Skip to content

Add saturating_add() and saturating_sub() to integers #352

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

Merged
Shnatsel merged 10 commits into from
Sep 3, 2026
Merged

Add saturating_add() and saturating_sub() to integers #352

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
38db25d
Initial pass at the saturating add. The codegen is good but the gener…
Shnatsel Aug 26, 2026
72d7e94
Refactor x86 saturating add generator to use a single large toplevel …
Shnatsel Aug 26, 2026
17d2d5c
Collapse three toplevel branches into one with shared structure
Shnatsel Aug 26, 2026
4ea5a4a
Use a use statement for nicer formatting
Shnatsel Aug 26, 2026
dea8646
Add saturating_sub
Shnatsel Aug 26, 2026
cc301a0
improve docs
Shnatsel Aug 26, 2026
711a596
address clippy lints
Shnatsel Aug 26, 2026
7b249a1
fix typo
Shnatsel Aug 26, 2026
10d2909
Merge branch 'main' into saturating-add
Shnatsel Sep 3, 2026
78e9100
Improve doc string
Shnatsel Sep 3, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,7 @@ You can find its changes [documented below](#070-2026-08-11).
### Added

- Added `reverse` for all SIMD vector and mask types.
- Added lane-wise `saturating_add` and `saturating_sub` for all integer vector types and backends.
- Added lane-wise `count_ones` and `count_zeros` operations for all integer vector types and backends.
- Added `mul_add_precise` and `mul_sub_precise` for floating-point vectors. They guarantee the infinite-precision product-plus-add rounded once, including on SIMD levels without hardware fused multiply-add instructions. They are not susceptible to the [bug](https://github.com/rust-lang/compiler-builtins/issues/1262) in Rust standard library, `std::simd` and musl libc that causes incorrect rounding for subnormal results. SSE4.2 gets SIMD emulation of these operations for better performance. ([#323][], [#324][] by [@Shnatsel][])
- Documented the storage representation of the SIMD vector types. The documented representation will not change without a semver major version change.
Expand Down
Loading
Loading