This commit is a refactoring of the LTO backend in Rust to support compilations
with multiple codegen units. The immediate result of this PR is to remove the
artificial error emitted by rustc about `-C lto -C codegen-units-8`, but longer
term this is intended to lay the groundwork for LTO with incremental compilation
and ultimately be the underpinning of ThinLTO support.
The problem here that needed solving is that when rustc is producing multiple
codegen units in one compilation LTO needs to merge them all together.
Previously only upstream dependencies were merged and it was inherently relied
on that there was only one local codegen unit. Supporting this involved
refactoring the optimization backend architecture for rustc, namely splitting
the `optimize_and_codegen` function into `optimize` and `codegen`. After an LLVM
module has been optimized it may be blocked and queued up for LTO, and only
after LTO are modules code generated.
Non-LTO compilations should look the same as they do today backend-wise, we'll
spin up a thread for each codegen unit and optimize/codegen in that thread. LTO
compilations will, however, send the LLVM module back to the coordinator thread
once optimizations have finished. When all LLVM modules have finished optimizing
the coordinator will invoke the LTO backend, producing a further list of LLVM
modules. Currently this is always a list of one LLVM module. The coordinator
then spawns further work to run LTO and code generation passes over each module.
In the course of this refactoring a number of other pieces were refactored:
* Management of the bytecode encoding in rlibs was centralized into one module
instead of being scattered across LTO and linking.
* Some internal refactorings on the link stage of the compiler was done to work
directly from `CompiledModule` structures instead of lists of paths.
* The trans time-graph output was tweaked a little to include a name on each
bar and inflate the size of the bars a little
rustc: Default 32 codegen units at O0
This commit changes the default of rustc to use 32 codegen units when compiling
in debug mode, typically an opt-level=0 compilation. Since their inception
codegen units have matured quite a bit, gaining features such as:
* Parallel translation and codegen enabling codegen units to get worked on even
more quickly.
* Deterministic and reliable partitioning through the same infrastructure as
incremental compilation.
* Global rate limiting through the `jobserver` crate to avoid overloading the
system.
The largest benefit of codegen units has forever been faster compilation through
parallel processing of modules on the LLVM side of things, using all the cores
available on build machines that typically have many available. Some downsides
have been fixed through the features above, but the major downside remaining is
that using codegen units reduces opportunities for inlining and optimization.
This, however, doesn't matter much during debug builds!
In this commit the default number of codegen units for debug builds has been
raised from 1 to 32. This should enable most `cargo build` compiles that are
bottlenecked on translation and/or code generation to immediately see speedups
through parallelization on available cores.
Work is being done to *always* enable multiple codegen units (and therefore
parallel codegen) but it requires #44841 at least to be landed and stabilized,
but stay tuned if you're interested in that aspect!
Point at signature on unused lint
```
warning: struct is never used: `Struct`
--> $DIR/unused-warning-point-at-signature.rs:22:1
|
22 | struct Struct {
| ^^^^^^^^^^^^^
```
Fix#33961.
Point at parameter type on E0301
On "the parameter type `T` may not live long enough" error, point to the
parameter type suggesting lifetime bindings:
```
error[E0310]: the parameter type `T` may not live long enough
--> $DIR/lifetime-doesnt-live-long-enough.rs:28:5
|
27 | struct Foo<T> {
| - help: consider adding an explicit lifetime bound `T: 'static`...
28 | foo: &'static T
| ^^^^^^^^^^^^^^^
|
note: ...so that the reference type `&'static T` does not outlive the data it points at
--> $DIR/lifetime-doesnt-live-long-enough.rs:28:5
|
28 | foo: &'static T
| ^^^^^^^^^^^^^^^
```
Fix#36700.
Initial support for `..=` syntax
#28237
This PR adds `..=` as a synonym for `...` in patterns and expressions.
Since `...` in expressions was never stable, we now issue a warning.
cc @durka
r? @aturon
Fix capacity comparison in reserve
You can otherwise end up in a situation where you don't actually resize
but still call into handle_cap_increase which then corrupts head/tail.
