to this:
```
::core::cmp::Ord::cmp(&self.y, &other.y),
```
This is fairly straightforward thanks to the existing `cs_fold1`
function.
The commit also removes the `cs_fold` function which is no longer used.
(Note: there is some repetition now in `cs_cmp` and `cs_partial_cmp`. I
will remove that in a follow-up PR.)
It's common to see repeated assertions like this in derived `clone` and
`eq` methods:
```
let _: ::core::clone::AssertParamIsClone<u32>;
let _: ::core::clone::AssertParamIsClone<u32>;
```
This commit avoids them.
All derive ops currently use match-destructuring to access fields. This
is reasonable for enums, but sub-optimal for structs. E.g.:
```
fn eq(&self, other: &Point) -> bool {
match *other {
Self { x: ref __self_1_0, y: ref __self_1_1 } =>
match *self {
Self { x: ref __self_0_0, y: ref __self_0_1 } =>
(*__self_0_0) == (*__self_1_0) &&
(*__self_0_1) == (*__self_1_1),
},
}
}
```
This commit changes derive ops on structs to use field access instead, e.g.:
```
fn eq(&self, other: &Point) -> bool {
self.x == other.x && self.y == other.y
}
```
This is faster to compile, results in smaller binaries, and is simpler to
generate. Unfortunately, we have to keep the old pattern generating code around
for `repr(packed)` structs because something like `&self.x` (which doesn't show
up in `PartialEq` ops, but does show up in `Debug` and `Hash` ops) isn't
allowed. But this commit at least changes those cases to use let-destructuring
instead of match-destructuring, e.g.:
```
fn hash<__H: ::core:#️⃣:Hasher>(&self, state: &mut __H) -> () {
{
let Self(ref __self_0_0) = *self;
{ ::core:#️⃣:Hash::hash(&(*__self_0_0), state) }
}
}
```
There are some unnecessary blocks remaining in the generated code, but I
will fix them in a follow-up PR.
The existing derive code allows for various possibilities that aren't
needed in practice, which complicates the code. There are only a few
auto-derived traits and new ones are unlikely, so this commit simplifies
things.
- `PtrTy` has been eliminated. The `Raw` variant was never used, and the
lifetime for the `Borrowed` variant was always `None`. That left just
the mutability field, which has been inlined as necessary.
- `MethodDef::explicit_self` was a confusing `Option<Option<PtrTy>>`.
Indicating either `&self` or nothing. It's now a `bool`.
- `borrowed_self` is renamed as `self_ref`.
- `Ty::Ptr` is renamed to `Ty::Ref`.
The `&[ast::Variant]` field isn't used.
The `Vec<Ident>` field is only used for its length, but that's always
the same as the length of the `&[Ident]` and so isn't necessary.
[RFC 2011] Optimize non-consuming operators
Tracking issue: https://github.com/rust-lang/rust/issues/44838
Fifth step of https://github.com/rust-lang/rust/pull/96496
The most non-invasive approach that will probably have very little to no performance impact.
## Current behaviour
Captures are handled "on-the-fly", i.e., they are performed in the same place expressions are located.
```rust
// `let a = 1; let b = 2; assert!(a > 1 && b < 100);`
if !(
{ ***try capture `a` and then return `a`*** } > 1 && { ***try capture `b` and then return `b`*** } < 100
) {
panic!( ... );
}
```
As such, some overhead is likely to occur (Specially with very large chains of conditions).
## New behaviour for non-consuming operators
When an operator is known to not take `self`, then it is possible to capture variables **AFTER** the condition.
```rust
// `let a = 1; let b = 2; assert!(a > 1 && b < 100);`
if !( a > 1 && b < 100 ) {
{ ***try capture `a`*** }
{ ***try capture `b`*** }
panic!( ... );
}
```
So the possible impact on the runtime execution time will be diminished.
r? ````@oli-obk````
Currently the generated code for methods like `eq`, `ne`, and `partial_cmp`
includes stuff like this:
```
let __self_vi = ::core::intrinsics::discriminant_value(&*self);
let __arg_1_vi = ::core::intrinsics::discriminant_value(&*other);
if true && __self_vi == __arg_1_vi {
...
}
```
This commit removes the unnecessary `true &&`, and makes the generating
code a little easier to read in the process. It also fixes some errors
in comments.
macros: use typed identifiers in diag and subdiag derive
Using typed identifiers instead of strings with the Fluent identifiers in the diagnostic and subdiagnostic derives - this enables the diagnostic derive to benefit from the compile-time validation that comes with typed identifiers, namely that use of a non-existent Fluent identifier will not compile.
r? `````@oli-obk`````
Using typed identifiers instead of strings with the Fluent identifier
enables the diagnostic derive to benefit from the compile-time
validation that comes with typed identifiers - use of a non-existent
Fluent identifier will not compile.
