Commit Graph

103 Commits

Author SHA1 Message Date
Dylan DPC
b81c6cdb57
Rollup merge of #83916 - Amanieu:asm_anonconst, r=petrochenkov
Use AnonConst for asm! constants

This replaces the old system which used explicit promotion. See #83169 for more background.

The syntax for `const` operands is still the same as before: `const <expr>`.

Fixes #83169

Because the implementation is heavily based on inline consts, we suffer from the same issues:
- We lose the ability to use expressions derived from generics. See the deleted tests in `src/test/ui/asm/const.rs`.
- We are hitting the same ICEs as inline consts, for example #78174. It is unlikely that we will be able to stabilize this before inline consts are stabilized.
2021-04-07 13:07:14 +02:00
Amanieu d'Antras
32be124e30 Use AnonConst for asm! constants 2021-04-06 12:35:41 +01:00
Vadim Petrochenkov
fbf1bec482 resolve/expand: Cache intermediate results of #[derive] expansion 2021-04-04 17:51:41 +03:00
Josh Stone
72ebebe474 Use iter::zip in compiler/ 2021-03-26 09:32:31 -07:00
Dylan DPC
b0bec95534
Rollup merge of #83486 - Aaron1011:fix/global-alloc-error, r=petrochenkov
Don't ICE when using `#[global_alloc]` on a non-item statement

Fixes #83469

We need to return an `Annotatable::Stmt` if we were passed an
`Annotatable::Stmt`
2021-03-26 02:34:45 +01:00
Aaron Hill
8ecd931a8e
Don't ICE when using #[global_alloc] on a non-item statement
Fixes #83469

We need to return an `Annotatable::Stmt` if we were passed an
`Annotatable::Stmt`
2021-03-25 15:41:31 -04:00
Amanieu d'Antras
5dabc80796 Refactor #82270 as lint instead of an error 2021-03-25 13:12:29 +00:00
mark
db5629adcb stabilize or_patterns 2021-03-19 19:45:32 -05:00
Dylan DPC
16f6583f2d
Rollup merge of #82270 - asquared31415:asm-syntax-directive-errors, r=nagisa
Emit error when trying to use assembler syntax directives in `asm!`

The `.intel_syntax` and `.att_syntax` assembler directives should not be used, in favor of not specifying a syntax for intel, and in favor of the explicit `att_syntax` option using the inline assembly options.

Closes #79869
2021-03-18 00:28:06 +01:00
Vadim Petrochenkov
b25d3ba781 ast/hir: Rename field-related structures
StructField -> FieldDef ("field definition")
Field -> ExprField ("expression field", not "field expression")
FieldPat -> PatField ("pattern field", not "field pattern")

Also rename visiting and other methods working on them.
2021-03-16 11:41:24 +03:00
Aaron Hill
18f89790da
Bump recursion_limit in a few places
This is needed to get rustdoc to succeed on `dist-x86_64-linux-alt`
2021-03-14 23:02:01 -04:00
Dylan DPC
759204ffc4
Rollup merge of #82217 - m-ou-se:edition-prelude, r=nikomatsakis
Edition-specific preludes

This changes `{std,core}::prelude` to export edition-specific preludes under `rust_2015`, `rust_2018` and `rust_2021`. (As suggested in https://github.com/rust-lang/rust/issues/51418#issuecomment-395630382.) For now they all just re-export `v1::*`, but this allows us to add things to the 2021edition prelude soon.

This also changes the compiler to make the automatically injected prelude import dependent on the selected edition.

cc `@rust-lang/libs` `@djc`
2021-03-10 17:55:38 +01:00
asquared31415
05ae66607f Move default inline asm dialect to Session 2021-03-08 12:16:12 -05:00
Dylan DPC
9c310571a8
Rollup merge of #82682 - petrochenkov:cfgeval, r=Aaron1011
Implement built-in attribute macro `#[cfg_eval]` + some refactoring

This PR implements a built-in attribute macro `#[cfg_eval]` as it was suggested in https://github.com/rust-lang/rust/pull/79078 to avoid `#[derive()]` without arguments being abused as a way to configure input for other attributes.

The macro is used for eagerly expanding all `#[cfg]` and `#[cfg_attr]` attributes in its input ("fully configuring" the input).
The effect is identical to effect of `#[derive(Foo, Bar)]` which also fully configures its input before passing it to macros `Foo` and `Bar`, but unlike `#[derive]` `#[cfg_eval]` can be applied to any syntax nodes supporting macro attributes, not only certain items.

