add a test for target-feature-ABI warnings in closures and when calling extern fn
Also update the comment regarding the inheritance of target features into closures, to make it more clear that we really shouldn't do this right now.
Using `cc_args` panics when using `-Clinker-flavor=ld`, because the
arguments are in a form tailored for `-Clinker-flavor=gcc`.
So instead, we use `link_args` and let that wrap the arguments with the
appropriate `-Wl` or `-Xlinker` when needed.
[AIX] Add option -X32_64 to the "strip" command
The AIX `strip` utility requires option `-X` to specify the object mode. This patch adds the `-X32_64` option to the `strip` command so that it can handle both 32-bit and 64-bit objects. The parameter `option` of function `strip_symbols_with_external_utility`, previously a single string, has been changed to `options`, an array of string slices, to accommodate multiple `strip` options.
The deployment target environment variable is OS-specific, and if you're
in a place where you're asking `rustc` for the deployment target, you're
likely to also wanna know the environment variable.
[AIX] change system dynamic library format
Historically on AIX, almost all dynamic libraries are distributed in `.a` Big Archive Format which can consists of both static and shared objects in the same archive (e.g. `libc++abi.a(libc++abi.so.1)`). During the initial porting process, the dynamic libraries are kept as `.a` to simplify the migration, but semantically having an XCOFF object under the archive extension is wrong. For crate type `cdylib` we want to be able to distribute the libraries as archives as well.
We are migrating to archives with the following format:
```
$ ar -t lib<name>.a
lib<name>.so
```
where each archive contains a single member that is a shared XCOFF object that can be loaded.
Rollup of 6 pull requests
Successful merges:
- #129838 (uefi: process: Add args support)
- #130800 (Mark `get_mut` and `set_position` in `std::io::Cursor` as const.)
- #132708 (Point at `const` definition when used instead of a binding in a `let` statement)
- #133226 (Make `PointerLike` opt-in instead of built-in)
- #133244 (Account for `wasm32v1-none` when exporting TLS symbols)
- #133257 (Add `UnordMap::clear` method)
r? `@ghost`
`@rustbot` modify labels: rollup
take 2
open up coroutines
tweak the wordings
the lint works up until 2021
We were missing one case, for ADTs, which was
causing `Result` to yield incorrect results.
only include field spans with significant types
deduplicate and eliminate field spans
switch to emit spans to impl Drops
Co-authored-by: Niko Matsakis <nikomat@amazon.com>
collect drops instead of taking liveness diff
apply some suggestions and add explantory notes
small fix on the cache
let the query recurse through coroutine
new suggestion format with extracted variable name
fine-tune the drop span and messages
bugfix on runtime borrows
tweak message wording
filter out ecosystem types earlier
apply suggestions
clippy
check lint level at session level
further restrict applicability of the lint
translate bid into nop for stable mir
detect cycle in type structure
The maximum discriminator value LLVM can currently encode is 2^12. If macro use
results in more than 2^12 calls to the same function attributed to the same
callsite, and those calls are MIR-inlined, we will require more than the maximum
discriminator value to completely represent the debug information. Once we reach
that point drop the debug info instead.
the behavior of the type system not only depends on the current
assumptions, but also the currentnphase of the compiler. This is
mostly necessary as we need to decide whether and how to reveal
opaque types. We track this via the `TypingMode`.
Over in Zed we've noticed that loading crates for a large-ish workspace can take almost 200ms. We've pinned it down to how rustc searches for paths, as it performs a linear search over the list of candidate paths. In our case the candidate list had about 20k entries which we had to iterate over for each dependency being loaded.
This commit introduces a simple FilesIndex that's just a sorted Vec under the hood. Since crates are looked up by both prefix and suffix, we perform a range search on said Vec (which constraints the search space based on prefix) and follow up with a linear scan of entries with matching suffixes.
FilesIndex is also pre-filtered before any queries are performed using available target information; query prefixes/sufixes are based on the target we are compiling for, so we can remove entries that can never match up front.
Overall, this commit brings down build time for us in dev scenarios by about 6%.
100ms might not seem like much, but this is a constant cost that each of our workspace crates has to pay, even when said crate is miniscule.
CFI: Append debug location to CFI blocks
Currently we're not appending debug locations to the inserted CFI blocks. This shows up in #132615 and #100783. This change fixes that by passing down the debug location to the CFI type-test generation and appending it to the blocks.
Credits also belong to `@jakos-sec` who worked with me on this.
Add a default implementation for CodegenBackend::link
As a side effect this should add raw-dylib support to cg_gcc as the default ArchiveBuilderBuilder that is used implements create_dll_import_lib. I haven't tested if the raw-dylib support actually works however.
After link_binary the temporary files referenced by CodegenResults are
deleted, so calling link_binary again with the same CodegenResults
should not be allowed.
As a side effect this should add raw-dylib support to cg_gcc as the
default ArchiveBuilderBuilder that is used implements
create_dll_import_lib. I haven't tested if the raw-dylib support
actually works however.
Use lld with non-LLVM backends
On arm64, Cranelift used to produce object files that don't work with lld. This has since been fixed. The GCC backend should always produce object files that work with lld unless lld for whatever reason drops GCC support. Most of the other more niche backends don't use cg_ssa's linker code at all. If they do and don't work with lld, they can always disable lld usage using a cli argument.
Without this commit using cg_clif is by default in a non-trivial amount of cases a perf regression on Linux due to ld.bfd being a fair bit slower than lld. It is possible to explicitly enable it without this commit, but most users are unlikely to do this.
On arm64, Cranelift used to produce object files that don't work with
lld. This has since been fixed. The GCC backend should always produce
object files that work with lld unless lld for whatever reason drops GCC
support. Most of the other more niche backends don't use cg_ssa's linker
code at all. If they do and don't work with lld, they can always disable
lld usage using a cli argument.
Without this commit using cg_clif is by default in a non-trivial amount
of cases a perf regression on Linux due to ld.bfd being a fair bit
slower than lld. It is possible to explicitly enable it without this
commit, but most users are unlikely to do this.
Set "symbol name" in raw-dylib import libraries to the decorated name
`windows-rs` received a bug report that mixing raw-dylib generated and the Windows SDK import libraries was causing linker failures: <https://github.com/microsoft/windows-rs/issues/3285>
The root cause turned out to be #124958, that is we are not including the decorated name in the import library and so the import name type is also not being correctly set.
This change modifies the generation of import libraries to set the "symbol name" to the fully decorated name and correctly marks the import as being data vs function.
Note that this also required some changes to how the symbol is named within Rust: for MSVC we now need to use the decorated name but for MinGW we still need to use partially decorated (or undecorated) name.
Fixes#124958
Passing i686 MSVC and MinGW build: <https://github.com/rust-lang/rust/actions/runs/11000433888?pr=130586>
r? `@ChrisDenton`
bootstrap/codegen_ssa: ship llvm-strip and use it for -Cstrip
Fixes#131206.
- Includes `llvm-strip` (a symlink to `llvm-objcopy`) in the compiler dist artifact so that it can be used for `-Cstrip` instead of the system tooling.
- Uses `llvm-strip` instead of `/usr/bin/strip` for macOS. macOS needs a specific linker and the system one is preferred, hence #130781 but that doesn't work when cross-compiling, so use the `llvm-strip` utility instead.
cc #123151
mark some target features as 'forbidden' so they cannot be (un)set with -Ctarget-feature
The context for this is https://github.com/rust-lang/rust/issues/116344: some target features change the way floats are passed between functions. Changing those target features is unsound as code compiled for the same target may now use different ABIs.
So this introduces a new concept of "forbidden" target features (on top of the existing "stable " and "unstable" categories), and makes it a hard error to (un)set such a target feature. For now, the x86 and ARM feature `soft-float` is on that list. We'll have to make some effort to collect more relevant features, and similar features from other targets, but that can happen after the basic infrastructure for this landed. (These features are being collected in https://github.com/rust-lang/rust/issues/131799.)
I've made this a warning for now to give people some time to speak up if this would break something.
MCP: https://github.com/rust-lang/compiler-team/issues/780
Remove unnecessary pub enum glob-imports from `rustc_middle::ty`
We used to have an idiom in the compiler where we'd prefix or suffix all the variants of an enum, for example `BoundRegionKind`, with something like `Br`, and then *glob-import* that enum variant directly.
`@noratrieb` brought this up, and I think that it's easier to read when we just use the normal style `EnumName::Variant`.
This PR is a bit large, but it's just naming.
The only somewhat opinionated change that this PR does is rename `BorrowKind::Imm` to `BorrowKind::Immutable` and same for the other variants. I think these enums are used sparingly enough that the extra length is fine.
r? `@noratrieb` or reassign
Reduce dependence on the target name
The target name can be anything with custom target specs. Matching on fields inside the target spec is much more robust than matching on the target name.
Also remove the unused is_builtin target spec field.
Generate correct symbols.o for sparc-unknown-none-elf
This fixes#130172 by selecting the correct ELF Machine type for sparc-unknown-none-elf (which has a baseline of SPARC V7).
The target name can be anything with custom target specs. Matching on
fields inside the target spec is much more robust than matching on the
target name.
