double bst_compute_121_m128_aligned4( void*_bst_obj, double* p, double* q, size_t nn ) {
segments_t* mem = (segments_t*) _bst_obj;
int n, i, r, l_end, l_end_pre, j;
double t, e_tmp;
double* e = mem->e, *w = mem->w;
int* root = mem->r;
__m128d v_tmp;
__m128d v00, v01, v02, v03;
__m128d v10, v11, v12, v13;
__m128i v_cur_roots, v_old_roots, v_new_roots;
__m128 v_rootmask;
// initialization
// mem->n = nn;
n = nn; // subtractions with n potentially negative. say hello to all the bugs
int idx1, idx2, idx3, pad, pad_r;
idx1 = (n+1)*(n+2)/2 + n/2;
e[idx1] = q[n];
idx1++;
pad = 1;
// pad contains the padding for row i+1
// for row n it's always 1
for (i = n-1; i >= 0; --i) {
idx1 -= 2*(n-i)+1 + pad;
idx2 = idx1 + 1;
e[idx1] = q[i];
w[idx1] = q[i];
for (j = i+1; j < n+1; ++j,++idx2) {
e[idx2] = INFINITY;
w[idx2] = w[idx2-1] + p[j-1] + q[j];
}
// idx2 now points to the beginning of the next line.
idx2 += pad; // padding of line i+1
idx3 = idx1;
pad_r = pad; // padding of line r
for (r = i; r < n; ++r) {
pad_r = !pad_r; // padding of line r+1
// idx2 = IDX(r+1, r+1);
idx1 = idx3;
l_end = idx2 + (n-r);
e_tmp = e[idx1++];
// calculate until a multiple of 8 doubles is left
// 8 = 4 * 2 128-bit vectors
l_end_pre = idx2 + ((n-r)&3);
for( ; (idx2 < l_end_pre) && (idx2 < l_end); ++idx2 ) {
t = e_tmp + e[idx2] + w[idx1];
if (t < e[idx1]) {
e[idx1] = t;
root[idx1] = r;
}
idx1++;
}
v_tmp = _mm_set_pd( e_tmp, e_tmp );
// execute the shit for 4 vectors of size 2
v_cur_roots = _mm_set_epi32(r, r, r, r);
for( ; idx2 < l_end; idx2 += 4 ) {
v01 = _mm_load_pd( &w[idx1 ] );
v11 = _mm_load_pd( &w[idx1+2] );
v00 = _mm_load_pd( &e[idx2 ] );
v01 = _mm_add_pd( v01, v_tmp ); // supoptimal for raw-dependency
v10 = _mm_load_pd( &e[idx2+2] );
v11 = _mm_add_pd( v11, v_tmp );
v01 = _mm_add_pd( v01, v00 );
v03 = _mm_load_pd( &e[idx1 ] );
v11 = _mm_add_pd( v11, v10 );
v13 = _mm_load_pd( &e[idx1+2] );
v02 = _mm_cmplt_pd( v01, v03 );
v12 = _mm_cmplt_pd( v11, v13 );
v00 = _mm_or_pd( _mm_and_pd( v02, v01 ), _mm_andnot_pd( v02, v03 ));
v10 = _mm_or_pd( _mm_and_pd( v12, v11 ), _mm_andnot_pd( v12, v13 ));
_mm_store_pd( &e[idx1 ], v00 );
_mm_store_pd( &e[idx1+2], v10 );
v_rootmask = _mm_shuffle_ps(
_mm_castpd_ps( v02 ),
_mm_castpd_ps( v12 ),
_MM_SHUFFLE(0,2,0,2) );
v_old_roots = _mm_lddqu_si128( &root[idx1] );
v_new_roots = _mm_or_si128(
_mm_and_si128( v_cur_roots,
_mm_castps_si128( v_rootmask ) ),
_mm_andnot_si128( v_old_roots,
_mm_castps_si128( v_rootmask ) )
);
_mm_storeu_si128( &root[idx1], v_new_roots );
idx1 += 4;
}
idx2 += pad_r;
idx3++;
}
pad = !pad;
// every other line as padding 0, or 1, respectively
}
// if n is even, the total number of entries in the first
// row of the table is odd, so we need padding
return e[n + !(n&1)];
}
Point::Point(double x, double y) #ifdef __SSE3__ : v(_mm_set_pd(y, x)) #else : x(x), y(y)
int sse3_ChirpData_ak8(
sah_complex * cx_DataArray,
sah_complex * cx_ChirpDataArray,
int chirp_rate_ind,
double chirp_rate,
int ul_NumDataPoints,
double sample_rate
) {
#ifdef USE_MANUAL_CALLSTACK
call_stack.enter("sse3_ChirpData_ak8()");
#endif
int i;
if (chirp_rate_ind == 0) {
memcpy(cx_ChirpDataArray, cx_DataArray, (int)ul_NumDataPoints * sizeof(sah_complex) );
#ifdef USE_MANUAL_CALLSTACK
call_stack.exit();
#endif
return 0;
}
int vEnd;
double srate = chirp_rate * 0.5 / (sample_rate * sample_rate);
__m128d rate = _mm_set1_pd(chirp_rate * 0.5 / (sample_rate * sample_rate));
__m128d roundVal = _mm_set1_pd(srate >= 0.0 ? TWO_TO_52 : -TWO_TO_52);
__m128d DFOUR = _mm_set_pd(4.0, 4.0);
// main vectorised loop
vEnd = ul_NumDataPoints - (ul_NumDataPoints & 3);
__m128d di1 = _mm_set_pd(2.0, 0.0); // set time patterns for eventual moveldup/movehdup
__m128d di2 = _mm_set_pd(3.0, 1.0);
for (i = 0; i < vEnd; i += 4) {
const float *d = (const float *) (cx_DataArray + i);
float *cd = (float *) (cx_ChirpDataArray + i);
__m128d a1, a2;
__m128 d1, d2;
__m128 cd1, cd2;
__m128 td1, td2;
__m128 x;
__m128 y;
__m128 z;
__m128 s;
__m128 c;
__m128 m;
// load the signal to be chirped
d1 = _mm_load_ps(d);
d2 = _mm_load_ps(d+4);
// calculate the input angle
a1 = _mm_mul_pd(_mm_mul_pd(di1, di1), rate);
a2 = _mm_mul_pd(_mm_mul_pd(di2, di2), rate);
// update times for next
di1 = _mm_add_pd(di1, DFOUR);
di2 = _mm_add_pd(di2, DFOUR);
// reduce the angle to the range (-0.5, 0.5)
a1 = _mm_sub_pd(a1, _mm_sub_pd(_mm_add_pd(a1, roundVal), roundVal));
a2 = _mm_sub_pd(a2, _mm_sub_pd(_mm_add_pd(a2, roundVal), roundVal));
// convert pair of packed double into packed single
x = _mm_movelh_ps(_mm_cvtpd_ps(a1), _mm_cvtpd_ps(a2)); // 3 1 2 0
// square to the range [0, 0.25)
y = _mm_mul_ps(x, x);
// perform the initial polynomial approximations, Estrin's method
z = _mm_mul_ps(y, y);
s = _mm_mul_ps(_mm_add_ps(_mm_mul_ps(_mm_add_ps(_mm_mul_ps(y, SS4F),
SS3F),
z),
_mm_add_ps(_mm_mul_ps(y, SS2F),
SS1F)),
x);
c = _mm_add_ps(_mm_mul_ps(_mm_add_ps(_mm_mul_ps(y, CC3F),
CC2F),
z),
_mm_add_ps(_mm_mul_ps(y, CC1F),
ONE));
// perform first angle doubling
x = _mm_sub_ps(_mm_mul_ps(c, c), _mm_mul_ps(s, s));
y = _mm_mul_ps(_mm_mul_ps(s, c), TWO);
// calculate scaling factor to correct the magnitude
m = _mm_sub_ps(_mm_sub_ps(TWO, _mm_mul_ps(x, x)), _mm_mul_ps(y, y));
// perform second angle doubling
c = _mm_sub_ps(_mm_mul_ps(x, x), _mm_mul_ps(y, y));
s = _mm_mul_ps(_mm_mul_ps(y, x), TWO);
// correct the magnitude (final sine / cosine approximations)
c = _mm_mul_ps(c, m); // c3 c1 c2 c0
s = _mm_mul_ps(s, m);
// chirp the data
cd1 = _mm_moveldup_ps(c); // c1 c1 c0 c0
cd2 = _mm_movehdup_ps(c); // c3 c3 c2 c2
cd1 = _mm_mul_ps(cd1, d1); // c1.i1 c1.r1 c0.i0 c0.r0
cd2 = _mm_mul_ps(cd2, d2); // c3.i3 c3.r3 c2.i2 c2.r2
d1 = _mm_shuffle_ps(d1, d1, 0xb1);
d2 = _mm_shuffle_ps(d2, d2, 0xb1);
td1 = _mm_moveldup_ps(s);
td2 = _mm_movehdup_ps(s);
td1 = _mm_mul_ps(td1, d1);
td2 = _mm_mul_ps(td2, d2);
cd1 = _mm_addsub_ps(cd1, td1);
cd2 = _mm_addsub_ps(cd2, td2);
// store chirped values
_mm_stream_ps(cd, cd1);
_mm_stream_ps(cd+4, cd2);
}
// handle tail elements with scalar code
for (; i < ul_NumDataPoints; ++i) {
double angle = srate * i * i * 0.5;
double s = sin(angle);
double c = cos(angle);
float re = cx_DataArray[i][0];
float im = cx_DataArray[i][1];
cx_ChirpDataArray[i][0] = re * c - im * s;
cx_ChirpDataArray[i][1] = re * s + im * c;
}
analysis_state.FLOP_counter+=12.0*ul_NumDataPoints;
#ifdef USE_MANUAL_CALLSTACK
call_stack.exit();
#endif
return 0;
}
static void
mlib_ImageCopyMask_Fp_d64(
const mlib_d64 *src,
mlib_s32 slb,
const mlib_d64 *mask,
mlib_s32 mlb,
mlib_d64 *dst,
mlib_s32 dlb,
mlib_s32 xsize,
mlib_s32 ysize,
mlib_s32 nchan,
const mlib_d64 *thresh)
{
mlib_s32 i, j, nsize;
mlib_d64 thresh0, thresh1, thresh2, thresh3;
__m128d threshs, threshu, threshv;
nsize = xsize * nchan;
switch (nchan) {
case 1:
thresh0 = thresh[0];
threshs = _mm_set1_pd(thresh0);
if ((((mlib_addr)dst | dlb |
(mlib_addr)src | slb |
(mlib_addr)mask | mlb) & 0xf) == 0) {
for (j = 0; j < ysize; j ++) {
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 2); i += 2) {
MLIB_S_COPYIMAGEMASK_D64(
_mm_store_pd,
_mm_load_pd,
_mm_load_pd,
_mm_load_pd);
}
for (; i < nsize; i ++) {
if (mask[i] <= thresh0) {
dst[i] = src[i];
}
}
src = (mlib_d64 *)((mlib_u8 *)src + slb);
mask = (mlib_d64 *)((mlib_u8 *)mask + mlb);
dst = (mlib_d64 *)((mlib_u8 *)dst + dlb);
}
} else {
for (j = 0; j < ysize; j ++) {
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 2); i += 2) {
MLIB_S_COPYIMAGEMASK_D64(
_mm_storeu_pd,
_mm_loadu_pd,
_mm_loadu_pd,
_mm_loadu_pd);
}
for (; i < nsize; i ++) {
if (mask[i] <= thresh0) {
dst[i] = src[i];
}
}
src = (mlib_d64 *)((mlib_u8 *)src + slb);
mask = (mlib_d64 *)((mlib_u8 *)mask + mlb);
dst = (mlib_d64 *)((mlib_u8 *)dst + dlb);
}
}
break;
case 2:
thresh0 = thresh[0];
thresh1 = thresh[1];
threshs = _mm_set_pd(thresh1, thresh0);
if ((((mlib_addr)dst | dlb |
(mlib_addr)src | slb |
(mlib_addr)mask | mlb) & 0xf) == 0) {
for (j = 0; j < ysize; j ++) {
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 2); i += 2) {
MLIB_S_COPYIMAGEMASK_D64(
_mm_store_pd,
_mm_load_pd,
_mm_load_pd,
_mm_load_pd);
}
for (; i < nsize; i += 2) {
if (mask[i] <= thresh0) {
dst[i] = src[i];
}
if (mask[i + 1] <= thresh1) {
dst[i + 1] = src[i + 1];
}
}
src = (mlib_d64 *)((mlib_u8 *)src + slb);
mask = (mlib_d64 *)((mlib_u8 *)mask + mlb);
dst = (mlib_d64 *)((mlib_u8 *)dst + dlb);
}
} else {
for (j = 0; j < ysize; j ++) {
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 2); i += 2) {
MLIB_S_COPYIMAGEMASK_D64(
_mm_storeu_pd,
_mm_loadu_pd,
_mm_loadu_pd,
_mm_loadu_pd);
}
for (; i < nsize; i += 2) {
if (mask[i] <= thresh0) {
dst[i] = src[i];
}
if (mask[i + 1] <= thresh1) {
dst[i + 1] = src[i + 1];
}
}
src = (mlib_d64 *)((mlib_u8 *)src + slb);
mask = (mlib_d64 *)((mlib_u8 *)mask + mlb);
dst = (mlib_d64 *)((mlib_u8 *)dst + dlb);
}
}
break;
case 3:
thresh0 = thresh[0];
thresh1 = thresh[1];
thresh2 = thresh[2];
threshs = _mm_set_pd(thresh1, thresh0);
threshu = _mm_set_pd(thresh0, thresh2);
threshv = _mm_set_pd(thresh2, thresh1);
if ((((mlib_addr)dst | dlb |
(mlib_addr)src | slb |
(mlib_addr)mask | mlb) & 0xf) == 0) {
for (j = 0; j < ysize; j ++) {
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 6); i += 6) {
MLIB_S_COPYIMAGEMASK3_D64(
_mm_store_pd,
_mm_load_pd,
_mm_load_pd,
_mm_load_pd);
}
for (; i < nsize; i += 3) {
if (mask[i] <= thresh0) {
dst[i] = src[i];
}
if (mask[i + 1] <= thresh1) {
dst[i + 1] = src[i + 1];
}
if (mask[i + 2] <= thresh2) {
dst[i + 2] = src[i + 2];
}
}
src = (mlib_d64 *)((mlib_u8 *)src + slb);
mask = (mlib_d64 *)((mlib_u8 *)mask + mlb);
dst = (mlib_d64 *)((mlib_u8 *)dst + dlb);
}
} else {
for (j = 0; j < ysize; j ++) {
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 6); i += 6) {
MLIB_S_COPYIMAGEMASK3_D64(
_mm_storeu_pd,
_mm_loadu_pd,
_mm_loadu_pd,
_mm_loadu_pd);
}
for (; i < nsize; i += 3) {
if (mask[i] <= thresh0) {
dst[i] = src[i];
}
if (mask[i + 1] <= thresh1) {
dst[i + 1] = src[i + 1];
}
if (mask[i + 2] <= thresh2) {
dst[i + 2] = src[i + 2];
}
}
src = (mlib_d64 *)((mlib_u8 *)src + slb);
mask = (mlib_d64 *)((mlib_u8 *)mask + mlb);
dst = (mlib_d64 *)((mlib_u8 *)dst + dlb);
}
}
break;
case 4:
thresh0 = thresh[0];
thresh1 = thresh[1];
thresh2 = thresh[2];
thresh3 = thresh[3];
threshs = _mm_set_pd(thresh1, thresh0);
threshu = _mm_set_pd(thresh3, thresh2);
if ((((mlib_addr)dst | dlb |
(mlib_addr)src | slb |
(mlib_addr)mask | mlb) & 0xf) == 0) {
for (j = 0; j < ysize; j ++) {
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 4); i += 4) {
MLIB_S_COPYIMAGEMASK4_D64(
_mm_store_pd,
_mm_load_pd,
_mm_load_pd,
_mm_load_pd);
}
for (; i < nsize; i += 4) {
if (mask[i] <= thresh0) {
dst[i] = src[i];
}
if (mask[i + 1] <= thresh1) {
dst[i + 1] = src[i + 1];
}
if (mask[i + 2] <= thresh2) {
dst[i + 1] = src[i + 2];
}
if (mask[i + 3] <= thresh3) {
dst[i + 1] = src[i + 3];
}
}
src = (mlib_d64 *)((mlib_u8 *)src + slb);
mask = (mlib_d64 *)((mlib_u8 *)mask + mlb);
dst = (mlib_d64 *)((mlib_u8 *)dst + dlb);
}
} else {
for (j = 0; j < ysize; j ++) {
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 4); i += 4) {
MLIB_S_COPYIMAGEMASK4_D64(
_mm_storeu_pd,
_mm_loadu_pd,
_mm_loadu_pd,
_mm_loadu_pd);
}
for (; i < nsize; i += 4) {
if (mask[i] <= thresh0) {
dst[i] = src[i];
}
if (mask[i + 1] <= thresh1) {
dst[i + 1] = src[i + 1];
}
if (mask[i + 2] <= thresh2) {
dst[i + 1] = src[i + 2];
}
if (mask[i + 3] <= thresh3) {
dst[i + 1] = src[i + 3];
}
}
src = (mlib_d64 *)((mlib_u8 *)src + slb);
mask = (mlib_d64 *)((mlib_u8 *)mask + mlb);
dst = (mlib_d64 *)((mlib_u8 *)dst + dlb);
}
}
break;
}
}
void exchlaplacecoeffData_2(unsigned int slot) {
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((!neighbor_isValid[1][0])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S710, S704, S707, S701, S709, S700, S703, S706, S708, S702, S705 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+394)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+402)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+394)] = 0.000000e+00;
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+226)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+234)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+226)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+170)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+178)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+170)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+58)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+66)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+58)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+450)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+458)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+450)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+114)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+122)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+114)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
yPos = ((((i1-1)/4.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(1.000000e+00);
__m128d vec2 = _mm_set1_pd(4.000000e+00);
__m128d vec5 = _mm_set1_pd(yPos);
for (; (i1<3); i1 += 4) {
/* yPos = ((((i1-1)/4.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec3 = _mm_load1_pd((&posEnd[1]));
__m128d vec3_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec4 = _mm_load1_pd((&posBegin[1]));
__m128d vec4_2 = _mm_load1_pd((&posBegin[1]));
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0, vec1), vec2), _mm_sub_pd(vec3, vec4)), vec4);
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0_2, vec1), vec2), _mm_sub_pd(vec3_2, vec4_2)), vec4_2);
}
for (; (i1<6); i1 += 1) {
yPos = ((((i1-1)/4.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+2)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+10)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+2)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+338)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+346)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+338)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+282)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+290)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+282)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
xPos = posBegin[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<3); i1 += 4) {
/* xPos = posBegin[0]; */
__m128d vec0 = _mm_load1_pd((&posBegin[0]));
__m128d vec0_2 = _mm_load1_pd((&posBegin[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<6); i1 += 1) {
xPos = posBegin[0];
}
}
}
}
}
if ((!neighbor_isValid[1][1])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S716, S719, S713, S721, S715, S718, S712, S711, S720, S714, S717 */
{
{
{
{
{
{
{
{
{
{
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
yPos = ((((i1-1)/4.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(1.000000e+00);
__m128d vec2 = _mm_set1_pd(4.000000e+00);
__m128d vec5 = _mm_set1_pd(yPos);
for (; (i1<3); i1 += 4) {
/* yPos = ((((i1-1)/4.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec3 = _mm_load1_pd((&posEnd[1]));
__m128d vec3_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec4 = _mm_load1_pd((&posBegin[1]));
__m128d vec4_2 = _mm_load1_pd((&posBegin[1]));
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0, vec1), vec2), _mm_sub_pd(vec3, vec4)), vec4);
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0_2, vec1), vec2), _mm_sub_pd(vec3_2, vec4_2)), vec4_2);
}
for (; (i1<6); i1 += 1) {
yPos = ((((i1-1)/4.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+454)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+462)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+454)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+230)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+238)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+230)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+118)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+126)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+118)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
xPos = posEnd[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<3); i1 += 4) {
/* xPos = posEnd[0]; */
__m128d vec0 = _mm_load1_pd((&posEnd[0]));
__m128d vec0_2 = _mm_load1_pd((&posEnd[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<6); i1 += 1) {
xPos = posEnd[0];
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+286)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+294)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+286)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+342)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+350)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+342)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+398)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+406)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+398)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+174)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+182)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+174)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+62)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+70)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+62)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+6)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[((i1*8)+14)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)+6)] = 0.000000e+00;
}
}
}
}
}
if ((!neighbor_isValid[1][2])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S722, S731, S725, S728, S727, S730, S724, S732, S726, S729, S723 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+344)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+345)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+344)] = 0.000000e+00;
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+400)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+401)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+400)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+120)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+121)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+120)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+8)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+9)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+8)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+64)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+65)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+64)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+456)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+457)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+456)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+232)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+233)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+232)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+288)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+289)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+288)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=5); i2 += 2) {