`cfg_eval` was the first name suggested in https://github.com/rust-lang/rust/pull/79078, but other alternatives are also possible, e.g. `cfg_expand`.

```rust
#[cfg_eval]
#[my_attr] // Receives `struct S {}` as input, the field is configured away by `#[cfg_eval]`
struct S {
    #[cfg(FALSE)]
    field: u8,
}
```

Tracking issue: https://github.com/rust-lang/rust/issues/82679
2021-03-08 13:13:23 +01:00
Vadim Petrochenkov
5d27728141 rustc_builtin_macros: Share some more logic between derive and cfg_eval 2021-03-07 01:12:18 +03:00
Vadim Petrochenkov
10ed08f5b6 cfg_eval: Configure everything through mutable visitor methods
This is simpler and mirrors what invocation collector does
2021-03-07 00:23:02 +03:00
Vadim Petrochenkov
f9019b7086 Move full configuration logic from rustc_expand to rustc_builtin_macros
This logic is applicable to two specific macros and not to the expansion infrastructure in general.
2021-03-07 00:17:31 +03:00
Vadim Petrochenkov
5dad6c2575 Implement built-in attribute macro #[cfg_eval] 2021-03-06 23:03:19 +03:00
Vadim Petrochenkov
069e612e73 rustc_ast: Replace AstLike::finalize_tokens with a getter tokens_mut 2021-03-06 21:19:31 +03:00
Vadim Petrochenkov
46b67aa74d expand: Some more consistent naming in module loading 2021-03-05 01:33:43 +03:00
Vadim Petrochenkov
39052c55bb expand: Move module file path stack from global session to expansion data
Also don't push the paths on the stack directly in `fn parse_external_mod`, return them instead.
2021-03-05 01:33:43 +03:00
Mara Bos
d3b564c3d7 Pick the injected prelude based on the edition. 2021-02-25 12:47:58 +01:00
Dániel Buga
10f234240d Remove some P-s 2021-02-20 10:51:26 +01:00
asquared31415
39dcd01bf5 Take into account target default syntax 2021-02-20 01:17:18 -05:00
asquared31415
12c6a12d62 Emit error when trying to use assembler syntax directives in asm! 2021-02-18 14:27:11 -05:00
Vadim Petrochenkov
4a88165124 ast: Keep expansion status for out-of-line module items
Also remove `ast::Mod` which is mostly redundant now
2021-02-18 13:07:49 +03:00
Vadim Petrochenkov
eb65f15c78 ast: Stop using Mod in Crate
Crate root is sufficiently different from `mod` items, at least at syntactic level.

Also remove customization point for "`mod` item or crate root" from AST visitors.
2021-02-18 13:07:49 +03:00
Matthias Krüger
4390a61b64 avoid full-slicing slices
If we already have a slice, there is no need to get another full-range slice from that, just use the original.
clippy::redundant_slicing
2021-02-16 00:31:11 +01:00
klensy
93c8ebe022 bumped smallvec deps 2021-02-14 18:03:11 +03:00
Tomasz Miąsko
1cf95059eb Borrow builder only once in debug derive 2021-02-10 00:00:00 +00:00
Skgland
0375022c73
fix derive(RustcEncodable, RustcDecodable) 2021-02-09 13:42:36 +01:00
Skgland
091ef95f8e
use ufcs in derive(RustDecodable) 2021-02-09 13:42:36 +01:00
Skgland
525fc4b8e4
use ufcs in derive(RustEncodable) 2021-02-09 13:42:36 +01:00
Skgland
2c33b070ad
use ufcs in derive(Ord) and derive(PartialOrd) 2021-02-09 13:42:35 +01:00
Tomasz Miąsko
e4efccd4a6 Fix derived PartialOrd operators
The derived implementation of `partial_cmp` compares matching fields one
by one, stopping the computation when the result of a comparison is not
equal to `Some(Equal)`.

On the other hand the derived implementation for `lt`, `le`, `gt` and
`ge` continues the computation when the result of a field comparison is
`None`, consequently those operators are not transitive and inconsistent
with `partial_cmp`.