Move versioned Apple LLVM targets from `rustc_target` to `rustc_codegen_ssa`
Fully specified LLVM targets contain the OS version on macOS/iOS/tvOS/watchOS/visionOS, and this version depends on the deployment target environment variables like `MACOSX_DEPLOYMENT_TARGET`, `IPHONEOS_DEPLOYMENT_TARGET` etc.
We would like to move this to later in the compilation pipeline, both because it feels impure to access environment variables when fetching target information, but mostly because we need access to more information from https://github.com/rust-lang/rust/pull/130883 to do https://github.com/rust-lang/rust/issues/118204. See also https://github.com/rust-lang/rust/pull/129342#issuecomment-2335156119 for some discussion.
The first and second commit does the actual refactor, it should be a non-functional change, the third commit adds diagnostics for invalid deployment targets, which are now possible to do because we have access to the session.
Tested with the same commands as in https://github.com/rust-lang/rust/pull/130435.
r? ``````@petrochenkov``````
Remove support for `-Zprofile` (gcov-style coverage instrumentation)
Tracking issue: #42524
MCP: https://github.com/rust-lang/compiler-team/issues/798
---
This PR removes the unstable `-Zprofile` flag, which enables ”gcov-style” coverage instrumentation, along with its associated `-Zprofile-emit` configuration flag.
(The profile flag predates and is almost entirely separate from the stable `-Cinstrument-coverage` flag.)
Notably, the `-Zprofile` flag:
- Is largely untested in-tree, having only one run-make test that does not check whether its output is correct or useful.
- Has no known maintainer.
- Has seen no push towards stabilization.
- Has at least one severe regression reported in 2022 that apparently remains unaddressed.
- #100125
- Is confusingly named, since it appears to be more about coverage than performance profiling, and has nothing to do with PGO.
- Is fundamentally limited by relying on counters auto-inserted by LLVM, with no knowledge of Rust beyond debuginfo.
The OS version depends on the deployment target environment variables,
the access of which we want to move to later in the compilation pipeline
that has access to more information, for example `env_depinfo`.
Add `lp64e` RISC-V ABI
This PR adds support for the `lp64e` RISC-V ABI, which is the 64-bit equivalent of the `ilp32e` ABI that is already supported.
For reference, this ABI was originally added to LLVM in [this PR](https://reviews.llvm.org/D70401).
Rename `rustc_abi::Abi` to `BackendRepr`
Remove the confabulation of `rustc_abi::Abi` with what "ABI" actually means by renaming it to `BackendRepr`, and rename `Abi::Aggregate` to `BackendRepr::Memory`. The type never actually represented how things are passed, as that has to have `PassMode` considered, at minimum, but rather it just is how we represented some things to the backend. This conflation arose because LLVM, the primary backend at the time, would lower certain IR forms using certain ABIs. Even that only somewhat was true, as it broke down when one ventured significantly afield of what is described by the System V AMD64 ABI either by using different architectures, ABI-modifying IR annotations, the same architecture **with different ISA extensions enabled**, or other... unexpected delights.
Unfortunately both names are still somewhat of a misnomer right now, as people have written code for years based on this misunderstanding. Still, their original names are even moreso, and for better or worse, this backend code hasn't received as much maintenance as the rest of the compiler, lately. Actually arriving at a correct end-state will simply require us to disentangle a lot of code in order to fix, much of it pointlessly repeated in several places. Thus this is not an "actual fix", just a way to deflect further misunderstandings.
The initial naming of "Abi" was an awful mistake, conveying wrong ideas
about how psABIs worked and even more about what the enum meant.
It was only meant to represent the way the value would be described to
a codegen backend as it was lowered to that intermediate representation.
It was never meant to mean anything about the actual psABI handling!
The conflation is because LLVM typically will associate a certain form
with a certain ABI, but even that does not hold when the special cases
that actually exist arise, plus the IR annotations that modify the ABI.
Reframe `rustc_abi::Abi` as the `BackendRepr` of the type, and rename
`BackendRepr::Aggregate` as `BackendRepr::Memory`. Unfortunately, due to
the persistent misunderstandings, this too is now incorrect:
- Scattered ABI-relevant code is entangled with BackendRepr
- We do not always pre-compute a correct BackendRepr that reflects how
we "actually" want this value to be handled, so we leave the backend
interface to also inject various special-cases here
- In some cases `BackendRepr::Memory` is a "real" aggregate, but in
others it is in fact using memory, and in some cases it is a scalar!
Our rustc-to-backend lowering code handles this sort of thing right now.
That will eventually be addressed by lifting duplicated lowering code
to either rustc_codegen_ssa or rustc_target as appropriate.
- removed extra bits from predicates queries that are no longer needed in the new system
- removed the need for `non_erasable_generics` to take in tcx and DefId, removed unused arguments in callers
"innermost", "outermost", "leftmost", and "rightmost" don't need hyphens
These are all standard dictionary words and don't require hyphenation.
-----
Encountered an instance of this in error messages and it bugged me, so I
figured I'd fix it across the entire codebase.
stabilize Strict Provenance and Exposed Provenance APIs
Given that [RFC 3559](https://rust-lang.github.io/rfcs/3559-rust-has-provenance.html) has been accepted, t-lang has approved the concept of provenance to exist in the language. So I think it's time that we stabilize the strict provenance and exposed provenance APIs, and discuss provenance explicitly in the docs:
```rust
// core::ptr
pub const fn without_provenance<T>(addr: usize) -> *const T;
pub const fn dangling<T>() -> *const T;
pub const fn without_provenance_mut<T>(addr: usize) -> *mut T;
pub const fn dangling_mut<T>() -> *mut T;
pub fn with_exposed_provenance<T>(addr: usize) -> *const T;
pub fn with_exposed_provenance_mut<T>(addr: usize) -> *mut T;
impl<T: ?Sized> *const T {
pub fn addr(self) -> usize;
pub fn expose_provenance(self) -> usize;
pub fn with_addr(self, addr: usize) -> Self;
pub fn map_addr(self, f: impl FnOnce(usize) -> usize) -> Self;
}
impl<T: ?Sized> *mut T {
pub fn addr(self) -> usize;
pub fn expose_provenance(self) -> usize;
pub fn with_addr(self, addr: usize) -> Self;
pub fn map_addr(self, f: impl FnOnce(usize) -> usize) -> Self;
}
impl<T: ?Sized> NonNull<T> {
pub fn addr(self) -> NonZero<usize>;
pub fn with_addr(self, addr: NonZero<usize>) -> Self;
pub fn map_addr(self, f: impl FnOnce(NonZero<usize>) -> NonZero<usize>) -> Self;
}
```
I also did a pass over the docs to adjust them, because this is no longer an "experiment". The `ptr` docs now discuss the concept of provenance in general, and then they go into the two families of APIs for dealing with provenance: Strict Provenance and Exposed Provenance. I removed the discussion of how pointers also have an associated "address space" -- that is not actually tracked in the pointer value, it is tracked in the type, so IMO it just distracts from the core point of provenance. I also adjusted the docs for `with_exposed_provenance` to make it clear that we cannot guarantee much about this function, it's all best-effort.
There are two unstable lints associated with the strict_provenance feature gate; I moved them to a new [strict_provenance_lints](https://github.com/rust-lang/rust/issues/130351) feature since I didn't want this PR to have an even bigger FCP. ;)
`@rust-lang/opsem` Would be great to get some feedback on the docs here. :)
Nominating for `@rust-lang/libs-api.`
Part of https://github.com/rust-lang/rust/issues/95228.
[FCP comment](https://github.com/rust-lang/rust/pull/130350#issuecomment-2395114536)
Create `_imp__` symbols also when doing ThinLTO
When generating a rlib crate on Windows we create `dllimport` / `_imp__` symbols for each global. This effectively makes the rlib contain an import library for itself and allows them to both be dynamically and statically linked. However when doing ThinLTO we do not generate these and thus we end up with missing symbols. Microsoft's `link` can fix these up (and emits warnings), but `lld` seems to currently be unable to.
This PR also does this generation for ThinLTO avoiding those issues with `lld` and also avoids the warnings on `link`.
This is an workaround for https://github.com/rust-lang/rust/issues/81408.
cc `@lqd`
codegen_ssa: consolidate tied target checks
Fixes#105110.
Fixes#105111.
`rustc_codegen_llvm` and `rustc_codegen_gcc` duplicated logic for checking if tied target features were partially enabled. This PR consolidates these checks into `rustc_codegen_ssa` in the `codegen_fn_attrs` query, which also is run pre-monomorphisation for each function, which ensures that this check is run for unused functions, as would be expected.
Also adds a test confirming that enabling one tied feature doesn't imply another - the appropriate error for this was already being emitted. I did a bisect and narrowed it down to two patches it was likely to be - something in #128796, probably #128221 or #128679.
- fix for divergence
- fix error message
- fix another cranelift test
- fix some cranelift things
- don't set the NORETURN option for naked asm
- fix use of naked_asm! in doc comment
- fix use of naked_asm! in run-make test
- use `span_bug` in unreachable branch
Implement RFC3695 Allow boolean literals as cfg predicates
This PR implements https://github.com/rust-lang/rfcs/pull/3695: allow boolean literals as cfg predicates, i.e. `cfg(true)` and `cfg(false)`.
r? `@nnethercote` *(or anyone with parser knowledge)*
cc `@clubby789`
Apple: Do not specify an SDK version in `rlib` object files
This was added in https://github.com/rust-lang/rust/pull/114114, but is unnecessary, since it ends up being overwritten when linking anyhow, and it feels wrong to embed some arbitrary SDK version in here. The object files produced by LLVM also do not set this, and the tooling shows `n/a` when it's `0`, so it seems to genuinely be optional in object files.