Signed-off-by: David Wood <david.wood@huawei.com>
Fixup missing renames from `#[main]` to `#[rustc_main]`
In #84217 `#[main]` was removed and replaced with `#[rustc_main]`. In some places the rename was forgotten, which makes the current code confusing, because at first glance it seems that `#[main]` is still around. Perform the renames also in these places.
I noticed this (after first being confused by it) when working on #97802.
r? `@petrochenkov`
(since you reviewed the other PR)
This commit adds new methods that combine sequences of existing
formatting methods.
- `Formatter::debug_{tuple,struct}_field[12345]_finish`, equivalent to a
`Formatter::debug_{tuple,struct}` + N x `Debug{Tuple,Struct}::field` +
`Debug{Tuple,Struct}::finish` call sequence.
- `Formatter::debug_{tuple,struct}_fields_finish` is similar, but can
handle any number of fields by using arrays.
These new methods are all marked as `doc(hidden)` and unstable. They are
intended for the compiler's own use.
Special-casing up to 5 fields gives significantly better performance
results than always using arrays (as was tried in #95637).
The commit also changes the `Debug` deriving code to use these new methods. For
example, where the old `Debug` code for a struct with two fields would be like
this:
```
fn fmt(&self, f: &mut ::core::fmt::Formatter) -> ::core::fmt::Result {
match *self {
Self {
f1: ref __self_0_0,
f2: ref __self_0_1,
} => {
let debug_trait_builder = &mut ::core::fmt::Formatter::debug_struct(f, "S2");
let _ = ::core::fmt::DebugStruct::field(debug_trait_builder, "f1", &&(*__self_0_0));
let _ = ::core::fmt::DebugStruct::field(debug_trait_builder, "f2", &&(*__self_0_1));
::core::fmt::DebugStruct::finish(debug_trait_builder)
}
}
}
```
the new code is like this:
```
fn fmt(&self, f: &mut ::core::fmt::Formatter) -> ::core::fmt::Result {
match *self {
Self {
f1: ref __self_0_0,
f2: ref __self_0_1,
} => ::core::fmt::Formatter::debug_struct_field2_finish(
f,
"S2",
"f1",
&&(*__self_0_0),
"f2",
&&(*__self_0_1),
),
}
}
```
This shrinks the code produced for `Debug` instances
considerably, reducing compile times and binary sizes.
Co-authored-by: Scott McMurray <scottmcm@users.noreply.github.com>
In fc357039f9 `#[main]` was removed and replaced with `#[rustc_main]`.
In some place the rename was forgotten, which makes the current code
confusing, because at first glance it seems that `#[main]` is still
around. Perform the renames also in these places.
proc_macro: don't pass a client-side function pointer through the server.
Before this PR, `proc_macro::bridge::Client<F>` contained both:
* the C ABI entry-point `run`, that the server can call to start the client
* some "payload" `f: F` passed to that entry-point
* in practice, this was always a (client-side Rust ABI) `fn` pointer to the actual function the proc macro author wrote, i.e. `#[proc_macro] fn foo(input: TokenStream) -> TokenStream`
In other words, the client was passing one of its (Rust) `fn` pointers to the server, which was passing it back to the client, for the client to call (see later below for why that was ever needed).
I was inspired by `@nnethercote's` attempt to remove the `get_handle_counters` field from `Client` (see https://github.com/rust-lang/rust/pull/97004#issuecomment-1139273301), which combined with removing the `f` ("payload") field, could theoretically allow for a `#[repr(transparent)]` `Client` that mostly just newtypes the C ABI entry-point `fn` pointer <sub>(and in the context of e.g. wasm isolation, that's *all* you want, since you can reason about it from outside the wasm VM, as just a 32-bit "function table index", that you can pass to the wasm VM to call that function)</sub>.
<hr/>
So this PR removes that "payload". But it's not a simple refactor: the reason the field existed in the first place is because monomorphizing over a function type doesn't let you call the function without having a value of that type, because function types don't implement anything like `Default`, i.e.:
```rust
extern "C" fn ffi_wrapper<A, R, F: Fn(A) -> R>(arg: A) -> R {
let f: F = ???; // no way to get a value of `F`
f(arg)
}
```
That could be solved with something like this, if it was allowed:
```rust
extern "C" fn ffi_wrapper<
A, R,
F: Fn(A) -> R,
const f: F // not allowed because the type is a generic param
>(arg: A) -> R {
f(arg)
}
```
Instead, this PR contains a workaround in `proc_macro::bridge::selfless_reify` (see its module-level comment for more details) that can provide something similar to the `ffi_wrapper` example above, but limited to `F` being `Copy` and ZST (and requiring an `F` value to prove the caller actually can create values of `F` and it's not uninhabited or some other unsound situation).
<hr/>
Hopefully this time we don't have a performance regression, and this has a chance to land.
cc `@mystor` `@bjorn3`
