External/cglm
1
0
Fork 0
mirror of https://github.com/recp/cglm.git synced 2025-10-04 01:00:46 +00:00
Code Issues Packages Projects Releases Wiki Activity

simd, sse: use 0x80000000 insteaf of -0.f to fix fastmath on msvc

Browse Source
This commit is contained in:
Recep Aslantas
2023-04-21 20:36:50 +03:00
parent 391d8670c2
commit 7307b1cbbe
1 changed files with 4 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
include/cglm/simd/x86.h
View File

@@ -65,11 +65,12 @@
#define glmm_float32x4_SIGNMASK_NPPN GLMM__SIGNMASKf(GLMM_NEGZEROf, 0, 0, GLMM_NEGZEROf) #define glmm_float32x4_SIGNMASK_NPPN GLMM__SIGNMASKf(GLMM_NEGZEROf, 0, 0, GLMM_NEGZEROf)
#define glmm_float32x4_SIGNMASK_NEG _mm_set1_ps(GLMM_NEGZEROf) #define glmm_float32x4_SIGNMASK_NEG _mm_set1_ps(GLMM_NEGZEROf)
#define glmm_float32x8_SIGNMASK_NEG _mm256_castsi256_ps(_mm256_set1_epi32(GLMM_NEGZEROf));
static inline static inline
__m128 __m128
glmm_abs(__m128 x) { glmm_abs(__m128 x) {
return _mm_andnot_ps(_mm_set1_ps(0x80000000/*-0.0f*/), x); return _mm_andnot_ps(glmm_float32x4_SIGNMASK_NEG, x);
} }
static inline static inline
@@ -269,7 +270,7 @@ glmm_fnmsub(__m128 a, __m128 b, __m128 c) {
return _mm_fnmsub_ps(a, b, c); return _mm_fnmsub_ps(a, b, c);
#else #else
return _mm_xor_ps(_mm_add_ps(_mm_mul_ps(a, b), c), return _mm_xor_ps(_mm_add_ps(_mm_mul_ps(a, b), c),
_mm_set1_ps(0x80000000/*-0.0f*/)); glmm_float32x4_SIGNMASK_NEG);
#endif #endif
} }
@@ -311,7 +312,7 @@ glmm256_fnmsub(__m256 a, __m256 b, __m256 c) {
return _mm256_fmsub_ps(a, b, c); return _mm256_fmsub_ps(a, b, c);
#else #else
return _mm256_xor_ps(_mm256_sub_ps(_mm256_mul_ps(a, b), c), return _mm256_xor_ps(_mm256_sub_ps(_mm256_mul_ps(a, b), c),
_mm256_set1_ps(0x80000000/*-0.0f*/)); glmm_float32x8_SIGNMASK_NEG);
#endif #endif
} }
#endif #endif