mirror of
https://github.com/gnss-sdr/gnss-sdr
synced 2024-06-18 02:59:58 +00:00
68fdedb224
Manual usage of _mm256_zeroupper() is not required and leads to miscompilation with GCC 10.2 and optimization level -O3
487 lines
19 KiB
C++
487 lines
19 KiB
C++
/*!
|
|
* \file volk_gnsssdr_32fc_32f_rotator_dot_prod_32fc_xn.h
|
|
* \brief VOLK_GNSSSDR kernel: multiplies N complex (32-bit float per component) vectors
|
|
* by a common vector, phase rotated and accumulates the results in N float complex outputs.
|
|
* \authors <ul>
|
|
* <li> Cillian O'Driscoll 2016. cillian.odriscoll(at)gmail.com
|
|
* </ul>
|
|
*
|
|
* VOLK_GNSSSDR kernel that multiplies N 32 bits complex vectors by a common vector, which is
|
|
* phase-rotated by phase offset and phase increment, and accumulates the results
|
|
* in N 32 bits float complex outputs.
|
|
* It is optimized to perform the N tap correlation process in GNSS receivers.
|
|
*
|
|
* -----------------------------------------------------------------------------
|
|
*
|
|
* GNSS-SDR is a Global Navigation Satellite System software-defined receiver.
|
|
* This file is part of GNSS-SDR.
|
|
*
|
|
* Copyright (C) 2010-2020 (see AUTHORS file for a list of contributors)
|
|
* SPDX-License-Identifier: GPL-3.0-or-later
|
|
*
|
|
* -----------------------------------------------------------------------------
|
|
*/
|
|
|
|
/*!
|
|
* \page volk_gnsssdr_32fc_32f_rotator_dot_prod_32fc_xn
|
|
*
|
|
* \b Overview
|
|
*
|
|
* Rotates and multiplies the reference complex vector with an arbitrary number of other real vectors,
|
|
* accumulates the results and stores them in the output vector.
|
|
* The rotation is done at a fixed rate per sample, from an initial \p phase offset.
|
|
* This function can be used for Doppler wipe-off and multiple correlator.
|
|
*
|
|
* <b>Dispatcher Prototype</b>
|
|
* \code
|
|
* void volk_gnsssdr_32fc_32f_rotator_dot_prod_32fc_xn(lv_32fc_t* result, const lv_32fc_t* in_common, const lv_32fc_t phase_inc, lv_32fc_t* phase, const float** in_a, int num_a_vectors, unsigned int num_points);
|
|
* \endcode
|
|
*
|
|
* \b Inputs
|
|
* \li in_common: Pointer to one of the vectors to be rotated, multiplied and accumulated (reference vector).
|
|
* \li phase_inc: Phase increment = lv_cmake(cos(phase_step_rad), sin(phase_step_rad))
|
|
* \li phase: Initial phase = lv_cmake(cos(initial_phase_rad), sin(initial_phase_rad))
|
|
* \li in_a: Pointer to an array of pointers to multiple vectors to be multiplied and accumulated.
|
|
* \li num_a_vectors: Number of vectors to be multiplied by the reference vector and accumulated.
|
|
* \li num_points: Number of complex values to be multiplied together, accumulated and stored into \p result.
|
|
*
|
|
* \b Outputs
|
|
* \li phase: Final phase.
|
|
* \li result: Vector of \p num_a_vectors components with the multiple vectors of \p in_a rotated, multiplied by \p in_common and accumulated.
