rustc_target: callconv: powerpc64: Use the ABI set in target options instead of guessing by Gelbpunkt · Pull Request #150468 · rust-lang/rust

Gelbpunkt · 2025-12-28T21:51:29Z

All PowerPC64 targets except AIX explicitly set the ABI in the target options. We can therefore stop hardcoding the ABI to be used based on the target environment or OS, except for the AIX special case.

The fallback based on endianness is kept for the sake of compatibility with custom targets.

This makes it so that big endian targets not explicitly accounted for before (powerpc64-unknown-openbsd) and targets that don't use the expected default ABI (big-endian ELFv2 Glibc targets) use the correct ABI in the calling convention code.

The second commit is a tiny change to validate the llvm_abiname set on PowerPC64(LE) targets. See the commit messages for details.

CC @RalfJung who pointed out the missing llvm_abiname validation

rustbot · 2025-12-28T21:51:32Z

These commits modify compiler targets.
(See the Target Tier Policy.)

rustbot · 2025-12-28T21:51:35Z

r? @chenyukang

rustbot has assigned @chenyukang.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

chenyukang · 2026-01-13T08:21:38Z

@rustbot reroll

jieyouxu · 2026-01-13T08:25:55Z

@rustbot reroll

compiler/rustc_target/src/spec/mod.rs

…instead of guessing All PowerPC64 targets except AIX explicitly set the ABI in the target options. We can therefore stop hardcoding the ABI to be used based on the target environment or OS, except for the AIX special case. The fallback based on endianness is kept for the sake of compatibility with custom targets. This makes it so that big endian targets not explicitly accounted for before (powerpc64-unknown-openbsd) and targets that don't use the expected default ABI (ELFv2 Glibc targets) use the correct ABI in the calling convention code.

rustbot · 2026-01-13T14:56:13Z

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

compiler/rustc_target/src/spec/mod.rs

… PowerPC64(LE) PowerPC64 ELF targets (effectively anything that isn't AIX) use either the ELFv1 or ELFv2 ABI. The ELFv1 ABI is only specified for big endian targets, while ELFv2 can be used by both little- and big-endian targets. Make sure that, if an LLVM ABI is set, it is set to one of the two. AIX does not set an LLVM ABI name, so ensure that AIX targets don't set anything other than an empty ABI name.

fee1-dead · 2026-01-18T01:37:07Z

@rustbot reroll

RalfJung · 2026-01-18T14:01:47Z

r? me

This LGTM from by 10k foot view without knowing anything about the details of these ABIs ;) . However, do we have to worry about this breaking the ABI for anyone already using these targets?

Also, @workingjubilee I wouldn't mind a second opinion. :)

Gelbpunkt · 2026-01-18T14:22:18Z

However, do we have to worry about this breaking the ABI for anyone already using these targets?

Previously, this hardcoded the list of big-endian targets using ELFv2 to musl and FreeBSD, which was not kept up to date (OpenBSD is missing from here). Since all PowerPC64 targets now specify their ABI, we can stop hardcoding the list. It will not break the ABI for any of them, AFAICT this is strictly a fix for the OpenBSD target and makes it so there's one less part of the code we need to worry about when changing or adding new targets.

jhpratt · 2026-02-22T06:40:23Z

yielding to encompassing rollup

@bors yield

rust-bors · 2026-02-22T06:40:26Z

Auto build cancelled. Cancelled workflows:

https://github.com/rust-lang/rust/actions/runs/22272126794

The next pull request likely to be tested is #152957.

Rollup of 13 pull requests Successful merges: - #150468 (rustc_target: callconv: powerpc64: Use the ABI set in target options instead of guessing) - #151628 (Fix ICE in const eval of packed SIMD types with non-power-of-two element counts) - #151871 (don't use env with infer vars) - #152591 (Simplify internals of `{Rc,Arc}::default`) - #152865 (Fixed ByteStr not padding within its Display trait when no specific alignment is mentioned) - #152908 (Enable rust.remap-debuginfo in the dist profile) - #152705 (Test(lib/win/proc): Skip `raw_attributes` doctest under Win7) - #152767 (fix typo in `carryless_mul` macro invocation) - #152837 (fix(codegen): Use `body_codegen_attrs` For Caller In `adjust_target_feature_sig`) - #152871 (Fix warnings in rs{begin,end}.rs files) - #152921 (Add build.rustdoc option to bootstrap config) - #152933 (Start migration for `LintDiagnostic` items by adding API and migrating `LinkerOutput` lint) - #152937 (remove unneeded reboxing)

rust-bors · 2026-02-22T08:18:42Z

⌛ Testing commit 168f324 with merge 7c35f94...