xPos = ((((i2-2)/4.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2-1)/4.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=6); i2 += 1) {
xPos = ((((i2-2)/4.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+176)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+177)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+176)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=5); i2 += 2) {
yPos = posBegin[1];
yPos = posBegin[1];
}
for (; (i2<=6); i2 += 1) {
yPos = posBegin[1];
}
}
}
}
}
if ((!neighbor_isValid[1][3])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S743, S737, S733, S742, S736, S739, S738, S741, S735, S740, S734 */
{
{
{
{
{
{
{
{
{
{
{
int i2 = 2;
for (; (i2<=5); i2 += 2) {
yPos = posEnd[1];
yPos = posEnd[1];
}
for (; (i2<=6); i2 += 1) {
yPos = posEnd[1];
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+376)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+377)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+376)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+488)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+489)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+488)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+40)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+41)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+40)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+208)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+209)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+208)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+152)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+153)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+152)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+320)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+321)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+320)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+432)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+433)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+432)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+96)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+97)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+96)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=5); i2 += 2) {
xPos = ((((i2-2)/4.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2-1)/4.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=6); i2 += 1) {
xPos = ((((i2-2)/4.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
}
{
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_LaplaceCoeff_2_p1[(i2+264)] = 0.000000e+00;
fieldData_LaplaceCoeff_2_p1[(i2+265)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_LaplaceCoeff_2_p1[(i2+264)] = 0.000000e+00;
}
}
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
/* Statements in this Scop: S744 */
for (int i0 = 0; (i0<=8); i0 += 1) {
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][(i0*56)]);
double* buffer_Send_1_p1 = (&buffer_Send[1][(i0*5)]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
buffer_Send_1_p1[(i1-1)] = fieldData_LaplaceCoeff_2_p1[((i1*8)+6)];
buffer_Send_1_p1[i1] = fieldData_LaplaceCoeff_2_p1[((i1*8)+14)];
}
for (; (i1<=5); i1 += 1) {
buffer_Send_1_p1[(i1-1)] = fieldData_LaplaceCoeff_2_p1[((i1*8)+6)];
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
MPI_Isend(buffer_Send[1], 45, MPI_DOUBLE, neighbor_remoteRank[1][1], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][1]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[1]);
reqOutstanding_Send[1] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
MPI_Irecv(buffer_Recv[0], 45, MPI_DOUBLE, neighbor_remoteRank[1][0], ((unsigned int)(neighbor_fragCommId[1][0]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[0]) {
waitForMPIReq(&mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
/* Statements in this Scop: S745 */
for (int i0 = 0; (i0<=8); i0 += 1) {
double* buffer_Recv_0_p1 = (&buffer_Recv[0][(i0*5)]);
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][(i0*56)]);
int i1 = 3;
for (; (i1<=6); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)-14)] = buffer_Recv_0_p1[(i1-3)];
fieldData_LaplaceCoeff_2_p1[((i1*8)-6)] = buffer_Recv_0_p1[(i1-2)];
}
for (; (i1<=7); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)-14)] = buffer_Recv_0_p1[(i1-3)];
}
}
}
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[1]) {
waitForMPIReq(&mpiRequest_Send[1]);
reqOutstanding_Send[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][3]&&neighbor_isRemote[1][3])) {
MPI_Isend(&fieldData_LaplaceCoeff[2][42], 1, mpiDatatype_9_5_56, neighbor_remoteRank[1][3], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][3]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[3]);
reqOutstanding_Send[3] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][2]&&neighbor_isRemote[1][2])) {
MPI_Irecv(&fieldData_LaplaceCoeff[2][10], 1, mpiDatatype_9_5_56, neighbor_remoteRank[1][2], ((unsigned int)(neighbor_fragCommId[1][2]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[2]) {
waitForMPIReq(&mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[3]) {
waitForMPIReq(&mpiRequest_Send[3]);
reqOutstanding_Send[3] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
/* Statements in this Scop: S746 */
for (int i0 = 0; (i0<=8); i0 += 1) {
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][(i0*56)]);
double* buffer_Send_0_p1 = (&buffer_Send[0][(i0*7)]);
int i1 = 0;
for (; (i1<=5); i1 += 2) {
buffer_Send_0_p1[i1] = fieldData_LaplaceCoeff_2_p1[((i1*8)+3)];
buffer_Send_0_p1[(i1+1)] = fieldData_LaplaceCoeff_2_p1[((i1*8)+11)];
}
for (; (i1<=6); i1 += 1) {
buffer_Send_0_p1[i1] = fieldData_LaplaceCoeff_2_p1[((i1*8)+3)];
}
}
}
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
/* Statements in this Scop: S747 */
for (int i0 = 0; (i0<=8); i0 += 1) {
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][(i0*56)]);
double* buffer_Send_1_p1 = (&buffer_Send[1][(i0*7)]);
int i1 = 0;
for (; (i1<=5); i1 += 2) {
buffer_Send_1_p1[i1] = fieldData_LaplaceCoeff_2_p1[((i1*8)+5)];
buffer_Send_1_p1[(i1+1)] = fieldData_LaplaceCoeff_2_p1[((i1*8)+13)];
}
for (; (i1<=6); i1 += 1) {
buffer_Send_1_p1[i1] = fieldData_LaplaceCoeff_2_p1[((i1*8)+5)];
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
MPI_Isend(buffer_Send[0], 63, MPI_DOUBLE, neighbor_remoteRank[1][0], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][0]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[0]);
reqOutstanding_Send[0] = true;
}
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
MPI_Isend(buffer_Send[1], 63, MPI_DOUBLE, neighbor_remoteRank[1][1], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][1]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[1]);
reqOutstanding_Send[1] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
MPI_Irecv(buffer_Recv[0], 63, MPI_DOUBLE, neighbor_remoteRank[1][0], ((unsigned int)(neighbor_fragCommId[1][0]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = true;
}
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
MPI_Irecv(buffer_Recv[1], 63, MPI_DOUBLE, neighbor_remoteRank[1][1], ((unsigned int)(neighbor_fragCommId[1][1]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[1]);
reqOutstanding_Recv[1] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[0]) {
waitForMPIReq(&mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = false;
}
if (reqOutstanding_Recv[1]) {
waitForMPIReq(&mpiRequest_Recv[1]);
reqOutstanding_Recv[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
/* Statements in this Scop: S748 */
for (int i0 = 0; (i0<=8); i0 += 1) {
double* buffer_Recv_0_p1 = (&buffer_Recv[0][(i0*7)]);
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][(i0*56)]);
int i1 = 1;
for (; (i1<=6); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)-7)] = buffer_Recv_0_p1[(i1-1)];
fieldData_LaplaceCoeff_2_p1[((i1*8)+1)] = buffer_Recv_0_p1[i1];
}
for (; (i1<=7); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)-7)] = buffer_Recv_0_p1[(i1-1)];
}
}
}
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
/* Statements in this Scop: S749 */
for (int i0 = 0; (i0<=8); i0 += 1) {
double* buffer_Recv_1_p1 = (&buffer_Recv[1][(i0*7)]);
double* fieldData_LaplaceCoeff_2_p1 = (&fieldData_LaplaceCoeff[2][(i0*56)]);
int i1 = 7;
for (; (i1<=12); i1 += 2) {
fieldData_LaplaceCoeff_2_p1[((i1*8)-49)] = buffer_Recv_1_p1[(i1-7)];
fieldData_LaplaceCoeff_2_p1[((i1*8)-41)] = buffer_Recv_1_p1[(i1-6)];
}
for (; (i1<=13); i1 += 1) {
fieldData_LaplaceCoeff_2_p1[((i1*8)-49)] = buffer_Recv_1_p1[(i1-7)];
}
}
}
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[0]) {
waitForMPIReq(&mpiRequest_Send[0]);
reqOutstanding_Send[0] = false;
}
if (reqOutstanding_Send[1]) {
waitForMPIReq(&mpiRequest_Send[1]);
reqOutstanding_Send[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][2]&&neighbor_isRemote[1][2])) {
MPI_Isend(&fieldData_LaplaceCoeff[2][17], 1, mpiDatatype_9_7_56, neighbor_remoteRank[1][2], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][2]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[2]);
reqOutstanding_Send[2] = true;
}
if ((neighbor_isValid[1][3]&&neighbor_isRemote[1][3])) {
MPI_Isend(&fieldData_LaplaceCoeff[2][33], 1, mpiDatatype_9_7_56, neighbor_remoteRank[1][3], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][3]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[3]);
reqOutstanding_Send[3] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][2]&&neighbor_isRemote[1][2])) {
MPI_Irecv(&fieldData_LaplaceCoeff[2][1], 1, mpiDatatype_9_7_56, neighbor_remoteRank[1][2], ((unsigned int)(neighbor_fragCommId[1][2]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = true;
}
if ((neighbor_isValid[1][3]&&neighbor_isRemote[1][3])) {
MPI_Irecv(&fieldData_LaplaceCoeff[2][49], 1, mpiDatatype_9_7_56, neighbor_remoteRank[1][3], ((unsigned int)(neighbor_fragCommId[1][3]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[3]);
reqOutstanding_Recv[3] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[2]) {
waitForMPIReq(&mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = false;
}
if (reqOutstanding_Recv[3]) {
waitForMPIReq(&mpiRequest_Recv[3]);
reqOutstanding_Recv[3] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[2]) {
waitForMPIReq(&mpiRequest_Send[2]);
reqOutstanding_Send[2] = false;
}
if (reqOutstanding_Send[3]) {
waitForMPIReq(&mpiRequest_Send[3]);
reqOutstanding_Send[3] = false;
}
}
}
}
void
mlib_s_ImageScalarBlend_s32(
mlib_s32 *dst,
mlib_s32 dlb,
const mlib_s32 *src1,
mlib_s32 slb1,
const mlib_s32 *src2,
mlib_s32 slb2,
const mlib_s32 *alpha,
mlib_s32 xsize,
mlib_s32 ysize,
mlib_s32 nchan)
{
mlib_s32 i, j, nsize;
__m128i *srcPtr1, *srcPtr2, *dstPtr;
mlib_s32 *dl = dst;
mlib_s32 *sl1 = (mlib_s32 *)src1, *sl2 = (mlib_s32 *)src2;
__m128d alphas0, alphas1, alphau0, alphau1, alphav0, alphav1;
__m128d betas0, betas1, betau0, betau1, betav0, betav1;
__m128d ones = _mm_set1_pd(1.0f);
mlib_s32 res, sdata1, sdata2;
mlib_d64 a0, a1, a2, a3;
nsize = xsize * nchan;
switch (nchan) {
case 1:
a0 = -(alpha[0] & MASK) / (mlib_d64)MLIB_S32_MIN;
alphas0 = _mm_set1_pd(a0);
alphas1 = _mm_set1_pd(a0);
betas0 = _mm_sub_pd(ones, alphas0);
betas1 = _mm_sub_pd(ones, alphas1);
if ((((mlib_addr)dst | dlb |
(mlib_addr)src1 | slb1 |
(mlib_addr)src2 | slb2) & 0xf) == 0) {
for (j = 0; j < ysize; j ++) {
srcPtr1 = (__m128i *)sl1;
srcPtr2 = (__m128i *)sl2;
dstPtr = (__m128i *)dl;
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 4); i += 4) {
MLIB_S_IMAGESCALARBLEND_S32(
_mm_store_si128,
_mm_load_si128,
_mm_load_si128);
}
for (; i < nsize; i++) {
MLIB_C_IMAGESCALARBLEND_S32_1(
sl1, sl2, dl);
}
sl1 = (mlib_s32 *)((mlib_u8 *)sl1 + slb1);
sl2 = (mlib_s32 *)((mlib_u8 *)sl2 + slb2);
dl = (mlib_s32 *)((mlib_u8 *)dl + dlb);
}
} else
if ((((mlib_addr)src1 | slb1 |
(mlib_addr)src2 | slb2) & 0xf) == 0) {
for (j = 0; j < ysize; j ++) {
srcPtr1 = (__m128i *)sl1;
srcPtr2 = (__m128i *)sl2;
dstPtr = (__m128i *)dl;
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 4); i += 4) {
MLIB_S_IMAGESCALARBLEND_S32(
_mm_storeu_si128,
_mm_load_si128,
_mm_load_si128);
}
for (; i < nsize; i++) {
MLIB_C_IMAGESCALARBLEND_S32_1(
sl1, sl2, dl);
}
sl1 = (mlib_s32 *)((mlib_u8 *)sl1 + slb1);
sl2 = (mlib_s32 *)((mlib_u8 *)sl2 + slb2);
dl = (mlib_s32 *)((mlib_u8 *)dl + dlb);
}
} else {
for (j = 0; j < ysize; j ++) {
srcPtr1 = (__m128i *)sl1;
srcPtr2 = (__m128i *)sl2;
dstPtr = (__m128i *)dl;
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 4); i += 4) {
MLIB_S_IMAGESCALARBLEND_S32(
_mm_storeu_si128,
_mm_loadu_si128,
_mm_loadu_si128);
}
for (; i < nsize; i++) {
MLIB_C_IMAGESCALARBLEND_S32_1(
sl1, sl2, dl);
}
sl1 = (mlib_s32 *)((mlib_u8 *)sl1 + slb1);
sl2 = (mlib_s32 *)((mlib_u8 *)sl2 + slb2);
dl = (mlib_s32 *)((mlib_u8 *)dl + dlb);
}
}
break;
case 2:
a0 = -(alpha[0] & MASK) / (mlib_d64)MLIB_S32_MIN;
a1 = -(alpha[1] & MASK) / (mlib_d64)MLIB_S32_MIN;
alphas0 = _mm_set_pd(a1, a0);
alphas1 = _mm_set_pd(a1, a0);
betas0 = _mm_sub_pd(ones, alphas0);
betas1 = _mm_sub_pd(ones, alphas1);
if ((((mlib_addr)dst | dlb |
(mlib_addr)src1 | slb1 |
(mlib_addr)src2 | slb2) & 0xf) == 0) {
for (j = 0; j < ysize; j ++) {
srcPtr1 = (__m128i *)sl1;
srcPtr2 = (__m128i *)sl2;
dstPtr = (__m128i *)dl;
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 4); i += 4) {
MLIB_S_IMAGESCALARBLEND_S32(
_mm_store_si128,
_mm_load_si128,
_mm_load_si128);
}
for (; i < nsize; i += 2) {
MLIB_C_IMAGESCALARBLEND_S32_2(
sl1, sl2, dl);
}
sl1 = (mlib_s32 *)((mlib_u8 *)sl1 + slb1);
sl2 = (mlib_s32 *)((mlib_u8 *)sl2 + slb2);
dl = (mlib_s32 *)((mlib_u8 *)dl + dlb);
}
} else
if ((((mlib_addr)src1 | slb1 |
(mlib_addr)src2 | slb2) & 0xf) == 0) {
for (j = 0; j < ysize; j ++) {
srcPtr1 = (__m128i *)sl1;
srcPtr2 = (__m128i *)sl2;
dstPtr = (__m128i *)dl;
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 4); i += 4) {
MLIB_S_IMAGESCALARBLEND_S32(
_mm_storeu_si128,
_mm_load_si128,
_mm_load_si128);
}
for (; i < nsize; i += 2) {
MLIB_C_IMAGESCALARBLEND_S32_2(
sl1, sl2, dl);
}
sl1 = (mlib_s32 *)((mlib_u8 *)sl1 + slb1);
sl2 = (mlib_s32 *)((mlib_u8 *)sl2 + slb2);
dl = (mlib_s32 *)((mlib_u8 *)dl + dlb);
}
} else {
for (j = 0; j < ysize; j ++) {
srcPtr1 = (__m128i *)sl1;
srcPtr2 = (__m128i *)sl2;
dstPtr = (__m128i *)dl;
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 4); i += 4) {
MLIB_S_IMAGESCALARBLEND_S32(
_mm_storeu_si128,
_mm_loadu_si128,
_mm_loadu_si128);
}
for (; i < nsize; i += 2) {
MLIB_C_IMAGESCALARBLEND_S32_2(
sl1, sl2, dl);
}
sl1 = (mlib_s32 *)((mlib_u8 *)sl1 + slb1);
sl2 = (mlib_s32 *)((mlib_u8 *)sl2 + slb2);
dl = (mlib_s32 *)((mlib_u8 *)dl + dlb);
}
}
break;
case 3:
a0 = -(alpha[0] & MASK) / (mlib_d64)MLIB_S32_MIN;
a1 = -(alpha[1] & MASK) / (mlib_d64)MLIB_S32_MIN;
a2 = -(alpha[2] & MASK) / (mlib_d64)MLIB_S32_MIN;
alphas0 = _mm_set_pd(a1, a0);
alphas1 = _mm_set_pd(a0, a2);
alphau0 = _mm_set_pd(a2, a1);
alphau1 = _mm_set_pd(a1, a0);
alphav0 = _mm_set_pd(a0, a2);
alphav1 = _mm_set_pd(a2, a1);
betas0 = _mm_sub_pd(ones, alphas0);
betas1 = _mm_sub_pd(ones, alphas1);
betau0 = _mm_sub_pd(ones, alphau0);
betau1 = _mm_sub_pd(ones, alphau1);
betav0 = _mm_sub_pd(ones, alphav0);
betav1 = _mm_sub_pd(ones, alphav1);
if ((((mlib_addr)dst | dlb |
(mlib_addr)src1 | slb1 |
(mlib_addr)src2 | slb2) & 0xf) == 0) {
for (j = 0; j < ysize; j ++) {
srcPtr1 = (__m128i *)sl1;
srcPtr2 = (__m128i *)sl2;
dstPtr = (__m128i *)dl;
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 12); i += 12) {
MLIB_S_IMAGESCALARBLEND3_S32(
_mm_store_si128,
_mm_load_si128,
_mm_load_si128);
}
for (; i < nsize; i += 3) {
MLIB_C_IMAGESCALARBLEND_S32_3(
sl1, sl2, dl);
}
sl1 = (mlib_s32 *)((mlib_u8 *)sl1 + slb1);
sl2 = (mlib_s32 *)((mlib_u8 *)sl2 + slb2);
dl = (mlib_s32 *)((mlib_u8 *)dl + dlb);
}
} else
if ((((mlib_addr)src1 | slb1 |
(mlib_addr)src2 | slb2) & 0xf) == 0) {
for (j = 0; j < ysize; j ++) {
srcPtr1 = (__m128i *)sl1;
srcPtr2 = (__m128i *)sl2;
dstPtr = (__m128i *)dl;
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 12); i += 12) {
MLIB_S_IMAGESCALARBLEND3_S32(
_mm_storeu_si128,
_mm_load_si128,
_mm_load_si128);
}
for (; i < nsize; i += 3) {
MLIB_C_IMAGESCALARBLEND_S32_3(
sl1, sl2, dl);
}
sl1 = (mlib_s32 *)((mlib_u8 *)sl1 + slb1);
sl2 = (mlib_s32 *)((mlib_u8 *)sl2 + slb2);
dl = (mlib_s32 *)((mlib_u8 *)dl + dlb);
}
} else {
for (j = 0; j < ysize; j ++) {
srcPtr1 = (__m128i *)sl1;
srcPtr2 = (__m128i *)sl2;
dstPtr = (__m128i *)dl;
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 12); i += 12) {
MLIB_S_IMAGESCALARBLEND3_S32(
_mm_storeu_si128,
_mm_loadu_si128,
_mm_loadu_si128);
}
for (; i < nsize; i += 3) {
MLIB_C_IMAGESCALARBLEND_S32_3(
sl1, sl2, dl);
}
sl1 = (mlib_s32 *)((mlib_u8 *)sl1 + slb1);
sl2 = (mlib_s32 *)((mlib_u8 *)sl2 + slb2);
dl = (mlib_s32 *)((mlib_u8 *)dl + dlb);
}
}
break;
case 4:
a0 = -(alpha[0] & MASK) / (mlib_d64)MLIB_S32_MIN;
a1 = -(alpha[1] & MASK) / (mlib_d64)MLIB_S32_MIN;
a2 = -(alpha[2] & MASK) / (mlib_d64)MLIB_S32_MIN;
a3 = -(alpha[3] & MASK) / (mlib_d64)MLIB_S32_MIN;
alphas0 = _mm_set_pd(a1, a0);
alphas1 = _mm_set_pd(a3, a2);
betas0 = _mm_sub_pd(ones, alphas0);
betas1 = _mm_sub_pd(ones, alphas1);
if ((((mlib_addr)dst | dlb |
(mlib_addr)src1 | slb1 |
(mlib_addr)src2 | slb2) & 0xf) == 0) {
for (j = 0; j < ysize; j ++) {
srcPtr1 = (__m128i *)sl1;
srcPtr2 = (__m128i *)sl2;
dstPtr = (__m128i *)dl;
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 4); i += 4) {
MLIB_S_IMAGESCALARBLEND_S32(
_mm_store_si128,
_mm_load_si128,
_mm_load_si128);
}
sl1 = (mlib_s32 *)((mlib_u8 *)sl1 + slb1);
sl2 = (mlib_s32 *)((mlib_u8 *)sl2 + slb2);
dl = (mlib_s32 *)((mlib_u8 *)dl + dlb);
}
} else
if ((((mlib_addr)src1 | slb1 |
(mlib_addr)src2 | slb2) & 0xf) == 0) {
for (j = 0; j < ysize; j ++) {
srcPtr1 = (__m128i *)sl1;
srcPtr2 = (__m128i *)sl2;
dstPtr = (__m128i *)dl;
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 4); i += 4) {
MLIB_S_IMAGESCALARBLEND_S32(
_mm_storeu_si128,
_mm_load_si128,
_mm_load_si128);
}
sl1 = (mlib_s32 *)((mlib_u8 *)sl1 + slb1);
sl2 = (mlib_s32 *)((mlib_u8 *)sl2 + slb2);
dl = (mlib_s32 *)((mlib_u8 *)dl + dlb);
}
} else {
for (j = 0; j < ysize; j ++) {
srcPtr1 = (__m128i *)sl1;
srcPtr2 = (__m128i *)sl2;
dstPtr = (__m128i *)dl;
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i <= (nsize - 4); i += 4) {
MLIB_S_IMAGESCALARBLEND_S32(
_mm_storeu_si128,
_mm_loadu_si128,
_mm_loadu_si128);
}
sl1 = (mlib_s32 *)((mlib_u8 *)sl1 + slb1);
