Make `exp`, `exp2`, `log`, `log2`, `log10` generic by N1ark · Pull Request #160989 · rust-lang/rust

N1ark · 2026-08-12T14:23:12Z

Rebased and smaller version of #153934

Following fabs, make the exp, exp2, log, log2 and log10 intrinsics generic over the float type, rather than having four variants per float type.

The first two commits are purely stylistic:

reorganised Cranelift code to make following changes simpler
moved a misplaced comment in compiler/rustc_codegen_llvm/src/intrinsic.rs that caused x fmt to give up

The last commit actually makes them generic! Most code is a bit simpler, and this will also hopefully simplify adding support for these intrinsics for the future bf16 type :)

Unfortunately both GCC and Cranelift backend changes are a bit churny. Their code is a bit, opaque, to put it kindly, and I didn't want to refactor those here.

r? @folkertdev
cc @RalfJung

Disclosure: I used an LLM to figure out the macro code.

rustbot · 2026-08-12T14:23:16Z

rustc_codegen_cranelift is developed in its own repository. If possible, consider making this change to rust-lang/rustc_codegen_cranelift instead.

cc @bjorn3

Some changes occurred to the intrinsics. Make sure the CTFE / Miri interpreter
gets adapted for the changes, if necessary.

cc @rust-lang/miri, @RalfJung, @oli-obk, @lcnr

miri is developed in its own repository. If the Miri part of this change can be broken out, consider making this change to rust-lang/miri instead. However, if Miri needs adjusting for rustc changes, just ignore this message.

cc @rust-lang/miri

rustc_codegen_gcc is developed in its own repository. If possible, consider making this change to rust-lang/rustc_codegen_gcc instead.

cc @antoyo, @GuillaumeGomez

folkertdev · 2026-08-12T14:33:38Z

I suspect it would be better to submit 295e7a3 to the cg_clif repo for better CI coverage.

rustbot · 2026-08-12T14:55:39Z

clippy is developed in its own repository. If possible, consider making this change to rust-lang/rust-clippy instead.

cc @rust-lang/clippy

rust-log-analyzer · 2026-08-12T15:44:56Z

The job x86_64-gnu-miri failed! Check out the build log: (web) (plain enhanced) (plain)

Click to see the possible cause of the failure (guessed by this bot)

tests/pass/shims/x86/rounding-error.rs ... ok
tests/pass/shims/x86/intrinsics-x86-gfni.rs ... ok

FAILED TEST: tests/pass/float.rs
command: MIRI_ENV_VAR_TEST="0" MIRI_TEMP="/tmp/miri-uitest-cHeUEp" RUST_BACKTRACE="1" "/checkout/obj/build/x86_64-unknown-linux-gnu/stage2-tools/x86_64-unknown-linux-gnu/release/miri" "--error-format=json" "-Dwarnings" "-Dunused" "-Ainternal_features" "-Zui-testing" "--sysroot=/checkout/obj/build/x86_64-unknown-linux-gnu/miri-sysroot" "--out-dir" "/checkout/obj/build/x86_64-unknown-linux-gnu/stage2-tools/x86_64-unknown-linux-gnu/tmp/miri_ui/0/tests/pass" "tests/pass/float.rs" "--edition" "2021"

error: test got exit status: 1, but expected 0
 = note: compilation failed, but was expected to succeed

error: no output was expected
Execute `./miri test --bless` to update `tests/pass/float.stderr` to the actual output
+++ <stderr output>
error: unsupported operation: Miri can only use intrinsic fallback bodies that exactly reflect the specification: they fully check for UB and are as non-deterministic as possible. After verifying that `exp` does so, add the `#[miri::intrinsic_fallback_is_spec]` attribute to it; also ping @rust-lang/miri when you do that
##[error]  --> /checkout/library/std/src/num/f16.rs:84:9
   |
LL |         intrinsics::exp(self)
   |         ^^^^^^^^^^^^^^^^^^^^^ unsupported operation occurred here
   |
   = help: this is likely not a bug in the program; it indicates that the program performed an operation that Miri does not support
   = note: stack backtrace:
           0: std::f16::<impl f16>::exp
               at /checkout/library/std/src/num/f16.rs:84:9: 84:30
           1: libm
               at tests/pass/float.rs:1136:23: 1136:33
           2: main
               at tests/pass/float.rs:78:5: 78:11
---



full stderr:
error: unsupported operation: Miri can only use intrinsic fallback bodies that exactly reflect the specification: they fully check for UB and are as non-deterministic as possible. After verifying that `exp` does so, add the `#[miri::intrinsic_fallback_is_spec]` attribute to it; also ping @rust-lang/miri when you do that
##[error]  --> /checkout/library/std/src/num/f16.rs:84:9
   |
LL |         intrinsics::exp(self)
   |         ^^^^^^^^^^^^^^^^^^^^^ unsupported operation occurred here
   |
   = help: this is likely not a bug in the program; it indicates that the program performed an operation that Miri does not support
   = note: stack backtrace:
           0: std::f16::<impl f16>::exp
               at /checkout/library/std/src/num/f16.rs:84:9: 84:30
           1: libm
               at tests/pass/float.rs:1136:23: 1136:33
           2: main
               at tests/pass/float.rs:78:5: 78:11
---