Workflow: https://github.com/rust-lang/rust/actions/runs/22273503822

@RalfJung

… r=RalfJung rustc_target: callconv: powerpc64: Use the ABI set in target options instead of guessing All PowerPC64 targets except AIX explicitly set the ABI in the target options. We can therefore stop hardcoding the ABI to be used based on the target environment or OS, except for the AIX special case. The fallback based on endianness is kept for the sake of compatibility with custom targets. This makes it so that big endian targets not explicitly accounted for before (powerpc64-unknown-openbsd) and targets that don't use the expected default ABI (big-endian ELFv2 Glibc targets) use the correct ABI in the calling convention code. The second commit is a tiny change to validate the `llvm_abiname` set on PowerPC64(LE) targets. See the commit messages for details. CC @RalfJung who pointed out the missing `llvm_abiname` validation

@RalfJung

…ec-elf-abi, r=RalfJung rustc_target: callconv: powerpc64: Use the ABI set in target options instead of guessing All PowerPC64 targets except AIX explicitly set the ABI in the target options. We can therefore stop hardcoding the ABI to be used based on the target environment or OS, except for the AIX special case. The fallback based on endianness is kept for the sake of compatibility with custom targets. This makes it so that big endian targets not explicitly accounted for before (powerpc64-unknown-openbsd) and targets that don't use the expected default ABI (big-endian ELFv2 Glibc targets) use the correct ABI in the calling convention code. The second commit is a tiny change to validate the `llvm_abiname` set on PowerPC64(LE) targets. See the commit messages for details. CC @RalfJung who pointed out the missing `llvm_abiname` validation

jhpratt · 2026-02-22T08:20:49Z

once again

@bors yield

rust-bors · 2026-02-22T08:20:52Z

Auto build cancelled. Cancelled workflows:

https://github.com/rust-lang/rust/actions/runs/22273503822

The next pull request likely to be tested is #152960.

Rollup of 15 pull requests Successful merges: - #150468 (rustc_target: callconv: powerpc64: Use the ABI set in target options instead of guessing) - #151628 (Fix ICE in const eval of packed SIMD types with non-power-of-two element counts) - #151871 (don't use env with infer vars) - #152591 (Simplify internals of `{Rc,Arc}::default`) - #152865 (Fixed ByteStr not padding within its Display trait when no specific alignment is mentioned) - #152908 (Enable rust.remap-debuginfo in the dist profile) - #147859 (reduce the amount of panics in `{TokenStream, Literal}::from_str` calls) - #152705 (Test(lib/win/proc): Skip `raw_attributes` doctest under Win7) - #152767 (fix typo in `carryless_mul` macro invocation) - #152837 (fix(codegen): Use `body_codegen_attrs` For Caller In `adjust_target_feature_sig`) - #152871 (Fix warnings in rs{begin,end}.rs files) - #152879 (Remove `impl IntoQueryParam<P> for &'a P`.) - #152933 (Start migration for `LintDiagnostic` items by adding API and migrating `LinkerOutput` lint) - #152937 (remove unneeded reboxing) - #152953 (Fix typo in armv7a-vex-v5.md)

rust-bors · 2026-02-22T10:05:37Z

⌛ Testing commit 168f324 with merge e59a71a...

Workflow: https://github.com/rust-lang/rust/actions/runs/22275041573

@RalfJung

… r=RalfJung rustc_target: callconv: powerpc64: Use the ABI set in target options instead of guessing All PowerPC64 targets except AIX explicitly set the ABI in the target options. We can therefore stop hardcoding the ABI to be used based on the target environment or OS, except for the AIX special case. The fallback based on endianness is kept for the sake of compatibility with custom targets. This makes it so that big endian targets not explicitly accounted for before (powerpc64-unknown-openbsd) and targets that don't use the expected default ABI (big-endian ELFv2 Glibc targets) use the correct ABI in the calling convention code. The second commit is a tiny change to validate the `llvm_abiname` set on PowerPC64(LE) targets. See the commit messages for details. CC @RalfJung who pointed out the missing `llvm_abiname` validation

@RalfJung

…ec-elf-abi, r=RalfJung rustc_target: callconv: powerpc64: Use the ABI set in target options instead of guessing All PowerPC64 targets except AIX explicitly set the ABI in the target options. We can therefore stop hardcoding the ABI to be used based on the target environment or OS, except for the AIX special case. The fallback based on endianness is kept for the sake of compatibility with custom targets. This makes it so that big endian targets not explicitly accounted for before (powerpc64-unknown-openbsd) and targets that don't use the expected default ABI (big-endian ELFv2 Glibc targets) use the correct ABI in the calling convention code. The second commit is a tiny change to validate the `llvm_abiname` set on PowerPC64(LE) targets. See the commit messages for details. CC @RalfJung who pointed out the missing `llvm_abiname` validation

JonathanBrouwer · 2026-02-22T10:31:36Z

once again

@bors yield

rust-bors · 2026-02-22T10:31:39Z

Auto build cancelled. Cancelled workflows:

https://github.com/rust-lang/rust/actions/runs/22275041573

The next pull request likely to be tested is #152965.

