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

simd, sse2: make alignment optional for load operations

Browse Source
This commit is contained in:
Recep Aslantas
2018-05-08 15:25:23 +03:00
parent 0f339c5c03
commit 252bf925fc
9 changed files with 118 additions and 110 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
include/cglm/simd/intrin.h
View File

@@ -35,7 +35,7 @@
_mm_shuffle1_ps(_mm_shuffle_ps(a, b, _MM_SHUFFLE(z0, y0, x0, w0)), \
z1, y1, x1, w1)
CGLM_INLINE
static inline
__m128
glm_simd_dot(__m128 a, __m128 b) {
__m128 x0;
@@ -44,7 +44,7 @@ glm_simd_dot(__m128 a, __m128 b) {
return _mm_add_ps(x0, _mm_shuffle1_ps(x0, 0, 1, 0, 1));
}
CGLM_INLINE
static inline
__m128
glm_simd_norm(__m128 a) {
return _mm_sqrt_ps(glm_simd_dot(a, a));
@@ -69,6 +69,14 @@ glm_simd_store_v3(__m128 vx, vec3 v) {
_mm_store_ss(&v[2], _mm_shuffle1_ps(vx, 2, 2, 2, 2));
}
#ifdef CGLM_ALL_UNALIGNED
#define glmm_load(p) _mm_loadu_ps(p)
#define glmm_store(p, a) _mm_storeu_ps(p, a)
#else
#define glmm_load(p) _mm_load_ps(p)
#define glmm_store(p, a) _mm_store_ps(p, a)
#endif
#endif
/* x86, x64 */

38
include/cglm/simd/sse2/affine.h
View File

@@ -18,30 +18,30 @@ glm_mul_sse2(mat4 m1, mat4 m2, mat4 dest) {
/* D = R * L (Column-Major) */
__m128 l0, l1, l2, l3, r;
l0 = _mm_load_ps(m1[0]);
l1 = _mm_load_ps(m1[1]);
l2 = _mm_load_ps(m1[2]);
l3 = _mm_load_ps(m1[3]);
l0 = glmm_load(m1[0]);
l1 = glmm_load(m1[1]);
l2 = glmm_load(m1[2]);
l3 = glmm_load(m1[3]);
r = _mm_load_ps(m2[0]);
r = glmm_load(m2[0]);
_mm_store_ps(dest[0],
_mm_add_ps(_mm_add_ps(_mm_mul_ps(_mm_shuffle1_ps1(r, 0), l0),
_mm_mul_ps(_mm_shuffle1_ps1(r, 1), l1)),
_mm_mul_ps(_mm_shuffle1_ps1(r, 2), l2)));
r = _mm_load_ps(m2[1]);
r = glmm_load(m2[1]);
_mm_store_ps(dest[1],
_mm_add_ps(_mm_add_ps(_mm_mul_ps(_mm_shuffle1_ps1(r, 0), l0),
_mm_mul_ps(_mm_shuffle1_ps1(r, 1), l1)),
_mm_mul_ps(_mm_shuffle1_ps1(r, 2), l2)));
r = _mm_load_ps(m2[2]);
r = glmm_load(m2[2]);
_mm_store_ps(dest[2],
_mm_add_ps(_mm_add_ps(_mm_mul_ps(_mm_shuffle1_ps1(r, 0), l0),
_mm_mul_ps(_mm_shuffle1_ps1(r, 1), l1)),
_mm_mul_ps(_mm_shuffle1_ps1(r, 2), l2)));
r = _mm_load_ps(m2[3]);
r = glmm_load(m2[3]);
_mm_store_ps(dest[3],
_mm_add_ps(_mm_add_ps(_mm_mul_ps(_mm_shuffle1_ps1(r, 0), l0),
_mm_mul_ps(_mm_shuffle1_ps1(r, 1), l1)),
@@ -55,24 +55,24 @@ glm_mul_rot_sse2(mat4 m1, mat4 m2, mat4 dest) {
/* D = R * L (Column-Major) */
__m128 l0, l1, l2, l3, r;
l0 = _mm_load_ps(m1[0]);
l1 = _mm_load_ps(m1[1]);
l2 = _mm_load_ps(m1[2]);
l3 = _mm_load_ps(m1[3]);
l0 = glmm_load(m1[0]);
l1 = glmm_load(m1[1]);
l2 = glmm_load(m1[2]);
l3 = glmm_load(m1[3]);
r = _mm_load_ps(m2[0]);
r = glmm_load(m2[0]);
_mm_store_ps(dest[0],
_mm_add_ps(_mm_add_ps(_mm_mul_ps(_mm_shuffle1_ps1(r, 0), l0),
_mm_mul_ps(_mm_shuffle1_ps1(r, 1), l1)),
_mm_mul_ps(_mm_shuffle1_ps1(r, 2), l2)));
r = _mm_load_ps(m2[1]);
r = glmm_load(m2[1]);
_mm_store_ps(dest[1],
_mm_add_ps(_mm_add_ps(_mm_mul_ps(_mm_shuffle1_ps1(r, 0), l0),
_mm_mul_ps(_mm_shuffle1_ps1(r, 1), l1)),
_mm_mul_ps(_mm_shuffle1_ps1(r, 2), l2)));
r = _mm_load_ps(m2[2]);
r = glmm_load(m2[2]);
_mm_store_ps(dest[2],
_mm_add_ps(_mm_add_ps(_mm_mul_ps(_mm_shuffle1_ps1(r, 0), l0),
_mm_mul_ps(_mm_shuffle1_ps1(r, 1), l1)),
@@ -86,10 +86,10 @@ void
glm_inv_tr_sse2(mat4 mat) {
__m128 r0, r1, r2, r3, x0, x1;
r0 = _mm_load_ps(mat[0]);
r1 = _mm_load_ps(mat[1]);
r2 = _mm_load_ps(mat[2]);
r3 = _mm_load_ps(mat[3]);
r0 = glmm_load(mat[0]);
r1 = glmm_load(mat[1]);
r2 = glmm_load(mat[2]);
r3 = glmm_load(mat[3]);
x1 = _mm_set_ps(1.0f, 0.0f, 0.0f, 0.0f);
_MM_TRANSPOSE4_PS(r0, r1, r2, x1);