FAILED TEST: tests/pass/integer-ops.rs
command: MIRI_ENV_VAR_TEST="0" MIRI_TEMP="/tmp/miri-uitest-cHeUEp" RUST_BACKTRACE="1" "/checkout/obj/build/x86_64-unknown-linux-gnu/stage2-tools/x86_64-unknown-linux-gnu/release/miri" "--error-format=json" "-Dwarnings" "-Dunused" "-Ainternal_features" "-Zui-testing" "--sysroot=/checkout/obj/build/x86_64-unknown-linux-gnu/miri-sysroot" "--out-dir" "/checkout/obj/build/x86_64-unknown-linux-gnu/stage2-tools/x86_64-unknown-linux-gnu/tmp/miri_ui/0/tests/pass" "tests/pass/integer-ops.rs" "-Coverflow-checks=off" "--edition" "2021"

error: test got exit status: 1, but expected 0
 = note: compilation failed, but was expected to succeed

error: no output was expected
Execute `./miri test --bless` to update `tests/pass/integer-ops.stderr` to the actual output
+++ <stderr output>
error: unsupported operation: Miri can only use intrinsic fallback bodies that exactly reflect the specification: they fully check for UB and are as non-deterministic as possible. After verifying that `log` does so, add the `#[miri::intrinsic_fallback_is_spec]` attribute to it; also ping @rust-lang/miri when you do that
##[error]  --> /checkout/library/std/src/num/f32.rs:474:9
   |
LL |         intrinsics::log(self)
   |         ^^^^^^^^^^^^^^^^^^^^^ unsupported operation occurred here
   |
   = help: this is likely not a bug in the program; it indicates that the program performed an operation that Miri does not support
   = note: stack backtrace:
           0: std::f32::<impl f32>::ln
               at /checkout/library/std/src/num/f32.rs:474:9: 474:30
           1: std::f32::<impl f32>::log
               at /checkout/library/std/src/num/f32.rs:511:9: 511:18
           2: main
               at tests/pass/integer-ops.rs:260:45: 260:65

note: some details are omitted, run with `MIRIFLAGS=-Zmiri-backtrace=full` for a verbose backtrace

error: aborting due to 1 previous error



full stderr:
error: unsupported operation: Miri can only use intrinsic fallback bodies that exactly reflect the specification: they fully check for UB and are as non-deterministic as possible. After verifying that `log` does so, add the `#[miri::intrinsic_fallback_is_spec]` attribute to it; also ping @rust-lang/miri when you do that
##[error]  --> /checkout/library/std/src/num/f32.rs:474:9
   |
LL |         intrinsics::log(self)
   |         ^^^^^^^^^^^^^^^^^^^^^ unsupported operation occurred here
   |
   = help: this is likely not a bug in the program; it indicates that the program performed an operation that Miri does not support
   = note: stack backtrace:
           0: std::f32::<impl f32>::ln
               at /checkout/library/std/src/num/f32.rs:474:9: 474:30
           1: std::f32::<impl f32>::log
               at /checkout/library/std/src/num/f32.rs:511:9: 511:18
           2: main
               at tests/pass/integer-ops.rs:260:45: 260:65

note: some details are omitted, run with `MIRIFLAGS=-Zmiri-backtrace=full` for a verbose backtrace
---

Location:
   /cargo/registry/src/index.crates.io-1949cf8c6b5b557f/ui_test-0.30.7/src/lib.rs:365

Backtrace omitted. Run with RUST_BACKTRACE=1 environment variable to display it.
Run with RUST_BACKTRACE=full to include source snippets.
error: test failed, to rerun pass `--test ui`

Caused by:
  process didn't exit successfully: `/checkout/obj/build/x86_64-unknown-linux-gnu/stage2-tools/x86_64-unknown-linux-gnu/release/build/miri/855274d73082e308/out/ui-855274d73082e308` (exit status: 1)
Bootstrap failed while executing `test --stage 2 miri cargo-miri`
Currently active steps:
test::Miri { target: x86_64-unknown-linux-gnu } at src/bootstrap/src/core/build_steps/test.rs:697
Command `/checkout/obj/build/x86_64-unknown-linux-gnu/stage0/bin/cargo test --target x86_64-unknown-linux-gnu -Zbinary-dep-depinfo -j 4 -Zroot-dir=/checkout --locked --color=always --profile=release --manifest-path /checkout/src/tools/miri/Cargo.toml -- [workdir=/checkout]` failed with exit code 1
Created at: src/bootstrap/src/core/build_steps/tool.rs:197:21
Executed at: src/bootstrap/src/core/build_steps/test.rs:771:19

Command has failed. Rerun with -v to see more details.
Build completed unsuccessfully in 0:44:55
  local time: Wed Aug 12 15:44:44 UTC 2026
  network time: Wed, 12 Aug 2026 15:44:44 GMT
##[error]Process completed with exit code 1.
##[group]Run echo "disk usage:"

bjorn3 · 2026-08-12T15:47:45Z

-            CValue::by_val(codegen_f16_f128::fma_f16(fx, args[0], args[1], args[2]), layout)
-        }
+    let val = match intrinsic {
+        sym::fmaf16 | sym::fmuladdf16 => codegen_f16_f128::fma_f16(fx, args[0], args[1], args[2]),


This conflicts with rust-lang/rustc_codegen_cranelift#1675. I'm planning to do a sync soon.