…uwer Rollup of 14 pull requests Successful merges: - #150468 (rustc_target: callconv: powerpc64: Use the ABI set in target options instead of guessing) - #151628 (Fix ICE in const eval of packed SIMD types with non-power-of-two element counts) - #151871 (don't use env with infer vars) - #152591 (Simplify internals of `{Rc,Arc}::default`) - #152865 (Fixed ByteStr not padding within its Display trait when no specific alignment is mentioned) - #147859 (reduce the amount of panics in `{TokenStream, Literal}::from_str` calls) - #152705 (Test(lib/win/proc): Skip `raw_attributes` doctest under Win7) - #152767 (fix typo in `carryless_mul` macro invocation) - #152837 (fix(codegen): Use `body_codegen_attrs` For Caller In `adjust_target_feature_sig`) - #152871 (Fix warnings in rs{begin,end}.rs files) - #152879 (Remove `impl IntoQueryParam<P> for &'a P`.) - #152933 (Start migration for `LintDiagnostic` items by adding API and migrating `LinkerOutput` lint) - #152937 (remove unneeded reboxing) - #152953 (Fix typo in armv7a-vex-v5.md)

@RalfJung

Rollup merge of #150468 - Gelbpunkt:rustc-target-callconv-spec-elf-abi, r=RalfJung rustc_target: callconv: powerpc64: Use the ABI set in target options instead of guessing All PowerPC64 targets except AIX explicitly set the ABI in the target options. We can therefore stop hardcoding the ABI to be used based on the target environment or OS, except for the AIX special case. The fallback based on endianness is kept for the sake of compatibility with custom targets. This makes it so that big endian targets not explicitly accounted for before (powerpc64-unknown-openbsd) and targets that don't use the expected default ABI (big-endian ELFv2 Glibc targets) use the correct ABI in the calling convention code. The second commit is a tiny change to validate the `llvm_abiname` set on PowerPC64(LE) targets. See the commit messages for details. CC @RalfJung who pointed out the missing `llvm_abiname` validation

Gelbpunkt · 2026-02-22T21:27:04Z

Also, @Gelbpunkt do you think this needs to be mentioned in the release notes?

I'm a bit late since you already added the label, sorry! It doesn't hurt to mention it, specifically since this should be a change in behavior for OpenBSD, even if it is a fix. Thanks for getting back to the PR!

…uwer Rollup of 14 pull requests Successful merges: - rust-lang/rust#150468 (rustc_target: callconv: powerpc64: Use the ABI set in target options instead of guessing) - rust-lang/rust#151628 (Fix ICE in const eval of packed SIMD types with non-power-of-two element counts) - rust-lang/rust#151871 (don't use env with infer vars) - rust-lang/rust#152591 (Simplify internals of `{Rc,Arc}::default`) - rust-lang/rust#152865 (Fixed ByteStr not padding within its Display trait when no specific alignment is mentioned) - rust-lang/rust#147859 (reduce the amount of panics in `{TokenStream, Literal}::from_str` calls) - rust-lang/rust#152705 (Test(lib/win/proc): Skip `raw_attributes` doctest under Win7) - rust-lang/rust#152767 (fix typo in `carryless_mul` macro invocation) - rust-lang/rust#152837 (fix(codegen): Use `body_codegen_attrs` For Caller In `adjust_target_feature_sig`) - rust-lang/rust#152871 (Fix warnings in rs{begin,end}.rs files) - rust-lang/rust#152879 (Remove `impl IntoQueryParam<P> for &'a P`.) - rust-lang/rust#152933 (Start migration for `LintDiagnostic` items by adding API and migrating `LinkerOutput` lint) - rust-lang/rust#152937 (remove unneeded reboxing) - rust-lang/rust#152953 (Fix typo in armv7a-vex-v5.md)

RalfJung · 2026-02-23T18:00:12Z

compiler/rustc_target/src/callconv/powerpc64.rs

+    let abi = if cx.target_spec().options.abi == Abi::ElfV2 {
        ELFv2
+    } else if cx.target_spec().options.abi == Abi::ElfV1 {
+        ELFv1


Oh, I only just realized that this could also check llvm_abiname -- which is what actually gets sanity-checked in the target spec logic, and it is what actually tells LLVM what to do, so it's the more reliable source of truth.

