bors [Sun, 2 Jan 2022 06:28:34 +0000 (06:28 +0000)]
Auto merge of #92034 - petrochenkov:nolinknores, r=joshtriplett
Remove effect of `#[no_link]` attribute on name resolution
Previously it hid all non-macro names from other crates.
This has no relation to linking and can change name resolution behavior in some cases (e.g. glob conflicts), in addition to just producing the "unresolved name" errors.
I can kind of understand the possible reasoning behind the current behavior - if you can use names from a `no_link` crates then you can use, for example, functions too, but whether it will actually work or produce link-time errors will depend on random factors like inliner behavior.
(^^^ This is not the actual reason why the current behavior exist, I've looked through git history and it's mostly accidental.)
I think this risk is ok for such an obscure attribute, and we don't need to specifically prevent use of non-macro items from such crates.
(I'm not actually sure why would anyone use `#[no_link]` on a crate, even if it's macro only, if you aware of any use cases, please share. IIRC, at some point it was used for crates implementing custom derives - the now removed legacy ones, not the current proc macros.)
Extracted from https://github.com/rust-lang/rust/pull/91795.
bors [Sun, 2 Jan 2022 00:20:04 +0000 (00:20 +0000)]
Auto merge of #92482 - matthiaskrgr:rollup-uso1zi0, r=matthiaskrgr
Rollup of 7 pull requests
Successful merges:
- #84083 (Clarify the guarantees that ThreadId does and doesn't make.)
- #91593 (Remove unnecessary bounds for some Hash{Map,Set} methods)
- #92297 (Reduce compile time of rustbuild)
- #92332 (Add test for where clause order)
- #92438 (Enforce formatting for rustc_codegen_cranelift)
- #92463 (Remove pronunciation guide from Vec<T>)
- #92468 (Emit an error for `--cfg=)`)
Matthias Krüger [Sat, 1 Jan 2022 21:49:53 +0000 (22:49 +0100)]
Rollup merge of #92468 - NieDzejkob:silent-cfg, r=petrochenkov
Emit an error for `--cfg=)`
Fixes #73026
See also: #64467, #89468
The issue stems from a `FatalError` being silently raised in
`panictry_buffer`. Normally this is not a problem, because
`panictry_buffer` emits the causes of the error, but they are not
themselves fatal, so they get filtered out by the silent emitter.
To fix this, we use a parser entrypoint which doesn't use
`panictry_buffer`, and we handle the error ourselves.
Matthias Krüger [Sat, 1 Jan 2022 21:49:52 +0000 (22:49 +0100)]
Rollup merge of #92463 - thomcc:thats-not-how-its-pronounced, r=joshtriplett
Remove pronunciation guide from Vec<T>
I performed an extremely scientific poll on twitter, and determined this is not how it's pronounced: https://twitter.com/at_tcsc/status/1476643344285581315
Matthias Krüger [Sat, 1 Jan 2022 21:49:48 +0000 (22:49 +0100)]
Rollup merge of #91593 - upsuper-forks:hashmap-set-methods-bound, r=dtolnay
Remove unnecessary bounds for some Hash{Map,Set} methods
This PR moves `HashMap::{into_keys,into_values,retain}` and `HashSet::retain` from `impl` blocks with `K: Eq + Hash, S: BuildHasher` into the blocks without them. It doesn't seem to me there is any reason these methods need to be bounded by that. This change brings `HashMap::{into_keys,into_values}` on par with `HashMap::{keys,values,values_mut}` which are not bounded either.
Matthias Krüger [Sat, 1 Jan 2022 21:49:47 +0000 (22:49 +0100)]
Rollup merge of #84083 - ltratt:threadid_doc_tweak, r=dtolnay
Clarify the guarantees that ThreadId does and doesn't make.
The existing documentation does not spell out whether `ThreadId`s are unique during the lifetime of a thread or of a process. I had to examine the source code to realise (pleasingly!) that they're unique for the lifetime of a process. That seems worth documenting clearly, as it's a strong guarantee.