sl2 = (mlib_s32 *)((mlib_u8 *)sl2 + slb2);
dl = (mlib_s32 *)((mlib_u8 *)dl + dlb);
}
}
break;
}
}
void
mlib_ImageMinimum_D64_124(
const mlib_image *img,
mlib_d64 *min)
{
/* src address */
mlib_d64 *sp, *sl;
/* src data */
__m128d sd0, sd1;
/* min values */
__m128d min0, min1;
/* loop variables */
mlib_s32 n1, i;
/* height of image */
mlib_s32 height = mlib_ImageGetHeight(img);
/* elements to next row */
mlib_s32 src_stride = mlib_ImageGetStride(img);
/* number of image channels */
mlib_s32 channels = mlib_ImageGetChannels(img);
mlib_s32 size_row = mlib_ImageGetWidth(img) * channels;
sp = sl = mlib_ImageGetData(img);
if (src_stride == size_row) {
size_row *= height;
height = 1;
}
/* min values */
min0 = _mm_set1_pd(MLIB_D64_MAX);
min1 = _mm_set1_pd(MLIB_D64_MAX);
mlib_d64 rez[4] = {
MLIB_D64_MAX, MLIB_D64_MAX,
MLIB_D64_MAX, MLIB_D64_MAX};
for (; height > 0; height--) {
n1 = size_row;
for (; n1 > 3; n1 -= 4) {
sd0 = _mm_loadu_pd(sp);
sp += 2;
sd1 = _mm_loadu_pd(sp);
sp += 2;
MLIB_S_IMAGE_MINIMUM_D64(min0, min0, sd0);
MLIB_S_IMAGE_MINIMUM_D64(min1, min1, sd1);
}
if (n1 > 0) {
mlib_d64 sd2[4] = {
MLIB_D64_MAX, MLIB_D64_MAX,
MLIB_D64_MAX, MLIB_D64_MAX};
for (int i = 0; i < n1; i++) {
sd2[i] = *sp++;
}
sd0 = _mm_set_pd(sd2[3], sd2[2]);
sd1 = _mm_set_pd(sd2[1], sd2[0]);
MLIB_S_IMAGE_MINIMUM_D64(min0, min0, sd0);
MLIB_S_IMAGE_MINIMUM_D64(min1, min1, sd1);
}
sp = sl = (mlib_d64 *) ((mlib_u8 *)sl + src_stride);
}
switch (channels) {
case 1: {
mlib_d64 tmpRes0[2];
mlib_d64 tmpRes1[2];
_mm_storeu_pd(tmpRes0, min0);
_mm_storeu_pd(tmpRes1, min1);
tmpRes0[0] =
(tmpRes0[0] < tmpRes0[1]) ? tmpRes0[0] : tmpRes0[1];
tmpRes1[0] =
(tmpRes1[0] < tmpRes1[1]) ? tmpRes1[0] : tmpRes1[1];
min[0] = (tmpRes0[0] < tmpRes1[0]) ? tmpRes0[0] : tmpRes1[0];
break;
}
case 2: {
mlib_d64 tmpRes0[2];
mlib_d64 tmpRes1[2];
_mm_storeu_pd(tmpRes0, min0);
_mm_storeu_pd(tmpRes1, min1);
min[0] = (tmpRes0[0] < tmpRes1[0]) ? tmpRes0[0] : tmpRes1[0];
min[1] = (tmpRes0[1] < tmpRes1[1]) ? tmpRes0[1] : tmpRes1[1];
break;
}
case 4: {
_mm_storeu_pd(min, min0);
_mm_storeu_pd((min + 2), min1);
break;
}
}
}
int
AnalyzeSamples ( const Float_t* left_samples, const Float_t* right_samples, size_t num_samples, int num_channels )
{
const Float_t* curleft;
const Float_t* curright;
long batchsamples;
long cursamples;
long cursamplepos;
int i;
#ifdef HAVE_SSE2
__m128d __temp;
__declspec(align(16)) Float_t __temp2[2];
#endif
if ( num_samples == 0 )
return GAIN_ANALYSIS_OK;
cursamplepos = 0;
batchsamples = (long)num_samples;
switch ( num_channels) {
case 1: right_samples = left_samples;
case 2: break;
default: return GAIN_ANALYSIS_ERROR;
}
if ( num_samples < MAX_ORDER ) {
memcpy ( linprebuf + MAX_ORDER, left_samples , num_samples * sizeof(Float_t) );
memcpy ( rinprebuf + MAX_ORDER, right_samples, num_samples * sizeof(Float_t) );
}
else {
memcpy ( linprebuf + MAX_ORDER, left_samples, MAX_ORDER * sizeof(Float_t) );
memcpy ( rinprebuf + MAX_ORDER, right_samples, MAX_ORDER * sizeof(Float_t) );
}
while ( batchsamples > 0 ) {
cursamples = batchsamples > sampleWindow-totsamp ? sampleWindow - totsamp : batchsamples;
if ( cursamplepos < MAX_ORDER ) {
curleft = linpre+cursamplepos;
curright = rinpre+cursamplepos;
if (cursamples > MAX_ORDER - cursamplepos )
cursamples = MAX_ORDER - cursamplepos;
}
else {
curleft = left_samples + cursamplepos;
curright = right_samples + cursamplepos;
}
YULE_FILTER ( curleft , lstep + totsamp, cursamples, ABYule[freqindex]);
YULE_FILTER ( curright, rstep + totsamp, cursamples, ABYule[freqindex]);
BUTTER_FILTER ( lstep + totsamp, lout + totsamp, cursamples, ABButter[freqindex]);
BUTTER_FILTER ( rstep + totsamp, rout + totsamp, cursamples, ABButter[freqindex]);
curleft = lout + totsamp; // Get the squared values
curright = rout + totsamp;
#ifdef HAVE_SSE2
i = cursamples % 16;
while (i--)
{
__temp = _mm_set_pd (*curleft++, *curright++);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
}
i = cursamples / 16;
while (i--)
{
__temp = _mm_set_pd (curleft[0], curright[0]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[1], curright[1]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[2], curright[2]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[3], curright[3]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[4], curright[4]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[5], curright[5]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[6], curright[6]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[7], curright[7]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[8], curright[8]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[9], curright[9]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[10], curright[10]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[11], curright[11]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[12], curright[12]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[13], curright[13]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[14], curright[14]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
__temp = _mm_set_pd (curleft[15], curright[15]);
__temp = _mm_mul_pd(__temp, __temp);
lrsum = _mm_add_pd(lrsum, __temp);
curleft += 16;
curright += 16;
}
#else
i = cursamples % 16;
while (i--)
{
lsum += fsqr(*curleft++);
rsum += fsqr(*curright++);
}
i = cursamples / 16;
while (i--)
{
lsum += fsqr(curleft[0])
+ fsqr(curleft[1])
+ fsqr(curleft[2])
+ fsqr(curleft[3])
+ fsqr(curleft[4])
+ fsqr(curleft[5])
+ fsqr(curleft[6])
+ fsqr(curleft[7])
+ fsqr(curleft[8])
+ fsqr(curleft[9])
+ fsqr(curleft[10])
+ fsqr(curleft[11])
+ fsqr(curleft[12])
+ fsqr(curleft[13])
+ fsqr(curleft[14])
+ fsqr(curleft[15]);
curleft += 16;
rsum += fsqr(curright[0])
+ fsqr(curright[1])
+ fsqr(curright[2])
+ fsqr(curright[3])
+ fsqr(curright[4])
+ fsqr(curright[5])
+ fsqr(curright[6])
+ fsqr(curright[7])
+ fsqr(curright[8])
+ fsqr(curright[9])
+ fsqr(curright[10])
+ fsqr(curright[11])
+ fsqr(curright[12])
+ fsqr(curright[13])
+ fsqr(curright[14])
+ fsqr(curright[15]);
curright += 16;
}
#endif
batchsamples -= cursamples;
cursamplepos += cursamples;
totsamp += cursamples;
if ( totsamp == sampleWindow ) { // Get the Root Mean Square (RMS) for this set of samples
double val;
int ival;
#ifdef HAVE_SSE2
_mm_store_pd (__temp2, lrsum);
val = (Float_t)STEPS_per_dB * 10. * log10 ( (__temp2[0]+__temp2[1]) / totsamp * 0.5 + 1.e-37 );
#else
val = (Float_t)STEPS_per_dB * 10. * log10 ( (lsum+rsum) / totsamp * 0.5 + 1.e-37 );
#endif
ival = (int) val;
if ( ival < 0 ) ival = 0;
if ( ival >= (int)(sizeof(A)/sizeof(*A)) ) ival = sizeof(A)/sizeof(*A) - 1;
A [ival]++;
#ifdef HAVE_SSE2
lrsum = _mm_setzero_pd();
#else
lsum = rsum = 0.;
#endif
memmove ( loutbuf , loutbuf + totsamp, MAX_ORDER * sizeof(Float_t) );
memmove ( routbuf , routbuf + totsamp, MAX_ORDER * sizeof(Float_t) );
memmove ( lstepbuf, lstepbuf + totsamp, MAX_ORDER * sizeof(Float_t) );
memmove ( rstepbuf, rstepbuf + totsamp, MAX_ORDER * sizeof(Float_t) );
totsamp = 0;
}
if ( totsamp > sampleWindow ) // somehow I really screwed up: Error in programming! Contact author about totsamp > sampleWindow
return GAIN_ANALYSIS_ERROR;
}
if ( num_samples < MAX_ORDER ) {
memmove ( linprebuf, linprebuf + num_samples, (MAX_ORDER-num_samples) * sizeof(Float_t) );
memmove ( rinprebuf, rinprebuf + num_samples, (MAX_ORDER-num_samples) * sizeof(Float_t) );
memcpy ( linprebuf + MAX_ORDER - num_samples, left_samples, num_samples * sizeof(Float_t) );
memcpy ( rinprebuf + MAX_ORDER - num_samples, right_samples, num_samples * sizeof(Float_t) );
}
else {
memcpy ( linprebuf, left_samples + num_samples - MAX_ORDER, MAX_ORDER * sizeof(Float_t) );
memcpy ( rinprebuf, right_samples + num_samples - MAX_ORDER, MAX_ORDER * sizeof(Float_t) );
}
return GAIN_ANALYSIS_OK;
}
mlib_status
__mlib_SignalLMSFilter_F32S_F32S(
mlib_f32 *dst,
const mlib_f32 *src,
const mlib_f32 *ref,
void *filter,
mlib_s32 n)
{
LMS_Filter *pLMS_Filter = (LMS_Filter *) filter;
mlib_d64 *W[2], *srcBuffer[2];
mlib_d64 U = pLMS_Filter->U;
mlib_d64 D, Y;
mlib_s32 i, t, j, off, count;
mlib_s32 tap, bufferSize;
mlib_f32 *psrc, *pdst, *pref;
__m128d srcs1, srcs2, srcs3, srcs4;
__m128d SBE0, SD, temp;
__m128d *SW;
if (src == NULL || dst == NULL || ref == NULL)
return (MLIB_NULLPOINTER);
if (n <= 0)
return (MLIB_FAILURE);
tap = pLMS_Filter->tap;
bufferSize = pLMS_Filter->bufferSize;
SW = __mlib_malloc(tap * sizeof (__m128d));
W[0] = pLMS_Filter->data;
srcBuffer[0] = (W[0] + 2 * tap);
W[1] = srcBuffer[0] + 2 * bufferSize;
srcBuffer[1] = (W[1] + 2 * tap);
psrc = (mlib_f32 *)src;
pdst = (mlib_f32 *)dst;
pref = (mlib_f32 *)ref;
for (off = 0; off < n; off += count) {
count = n - off;
if (count > bufferSize) {
count = bufferSize;
}
for (i = 0; i < count; i++) {
srcBuffer[0][i] = (mlib_d64)psrc[2 * i];
srcBuffer[1][i] = (mlib_d64)psrc[2 * i + 1];
}
for (j = 0; j < 2; j++) {
mlib_d64 *psrcBuffer = srcBuffer[j];
mlib_d64 *pW = W[j];
mlib_d64 BE0 = pLMS_Filter->BE[j];
for (i = 0; i < (tap / 2); i++) {
SW[i] = _mm_set_pd(pW[i * 2], pW[i * 2 + 1]);
}
for (t = 0; t < count; t++) {
D = 0;
Y = pref[2 * t + j];
SBE0 = _mm_set1_pd(BE0);
SD = _mm_setzero_pd();
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i < (tap / 2); i++) {
srcs1 = _mm_loadu_pd(
psrcBuffer + t - i * 2 - 2);
srcs2 = _mm_loadu_pd(
psrcBuffer + t - i * 2 - 1);
temp = _mm_mul_pd(SBE0, srcs1);
SW[i] = _mm_add_pd(SW[i], temp);
SD = _mm_add_pd(SD,
_mm_mul_pd(srcs2, SW[i]));
}
mlib_d64 TD[2];
_mm_storeu_pd(TD, SD);
D += TD[0] + TD[1];
for (i = i * 2; i < tap; i++) {
pW[i] += BE0 * psrcBuffer[t - i - 1];
D += psrcBuffer[t - i] * pW[i];
}
BE0 = (Y - D) * U;
pdst[2 * t + j] = (mlib_f32)D;
}
mlib_d64 TW[2];
for (i = 0; i < (tap / 2); i++) {
_mm_storeu_pd(TW, SW[i]);
pW[i * 2] = TW[1];
pW[i * 2 + 1] = TW[0];
}
for (i = 0; i < tap; i++) {
psrcBuffer[i - tap] =
psrcBuffer[count + (i - tap)];
}
pLMS_Filter->BE[j] = BE0;
}
psrc += 2 * count;
pdst += 2 * count;
pref += 2 * count;
}
__mlib_free(SW);
return (MLIB_SUCCESS);
}
/* mix local carrier -----------------------------------------------------------
* mix local carrier to data
* args : char *data I data
* int dtype I data type (0:real,1:complex)
* double ti I sampling interval (s)
* int n I number of samples
* double freq I carrier frequency (Hz)
* double phi0 I initial phase (rad)
* short *I,*Q O carrier mixed data I, Q component
* return : double phase remainder
*-----------------------------------------------------------------------------*/
extern double mixcarr(const char *data, int dtype, double ti, int n,
double freq, double phi0, short *II, short *QQ)
{
const char *p;
double phi,ps,prem;
#if !defined(SSE2_ENABLE)
static short cost[CDIV]={0},sint[CDIV]={0};
int i,index;
/* initialize local carrier table */
if (!cost[0]) {
for (i=0;i<CDIV;i++) {
cost[i]=(short)floor((cos(DPI/CDIV*i)/CSCALE+0.5));
sint[i]=(short)floor((sin(DPI/CDIV*i)/CSCALE+0.5));
}
}
phi=phi0*CDIV/DPI;
ps=freq*CDIV*ti; /* phase step */
if (dtype==DTYPEIQ) { /* complex */
for (p=data;p<data+n*2;p+=2,II++,QQ++,phi+=ps) {
index=((int)phi)&CMASK;
*II=cost[index]*p[0]-sint[index]*p[1];
*QQ=sint[index]*p[0]+cost[index]*p[1];
}
}
if (dtype==DTYPEI) { /* real */
for (p=data;p<data+n;p++,II++,QQ++,phi+=ps) {
index=((int)phi)&CMASK;
*II=cost[index]*p[0];
*QQ=sint[index]*p[0];
}
}
prem=phi*DPI/CDIV;
while(prem>DPI) prem-=DPI;
return prem;
#else
static char cost[16]={0},sint[16]={0};
short I1[16]={0},I2[16]={0},Q1[16]={0},Q2[16]={0};
int i;
__m128d xmm1,xmm2,xmm3,xmm4,xmm5,xmm6,xmm7,xmm8,xmm9;
__m128i dat1,dat2,dat3,dat4,ind1,ind2,xcos,xsin;
__m128i zero=_mm_setzero_si128();
__m128i mask4=_mm_set1_epi32(15);
__m128i mask8=_mm_set1_epi16(255);
if (!cost[0]) {
for (i=0;i<16;i++) {
cost[i]=(char)floor((cos(DPI/16*i)/CSCALE+0.5));
sint[i]=(char)floor((sin(DPI/16*i)/CSCALE+0.5));
}
}
phi=phi0/DPI*16-floor(phi0/DPI)*16;
ps=freq*16*ti;
xmm1=_mm_set_pd(phi+ps,phi); phi+=ps*2;
xmm2=_mm_set_pd(phi+ps,phi); phi+=ps*2;
xmm3=_mm_set_pd(phi+ps,phi); phi+=ps*2;
xmm4=_mm_set_pd(phi+ps,phi); phi+=ps*2;
xmm5=_mm_set_pd(phi+ps,phi); phi+=ps*2;
xmm6=_mm_set_pd(phi+ps,phi); phi+=ps*2;
xmm7=_mm_set_pd(phi+ps,phi); phi+=ps*2;
xmm8=_mm_set_pd(phi+ps,phi); phi+=ps*2;
xmm9=_mm_set1_pd(ps*16);
xcos=_mm_loadu_si128((__m128i *)cost);
xsin=_mm_loadu_si128((__m128i *)sint);
if (dtype==DTYPEIQ) { /* complex */
for (p=data;p<data+n*2;p+=32,II+=16,QQ+=16) {
LOAD_INT8C(dat1,dat2,p ,zero,mask8);
LOAD_INT8C(dat3,dat4,p+16,zero,mask8);
DBLTOINT16(ind1,xmm1,xmm2,xmm3,xmm4,mask4);
DBLTOINT16(ind2,xmm5,xmm6,xmm7,xmm8,mask4);
ind1=_mm_packus_epi16(ind1,ind2);
MIX_INT8(I1,dat1,dat3,xcos,ind1,zero);
MIX_INT8(I2,dat1,dat3,xsin,ind1,zero);
MIX_INT8(Q1,dat2,dat4,xsin,ind1,zero);
MIX_INT8(Q2,dat2,dat4,xcos,ind1,zero);
for (i=0;i<16;i++) {
II[i]=I1[i]-Q1[i];
QQ[i]=I2[i]+Q2[i];
}
xmm1=_mm_add_pd(xmm1,xmm9);
xmm2=_mm_add_pd(xmm2,xmm9);
xmm3=_mm_add_pd(xmm3,xmm9);
xmm4=_mm_add_pd(xmm4,xmm9);
xmm5=_mm_add_pd(xmm5,xmm9);
xmm6=_mm_add_pd(xmm6,xmm9);
xmm7=_mm_add_pd(xmm7,xmm9);
xmm8=_mm_add_pd(xmm8,xmm9);
}
}
if (dtype==DTYPEI) { /* real */
for (p=data;p<data+n;p+=16,II+=16,QQ+=16) {
LOAD_INT8(dat1,dat2,p,zero);
DBLTOINT16(ind1,xmm1,xmm2,xmm3,xmm4,mask4);
DBLTOINT16(ind2,xmm5,xmm6,xmm7,xmm8,mask4);
ind1=_mm_packus_epi16(ind1,ind2);
MIX_INT8(II,dat1,dat2,xcos,ind1,zero);
MIX_INT8(QQ,dat1,dat2,xsin,ind1,zero);
xmm1=_mm_add_pd(xmm1,xmm9);
xmm2=_mm_add_pd(xmm2,xmm9);
xmm3=_mm_add_pd(xmm3,xmm9);
xmm4=_mm_add_pd(xmm4,xmm9);
xmm5=_mm_add_pd(xmm5,xmm9);
xmm6=_mm_add_pd(xmm6,xmm9);
xmm7=_mm_add_pd(xmm7,xmm9);
xmm8=_mm_add_pd(xmm8,xmm9);
}
}
prem=phi0+freq*ti*n*DPI;
while(prem>DPI) prem-=DPI;
return prem;
#endif
}
void exchrhs_gmrfData_8(unsigned int slot) {
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((!neighbor_isValid[0][0])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S589, S588, S590 */
{
{
{
int i1 = 0;
for (; (i1<(1&(~1))); i1 += 1) {
yPos = (((i1/2.560000e+02)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(2.560000e+02);
__m128d vec4 = _mm_set1_pd(yPos);
for (; (i1<254); i1 += 4) {
/* yPos = (((i1/2.560000e+02)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec2 = _mm_load1_pd((&posEnd[1]));
__m128d vec2_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec3 = _mm_load1_pd((&posBegin[1]));
__m128d vec3_2 = _mm_load1_pd((&posBegin[1]));
vec4 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(vec0, vec1), _mm_sub_pd(vec2, vec3)), vec3);
vec4 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(vec0_2, vec1), _mm_sub_pd(vec2_2, vec3_2)), vec3_2);
}
for (; (i1<257); i1 += 1) {
yPos = (((i1/2.560000e+02)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
{
int i1 = 0;
for (; (i1<(1&(~1))); i1 += 1) {
xPos = posBegin[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<254); i1 += 4) {
/* xPos = posBegin[0]; */
__m128d vec0 = _mm_load1_pd((&posBegin[0]));
__m128d vec0_2 = _mm_load1_pd((&posBegin[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<257); i1 += 1) {
xPos = posBegin[0];
}
}
}
{
double* fieldData_RHS_GMRF_8_p1 = (&fieldData_RHS_GMRF[8][0]);
int i1 = 0;
for (; (i1<=255); i1 += 2) {
fieldData_RHS_GMRF_8_p1[(i1*258)] = 0.000000e+00;
fieldData_RHS_GMRF_8_p1[((i1*258)+258)] = 0.000000e+00;
}
for (; (i1<=256); i1 += 1) {
fieldData_RHS_GMRF_8_p1[(i1*258)] = 0.000000e+00;
}
}
}
}
}
if ((!neighbor_isValid[0][1])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S592, S591, S593 */
{
{
{
int i1 = 0;
for (; (i1<(1&(~1))); i1 += 1) {
yPos = (((i1/2.560000e+02)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(2.560000e+02);
__m128d vec4 = _mm_set1_pd(yPos);
for (; (i1<254); i1 += 4) {
/* yPos = (((i1/2.560000e+02)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec2 = _mm_load1_pd((&posEnd[1]));
__m128d vec2_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec3 = _mm_load1_pd((&posBegin[1]));
__m128d vec3_2 = _mm_load1_pd((&posBegin[1]));
vec4 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(vec0, vec1), _mm_sub_pd(vec2, vec3)), vec3);
vec4 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(vec0_2, vec1), _mm_sub_pd(vec2_2, vec3_2)), vec3_2);
}
for (; (i1<257); i1 += 1) {
yPos = (((i1/2.560000e+02)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
{
double* fieldData_RHS_GMRF_8_p1 = (&fieldData_RHS_GMRF[8][0]);
int i1 = 0;
for (; (i1<=255); i1 += 2) {
fieldData_RHS_GMRF_8_p1[((i1*258)+256)] = 0.000000e+00;
fieldData_RHS_GMRF_8_p1[((i1*258)+514)] = 0.000000e+00;
}
for (; (i1<=256); i1 += 1) {
fieldData_RHS_GMRF_8_p1[((i1*258)+256)] = 0.000000e+00;
}
}
}
{
int i1 = 0;
for (; (i1<(1&(~1))); i1 += 1) {
xPos = posEnd[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<254); i1 += 4) {
/* xPos = posEnd[0]; */
__m128d vec0 = _mm_load1_pd((&posEnd[0]));
__m128d vec0_2 = _mm_load1_pd((&posEnd[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<257); i1 += 1) {
xPos = posEnd[0];
}
}
}
}
}
if ((!neighbor_isValid[0][2])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S595, S594, S596 */
{
{
{
double* fieldData_RHS_GMRF_8_p1 = (&fieldData_RHS_GMRF[8][0]);
int i2 = 0;
for (; (i2<=255); i2 += 2) {
fieldData_RHS_GMRF_8_p1[i2] = 0.000000e+00;
fieldData_RHS_GMRF_8_p1[(i2+1)] = 0.000000e+00;
}
for (; (i2<=256); i2 += 1) {
fieldData_RHS_GMRF_8_p1[i2] = 0.000000e+00;
}
}
{
int i2 = 0;
for (; (i2<=255); i2 += 2) {
xPos = (((i2/2.560000e+02)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2+1)/2.560000e+02)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=256); i2 += 1) {
xPos = (((i2/2.560000e+02)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
}
{
int i2 = 0;
for (; (i2<=255); i2 += 2) {
yPos = posBegin[1];
yPos = posBegin[1];
}
for (; (i2<=256); i2 += 1) {
yPos = posBegin[1];
}
}
}
}
}
if ((!neighbor_isValid[0][3])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S598, S597, S599 */
{
{
{
int i2 = 0;
for (; (i2<=255); i2 += 2) {
xPos = (((i2/2.560000e+02)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2+1)/2.560000e+02)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=256); i2 += 1) {
xPos = (((i2/2.560000e+02)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
{
int i2 = 0;
for (; (i2<=255); i2 += 2) {
yPos = posEnd[1];
yPos = posEnd[1];
}
for (; (i2<=256); i2 += 1) {
yPos = posEnd[1];
}
}
}
{
double* fieldData_RHS_GMRF_8_p1 = (&fieldData_RHS_GMRF[8][0]);
int i2 = 0;
for (; (i2<=255); i2 += 2) {
fieldData_RHS_GMRF_8_p1[(i2+66048)] = 0.000000e+00;
fieldData_RHS_GMRF_8_p1[(i2+66049)] = 0.000000e+00;
}
for (; (i2<=256); i2 += 1) {
fieldData_RHS_GMRF_8_p1[(i2+66048)] = 0.000000e+00;
}
}
}
}
}
}
}
}
int main()
{
#ifndef __EMSCRIPTEN__
_MM_SET_FLUSH_ZERO_MODE(_MM_FLUSH_ZERO_ON);
#endif
printf ("{ \"workload\": %u, \"results\": [\n", N);
assert(N%2 == 0); // Don't care about the tail for now.