I've also added a test for the different places the SDK version shows up, and documented a bit more in the code how SDK versions work.
See https://github.com/rust-lang/rust/issues/129432 for the bigger picture.
Tested with (excludes the same few targets as in https://github.com/rust-lang/rust/pull/130435):
```console
./x test tests/run-make/apple-sdk-version --target aarch64-apple-darwin,aarch64-apple-ios,aarch64-apple-ios-macabi,aarch64-apple-ios-sim,aarch64-apple-tvos,aarch64-apple-tvos-sim,aarch64-apple-visionos,aarch64-apple-visionos-sim,aarch64-apple-watchos,aarch64-apple-watchos-sim,arm64_32-apple-watchos,armv7k-apple-watchos,armv7s-apple-ios,x86_64-apple-darwin,x86_64-apple-ios,x86_64-apple-ios-macabi,x86_64-apple-tvos,x86_64-apple-watchos-sim,x86_64h-apple-darwin
IPHONEOS_DEPLOYMENT_TARGET=10.0 ./x test tests/run-make/apple-sdk-version --target=i386-apple-ios
```
CC `@BlackHoleFox,` you [originally commented on these values](https://github.com/rust-lang/rust/pull/114114#discussion_r1300599445).
`@rustbot` label O-apple
Relax a debug assertion for dyn principal *equality* in codegen
Maybe this sucks and I should just bite the bullet and use `infcx.sub` here. Thoughts?
r? lcnr
Fixes#130855
Only add an automatic SONAME for Rust dylibs
#126094 added an automatic relative `SONAME` to all dynamic libraries, but it was really only needed for Rust `--crate-type="dylib"`. In Fedora, it was a surprise to see `SONAME` on `"cdylib"` libraries like Python modules, especially because that generates an undesirable RPM `Provides`. We can instead add a `SONAME` just for Rust dylibs by passing the crate-type argument farther.
Ref: https://bugzilla.redhat.com/show_bug.cgi?id=2314879
Reorder stack spills so that constants come later.
Currently constants are "pulled forward" and have their stack spills emitted first. This confuses LLVM as to where to place breakpoints at function entry, and results in argument values being wrong in the debugger. It's straightforward to avoid emitting the stack spills for constants until arguments/etc have been introduced in debug_introduce_locals, so do that.
Example LLVM IR (irrelevant IR elided):
Before:
```
define internal void `@_ZN11rust_1289457binding17h2c78f956ba4bd2c3E(i64` %a, i64 %b, double %c) unnamed_addr #0 !dbg !178 { start:
%c.dbg.spill = alloca [8 x i8], align 8
%b.dbg.spill = alloca [8 x i8], align 8
%a.dbg.spill = alloca [8 x i8], align 8
%x.dbg.spill = alloca [4 x i8], align 4
store i32 0, ptr %x.dbg.spill, align 4, !dbg !192 ; LLVM places breakpoint here.
#dbg_declare(ptr %x.dbg.spill, !190, !DIExpression(), !192)
store i64 %a, ptr %a.dbg.spill, align 8
#dbg_declare(ptr %a.dbg.spill, !187, !DIExpression(), !193)
store i64 %b, ptr %b.dbg.spill, align 8
#dbg_declare(ptr %b.dbg.spill, !188, !DIExpression(), !194)
store double %c, ptr %c.dbg.spill, align 8
#dbg_declare(ptr %c.dbg.spill, !189, !DIExpression(), !195)
ret void, !dbg !196
}
```
After:
```
define internal void `@_ZN11rust_1289457binding17h2c78f956ba4bd2c3E(i64` %a, i64 %b, double %c) unnamed_addr #0 !dbg !178 { start:
%x.dbg.spill = alloca [4 x i8], align 4
%c.dbg.spill = alloca [8 x i8], align 8
%b.dbg.spill = alloca [8 x i8], align 8
%a.dbg.spill = alloca [8 x i8], align 8
store i64 %a, ptr %a.dbg.spill, align 8
#dbg_declare(ptr %a.dbg.spill, !187, !DIExpression(), !192)
store i64 %b, ptr %b.dbg.spill, align 8
#dbg_declare(ptr %b.dbg.spill, !188, !DIExpression(), !193)
store double %c, ptr %c.dbg.spill, align 8
#dbg_declare(ptr %c.dbg.spill, !189, !DIExpression(), !194)
store i32 0, ptr %x.dbg.spill, align 4, !dbg !195 ; LLVM places breakpoint here.
#dbg_declare(ptr %x.dbg.spill, !190, !DIExpression(), !195)
ret void, !dbg !196
}
```
Note in particular the position of the "LLVM places breakpoint here" comment relative to the stack spills for the function arguments. LLVM assumes that the first instruction with with a debug location is the end of the prologue. As LLVM does not currently offer front ends any direct control over the placement of the prologue end reordering the IR is the only mechanism available to fix argument values at function entry in the presence of MIR optimizations like SingleUseConsts. Fixes#128945
r? `@michaelwoerister`
Fix up setting strip = true in Cargo.toml makes build scripts fail in…
Fix issue: https://github.com/rust-lang/rust/issues/110536
Strip binary is PATH dependent which breaks builds in MacOS.
For example, on my Mac, the output of 'which strip' is '/opt/homebrew/opt/binutils/bin/strip', which leads to incorrect 'strip' results. Therefore, just like on other systems, it is also necessary to specify 'stripcmd' on macOS. However, it seems that there is a bug in binutils [bugzilla-Bug 31571](https://sourceware.org/bugzilla/show_bug.cgi?id=31571), which leads to the problem mentioned above.
Rollup of 6 pull requests
Successful merges:
- #130549 (Add RISC-V vxworks targets)
- #130595 (Initial std library support for NuttX)
- #130734 (Fix: ices on virtual-function-elimination about principal trait)
- #130787 (Ban combination of GCE and new solver)
- #130809 (Update llvm triple for OpenHarmony targets)
- #130810 (Don't trap into the debugger on panics under Linux)
r? `@ghost`
`@rustbot` modify labels: rollup
Fix: ices on virtual-function-elimination about principal trait
Extract `load_vtable` function to ensure the `virtual_function_elimination` option is always checked.
It's okay not to use `llvm.type.checked.load` to load the vtable if there is no principal trait.
Fixes#123955Fixes#124092
Add `File` constructors that return files wrapped with a buffer
In addition to the light convenience, these are intended to raise visibility that buffering is something you should consider when opening a file, since unbuffered I/O is a common performance footgun to Rust newcomers.
ACP: https://github.com/rust-lang/libs-team/issues/446
Tracking Issue: #130804
`rustc_codegen_llvm` and `rustc_codegen_gcc` duplicated logic for
checking if tied target features were partially enabled. This commit
consolidates these checks into `rustc_codegen_ssa` in the
`codegen_fn_attrs` query, which also is run pre-monomorphisation for
each function, which ensures that this check is run for unused functions,
as would be expected.
llvm: replace some deprecated functions
`LLVMMDStringInContext` and `LLVMMDNodeInContext` are deprecated, replace them with `LLVMMDStringInContext2` and `LLVMMDNodeInContext2`.
Also replace `Value` with `Metadata` in some function signatures for better consistency.
add `extern "C-cmse-nonsecure-entry" fn`
tracking issue #75835
in https://github.com/rust-lang/rust/issues/75835#issuecomment-1183517255 it was decided that using an abi, rather than an attribute, was the right way to go for this feature.
This PR adds that ABI and removes the `#[cmse_nonsecure_entry]` attribute. All relevant tests have been updated, some are now obsolete and have been removed.
Error 0775 is no longer generated. It contains the list of targets that support the CMSE feature, and maybe we want to still use this? right now a generic "this abi is not supported on this platform" error is returned when this abi is used on an unsupported platform. On the other hand, users of this abi are likely to be experienced rust users, so maybe the generic error is good enough.
Don't alloca for unused locals
We already have a concept of mono-unreachable basic blocks; this is primarily useful for ensuring that we do not compile code under an `if false`. But since we never gave locals the same analysis, a large local only used under an `if false` will still have stack space allocated for it.
There are 3 places we traverse MIR during monomorphization: Inside the collector, `non_ssa_locals`, and the walk to generate code. Unfortunately, https://github.com/rust-lang/rust/pull/129283#issuecomment-2297925578 indicates that we cannot afford the expense of tracking reachable locals during the collector's traversal, so we do need at least two mono-reachable traversals. And of course caching is of no help here because the benchmarks that regress are incr-unchanged; they don't do any codegen.
This fixes the second problem in https://github.com/rust-lang/rust/issues/129282, and brings us anther step toward `const if` at home.
It's crazy to have the integer methods in something close to random
order.
The reordering makes the gaps clear: `const_i64`, `const_i128`,
`const_isize`, and `const_u16`. I guess they just aren't needed.