Examining the way `ThreadId`s are created also made me realise that the `as_u64` method on `ThreadId` could be a trap for the unwary on those platforms where the platform's notion of a thread identifier is also a 64 bit integer (particularly if they happen to use a similar identifier scheme to `ThreadId`). I therefore think it's worth being even clearer that there's no relationship between the two.
bors [Sat, 1 Jan 2022 20:45:37 +0000 (20:45 +0000)]
Auto merge of #92396 - xfix:remove-commandenv-apply, r=Mark-Simulacrum
Remove CommandEnv::apply
It's not being used and uses unsound set_var and remove_var functions. This is an internal function that isn't exported (even with `process_internals` feature), so this shouldn't break anything.
Also see #92365. Note that this isn't the only use of those methods in standard library, so that particular pull request will need more changes than just this to work (in particular, `test_capture_env_at_spawn` is using `set_var` and `remove_var`).
bjorn3 [Sun, 26 Dec 2021 14:35:50 +0000 (15:35 +0100)]
Make the rustc and rustdoc wrapper not depend on libbootstrap
This slightly improves compilation time by reducing linking time
(saving about a 1/10 of the the total compilation time after
changing rustbuild) and slightly reduces disk usage (from 16MB for
the rustc wrapper to 4MB).
bjorn3 [Sun, 26 Dec 2021 14:05:12 +0000 (15:05 +0100)]
Avoid the merge derive macro in rustbuild
The task of the macro is simple enough that a decl macro is almost ten
times shorter than the original proc macro. The proc macro is 159 lines
while the decl macro is just 18 lines.
This reduces the amount of dependencies of rustbuild from 45 to 37. It
also slight reduces compilation time from 47s to 44s for debug builds.
bors [Sat, 1 Jan 2022 13:28:13 +0000 (13:28 +0000)]
Auto merge of #92419 - erikdesjardins:coldland, r=nagisa
Mark drop calls in landing pads `cold` instead of `noinline`
Now that deferred inlining has been disabled in LLVM (#92110), this shouldn't cause catastrophic size blowup.
I confirmed that the test cases from https://github.com/rust-lang/rust/issues/41696#issuecomment-298696944 still compile quickly (<1s) after this change. ~Although note that I wasn't able to reproduce the original issue using a recent rustc/llvm with deferred inlining enabled, so those tests may no longer be representative. I was also unable to create a modified test case that reproduced the original issue.~ (edit: I reproduced it on CI by accident--the first commit timed out on the LLVM 12 builder, because I forgot to make it conditional on LLVM version)
r? `@nagisa`
cc `@arielb1` (this effectively reverts #42771 "mark calls in the unwind path as !noinline")
cc `@RalfJung` (fixes #46515)
bors [Sat, 1 Jan 2022 09:57:35 +0000 (09:57 +0000)]
Auto merge of #92471 - matthiaskrgr:rollup-lmduxwh, r=matthiaskrgr
Rollup of 7 pull requests
Successful merges:
- #88310 (Lock bootstrap (x.py) build directory)
- #92097 (Implement split_at_spare_mut without Deref to a slice so that the spare slice is valid)
- #92412 (Fix double space in pretty printed TryBlock)
- #92420 (Fix whitespace in pretty printed PatKind::Range)
- #92457 (Sync rustc_codegen_gcc)
- #92460 ([rustc_builtin_macros] add indices to format_foreign::printf::Substitution::Escape)
- #92469 (Make tidy check for magic numbers that spell things)
Matthias Krüger [Sat, 1 Jan 2022 09:48:57 +0000 (10:48 +0100)]
Rollup merge of #92460 - dwrensha:fix-92267, r=petrochenkov
[rustc_builtin_macros] add indices to format_foreign::printf::Substitution::Escape
Fixes #92267.
The problem was that the escape string "%%" does not need to appear at the very beginning of the format string, but
the iterator implementation assumed that it did.
The solution follows the pattern used by `format_foregin::shell::Subtitution::Escape`: https://github.com/rust-lang/rust/blob/8ed935e92dfb09ae388344b12284bf5110cf9265/compiler/rustc_builtin_macros/src/format_foreign.rs#L629
Matthias Krüger [Sat, 1 Jan 2022 09:48:54 +0000 (10:48 +0100)]
Rollup merge of #92097 - saethlin:split-without-deref, r=the8472
Implement split_at_spare_mut without Deref to a slice so that the spare slice is valid
~I'm not sure I understand what's going on here correctly. And I'm pretty sure this safety comment needs to be changed. I'm just referring to the same thing that `as_mut_ptr_range` does.~ (Thanks `@RalfJung` for the guidance and clearing things up)
I tried to run https://github.com/rust-lang/miri-test-libstd on alloc with -Zmiri-track-raw-pointers, and got a failure on the test `vec::test_extend_from_within`.