Closes#44800
Not totally sure the right way to write a test for this - there are some debug asserts the old bad behavior will hit but we don't build the stdlib with debug assertions by default.
r? @Gankro
This commit changes the default of rustc to use 32 codegen units when compiling
in debug mode, typically an opt-level=0 compilation. Since their inception
codegen units have matured quite a bit, gaining features such as:
* Parallel translation and codegen enabling codegen units to get worked on even
more quickly.
* Deterministic and reliable partitioning through the same infrastructure as
incremental compilation.
* Global rate limiting through the `jobserver` crate to avoid overloading the
system.
The largest benefit of codegen units has forever been faster compilation through
parallel processing of modules on the LLVM side of things, using all the cores
available on build machines that typically have many available. Some downsides
have been fixed through the features above, but the major downside remaining is
that using codegen units reduces opportunities for inlining and optimization.
This, however, doesn't matter much during debug builds!
In this commit the default number of codegen units for debug builds has been
raised from 1 to 32. This should enable most `cargo build` compiles that are
bottlenecked on translation and/or code generation to immediately see speedups
through parallelization on available cores.
Work is being done to *always* enable multiple codegen units (and therefore
parallel codegen) but it requires #44841 at least to be landed and stabilized,
but stay tuned if you're interested in that aspect!
Add suggestions for misspelled method names
Use the syntax::util::lev_distance module to provide suggestions when a
named method cannot be found.
Part of #30197
Require rlibs for dependent crates when linking static executables
This handles the case for `CrateTypeExecutable` and `+crt_static`. I reworked the match block to avoid duplicating the `attempt_static` and error checking code again (this case would have been a copy of the `CrateTypeCdylib`/`CrateTypeStaticlib` case).
On `linux-musl` targets where `std` was built with `crt_static = false` in `config.toml`, this change brings the test suite from entirely failing to mostly passing.
This change should not affect behavior for other crate types, or for targets which do not respect `+crt_static`.
Move effect-checking to MIR
This allows emitting lints from MIR and moves the effect-checking pass to work on it.
I'll make `repr(packed)` misuse unsafe in a separate PR.
r? @eddyb
On "the parameter type `T` may not live long enough" error, point to the
parameter type suggesting lifetime bindings:
```
error[E0310]: the parameter type `T` may not live long enough
--> $DIR/lifetime-doesnt-live-long-enough.rs:28:5
|
27 | struct Foo<T> {
| - help: consider adding an explicit lifetime bound `T: 'static`...
28 | foo: &'static T
| ^^^^^^^^^^^^^^^
|
note: ...so that the reference type `&'static T` does not outlive the data it points at
--> $DIR/lifetime-doesnt-live-long-enough.rs:28:5
|
28 | foo: &'static T
| ^^^^^^^^^^^^^^^
```
The convention for suggesting close matches is to provide at most one match (the
closest one). Change the suggestions for misspelt method names to obey that.
add comparison operators to must-use lint (under `fn_must_use` feature)
Although RFC 1940 is about annotating functions with `#[must_use]`, a
key part of the motivation was linting unused equality operators.
(See
https://github.com/rust-lang/rfcs/pull/1812#issuecomment-265695898—it
seems to have not been clear to discussants at the time that marking the
comparison methods as `must_use` would not give us the lints on
comparison operators, at least in (what the present author understood
as) the most straightforward implementation, as landed in #43728
(3645b062).)
To rectify the situation, we here lint unused comparison operators as
part of the unused-must-use lint (feature gated by the `fn_must_use`
feature flag, which now arguably becomes a slight (tolerable in the
opinion of the present author) misnomer).
This is in the matter of #43302.
cc @crumblingstatue
Improve diagnostics when attempting to match tuple enum variant with struct pattern
Adds an extra note as below to explain that a tuple pattern was probably intended.
```
error[E0026]: variant `X::Y` does not have a field named `data`
--> src/main.rs:18:16
|
18 | X::Y { data } => println!("The data is {}", data)
| ^^^^ variant `X::Y` does not have field `data`
error[E0027]: pattern does not mention field `0`
--> src/main.rs:18:9
|
18 | X::Y { data } => println!("The data is {}", data)
| ^^^^^^^^^^^^^ missing field `0`
|
= note: trying to match a tuple variant with a struct variant pattern
```
Fixes#41314.
