Lower to a memset(undef) when Rvalue::Repeat repeats uninit
Fixes https://github.com/rust-lang/rust/issues/138625.
It is technically correct to just do nothing. But if we actually do nothing, we may miss that this is de-initializing something, so instead we just lower to a single memset that writes undef. This is still superior to the memcpy loop, in both quality of code we hand to the backend and LLVM's final output.
Lower BinOp::Cmp to llvm.{s,u}cmp.* intrinsics
Lowers `mir::BinOp::Cmp` (`three_way_compare` intrinsic) to the corresponding LLVM `llvm.{s,u}cmp.i8.*` intrinsics.
These are the intrinsics mentioned in https://github.com/rust-lang/rust/pull/118310, which are now available in LLVM 19.
I couldn't find any follow-up PRs/discussions about this, please let me know if I missed something.
r? `@scottmcm`
`with_scope` is only ever used for ast modules
Thus I renamed it to match other similar functions (`with_mod_rib`) and made it panic if used on non-modules
match lowering cleanup: remove unused unsizing logic from `non_scalar_compare`
Since array and slice constants are now translated to array and slice patterns, `non_scalar_compare` is only used for string comparisons. This specializes it to strings, renames it, and removes the unused array-unsizing logic.
This also updates the doc comments for `thir::PatKind::Constant` and `TestKind::Eq`, which referred to them being used for slice references.
r? ````@oli-obk````
Optimize hash map operations in the query system
This optimizes hash map operations in the query system by explicitly passing hashes and using more optimal operations. `find_or_find_insert_slot` in particular saves a hash table lookup over `entry`. It's not yet available in a safe API, but will be in https://github.com/rust-lang/hashbrown/pull/466.
<table><tr><td rowspan="2">Benchmark</td><td colspan="1"><b>Before</b></th><td colspan="2"><b>After</b></th></tr><tr><td align="right">Time</td><td align="right">Time</td><td align="right">%</th></tr><tr><td>🟣 <b>clap</b>:check</td><td align="right">1.6189s</td><td align="right">1.6129s</td><td align="right"> -0.37%</td></tr><tr><td>🟣 <b>hyper</b>:check</td><td align="right">0.2353s</td><td align="right">0.2337s</td><td align="right"> -0.67%</td></tr><tr><td>🟣 <b>regex</b>:check</td><td align="right">0.9344s</td><td align="right">0.9289s</td><td align="right"> -0.59%</td></tr><tr><td>🟣 <b>syn</b>:check</td><td align="right">1.4693s</td><td align="right">1.4652s</td><td align="right"> -0.28%</td></tr><tr><td>🟣 <b>syntex_syntax</b>:check</td><td align="right">5.6606s</td><td align="right">5.6439s</td><td align="right"> -0.30%</td></tr><tr><td>Total</td><td align="right">9.9185s</td><td align="right">9.8846s</td><td align="right"> -0.34%</td></tr><tr><td>Summary</td><td align="right">1.0000s</td><td align="right">0.9956s</td><td align="right"> -0.44%</td></tr></table>
r? `@cjgillot`
They're dodgy, covering all the keywords, including weak ones, and
edition-specific ones without considering the edition. They have a
single use in rustfmt. This commit changes that use to
`is_reserved_ident`, which is a much more widely used alternative and is
good enough, judging by the lack of effect on the test suite.
Tweaks to writeback and `Obligation -> Goal` conversion
Each of these commits are self-contained, but are prerequisites that I'd like to land before #138845, which still needs some cleaning.
The ""most controversial"" one is probably [Explicitly don't fold coroutine obligations in writeback](e7d27bae27), which I prefer because I think using `fold_predicate` to control against not normalizing predicates seems... easy to mess up 🤔, and we could have *other things* that we don't want to normalize.
Explicitly noting whether we want `resolve` to normalize is a lot clearer (and currently in writeback is limited to resolving stalled coroutine obligations), since we can attach it to a comment that explains *why*.
Cache current_dll_path output
Computing the current dll path is somewhat expensive relative to other work when compiling `fn main() {}` as `dladdr` needs to iterate over the symbol table of librustc_driver.so until it finds a match.
Don't attempt to export compiler-builtins symbols from rust dylibs
They are marked with hidden visibility to prevent them from getting exported, so we shouldn't ask the linker to export them anyway. The only thing that does it cause a warning on macOS.
Part of https://github.com/rust-lang/rust/issues/136096
cc `@jyn514`
Tweak type flags, fix missing flags from coroutine kind ty
Firstly, make sure we visit the coroutine kind ty. Since this kind ty is either infer (before upvar computation), or `()` or `i8`/`i16`/`i32`, this isn't really that big of a deal, since other types in the coroutine will also be infer, so we're not misreporting `ty.has_infer()` or anything, but it's still wrong not to do this.