66
include/cglm/simd/sse2/mat4.h
View File

@@ -20,10 +20,10 @@ glm_mat4_scale_sse2(mat4 m, float s){
__m128 x0;
x0 = _mm_set1_ps(s);
_mm_store_ps(m[0], _mm_mul_ps(_mm_load_ps(m[0]), x0));
_mm_store_ps(m[1], _mm_mul_ps(_mm_load_ps(m[1]), x0));
_mm_store_ps(m[2], _mm_mul_ps(_mm_load_ps(m[2]), x0));
_mm_store_ps(m[3], _mm_mul_ps(_mm_load_ps(m[3]), x0));
_mm_store_ps(m[0], _mm_mul_ps(glmm_load(m[0]), x0));
_mm_store_ps(m[1], _mm_mul_ps(glmm_load(m[1]), x0));
_mm_store_ps(m[2], _mm_mul_ps(glmm_load(m[2]), x0));
_mm_store_ps(m[3], _mm_mul_ps(glmm_load(m[3]), x0));
}
CGLM_INLINE
@@ -31,10 +31,10 @@ void
glm_mat4_transp_sse2(mat4 m, mat4 dest){
__m128 r0, r1, r2, r3;
r0 = _mm_load_ps(m[0]);
r1 = _mm_load_ps(m[1]);
r2 = _mm_load_ps(m[2]);
r3 = _mm_load_ps(m[3]);
r0 = glmm_load(m[0]);
r1 = glmm_load(m[1]);
r2 = glmm_load(m[2]);
r3 = glmm_load(m[3]);
_MM_TRANSPOSE4_PS(r0, r1, r2, r3);
@@ -51,31 +51,31 @@ glm_mat4_mul_sse2(mat4 m1, mat4 m2, mat4 dest) {
__m128 l0, l1, l2, l3, r;
l0 = _mm_load_ps(m1[0]);
l1 = _mm_load_ps(m1[1]);
l2 = _mm_load_ps(m1[2]);
l3 = _mm_load_ps(m1[3]);
l0 = glmm_load(m1[0]);
l1 = glmm_load(m1[1]);
l2 = glmm_load(m1[2]);
l3 = glmm_load(m1[3]);
r = _mm_load_ps(m2[0]);
r = glmm_load(m2[0]);
_mm_store_ps(dest[0],
_mm_add_ps(_mm_add_ps(_mm_mul_ps(_mm_shuffle1_ps1(r, 0), l0),
_mm_mul_ps(_mm_shuffle1_ps1(r, 1), l1)),
_mm_add_ps(_mm_mul_ps(_mm_shuffle1_ps1(r, 2), l2),
_mm_mul_ps(_mm_shuffle1_ps1(r, 3), l3))));
r = _mm_load_ps(m2[1]);
r = glmm_load(m2[1]);
_mm_store_ps(dest[1],
_mm_add_ps(_mm_add_ps(_mm_mul_ps(_mm_shuffle1_ps1(r, 0), l0),
_mm_mul_ps(_mm_shuffle1_ps1(r, 1), l1)),
_mm_add_ps(_mm_mul_ps(_mm_shuffle1_ps1(r, 2), l2),
_mm_mul_ps(_mm_shuffle1_ps1(r, 3), l3))));
r = _mm_load_ps(m2[2]);
r = glmm_load(m2[2]);
_mm_store_ps(dest[2],
_mm_add_ps(_mm_add_ps(_mm_mul_ps(_mm_shuffle1_ps1(r, 0), l0),
_mm_mul_ps(_mm_shuffle1_ps1(r, 1), l1)),
_mm_add_ps(_mm_mul_ps(_mm_shuffle1_ps1(r, 2), l2),
_mm_mul_ps(_mm_shuffle1_ps1(r, 3), l3))));
r = _mm_load_ps(m2[3]);
r = glmm_load(m2[3]);
_mm_store_ps(dest[3],
_mm_add_ps(_mm_add_ps(_mm_mul_ps(_mm_shuffle1_ps1(r, 0), l0),
_mm_mul_ps(_mm_shuffle1_ps1(r, 1), l1)),
@@ -88,15 +88,15 @@ void
glm_mat4_mulv_sse2(mat4 m, vec4 v, vec4 dest) {
__m128 x0, x1, x2;
x0 = _mm_load_ps(v);
x1 = _mm_add_ps(_mm_mul_ps(_mm_load_ps(m[0]),
