rust/compiler/rustc_codegen_ssa/src/mir/operand.rs, branch 1.90.0https://github.com/rust-lang/rust
http://git.dreamy.place/mirrors/rust/atom?h=1.90.02025-07-27T06:37:55+00:00Auto merge of #144347 - scottmcm:ssa-enums-v0, r=WaffleLapkin2025-07-27T06:37:55+00:00borsbors@rust-lang.org2025-07-27T06:37:55+00:00urn:sha1:86ef32029427cfc4161a3fd7a51992302f7c5552
No longer need `alloca`s for consuming `Result<!, i32>` and similar
In optimized builds GVN gets rid of these already, but in `opt-level=0` we actually make `alloca`s for this, which particularly impacts `?`-style things that use actually-only-one-variant types like this.
While doing so, rewrite `LocalAnalyzer::process_place` to be non-recursive, solving a 6+ year old FIXME.
r? codegen
Auto merge of #143860 - scottmcm:transmute-always-rvalue, r=WaffleLapkin2025-07-26T22:45:18+00:00borsbors@rust-lang.org2025-07-26T22:45:18+00:00urn:sha1:283a0746a244a88503fed61844f44df925ccdbb6
Let `codegen_transmute_operand` just handle everything
When combined with rust-lang/rust#143720, this means `rvalue_creates_operand` can just return `true` for *every* `Rvalue`. (A future PR could consider removing it, though just letting it optimize out is fine for now.)
It's nicer anyway, IMHO, because it avoids needing the layout checks to be consistent in the two places, and thus is an overall reduction in code. Plus it's a more helpful building block when used in other places this way.
(TBH, it probably would have been better to have it this way the whole time, but I clearly didn't understand `rvalue_creates_operand` when I originally wrote rust-lang/rust#109843.)
Remove useless lifetime parameter.2025-07-23T23:54:37+00:00Camille GILLOTgillot.camille@gmail.com2025-07-03T18:41:12+00:00urn:sha1:0460c92d527dbada3fde227d3f01e6d1e132a186Give an AllocId to ConstValue::Slice.2025-07-23T23:54:37+00:00Camille GILLOTgillot.camille@gmail.com2023-10-13T20:20:57+00:00urn:sha1:9ff071219bca913e45235568defadd5ab840c50aLet `codegen_transmute_operand` just handle everything2025-07-23T15:25:13+00:00Scott McMurrayscottmcm@users.noreply.github.com2025-07-12T23:38:46+00:00urn:sha1:231dddde3e686c0903a28ce605b8c5d505a2ba21
When combined with 143720, this means `rvalue_creates_operand` can just return `true` for *every* `Rvalue`. (A future PR could consider removing it, though just letting it optimize out is fine for now.)
It's nicer anyway, IMHO, because it avoids needing the layout checks to be consistent in the two places, and thus is an overall reduction in code. Plus it's a more helpful building block when used in other places this way.
No longer need `alloca`s for consuming `Result<!, i32>` and similar2025-07-23T07:09:36+00:00Scott McMurrayscottmcm@users.noreply.github.com2025-07-23T06:23:02+00:00urn:sha1:6a5c7e04156125cb8c1ff78eae95e6ff076f51ed
In optimized builds GVN gets rid of these already, but in `opt-level=0` we actually make `alloca`s for this, which particularly impacts `?`-style things that use actually-only-one-variant types like this.
Ban projecting into SIMD types [MCP838]2025-07-20T17:22:09+00:00Scott McMurrayscottmcm@users.noreply.github.com2025-03-07T03:13:46+00:00urn:sha1:41ce1ed252f194756fb2f3e3e92bbdfb3940088dImprove comments inside `codegen_get_discr`2025-07-16T05:30:46+00:00Scott McMurrayscottmcm@users.noreply.github.com2025-07-16T05:28:05+00:00urn:sha1:4fa23d96bc9334716b8f50511f5a920b46d349ddSimplify codegen for niche-encoded variant tests2025-07-12T11:53:24+00:00Scott McMurrayscottmcm@users.noreply.github.com2025-07-11T12:07:37+00:00urn:sha1:d5bcfb334b616e01c3589ba7b1697c7ebc47a024Auto merge of #142911 - mejrs:unsized, r=compiler-errors2025-07-11T05:27:32+00:00borsbors@rust-lang.org2025-07-11T05:27:32+00:00urn:sha1:855e0fe46e68d94e9f6147531b75ac2d488c548e
Remove support for dynamic allocas
Followup to rust-lang/rust#141811