double *src = get_src_d();//(float*)aligned_alloc(16, N*sizeof(float));
for(int i = 0; i < N; ++i)
src[i] = (double)rand() / RAND_MAX;
double *src2 = get_src2_d();//(float*)aligned_alloc(16, N*sizeof(float));
for(int i = 0; i < N; ++i)
src2[i] = (double)rand() / RAND_MAX;
double *dst = get_dst_d();//(float*)aligned_alloc(16, N*sizeof(float));
float scalarTime;
SETCHART("load");
START();
for(int i = 0; i < N; ++i)
dst[i] = src[i];
ENDSCALAR(checksum_dst(dst), "scalar");
LS_TEST("_mm_load_pd", _mm_load_pd, 0, _mm_store_pd, double*, 0, 2);
LS_TEST("_mm_load_pd1", _mm_load_pd1, 1, _mm_store_pd, double*, 0, 2);
LS_TEST("_mm_load_sd", _mm_load_sd, 1, _mm_store_pd, double*, 0, 2);
// _mm_load_si128
LS_TEST("_mm_load1_pd", _mm_load1_pd, 1, _mm_store_pd, double*, 0, 2);
__m128d tempReg = _mm_set_pd(1.0, 2.0);
LSH_TEST("_mm_loadh_pd", tempReg, _mm_loadh_pd, double*, 1, _mm_store_pd, double*, 0, 2);
// _mm_loadl_epi64
LSH_TEST("_mm_loadl_pd", tempReg, _mm_loadh_pd, double*, 1, _mm_store_pd, double*, 0, 2);
LS_TEST("_mm_loadr_pd", _mm_loadr_pd, 0, _mm_store_pd, double*, 0, 2);
LS_TEST("_mm_loadu_pd", _mm_loadu_pd, 1, _mm_store_pd, double*, 0, 2);
// _mm_loadu_si128
SETCHART("set");
/* _mm_set_epi16
_mm_set_epi32
_mm_set_epi64
_mm_set_epi64x
_mm_set_epi8 */
SS_TEST_D("_mm_set_pd", _mm_set_pd(src[i+2], src[i+0]));
//SS_TEST_D("_mm_set_pd1", _mm_set_pd1(src[i]));
SS_TEST_D("_mm_set_sd", _mm_set_sd(src[i]));
/* _mm_set1_epi16
_mm_set1_epi32
_mm_set1_epi64
_mm_set1_epi64x
_mm_set1_epi8 */
SS_TEST_D("_mm_set1_pd", _mm_set1_pd(src[i]));
/* _mm_setr_epi16
_mm_setr_epi32
_mm_setr_epi64
_mm_setr_epi8 */
SS_TEST_D("_mm_setr_pd", _mm_set_pd(src[i+2], src[i+0]));
SS_TEST_D("_mm_setzero_pd", _mm_setzero_pd());
// _mm_setzero_si128
SETCHART("move");
// _mm_move_epi64
SS_TEST_D("_mm_move_sd", _mm_move_sd(_mm_load_pd(src+i), _mm_load_pd(src2+i)));
SETCHART("store");
// _mm_maskmoveu_si128
LS_TEST("_mm_store_pd", _mm_load_pd, 0, _mm_store_pd, double*, 0, 2);
// LS_TEST("_mm_store_pd1", _mm_load_pd, 0, _mm_store_pd1, double*, 0);
LS_TEST("_mm_store_sd", _mm_load_pd, 0, _mm_store_sd, double*, 1, 2);
// _mm_store_si128
// _mm_store1_pd
LS64_TEST("_mm_storeh_pi", _mm_load_pd, 0, _mm_storeh_pi, 1, 2);
// _mm_storel_epi64
LS64_TEST("_mm_storel_pi", _mm_load_pd, 0, _mm_storel_pi, 1, 2);
LS_TEST("_mm_storer_pd", _mm_load_pd, 0, _mm_storer_pd, double*, 0, 2);
LS_TEST("_mm_storeu_pd", _mm_load_pd, 0, _mm_storeu_pd, double*, 1, 2);
// _mm_storeu_si128
LS_TEST("_mm_stream_pd", _mm_load_pd, 0, _mm_stream_pd, double*, 0, 2);
// _mm_stream_si128
// _mm_stream_si32
// _mm_stream_si64
SETCHART("arithmetic");
// _mm_add_epi16
// _mm_add_epi32
// _mm_add_epi64
// _mm_add_epi8
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] += src2[0]; dst[1] += src2[1]; dst[2] += src2[2]; dst[3] += src2[3]; } ENDSCALAR(checksum_dst(dst), "scalar add");
BINARYOP_TEST_D("_mm_add_pd", _mm_add_pd, _mm_load_pd(src), _mm_load_pd(src2));
BINARYOP_TEST_D("_mm_add_sd", _mm_add_sd, _mm_load_pd(src), _mm_load_pd(src2));
// _mm_adds_epi16
// _mm_adds_epi8
// _mm_adds_epu16
// _mm_adds_epu8
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] /= src2[0]; dst[1] /= src2[1]; dst[2] /= src2[2]; dst[3] /= src2[3]; } ENDSCALAR(checksum_dst(dst), "scalar div");
BINARYOP_TEST_D("_mm_div_pd", _mm_div_pd, _mm_load_pd(src), _mm_load_pd(src2));
BINARYOP_TEST_D("_mm_div_sd", _mm_div_sd, _mm_load_pd(src), _mm_load_pd(src2));
// _mm_madd_epi16
// _mm_mul_epu32
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] *= src2[0]; dst[1] *= src2[1]; dst[2] *= src2[2]; dst[3] *= src2[3]; } ENDSCALAR(checksum_dst(dst), "scalar mul");
BINARYOP_TEST_D("_mm_mul_pd", _mm_mul_pd, _mm_load_pd(src), _mm_load_pd(src2));
BINARYOP_TEST_D("_mm_mul_sd", _mm_mul_sd, _mm_load_pd(src), _mm_load_pd(src2));
// _mm_mulhi_epi16
// _mm_mulhi_epu16
// _mm_mullo_epi16
// _mm_sad_epu8
// _mm_sub_epi16
// _mm_sub_epi32
// _mm_sub_epi64
// _mm_sub_epi8
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] -= src2[0]; dst[1] -= src2[1]; dst[2] -= src2[2]; dst[3] -= src2[3]; } ENDSCALAR(checksum_dst(dst), "scalar sub");
BINARYOP_TEST_D("_mm_sub_pd", _mm_sub_pd, _mm_load_pd(src), _mm_load_pd(src2));
BINARYOP_TEST_D("_mm_sub_sd", _mm_sub_sd, _mm_load_pd(src), _mm_load_pd(src2));
// _mm_subs_epi16
// _mm_subs_epi8
// _mm_subs_epu16
// _mm_subs_epu8
SETCHART("roots");
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = sqrt(dst[0]); dst[1] = sqrt(dst[1]); dst[2] = sqrt(dst[2]); dst[3] = sqrt(dst[3]); } ENDSCALAR(checksum_dst(dst), "scalar sqrt");
UNARYOP_TEST_D("_mm_sqrt_pd", _mm_sqrt_pd, _mm_load_pd(src));
// UNARYOP_TEST_D("_mm_sqrt_sd", _mm_sqrt_sd, _mm_load_pd(src));
SETCHART("logical");
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = ucastd(dcastu(dst[0]) & dcastu(src2[0])); dst[1] = ucastd(dcastu(dst[1]) & dcastu(src2[1])); dst[2] = ucastd(dcastu(dst[2]) & dcastu(src2[2])); dst[3] = ucastd(dcastu(dst[3]) & dcastu(src2[3])); } ENDSCALAR(checksum_dst(dst), "scalar and");
BINARYOP_TEST_D("_mm_and_pd", _mm_and_pd, _mm_load_pd(src), _mm_load_pd(src2));
// _mm_and_si128
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = ucastd((~dcastu(dst[0])) & dcastu(src2[0])); dst[1] = ucastd((~dcastu(dst[1])) & dcastu(src2[1])); dst[2] = ucastd((~dcastu(dst[2])) & dcastu(src2[2])); dst[3] = ucastd((~dcastu(dst[3])) & dcastu(src2[3])); } ENDSCALAR(checksum_dst(dst), "scalar andnot");
BINARYOP_TEST_D("_mm_andnot_pd", _mm_andnot_pd, _mm_load_pd(src), _mm_load_pd(src2));
// _mm_andnot_si128
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = ucastd(dcastu(dst[0]) | dcastu(src2[0])); dst[1] = ucastd(dcastu(dst[1]) | dcastu(src2[1])); dst[2] = ucastd(dcastu(dst[2]) | dcastu(src2[2])); dst[3] = ucastd(dcastu(dst[3]) | dcastu(src2[3])); } ENDSCALAR(checksum_dst(dst), "scalar or");
BINARYOP_TEST_D("_mm_or_pd", _mm_or_pd, _mm_load_pd(src), _mm_load_pd(src2));
// _mm_or_si128
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = ucastd(dcastu(dst[0]) ^ dcastu(src2[0])); dst[1] = ucastd(dcastu(dst[1]) ^ dcastu(src2[1])); dst[2] = ucastd(dcastu(dst[2]) ^ dcastu(src2[2])); dst[3] = ucastd(dcastu(dst[3]) ^ dcastu(src2[3])); } ENDSCALAR(checksum_dst(dst), "scalar xor");
BINARYOP_TEST_D("_mm_xor_pd", _mm_xor_pd, _mm_load_pd(src), _mm_load_pd(src2));
// _mm_xor_si128
SETCHART("cmp");
// _mm_cmpeq_epi16
// _mm_cmpeq_epi32
// _mm_cmpeq_epi8
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = (dst[0] == src2[0]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[1] = (dst[1] == src2[1]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[2] = (dst[2] == src2[2]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[3] = (dst[3] == src2[3]) ? ucastd(0xFFFFFFFFU) : 0.f; } ENDSCALAR(checksum_dst(dst), "scalar cmp==");
BINARYOP_TEST_D("_mm_cmpeq_pd", _mm_cmpeq_pd, _mm_load_pd(src), _mm_load_pd(src2));
BINARYOP_TEST_D("_mm_cmpeq_sd", _mm_cmpeq_sd, _mm_load_pd(src), _mm_load_pd(src2));
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = (dst[0] >= src2[0]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[1] = (dst[1] >= src2[1]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[2] = (dst[2] >= src2[2]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[3] = (dst[3] >= src2[3]) ? ucastd(0xFFFFFFFFU) : 0.f; } ENDSCALAR(checksum_dst(dst), "scalar cmp>=");
BINARYOP_TEST_D("_mm_cmpge_pd", _mm_cmpge_pd, _mm_load_pd(src), _mm_load_pd(src2));
BINARYOP_TEST_D("_mm_cmpge_sd", _mm_cmpge_sd, _mm_load_pd(src), _mm_load_pd(src2));
// _mm_cmpgt_epi16
// _mm_cmpgt_epi32
// _mm_cmpgt_epi8
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = (dst[0] > src2[0]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[1] = (dst[1] > src2[1]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[2] = (dst[2] > src2[2]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[3] = (dst[3] > src2[3]) ? ucastd(0xFFFFFFFFU) : 0.f; } ENDSCALAR(checksum_dst(dst), "scalar cmp>");
BINARYOP_TEST_D("_mm_cmpgt_pd", _mm_cmpgt_pd, _mm_load_pd(src), _mm_load_pd(src2));
BINARYOP_TEST_D("_mm_cmpgt_sd", _mm_cmpgt_sd, _mm_load_pd(src), _mm_load_pd(src2));
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = (dst[0] <= src2[0]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[1] = (dst[1] <= src2[1]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[2] = (dst[2] <= src2[2]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[3] = (dst[3] <= src2[3]) ? ucastd(0xFFFFFFFFU) : 0.f; } ENDSCALAR(checksum_dst(dst), "scalar cmp<=");
BINARYOP_TEST_D("_mm_cmple_pd", _mm_cmple_pd, _mm_load_pd(src), _mm_load_pd(src2));
BINARYOP_TEST_D("_mm_cmple_sd", _mm_cmple_sd, _mm_load_pd(src), _mm_load_pd(src2));
// _mm_cmplt_epi16
// _mm_cmplt_epi32
// _mm_cmplt_epi8
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = (dst[0] < src2[0]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[1] = (dst[1] < src2[1]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[2] = (dst[2] < src2[2]) ? ucastd(0xFFFFFFFFU) : 0.f; dst[3] = (dst[3] < src2[3]) ? ucastd(0xFFFFFFFFU) : 0.f; } ENDSCALAR(checksum_dst(dst), "scalar cmp<");
BINARYOP_TEST_D("_mm_cmplt_pd", _mm_cmplt_pd, _mm_load_pd(src), _mm_load_pd(src2));
BINARYOP_TEST_D("_mm_cmplt_sd", _mm_cmplt_sd, _mm_load_pd(src), _mm_load_pd(src2));
/*_mm_cmpneq_pd
_mm_cmpneq_sd
_mm_cmpnge_pd
_mm_cmpnge_sd
_mm_cmpngt_pd
_mm_cmpngt_sd
_mm_cmpnle_pd
_mm_cmpnle_sd
_mm_cmpnlt_pd
_mm_cmpnlt_sd*/
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = (!Isnan(dst[0]) && !Isnan(src2[0])) ? ucastd(0xFFFFFFFFU) : 0.f; dst[1] = (!Isnan(dst[1]) && !Isnan(src2[1])) ? ucastd(0xFFFFFFFFU) : 0.f; dst[2] = (!Isnan(dst[2]) && !Isnan(src2[2])) ? ucastd(0xFFFFFFFFU) : 0.f; dst[3] = (!Isnan(dst[3]) && !Isnan(src2[3])) ? ucastd(0xFFFFFFFFU) : 0.f; } ENDSCALAR(checksum_dst(dst), "scalar cmpord");
BINARYOP_TEST_D("_mm_cmpord_pd", _mm_cmpord_pd, _mm_load_pd(src), _mm_load_pd(src2));
BINARYOP_TEST_D("_mm_cmpord_sd", _mm_cmpord_sd, _mm_load_pd(src), _mm_load_pd(src2));
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = (Isnan(dst[0]) || Isnan(src2[0])) ? ucastd(0xFFFFFFFFU) : 0.f; dst[1] = (Isnan(dst[1]) || Isnan(src2[1])) ? ucastd(0xFFFFFFFFU) : 0.f; dst[2] = (Isnan(dst[2]) || Isnan(src2[2])) ? ucastd(0xFFFFFFFFU) : 0.f; dst[3] = (Isnan(dst[3]) || Isnan(src2[3])) ? ucastd(0xFFFFFFFFU) : 0.f; } ENDSCALAR(checksum_dst(dst), "scalar cmpunord");
BINARYOP_TEST_D("_mm_cmpunord_pd", _mm_cmpunord_pd, _mm_load_pd(src), _mm_load_pd(src2));
BINARYOP_TEST_D("_mm_cmpunord_sd", _mm_cmpunord_sd, _mm_load_pd(src), _mm_load_pd(src2));
SETCHART("max");
// _mm_max_epi16
// _mm_max_epu8
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = Max(dst[0], src2[0]); dst[1] = Max(dst[1], src2[1]); dst[2] = Max(dst[2], src2[2]); dst[3] = Max(dst[3], src2[3]); } ENDSCALAR(checksum_dst(dst), "scalar max");
BINARYOP_TEST_D("_mm_max_pd", _mm_max_pd, _mm_load_pd(src), _mm_load_pd(src2));
BINARYOP_TEST_D("_mm_max_sd", _mm_max_sd, _mm_load_pd(src), _mm_load_pd(src2));
// _mm_min_epi16
// _mm_min_epu8
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = Min(dst[0], src2[0]); dst[1] = Min(dst[1], src2[1]); dst[2] = Min(dst[2], src2[2]); dst[3] = Min(dst[3], src2[3]); } ENDSCALAR(checksum_dst(dst), "scalar min");
BINARYOP_TEST_D("_mm_min_pd", _mm_min_pd, _mm_load_pd(src), _mm_load_pd(src2));
BINARYOP_TEST_D("_mm_min_sd", _mm_min_sd, _mm_load_pd(src), _mm_load_pd(src2));
SETCHART("shuffle");
// _mm_extract_epi16
// _mm_insert_epi16
// _mm_shuffle_epi32
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[3] = dst[1]; dst[2] = dst[0]; dst[1] = src2[3]; dst[0] = src2[2]; } ENDSCALAR(checksum_dst(dst), "scalar shuffle");
// BINARYOP_TEST_D("_mm_shuffle_pd", _mm_shuffle_pd, _mm_load_pd(src), _mm_load_pd(src2));
START();
__m128 o0 = _mm_load_pd(src);
__m128 o1 = _mm_load_pd(src2);
for(int i = 0; i < N; i += 4)
o0 = _mm_shuffle_pd(o0, o1, _MM_SHUFFLE(1, 0, 3, 2));
_mm_store_pd(dst, o0);
END(checksum_dst(dst), "_mm_shuffle_pd");
// _mm_shufflehi_epi16
// _mm_shufflelo_epi16
// _mm_unpackhi_epi16
// _mm_unpackhi_epi32
// _mm_unpackhi_epi64
// _mm_unpackhi_epi8
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[0] = dst[2]; dst[1] = src2[2]; dst[2] = dst[3]; dst[3] = src2[3]; } ENDSCALAR(checksum_dst(dst), "scalar unpackhi_pd");
BINARYOP_TEST_D("_mm_unpackhi_pd", _mm_unpackhi_pd, _mm_load_pd(src), _mm_load_pd(src2));
// _mm_unpacklo_epi16
// _mm_unpacklo_epi32
// _mm_unpacklo_epi64
// _mm_unpacklo_epi8
START(); dst[0] = src[0]; dst[1] = src[1]; dst[2] = src[2]; dst[3] = src[3]; for(int i = 0; i < N; ++i) { dst[2] = dst[1]; dst[1] = dst[0]; dst[0] = src2[0]; dst[3] = src2[1]; } ENDSCALAR(checksum_dst(dst), "scalar unpacklo_pd");
BINARYOP_TEST_D("_mm_unpacklo_pd", _mm_unpacklo_pd, _mm_load_pd(src), _mm_load_pd(src2));
printf("]}\n");
/*
printf("Finished!\n");
printf("Total time spent in scalar intrinsics: %f msecs.\n", (double)scalarTotalTicks * 1000.0 / ticks_per_sec());
printf("Total time spent in SSE1 intrinsics: %f msecs.\n", (double)simdTotalTicks * 1000.0 / ticks_per_sec());
if (scalarTotalTicks > simdTotalTicks)
printf("SSE1 was %.3fx faster than scalar!\n", (double)scalarTotalTicks / simdTotalTicks);
else
printf("SSE1 was %.3fx slower than scalar!\n", (double)simdTotalTicks / scalarTotalTicks);
*/
#ifdef __EMSCRIPTEN__
fprintf(stderr,"User Agent: %s\n", emscripten_run_script_string("navigator.userAgent"));
printf("/*Test finished! Now please close Firefox to continue with benchmark_sse2.py.*/\n");
#endif
exit(0);
}
// =============================================================================
//
// sse3_vChirpData
// version by: Alex Kan
// http://tbp.berkeley.edu/~alexkan/seti/
//
int sse3_ChirpData_ak(
sah_complex * cx_DataArray,
sah_complex * cx_ChirpDataArray,
int chirp_rate_ind,
double chirp_rate,
int ul_NumDataPoints,
double sample_rate
) {
int i;
#ifdef USE_MANUAL_CALLSTACK
call_stack.enter("sse3_ChirpData_ak()");
#endif
if (chirp_rate_ind == 0) {
memcpy(cx_ChirpDataArray, cx_DataArray, (int)ul_NumDataPoints * sizeof(sah_complex) );
#ifdef USE_MANUAL_CALLSTACK
call_stack.exit();
#endif
return 0;
}
int vEnd;
double srate = chirp_rate * 0.5 / (sample_rate * sample_rate);
__m128d rate = _mm_set1_pd(chirp_rate * 0.5 / (sample_rate * sample_rate));
__m128d roundVal = _mm_set1_pd(srate >= 0.0 ? TWO_TO_52 : -TWO_TO_52);
// main vectorised loop
vEnd = ul_NumDataPoints - (ul_NumDataPoints & 3);
for (i = 0; i < vEnd; i += 4) {
const float *data = (const float *) (cx_DataArray + i);
float *chirped = (float *) (cx_ChirpDataArray + i);
__m128d di = _mm_set1_pd(i);
__m128d a1 = _mm_add_pd(_mm_set_pd(1.0, 0.0), di);
__m128d a2 = _mm_add_pd(_mm_set_pd(3.0, 2.0), di);
__m128 d1, d2;
__m128 cd1, cd2;
__m128 td1, td2;
__m128 x;
__m128 y;
__m128 s;
__m128 c;
__m128 m;
// load the signal to be chirped
prefetchnta((const void *)( data+32 ));
d1 = _mm_load_ps(data);
d2 = _mm_load_ps(data+4);
// calculate the input angle
a1 = _mm_mul_pd(_mm_mul_pd(a1, a1), rate);
a2 = _mm_mul_pd(_mm_mul_pd(a2, a2), rate);
// reduce the angle to the range (-0.5, 0.5)
a1 = _mm_sub_pd(a1, _mm_sub_pd(_mm_add_pd(a1, roundVal), roundVal));
a2 = _mm_sub_pd(a2, _mm_sub_pd(_mm_add_pd(a2, roundVal), roundVal));
// convert pair of packed double into packed single
x = _mm_movelh_ps(_mm_cvtpd_ps(a1), _mm_cvtpd_ps(a2));
// square to the range [0, 0.25)
y = _mm_mul_ps(x, x);
// perform the initial polynomial approximations
s = _mm_mul_ps(_mm_add_ps(_mm_mul_ps(_mm_add_ps(_mm_mul_ps(_mm_add_ps(_mm_mul_ps(y, SS4),
SS3),
y),
SS2),
y),
SS1),
x);
c = _mm_add_ps(_mm_mul_ps(_mm_add_ps(_mm_mul_ps(_mm_add_ps(_mm_mul_ps(y, CC3),
CC2),
y),
CC1),
y),
ONE);
// perform first angle doubling
x = _mm_sub_ps(_mm_mul_ps(c, c), _mm_mul_ps(s, s));
y = _mm_mul_ps(_mm_mul_ps(s, c), TWO);
// calculate scaling factor to correct the magnitude
// m1 = vec_nmsub(y1, y1, vec_nmsub(x1, x1, TWO));
// m2 = vec_nmsub(y2, y2, vec_nmsub(x2, x2, TWO));
m = vec_recip3(_mm_add_ps(_mm_mul_ps(x, x), _mm_mul_ps(y, y)));
// perform second angle doubling
c = _mm_sub_ps(_mm_mul_ps(x, x), _mm_mul_ps(y, y));
s = _mm_mul_ps(_mm_mul_ps(y, x), TWO);
// correct the magnitude (final sine / cosine approximations)
s = _mm_mul_ps(s, m);
c = _mm_mul_ps(c, m);
// chirp the data
cd1 = _mm_shuffle_ps(c, c, 0x50);
cd2 = _mm_shuffle_ps(c, c, 0xfa);
cd1 = _mm_mul_ps(cd1, d1);
cd2 = _mm_mul_ps(cd2, d2);
d1 = _mm_shuffle_ps(d1, d1, 0xb1);
d2 = _mm_shuffle_ps(d2, d2, 0xb1);
td1 = _mm_shuffle_ps(s, s, 0x50);
td2 = _mm_shuffle_ps(s, s, 0xfa);
td1 = _mm_mul_ps(td1, d1);
td2 = _mm_mul_ps(td2, d2);
cd1 = _mm_addsub_ps(cd1, td1);
cd2 = _mm_addsub_ps(cd2, td2);
// store chirped values
_mm_stream_ps(chirped, cd1);
_mm_stream_ps(chirped+4, cd2);
}
_mm_sfence();
// handle tail elements with scalar code
for ( ; i < ul_NumDataPoints; ++i) {
double angle = srate * i * i * 0.5;
double s = sin(angle);
double c = cos(angle);
float re = cx_DataArray[i][0];
float im = cx_DataArray[i][1];
cx_ChirpDataArray[i][0] = re * c - im * s;
cx_ChirpDataArray[i][1] = re * s + im * c;
}
analysis_state.FLOP_counter+=12.0*ul_NumDataPoints;
#ifdef USE_MANUAL_CALLSTACK
call_stack.exit();
#endif
return 0;
}
void exchlaplacecoeff_gmrfData_0(unsigned int slot) {
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((!neighbor_isValid[0][0])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S1053, S1056, S1059, S1050, S1058, S1052, S1055, S1060, S1054, S1057, S1051 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+26)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+32)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+26)] = 0.000000e+00;
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+146)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+152)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+146)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+98)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+104)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+98)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+74)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+80)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+74)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
xPos = posBegin[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<0); i1 += 4) {
/* xPos = posBegin[0]; */
__m128d vec0 = _mm_load1_pd((&posBegin[0]));
__m128d vec0_2 = _mm_load1_pd((&posBegin[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<3); i1 += 1) {
xPos = posBegin[0];
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+122)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+128)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+122)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+170)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+176)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+170)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+194)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+200)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+194)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
yPos = ((((i1-1)/1.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(1.000000e+00);
__m128d vec2 = _mm_set1_pd(1.000000e+00);
__m128d vec5 = _mm_set1_pd(yPos);
for (; (i1<0); i1 += 4) {
/* yPos = ((((i1-1)/1.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec3 = _mm_load1_pd((&posEnd[1]));
__m128d vec3_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec4 = _mm_load1_pd((&posBegin[1]));
__m128d vec4_2 = _mm_load1_pd((&posBegin[1]));
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0, vec1), vec2), _mm_sub_pd(vec3, vec4)), vec4);
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0_2, vec1), vec2), _mm_sub_pd(vec3_2, vec4_2)), vec4_2);
}
for (; (i1<3); i1 += 1) {
yPos = ((((i1-1)/1.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+2)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+8)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+2)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+50)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+56)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+50)] = 0.000000e+00;
}
}
}
}
}
if ((!neighbor_isValid[0][1])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S1071, S1065, S1068, S1062, S1070, S1064, S1067, S1061, S1069, S1063, S1066 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+195)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+201)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+195)] = 0.000000e+00;
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+51)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+57)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+51)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+75)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+81)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+75)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+3)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+9)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+3)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+171)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+177)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+171)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
yPos = ((((i1-1)/1.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(1.000000e+00);
__m128d vec2 = _mm_set1_pd(1.000000e+00);
__m128d vec5 = _mm_set1_pd(yPos);
for (; (i1<0); i1 += 4) {
/* yPos = ((((i1-1)/1.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec3 = _mm_load1_pd((&posEnd[1]));
__m128d vec3_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec4 = _mm_load1_pd((&posBegin[1]));
__m128d vec4_2 = _mm_load1_pd((&posBegin[1]));
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0, vec1), vec2), _mm_sub_pd(vec3, vec4)), vec4);
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0_2, vec1), vec2), _mm_sub_pd(vec3_2, vec4_2)), vec4_2);
}
for (; (i1<3); i1 += 1) {
yPos = ((((i1-1)/1.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+99)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+105)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+99)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+123)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+129)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+123)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+147)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+153)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+147)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
xPos = posEnd[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<0); i1 += 4) {
/* xPos = posEnd[0]; */
__m128d vec0 = _mm_load1_pd((&posEnd[0]));
__m128d vec0_2 = _mm_load1_pd((&posEnd[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<3); i1 += 1) {
xPos = posEnd[0];
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+27)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+33)] = 0.000000e+00;
}
for (; (i1<=2); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+27)] = 0.000000e+00;
}
}
}
}
}
if ((!neighbor_isValid[0][2])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S1080, S1074, S1077, S1082, S1076, S1079, S1073, S1072, S1081, S1075, S1078 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+126)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+127)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+126)] = 0.000000e+00;
}
}
{
int i2 = 2;
for (; (i2<=2); i2 += 2) {
xPos = ((((i2-2)/1.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2-1)/1.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=3); i2 += 1) {
xPos = ((((i2-2)/1.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+198)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+199)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+198)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=2); i2 += 2) {
yPos = posBegin[1];
yPos = posBegin[1];
}
for (; (i2<=3); i2 += 1) {
yPos = posBegin[1];