Implement a Method to Seal `DiagInner`'s Suggestions
This PR adds a method on `DiagInner` called `.seal_suggestions()` to prevent new suggestions from being added while preserving existing suggestions.
This is useful because currently there is no way to prevent new suggestions from being added to a diagnostic. `.disable_suggestions()` is the closest but it gets rid of all suggestions before and after the call.
Therefore, `.seal_suggestions()` can be used when, for example, misspelled keyword is detected and reported. In such cases, we may want to prevent other suggestions from being added to the diagnostic, as they would likely be meaningless once the misspelled keyword is identified. For context: https://github.com/rust-lang/rust/pull/129899#discussion_r1741307132
To store an additional state, the type of the `suggestions` field in `DiagInner` was changed into a three variant enum. While this change affects files across different crates, care was taken to preserve the existing code's semantics. This is validated by the fact that all UI tests pass without any modifications.
r? chenyukang
Currently constants are "pulled forward" and have their stack spills emitted
first. This confuses LLVM as to where to place breakpoints at function
entry, and results in argument values being wrong in the debugger. It's
straightforward to avoid emitting the stack spills for constants until
arguments/etc have been introduced in debug_introduce_locals, so do that.
Example LLVM IR (irrelevant IR elided):
Before:
define internal void @_ZN11rust_1289457binding17h2c78f956ba4bd2c3E(i64 %a, i64 %b, double %c) unnamed_addr #0 !dbg !178 {
start:
%c.dbg.spill = alloca [8 x i8], align 8
%b.dbg.spill = alloca [8 x i8], align 8
%a.dbg.spill = alloca [8 x i8], align 8
%x.dbg.spill = alloca [4 x i8], align 4
store i32 0, ptr %x.dbg.spill, align 4, !dbg !192 ; LLVM places breakpoint here.
#dbg_declare(ptr %x.dbg.spill, !190, !DIExpression(), !192)
store i64 %a, ptr %a.dbg.spill, align 8
#dbg_declare(ptr %a.dbg.spill, !187, !DIExpression(), !193)
store i64 %b, ptr %b.dbg.spill, align 8
#dbg_declare(ptr %b.dbg.spill, !188, !DIExpression(), !194)
store double %c, ptr %c.dbg.spill, align 8
#dbg_declare(ptr %c.dbg.spill, !189, !DIExpression(), !195)
ret void, !dbg !196
}
After:
define internal void @_ZN11rust_1289457binding17h2c78f956ba4bd2c3E(i64 %a, i64 %b, double %c) unnamed_addr #0 !dbg !178 {
start:
%x.dbg.spill = alloca [4 x i8], align 4
%c.dbg.spill = alloca [8 x i8], align 8
%b.dbg.spill = alloca [8 x i8], align 8
%a.dbg.spill = alloca [8 x i8], align 8
store i64 %a, ptr %a.dbg.spill, align 8
#dbg_declare(ptr %a.dbg.spill, !187, !DIExpression(), !192)
store i64 %b, ptr %b.dbg.spill, align 8
#dbg_declare(ptr %b.dbg.spill, !188, !DIExpression(), !193)
store double %c, ptr %c.dbg.spill, align 8
#dbg_declare(ptr %c.dbg.spill, !189, !DIExpression(), !194)
store i32 0, ptr %x.dbg.spill, align 4, !dbg !195 ; LLVM places breakpoint here.
#dbg_declare(ptr %x.dbg.spill, !190, !DIExpression(), !195)
ret void, !dbg !196
}
Note in particular the position of the "LLVM places breakpoint here" comment
relative to the stack spills for the function arguments. LLVM assumes that
the first instruction with with a debug location is the end of the prologue.
As LLVM does not currently offer front ends any direct control over the
placement of the prologue end reordering the IR is the only mechanism available
to fix argument values at function entry in the presence of MIR optimizations
like SingleUseConsts. Fixes#128945
`'mir` is not a good lifetime name in `LocalAnalyzer`, because it's used
on two unrelated fields. `'a` is more standard for a situation like this
(e.g. #130022).
Supertraits of `BuilderMethods` are all called `XyzBuilderMethods`.
Supertraits of `CodegenMethods` are all called `XyzMethods`. This commit
changes the latter to `XyzCodegenMethods`, for consistency.
It's a trait that aggregates five other traits. But consider the places
that use it.
- `BuilderMethods`: requires three of the five traits.
- `CodegenMethods`: requires zero(!) of the five traits.
- `BaseTypeMethods`: requires two of the five traits.
- `LayoutTypeMethods`: requires three of the five traits.
- `TypeMembershipMethods`: requires one of the five traits.
This commit just removes it, which makes everything simpler.
It has `Backend` and `Deref` boudns, plus an associated type
`CodegenCx`, and it has a single use. This commit "inlines" it into
`BuilderMethods`, which makes the complicated backend trait situation a
little simpler.
It only needs `Self::Value` and `Self::Type`, so it can be a subtrait of
`BackendTypes`. That is a smaller and simpler trait than `HasCodegen`
(which includes `BackendTypes` and a lot more).
linker: Allow MSVC to use import libraries following the Meson/MinGW convention
Hi all,
This PR implements support for `MsvcLinker` to use import libraries following Meson and the MinGW toolchain's naming convention. Meson [follows the `libfoo.dll.a` naming convention](https://mesonbuild.com/FAQ.html#why-does-building-my-project-with-msvc-output-static-libraries-called-libfooa) to disambiguate between static and import libraries.
This support already existed for static libraries (see #100101), but not for dynamic libraries. The latter case was added by duplicating the logic in `native_libs::find_native_static_library`, but a separate case was added in `link_dylib_by_name` for the Windows CRT libraries which must be handled by the linker itself.
See for prerequisites #129366, #126094, and #128370.
All feedback is appreciated!
Fixes#122455
cc `@sdroege` `@nirbheek`
simd_shuffle: require index argument to be a vector
Remove some codegen hacks by forcing the SIMD shuffle `index` argument to be a vector, which means (thanks to https://github.com/rust-lang/rust/pull/128537) that it will automatically be passed as an immediate in LLVM. The only special-casing we still have is for the extra sanity-checks we add that ensure that the indices are all in-bounds. (And the GCC backend needs to do a bunch of work since the Rust intrinsic is modeled after what LLVM expects, which seems to be quite different from what GCC expects.)
Fixes https://github.com/rust-lang/rust/issues/128738, see that issue for more context.
Don't leave debug locations for constants sitting on the builder indefinitely
Because constants are currently emitted *before* the prologue, leaving the debug location on the IRBuilder spills onto other instructions in the prologue and messes up both line numbers as well as the point LLVM chooses to be the prologue end.
Example LLVM IR (irrelevant IR elided):
Before:
```
define internal { i64, i64 } `@_ZN3tmp3Foo18var_return_opt_try17he02116165b0fc08cE(ptr` align 8 %self) !dbg !347 { start:
%self.dbg.spill = alloca [8 x i8], align 8
%_0 = alloca [16 x i8], align 8
%residual.dbg.spill = alloca [0 x i8], align 1
#dbg_declare(ptr %residual.dbg.spill, !353, !DIExpression(), !357)
store ptr %self, ptr %self.dbg.spill, align 8, !dbg !357
#dbg_declare(ptr %self.dbg.spill, !350, !DIExpression(), !358)
```
After:
```
define internal { i64, i64 } `@_ZN3tmp3Foo18var_return_opt_try17h00b17d08874ddd90E(ptr` align 8 %self) !dbg !347 { start:
%self.dbg.spill = alloca [8 x i8], align 8
%_0 = alloca [16 x i8], align 8
%residual.dbg.spill = alloca [0 x i8], align 1
#dbg_declare(ptr %residual.dbg.spill, !353, !DIExpression(), !357)
store ptr %self, ptr %self.dbg.spill, align 8
#dbg_declare(ptr %self.dbg.spill, !350, !DIExpression(), !358)
```
Note in particular how !357 from %residual.dbg.spill's dbg_declare no longer falls through onto the store to %self.dbg.spill. This fixes argument values at entry when the constant is a ZST (e.g. `<Option as Try>::Residual`). This fixes#130003 (but note that it does *not* fix issues with argument values and non-ZST constants, which emit their own stores that have debug info on them, like #128945).
r? `@michaelwoerister`
Simplify some nested `if` statements
Applies some but not all instances of `clippy::collapsible_if`. Some ended up looking worse afterwards, though, so I left those out. Also applies instances of `clippy::collapsible_else_if`
Review with whitespace disabled please.
Apple: Refactor deployment target version parsing
Refactor deployment target parsing to make it easier to do https://github.com/rust-lang/rust/pull/129342 (I wanted to make sure of all the places that `std::env::var` is called).
Specifically, my goal was to minimize the amount of target-specific configuration, so to that end I renamed the `opts` function that generates the `TargetOptions` to `base`, and made it return the LLVM target and `target_arch` too. In the future, I would like to move even more out of the target files and into `spec::apple`, as it makes it easier for me to maintain.
For example, this fixed a bug in `aarch64-apple-watchos`, which wasn't passing the deployment target as part of the LLVM triple. This (probably) fixes https://github.com/rust-lang/rust/issues/123582 and fixes https://github.com/rust-lang/rust/issues/107630.
We also now parse the patch version of deployment targets, allowing the user to specify e.g. `MACOSX_DEPLOYMENT_TARGET=10.12.6`.