I minimized the test failure into this program:
```rust
#![feature(vec_split_at_spare)]
fn main() {
Vec::<i32>::with_capacity(1).split_at_spare_mut();
}
```
The problem is that the existing implementation is actually getting a pointer range where both pointers are derived from the initialized region of the Vec's allocation, but we need the second one to be valid for the region between len and capacity. (thanks Ralf for clearing this up)
bors [Sat, 1 Jan 2022 05:21:22 +0000 (05:21 +0000)]
Auto merge of #92374 - ehuss:update-libssh2-sys, r=Mark-Simulacrum
Update libssh2-sys
Updates libssh2-sys from 0.2.19 to 0.2.23. This brings in libssh2 1.10 ([RELEASE-NOTES](https://github.com/libssh2/libssh2/blob/libssh2-1.10.0/RELEASE-NOTES)). One of the major changes is to add support for OpenSSH agent on Windows.
Jakub Kądziołka [Sat, 1 Jan 2022 04:12:56 +0000 (05:12 +0100)]
Emit an error for `--cfg=)`
Fixes #73026
See also: #64467, #89468
The issue stems from a `FatalError` being silently raised in
`panictry_buffer`. Normally this is not a problem, because
`panictry_buffer` emits the causes of the error, but they are not
themselves fatal, so they get filtered out by the silent emitter.
To fix this, we use a parser entrypoint which doesn't use
`panictry_buffer`, and we handle the error ourselves.
bors [Sat, 1 Jan 2022 02:03:23 +0000 (02:03 +0000)]
Auto merge of #92294 - Kobzol:rustdoc-meta-kind, r=GuillaumeGomez
Add Attribute::meta_kind
The `AttrItem::meta` function is being called on a lot of places, however almost always the caller is only interested in the `kind` of the result `MetaItem`. Before, the `path` had to be cloned in order to get the kind, now it does not have to be.
There is a larger related "problem". In a lot of places, something wants to know contents of attributes. This is accessed through `Attribute::meta_item_list`, which calls `AttrItem::meta` (now `AttrItem::meta_kind`), among other methods. When this function is called, the meta item list has to be recreated from scratch. Everytime something asks a simple question (like is this item/list of attributes `#[doc(hidden)]`?), the tokens of the attribute(s) are cloned, parsed and the results are allocated on the heap. That seems really unnecessary. What would be the best way to cache this? Turn `meta_item_list` into a query perhaps? Related PR: https://github.com/rust-lang/rust/pull/92227
bors [Fri, 31 Dec 2021 22:57:51 +0000 (22:57 +0000)]
Auto merge of #92465 - matthiaskrgr:rollup-yuary84, r=matthiaskrgr
Rollup of 7 pull requests
Successful merges:
- #90383 (Extend check for UnsafeCell in consts to cover unions)
- #91375 (config.rs: Add support for a per-target default_linker option.)
- #91480 (rustdoc: use smaller number of colors to distinguish items)
- #92338 (Add try_reserve and try_reserve_exact for OsString)
- #92405 (Add a couple needs-asm-support headers to tests)
- #92435 (Sync rustc_codegen_cranelift)
- #92440 (Fix mobile toggles position)
Matthias Krüger [Fri, 31 Dec 2021 22:14:48 +0000 (23:14 +0100)]
Rollup merge of #92435 - bjorn3:sync_cg_clif-2021-12-30, r=bjorn3
Sync rustc_codegen_cranelift
The main highlight this sync is enforcing rustfmt and lack of warnings on cg_clif's CI. I will open a separate PR to remove the cg_clif exceptions for them from this repo.
Matthias Krüger [Fri, 31 Dec 2021 22:14:45 +0000 (23:14 +0100)]
Rollup merge of #91480 - jsha:fewer-colors, r=GuillaumeGomez
rustdoc: use smaller number of colors to distinguish items
This reduces visual distractions when reading method signatures.