Fix the inconsistency by using the default implementation that fall-backs
to the `partial_cmp`. This also avoids creating very deeply nested
closures that were quite costly to compile.
2021-02-09 08:15:37 +01:00
Vadim Petrochenkov
dbdbd30bf2 expand/resolve: Turn #[derive] into a regular macro attribute 2021-02-07 20:08:45 +03:00
bors
186f7ae5b0 Auto merge of #81294 - pnkfelix:issue-81211-use-ufcs-in-derive-debug, r=oli-obk
Use ufcs in derive(Debug)

Cc #81211.

(Arguably this *is* the fix for it.)
2021-02-03 15:12:19 +00:00
Jack Huey
7f2eeb10c7
Rollup merge of #81647 - m-ou-se:assert-2021-fix, r=petrochenkov
Fix bug with assert!() calling the wrong edition of panic!().

The span of `panic!` produced by the `assert` macro did not carry the right edition. This changes `assert` to call the right version.

Also adds tests for the 2021 edition of panic and assert, that would've caught this.
2021-02-02 16:01:46 -05:00
bors
3182375e06 Auto merge of #81405 - bugadani:ast, r=cjgillot
Box the biggest ast::ItemKind variants

This PR is a different approach on https://github.com/rust-lang/rust/pull/81400, aiming to save memory in humongous ASTs.

The three affected item kind enums are:
 - `ast::ItemKind` (208 -> 112 bytes)
 - `ast::AssocItemKind` (176 -> 72 bytes)
 - `ast::ForeignItemKind` (176 -> 72 bytes)
2021-02-02 17:34:08 +00:00
Mark Rousskov
d5b760ba62 Bump rustfmt version
Also switches on formatting of the mir build module
2021-02-02 09:09:52 -05:00
Mara Bos
ed1de99b4f Fix bug with assert!() calling the wrong edition of panic!().
The span of `panic!` produced by the `assert` macro did not carry the
right edition. This changes `assert` to call the right version.
2021-02-01 23:23:27 +01:00
Felix S. Klock II
a7745d965e placate tidy. 2021-02-01 17:08:38 -05:00
Felix S. Klock II
2307d08d2f Use UFCS instead of method calls in derive(Debug). See issue 81211 for discussion. 2021-02-01 17:08:37 -05:00
bors
e0d9f79399 Auto merge of #80851 - m-ou-se:panic-2021, r=petrochenkov
Implement Rust 2021 panic

This implements the Rust 2021 versions of `panic!()`. See https://github.com/rust-lang/rust/issues/80162 and https://github.com/rust-lang/rfcs/pull/3007.

It does so by replacing `{std, core}::panic!()` by a bulitin macro that expands to either `$crate::panic::panic_2015!(..)` or `$crate::panic::panic_2021!(..)` depending on the edition of the caller.

This does not yet make std's panic an alias for core's panic on Rust 2021 as the RFC proposes. That will be a separate change: c5273bdfb2 That change is blocked on figuring out what to do with https://github.com/rust-lang/rust/issues/80846 first.
2021-02-01 10:25:31 +00:00
Dániel Buga
b87e1ecdf0 Box the biggest ast::ItemKind variants 2021-02-01 09:23:39 +01:00
Mara Bos
d5414f9a9f Implement new panic!() behaviour for Rust 2021. 2021-01-25 13:48:11 +01:00
Jonas Schievink
e8ef15d9d1
Rollup merge of #80855 - m-ou-se:assert-2021, r=petrochenkov
Expand assert!(expr, args..) to include $crate for hygiene on 2021.

This makes `assert!(expr, args..)` properly hygienic in Rust 2021.

This is part of rust-lang/rfcs#3007, see #80162.

Before edition 2021, this was a breaking change, as `std::panic` and `core::panic` are different. In edition 2021 they will be identical, making it possible to apply proper hygiene here.
2021-01-24 22:09:53 +01:00
Mara Bos
a730970dff Only call span.rust_2021() when necessary. 2021-01-24 14:19:05 +01:00
Aaron Hill
11b1e37016
Force token collection to run when parsing nonterminals
Fixes #81007

Previously, we would fail to collect tokens in the proper place when
only builtin attributes were present. As a result, we would end up with
attribute tokens in the collected `TokenStream`, leading to duplication
when we attempted to prepend the attributes from the AST node.

We now explicitly track when token collection must be performed due to
nomterminal parsing.
2021-01-20 18:09:32 -05:00
Vadim Petrochenkov
f9b5859173 resolve: Simplify built-in macro table 2021-01-10 14:48:47 +03:00