Record semantic types for all syntactic types in bodies
... and use recorded types in type privacy checking (types are recorded after inference, so there are no `_`s left).
Also use `hir_ty_to_ty` for types in signatures in type privacy checking.
This could also be potentially useful for save-analysis and diagnostics.
Fixes https://github.com/rust-lang/rust/pull/42125#issuecomment-305987755
r? @eddyb
Less confusing placeholder when RefCell is exclusively borrowed
Based on ExpHP's comment in [*RefCell.borrow_mut get strange result*](https://users.rust-lang.org/t/refcell-borrow-mut-get-strange-result/12994):
> it would perhaps be nicer if it didn't put something that could be misinterpreted as a valid string value
The previous Debug implementation would show:
RefCell { value: "<borrowed>" }
The new one is:
RefCell { value: <borrowed> }
rustc: Don't use DelimToken::None if possible
This commit fixes a regression from #44601 where lowering attribute to HIR now
involves expanding interpolated tokens to their actual tokens. In that commit
all interpolated tokens were surrounded with a `DelimToken::None` group of
tokens, but this ended up causing regressions like #44730 where the various
attribute parsers in `syntax/attr.rs` weren't ready to cope with
`DelimToken::None`. Instead of fixing the parser in `attr.rs` this commit
instead opts to just avoid the `DelimToken::None` in the first place, ensuring
that the token stream should look the same as it did before where possible.
Closes#44730
only set non-ADT derive error once per attribute, not per trait
I found the expansion code very hard to follow, leaving me unsure as to whether this might somehow be done better, but this patch does give us the behavior requested in #43927 (up to exact choice of span; here, it's the entire attribute, not just the `derive` token).
(Note to GitHub robots: _resolves #43927_.)
r? @jseyfried
Although RFC 1940 is about annotating functions with `#[must_use]`, a
key part of the motivation was linting unused equality operators.
(See
https://github.com/rust-lang/rfcs/pull/1812#issuecomment-265695898—it
seems to have not been clear to discussants at the time that marking the
comparison methods as `must_use` would not give us the lints on
comparison operators, at least in (what the present author understood
as) the most straightforward implementation, as landed in #43728
(3645b062).)
To rectify the situation, we here lint unused comparison operators as
part of the unused-must-use lint (feature gated by the `fn_must_use`
feature flag, which now arguably becomes a slight (tolerable in the
opinion of the present author) misnomer).
This is in the matter of #43302.
Add ..= to the parser
Add ..= to libproc_macro
Add ..= to ICH
Highlight ..= in rustdoc
Update impl Debug for RangeInclusive to ..=
Replace `...` to `..=` in range docs
Make the dotdoteq warning point to the ...
Add warning for ... in expressions
Updated more tests to the ..= syntax
Updated even more tests to the ..= syntax
Updated the inclusive_range entry in unstable book
Based on ExpHP's comment in
https://users.rust-lang.org/t/refcell-borrow-mut-get-strange-result/12994
> it would perhaps be nicer if it didn't put something that could be
> misinterpreted as a valid string value
The previous Debug implementation would show:
RefCell { value: "<borrowed>" }
The new one is:
RefCell { value: <borrowed> }
A slight eccentricity of this change is that now non-ADT-derive errors prevent
derive-macro-not-found errors from surfacing (see changes to the
gating-of-derive compile-fail tests).
Resolves#43927.
don't suggest placing `use` statements into expanded code
r? @nrc
fixes#44210
```rust
#[derive(Debug)]
struct Foo;
type X = Path;
```
will try to place `use std::path::Path;` between `#[derive(Debug)]` and `struct Foo;`
I am not sure how to obtain a span before the first attribute, because derive attributes are removed during expansion.
It would be trivial to detect this case and place the `use` after the item, but that would be somewhat weird I think.
Alex Crichton