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+30)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+31)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+30)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+174)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+175)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+174)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+78)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+79)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+78)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+54)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+55)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+54)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+150)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+151)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+150)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+6)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+7)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+6)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+102)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+103)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+102)] = 0.000000e+00;
}
}
}
}
}
if ((!neighbor_isValid[0][3])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S1083, S1092, S1086, S1089, S1088, S1091, S1085, S1090, S1093, S1087, S1084 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+12)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+13)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+12)] = 0.000000e+00;
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+60)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+61)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+60)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+204)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+205)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+204)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+132)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+133)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+132)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+84)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+85)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+84)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=2); i2 += 2) {
yPos = posEnd[1];
yPos = posEnd[1];
}
for (; (i2<=3); i2 += 1) {
yPos = posEnd[1];
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+36)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+37)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+36)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=2); i2 += 2) {
xPos = ((((i2-2)/1.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2-1)/1.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=3); i2 += 1) {
xPos = ((((i2-2)/1.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+180)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+181)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+180)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+156)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+157)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+156)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][0]);
int i2 = 2;
for (; (i2<=2); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+108)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+109)] = 0.000000e+00;
}
for (; (i2<=3); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[(i2+108)] = 0.000000e+00;
}
}
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
/* Statements in this Scop: S1094 */
for (int i0 = 0; (i0<=8); i0 += 1) {
double* buffer_Send_1_p1 = (&buffer_Send[1][(i0*2)]);
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][(i0*24)]);
int i1 = 1;
for (; (i1<=1); i1 += 2) {
buffer_Send_1_p1[(i1-1)] = fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+3)];
buffer_Send_1_p1[i1] = fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+9)];
}
for (; (i1<=2); i1 += 1) {
buffer_Send_1_p1[(i1-1)] = fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+3)];
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
MPI_Isend(buffer_Send[1], 18, MPI_DOUBLE, neighbor_remoteRank[0][1], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][1]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[1]);
reqOutstanding_Send[1] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
MPI_Irecv(buffer_Recv[0], 18, MPI_DOUBLE, neighbor_remoteRank[0][0], ((unsigned int)(neighbor_fragCommId[0][0]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Recv[0]) {
waitForMPIReq(&mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
/* Statements in this Scop: S1095 */
for (int i0 = 0; (i0<=8); i0 += 1) {
double* buffer_Recv_0_p1 = (&buffer_Recv[0][(i0*2)]);
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][(i0*24)]);
int i1 = 3;
for (; (i1<=3); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)-10)] = buffer_Recv_0_p1[(i1-3)];
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)-4)] = buffer_Recv_0_p1[(i1-2)];
}
for (; (i1<=4); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)-10)] = buffer_Recv_0_p1[(i1-3)];
}
}
}
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Send[1]) {
waitForMPIReq(&mpiRequest_Send[1]);
reqOutstanding_Send[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][3]&&neighbor_isRemote[0][3])) {
MPI_Isend(&fieldData_LaplaceCoeff_GMRF[0][14], 1, mpiDatatype_9_2_24, neighbor_remoteRank[0][3], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][3]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[3]);
reqOutstanding_Send[3] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][2]&&neighbor_isRemote[0][2])) {
MPI_Irecv(&fieldData_LaplaceCoeff_GMRF[0][8], 1, mpiDatatype_9_2_24, neighbor_remoteRank[0][2], ((unsigned int)(neighbor_fragCommId[0][2]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Recv[2]) {
waitForMPIReq(&mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Send[3]) {
waitForMPIReq(&mpiRequest_Send[3]);
reqOutstanding_Send[3] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
/* Statements in this Scop: S1096 */
for (int i0 = 0; (i0<=8); i0 += 1) {
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][(i0*24)]);
double* buffer_Send_0_p1 = (&buffer_Send[0][(i0*4)]);
int i1 = 0;
for (; (i1<=2); i1 += 2) {
buffer_Send_0_p1[i1] = fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+3)];
buffer_Send_0_p1[(i1+1)] = fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+9)];
}
for (; (i1<=3); i1 += 1) {
buffer_Send_0_p1[i1] = fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+3)];
}
}
}
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
/* Statements in this Scop: S1097 */
for (int i0 = 0; (i0<=8); i0 += 1) {
double* buffer_Send_1_p1 = (&buffer_Send[1][(i0*4)]);
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][(i0*24)]);
int i1 = 0;
for (; (i1<=2); i1 += 2) {
buffer_Send_1_p1[i1] = fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+2)];
buffer_Send_1_p1[(i1+1)] = fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+8)];
}
for (; (i1<=3); i1 += 1) {
buffer_Send_1_p1[i1] = fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+2)];
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
MPI_Isend(buffer_Send[0], 36, MPI_DOUBLE, neighbor_remoteRank[0][0], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][0]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[0]);
reqOutstanding_Send[0] = true;
}
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
MPI_Isend(buffer_Send[1], 36, MPI_DOUBLE, neighbor_remoteRank[0][1], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][1]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[1]);
reqOutstanding_Send[1] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
MPI_Irecv(buffer_Recv[0], 36, MPI_DOUBLE, neighbor_remoteRank[0][0], ((unsigned int)(neighbor_fragCommId[0][0]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = true;
}
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
MPI_Irecv(buffer_Recv[1], 36, MPI_DOUBLE, neighbor_remoteRank[0][1], ((unsigned int)(neighbor_fragCommId[0][1]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[1]);
reqOutstanding_Recv[1] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Recv[0]) {
waitForMPIReq(&mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = false;
}
if (reqOutstanding_Recv[1]) {
waitForMPIReq(&mpiRequest_Recv[1]);
reqOutstanding_Recv[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
/* Statements in this Scop: S1098 */
for (int i0 = 0; (i0<=8); i0 += 1) {
double* buffer_Recv_0_p1 = (&buffer_Recv[0][(i0*4)]);
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][(i0*24)]);
int i1 = 1;
for (; (i1<=3); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)-5)] = buffer_Recv_0_p1[(i1-1)];
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)+1)] = buffer_Recv_0_p1[i1];
}
for (; (i1<=4); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)-5)] = buffer_Recv_0_p1[(i1-1)];
}
}
}
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
/* Statements in this Scop: S1099 */
for (int i0 = 0; (i0<=8); i0 += 1) {
double* buffer_Recv_1_p1 = (&buffer_Recv[1][(i0*4)]);
double* fieldData_LaplaceCoeff_GMRF_0_p1 = (&fieldData_LaplaceCoeff_GMRF[0][(i0*24)]);
int i1 = 4;
for (; (i1<=6); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)-20)] = buffer_Recv_1_p1[(i1-4)];
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)-14)] = buffer_Recv_1_p1[(i1-3)];
}
for (; (i1<=7); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_0_p1[((i1*6)-20)] = buffer_Recv_1_p1[(i1-4)];
}
}
}
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Send[0]) {
waitForMPIReq(&mpiRequest_Send[0]);
reqOutstanding_Send[0] = false;
}
if (reqOutstanding_Send[1]) {
waitForMPIReq(&mpiRequest_Send[1]);
reqOutstanding_Send[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][2]&&neighbor_isRemote[0][2])) {
MPI_Isend(&fieldData_LaplaceCoeff_GMRF[0][13], 1, mpiDatatype_9_4_24, neighbor_remoteRank[0][2], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][2]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[2]);
reqOutstanding_Send[2] = true;
}
if ((neighbor_isValid[0][3]&&neighbor_isRemote[0][3])) {
MPI_Isend(&fieldData_LaplaceCoeff_GMRF[0][7], 1, mpiDatatype_9_4_24, neighbor_remoteRank[0][3], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][3]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[3]);
reqOutstanding_Send[3] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][2]&&neighbor_isRemote[0][2])) {
MPI_Irecv(&fieldData_LaplaceCoeff_GMRF[0][1], 1, mpiDatatype_9_4_24, neighbor_remoteRank[0][2], ((unsigned int)(neighbor_fragCommId[0][2]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = true;
}
if ((neighbor_isValid[0][3]&&neighbor_isRemote[0][3])) {
MPI_Irecv(&fieldData_LaplaceCoeff_GMRF[0][19], 1, mpiDatatype_9_4_24, neighbor_remoteRank[0][3], ((unsigned int)(neighbor_fragCommId[0][3]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[3]);
reqOutstanding_Recv[3] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Recv[2]) {
waitForMPIReq(&mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = false;
}
if (reqOutstanding_Recv[3]) {
waitForMPIReq(&mpiRequest_Recv[3]);
reqOutstanding_Recv[3] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Send[2]) {
waitForMPIReq(&mpiRequest_Send[2]);
reqOutstanding_Send[2] = false;
}
if (reqOutstanding_Send[3]) {
waitForMPIReq(&mpiRequest_Send[3]);
reqOutstanding_Send[3] = false;
}
}
}
}
void thread_ibs_num(size_t i, size_t n)
{
const size_t npack = nBlock >> 3;
const size_t npack2 = npack * 2;
C_UInt8 *Base = Geno1b.Get();
IdMatTri I = Array_Thread_MatIdx[i];
C_Int64 N = Array_Thread_MatCnt[i];
TS_KINGHomo *p = ptrKING + I.Offset();
for (; N > 0; N--, ++I, p++)
{
C_UInt8 *p1 = Base + I.Row() * npack2;
C_UInt8 *p2 = Base + I.Column() * npack2;
double *pAF = AF_1_AF.Get();
double *pAF2 = AF_1_AF_2.Get();
ssize_t m = npack;
#if defined(COREARRAY_SIMD_SSE2)
{
POPCNT_SSE2_HEAD
__m128i ibs0_sum, sumsq_sum;
ibs0_sum = sumsq_sum = _mm_setzero_si128();
__m128d sq_sum, sq_sum2;
sq_sum = sq_sum2 = _mm_setzero_pd();
for (; m > 0; m-=16)
{
__m128i g1_1 = _mm_load_si128((__m128i*)p1);
__m128i g1_2 = _mm_load_si128((__m128i*)(p1 + npack));
__m128i g2_1 = _mm_load_si128((__m128i*)p2);
__m128i g2_2 = _mm_load_si128((__m128i*)(p2 + npack));
p1 += 16; p2 += 16;
__m128i mask = (g1_1 | ~g1_2) & (g2_1 | ~g2_2);
__m128i ibs0 = (~((g1_1 ^ ~g2_1) | (g1_2 ^ ~g2_2))) & mask;
__m128i het = ((g1_1 ^ g1_2) ^ (g2_1 ^ g2_2)) & mask;
POPCNT_SSE2_RUN(ibs0)
ibs0_sum = _mm_add_epi32(ibs0_sum, ibs0);
POPCNT_SSE2_RUN(het)
sumsq_sum = _mm_add_epi32(_mm_add_epi32(sumsq_sum, het),
_mm_slli_epi32(ibs0, 2));
C_UInt64 m1 = _mm_cvtsi128_si64(mask);
C_UInt64 m2 = _mm_cvtsi128_si64(_mm_shuffle_epi32(mask,
_MM_SHUFFLE(1,0,3,2)));
for (size_t k=32; k > 0; k--)
{
switch (m1 & 0x03)
{
case 3:
sq_sum = _mm_add_pd(sq_sum, _mm_load_pd(pAF));
sq_sum2 = _mm_add_pd(sq_sum2, _mm_load_pd(pAF2));
break;
case 1:
sq_sum = _mm_add_pd(sq_sum, _mm_set_pd(0, pAF[0]));
sq_sum2 = _mm_add_pd(sq_sum2, _mm_set_pd(0, pAF2[0]));
break;
case 2:
sq_sum = _mm_add_pd(sq_sum, _mm_set_pd(pAF[1], 0));
sq_sum2 = _mm_add_pd(sq_sum2, _mm_set_pd(pAF2[1], 0));
break;
}
pAF += 2; pAF2 += 2; m1 >>= 2;
}
for (size_t k=32; k > 0; k--)
{
switch (m2 & 0x03)
{
case 3:
sq_sum = _mm_add_pd(sq_sum, _mm_load_pd(pAF));
sq_sum2 = _mm_add_pd(sq_sum2, _mm_load_pd(pAF2));
break;
case 1:
sq_sum = _mm_add_pd(sq_sum, _mm_set_pd(0, pAF[0]));
sq_sum2 = _mm_add_pd(sq_sum2, _mm_set_pd(0, pAF2[0]));
break;
case 2:
sq_sum = _mm_add_pd(sq_sum, _mm_set_pd(pAF[1], 0));
sq_sum2 = _mm_add_pd(sq_sum2, _mm_set_pd(pAF2[1], 0));
break;
}
pAF += 2; pAF2 += 2; m2 >>= 2;
}
}
p->IBS0 += vec_sum_i32(ibs0_sum);
p->SumSq += vec_sum_i32(sumsq_sum);
p->SumAFreq += vec_sum_f64(sq_sum);
p->SumAFreq2 += vec_sum_f64(sq_sum2);
}
#else
for (; m > 0; m-=8)
{
C_UInt64 g1_1 = *((C_UInt64*)p1);
C_UInt64 g1_2 = *((C_UInt64*)(p1 + npack));
C_UInt64 g2_1 = *((C_UInt64*)p2);
C_UInt64 g2_2 = *((C_UInt64*)(p2 + npack));
p1 += 8; p2 += 8;
C_UInt64 mask = (g1_1 | ~g1_2) & (g2_1 | ~g2_2);
C_UInt64 ibs0 = (~((g1_1 ^ ~g2_1) | (g1_2 ^ ~g2_2))) & mask;
C_UInt64 het = ((g1_1 ^ g1_2) ^ (g2_1 ^ g2_2)) & mask;
p->IBS0 += POPCNT_U64(ibs0);
p->SumSq += POPCNT_U64(het) + POPCNT_U64(ibs0)*4;
double sum=0, sum2=0;
for (size_t k=64; k > 0; k--)
{
if (mask & 0x01)
{ sum += (*pAF); sum2 += (*pAF2); }
pAF ++; pAF2 ++;
mask >>= 1;
}
p->SumAFreq += sum;
p->SumAFreq2 += sum2;
}
#endif
}
}
void exchsolution_gmrfData_1(unsigned int slot) {
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((!neighbor_isValid[0][0])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S493, S492, S494 */
{
{
{
double* fieldData_Solution_GMRF_1_p1 = (&fieldData_Solution_GMRF[1][0]);
int i1 = 1;
for (; (i1<=2); i1 += 2) {
fieldData_Solution_GMRF_1_p1[((i1*6)+2)] = 0.000000e+00;
fieldData_Solution_GMRF_1_p1[((i1*6)+8)] = 0.000000e+00;
}
for (; (i1<=3); i1 += 1) {
fieldData_Solution_GMRF_1_p1[((i1*6)+2)] = 0.000000e+00;
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
xPos = posBegin[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<1); i1 += 4) {
/* xPos = posBegin[0]; */
__m128d vec0 = _mm_load1_pd((&posBegin[0]));
__m128d vec0_2 = _mm_load1_pd((&posBegin[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<4); i1 += 1) {
xPos = posBegin[0];
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
yPos = ((((i1-1)/2.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(1.000000e+00);
__m128d vec2 = _mm_set1_pd(2.000000e+00);
__m128d vec5 = _mm_set1_pd(yPos);
for (; (i1<1); i1 += 4) {
/* yPos = ((((i1-1)/2.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec3 = _mm_load1_pd((&posEnd[1]));
__m128d vec3_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec4 = _mm_load1_pd((&posBegin[1]));
__m128d vec4_2 = _mm_load1_pd((&posBegin[1]));
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0, vec1), vec2), _mm_sub_pd(vec3, vec4)), vec4);
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0_2, vec1), vec2), _mm_sub_pd(vec3_2, vec4_2)), vec4_2);
}
for (; (i1<4); i1 += 1) {
yPos = ((((i1-1)/2.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
}
}
}
if ((!neighbor_isValid[0][1])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S496, S495, S497 */
{
{
{
double* fieldData_Solution_GMRF_1_p1 = (&fieldData_Solution_GMRF[1][0]);
int i1 = 1;
for (; (i1<=2); i1 += 2) {
fieldData_Solution_GMRF_1_p1[((i1*6)+4)] = 0.000000e+00;
fieldData_Solution_GMRF_1_p1[((i1*6)+10)] = 0.000000e+00;
}
for (; (i1<=3); i1 += 1) {
fieldData_Solution_GMRF_1_p1[((i1*6)+4)] = 0.000000e+00;
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
xPos = posEnd[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<1); i1 += 4) {
/* xPos = posEnd[0]; */
__m128d vec0 = _mm_load1_pd((&posEnd[0]));
__m128d vec0_2 = _mm_load1_pd((&posEnd[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<4); i1 += 1) {
xPos = posEnd[0];
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
yPos = ((((i1-1)/2.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(1.000000e+00);
__m128d vec2 = _mm_set1_pd(2.000000e+00);
__m128d vec5 = _mm_set1_pd(yPos);
for (; (i1<1); i1 += 4) {
/* yPos = ((((i1-1)/2.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec3 = _mm_load1_pd((&posEnd[1]));
__m128d vec3_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec4 = _mm_load1_pd((&posBegin[1]));
__m128d vec4_2 = _mm_load1_pd((&posBegin[1]));
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0, vec1), vec2), _mm_sub_pd(vec3, vec4)), vec4);
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0_2, vec1), vec2), _mm_sub_pd(vec3_2, vec4_2)), vec4_2);
}
for (; (i1<4); i1 += 1) {
yPos = ((((i1-1)/2.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
}
}
}
if ((!neighbor_isValid[0][2])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S500, S499, S498 */
{
{
{
int i2 = 2;
for (; (i2<=3); i2 += 2) {
xPos = ((((i2-2)/2.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2-1)/2.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=4); i2 += 1) {
xPos = ((((i2-2)/2.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
{
double* fieldData_Solution_GMRF_1_p1 = (&fieldData_Solution_GMRF[1][0]);
int i2 = 2;
for (; (i2<=3); i2 += 2) {
fieldData_Solution_GMRF_1_p1[(i2+6)] = 0.000000e+00;
fieldData_Solution_GMRF_1_p1[(i2+7)] = 0.000000e+00;
}
for (; (i2<=4); i2 += 1) {
fieldData_Solution_GMRF_1_p1[(i2+6)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=3); i2 += 2) {
yPos = posBegin[1];
yPos = posBegin[1];
}
for (; (i2<=4); i2 += 1) {
yPos = posBegin[1];
}
}
}
}
}
if ((!neighbor_isValid[0][3])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S503, S502, S501 */
{
{
{
int i2 = 2;
for (; (i2<=3); i2 += 2) {
xPos = ((((i2-2)/2.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2-1)/2.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=4); i2 += 1) {
xPos = ((((i2-2)/2.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
{
int i2 = 2;
for (; (i2<=3); i2 += 2) {
yPos = posEnd[1];
yPos = posEnd[1];
}
for (; (i2<=4); i2 += 1) {
yPos = posEnd[1];
}
}
}
{
double* fieldData_Solution_GMRF_1_p1 = (&fieldData_Solution_GMRF[1][0]);
int i2 = 2;
for (; (i2<=3); i2 += 2) {
fieldData_Solution_GMRF_1_p1[(i2+18)] = 0.000000e+00;
fieldData_Solution_GMRF_1_p1[(i2+19)] = 0.000000e+00;
}
for (; (i2<=4); i2 += 1) {
fieldData_Solution_GMRF_1_p1[(i2+18)] = 0.000000e+00;
}
}
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
MPI_Isend(&fieldData_Solution_GMRF[1][10], 1, mpiDatatype_3_1_6, neighbor_remoteRank[0][1], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][1]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[1]);
reqOutstanding_Send[1] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
MPI_Irecv(&fieldData_Solution_GMRF[1][8], 1, mpiDatatype_3_1_6, neighbor_remoteRank[0][0], ((unsigned int)(neighbor_fragCommId[0][0]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Recv[0]) {
waitForMPIReq(&mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Send[1]) {
waitForMPIReq(&mpiRequest_Send[1]);
reqOutstanding_Send[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][3]&&neighbor_isRemote[0][3])) {
MPI_Isend(&fieldData_Solution_GMRF[1][20], 1, mpiDatatype_1_3_6, neighbor_remoteRank[0][3], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][3]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[3]);
reqOutstanding_Send[3] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][2]&&neighbor_isRemote[0][2])) {
MPI_Irecv(&fieldData_Solution_GMRF[1][8], 1, mpiDatatype_1_3_6, neighbor_remoteRank[0][2], ((unsigned int)(neighbor_fragCommId[0][2]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Recv[2]) {
waitForMPIReq(&mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Send[3]) {
waitForMPIReq(&mpiRequest_Send[3]);
reqOutstanding_Send[3] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
MPI_Isend(&fieldData_Solution_GMRF[1][3], 1, mpiDatatype_5_1_6, neighbor_remoteRank[0][0], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][0]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[0]);
reqOutstanding_Send[0] = true;
}
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
MPI_Isend(&fieldData_Solution_GMRF[1][3], 1, mpiDatatype_5_1_6, neighbor_remoteRank[0][1], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][1]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[1]);
reqOutstanding_Send[1] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
MPI_Irecv(&fieldData_Solution_GMRF[1][1], 1, mpiDatatype_5_1_6, neighbor_remoteRank[0][0], ((unsigned int)(neighbor_fragCommId[0][0]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = true;
}
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
MPI_Irecv(&fieldData_Solution_GMRF[1][5], 1, mpiDatatype_5_1_6, neighbor_remoteRank[0][1], ((unsigned int)(neighbor_fragCommId[0][1]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[1]);
reqOutstanding_Recv[1] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Recv[0]) {
waitForMPIReq(&mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = false;
}
if (reqOutstanding_Recv[1]) {
waitForMPIReq(&mpiRequest_Recv[1]);
reqOutstanding_Recv[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Send[0]) {
waitForMPIReq(&mpiRequest_Send[0]);
reqOutstanding_Send[0] = false;
}
if (reqOutstanding_Send[1]) {
waitForMPIReq(&mpiRequest_Send[1]);
reqOutstanding_Send[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][2]&&neighbor_isRemote[0][2])) {
MPI_Isend(&fieldData_Solution_GMRF[1][13], 1, mpiDatatype_1_5_6, neighbor_remoteRank[0][2], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][2]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[2]);
reqOutstanding_Send[2] = true;
}
if ((neighbor_isValid[0][3]&&neighbor_isRemote[0][3])) {
MPI_Isend(&fieldData_Solution_GMRF[1][13], 1, mpiDatatype_1_5_6, neighbor_remoteRank[0][3], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][3]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[3]);
reqOutstanding_Send[3] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][2]&&neighbor_isRemote[0][2])) {
MPI_Irecv(&fieldData_Solution_GMRF[1][1], 1, mpiDatatype_1_5_6, neighbor_remoteRank[0][2], ((unsigned int)(neighbor_fragCommId[0][2]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = true;
}
if ((neighbor_isValid[0][3]&&neighbor_isRemote[0][3])) {
MPI_Irecv(&fieldData_Solution_GMRF[1][25], 1, mpiDatatype_1_5_6, neighbor_remoteRank[0][3], ((unsigned int)(neighbor_fragCommId[0][3]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[3]);
reqOutstanding_Recv[3] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Recv[2]) {
waitForMPIReq(&mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = false;
}
if (reqOutstanding_Recv[3]) {
waitForMPIReq(&mpiRequest_Recv[3]);
reqOutstanding_Recv[3] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Send[2]) {
waitForMPIReq(&mpiRequest_Send[2]);
reqOutstanding_Send[2] = false;
}
if (reqOutstanding_Send[3]) {
waitForMPIReq(&mpiRequest_Send[3]);
reqOutstanding_Send[3] = false;
}
}
}
}
// =============================================================================
//
// sse2_vChirpData
// version by: Alex Kan - SSE2 mods (haddsum removal) BH
// http://tbp.berkeley.edu/~alexkan/seti/
//
int sse2_ChirpData_ak(
sah_complex * cx_DataArray,
sah_complex * cx_ChirpDataArray,
int chirp_rate_ind,
double chirp_rate,
int ul_NumDataPoints,
double sample_rate
) {
int i;
if (chirp_rate_ind == 0) {
memcpy(cx_ChirpDataArray, cx_DataArray, (int)ul_NumDataPoints * sizeof(sah_complex) );
return 0;
}
int vEnd;
double srate = chirp_rate * 0.5 / (sample_rate * sample_rate);