|
|
*
|
|
*/
|
|
|
|
#ifndef INCLUDED_volk_gnsssdr_32fc_32f_rotator_dot_prod_32fc_xn_H
|
|
#define INCLUDED_volk_gnsssdr_32fc_32f_rotator_dot_prod_32fc_xn_H
|
|
|
|
|
|
#include <volk_gnsssdr/saturation_arithmetic.h>
|
|
#include <volk_gnsssdr/volk_gnsssdr.h>
|
|
#include <volk_gnsssdr/volk_gnsssdr_complex.h>
|
|
#include <volk_gnsssdr/volk_gnsssdr_malloc.h>
|
|
#include <math.h>
|
|
|
|
#ifdef LV_HAVE_GENERIC
|
|
|
|
static inline void volk_gnsssdr_32fc_32f_rotator_dot_prod_32fc_xn_generic(lv_32fc_t* result, const lv_32fc_t* in_common, const lv_32fc_t phase_inc, lv_32fc_t* phase, const float** in_a, int num_a_vectors, unsigned int num_points)
|
|
{
|
|
lv_32fc_t tmp32_1, tmp32_2;
|
|
int n_vec;
|
|
unsigned int n;
|
|
for (n_vec = 0; n_vec < num_a_vectors; n_vec++)
|
|
{
|
|
result[n_vec] = lv_cmake(0.0f, 0.0f);
|
|
}
|
|
for (n = 0; n < num_points; n++)
|
|
{
|
|
tmp32_1 = *in_common++ * (*phase); // if(n<10 || n >= 8108) printf("generic phase %i: %f,%f\n", n,lv_creal(*phase),lv_cimag(*phase));
|
|
|
|
// Regenerate phase
|
|
if (n % 256 == 0)
|
|
{
|
|
// printf("Phase before regeneration %i: %f,%f Modulus: %f\n", n,lv_creal(*phase),lv_cimag(*phase), cabsf(*phase));
|
|
#ifdef __cplusplus
|
|
(*phase) /= std::abs((*phase));
|
|
#else
|
|
(*phase) /= hypotf(lv_creal(*phase), lv_cimag(*phase));
|
|
#endif
|
|
// printf("Phase after regeneration %i: %f,%f Modulus: %f\n", n,lv_creal(*phase),lv_cimag(*phase), cabsf(*phase));
|
|
}
|
|
|
|
(*phase) *= phase_inc;
|
|
for (n_vec = 0; n_vec < num_a_vectors; n_vec++)
|
|
{
|
|
tmp32_2 = tmp32_1 * in_a[n_vec][n];
|
|
result[n_vec] += tmp32_2;
|
|
}
|
|
}
|
|
}
|
|
|
|
#endif /* LV_HAVE_GENERIC */
|
|
|
|
|
|
#ifdef LV_HAVE_GENERIC
|
|
|
|
static inline void volk_gnsssdr_32fc_32f_rotator_dot_prod_32fc_xn_generic_reload(lv_32fc_t* result, const lv_32fc_t* in_common, const lv_32fc_t phase_inc, lv_32fc_t* phase, const float** in_a, int num_a_vectors, unsigned int num_points)
|
|
{
|
|
lv_32fc_t tmp32_1, tmp32_2;
|
|
const unsigned int ROTATOR_RELOAD = 256;
|
|
int n_vec;
|
|
unsigned int n;
|
|
unsigned int j;
|
|
for (n_vec = 0; n_vec < num_a_vectors; n_vec++)
|
|
{
|
|
result[n_vec] = lv_cmake(0.0f, 0.0f);
|
|
}
|
|
|
|
for (n = 0; n < num_points / ROTATOR_RELOAD; n++)
|
|
{
|
|
for (j = 0; j < ROTATOR_RELOAD; j++)
|
|
{
|
|
tmp32_1 = *in_common++ * (*phase);
|
|
(*phase) *= phase_inc;
|
|
