lane -> element for core::simd::Simd
#338
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Also yes, I realize it's slightly confusing to not change the actual function names where applicable, yet, but we were already inconsistent. I feel hashing out a bit of how the documentation should look is a necessary first step to updating any of the function names still not yet renamed. |
programmerjake
approved these changes
Apr 10, 2023
Lokathor
reviewed
Apr 10, 2023
programmerjake
approved these changes
Apr 11, 2023
workingjubilee
force-pushed
the
simdty-docs-using-elements
branch
from
c3d6f33 to
aebf287
Compare
workingjubilee
commented
Apr 11, 2023
crates/core_simd/src/vector.rs
Outdated
Comment on lines
34
to
42
| /// More complex elementwise operations may join vectors with masks which "enable" or "disable" | ||
| /// appropriate elements. These often involve a `Simd<T, N>` where `T` is `usize`, `*const T`, | ||
| /// or `*mut T`, and "enabling" means to use the index or pointer. | ||
| /// The resulting operation is like a mutating iteration: | ||
| /// | ||
| /// ```rust | ||
| /// # #![feature(portable_simd)] | ||
| /// # use core::simd::Simd; | ||
| /// let source: Vec<i32> = (0..100).into_iter().collect(); |
There was a problem hiding this comment.
This example is clearly unfinished, I am removing it for now since it's entirely greenfield.
workingjubilee
force-pushed
the
simdty-docs-using-elements
branch
from
10d53e4 to
db04eda
Compare
A while ago we began saying T, N instead of T, LANES in reference to Simd. At some point that leaked in to us checking in code with const N: usize. After a while, we had a discussion and agreed that "lanes", while common, is unnecessary jargon for Rust learners who aren't familiar with SIMD, and is fully interchangeable with terms for arrays like element and index. But we never acted on that. Let's update the main type's docs, at least. The example tweaks also enable removing a slated-for-removal nightly fn.
Saying "elementwise (non-)equal" may suggest it returns a vector. The comments should be clear that it instead reduces to a scalar. Co-authored-by: Jacob Lifshay <[email protected]>
workingjubilee
force-pushed
the
simdty-docs-using-elements
branch
from
db04eda to
4064678
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
A while ago we began saying T, N instead of T, LANES in reference to Simd. At some point that leaked in to us checking in code with const N: usize. After a while, we had a discussion and agreed that "lanes", while common, is unnecessary jargon for Rust learners who aren't familiar with SIMD, and is fully interchangeable with terms for arrays like element and index.
But we never acted on that. Let's update the main type's docs, at least. VL has been proposed here because Simd appears beside [T; N] often, so getting ahead on assigning a name to Simd's specific N may help? The example tweaks enable removing a slated-for-removal nightly fn.
I have the rest of this as it would go across the rest of the module's API docs waiting in the wings, but I figured that any kibitzing over how this should be done should be handled on Simd's docs because it's pretty procedurally applied elsewhere. There's basically "how we talk about Simd" and "how we talk about Mask" as the two major points of interest, so once both of those are done, and they can be mostly-separate discussions, it's robotic.