__m128d rate = _mm_set1_pd(chirp_rate * 0.5 / (sample_rate * sample_rate));
__m128d roundVal = _mm_set1_pd(srate >= 0.0 ? TWO_TO_52 : -TWO_TO_52);
// main vectorised loop
vEnd = ul_NumDataPoints - (ul_NumDataPoints & 3);
for (i = 0; i < vEnd; i += 4) {
const float *data = (const float *) (cx_DataArray + i);
float *chirped = (float *) (cx_ChirpDataArray + i);
__m128d di = _mm_set1_pd(i);
__m128d a1 = _mm_add_pd(_mm_set_pd(1.0, 0.0), di);
__m128d a2 = _mm_add_pd(_mm_set_pd(3.0, 2.0), di);
__m128d x1, y1;
__m128 d1, d2;
__m128 cd1, cd2;
__m128 td1, td2;
__m128 x;
__m128 y;
__m128 s;
__m128 c;
__m128 m;
// load the signal to be chirped
prefetchnta((const void *)( data+32 ));
d1 = _mm_load_ps(data);
d2 = _mm_load_ps(data+4);
// calculate the input angle
a1 = _mm_mul_pd(a1, a1);
a2 = _mm_mul_pd(a2, a2);
a1 = _mm_mul_pd(a1, rate);
a2 = _mm_mul_pd(a2, rate);
// reduce the angle to the range (-0.5, 0.5)
x1 = _mm_add_pd(a1, roundVal);
y1 = _mm_add_pd(a2, roundVal);
x1 = _mm_sub_pd(x1, roundVal);
y1 = _mm_sub_pd(y1, roundVal);
a1 = _mm_sub_pd(a1, x1);
a2 = _mm_sub_pd(a2, y1);
// convert pair of packed double into packed single
x = _mm_movelh_ps(_mm_cvtpd_ps(a1), _mm_cvtpd_ps(a2));
// square to the range [0, 0.25)
y = _mm_mul_ps(x, x);
// perform the initial polynomial approximations
s = _mm_mul_ps(y, SS4);
c = _mm_mul_ps(y, CC3);
s = _mm_add_ps(s, SS3);
c = _mm_add_ps(c, CC2);
s = _mm_mul_ps(s, y);
c = _mm_mul_ps(c, y);
s = _mm_add_ps(s, SS2);
c = _mm_add_ps(c, CC1);
s = _mm_mul_ps(s, y);
c = _mm_mul_ps(c, y);
s = _mm_add_ps(s, SS1);
s = _mm_mul_ps(s, x);
c = _mm_add_ps(c, ONE);
// perform first angle doubling
x = _mm_sub_ps(_mm_mul_ps(c, c), _mm_mul_ps(s, s));
y = _mm_mul_ps(_mm_mul_ps(s, c), TWO);
// calculate scaling factor to correct the magnitude
// m1 = vec_nmsub(y1, y1, vec_nmsub(x1, x1, TWO));
// m2 = vec_nmsub(y2, y2, vec_nmsub(x2, x2, TWO));
m = vec_recip2(_mm_add_ps(_mm_mul_ps(x, x), _mm_mul_ps(y, y)));
// perform second angle doubling
c = _mm_sub_ps(_mm_mul_ps(x, x), _mm_mul_ps(y, y));
s = _mm_mul_ps(_mm_mul_ps(y, x), TWO);
// correct the magnitude (final sine / cosine approximations)
c = _mm_mul_ps(c, m);
s = _mm_mul_ps(s, m);
/* c1 c2 c3 c4
s1 s2 s3 s4
R1 i1 R2 I2 R3 i3 R4 i4
R1 * c1 + (i1 * s1 * -1)
i1 * c1 + R1 * s1
R2 * c2 + (i2 * s2 * -1)
i2 * c2 + R2 * s2
*/
x = d1;
y = d2;
x = _mm_shuffle_ps(x, x, 0xB1);
y = _mm_shuffle_ps(y, y, 0xB1);
x = _mm_mul_ps(x, R_NEG);
y = _mm_mul_ps(y, R_NEG);
cd1 = _mm_shuffle_ps(c, c, 0x50); // 01 01 00 00 AaBb => BBbb => c3c3c4c4
cd2 = _mm_shuffle_ps(c, c, 0xfa); // 11 11 10 10 AaBb => AAaa => c1c1c2c2
td1 = _mm_shuffle_ps(s, s, 0x50);
td2 = _mm_shuffle_ps(s, s, 0xfa);
cd1 = _mm_mul_ps(cd1, d1);
cd2 = _mm_mul_ps(cd2, d2);
td1 = _mm_mul_ps(td1, x);
td2 = _mm_mul_ps(td2, y);
cd1 = _mm_add_ps(cd1, td1);
cd2 = _mm_add_ps(cd2, td2);
// store chirped values
_mm_stream_ps(chirped+0, cd1);
_mm_stream_ps(chirped+4, cd2);
}
_mm_sfence();
if( i < ul_NumDataPoints) {
// use original routine to finish up any tailings (max stride-1 elements)
v_ChirpData(cx_DataArray+i, cx_ChirpDataArray+i
, chirp_rate_ind, chirp_rate, ul_NumDataPoints-i, sample_rate);
}
analysis_state.FLOP_counter+=12.0*ul_NumDataPoints;
return 0;
}
void exchsolutionData_2(unsigned int slot) {
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((!neighbor_isValid[1][0])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S397, S396, S398 */
{
{
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
yPos = ((((i1-1)/4.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(1.000000e+00);
__m128d vec2 = _mm_set1_pd(4.000000e+00);
__m128d vec5 = _mm_set1_pd(yPos);
for (; (i1<3); i1 += 4) {
/* yPos = ((((i1-1)/4.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec3 = _mm_load1_pd((&posEnd[1]));
__m128d vec3_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec4 = _mm_load1_pd((&posBegin[1]));
__m128d vec4_2 = _mm_load1_pd((&posBegin[1]));
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0, vec1), vec2), _mm_sub_pd(vec3, vec4)), vec4);
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0_2, vec1), vec2), _mm_sub_pd(vec3_2, vec4_2)), vec4_2);
}
for (; (i1<6); i1 += 1) {
yPos = ((((i1-1)/4.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
{
double* fieldData_Solution_2_p1 = (&fieldData_Solution[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_Solution_2_p1[((i1*8)+2)] = 0.000000e+00;
fieldData_Solution_2_p1[((i1*8)+10)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_Solution_2_p1[((i1*8)+2)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
xPos = posBegin[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<3); i1 += 4) {
/* xPos = posBegin[0]; */
__m128d vec0 = _mm_load1_pd((&posBegin[0]));
__m128d vec0_2 = _mm_load1_pd((&posBegin[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<6); i1 += 1) {
xPos = posBegin[0];
}
}
}
}
}
if ((!neighbor_isValid[1][1])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S401, S400, S399 */
{
{
{
double* fieldData_Solution_2_p1 = (&fieldData_Solution[2][0]);
int i1 = 1;
for (; (i1<=4); i1 += 2) {
fieldData_Solution_2_p1[((i1*8)+6)] = 0.000000e+00;
fieldData_Solution_2_p1[((i1*8)+14)] = 0.000000e+00;
}
for (; (i1<=5); i1 += 1) {
fieldData_Solution_2_p1[((i1*8)+6)] = 0.000000e+00;
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
xPos = posEnd[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<3); i1 += 4) {
/* xPos = posEnd[0]; */
__m128d vec0 = _mm_load1_pd((&posEnd[0]));
__m128d vec0_2 = _mm_load1_pd((&posEnd[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<6); i1 += 1) {
xPos = posEnd[0];
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
yPos = ((((i1-1)/4.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(1.000000e+00);
__m128d vec2 = _mm_set1_pd(4.000000e+00);
__m128d vec5 = _mm_set1_pd(yPos);
for (; (i1<3); i1 += 4) {
/* yPos = ((((i1-1)/4.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec3 = _mm_load1_pd((&posEnd[1]));
__m128d vec3_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec4 = _mm_load1_pd((&posBegin[1]));
__m128d vec4_2 = _mm_load1_pd((&posBegin[1]));
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0, vec1), vec2), _mm_sub_pd(vec3, vec4)), vec4);
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0_2, vec1), vec2), _mm_sub_pd(vec3_2, vec4_2)), vec4_2);
}
for (; (i1<6); i1 += 1) {
yPos = ((((i1-1)/4.000000e+00)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
}
}
}
if ((!neighbor_isValid[1][2])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S404, S403, S402 */
{
{
{
double* fieldData_Solution_2_p1 = (&fieldData_Solution[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_Solution_2_p1[(i2+8)] = 0.000000e+00;
fieldData_Solution_2_p1[(i2+9)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_Solution_2_p1[(i2+8)] = 0.000000e+00;
}
}
{
int i2 = 2;
for (; (i2<=5); i2 += 2) {
xPos = ((((i2-2)/4.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2-1)/4.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=6); i2 += 1) {
xPos = ((((i2-2)/4.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
}
{
int i2 = 2;
for (; (i2<=5); i2 += 2) {
yPos = posBegin[1];
yPos = posBegin[1];
}
for (; (i2<=6); i2 += 1) {
yPos = posBegin[1];
}
}
}
}
}
if ((!neighbor_isValid[1][3])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S407, S406, S405 */
{
{
{
double* fieldData_Solution_2_p1 = (&fieldData_Solution[2][0]);
int i2 = 2;
for (; (i2<=5); i2 += 2) {
fieldData_Solution_2_p1[(i2+40)] = 0.000000e+00;
fieldData_Solution_2_p1[(i2+41)] = 0.000000e+00;
}
for (; (i2<=6); i2 += 1) {
fieldData_Solution_2_p1[(i2+40)] = 0.000000e+00;
}
}
{
int i2 = 2;
for (; (i2<=5); i2 += 2) {
xPos = ((((i2-2)/4.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2-1)/4.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=6); i2 += 1) {
xPos = ((((i2-2)/4.000000e+00)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
}
{
int i2 = 2;
for (; (i2<=5); i2 += 2) {
yPos = posEnd[1];
yPos = posEnd[1];
}
for (; (i2<=6); i2 += 1) {
yPos = posEnd[1];
}
}
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
MPI_Isend(&fieldData_Solution[2][14], 1, mpiDatatype_5_1_8, neighbor_remoteRank[1][1], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][1]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[1]);
reqOutstanding_Send[1] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
MPI_Irecv(&fieldData_Solution[2][10], 1, mpiDatatype_5_1_8, neighbor_remoteRank[1][0], ((unsigned int)(neighbor_fragCommId[1][0]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[0]) {
waitForMPIReq(&mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[1]) {
waitForMPIReq(&mpiRequest_Send[1]);
reqOutstanding_Send[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][3]&&neighbor_isRemote[1][3])) {
MPI_Isend(&fieldData_Solution[2][42], 1, mpiDatatype_1_5_8, neighbor_remoteRank[1][3], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][3]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[3]);
reqOutstanding_Send[3] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][2]&&neighbor_isRemote[1][2])) {
MPI_Irecv(&fieldData_Solution[2][10], 1, mpiDatatype_1_5_8, neighbor_remoteRank[1][2], ((unsigned int)(neighbor_fragCommId[1][2]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[2]) {
waitForMPIReq(&mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[3]) {
waitForMPIReq(&mpiRequest_Send[3]);
reqOutstanding_Send[3] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
MPI_Isend(&fieldData_Solution[2][3], 1, mpiDatatype_7_1_8, neighbor_remoteRank[1][0], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][0]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[0]);
reqOutstanding_Send[0] = true;
}
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
MPI_Isend(&fieldData_Solution[2][5], 1, mpiDatatype_7_1_8, neighbor_remoteRank[1][1], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][1]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[1]);
reqOutstanding_Send[1] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
MPI_Irecv(&fieldData_Solution[2][1], 1, mpiDatatype_7_1_8, neighbor_remoteRank[1][0], ((unsigned int)(neighbor_fragCommId[1][0]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = true;
}
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
MPI_Irecv(&fieldData_Solution[2][7], 1, mpiDatatype_7_1_8, neighbor_remoteRank[1][1], ((unsigned int)(neighbor_fragCommId[1][1]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[1]);
reqOutstanding_Recv[1] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[0]) {
waitForMPIReq(&mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = false;
}
if (reqOutstanding_Recv[1]) {
waitForMPIReq(&mpiRequest_Recv[1]);
reqOutstanding_Recv[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[0]) {
waitForMPIReq(&mpiRequest_Send[0]);
reqOutstanding_Send[0] = false;
}
if (reqOutstanding_Send[1]) {
waitForMPIReq(&mpiRequest_Send[1]);
reqOutstanding_Send[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][2]&&neighbor_isRemote[1][2])) {
MPI_Isend(&fieldData_Solution[2][17], 1, mpiDatatype_1_7_8, neighbor_remoteRank[1][2], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][2]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[2]);
reqOutstanding_Send[2] = true;
}
if ((neighbor_isValid[1][3]&&neighbor_isRemote[1][3])) {
MPI_Isend(&fieldData_Solution[2][33], 1, mpiDatatype_1_7_8, neighbor_remoteRank[1][3], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][3]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[3]);
reqOutstanding_Send[3] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][2]&&neighbor_isRemote[1][2])) {
MPI_Irecv(&fieldData_Solution[2][1], 1, mpiDatatype_1_7_8, neighbor_remoteRank[1][2], ((unsigned int)(neighbor_fragCommId[1][2]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = true;
}
if ((neighbor_isValid[1][3]&&neighbor_isRemote[1][3])) {
MPI_Irecv(&fieldData_Solution[2][49], 1, mpiDatatype_1_7_8, neighbor_remoteRank[1][3], ((unsigned int)(neighbor_fragCommId[1][3]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[3]);
reqOutstanding_Recv[3] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[2]) {
waitForMPIReq(&mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = false;
}
if (reqOutstanding_Recv[3]) {
waitForMPIReq(&mpiRequest_Recv[3]);
reqOutstanding_Recv[3] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[2]) {
waitForMPIReq(&mpiRequest_Send[2]);
reqOutstanding_Send[2] = false;
}
if (reqOutstanding_Send[3]) {
waitForMPIReq(&mpiRequest_Send[3]);
reqOutstanding_Send[3] = false;
}
}
}
}
void exchlaplacecoeff_gmrfData_5(unsigned int slot) {
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((!neighbor_isValid[0][0])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S1306, S1309, S1300, S1308, S1302, S1305, S1310, S1304, S1307, S1301, S1303 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+6302)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+6338)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+6302)] = 0.000000e+00;
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+3782)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+3818)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+3782)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+7562)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+7598)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+7562)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+2)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+38)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+2)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+2522)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+2558)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+2522)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
xPos = posBegin[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<31); i1 += 4) {
/* xPos = posBegin[0]; */
__m128d vec0 = _mm_load1_pd((&posBegin[0]));
__m128d vec0_2 = _mm_load1_pd((&posBegin[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<34); i1 += 1) {
xPos = posBegin[0];
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+8822)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+8858)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+8822)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+1262)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+1298)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+1262)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
yPos = ((((i1-1)/3.200000e+01)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(1.000000e+00);
__m128d vec2 = _mm_set1_pd(3.200000e+01);
__m128d vec5 = _mm_set1_pd(yPos);
for (; (i1<31); i1 += 4) {
/* yPos = ((((i1-1)/3.200000e+01)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec3 = _mm_load1_pd((&posEnd[1]));
__m128d vec3_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec4 = _mm_load1_pd((&posBegin[1]));
__m128d vec4_2 = _mm_load1_pd((&posBegin[1]));
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0, vec1), vec2), _mm_sub_pd(vec3, vec4)), vec4);
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0_2, vec1), vec2), _mm_sub_pd(vec3_2, vec4_2)), vec4_2);
}
for (; (i1<34); i1 += 1) {
yPos = ((((i1-1)/3.200000e+01)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+5042)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+5078)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+5042)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+10082)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+10118)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+10082)] = 0.000000e+00;
}
}
}
}
}
if ((!neighbor_isValid[0][1])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S1312, S1320, S1314, S1317, S1311, S1319, S1313, S1316, S1321, S1315, S1318 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+1294)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+1330)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+1294)] = 0.000000e+00;
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+5074)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+5110)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+5074)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+8854)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+8890)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+8854)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+2554)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+2590)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+2554)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
yPos = ((((i1-1)/3.200000e+01)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(1.000000e+00);
__m128d vec2 = _mm_set1_pd(3.200000e+01);
__m128d vec5 = _mm_set1_pd(yPos);
for (; (i1<31); i1 += 4) {
/* yPos = ((((i1-1)/3.200000e+01)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec3 = _mm_load1_pd((&posEnd[1]));
__m128d vec3_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec4 = _mm_load1_pd((&posBegin[1]));
__m128d vec4_2 = _mm_load1_pd((&posBegin[1]));
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0, vec1), vec2), _mm_sub_pd(vec3, vec4)), vec4);
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0_2, vec1), vec2), _mm_sub_pd(vec3_2, vec4_2)), vec4_2);
}
for (; (i1<34); i1 += 1) {
yPos = ((((i1-1)/3.200000e+01)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+34)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+70)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+34)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+3814)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+3850)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+3814)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+7594)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+7630)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+7594)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+6334)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+6370)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+6334)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
xPos = posEnd[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<31); i1 += 4) {
/* xPos = posEnd[0]; */
__m128d vec0 = _mm_load1_pd((&posEnd[0]));
__m128d vec0_2 = _mm_load1_pd((&posEnd[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<34); i1 += 1) {
xPos = posEnd[0];
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i1 = 1;
for (; (i1<=32); i1 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+10114)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+10150)] = 0.000000e+00;
}
for (; (i1<=33); i1 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i1*36)+10114)] = 0.000000e+00;
}
}
}
}
}
if ((!neighbor_isValid[0][2])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S1327, S1332, S1326, S1329, S1323, S1322, S1331, S1325, S1328, S1330, S1324 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+5076)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+5077)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+5076)] = 0.000000e+00;
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+8856)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+8857)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+8856)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+6336)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+6337)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+6336)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+7596)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+7597)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+7596)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=33); i2 += 2) {
xPos = ((((i2-2)/3.200000e+01)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2-1)/3.200000e+01)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=34); i2 += 1) {
xPos = ((((i2-2)/3.200000e+01)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+36)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+37)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+36)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+2556)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+2557)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+2556)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+10116)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+10117)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+10116)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+3816)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+3817)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+3816)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+1296)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+1297)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+1296)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=33); i2 += 2) {
yPos = posBegin[1];
yPos = posBegin[1];
}
for (; (i2<=34); i2 += 1) {
yPos = posBegin[1];
}
}
}
}
}
if ((!neighbor_isValid[0][3])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S1338, S1341, S1335, S1340, S1343, S1337, S1334, S1333, S1342, S1336, S1339 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+7488)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+7489)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+7488)] = 0.000000e+00;
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+2448)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+2449)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+2448)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+11268)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+11269)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+11268)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=33); i2 += 2) {
yPos = posEnd[1];
yPos = posEnd[1];
}
for (; (i2<=34); i2 += 1) {
yPos = posEnd[1];
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+10008)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+10009)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+10008)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+6228)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+6229)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+6228)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+8748)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+8749)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+8748)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=33); i2 += 2) {
xPos = ((((i2-2)/3.200000e+01)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2-1)/3.200000e+01)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=34); i2 += 1) {
xPos = ((((i2-2)/3.200000e+01)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+3708)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+3709)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+3708)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+1188)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+1189)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+1188)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][0]);
int i2 = 2;
for (; (i2<=33); i2 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+4968)] = 0.000000e+00;
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+4969)] = 0.000000e+00;
}
for (; (i2<=34); i2 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[(i2+4968)] = 0.000000e+00;
}
}
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
/* Statements in this Scop: S1344 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][(i3*1260)]);
double* buffer_Send_1_p1 = (&buffer_Send[1][(i3*33)]);
int i4 = 1;
for (; (i4<=32); i4 += 2) {
buffer_Send_1_p1[(i4-1)] = fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)+34)];
buffer_Send_1_p1[i4] = fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)+70)];
}
for (; (i4<=33); i4 += 1) {
buffer_Send_1_p1[(i4-1)] = fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)+34)];
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
MPI_Isend(buffer_Send[1], 297, MPI_DOUBLE, neighbor_remoteRank[0][1], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][1]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[1]);
reqOutstanding_Send[1] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
MPI_Irecv(buffer_Recv[0], 297, MPI_DOUBLE, neighbor_remoteRank[0][0], ((unsigned int)(neighbor_fragCommId[0][0]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Recv[0]) {