for (n_vec = 0; n_vec < num_a_vectors; n_vec++)
|
|
{
|
|
tmp32_2 = tmp32_1 * in_a[n_vec][n * ROTATOR_RELOAD + j];
|
|
result[n_vec] += tmp32_2;
|
|
}
|
|
}
|
|
/* Regenerate phase */
|
|
#ifdef __cplusplus
|
|
(*phase) /= std::abs((*phase));
|
|
#else
|
|
// (*phase) /= cabsf((*phase));
|
|
(*phase) /= hypotf(lv_creal(*phase), lv_cimag(*phase));
|
|
#endif
|
|
}
|
|
|
|
for (j = 0; j < num_points % ROTATOR_RELOAD; j++)
|
|
{
|
|
tmp32_1 = *in_common++ * (*phase);
|
|
(*phase) *= phase_inc;
|
|
for (n_vec = 0; n_vec < num_a_vectors; n_vec++)
|
|
{
|
|
tmp32_2 = tmp32_1 * in_a[n_vec][(num_points / ROTATOR_RELOAD) * ROTATOR_RELOAD + j];
|
|
result[n_vec] += tmp32_2;
|
|
}
|
|
}
|
|
}
|
|
|
|
#endif /* LV_HAVE_GENERIC */
|
|
|
|
#ifdef LV_HAVE_AVX
|
|
#include <volk_gnsssdr/volk_gnsssdr_avx_intrinsics.h>
|
|
#include <immintrin.h>
|
|
static inline void volk_gnsssdr_32fc_32f_rotator_dot_prod_32fc_xn_u_avx(lv_32fc_t* result, const lv_32fc_t* in_common, const lv_32fc_t phase_inc, lv_32fc_t* phase, const float** in_a, int num_a_vectors, unsigned int num_points)
|
|
{
|
|
#ifndef WIN32
|
|
unsigned int number = 0;
|
|
int vec_ind = 0;
|
|
unsigned int i = 0;
|
|
const unsigned int sixteenthPoints = num_points / 16;
|
|
|
|
const float* aPtr = (float*)in_common;
|
|
const float* bPtr[num_a_vectors];
|
|
for (vec_ind = 0; vec_ind < num_a_vectors; ++vec_ind)
|
|
{
|
|
bPtr[vec_ind] = in_a[vec_ind];
|
|
}
|
|
|
|
lv_32fc_t _phase = (*phase);
|
|
lv_32fc_t wo;
|
|
|
|
__m256 a0Val, a1Val, a2Val, a3Val;
|
|
__m256 b0Val[num_a_vectors], b1Val[num_a_vectors], b2Val[num_a_vectors], b3Val[num_a_vectors];
|
|
__m256 x0Val[num_a_vectors], x1Val[num_a_vectors], x0loVal[num_a_vectors], x0hiVal[num_a_vectors], x1loVal[num_a_vectors], x1hiVal[num_a_vectors];
|
|
__m256 c0Val[num_a_vectors], c1Val[num_a_vectors], c2Val[num_a_vectors], c3Val[num_a_vectors];
|
|
|
|
__m256 dotProdVal0[num_a_vectors];
|
|
__m256 dotProdVal1[num_a_vectors];
|
|
__m256 dotProdVal2[num_a_vectors];
|
|
__m256 dotProdVal3[num_a_vectors];
|
|
|
|
for (vec_ind = 0; vec_ind < num_a_vectors; vec_ind++)
|
|
{
|
|
dotProdVal0[vec_ind] = _mm256_setzero_ps();
|
|
dotProdVal1[vec_ind] = _mm256_setzero_ps();
|
|
dotProdVal2[vec_ind] = _mm256_setzero_ps();
|
|
dotProdVal3[vec_ind] = _mm256_setzero_ps();
|
|
}
|
|
|
|
// Set up the complex rotator
|
|
__m256 z0, z1, z2, z3;
|
|
__VOLK_ATTR_ALIGNED(32)
|
