rust/compiler/rustc_mir_transform/src/inline.rs, branch perf-tmphttps://github.com/rust-lang/rust
http://git.dreamy.place/mirrors/rust/atom?h=perf-tmp2025-10-02T06:55:51+00:00mir-opt: Eliminate trivial unnecessary storage annotations2025-10-02T06:55:51+00:00dianqkdianqk@dianqk.net2025-07-15T14:54:54+00:00urn:sha1:85b2f706939528b5796d98e82c183637f8338cd1remove rustc_attr_data_structures2025-07-31T12:19:27+00:00Jana Dönszelmannjana@donsz.nl2025-07-31T09:00:40+00:00urn:sha1:e1d3ad89c7a2ad4f5d944a7fee1298ffe8c99645Auto merge of #142890 - kornelski:unused-var-debug, r=saethlin2025-07-03T23:17:03+00:00borsbors@rust-lang.org2025-07-03T23:17:03+00:00urn:sha1:837c5dd7de03aa97190593aef4e70d53e1bb574b
MIR inliner maintains unused var_debug_info
Only `full` debuginfo level promises variable-level debug information, but the MIR inline pass needlessly preserved the local variable debug info for the `limited` level too.
mir: Add a `new` method to `statement`2025-06-29T12:13:36+00:00dianqkdianqk@dianqk.net2025-06-08T07:30:18+00:00urn:sha1:9f9cd5e2837f57504b92b5acbadb48ff0cb9e29c
Avoid introducing a large number of changes when adding optional initialization fields.
Keep inlined var_debug_info only when full debug info is used2025-06-28T11:22:49+00:00Kornelkornel@geekhood.net2025-02-08T01:33:38+00:00urn:sha1:c9ef11695ffee0dc17f795b51c19f333abdb08a2Only store the LocalDefId instead of the whole instance.2025-06-23T08:44:31+00:00Camille GILLOTgillot.camille@gmail.com2025-06-19T08:47:39+00:00urn:sha1:2074013f44e4b12a0949182d050d3470bd76e4deOnly compute recursive callees once.2025-06-22T20:08:49+00:00Camille GILLOTgillot.camille@gmail.com2025-06-17T13:24:38+00:00urn:sha1:09aab29ebf93fbb546d409d1e127392abf36cd8adon't depend on rustc_attr_parsing if rustc_data_structures will do2025-05-09T21:16:55+00:00mejrs59372212+mejrs@users.noreply.github.com2025-05-09T21:16:55+00:00urn:sha1:684b7b70f4940a0abfecd3d95fe339960d6048d4AsyncDrop implementation using shim codegen of async_drop_in_place::{closure}, scoped async drop added.2025-04-28T09:23:13+00:00Andrew Zhoginandrew.zhogin@gmail.com2024-08-26T13:45:15+00:00urn:sha1:c366756a8537ef998d32c1ef57098d5aae7ca92fRollup merge of #139042 - compiler-errors:do-not-optimize-switchint, r=saethlin2025-04-19T19:30:46+00:00Chris Dentonchris@chrisdenton.dev2025-04-19T19:30:46+00:00urn:sha1:5d2375f789e88d0ac6dbbd828082559a5d2fa8c6
Do not remove trivial `SwitchInt` in analysis MIR
This PR ensures that we don't prematurely remove trivial `SwitchInt` terminators which affects both the borrow-checking and runtime semantics (i.e. UB) of the code. Previously the `SimplifyCfg` optimization was removing `SwitchInt` terminators when they was "trivial", i.e. when all arms branched to the same basic block, even if that `SwitchInt` terminator had the side-effect of reading an operand which (for example) may not be initialized or may point to an invalid place in memory.
This behavior is unlike all other optimizations, which are only applied after "analysis" (i.e. borrow-checking) is finished, and which Miri disables to make sure the compiler doesn't silently remove UB.
Fixing this code "breaks" (i.e. unmasks) code that used to borrow-check but no longer does, like:
```rust
fn foo() {
let x;
let (0 | _) = x;
}
```
This match expression should perform a read because `_` does not shadow the `0` literal pattern, and the compiler should have to read the match scrutinee to compare it to 0. I've checked that this behavior does not actually manifest in practice via a crater run which came back clean: https://github.com/rust-lang/rust/pull/139042#issuecomment-2767436367
As a side-note, it may be tempting to suggest that this is actually a good thing or that we should preserve this behavior. If we wanted to make this work (i.e. trivially optimize out reads from matches that are redundant like `0 | _`), then we should be enabling this behavior *after* fixing this. However, I think it's kinda unprincipled, and for example other variations of the code don't even work today, e.g.:
```rust
fn foo() {
let x;
let (0.. | _) = x;
}
```