[core/simd]: Write package documentation #4545
Open
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
Probably caused by actions/upload-artifact#602 Possibly fixes odin-lang#4384
Co-authored-by: Feoramund <[email protected]>
Co-authored-by: Feoramund <[email protected]>
Do not memory map files; Windows cannot write to a file that is memory mapped. Write cache after llvm building. This ensures the cache won't have a false positive if llvm fails.
This returns the number of runes in a u16 slice
* Fix odin-lang#4508 for abs, min, max and the rest of the builtins. None of these segfault now: ```odin package bug main :: proc() { p :: proc() {} // _ = len(p()) // _ = cap(p()) // _ = size_of(p()) // _ = align_of(p()) // T :: struct {} // _ = offset_of(p()) // _ = offset_of(T, p()) // _ = offset_of(p(), foo) // _ = offset_of(p(), "") // _ = type_of(p()) // _ = type_info_of(p()) // _ = typeid_of(p()) // A: [4]int // _ = swizzle(p()) // :: proc(x: [N]T, indices: ..int) -> [len(indices)]T --- // _ = swizzle(A, p()) // :: proc(x: [N]T, indices: ..int) -> [len(indices)]T --- // _ = complex(p(), p()) // _ = quaternion(p(), p(), p(), p()) // _ = quaternion(w=p(), x=p(), y=p(), z=p()) // _ = real(p()) // _ = imag(p()) // _ = jmag(p()) // _ = kmag(p()) // _ = conj(p()) // _ = expand_values(p()) // _ = min(p()) // _ = max(p()) // _ = abs(p()) // _ = clamp(p(), p(), p()) // _ = soa_zip(p()) // _ = soa_unzip(p()) } ```
Co-authored-by: Laytan <[email protected]>
And also moved Windows file procs to `os_windows.odin`, in line with all the other platforms who didn't have a dedicated `file_<platform>.odin`
laytan
reviewed
Dec 3, 2024
core/simd/simd.odin
Outdated
| /* | ||
| Check if SIMD is emulated on a target platform. | ||
|
|
||
| This value is `true`, if the compile-time target has the hardware support for |
There was a problem hiding this comment.
Suggested change
| This value is `true`, if the compile-time target has the hardware support for | |
| This value is `false`, if the compile-time target has the hardware support for |
core/simd/simd.odin
Outdated
|
|
||
| This value is `true`, if the compile-time target has the hardware support for | ||
| at 128-bit (or wider) SIMD. If the compile-time target lacks the hardware support | ||
| for 128-bit SIMD, this value is `false`, and all SIMD operations will be |
There was a problem hiding this comment.
Suggested change
| for 128-bit SIMD, this value is `false`, and all SIMD operations will be | |
| for 128-bit SIMD, this value is `true`, and all SIMD operations will likely be |
Co-authored-by: Laytan <[email protected]>
flysand7
commented
Dec 4, 2024
| +-------+-------+-------+-------+ | ||
| res: | ||
| +-------+-------+-------+--------+ | ||
| | 0x44 | 0xaa | 0x06 | 0xfe | |
There was a problem hiding this comment.
Okay this one I have to refuse, I was assuming 1-byte lanes, which I should probably make explicit in the comments.
core/simd/simd.odin
Outdated
| /* | ||
| Saturated addition of SIMD vectors. | ||
|
|
||
| The *saturated sum* is a sum, that upon overflow or underflow, instead of |
There was a problem hiding this comment.
Yes, I should rewrite this.
core/simd/simd.odin
Outdated
| res := simd.gather(ptrs, defaults, mask) | ||
| fmt.println(res) | ||
|
|
||
| The code would print `<2, 127, 10, 127>`. First and the third positions came |
There was a problem hiding this comment.
They are indented with a tab. Thanks for Output: tip!
Co-authored-by: Laytan <[email protected]>
gingerBill
reviewed
Dec 4, 2024
| +------+------+------+------+ | ||
| | 0 | 1 | 0.33 | 0.2 | | ||
| +------+------+------+------+ | ||
| */ |
There was a problem hiding this comment.
1/0 may not result in 0. That's a poor example.
|
Did some fixes, I think this one's ready for the final review. |
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.
No description provided.