Furthermore, remove `HAS_TY_COROUTINE`, since nobody used it, and also remove special casing for `STILL_FURTHER_SPECIALIZABLE` since it's likely not important anymore? I have a vague recollection that it was important for polymorphization(?), but no tests seem to rely on this behavior.
r? lcnr or reassign
Add unstable `--print=supported-crate-types` option
MCP: https://github.com/rust-lang/compiler-team/issues/836
Tracking issue: https://github.com/rust-lang/rust/issues/138640
### Test coverage
Two tests:
1. `tests/ui/print-request/stability.rs` to check that `--print=supported-crate-types` is `-Zunstable-options`-gated
2. `tests/ui/print-request/supported-crate-types.rs` is added as a basic smoke test. Observe that the compiler stdout corresponds to the below *Example output* section (e.g. `proc-macro` is unsupported on `wasm32-unknown-unknown` currently).
### Example output
<details>
<summary>For `x86_64-unknown-linux-gnu`</summary>
Notice the presence of `{c,}dylib` and `proc-macro`:
```
bin
cdylib
dylib
lib
proc-macro
rlib
staticlib
```
</details>
<details>
<summary>For `wasm32-unknown-unknown`</summary>
Notice the absence of `dylib` and `proc-macro`:
```
bin
cdylib
lib
rlib
staticlib
```
</details>
<details>
<summary>For `x86_64-unknown-linux-musl`</summary>
Notice the absence of `{c,}dylib` but presence of `proc-macro`:
```
bin
lib
proc-macro
rlib
staticlib
```
</details>
### Documentation
I added an entry in the unstable book's print request section to document this `supported-crate-types` print request.
### Unresolved questions
- [ ] (Name bikeshedding) is `supported-crate-types` a good name for the print request? I'm inclined to say it's good enough for an unstable print request, but may be worth revisiting at stabilization time.
### Stability
This print request being added is *unstable* in this PR. A separate stabilization PR following the usual compiler flag stabilization procedure should be filed for stabilization after some baking time.
### Review remarks
Best reviewed commit-by-commit.
r? compiler
Move some driver code around
`--emit mir`, `#[rustc_symbol_name]` and `#[rustc_def_path]` now run before codegen and thus work even if codegen fails. This can help with debugging.
Reduce FormattingOptions to 64 bits
This is part of https://github.com/rust-lang/rust/issues/99012
This reduces FormattingOptions from 6-7 machine words (384 bits on 64-bit platforms, 224 bits on 32-bit platforms) to just 64 bits (a single register on 64-bit platforms).
Before:
```rust
pub struct FormattingOptions {
flags: u32, // only 6 bits used
fill: char,
align: Option<Alignment>,
width: Option<usize>,
precision: Option<usize>,
}
```
After:
```rust
pub struct FormattingOptions {
/// Bits:
/// - 0-20: fill character (21 bits, a full `char`)
/// - 21: `+` flag
/// - 22: `-` flag
/// - 23: `#` flag
/// - 24: `0` flag
/// - 25: `x?` flag
/// - 26: `X?` flag
/// - 27: Width flag (if set, the width field below is used)
/// - 28: Precision flag (if set, the precision field below is used)
/// - 29-30: Alignment (0: Left, 1: Right, 2: Center, 3: Unknown)
/// - 31: Always set to 1
flags: u32,
/// Width if width flag above is set. Otherwise, always 0.
width: u16,
/// Precision if precision flag above is set. Otherwise, always 0.
precision: u16,
}
```
Since array and slice constants are now lowered to array and slice
patterns, `non_scalar_compare` was only called for string comparisons.
This specializes it to strings, renames it, and removes the unused
array-unsizing logic.
This also updates some outdated doc comments.
Avoid no-op unlink+link dances in incr comp
Incremental compilation scales quite poorly with the number of CGUs. This PR improves one reason for that.
The incr comp process hard-links all the files from an old session into a new one, then it runs the backend, which may just hard-link the new session files into the output directory. Then codegen hard-links all the output files back to the new session directory.
This PR (perhaps unimaginatively) fixes the silliness that ensues in the last step. The old `link_or_copy` implementation would be passed pairs of paths which are already the same inode, then it would blindly delete the destination and re-create the hard-link that it just deleted. This PR lets us skip both those operations. We don't skip the other two hard-links.
`cargo +stage1 b && touch crates/core/main.rs && strace -cfw -elink,linkat,unlink,unlinkat cargo +stage1 b` before and then after on `ripgrep-13.0.0`:
```
% time seconds usecs/call calls errors syscall
------ ----------- ----------- --------- --------- ----------------
52.56 0.024950 25 978 485 unlink
34.38 0.016318 22 727 linkat
13.06 0.006200 24 249 unlinkat
------ ----------- ----------- --------- --------- ----------------
100.00 0.047467 24 1954 485 total
```
```
% time seconds usecs/call calls errors syscall
------ ----------- ----------- --------- --------- ----------------
42.83 0.014521 57 252 unlink
38.41 0.013021 26 486 linkat
18.77 0.006362 25 249 unlinkat
------ ----------- ----------- --------- --------- ----------------
100.00 0.033904 34 987 total
```
This reduces the number of hard-links that are causing perf troubles, noted in https://github.com/rust-lang/rust/issues/64291 and https://github.com/rust-lang/rust/issues/137560