View changes since the review

oh then we can hold off from merging this until you've synced, i dont mind

@rustbot author

RalfJung

LLVM codegen changes LGTM.
Miri changes are missing, as CI is pointing out. :)
Personally I am fine with these macros, but it's clearly a tradeoff.

View changes since this review

RalfJung · 2026-08-16T18:10:43Z

+        const _: () = {
+            type $generic = $concrete;


If these are already in a module, why do you put them additionally into an anonymous const?

because we need to bind T for that impl, since the generated trait impls look like this:

impl Dispatch<f32> for f32 { #[inline] fn dispatch(x: T) -> T { libm::likely_available::expf(x) } }

Hm... I see, clever. The generic parameter in Dispatch is also part of this trick to keep the T working.

That's definitely worth a comment or three to explain what is going on.

RalfJung · 2026-08-16T18:12:00Z

+   |
+help: consider importing this function
+   |
+LL + use std::intrinsics::log;


This seems like a bad side-effect. Do we have an open issue to track the problem that apparently we recommend importing unstable functions?

not that i could fine, but not sure. agreed this seems pretty bad though, especially for a name as common as log

Hm, something odd is happening -- I don't get similar suggestions for existing intrinsics
Playground

RalfJung · 2026-08-16T18:19:20Z

moved a misplaced comment in compiler/rustc_codegen_llvm/src/intrinsic.rs that caused x fmt to give up

The comment was not misplaced, it was placed exactly where it should be. rustfmt just doesn't like comments in some places so we have to move the comment into a suboptimal place. :(

N1ark · 2026-08-17T08:48:02Z

moved a misplaced comment in compiler/rustc_codegen_llvm/src/intrinsic.rs that caused x fmt to give up

The comment was not misplaced, it was placed exactly where it should be. rustfmt just doesn't like comments in some places so we have to move the comment into a suboptimal place. :(

yes sorry, i meant misplaced from rustfmt's perspective ^^ i agree it's not ideal there, it just seemed clearer than inside the match arm:/

N1ark added 2 commits August 12, 2026 12:01

Cleanup cranelift codegen_float_intrinsic_call

295e7a3

Fix misplaced comment

84dc18d

rustbot added the A-LLVM Area: Code generation parts specific to LLVM. Both correctness bugs and optimization-related issues. label Aug 12, 2026

rustbot assigned folkertdev Aug 12, 2026

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. T-libs Relevant to the library team, which will review and decide on the PR/issue. labels Aug 12, 2026

This comment has been minimized.

Sign in to view

Make exp, exp2, log, log2, log10 generic

b3a59f6

N1ark force-pushed the 08-12-generic-float-intrinsincs branch from d98ca71 to b3a59f6 Compare August 12, 2026 14:55

rustbot added the T-clippy Relevant to the Clippy team. label Aug 12, 2026

bjorn3 reviewed Aug 12, 2026

View reviewed changes

rustbot added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Aug 12, 2026

RalfJung reviewed Aug 16, 2026

View reviewed changes

Uh oh!

Uh oh!

Conversation

N1ark commented Aug 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Aug 12, 2026

Uh oh!

folkertdev commented Aug 12, 2026

Uh oh!

This comment has been minimized.

rustbot commented Aug 12, 2026

Uh oh!

rust-log-analyzer commented Aug 12, 2026

Uh oh!

bjorn3 Aug 12, 2026 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

N1ark Aug 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

folkertdev Aug 12, 2026

Choose a reason for hiding this comment

Uh oh!

RalfJung left a comment • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

RalfJung Aug 16, 2026

Choose a reason for hiding this comment

Uh oh!

N1ark Aug 17, 2026

Choose a reason for hiding this comment

Uh oh!

RalfJung Aug 17, 2026

Choose a reason for hiding this comment

Uh oh!

RalfJung Aug 16, 2026

Choose a reason for hiding this comment

Uh oh!

N1ark Aug 17, 2026

Choose a reason for hiding this comment

Uh oh!

RalfJung Aug 17, 2026

Choose a reason for hiding this comment

Uh oh!

RalfJung commented Aug 16, 2026

Uh oh!

N1ark commented Aug 17, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

N1ark commented Aug 12, 2026 •

edited

Loading

bjorn3 Aug 12, 2026 •

edited by rustbot

Loading

N1ark Aug 12, 2026 •

edited

Loading

RalfJung left a comment •

edited by rustbot

Loading