Finally, this fixes the LLVM target name for visionOS, it should be `*-apple-xros` and not `*-apple-visionos`.
Since I have changed all the Apple targets here, I smoke-tested my changes by running the following:
```console
# Build each target
./x build library --target="aarch64-apple-darwin,aarch64-apple-ios,aarch64-apple-ios-macabi,aarch64-apple-ios-sim,aarch64-apple-tvos,aarch64-apple-tvos-sim,aarch64-apple-visionos,aarch64-apple-visionos-sim,aarch64-apple-watchos,aarch64-apple-watchos-sim,arm64_32-apple-watchos,arm64e-apple-ios,armv7k-apple-watchos,armv7s-apple-ios,i386-apple-ios,x86_64-apple-darwin,x86_64-apple-ios,x86_64-apple-ios-macabi,x86_64-apple-tvos,x86_64-apple-watchos-sim,x86_64h-apple-darwin"
# Test that we can still at least link basic projects
cargo new foobar && cd foobar && cargo +stage1 build --target=aarch64-apple-darwin --target=aarch64-apple-ios --target=aarch64-apple-ios-macabi --target=aarch64-apple-ios-sim --target=aarch64-apple-tvos --target=aarch64-apple-tvos-sim --target=aarch64-apple-visionos --target=aarch64-apple-visionos-sim --target=aarch64-apple-watchos --target=aarch64-apple-watchos-sim --target=arm64_32-apple-watchos --target=armv7s-apple-ios --target=i386-apple-ios --target=x86_64-apple-darwin --target=x86_64-apple-ios --target=x86_64-apple-ios-macabi --target=x86_64-apple-tvos --target=x86_64-apple-watchos-sim --target=x86_64h-apple-darwin
```
I couldn't build for the `arm64e-apple-darwin` target, the `armv7k-apple-watchos` and `arm64e-apple-ios` targets failed to link, and I know that the `i686-apple-darwin` target requires a bit of setup, but all of this is as it was before this PR.
r? thomcc
CC `@BlackHoleFox`
I would recommend using `rollup=never` when merging this, in case we need to bisect this later.
Because constants are currently emitted *before* the prologue, leaving the
debug location on the IRBuilder spills onto other instructions in the prologue
and messes up both line numbers as well as the point LLVM chooses to be the
prologue end.
Example LLVM IR (irrelevant IR elided):
Before:
define internal { i64, i64 } @_ZN3tmp3Foo18var_return_opt_try17he02116165b0fc08cE(ptr align 8 %self) !dbg !347 {
start:
%self.dbg.spill = alloca [8 x i8], align 8
%_0 = alloca [16 x i8], align 8
%residual.dbg.spill = alloca [0 x i8], align 1
#dbg_declare(ptr %residual.dbg.spill, !353, !DIExpression(), !357)
store ptr %self, ptr %self.dbg.spill, align 8, !dbg !357
#dbg_declare(ptr %self.dbg.spill, !350, !DIExpression(), !358)
After:
define internal { i64, i64 } @_ZN3tmp3Foo18var_return_opt_try17h00b17d08874ddd90E(ptr align 8 %self) !dbg !347 {
start:
%self.dbg.spill = alloca [8 x i8], align 8
%_0 = alloca [16 x i8], align 8
%residual.dbg.spill = alloca [0 x i8], align 1
#dbg_declare(ptr %residual.dbg.spill, !353, !DIExpression(), !357)
store ptr %self, ptr %self.dbg.spill, align 8
#dbg_declare(ptr %self.dbg.spill, !350, !DIExpression(), !358)
Note in particular how !357 from %residual.dbg.spill's dbg_declare no longer
falls through onto the store to %self.dbg.spill. This fixes argument values
at entry when the constant is a ZST (e.g. <Option as Try>::Residual). This
fixes#130003 (but note that it does *not* fix issues with argument values and
non-ZST constants, which emit their own stores that have debug info on them,
like #128945).
- Merge minimum OS version list into one function (makes it easier to
see the logic in it).
- Parse patch deployment target versions.
- Consistently specify deployment target in LLVM target (previously
omitted on `aarch64-apple-watchos`).
Update the `wasm-component-ld` binary dependency
This keeps it up-to-date by moving from 0.5.6 to 0.5.7. While here I've additionally updated some other wasm-related dependencies in the workspace to keep them up-to-date and try to avoid duplicate versions as well.
linker: Synchronize native library search in rustc and linker
Also search for static libraries with alternative naming (`libname.a`) on MSVC when producing executables or dynamic libraries, and not just rlibs.
This unblocks https://github.com/rust-lang/rust/pull/123436.
try-job: x86_64-msvc
This keeps it up-to-date by moving from 0.5.6 to 0.5.7. While here I've
additionally updated some other wasm-related dependencies in the
workspace to keep them up-to-date and try to avoid duplicate versions as
well.
rustc: Simplify getting sysroot library directory
It was very non-obvious that `sess.target_tlib_path`, `make_target_lib_path(...)`, and `sess.target_filesearch(...).search_paths()` result in the same sysroot library directory paths.
They are however, indeed the same, because `sess.target_tlib_path` is initialized to `make_target_lib_path(...)` on `Session` creation, and they are used interchangeably.
There are still some redundant calls to `make_target_lib_path` and other inconsistent ways to obtain sysroot directories, but fixing that requires some behavior changes, while this PR is a pure refactoring.
Some places in the compiler even disagree on the number of sysroots - 1 (explicit `--sysroot` *or* default sysroot), 2 (explicit `--sysroot` *and* default sysroot), or an unclear number of `sysroot_candidates` every of which is considered.
The logic currently using `sess.target_tlib_path` or equivalents assumes one sysroot.
Add `#[warn(unreachable_pub)]` to a bunch of compiler crates
By default `unreachable_pub` identifies things that need not be `pub` and tells you to make them `pub(crate)`. But sometimes those things don't need any kind of visibility. So they way I did these was to remove the visibility entirely for each thing the lint identifies, and then add `pub(crate)` back in everywhere the compiler said it was necessary. (Or occasionally `pub(super)` when context suggested that was appropriate.) Tedious, but results in more `pub` removal.
There are plenty more crates to do but this seems like enough for a first PR.
r? `@compiler-errors`
Allow rust staticlib to work with MSVC's /WHOLEARCHIVE
This fixes#129020 by renaming the `__NULL_IMPORT_DESCRIPTOR` to prevent conflicts.
try-job: dist-i686-msvc
Switch to using the v2 resolver in most workspaces
Pinning the resolver to v1 was done in 5abff3753a ("Explicit set workspace.resolver ...") in order to suppress warnings. Since there is no specific reason not to use the new resolver and since it fixes issues, change to `resolver = "2"` everywhere except library.
With the new resolver, a few dependencies get brought in twice with
different licenses. For example, all dependencies from `wasm-tools`
gained Apache-2.0 and MIT options, and with the v2 resolver we were
using one version from before and one version from after this change.
This made tidy's license check difficult.
Update some minimum versions to remove duplicate dependencies and smooth
out license checking.
Rework MIR inlining debuginfo so function parameters show up in debuggers.
Line numbers of multiply-inlined functions were fixed in #114643 by using a single DISubprogram. That, however, triggered assertions because parameters weren't deduplicated. The "solution" to that in #115417 was to insert a DILexicalScope below the DISubprogram and parent all of the parameters to that scope. That fixed the assertion, but debuggers (including gdb and lldb) don't recognize variables that are not parented to the subprogram itself as parameters, even if they are emitted with DW_TAG_formal_parameter.
Consider the program:
```rust
use std::env;
#[inline(always)]
fn square(n: i32) -> i32 {
n * n
}
#[inline(never)]
fn square_no_inline(n: i32) -> i32 {
n * n
}
fn main() {
let x = square(env::vars().count() as i32);
let y = square_no_inline(env::vars().count() as i32);
println!("{x} == {y}");
}
```
When making a release build with debug=2 and rustc 1.82.0-nightly (8b3870784 2024-08-07)
```
(gdb) r
Starting program: /ephemeral/tmp/target/release/tmp [Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
Breakpoint 1, tmp::square () at src/main.rs:5
5 n * n
(gdb) info args
No arguments.
(gdb) info locals
n = 31
(gdb) c
Continuing.
Breakpoint 2, tmp::square_no_inline (n=31) at src/main.rs:10
10 n * n
(gdb) info args
n = 31
(gdb) info locals
No locals.
```
This issue is particularly annoying because it removes arguments from stack traces.
The DWARF for the inlined function looks like this:
```
< 2><0x00002132 GOFF=0x00002132> DW_TAG_subprogram
DW_AT_linkage_name _ZN3tmp6square17hc507052ff3d2a488E
DW_AT_name square
DW_AT_decl_file 0x0000000f /ephemeral/tmp/src/main.rs
DW_AT_decl_line 0x00000004
DW_AT_type 0x00001a56<.debug_info+0x00001a56>
DW_AT_inline DW_INL_inlined
< 3><0x00002142 GOFF=0x00002142> DW_TAG_lexical_block
< 4><0x00002143 GOFF=0x00002143> DW_TAG_formal_parameter
DW_AT_name n
DW_AT_decl_file 0x0000000f /ephemeral/tmp/src/main.rs
DW_AT_decl_line 0x00000004
DW_AT_type 0x00001a56<.debug_info+0x00001a56>
< 4><0x0000214e GOFF=0x0000214e> DW_TAG_null
< 3><0x0000214f GOFF=0x0000214f> DW_TAG_null
```
That DW_TAG_lexical_block inhibits every debugger I've tested from recognizing 'n' as a parameter.