As discussed in https://github.com/rust-lang/rust/issues/59845#issuecomment-974757191, this categorizes items into one of six colors (down from thirteen):
- method, function (ochre `#AD7C37`)
- trait, trait alias (dark slate blue `#6E4FC9`)
- enum, struct, type alias, union, primitive (maroon `#AD378A`)
- static, module, keyword, associated type, foreign type (steel blue `#3873AD`)
- macro (green `#068000`)
- generic params, self, Self (unmarked black `#000000`)
I slightly tweaked the actual color values so they'd have the same lightness (previously the trait color stood out much more than the others). And I made the color for links in general consistently use steel blue (previously there was a slightly different color for "search-failed").
The ayu and dark themes have been updated according to the same logic. I haven't changed any of the color values in those themes, just their assignment to types.
Matthias Krüger [Fri, 31 Dec 2021 22:14:44 +0000 (23:14 +0100)]
Rollup merge of #91375 - Apteryks:per-target-default-linker, r=Mark-Simulacrum
config.rs: Add support for a per-target default_linker option.
* src/bootstrap/config.rs (Target) <default_linker>: New field.
(TomlTarget): Likewise.
* src/bootstrap/compile.rs (rustc_cargo_env): Prefer a
target-specified default_linker over a global one if available.
* config.toml.example: Adjust doc.
bors [Fri, 31 Dec 2021 19:54:10 +0000 (19:54 +0000)]
Auto merge of #90637 - Mark-Simulacrum:liveness-btree, r=lqd
Store liveness in interval sets for region inference
On the 100,000 line test case from https://github.com/rust-lang/rust/issues/90445, this reduces memory usage from 35 GB to 444 MB at peak (based on DHAT results, though with regular malloc), and yields a 9.4x speedup, with wall time going from 14.5 seconds to 1.5s. Performance results show that for the majority of real-world code this has little to no impact, but it's expected to generally scale better for auto-generated functions and other cases which stress this area of the compiler, as results on #90445 illustrate.
There may also be further room for improvement in future PRs making use of this data structures benefits over raw bitsets (which, at some level, are a less perfect fit for representing liveness, which is almost always composed of contiguous ranges, not point locations).
bors [Fri, 31 Dec 2021 16:38:23 +0000 (16:38 +0000)]
Auto merge of #92175 - Aaron1011:fix-missing-source-file, r=cjgillot
Import `SourceFile`s from crate before decoding foreign `Span`
Fixes #92163
Fixes #92014
When writing to the incremental cache, we encode all `Span`s
we encounter, regardless of whether or not their `SourceFile`
comes from the local crate, or from a foreign crate.
When we decode a `Span`, we use the `StableSourceFileId` we encoded
to locate the matching `SourceFile` in the current session. If this
id corresponds to a `SourceFile` from another crate, then we need to
have already imported that `SourceFile` into our current session.
This usually happens automatically during resolution / macro expansion,
when we try to resolve definitions from other crates. In certain cases,
however, we may try to load a `Span` from a transitive dependency
without having ever imported the `SourceFile`s from that crate, leading
to an ICE.
This PR fixes the issue by enconding the `SourceFile`'s `CrateNum`
when we encode a `Span`. During decoding, we call `imported_source_files()`
when we encounter a foreign `CrateNum`, which ensure that all
`SourceFile`s from that crate are imported into the current session.
worldeva [Tue, 24 Aug 2021 20:45:46 +0000 (15:45 -0500)]
Lock x.py build state
Prevent spurious build failures and other bugs caused by parallel runs of
x.py. We back the lock with sqlite, so that we have a cross-platform locking
strategy, and which can be done nearly first in the build process (from Python),
which helps move the lock as early as possible.
Mark Rousskov [Fri, 5 Nov 2021 21:27:53 +0000 (17:27 -0400)]
Use SparseIntervalMatrix instead of SparseBitMatrix
Region inference contains several bitsets which are filled with large intervals
representing liveness. These can cause excessive memory usage, and are
relatively slow when growing to large sizes compared to the IntervalSet.
Mark Rousskov [Fri, 5 Nov 2021 18:50:29 +0000 (14:50 -0400)]
Introduce IntervalSet
This is a compact, fast storage for variable-sized sets, typically consisting of
larger ranges. It is less efficient than a bitset if ranges are both small and
the domain size is small, but will still perform acceptably. With enormous
domain sizes and large ranges, the interval set performs much better, as it can
be much more densely packed in memory than the uncompressed bit set alternative.