|
lv_32fc_t phase_vec[16];
|
|
for (vec_ind = 0; vec_ind < 16; ++vec_ind)
|
|
{
|
|
phase_vec[vec_ind] = _phase;
|
|
_phase *= phase_inc;
|
|
}
|
|
|
|
z0 = _mm256_load_ps((float*)phase_vec);
|
|
z1 = _mm256_load_ps((float*)(phase_vec + 4));
|
|
z2 = _mm256_load_ps((float*)(phase_vec + 8));
|
|
z3 = _mm256_load_ps((float*)(phase_vec + 12));
|
|
|
|
lv_32fc_t dz = phase_inc;
|
|
dz *= dz;
|
|
dz *= dz;
|
|
dz *= dz;
|
|
dz *= dz; // dz = phase_inc^16;
|
|
|
|
for (vec_ind = 0; vec_ind < 4; ++vec_ind)
|
|
{
|
|
phase_vec[vec_ind] = dz;
|
|
}
|
|
|
|
__m256 dz_reg = _mm256_load_ps((float*)phase_vec);
|
|
dz_reg = _mm256_complexnormalise_ps(dz_reg);
|
|
|
|
for (; number < sixteenthPoints; number++)
|
|
{
|
|
a0Val = _mm256_loadu_ps(aPtr);
|
|
a1Val = _mm256_loadu_ps(aPtr + 8);
|
|
a2Val = _mm256_loadu_ps(aPtr + 16);
|
|
a3Val = _mm256_loadu_ps(aPtr + 24);
|
|
|
|
a0Val = _mm256_complexmul_ps(a0Val, z0);
|
|
a1Val = _mm256_complexmul_ps(a1Val, z1);
|
|
a2Val = _mm256_complexmul_ps(a2Val, z2);
|
|
a3Val = _mm256_complexmul_ps(a3Val, z3);
|
|
|
|
z0 = _mm256_complexmul_ps(z0, dz_reg);
|
|
z1 = _mm256_complexmul_ps(z1, dz_reg);
|
|
z2 = _mm256_complexmul_ps(z2, dz_reg);
|
|
z3 = _mm256_complexmul_ps(z3, dz_reg);
|
|
|
|
for (vec_ind = 0; vec_ind < num_a_vectors; ++vec_ind)
|
|
{
|
|
x0Val[vec_ind] = _mm256_loadu_ps(bPtr[vec_ind]); // t0|t1|t2|t3|t4|t5|t6|t7
|
|
x1Val[vec_ind] = _mm256_loadu_ps(bPtr[vec_ind] + 8);
|
|
x0loVal[vec_ind] = _mm256_unpacklo_ps(x0Val[vec_ind], x0Val[vec_ind]); // t0|t0|t1|t1|t4|t4|t5|t5
|
|
x0hiVal[vec_ind] = _mm256_unpackhi_ps(x0Val[vec_ind], x0Val[vec_ind]); // t2|t2|t3|t3|t6|t6|t7|t7
|
|
x1loVal[vec_ind] = _mm256_unpacklo_ps(x1Val[vec_ind], x1Val[vec_ind]);
|
|
x1hiVal[vec_ind] = _mm256_unpackhi_ps(x1Val[vec_ind], x1Val[vec_ind]);
|
|
|
|
// TODO: it may be possible to rearrange swizzling to better pipeline data
|
|
b0Val[vec_ind] = _mm256_permute2f128_ps(x0loVal[vec_ind], x0hiVal[vec_ind], 0x20); // t0|t0|t1|t1|t2|t2|t3|t3
|
|
b1Val[vec_ind] = _mm256_permute2f128_ps(x0loVal[vec_ind], x0hiVal[vec_ind], 0x31); // t4|t4|t5|t5|t6|t6|t7|t7
|
|
b2Val[vec_ind] = _mm256_permute2f128_ps(x1loVal[vec_ind], x1hiVal[vec_ind], 0x20);
|
|
b3Val[vec_ind] = _mm256_permute2f128_ps(x1loVal[vec_ind], x1hiVal[vec_ind], 0x31);
|
|
|
|
c0Val[vec_ind] = _mm256_mul_ps(a0Val, b0Val[vec_ind]);
|
|
