You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The WebAssembly spec already supports 128-bit fixed-width SIMD opcodes, see the spec and the clang builtins for details.
I'm wondering if the simdutf maintainers are interested in accepting a port to WebAssembly SIMD128. This won't be worked on immediately (for Safari it's only in TP and not release yet), but having a tracking ticket is good anyway, and I'm curious to hear what you folks think :)
The text was updated successfully, but these errors were encountered:
WASM is a blatant copy of very similar to NEON (just without half vectors), so it would probably be an easy port — especially since the emscripten toolchain includes a polyfill arm_neon.h from SIMDe.
I know that for xxHash, the arm_neon.h header had solid performance with literally no changes, so it could probably get away with only a few specializations. (e.g. alternative paths when half vectors are used)
The WebAssembly spec already supports 128-bit fixed-width SIMD opcodes, see the spec and the clang builtins for details.
I'm wondering if the simdutf maintainers are interested in accepting a port to WebAssembly SIMD128. This won't be worked on immediately (for Safari it's only in TP and not release yet), but having a tracking ticket is good anyway, and I'm curious to hear what you folks think :)
The text was updated successfully, but these errors were encountered: