RESOLVED FIXED 248549
[WebAssembly SIMD] Support integer and floating-point arithmetic on Intel
https://bugs.webkit.org/show_bug.cgi?id=248549
Summary [WebAssembly SIMD] Support integer and floating-point arithmetic on Intel
David Degazio
Reported 2022-11-30 13:04:29 PST
Adds support for basic SIMD integer arithmetic and floating-point arithmetic instructions on Intel.
Attachments
David Degazio
Comment 1 2022-11-30 13:33:13 PST
David Degazio
Comment 2 2022-12-01 16:49:07 PST
For reference, this means adding support for the following instructions: - i8x16.add, i16x8.add, i32x4.add, i64x2.add - i8x16.sub, i16x8.sub, i32x4.sub, f64x2.sub - i16x8.mul, i32x4.mul, i64x2.mul - i32x4.dot_i16x8_s - i8x16.neg, i16x8.neg, i32x4.neg, i64x2.neg - f32x4.add, f64x2.add - f32x4.sub, f64x2.sub - f32x4.mul, f64x2.mul - f32x4.div, f64x2.div - f32x4.sqrt, f64x2.sqrt - f32x4.ceil, f64x2.ceil - f32x4.floor, f64x2.floor - f32x4.trunc, f64x2.trunc - f32x4.nearest, f64x2.nearest
Mark Lam
Comment 3 2022-12-07 10:56:45 PST
EWS
Comment 4 2022-12-07 15:02:35 PST
Committed 257511@main (155740a50205): <https://commits.webkit.org/257511@main> Reviewed commits have been landed. Closing PR #6984 and removing active labels.
Note You need to log in before you can comment on or make changes to this bug.