waitForMPIReq(&mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
/* Statements in this Scop: S1345 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][(i3*1260)]);
double* buffer_Recv_0_p1 = (&buffer_Recv[0][(i3*33)]);
int i4 = 3;
for (; (i4<=34); i4 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)-70)] = buffer_Recv_0_p1[(i4-3)];
fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)-34)] = buffer_Recv_0_p1[(i4-2)];
}
for (; (i4<=35); i4 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)-70)] = buffer_Recv_0_p1[(i4-3)];
}
}
}
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Send[1]) {
waitForMPIReq(&mpiRequest_Send[1]);
reqOutstanding_Send[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][3]&&neighbor_isRemote[0][3])) {
MPI_Isend(&fieldData_LaplaceCoeff_GMRF[5][1190], 1, mpiDatatype_9_33_1260, neighbor_remoteRank[0][3], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][3]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[3]);
reqOutstanding_Send[3] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][2]&&neighbor_isRemote[0][2])) {
MPI_Irecv(&fieldData_LaplaceCoeff_GMRF[5][38], 1, mpiDatatype_9_33_1260, neighbor_remoteRank[0][2], ((unsigned int)(neighbor_fragCommId[0][2]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Recv[2]) {
waitForMPIReq(&mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Send[3]) {
waitForMPIReq(&mpiRequest_Send[3]);
reqOutstanding_Send[3] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
/* Statements in this Scop: S1346 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][(i3*1260)]);
double* buffer_Send_0_p1 = (&buffer_Send[0][(i3*35)]);
int i4 = 0;
for (; (i4<=33); i4 += 2) {
buffer_Send_0_p1[i4] = fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)+3)];
buffer_Send_0_p1[(i4+1)] = fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)+39)];
}
for (; (i4<=34); i4 += 1) {
buffer_Send_0_p1[i4] = fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)+3)];
}
}
}
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
/* Statements in this Scop: S1347 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][(i3*1260)]);
double* buffer_Send_1_p1 = (&buffer_Send[1][(i3*35)]);
int i4 = 0;
for (; (i4<=33); i4 += 2) {
buffer_Send_1_p1[i4] = fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)+33)];
buffer_Send_1_p1[(i4+1)] = fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)+69)];
}
for (; (i4<=34); i4 += 1) {
buffer_Send_1_p1[i4] = fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)+33)];
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
MPI_Isend(buffer_Send[0], 315, MPI_DOUBLE, neighbor_remoteRank[0][0], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][0]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[0]);
reqOutstanding_Send[0] = true;
}
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
MPI_Isend(buffer_Send[1], 315, MPI_DOUBLE, neighbor_remoteRank[0][1], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][1]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[1]);
reqOutstanding_Send[1] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
MPI_Irecv(buffer_Recv[0], 315, MPI_DOUBLE, neighbor_remoteRank[0][0], ((unsigned int)(neighbor_fragCommId[0][0]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = true;
}
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
MPI_Irecv(buffer_Recv[1], 315, MPI_DOUBLE, neighbor_remoteRank[0][1], ((unsigned int)(neighbor_fragCommId[0][1]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[1]);
reqOutstanding_Recv[1] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Recv[0]) {
waitForMPIReq(&mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = false;
}
if (reqOutstanding_Recv[1]) {
waitForMPIReq(&mpiRequest_Recv[1]);
reqOutstanding_Recv[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][0]&&neighbor_isRemote[0][0])) {
/* Statements in this Scop: S1348 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][(i3*1260)]);
double* buffer_Recv_0_p1 = (&buffer_Recv[0][(i3*35)]);
int i4 = 1;
for (; (i4<=34); i4 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)-35)] = buffer_Recv_0_p1[(i4-1)];
fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)+1)] = buffer_Recv_0_p1[i4];
}
for (; (i4<=35); i4 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)-35)] = buffer_Recv_0_p1[(i4-1)];
}
}
}
if ((neighbor_isValid[0][1]&&neighbor_isRemote[0][1])) {
/* Statements in this Scop: S1349 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* fieldData_LaplaceCoeff_GMRF_5_p1 = (&fieldData_LaplaceCoeff_GMRF[5][(i3*1260)]);
double* buffer_Recv_1_p1 = (&buffer_Recv[1][(i3*35)]);
int i4 = 35;
for (; (i4<=68); i4 += 2) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)-1225)] = buffer_Recv_1_p1[(i4-35)];
fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)-1189)] = buffer_Recv_1_p1[(i4-34)];
}
for (; (i4<=69); i4 += 1) {
fieldData_LaplaceCoeff_GMRF_5_p1[((i4*36)-1225)] = buffer_Recv_1_p1[(i4-35)];
}
}
}
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Send[0]) {
waitForMPIReq(&mpiRequest_Send[0]);
reqOutstanding_Send[0] = false;
}
if (reqOutstanding_Send[1]) {
waitForMPIReq(&mpiRequest_Send[1]);
reqOutstanding_Send[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][2]&&neighbor_isRemote[0][2])) {
MPI_Isend(&fieldData_LaplaceCoeff_GMRF[5][73], 1, mpiDatatype_9_35_1260, neighbor_remoteRank[0][2], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][2]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[2]);
reqOutstanding_Send[2] = true;
}
if ((neighbor_isValid[0][3]&&neighbor_isRemote[0][3])) {
MPI_Isend(&fieldData_LaplaceCoeff_GMRF[5][1153], 1, mpiDatatype_9_35_1260, neighbor_remoteRank[0][3], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[0][3]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[3]);
reqOutstanding_Send[3] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if ((neighbor_isValid[0][2]&&neighbor_isRemote[0][2])) {
MPI_Irecv(&fieldData_LaplaceCoeff_GMRF[5][1], 1, mpiDatatype_9_35_1260, neighbor_remoteRank[0][2], ((unsigned int)(neighbor_fragCommId[0][2]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = true;
}
if ((neighbor_isValid[0][3]&&neighbor_isRemote[0][3])) {
MPI_Irecv(&fieldData_LaplaceCoeff_GMRF[5][1225], 1, mpiDatatype_9_35_1260, neighbor_remoteRank[0][3], ((unsigned int)(neighbor_fragCommId[0][3]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[3]);
reqOutstanding_Recv[3] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Recv[2]) {
waitForMPIReq(&mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = false;
}
if (reqOutstanding_Recv[3]) {
waitForMPIReq(&mpiRequest_Recv[3]);
reqOutstanding_Recv[3] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
;
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[0]) {
if (reqOutstanding_Send[2]) {
waitForMPIReq(&mpiRequest_Send[2]);
reqOutstanding_Send[2] = false;
}
if (reqOutstanding_Send[3]) {
waitForMPIReq(&mpiRequest_Send[3]);
reqOutstanding_Send[3] = false;
}
}
}
}
static inline void computeVectorGTRGAMMAPROT(double *lVector, int *eVector, double *gammaRates, int i, double qz, double rz,
traversalInfo *ti, double *EIGN, double *EI, double *EV, double *tipVector,
unsigned char **yVector, int mxtips)
{
double
*x1,
*x2,
*x3;
int
s,
pNumber = ti->pNumber,
rNumber = ti->rNumber,
qNumber = ti->qNumber,
index1[4],
index2[4];
x3 = &(lVector[80 * (pNumber - mxtips)]);
switch(ti->tipCase)
{
case TIP_TIP:
x1 = &(tipVector[20 * yVector[qNumber][i]]);
x2 = &(tipVector[20 * yVector[rNumber][i]]);
for(s = 0; s < 4; s++)
{
index1[s] = 0;
index2[s] = 0;
}
break;
case TIP_INNER:
x1 = &(tipVector[20 * yVector[qNumber][i]]);
x2 = &( lVector[80 * (rNumber - mxtips)]);
for(s = 0; s < 4; s++)
index1[s] = 0;
for(s = 0; s < 4; s++)
index2[s] = s;
break;
case INNER_INNER:
x1 = &(lVector[80 * (qNumber - mxtips)]);
x2 = &(lVector[80 * (rNumber - mxtips)]);
for(s = 0; s < 4; s++)
{
index1[s] = s;
index2[s] = s;
}
break;
default:
assert(0);
}
{
double
e1[20] __attribute__ ((aligned (BYTE_ALIGNMENT))),
e2[20] __attribute__ ((aligned (BYTE_ALIGNMENT))),
d1[20] __attribute__ ((aligned (BYTE_ALIGNMENT))),
d2[20] __attribute__ ((aligned (BYTE_ALIGNMENT))),
lz1, lz2;
int
l,
k,
scale,
j;
for(j = 0; j < 4; j++)
{
lz1 = qz * gammaRates[j];
lz2 = rz * gammaRates[j];
e1[0] = 1.0;
e2[0] = 1.0;
for(l = 1; l < 20; l++)
{
e1[l] = EXP(EIGN[l] * lz1);
e2[l] = EXP(EIGN[l] * lz2);
}
for(l = 0; l < 20; l+=2)
{
__m128d d1v = _mm_mul_pd(_mm_load_pd(&x1[20 * index1[j] + l]), _mm_load_pd(&e1[l]));
__m128d d2v = _mm_mul_pd(_mm_load_pd(&x2[20 * index2[j] + l]), _mm_load_pd(&e2[l]));
_mm_store_pd(&d1[l], d1v);
_mm_store_pd(&d2[l], d2v);
}
__m128d zero = _mm_setzero_pd();
for(l = 0; l < 20; l+=2)
_mm_store_pd(&x3[j * 20 + l], zero);
for(l = 0; l < 20; l++)
{
double *ev = &EV[l * 20];
__m128d ump_x1v = _mm_setzero_pd();
__m128d ump_x2v = _mm_setzero_pd();
__m128d x1px2v;
for(k = 0; k < 20; k+=2)
{
__m128d eiv = _mm_load_pd(&EI[20 * l + k]);
__m128d d1v = _mm_load_pd(&d1[k]);
__m128d d2v = _mm_load_pd(&d2[k]);
ump_x1v = _mm_add_pd(ump_x1v, _mm_mul_pd(d1v, eiv));
ump_x2v = _mm_add_pd(ump_x2v, _mm_mul_pd(d2v, eiv));
}
ump_x1v = _mm_hadd_pd(ump_x1v, ump_x1v);
ump_x2v = _mm_hadd_pd(ump_x2v, ump_x2v);
x1px2v = _mm_mul_pd(ump_x1v, ump_x2v);
for(k = 0; k < 20; k+=2)
{
__m128d ex3v = _mm_load_pd(&x3[j * 20 + k]);
__m128d EVV = _mm_load_pd(&ev[k]);
ex3v = _mm_add_pd(ex3v, _mm_mul_pd(x1px2v, EVV));
_mm_store_pd(&x3[j * 20 + k], ex3v);
}
}
}
scale = 1;
for(l = 0; scale && (l < 80); l++)
scale = ((x3[l] < minlikelihood) && (x3[l] > minusminlikelihood));
if(scale)
{
__m128d twoto = _mm_set_pd(twotothe256, twotothe256);
for(l = 0; l < 80; l+=2)
{
__m128d ex3v = _mm_mul_pd(_mm_load_pd(&x3[l]),twoto);
_mm_store_pd(&x3[l], ex3v);
}
*eVector = *eVector + 1;
}
return;
}
}
void exchlaplacecoeffData_6(unsigned int slot) {
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((!neighbor_isValid[1][0])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S902, S905, S908, S907, S901, S910, S904, S903, S906, S909, S900 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+4558)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+4626)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+4558)] = 0.000000e+00;
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
yPos = ((((i1-1)/6.400000e+01)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(1.000000e+00);
__m128d vec2 = _mm_set1_pd(6.400000e+01);
__m128d vec5 = _mm_set1_pd(yPos);
for (; (i1<63); i1 += 4) {
/* yPos = ((((i1-1)/6.400000e+01)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec3 = _mm_load1_pd((&posEnd[1]));
__m128d vec3_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec4 = _mm_load1_pd((&posBegin[1]));
__m128d vec4_2 = _mm_load1_pd((&posBegin[1]));
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0, vec1), vec2), _mm_sub_pd(vec3, vec4)), vec4);
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0_2, vec1), vec2), _mm_sub_pd(vec3_2, vec4_2)), vec4_2);
}
for (; (i1<66); i1 += 1) {
yPos = ((((i1-1)/6.400000e+01)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+9114)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+9182)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+9114)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+2)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+70)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+2)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+31894)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+31962)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+31894)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+36450)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+36518)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+36450)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
xPos = posBegin[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<63); i1 += 4) {
/* xPos = posBegin[0]; */
__m128d vec0 = _mm_load1_pd((&posBegin[0]));
__m128d vec0_2 = _mm_load1_pd((&posBegin[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<66); i1 += 1) {
xPos = posBegin[0];
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+18226)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+18294)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+18226)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+13670)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+13738)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+13670)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+27338)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+27406)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+27338)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+22782)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+22850)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+22782)] = 0.000000e+00;
}
}
}
}
}
if ((!neighbor_isValid[1][1])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S920, S914, S917, S911, S913, S916, S919, S921, S918, S912, S915 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+4622)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+4690)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+4622)] = 0.000000e+00;
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+31958)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+32026)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+31958)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+13734)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+13802)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+13734)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+66)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+134)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+66)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+22846)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+22914)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+22846)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
yPos = ((((i1-1)/6.400000e+01)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(1.000000e+00);
__m128d vec2 = _mm_set1_pd(6.400000e+01);
__m128d vec5 = _mm_set1_pd(yPos);
for (; (i1<63); i1 += 4) {
/* yPos = ((((i1-1)/6.400000e+01)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec3 = _mm_load1_pd((&posEnd[1]));
__m128d vec3_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec4 = _mm_load1_pd((&posBegin[1]));
__m128d vec4_2 = _mm_load1_pd((&posBegin[1]));
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0, vec1), vec2), _mm_sub_pd(vec3, vec4)), vec4);
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0_2, vec1), vec2), _mm_sub_pd(vec3_2, vec4_2)), vec4_2);
}
for (; (i1<66); i1 += 1) {
yPos = ((((i1-1)/6.400000e+01)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+18290)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+18358)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+18290)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+27402)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+27470)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+27402)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+36514)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+36582)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+36514)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i1 = 1;
for (; (i1<=64); i1 += 2) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+9178)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[((i1*68)+9246)] = 0.000000e+00;
}
for (; (i1<=65); i1 += 1) {
fieldData_LaplaceCoeff_6_p1[((i1*68)+9178)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
xPos = posEnd[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<63); i1 += 4) {
/* xPos = posEnd[0]; */
__m128d vec0 = _mm_load1_pd((&posEnd[0]));
__m128d vec0_2 = _mm_load1_pd((&posEnd[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<66); i1 += 1) {
xPos = posEnd[0];
}
}
}
}
}
if ((!neighbor_isValid[1][2])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S929, S923, S926, S931, S925, S928, S922, S930, S924, S927, S932 */
{
{
{
{
{
{
{
{
{
{
{
int i2 = 2;
for (; (i2<=65); i2 += 2) {
xPos = ((((i2-2)/6.400000e+01)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2-1)/6.400000e+01)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=66); i2 += 1) {
xPos = ((((i2-2)/6.400000e+01)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+18292)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+18293)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+18292)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+36516)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+36517)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+36516)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+31960)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+31961)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+31960)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+68)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+69)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+68)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+9180)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+9181)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+9180)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+22848)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+22849)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+22848)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+27404)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+27405)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+27404)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+4624)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+4625)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+4624)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+13736)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+13737)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+13736)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=65); i2 += 2) {
yPos = posBegin[1];
yPos = posBegin[1];
}
for (; (i2<=66); i2 += 1) {
yPos = posBegin[1];
}
}
}
}
}
if ((!neighbor_isValid[1][3])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S941, S935, S938, S943, S940, S934, S937, S942, S936, S939, S933 */
{
{
{
{
{
{
{
{
{
{
{
int i2 = 2;
for (; (i2<=65); i2 += 2) {
xPos = ((((i2-2)/6.400000e+01)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2-1)/6.400000e+01)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=66); i2 += 1) {
xPos = ((((i2-2)/6.400000e+01)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+36312)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+36313)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+36312)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+22644)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+22645)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+22644)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+13532)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+13533)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+13532)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+40868)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+40869)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+40868)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+18088)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+18089)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+18088)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+4420)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+4421)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+4420)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+31756)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+31757)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+31756)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+27200)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+27201)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+27200)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=65); i2 += 2) {
yPos = posEnd[1];
yPos = posEnd[1];
}
for (; (i2<=66); i2 += 1) {
yPos = posEnd[1];
}
}
}
{
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][0]);
int i2 = 2;
for (; (i2<=65); i2 += 2) {
fieldData_LaplaceCoeff_6_p1[(i2+8976)] = 0.000000e+00;
fieldData_LaplaceCoeff_6_p1[(i2+8977)] = 0.000000e+00;
}
for (; (i2<=66); i2 += 1) {
fieldData_LaplaceCoeff_6_p1[(i2+8976)] = 0.000000e+00;
}
}
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
/* Statements in this Scop: S944 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][(i3*4556)]);
double* buffer_Send_1_p1 = (&buffer_Send[1][(i3*65)]);
int i4 = 1;
for (; (i4<=64); i4 += 2) {
buffer_Send_1_p1[(i4-1)] = fieldData_LaplaceCoeff_6_p1[((i4*68)+66)];
buffer_Send_1_p1[i4] = fieldData_LaplaceCoeff_6_p1[((i4*68)+134)];
}
for (; (i4<=65); i4 += 1) {
buffer_Send_1_p1[(i4-1)] = fieldData_LaplaceCoeff_6_p1[((i4*68)+66)];
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
MPI_Isend(buffer_Send[1], 585, MPI_DOUBLE, neighbor_remoteRank[1][1], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][1]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[1]);
reqOutstanding_Send[1] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
MPI_Irecv(buffer_Recv[0], 585, MPI_DOUBLE, neighbor_remoteRank[1][0], ((unsigned int)(neighbor_fragCommId[1][0]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[0]) {
waitForMPIReq(&mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
/* Statements in this Scop: S945 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][(i3*4556)]);
double* buffer_Recv_0_p1 = (&buffer_Recv[0][(i3*65)]);
int i4 = 3;
for (; (i4<=66); i4 += 2) {
fieldData_LaplaceCoeff_6_p1[((i4*68)-134)] = buffer_Recv_0_p1[(i4-3)];
fieldData_LaplaceCoeff_6_p1[((i4*68)-66)] = buffer_Recv_0_p1[(i4-2)];
}
for (; (i4<=67); i4 += 1) {
fieldData_LaplaceCoeff_6_p1[((i4*68)-134)] = buffer_Recv_0_p1[(i4-3)];
}
}
}
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[1]) {
waitForMPIReq(&mpiRequest_Send[1]);
reqOutstanding_Send[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][3]&&neighbor_isRemote[1][3])) {
MPI_Isend(&fieldData_LaplaceCoeff[6][4422], 1, mpiDatatype_9_65_4556, neighbor_remoteRank[1][3], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][3]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[3]);
reqOutstanding_Send[3] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][2]&&neighbor_isRemote[1][2])) {
MPI_Irecv(&fieldData_LaplaceCoeff[6][70], 1, mpiDatatype_9_65_4556, neighbor_remoteRank[1][2], ((unsigned int)(neighbor_fragCommId[1][2]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[2]) {
waitForMPIReq(&mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[3]) {
waitForMPIReq(&mpiRequest_Send[3]);
reqOutstanding_Send[3] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
/* Statements in this Scop: S946 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][(i3*4556)]);
double* buffer_Send_0_p1 = (&buffer_Send[0][(i3*67)]);
int i4 = 0;
for (; (i4<=65); i4 += 2) {
buffer_Send_0_p1[i4] = fieldData_LaplaceCoeff_6_p1[((i4*68)+3)];
buffer_Send_0_p1[(i4+1)] = fieldData_LaplaceCoeff_6_p1[((i4*68)+71)];
}
for (; (i4<=66); i4 += 1) {
buffer_Send_0_p1[i4] = fieldData_LaplaceCoeff_6_p1[((i4*68)+3)];
}
}
}
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
/* Statements in this Scop: S947 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][(i3*4556)]);
double* buffer_Send_1_p1 = (&buffer_Send[1][(i3*67)]);
int i4 = 0;
for (; (i4<=65); i4 += 2) {
buffer_Send_1_p1[i4] = fieldData_LaplaceCoeff_6_p1[((i4*68)+65)];
buffer_Send_1_p1[(i4+1)] = fieldData_LaplaceCoeff_6_p1[((i4*68)+133)];
}
for (; (i4<=66); i4 += 1) {
buffer_Send_1_p1[i4] = fieldData_LaplaceCoeff_6_p1[((i4*68)+65)];
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
MPI_Isend(buffer_Send[0], 603, MPI_DOUBLE, neighbor_remoteRank[1][0], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][0]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[0]);
reqOutstanding_Send[0] = true;
}
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
MPI_Isend(buffer_Send[1], 603, MPI_DOUBLE, neighbor_remoteRank[1][1], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][1]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[1]);
