Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

simd_scatter: implement correct order of stores #3235

Closed
RalfJung opened this issue Dec 21, 2023 · 1 comment · Fixed by #3237
Closed

simd_scatter: implement correct order of stores #3235

RalfJung opened this issue Dec 21, 2023 · 1 comment · Fixed by #3237

Comments

@RalfJung
Copy link
Member

RalfJung commented Dec 21, 2023 •
edited
Loading

If we follow LLVM semantics, the stores have to happen "from least significant to most significant", which I think means the order differs between little-endian and big-endian.

Or we may want to make overlapping stores UB in our intrinsics.

Also see the Zulip discussion.
@RalfJung
Copy link
Member Author

Turns out it means "left-to-right" in the vector, so what we do is good. But we should add a test.

@RalfJung RalfJung mentioned this issue Dec 22, 2023
Merged
@bors bors closed this as completed in fcd54c5 Dec 22, 2023
RalfJung pushed a commit to RalfJung/rust that referenced this issue Dec 26, 2023
@bors
Auto merge of rust-lang#3237 - RalfJung:simd-loadstore, r=RalfJung
2c7e0fd 
implement and test simd_masked_load and simd_masked_store

also extend the scatter/gather tests

Fixes rust-lang/miri#3235
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

implement and test simd_masked_load and simd_masked_store RalfJung/miri
1 participant
@RalfJung