myfreeer 009405adcd wasm: prefer pmin/pmax ...

According to [emscripten](https://emscripten.org/docs/porting/simd.html) and [v8](b6520eda5e/src/compiler/backend/x64/code-generator-x64.cc (L2661-L2699)), `[f32x4|f64x2].[min|max]` compiles to much more instructions than `[f32x4|f64x2].[pmin|pmax]`.
It is defined in [spec](https://github.com/WebAssembly/spec/blob/main/proposals/simd/SIMD.md#floating-point-min-and-max) that the difference between pmin/pmax and min/max is NaN-propagating behavior, and the equivalent to the x86 `_mm_min_ps`/`_mm_max_ps` is pmin/pmax in [v8](b6520eda5e/src/compiler/backend/x64/code-generator-x64.cc (L2740-L2747)).
This should make functions with min/max faster on webassembly, and align with the existing behavior with x86 sse.

2023-12-02 09:54:49 +08:00

..

cglm

wasm: prefer pmin/pmax

2023-12-02 09:54:49 +08:00

module.modulemap

Added simple swiftpm package file with module map and empty C file to trick the compiler to build the empty C target.

2020-09-05 20:39:12 -07:00