This patch removes the additional lexical scope. Parameters can be easily deduplicated by a tuple of their scope and the argument index, at the trivial cost of taking a Hash + Eq bound on DIScope.
Use the `enum2$` Natvis visualiser for repr128 C-style enums
Use the preexisting `enum2$` Natvis visualiser to allow PDB debuggers to display fieldless `#[repr(u128)]]`/`#[repr(i128)]]` enums correctly.
Tracking issue: #56071
try-job: x86_64-msvc
Shrink `TyKind::FnPtr`.
By splitting the `FnSig` within `TyKind::FnPtr` into `FnSigTys` and `FnHeader`, which can be packed more efficiently. This reduces the size of the hot `TyKind` type from 32 bytes to 24 bytes on 64-bit platforms. This reduces peak memory usage by a few percent on some benchmarks. It also reduces cache misses and page faults similarly, though this doesn't translate to clear cycles or wall-time improvements on CI.
r? `@compiler-errors`
Line numbers of multiply-inlined functions were fixed in #114643 by using a
single DISubprogram. That, however, triggered assertions because parameters
weren't deduplicated. The "solution" to that in #115417 was to insert a
DILexicalScope below the DISubprogram and parent all of the parameters to that
scope. That fixed the assertion, but debuggers (including gdb and lldb) don't
recognize variables that are not parented to the subprogram itself as parameters,
even if they are emitted with DW_TAG_formal_parameter.
Consider the program:
use std::env;
fn square(n: i32) -> i32 {
n * n
}
fn square_no_inline(n: i32) -> i32 {
n * n
}
fn main() {
let x = square(env::vars().count() as i32);
let y = square_no_inline(env::vars().count() as i32);
println!("{x} == {y}");
}
When making a release build with debug=2 and rustc 1.82.0-nightly (8b3870784 2024-08-07)
(gdb) r
Starting program: /ephemeral/tmp/target/release/tmp
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
Breakpoint 1, tmp::square () at src/main.rs:5
5 n * n
(gdb) info args
No arguments.
(gdb) info locals
n = 31
(gdb) c
Continuing.
Breakpoint 2, tmp::square_no_inline (n=31) at src/main.rs:10
10 n * n
(gdb) info args
n = 31
(gdb) info locals
No locals.
This issue is particularly annoying because it removes arguments from stack traces.
The DWARF for the inlined function looks like this:
< 2><0x00002132 GOFF=0x00002132> DW_TAG_subprogram
DW_AT_linkage_name _ZN3tmp6square17hc507052ff3d2a488E
DW_AT_name square
DW_AT_decl_file 0x0000000f /ephemeral/tmp/src/main.rs
DW_AT_decl_line 0x00000004
DW_AT_type 0x00001a56<.debug_info+0x00001a56>
DW_AT_inline DW_INL_inlined
< 3><0x00002142 GOFF=0x00002142> DW_TAG_lexical_block
< 4><0x00002143 GOFF=0x00002143> DW_TAG_formal_parameter
DW_AT_name n
DW_AT_decl_file 0x0000000f /ephemeral/tmp/src/main.rs
DW_AT_decl_line 0x00000004
DW_AT_type 0x00001a56<.debug_info+0x00001a56>
< 4><0x0000214e GOFF=0x0000214e> DW_TAG_null
< 3><0x0000214f GOFF=0x0000214f> DW_TAG_null
That DW_TAG_lexical_block inhibits every debugger I've tested from recognizing
'n' as a parameter.
This patch removes the additional lexical scope. Parameters can be easily
deduplicated by a tuple of their scope and the argument index, at the trivial
cost of taking a Hash + Eq bound on DIScope.
const vector passed through to codegen
This allows constant vectors using a repr(simd) type to be propagated
through to the backend by reusing the functionality used to do a similar
thing for the simd_shuffle intrinsic
#118209
r? RalfJung
By splitting the `FnSig` within `TyKind::FnPtr` into `FnSigTys` and
`FnHeader`, which can be packed more efficiently. This reduces the size
of the hot `TyKind` type from 32 bytes to 24 bytes on 64-bit platforms.
This reduces peak memory usage by a few percent on some benchmarks. It
also reduces cache misses and page faults similarly, though this doesn't
translate to clear cycles or wall-time improvements on CI.
rustc_codegen_ssa: Set architecture for object crate for 32-bit SPARC
The `object` crate was recently updated to recognize the 32-bit SPARC ELF targets `EM_SPARC` and `EM_SPARC32PLUS`, so the proper architecture for 32-bit SPARC can now be set in `rustc_codegen_ssa`.
r? nagisa
Make create_dll_import_lib easier to implement
This will make it easier to implement raw-dylib support in cg_clif and cg_gcc. This PR doesn't yet include an create_dll_import_lib implementation for cg_clif as I need to correctly implement dllimport in cg_clif first before raw-dylib can work at all with cg_clif.
Required for https://github.com/rust-lang/rustc_codegen_cranelift/issues/1345
Version 0.3.1 has added support for writing import libraries. Version
0.3.2 fixed creating archives containing members of import libraries.
Version 0.3.3 fixed building on big-endian systems.
The object crate was recently updated to recognize the 32-bit SPARC
ELF targets EM_SPARC and EM_SPARC32PLUS, so the proper architecture
for 32-bit SPARC can now be set in rustc_codegen_ssa.
Add `Debug` impls to API types in `rustc_codegen_ssa`
Some types used in `rustc_codegen_ssa`'s interface traits are missing `Debug` impls. Though I did not smear `#[derive(Debug)]` all over the crate (some structs are quite large).
linker: Pass fewer search directories to the linker
- The logic for passing `-L` directories to the linker is consolidated in a single function, so the search priorities are immediately clear.
- Only `-Lnative=`, `-Lframework=` `-Lall=` directories are passed to linker, but not `-Lcrate=` and others. That's because only native libraries are looked up by name by linker, all Rust crates are passed using full paths, and their directories should not interfere with linker search paths.
- The main sysroot library directory shouldn't generally be passed because it shouldn't contain native libraries, except for one case which is now marked with a FIXME.
- This also helps with https://github.com/rust-lang/rust/pull/123436, in which we need to walk the same list of directories manually.
The next step is to migrate `find_native_static_library` to exactly the same set and order of search directories (which may be a bit annoying for the `iOSSupport` directories https://github.com/rust-lang/rust/pull/121430#issuecomment-2256372341).
Create COFF archives for non-LLVM backends
`ar_archive_writer` now supports creating COFF archives, so enable them for the non-LLVM backends when requested.
r? ``@bjorn3``
When an archive fails to build, print the path
Currently the output on failure is as follows:
Compiling block-buffer v0.10.4
Compiling crypto-common v0.1.6
Compiling digest v0.10.7
Compiling sha2 v0.10.8
Compiling xz2 v0.1.7
error: failed to build archive: No such file or directory
error: could not compile `bootstrap` (lib) due to 1 previous error
Change this to print which file is being constructed, to give some hint about what is going on.
error: failed to build archive at `path/to/output`: No such file or directory
deps: dedup object, wasmparser, wasm-encoder
* dedups one `object`, additional dupe will be removed, with next `thorin-dwp` update
* `wasmparser` pinned to minor versions, so full merge isn't possible
* same with `wasm-encoder`
Turned off some features for `wasmparser` (see features https://github.com/bytecodealliance/wasm-tools/blob/v1.208.1/crates/wasmparser/Cargo.toml) in `run-make-support`, looks working?
compiler: Never debug_assert in codegen
In the name of Turing and his Hoarey heralds, assert our truths before creating a monster!
The `rustc_codegen_llvm` and `rustc_codegen_ssa` crates are fairly critical for rustc's correctness. Small mistakes here can easily result in undefined behavior, since a "small mistake" can mean something like "link and execute the wrong code". We should probably run any and all asserts in these modules unconditionally on whether this is a "debug build", and damn the costs in performance.
...Especially because the costs in performance seem to be *nothing*. It is not clear how much correctness we gain here, but I'll take free correctness improvements.
rustc_target: add known safe s390x target features
This pull request adds known safe target features for s390x (aka IBM Z systems).
Currently, these features are unstable since stabilizing the target features requires submitting proposals.
The `vector` feature was added in IBM Z13 (`arch11`), and this is a SIMD feature for the newer IBM Z systems.
The `backchain` attribute is the IBM Z way of adding frame pointers like unwinding capabilities (the "frame-pointer" switch on IBM Z and IBM POWER platforms will add _emulated_ frame pointers to the binary, which profilers can't use for unwinding the stack).
Both attributes can be applied at the LLVM module or function levels. However, the `backchain` attribute has to be enabled for all the functions in the call stack to get a successful unwind process.
Add a hook for `should_codegen_locally`
This PR lifts the module-local function `should_codegen_locally` to `TyCtxt` as a hook.