x0 = glmm_load(v);
x1 = _mm_add_ps(_mm_mul_ps(glmm_load(m[0]),
_mm_shuffle1_ps1(x0, 0)),
_mm_mul_ps(_mm_load_ps(m[1]),
_mm_mul_ps(glmm_load(m[1]),
_mm_shuffle1_ps1(x0, 1)));
x2 = _mm_add_ps(_mm_mul_ps(_mm_load_ps(m[2]),
x2 = _mm_add_ps(_mm_mul_ps(glmm_load(m[2]),
_mm_shuffle1_ps1(x0, 2)),
_mm_mul_ps(_mm_load_ps(m[3]),
_mm_mul_ps(glmm_load(m[3]),
_mm_shuffle1_ps1(x0, 3)));
_mm_store_ps(dest, _mm_add_ps(x1, x2));
@@ -108,10 +108,10 @@ glm_mat4_det_sse2(mat4 mat) {
__m128 r0, r1, r2, r3, x0, x1, x2;
/* 127 <- 0, [square] det(A) = det(At) */
r0 = _mm_load_ps(mat[0]); /* d c b a */
r1 = _mm_load_ps(mat[1]); /* h g f e */
r2 = _mm_load_ps(mat[2]); /* l k j i */
r3 = _mm_load_ps(mat[3]); /* p o n m */
r0 = glmm_load(mat[0]); /* d c b a */
r1 = glmm_load(mat[1]); /* h g f e */
r2 = glmm_load(mat[2]); /* l k j i */
r3 = glmm_load(mat[3]); /* p o n m */
/*
t[1] = j * p - n * l;
@@ -166,10 +166,10 @@ glm_mat4_inv_fast_sse2(mat4 mat, mat4 dest) {
x0, x1, x2, x3, x4, x5, x6, x7;
/* 127 <- 0 */
r0 = _mm_load_ps(mat[0]); /* d c b a */
r1 = _mm_load_ps(mat[1]); /* h g f e */
r2 = _mm_load_ps(mat[2]); /* l k j i */
r3 = _mm_load_ps(mat[3]); /* p o n m */
r0 = glmm_load(mat[0]); /* d c b a */
r1 = glmm_load(mat[1]); /* h g f e */
r2 = glmm_load(mat[2]); /* l k j i */
r3 = glmm_load(mat[3]); /* p o n m */
x0 = _mm_shuffle_ps(r2, r3, _MM_SHUFFLE(3, 2, 3, 2)); /* p o l k */
x1 = _mm_shuffle1_ps(x0, 1, 3, 3, 3); /* l p p p */
@@ -290,10 +290,10 @@ glm_mat4_inv_sse2(mat4 mat, mat4 dest) {
x0, x1, x2, x3, x4, x5, x6, x7;
/* 127 <- 0 */
r0 = _mm_load_ps(mat[0]); /* d c b a */
r1 = _mm_load_ps(mat[1]); /* h g f e */
r2 = _mm_load_ps(mat[2]); /* l k j i */
r3 = _mm_load_ps(mat[3]); /* p o n m */
r0 = glmm_load(mat[0]); /* d c b a */
r1 = glmm_load(mat[1]); /* h g f e */
r2 = glmm_load(mat[2]); /* l k j i */
r3 = glmm_load(mat[3]); /* p o n m */
x0 = _mm_shuffle_ps(r2, r3, _MM_SHUFFLE(3, 2, 3, 2)); /* p o l k */
x1 = _mm_shuffle1_ps(x0, 1, 3, 3, 3); /* l p p p */

4
include/cglm/simd/sse2/quat.h
View File

@@ -24,8 +24,8 @@ glm_quat_mul_sse2(versor p, versor q, versor dest) {
__m128 xp, xq, x0, r;
xp = _mm_load_ps(p); /* 3 2 1 0 */
xq = _mm_load_ps(q);
xp = glmm_load(p); /* 3 2 1 0 */
xq = glmm_load(q);
r = _mm_mul_ps(_mm_shuffle1_ps1(xp, 3), xq);