reqOutstanding_Send[1] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
MPI_Irecv(buffer_Recv[0], 603, MPI_DOUBLE, neighbor_remoteRank[1][0], ((unsigned int)(neighbor_fragCommId[1][0]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = true;
}
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
MPI_Irecv(buffer_Recv[1], 603, MPI_DOUBLE, neighbor_remoteRank[1][1], ((unsigned int)(neighbor_fragCommId[1][1]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[1]);
reqOutstanding_Recv[1] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[0]) {
waitForMPIReq(&mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = false;
}
if (reqOutstanding_Recv[1]) {
waitForMPIReq(&mpiRequest_Recv[1]);
reqOutstanding_Recv[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
/* Statements in this Scop: S948 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][(i3*4556)]);
double* buffer_Recv_0_p1 = (&buffer_Recv[0][(i3*67)]);
int i4 = 1;
for (; (i4<=66); i4 += 2) {
fieldData_LaplaceCoeff_6_p1[((i4*68)-67)] = buffer_Recv_0_p1[(i4-1)];
fieldData_LaplaceCoeff_6_p1[((i4*68)+1)] = buffer_Recv_0_p1[i4];
}
for (; (i4<=67); i4 += 1) {
fieldData_LaplaceCoeff_6_p1[((i4*68)-67)] = buffer_Recv_0_p1[(i4-1)];
}
}
}
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
/* Statements in this Scop: S949 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* buffer_Recv_1_p1 = (&buffer_Recv[1][(i3*67)]);
double* fieldData_LaplaceCoeff_6_p1 = (&fieldData_LaplaceCoeff[6][(i3*4556)]);
int i4 = 67;
for (; (i4<=132); i4 += 2) {
fieldData_LaplaceCoeff_6_p1[((i4*68)-4489)] = buffer_Recv_1_p1[(i4-67)];
fieldData_LaplaceCoeff_6_p1[((i4*68)-4421)] = buffer_Recv_1_p1[(i4-66)];
}
for (; (i4<=133); i4 += 1) {
fieldData_LaplaceCoeff_6_p1[((i4*68)-4489)] = buffer_Recv_1_p1[(i4-67)];
}
}
}
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[0]) {
waitForMPIReq(&mpiRequest_Send[0]);
reqOutstanding_Send[0] = false;
}
if (reqOutstanding_Send[1]) {
waitForMPIReq(&mpiRequest_Send[1]);
reqOutstanding_Send[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][2]&&neighbor_isRemote[1][2])) {
MPI_Isend(&fieldData_LaplaceCoeff[6][137], 1, mpiDatatype_9_67_4556, neighbor_remoteRank[1][2], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][2]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[2]);
reqOutstanding_Send[2] = true;
}
if ((neighbor_isValid[1][3]&&neighbor_isRemote[1][3])) {
MPI_Isend(&fieldData_LaplaceCoeff[6][4353], 1, mpiDatatype_9_67_4556, neighbor_remoteRank[1][3], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][3]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[3]);
reqOutstanding_Send[3] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][2]&&neighbor_isRemote[1][2])) {
MPI_Irecv(&fieldData_LaplaceCoeff[6][1], 1, mpiDatatype_9_67_4556, neighbor_remoteRank[1][2], ((unsigned int)(neighbor_fragCommId[1][2]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = true;
}
if ((neighbor_isValid[1][3]&&neighbor_isRemote[1][3])) {
MPI_Irecv(&fieldData_LaplaceCoeff[6][4489], 1, mpiDatatype_9_67_4556, neighbor_remoteRank[1][3], ((unsigned int)(neighbor_fragCommId[1][3]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[3]);
reqOutstanding_Recv[3] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[2]) {
waitForMPIReq(&mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = false;
}
if (reqOutstanding_Recv[3]) {
waitForMPIReq(&mpiRequest_Recv[3]);
reqOutstanding_Recv[3] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[2]) {
waitForMPIReq(&mpiRequest_Send[2]);
reqOutstanding_Send[2] = false;
}
if (reqOutstanding_Send[3]) {
waitForMPIReq(&mpiRequest_Send[3]);
reqOutstanding_Send[3] = false;
}
}
}
}
mlib_status
mlib_ImageColorConvert2_D64(
const mlib_d64 *src,
mlib_s32 slb,
mlib_d64 *dst,
mlib_s32 dlb,
mlib_s32 xsize,
mlib_s32 ysize,
const mlib_d64 *fmat,
const mlib_d64 *offset)
{
/* pointers for pixel and line of source */
mlib_d64 *sa, *sl;
/* pointers for pixel and line of destination */
mlib_d64 *da, *dl;
/* indices */
mlib_s32 i, j;
/* coeff's */
mlib_d64 k02, k12, k22;
mlib_d64 p0, p1, p2, q2;
mlib_d64 off2;
/* intermediate */
__m128d t0, t1, t2, s0, s1, q;
__m128d p00, p11, p22;
/* packed kernel */
__m128d k0, k1, k2;
/* packed offset */
__m128d off;
/* load transposed kernel */
k0 = _mm_set_pd(fmat[3], fmat[0]);
k1 = _mm_set_pd(fmat[4], fmat[1]);
k2 = _mm_set_pd(fmat[5], fmat[2]);
/* load offset */
off = _mm_set_pd(offset[1], offset[0]);
/* keep kernel in regs */
k02 = fmat[6];
k12 = fmat[7];
k22 = fmat[8];
off2 = offset[2];
sa = sl = (void *)src;
da = dl = dst;
for (j = 0; j < ysize; j++) {
#ifdef __SUNPRO_C
#pragma pipeloop(0)
#endif /* __SUNPRO_C */
for (i = 0; i < 3 * xsize; i += 3) {
p0 = *sa;
sa ++;
p00 = _mm_set1_pd(p0);
p1 = *sa;
sa ++;
p11 = _mm_set1_pd(p1);
p2 = *sa;
sa ++;
p22 = _mm_set1_pd(p2);
t0 = _mm_mul_pd(p00, k0);
t1 = _mm_mul_pd(p11, k1);
t2 = _mm_mul_pd(p22, k2);
s0 = _mm_add_pd(t0, t1);
s1 = _mm_add_pd(t2, off);
q = _mm_add_pd(s0, s1);
q2 = p0 * k02 + p1 * k12 + p2 * k22 + off2;
_mm_storeu_pd(da, q);
da += 2;
*da = q2;
da ++;
}
/* set src ptrs to next row */
sl = sa = sl + slb;
/* set dst ptrs to next row */
dl = da = dl + dlb;
}
return (MLIB_SUCCESS);
}
void exchlaplacecoeffData_7(unsigned int slot) {
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((!neighbor_isValid[1][0])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S956, S958, S952, S955, S960, S954, S957, S951, S959, S950, S953 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+138338)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+138470)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+138338)] = 0.000000e+00;
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+34586)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+34718)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+34586)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+103754)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+103886)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+103754)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+2)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+134)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+2)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+69170)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+69302)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+69170)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+121046)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+121178)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+121046)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+51878)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+52010)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+51878)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
xPos = posBegin[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<127); i1 += 4) {
/* xPos = posBegin[0]; */
__m128d vec0 = _mm_load1_pd((&posBegin[0]));
__m128d vec0_2 = _mm_load1_pd((&posBegin[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<130); i1 += 1) {
xPos = posBegin[0];
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+86462)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+86594)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+86462)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+17294)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+17426)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+17294)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
yPos = ((((i1-1)/1.280000e+02)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(1.000000e+00);
__m128d vec2 = _mm_set1_pd(1.280000e+02);
__m128d vec5 = _mm_set1_pd(yPos);
for (; (i1<127); i1 += 4) {
/* yPos = ((((i1-1)/1.280000e+02)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec3 = _mm_load1_pd((&posEnd[1]));
__m128d vec3_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec4 = _mm_load1_pd((&posBegin[1]));
__m128d vec4_2 = _mm_load1_pd((&posBegin[1]));
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0, vec1), vec2), _mm_sub_pd(vec3, vec4)), vec4);
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0_2, vec1), vec2), _mm_sub_pd(vec3_2, vec4_2)), vec4_2);
}
for (; (i1<130); i1 += 1) {
yPos = ((((i1-1)/1.280000e+02)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
}
}
}
if ((!neighbor_isValid[1][1])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S962, S961, S970, S964, S967, S966, S969, S963, S971, S965, S968 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+121174)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+121306)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+121174)] = 0.000000e+00;
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+138466)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+138598)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+138466)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
yPos = ((((i1-1)/1.280000e+02)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
__m128d vec1 = _mm_set1_pd(1.000000e+00);
__m128d vec2 = _mm_set1_pd(1.280000e+02);
__m128d vec5 = _mm_set1_pd(yPos);
for (; (i1<127); i1 += 4) {
/* yPos = ((((i1-1)/1.280000e+02)*(posEnd[1]-posBegin[1]))+posBegin[1]); */
__m128d vec0 = _mm_set_pd(i1+1,i1);
__m128d vec0_2 = _mm_set_pd(i1+1,i1);
__m128d vec3 = _mm_load1_pd((&posEnd[1]));
__m128d vec3_2 = _mm_load1_pd((&posEnd[1]));
__m128d vec4 = _mm_load1_pd((&posBegin[1]));
__m128d vec4_2 = _mm_load1_pd((&posBegin[1]));
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0, vec1), vec2), _mm_sub_pd(vec3, vec4)), vec4);
vec5 = _mm_add_pd(_mm_mul_pd(_mm_div_pd(_mm_sub_pd(vec0_2, vec1), vec2), _mm_sub_pd(vec3_2, vec4_2)), vec4_2);
}
for (; (i1<130); i1 += 1) {
yPos = ((((i1-1)/1.280000e+02)*(posEnd[1]-posBegin[1]))+posBegin[1]);
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+103882)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+104014)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+103882)] = 0.000000e+00;
}
}
}
{
int i1 = 1;
for (; (i1<(2&(~1))); i1 += 1) {
xPos = posEnd[0];
}
__m128d vec1 = _mm_set1_pd(xPos);
for (; (i1<127); i1 += 4) {
/* xPos = posEnd[0]; */
__m128d vec0 = _mm_load1_pd((&posEnd[0]));
__m128d vec0_2 = _mm_load1_pd((&posEnd[0]));
vec1 = vec0;
vec1 = vec0_2;
}
for (; (i1<130); i1 += 1) {
xPos = posEnd[0];
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+130)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+262)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+130)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+69298)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+69430)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+69298)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+52006)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+52138)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+52006)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+86590)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+86722)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+86590)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+34714)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+34846)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+34714)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i1 = 1;
for (; (i1<=128); i1 += 2) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+17422)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[((i1*132)+17554)] = 0.000000e+00;
}
for (; (i1<=129); i1 += 1) {
fieldData_LaplaceCoeff_7_p1[((i1*132)+17422)] = 0.000000e+00;
}
}
}
}
}
if ((!neighbor_isValid[1][2])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S982, S976, S979, S973, S972, S981, S975, S978, S977, S980, S974 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+138468)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+138469)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+138468)] = 0.000000e+00;
}
}
{
int i2 = 2;
for (; (i2<=129); i2 += 2) {
xPos = ((((i2-2)/1.280000e+02)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2-1)/1.280000e+02)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=130); i2 += 1) {
xPos = ((((i2-2)/1.280000e+02)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+34716)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+34717)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+34716)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+17424)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+17425)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+17424)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+132)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+133)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+132)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+86592)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+86593)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+86592)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+52008)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+52009)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+52008)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+103884)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+103885)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+103884)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+121176)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+121177)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+121176)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=129); i2 += 2) {
yPos = posBegin[1];
yPos = posBegin[1];
}
for (; (i2<=130); i2 += 1) {
yPos = posBegin[1];
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+69300)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+69301)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+69300)] = 0.000000e+00;
}
}
}
}
}
if ((!neighbor_isValid[1][3])) {
{
double xPos;
double yPos;
/* Statements in this Scop: S988, S991, S985, S990, S984, S993, S987, S983, S992, S986, S989 */
{
{
{
{
{
{
{
{
{
{
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+138072)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+138073)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+138072)] = 0.000000e+00;
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+34320)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+34321)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+34320)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+155364)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+155365)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+155364)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+51612)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+51613)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+51612)] = 0.000000e+00;
}
}
}
{
int i2 = 2;
for (; (i2<=129); i2 += 2) {
yPos = posEnd[1];
yPos = posEnd[1];
}
for (; (i2<=130); i2 += 1) {
yPos = posEnd[1];
}
}
}
{
int i2 = 2;
for (; (i2<=129); i2 += 2) {
xPos = ((((i2-2)/1.280000e+02)*(posEnd[0]-posBegin[0]))+posBegin[0]);
xPos = ((((i2-1)/1.280000e+02)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
for (; (i2<=130); i2 += 1) {
xPos = ((((i2-2)/1.280000e+02)*(posEnd[0]-posBegin[0]))+posBegin[0]);
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+103488)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+103489)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+103488)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+68904)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+68905)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+68904)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+17028)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+17029)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+17028)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+120780)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+120781)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+120780)] = 0.000000e+00;
}
}
}
{
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][0]);
int i2 = 2;
for (; (i2<=129); i2 += 2) {
fieldData_LaplaceCoeff_7_p1[(i2+86196)] = 0.000000e+00;
fieldData_LaplaceCoeff_7_p1[(i2+86197)] = 0.000000e+00;
}
for (; (i2<=130); i2 += 1) {
fieldData_LaplaceCoeff_7_p1[(i2+86196)] = 0.000000e+00;
}
}
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
/* Statements in this Scop: S994 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* buffer_Send_1_p1 = (&buffer_Send[1][(i3*129)]);
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][(i3*17292)]);
int i4 = 1;
for (; (i4<=128); i4 += 2) {
buffer_Send_1_p1[(i4-1)] = fieldData_LaplaceCoeff_7_p1[((i4*132)+130)];
buffer_Send_1_p1[i4] = fieldData_LaplaceCoeff_7_p1[((i4*132)+262)];
}
for (; (i4<=129); i4 += 1) {
buffer_Send_1_p1[(i4-1)] = fieldData_LaplaceCoeff_7_p1[((i4*132)+130)];
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
MPI_Isend(buffer_Send[1], 1161, MPI_DOUBLE, neighbor_remoteRank[1][1], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][1]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[1]);
reqOutstanding_Send[1] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
MPI_Irecv(buffer_Recv[0], 1161, MPI_DOUBLE, neighbor_remoteRank[1][0], ((unsigned int)(neighbor_fragCommId[1][0]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[0]) {
waitForMPIReq(&mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
/* Statements in this Scop: S995 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* buffer_Recv_0_p1 = (&buffer_Recv[0][(i3*129)]);
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][(i3*17292)]);
int i4 = 3;
for (; (i4<=130); i4 += 2) {
fieldData_LaplaceCoeff_7_p1[((i4*132)-262)] = buffer_Recv_0_p1[(i4-3)];
fieldData_LaplaceCoeff_7_p1[((i4*132)-130)] = buffer_Recv_0_p1[(i4-2)];
}
for (; (i4<=131); i4 += 1) {
fieldData_LaplaceCoeff_7_p1[((i4*132)-262)] = buffer_Recv_0_p1[(i4-3)];
}
}
}
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[1]) {
waitForMPIReq(&mpiRequest_Send[1]);
reqOutstanding_Send[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][3]&&neighbor_isRemote[1][3])) {
MPI_Isend(&fieldData_LaplaceCoeff[7][17030], 1, mpiDatatype_9_129_17292, neighbor_remoteRank[1][3], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][3]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[3]);
reqOutstanding_Send[3] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][2]&&neighbor_isRemote[1][2])) {
MPI_Irecv(&fieldData_LaplaceCoeff[7][134], 1, mpiDatatype_9_129_17292, neighbor_remoteRank[1][2], ((unsigned int)(neighbor_fragCommId[1][2]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[2]) {
waitForMPIReq(&mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[3]) {
waitForMPIReq(&mpiRequest_Send[3]);
reqOutstanding_Send[3] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
/* Statements in this Scop: S996 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][(i3*17292)]);
double* buffer_Send_0_p1 = (&buffer_Send[0][(i3*131)]);
int i4 = 0;
for (; (i4<=129); i4 += 2) {
buffer_Send_0_p1[i4] = fieldData_LaplaceCoeff_7_p1[((i4*132)+3)];
buffer_Send_0_p1[(i4+1)] = fieldData_LaplaceCoeff_7_p1[((i4*132)+135)];
}
for (; (i4<=130); i4 += 1) {
buffer_Send_0_p1[i4] = fieldData_LaplaceCoeff_7_p1[((i4*132)+3)];
}
}
}
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
/* Statements in this Scop: S997 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* buffer_Send_1_p1 = (&buffer_Send[1][(i3*131)]);
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][(i3*17292)]);
int i4 = 0;
for (; (i4<=129); i4 += 2) {
buffer_Send_1_p1[i4] = fieldData_LaplaceCoeff_7_p1[((i4*132)+129)];
buffer_Send_1_p1[(i4+1)] = fieldData_LaplaceCoeff_7_p1[((i4*132)+261)];
}
for (; (i4<=130); i4 += 1) {
buffer_Send_1_p1[i4] = fieldData_LaplaceCoeff_7_p1[((i4*132)+129)];
}
}
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
MPI_Isend(buffer_Send[0], 1179, MPI_DOUBLE, neighbor_remoteRank[1][0], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][0]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[0]);
reqOutstanding_Send[0] = true;
}
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
MPI_Isend(buffer_Send[1], 1179, MPI_DOUBLE, neighbor_remoteRank[1][1], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][1]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[1]);
reqOutstanding_Send[1] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
MPI_Irecv(buffer_Recv[0], 1179, MPI_DOUBLE, neighbor_remoteRank[1][0], ((unsigned int)(neighbor_fragCommId[1][0]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = true;
}
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
MPI_Irecv(buffer_Recv[1], 1179, MPI_DOUBLE, neighbor_remoteRank[1][1], ((unsigned int)(neighbor_fragCommId[1][1]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[1]);
reqOutstanding_Recv[1] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[0]) {
waitForMPIReq(&mpiRequest_Recv[0]);
reqOutstanding_Recv[0] = false;
}
if (reqOutstanding_Recv[1]) {
waitForMPIReq(&mpiRequest_Recv[1]);
reqOutstanding_Recv[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][0]&&neighbor_isRemote[1][0])) {
/* Statements in this Scop: S998 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* buffer_Recv_0_p1 = (&buffer_Recv[0][(i3*131)]);
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][(i3*17292)]);
int i4 = 1;
for (; (i4<=130); i4 += 2) {
fieldData_LaplaceCoeff_7_p1[((i4*132)-131)] = buffer_Recv_0_p1[(i4-1)];
fieldData_LaplaceCoeff_7_p1[((i4*132)+1)] = buffer_Recv_0_p1[i4];
}
for (; (i4<=131); i4 += 1) {
fieldData_LaplaceCoeff_7_p1[((i4*132)-131)] = buffer_Recv_0_p1[(i4-1)];
}
}
}
if ((neighbor_isValid[1][1]&&neighbor_isRemote[1][1])) {
/* Statements in this Scop: S999 */
for (int i3 = 0; (i3<=8); i3 += 1) {
double* buffer_Recv_1_p1 = (&buffer_Recv[1][(i3*131)]);
double* fieldData_LaplaceCoeff_7_p1 = (&fieldData_LaplaceCoeff[7][(i3*17292)]);
int i4 = 131;
for (; (i4<=260); i4 += 2) {
fieldData_LaplaceCoeff_7_p1[((i4*132)-17161)] = buffer_Recv_1_p1[(i4-131)];
fieldData_LaplaceCoeff_7_p1[((i4*132)-17029)] = buffer_Recv_1_p1[(i4-130)];
}
for (; (i4<=261); i4 += 1) {
fieldData_LaplaceCoeff_7_p1[((i4*132)-17161)] = buffer_Recv_1_p1[(i4-131)];
}
}
}
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[0]) {
waitForMPIReq(&mpiRequest_Send[0]);
reqOutstanding_Send[0] = false;
}
if (reqOutstanding_Send[1]) {
waitForMPIReq(&mpiRequest_Send[1]);
reqOutstanding_Send[1] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
;
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][2]&&neighbor_isRemote[1][2])) {
MPI_Isend(&fieldData_LaplaceCoeff[7][265], 1, mpiDatatype_9_131_17292, neighbor_remoteRank[1][2], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][2]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[2]);
reqOutstanding_Send[2] = true;
}
if ((neighbor_isValid[1][3]&&neighbor_isRemote[1][3])) {
MPI_Isend(&fieldData_LaplaceCoeff[7][16897], 1, mpiDatatype_9_131_17292, neighbor_remoteRank[1][3], ((unsigned int)commId << 16) + ((unsigned int)(neighbor_fragCommId[1][3]) & 0x0000ffff), mpiCommunicator, &mpiRequest_Send[3]);
reqOutstanding_Send[3] = true;
}
}
}
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if ((neighbor_isValid[1][2]&&neighbor_isRemote[1][2])) {
MPI_Irecv(&fieldData_LaplaceCoeff[7][1], 1, mpiDatatype_9_131_17292, neighbor_remoteRank[1][2], ((unsigned int)(neighbor_fragCommId[1][2]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = true;
}
if ((neighbor_isValid[1][3]&&neighbor_isRemote[1][3])) {
MPI_Irecv(&fieldData_LaplaceCoeff[7][17161], 1, mpiDatatype_9_131_17292, neighbor_remoteRank[1][3], ((unsigned int)(neighbor_fragCommId[1][3]) << 16) + ((unsigned int)commId & 0x0000ffff), mpiCommunicator, &mpiRequest_Recv[3]);
reqOutstanding_Recv[3] = true;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Recv[2]) {
waitForMPIReq(&mpiRequest_Recv[2]);
reqOutstanding_Recv[2] = false;
}
if (reqOutstanding_Recv[3]) {
waitForMPIReq(&mpiRequest_Recv[3]);
reqOutstanding_Recv[3] = false;
}
}
}
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
;
;
}
}
;
;
for (int fragmentIdx = 0; fragmentIdx < 1; ++fragmentIdx) {
if (isValidForSubdomain[1]) {
if (reqOutstanding_Send[2]) {
waitForMPIReq(&mpiRequest_Send[2]);
reqOutstanding_Send[2] = false;
}
if (reqOutstanding_Send[3]) {
waitForMPIReq(&mpiRequest_Send[3]);
reqOutstanding_Send[3] = false;
}
}
}
}
