Can non-temporal loads use the LLVM intrinsic instead?
Open
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 694
- Forks
- 336
- Avg merge
- 5d 14h
- Merged PRs (30d)
- 9
Description
According to @Amanieu:
I checked and only non-temporal stores have special memory orderings on x86. x86 non-temporal loads work just like normal loads.
Using inline assembly produces suboptimal code, see https://godbolt.org/z/os8bG3Efr.
References:
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 locating the non-temporal load implementation and compare its generated code with the LLVM intrinsic approach discussed in the issue and linked references. Use the provided Godbolt example to verify that the change produces better code, then run the relevant existing tests and confirm non-temporal loads retain their expected behavior.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- compilers, performance
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 48/100