In addition to monomorphization, this function is used for checking the dependency of `compiler_builtins` on other libraries. Moving this function to the hooks also makes overriding it possible for the tools that use the rustc interface.
Clean up a few minor refs in `format!` macro, as it has a performance cost. Apparently the compiler is unable to inline `format!("{}", &variable)`, and does a run-time double-reference instead (format macro already does one level referencing). Inlining format args prevents accidental `&` misuse.
Sync ar_archive_writer to LLVM 18.1.3
From LLVM 15.0.0-rc3. This adds support for COFF archives containing Arm64EC object files and has various fixes for AIX big archive files.
Currently the output on failure is as follows:
Compiling block-buffer v0.10.4
Compiling crypto-common v0.1.6
Compiling digest v0.10.7
Compiling sha2 v0.10.8
Compiling xz2 v0.1.7
error: failed to build archive: No such file or directory
error: could not compile `bootstrap` (lib) due to 1 previous error
Print which file is being constructed to give some hint about what is
going on.
use "bootstrap" instead of "rustbuild" in comments and docs
Let's stick with the single name "bootstrap" to refer to the bootstrap project to avoid confusion. This should make it clearer, especially for new contributors.
linker: Link dylib crates by path
Linkers seem to support linking dynamic libraries by path.
Not sure why the previous scheme with splitting the path into a directory (passed with `-L`) and a name (passed with `-l`) was used (upd: likely due to https://github.com/rust-lang/rust/pull/126094#issuecomment-2155063414).
When we split a library path `some/dir/libfoo.so` into `-L some/dir` and `-l foo` we add `some/dir` to search directories for *all* libraries looked up by the linker, not just `foo`, and `foo` is also looked up in *all* search directories not just `some/dir`.
Technically we may find some unintended libraries this way.
Therefore linking dylibs via a full path is both simpler and more reliable.
It also makes the set of search directories more easily reproducible when we need to lookup some native library manually (like in https://github.com/rust-lang/rust/pull/123436).
Re-implement a type-size based limit
r? lcnr
This PR reintroduces the type length limit added in #37789, which was accidentally made practically useless by the caching changes to `Ty::walk` in #72412, which caused the `walk` function to no longer walk over identical elements.
Hitting this length limit is not fatal unless we are in codegen -- so it shouldn't affect passes like the mir inliner which creates potentially very large types (which we observed, for example, when the new trait solver compiles `itertools` in `--release` mode).
This also increases the type length limit from `1048576 == 2 ** 20` to `2 ** 24`, which covers all of the code that can be reached with craterbot-check. Individual crates can increase the length limit further if desired.
Perf regression is mild and I think we should accept it -- reinstating this limit is important for the new trait solver and to make sure we don't accidentally hit more type-size related regressions in the future.
Fixes#125460
Fix `FnMut::call_mut`/`Fn::call` shim for async closures that capture references
I adjusted async closures to be able to implement `Fn` and `FnMut` *even if* they capture references, as long as those references did not need to borrow data from the closure captures themselves. See #125259.
However, when I did this, I didn't actually relax an assertion in the `build_construct_coroutine_by_move_shim` shim code, which builds the `Fn`/`FnMut`/`FnOnce` implementations for async closures. Therefore, if we actually tried to *call* `FnMut`/`Fn` on async closures, it would ICE.
This PR adjusts this assertion to ensure that we only capture immutable references in closures if they implement `Fn`/`FnMut`. It also adds a bunch of tests and makes more of the async-closure tests into `build-pass` since we often care about these tests actually generating the right closure shims and stuff. I think it might be excessive to *always* use build-pass here, but 🤷 it's not that big of a deal.
Fixes#127019Fixes#127012
r? oli-obk
patchable-function-entry: Add unstable compiler flag and attribute
Tracking issue: #123115
Add the -Z patchable-function-entry compiler flag and the #[patchable_function_entry(prefix_nops = m, entry_nops = n)] attribute.
Rebased and adjusted the canditate implementation to match changes in the RFC.
coverage: Make `#[coverage(..)]` apply recursively to nested functions
This PR makes the (currently-unstable) `#[coverage(off)]` and `#[coverage(on)]` attributes apply recursively to all nested functions/closures, instead of just the function they are directly attached to.
Those attributes can now also be applied to modules and to impl/impl-trait blocks, where they have no direct effect, but will be inherited by all enclosed functions/closures/methods that don't override the inherited value.
---
Fixes#126625.
De-duplicate all consecutive native libs regardless of their options
Address https://github.com/rust-lang/rust/pull/126913#issuecomment-2188184011 by no longer de-duplicating based on the "options" but by only looking at the generated link args, as to avoid consecutive libs that originated from different native-lib with different options (like `raw-dylib` on Windows) but isn't relevant for `--print=native-static-libs`.
r? ``@petrochenkov``
coverage: Overhaul validation of the `#[coverage(..)]` attribute
This PR makes sweeping changes to how the (currently-unstable) coverage attribute is validated:
- Multiple coverage attributes on the same item/expression are now treated as an error.
- The attribute must always be `#[coverage(off)]` or `#[coverage(on)]`, and the error messages for this are more consistent.
- A trailing comma is still allowed after off/on, since that's part of the normal attribute syntax.
- Some places that silently ignored a coverage attribute now produce an error instead.
- These cases were all clearly bugs.
- Some places that ignored a coverage attribute (with a warning) now produce an error instead.
- These were originally added as lints, but I don't think it makes much sense to knowingly allow new attributes to be used in meaningless places.
- Some of these errors might soon disappear, if it's easy to extend recursive coverage attributes to things like modules and impl blocks.
---
One of the goals of this PR is to lay a more solid foundation for making the coverage attribute recursive, so that it applies to all nested functions/closures instead of just the one it is directly attached to.
Fixes#126658.
This PR incorporates #126659, which adds more tests for validation of the coverage attribute.
`@rustbot` label +A-code-coverage
Deprecate no-op codegen option `-Cinline-threshold=...`
This deprecates `-Cinline-threshold` since using it has no effect. This has been the case since the new LLVM pass manager started being used, more than 2 years ago.
Recommend using `-Cllvm-args=--inline-threshold=...` instead.
Closes#89742 which is E-help-wanted.
`PtrMetadata` doesn't care about `*const`/`*mut`/`&`/`&mut`, so GVN away those casts in its argument.
This includes updating MIR to allow calling PtrMetadata on references too, not just raw pointers. That means that `[T]::len` can be just `_0 = PtrMetadata(_1)`, for example.
# Conflicts:
# tests/mir-opt/pre-codegen/slice_index.slice_get_unchecked_mut_range.PreCodegen.after.panic-abort.mir
# tests/mir-opt/pre-codegen/slice_index.slice_get_unchecked_mut_range.PreCodegen.after.panic-unwind.mir
Clean up some comments near `use` declarations
#125443 will reformat all `use` declarations in the repository. There are a few edge cases involving comments on `use` declarations that require care. This PR cleans up some clumsy comment cases, taking us a step closer to #125443 being able to merge.
r? ``@lqd``
Most modules have such a blank line, but some don't. Inserting the blank
line makes it clearer that the `//!` comments are describing the entire
module, rather than the `use` declaration(s) that immediately follows.
We already do this for a number of crates, e.g. `rustc_middle`,
`rustc_span`, `rustc_metadata`, `rustc_span`, `rustc_errors`.
For the ones we don't, in many cases the attributes are a mess.
- There is no consistency about order of attribute kinds (e.g.
`allow`/`deny`/`feature`).
- Within attribute kind groups (e.g. the `feature` attributes),
sometimes the order is alphabetical, and sometimes there is no
particular order.
- Sometimes the attributes of a particular kind aren't even grouped
all together, e.g. there might be a `feature`, then an `allow`, then
another `feature`.
This commit extends the existing sorting to all compiler crates,
increasing consistency. If any new attribute line is added there is now
only one place it can go -- no need for arbitrary decisions.
Exceptions:
- `rustc_log`, `rustc_next_trait_solver` and `rustc_type_ir_macros`,
because they have no crate attributes.
- `rustc_codegen_gcc`, because it's quasi-external to rustc (e.g. it's
ignored in `rustfmt.toml`).
rustc_codegen_ssa: fix `get_rpath_relative_to_output` panic when lib only contains file name
<!--
If this PR is related to an unstable feature or an otherwise tracked effort,
please link to the relevant tracking issue here. If you don't know of a related
tracking issue or there are none, feel free to ignore this.
This PR will get automatically assigned to a reviewer. In case you would like
a specific user to review your work, you can assign it to them by using
r? <reviewer name>
-->
When compiles program with `-C rpath=yes` but with no output filename specified, or with filename ONLY, we will get an ICE for now. Fix it by treat empty `output` path in `get_rpath_relative_to_output` as current dir.
Before this patch:
```bash
rustc -C prefer_dynamic=yes -C rpath=yes -O h.rs # ICE, no output filename specified
rustc -o hello -C prefer_dynamic=yes -C rpath=yes -O h.rs # ICE, output filename has no path
rustc -o ./hello -C prefer_dynamic=yes -C rpath=yes -O h.rs # Works
```
All those examples work after the patch.
Close#119571.
Close#125785.
Align `Term` methods with `GenericArg` methods, add `Term::expect_*`
* `Term::ty` -> `Term::as_type`.