David Tolnay [Thu, 30 Dec 2021 20:40:51 +0000 (12:40 -0800)]
Fix some copy/paste hysteresis in OsString try_reserve docs
It appears `find_max_slow` comes from the BinaryHeap docs, where the
try_reserve example is a slow implementation of find_max. It has no
relevance to this code in OsString though.
Matthias Krüger [Thu, 30 Dec 2021 12:12:33 +0000 (13:12 +0100)]
Rollup merge of #91519 - petrochenkov:cratexp2, r=Aaron1011
ast: Avoid aborts on fatal errors thrown from mutable AST visitor
Set the node to some dummy value and rethrow the error instead.
When using the old aborting `visit_clobber` in `InvocationCollector::visit_crate` the next tests abort due to fatal errors:
```
ui\modules\path-invalid-form.rs
ui\modules\path-macro.rs
ui\modules\path-no-file-name.rs
ui\parser\issues\issue-5806.rs
ui\parser\mod_file_with_path_attr.rs
```
Follow up to https://github.com/rust-lang/rust/pull/91313.
bors [Thu, 30 Dec 2021 11:34:24 +0000 (11:34 +0000)]
Auto merge of #89336 - Aaron1011:variance-struct-diag, r=cjgillot
Refactor variance diagnostics to work with more types
Instead of special-casing mutable pointers/references, we
now support general generic types (currently, we handle
`ty::Ref`, `ty::RawPtr`, and `ty::Adt`)
When a `ty::Adt` is involved, we show an additional note
explaining which of the type's generic parameters is
invariant (e.g. the `T` in `Cell<T>`). Currently, we don't
explain *why* a particular generic parameter ends up becoming
invariant. In the general case, this could require printing
a long 'backtrace' of types, so doing this would be
more suitable for a follow-up PR.
We still only handle the case where our variance switches
to `ty::Invariant`.
bjorn3 [Thu, 30 Dec 2021 10:35:58 +0000 (11:35 +0100)]
Rename cargo executable to cargo-clif
This allows executing it like cargo clif build if you add it to your
PATH. It also fixes infinite recursion on Windows when invoking it as
Windows includes the current directory in PATH by default.
bors [Thu, 30 Dec 2021 02:53:09 +0000 (02:53 +0000)]
Auto merge of #91125 - eskarn:llvm-passes-plugin-support, r=nagisa
Allow loading LLVM plugins with both legacy and new pass manager
Opening a draft PR to get feedback and start discussion on this feature. There is already a codegen option `passes` which allow giving a list of LLVM pass names, however we currently can't use a LLVM pass plugin (as described here : https://llvm.org/docs/WritingAnLLVMPass.html), the only available passes are the LLVM built-in ones.
The proposed modification would be to add another codegen option `pass-plugins`, which can be set with a list of paths to shared library files. These libraries are loaded using the LLVM function `PassPlugin::Load`, which calls the expected symbol `lvmGetPassPluginInfo`, and register the pipeline parsing and optimization callbacks.
An example usage with a single plugin and 3 passes would look like this in the `.cargo/config`:
```toml
rustflags = [
"-C", "pass-plugins=/tmp/libLLVMPassPlugin",
"-C", "passes=pass1 pass2 pass3",
]
```
This would give the same functionality as the opt LLVM tool directly integrated in rust build system.
Additionally, we can also not specify the `passes` option, and use a plugin which inserts passes in the optimization pipeline, as one could do using clang.
Aaron Hill [Tue, 28 Sep 2021 01:22:44 +0000 (20:22 -0500)]
Refactor variance diagnostics to work with more types
Instead of special-casing mutable pointers/references, we
now support general generic types (currently, we handle
`ty::Ref`, `ty::RawPtr`, and `ty::Adt`)
When a `ty::Adt` is involved, we show an additional note
explaining which of the type's generic parameters is
invariant (e.g. the `T` in `Cell<T>`). Currently, we don't
explain *why* a particular generic parameter ends up becoming
invariant. In the general case, this could require printing
a long 'backtrace' of types, so doing this would be
more suitable for a follow-up PR.
We still only handle the case where our variance switches
to `ty::Invariant`.