c1Val[vec_ind] = _mm256_mul_ps(a1Val, b1Val[vec_ind]);
|
|
c2Val[vec_ind] = _mm256_mul_ps(a2Val, b2Val[vec_ind]);
|
|
c3Val[vec_ind] = _mm256_mul_ps(a3Val, b3Val[vec_ind]);
|
|
|
|
dotProdVal0[vec_ind] = _mm256_add_ps(c0Val[vec_ind], dotProdVal0[vec_ind]);
|
|
dotProdVal1[vec_ind] = _mm256_add_ps(c1Val[vec_ind], dotProdVal1[vec_ind]);
|
|
dotProdVal2[vec_ind] = _mm256_add_ps(c2Val[vec_ind], dotProdVal2[vec_ind]);
|
|
dotProdVal3[vec_ind] = _mm256_add_ps(c3Val[vec_ind], dotProdVal3[vec_ind]);
|
|
|
|
bPtr[vec_ind] += 16;
|
|
}
|
|
|
|
// Force the rotators back onto the unit circle
|
|
if ((number % 64) == 0)
|
|
{
|
|
z0 = _mm256_complexnormalise_ps(z0);
|
|
z1 = _mm256_complexnormalise_ps(z1);
|
|
z2 = _mm256_complexnormalise_ps(z2);
|
|
z3 = _mm256_complexnormalise_ps(z3);
|
|
}
|
|
|
|
aPtr += 32;
|
|
}
|
|
__VOLK_ATTR_ALIGNED(32)
|
|
lv_32fc_t dotProductVector[4];
|
|
|
|
for (vec_ind = 0; vec_ind < num_a_vectors; ++vec_ind)
|
|
{
|
|
dotProdVal0[vec_ind] = _mm256_add_ps(dotProdVal0[vec_ind], dotProdVal1[vec_ind]);
|
|
dotProdVal0[vec_ind] = _mm256_add_ps(dotProdVal0[vec_ind], dotProdVal2[vec_ind]);
|
|
dotProdVal0[vec_ind] = _mm256_add_ps(dotProdVal0[vec_ind], dotProdVal3[vec_ind]);
|
|
|
|
_mm256_store_ps((float*)dotProductVector, dotProdVal0[vec_ind]); // Store the results back into the dot product vector
|
|
|
|
result[vec_ind] = lv_cmake(0.0f, 0.0f);
|
|
for (i = 0; i < 4; ++i)
|
|
{
|
|
result[vec_ind] += dotProductVector[i];
|
|
}
|
|
}
|
|
|
|
z0 = _mm256_complexnormalise_ps(z0);
|
|
_mm256_store_ps((float*)phase_vec, z0);
|
|
_phase = phase_vec[0];
|
|
|
|
number = sixteenthPoints * 16;
|
|
for (; number < num_points; number++)
|
|
{
|
|
wo = in_common[number] * _phase;
|
|
_phase *= phase_inc;
|
|
|
|
for (vec_ind = 0; vec_ind < num_a_vectors; ++vec_ind)
|
|
{
|
|
result[vec_ind] += wo * in_a[vec_ind][number];
|
|
}
|
|
}
|
|
|
|
*phase = _phase;
|
|
#else
|
|
volk_gnsssdr_32fc_32f_rotator_dot_prod_32fc_xn_generic_reload(result, in_common, phase_inc, phase, in_a, num_a_vectors, num_points);
|
|
#endif
|
|
}
|
|
|
|
#endif /* LV_HAVE_AVX */
|
|
|
|
|
|
#ifdef LV_HAVE_AVX
|
|
#include <volk_gnsssdr/volk_gnsssdr_avx_intrinsics.h>
|
|
#include <immintrin.h>
|
|
static inline void volk_gnsssdr_32fc_32f_rotator_dot_prod_32fc_xn_a_avx(lv_32fc_t* result, const lv_32fc_t* in_common, const lv_32fc_t phase_inc, lv_32fc_t* phase, const float** in_a, int num_a_vectors, unsigned int num_points)