* `Term::ct` -> `Term::as_const`.
* Adds `Term::expect_type` and `Term::expect_const`, and uses them in favor of `.ty().unwrap()`, etc.
I could also shorten these to `as_ty` and then do `GenericArg::as_ty` as well, but I do think the `as_` is important to signal that this is a conversion method, and not a getter, like `Const::ty` is.
r? types
Show files produced by `--emit foo` in json artifact notifications
Right now it is possible to ask `rustc` to save some intermediate representation into one or more files with `--emit=foo`, but figuring out what exactly was produced is difficult. This pull request adds information about `llvm_ir` and `asm` intermediate files into notifications produced by `--json=artifacts`.
Related discussion: https://internals.rust-lang.org/t/easier-access-to-files-generated-by-emit-foo/20477
Motivation - `cargo-show-asm` parses those intermediate files and presents them in a user friendly way, but right now I have to apply some dirty hacks. Hacks make behavior confusing: https://github.com/hintron/computer-enhance/issues/35
This pull request introduces a new behavior: now `rustc` will emit a new artifact notification for every artifact type user asked to `--emit`, for example for `--emit asm` those will include all the `.s` files.
Most users won't notice this behavior, to be affected by it all of the following must hold:
- user must use `rustc` binary directly (when `cargo` invokes `rustc` - it consumes artifact notifications and doesn't emit anything)
- user must specify both `--emit xxx` and `--json artifacts`
- user must refuse to handle unknown artifact types
- user must disable incremental compilation (or deal with it better than cargo does, or use a workaround like `save-temps`) in order not to hit #88829 / #89149
This replaces the drop_in_place reference with null in vtables. On
librustc_driver.so, this drops about ~17k dynamic relocations from the
output, since many vtables can now be placed in read-only memory, rather
than having a relocated pointer included.
This makes a tradeoff by adding a null check at vtable call sites.
That's hard to avoid without changing the vtable format (e.g., to use a
pc-relative relocation instead of an absolute address, and avoid the
dynamic relocation that way). But it seems likely that the check is
cheap at runtime.
[perf] Delay the construction of early lint diag structs
Attacks some of the perf regressions from https://github.com/rust-lang/rust/pull/124417#issuecomment-2123700666.
See individual commits for details. The first three commits are not strictly necessary.
However, the 2nd one (06bc4fc671, *Remove `LintDiagnostic::msg`*) makes the main change way nicer to implement.
It's also pretty sweet on its own if I may say so myself.
Run rustfmt on files that need it.
Somehow these files aren't properly formatted. By default `x fmt` and `x tidy` only check files that have changed against master, so if an ill-formatted file somehow slips in it can stay that way as long as it doesn't get modified(?)
I found these when I ran `x fmt` explicitly on every `.rs` file in the repo, while working on
https://github.com/rust-lang/compiler-team/issues/750.
Somehow these files aren't properly formatted. By default `x fmt` and `x
tidy` only check files that have changed against master, so if an
ill-formatted file somehow slips in it can stay that way as long as it
doesn't get modified(?)
I found these when I ran `x fmt` explicitly on every `.rs` file in the
repo, while working on
https://github.com/rust-lang/compiler-team/issues/750.
Rollup of 6 pull requests
Successful merges:
- #125263 (rust-lld: fallback to rustc's sysroot if there's no path to the linker in the target sysroot)
- #125345 (rustc_codegen_llvm: add support for writing summary bitcode)
- #125362 (Actually use TAIT instead of emulating it)
- #125412 (Don't suggest adding the unexpected cfgs to the build-script it-self)
- #125445 (Migrate `run-make/rustdoc-with-short-out-dir-option` to `rmake.rs`)
- #125452 (Cleanup check-cfg handling in core and std)
r? `@ghost`
`@rustbot` modify labels: rollup
rustc_codegen_llvm: add support for writing summary bitcode
Typical uses of ThinLTO don't have any use for this as a standalone file, but distributed ThinLTO uses this to make the linker phase more efficient. With clang you'd do something like `clang -flto=thin -fthin-link-bitcode=foo.indexing.o -c foo.c` and then get both foo.o (full of bitcode) and foo.indexing.o (just the summary or index part of the bitcode). That's then usable by a two-stage linking process that's more friendly to distributed build systems like bazel, which is why I'm working on this area.
I talked some to `@teresajohnson` about naming in this area, as things seem to be a little confused between various blog posts and build systems. "bitcode index" and "bitcode summary" tend to be a little too ambiguous, and she tends to use "thin link bitcode" and "minimized bitcode" (which matches the descriptions in LLVM). Since the clang option is thin-link-bitcode, I went with that to try and not add a new spelling in the world.
Per `@dtolnay,` you can work around the lack of this by using `lld --thinlto-index-only` to do the indexing on regular .o files of bitcode, but that is a bit wasteful on actions when we already have all the information in rustc and could just write out the matching minimized bitcode. I didn't test that at all in our infrastructure, because by the time I learned that I already had this patch largely written.
rust-lld: fallback to rustc's sysroot if there's no path to the linker in the target sysroot
As seen in #125246, some sysroots don't expect to contain `rust-lld` and want to keep it that way, so we fallback to the default rustc sysroot if there is no path to the linker in any of the sysroot tools search paths. This is how we locate codegen-backends' dylibs already.
People also have requested an error if none of these search paths contain the self-contained linker directory, so there's also an error in that case.
r? `@petrochenkov` cc `@ehuss` `@RalfJung`
I'm not sure where we check for `rust-lld`'s existence on the targets where we use it by default, and if we just ignore it when missing or emit a warning (as I assume we don't emit an error), so I just checked for the existence of `gcc-ld`, where `cc` will look for the lld-wrapper binaries.
<sub>*Feel free to point out better ways to do this, it's the middle of the night here.*</sub>
Fixes#125246
Remove more `#[macro_use] extern crate tracing`
Because explicit importing of macros via use items is nicer (more standard and readable) than implicit importing via `#[macro_use]`. Continuing the work from #124511 and #124914.
r? `@jackh726`
If we don't do this, some versions of LLVM (at least 17, experimentally)
will double-emit some error messages, which is how I noticed this. Given
that it seems to be costing some extra work, let's only request the
summary bitcode production if we'll actually bother writing it down,
otherwise skip it.
self-contained linker: retry linking without `-fuse-ld=lld` on CCs that don't support it
For the self-contained linker, this PR applies [the strategy](https://github.com/rust-lang/rust/issues/125330#issuecomment-2125119838) of retrying the linking step when the driver doesn't support `-fuse-ld=lld`, but with the option removed. This is the same strategy we already use of retrying when e.g. `-no-pie` is not supported.
Fixes#125330
r? `@petrochenkov`
I have no idea how we could add a test here, much like we don't have one for `-no-pie` or `-static-pie` -- let me know if you have ideas -- but I tested on a CentOS7 image:
```console
[root@d25b38376ede tmp]# ../build/host/stage1/bin/rustc helloworld.rs
WARN rustc_codegen_ssa:🔙:link The linker driver does not support `-fuse-ld=lld`. Retrying without it.
[root@d25b38376ede tmp]# readelf -p .comment helloworld
String dump of section '.comment':
[ 0] GCC: (GNU) 4.8.5 20150623 (Red Hat 4.8.5-44)
[ 2d] rustc version 1.80.0-dev
```
I wasn't able to test with `cross` as the issue describes: I wasn't able to reproduce that behavior locally.
* instead simply set the primary message inside the lint decorator functions
* it used to be this way before [#]101986 which introduced `msg` to prevent
good path delayed bugs (which no longer exist) from firing under certain
circumstances when lints were suppressed / silenced
* this is no longer necessary for various reasons I presume
* it shaves off complexity and makes further changes easier to implement
Stop using `to_hir_binop` in codegen
This came up in https://github.com/rust-lang/rust/pull/125359#discussion_r1609401311 , and looking into it we can just use the `mir::BinOp`s directly instead of `hir::BinOpKind`s.
(AKA rather than going `mir::BinOp` → `hir::BinOpKind` → `IntPredicate`, just go `mir::BinOp` → `IntPredicate`.)
Typical uses of ThinLTO don't have any use for this as a standalone
file, but distributed ThinLTO uses this to make the linker phase more
efficient. With clang you'd do something like `clang -flto=thin
-fthin-link-bitcode=foo.indexing.o -c foo.c` and then get both foo.o
(full of bitcode) and foo.indexing.o (just the summary or index part of
the bitcode). That's then usable by a two-stage linking process that's
more friendly to distributed build systems like bazel, which is why I'm
working on this area.
I talked some to @teresajohnson about naming in this area, as things
seem to be a little confused between various blog posts and build
systems. "bitcode index" and "bitcode summary" tend to be a little too
ambiguous, and she tends to use "thin link bitcode" and "minimized
bitcode" (which matches the descriptions in LLVM). Since the clang
option is thin-link-bitcode, I went with that to try and not add a new
spelling in the world.
Per @dtolnay, you can work around the lack of this by using `lld
--thinlto-index-only` to do the indexing on regular .o files of
bitcode, but that is a bit wasteful on actions when we already have all
the information in rustc and could just write out the matching minimized
bitcode. I didn't test that at all in our infrastructure, because by the
time I learned that I already had this patch largely written.