Can you submit a PR to check llvm_abiname here instead?

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. labels Dec 28, 2025

rustbot assigned chenyukang Dec 28, 2025

Gelbpunkt force-pushed the rustc-target-callconv-spec-elf-abi branch from 2106bd3 to 35377f3 Compare December 28, 2025 21:56

RalfJung mentioned this pull request Jan 9, 2026

added support for s390x target features soft-float and packed-stack #150766

Closed

rustbot assigned jieyouxu and unassigned chenyukang Jan 13, 2026

rustbot assigned fee1-dead and unassigned jieyouxu Jan 13, 2026

taiki-e reviewed Jan 13, 2026

View reviewed changes

compiler/rustc_target/src/spec/mod.rs Outdated Show resolved Hide resolved

Gelbpunkt force-pushed the rustc-target-callconv-spec-elf-abi branch from 35377f3 to b2e1f69 Compare January 13, 2026 14:56

taiki-e reviewed Jan 13, 2026

View reviewed changes

compiler/rustc_target/src/spec/mod.rs Outdated Show resolved Hide resolved

Gelbpunkt force-pushed the rustc-target-callconv-spec-elf-abi branch from b2e1f69 to d40f0cd Compare January 13, 2026 15:33

taiki-e reviewed Jan 13, 2026

View reviewed changes

compiler/rustc_target/src/spec/mod.rs Outdated Show resolved Hide resolved

Gelbpunkt force-pushed the rustc-target-callconv-spec-elf-abi branch from d40f0cd to 168f324 Compare January 15, 2026 03:00

taiki-e approved these changes Jan 15, 2026

View reviewed changes

rustbot assigned davidtwco and unassigned fee1-dead Jan 18, 2026

rustbot assigned RalfJung and unassigned davidtwco Jan 18, 2026

jhpratt mentioned this pull request Feb 22, 2026

Rollup of 13 pull requests #152957

Closed

jhpratt mentioned this pull request Feb 22, 2026

Rollup of 15 pull requests #152960

Closed

This was referenced Feb 22, 2026

Rollup of 13 pull requests #152964

Closed

Rollup of 14 pull requests #152965

Merged

RalfJung added the relnotes Marks issues that should be documented in the release notes of the next release. label Feb 22, 2026

rustbot mentioned this pull request Feb 22, 2026

Tracking issue for release notes of #150468: rustc_target: callconv: powerpc64: Use the ABI set in target options instead of guessing #152975

Open

rust-bors bot merged commit 7360a83 into rust-lang:main Feb 22, 2026
11 of 12 checks passed

rustbot added this to the 1.95.0 milestone Feb 22, 2026

RalfJung reviewed Feb 23, 2026

View reviewed changes

Gelbpunkt mentioned this pull request Feb 24, 2026

rustc_target: callconv: powerpc64: Use llvm_abiname rather than target_abi for ABI determination #153035

Open

Uh oh!

Comments

Conversation

Gelbpunkt commented Dec 28, 2025 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Dec 28, 2025

Uh oh!

rustbot commented Dec 28, 2025

Uh oh!

chenyukang commented Jan 13, 2026

Uh oh!

jieyouxu commented Jan 13, 2026

Uh oh!

Uh oh!

rustbot commented Jan 13, 2026

Uh oh!

Uh oh!

Uh oh!

fee1-dead commented Jan 18, 2026

Uh oh!

RalfJung commented Jan 18, 2026

Uh oh!

Gelbpunkt commented Jan 18, 2026

Uh oh!

jhpratt commented Feb 22, 2026

Uh oh!

rust-bors bot commented Feb 22, 2026

Uh oh!

rust-bors bot commented Feb 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jhpratt commented Feb 22, 2026

Uh oh!

rust-bors bot commented Feb 22, 2026

Uh oh!

rust-bors bot commented Feb 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

JonathanBrouwer commented Feb 22, 2026

Uh oh!

rust-bors bot commented Feb 22, 2026

Uh oh!

Uh oh!

Gelbpunkt commented Feb 22, 2026

Uh oh!

RalfJung Feb 23, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

11 participants

Gelbpunkt commented Dec 28, 2025 •

edited by rustbot

Loading

rust-bors bot commented Feb 22, 2026 •

edited

Loading

rust-bors bot commented Feb 22, 2026 •

edited

Loading