YourWishes/cglm
1
0
Fork 0
mirror of https://github.com/recp/cglm.git synced 2026-02-17 03:39:05 +00:00
Code Issues Packages Projects Releases Wiki Activity

simd128: enable in mat4

Browse Source
This commit is contained in:
myfreeer
2023-03-06 16:40:45 +08:00
parent 5d60c17435
commit 84c521c203
3 changed files with 70 additions and 39 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
include/cglm/simd/wasm.h
View File

@@ -83,6 +83,18 @@ _mm_set_ps(float __z, float __y, float __x, float __w)
return (glmm_128)wasm_f32x4_make(__w, __x, __y, __z);
}
static inline glmm_128 __attribute__((__always_inline__, __nodebug__))
_mm_sqrt_ss(glmm_128 __a)
{
return wasm_i32x4_shuffle(__a, wasm_f32x4_sqrt(__a), 4, 1, 2, 3);
}
static __inline__ glmm_128 __attribute__((__always_inline__, __nodebug__))
_mm_rcp_ps(glmm_128 __a)
{
return (glmm_128)wasm_f32x4_div((v128_t)wasm_f32x4_splat(1.0f), (v128_t)__a);
}
#define _MM_TRANSPOSE4_PS(row0, row1, row2, row3) \
do { \
glmm_128 __row0 = (row0); \
@@ -184,12 +196,6 @@ glmm_dot(glmm_128 a, glmm_128 b) {
return _mm_cvtss_f32(glmm_vdots(a, b));
}
static inline glmm_128 __attribute__((__always_inline__, __nodebug__))
_mm_sqrt_ss(glmm_128 __a)
{
return wasm_i32x4_shuffle(__a, wasm_f32x4_sqrt(__a), 4, 1, 2, 3);
}
static inline
float
glmm_norm(glmm_128 a) {
@@ -234,11 +240,6 @@ glmm_div(glmm_128 a, glmm_128 b) {
return wasm_f32x4_div(a, b);
}
/* enable FMA macro for MSVC? */
#if defined(_MSC_VER) && !defined(__FMA__) && defined(__AVX2__)
# define __FMA__ 1
#endif
static inline
glmm_128
glmm_fmadd(glmm_128 a, glmm_128 b, glmm_128 c) {