Scalar vector operations can use NEON instructions, which have incorrect subnormal behavior
Open
Nobody has claimed this yet.
A-floating-point
A-LLVM
C-bug
T-compiler
- Dominant language
- Rust
- Stars
- 119k
- Forks
- 16.1k
- PR merge metrics
- PR metrics pending
Description
This is the Rust version of https://github.com/llvm/llvm-project/issues/106909: scalar f32/f64 operations can produce the wrong result for subnormals on ARM if NEON instructions are used.
This just got fixed in upstream LLVM, but we should also have a test covering this.
Cc @nikic
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reading the linked upstream LLVM issue and locating Rust's compiler test entry point for ARM NEON scalar f32/f64 operations. Add coverage for the incorrect subnormal result, with the test passing once the relevant behavior is verified against the upstream fix.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- compilers
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100