|
|
{
|
|
#ifndef WIN32
|
|
unsigned int number = 0;
|
|
int vec_ind = 0;
|
|
unsigned int i = 0;
|
|
const unsigned int sixteenthPoints = num_points / 16;
|
|
|
|
const float* aPtr = (float*)in_common;
|
|
const float* bPtr[num_a_vectors];
|
|
for (vec_ind = 0; vec_ind < num_a_vectors; ++vec_ind)
|
|
{
|
|
bPtr[vec_ind] = in_a[vec_ind];
|
|
}
|
|
|
|
lv_32fc_t _phase = (*phase);
|
|
lv_32fc_t wo;
|
|
|
|
__m256 a0Val, a1Val, a2Val, a3Val;
|
|
__m256 b0Val[num_a_vectors], b1Val[num_a_vectors], b2Val[num_a_vectors], b3Val[num_a_vectors];
|
|
__m256 x0Val[num_a_vectors], x1Val[num_a_vectors], x0loVal[num_a_vectors], x0hiVal[num_a_vectors], x1loVal[num_a_vectors], x1hiVal[num_a_vectors];
|
|
__m256 c0Val[num_a_vectors], c1Val[num_a_vectors], c2Val[num_a_vectors], c3Val[num_a_vectors];
|
|
|
|
__m256 dotProdVal0[num_a_vectors];
|
|
__m256 dotProdVal1[num_a_vectors];
|
|
__m256 dotProdVal2[num_a_vectors];
|
|
__m256 dotProdVal3[num_a_vectors];
|
|
|
|
for (vec_ind = 0; vec_ind < num_a_vectors; vec_ind++)
|
|
{
|
|
dotProdVal0[vec_ind] = _mm256_setzero_ps();
|
|
dotProdVal1[vec_ind] = _mm256_setzero_ps();
|
|
dotProdVal2[vec_ind] = _mm256_setzero_ps();
|
|
dotProdVal3[vec_ind] = _mm256_setzero_ps();
|
|
}
|
|
|
|
// Set up the complex rotator
|
|
__m256 z0, z1, z2, z3;
|
|
__VOLK_ATTR_ALIGNED(32)
|
|
lv_32fc_t phase_vec[16];
|
|
for (vec_ind = 0; vec_ind < 16; ++vec_ind)
|
|
{
|
|
phase_vec[vec_ind] = _phase;
|
|
_phase *= phase_inc;
|
|
}
|
|
|
|
z0 = _mm256_load_ps((float*)phase_vec);
|
|
z1 = _mm256_load_ps((float*)(phase_vec + 4));
|
|
z2 = _mm256_load_ps((float*)(phase_vec + 8));
|
|
z3 = _mm256_load_ps((float*)(phase_vec + 12));
|
|
|
|
lv_32fc_t dz = phase_inc;
|
|
dz *= dz;
|
|
dz *= dz;
|
|
dz *= dz;
|
|
dz *= dz; // dz = phase_inc^16;
|
|
|
|
for (vec_ind = 0; vec_ind < 4; ++vec_ind)
|
|
{
|
|
phase_vec[vec_ind] = dz;
|
|
}
|
|
|
|
__m256 dz_reg = _mm256_load_ps((float*)phase_vec);
|
|
dz_reg = _mm256_complexnormalise_ps(dz_reg);
|
|
|
|
for (; number < sixteenthPoints; number++)
|
|
{
|
|
a0Val = _mm256_load_ps(aPtr);
|
|
a1Val = _mm256_load_ps(aPtr + 8);
|
|
a2Val = _mm256_load_ps(aPtr + 16);
|
|
a3Val = _mm256_load_ps(aPtr + 24);
|
|
|
|
a0Val = _mm256_complexmul_ps(a0Val, z0);
|
|
a1Val = _mm256_complexmul_ps(a1Val, z1);
|
|
a2Val = _mm256_complexmul_ps(a2Val, z2);
|
|
a3Val = _mm256_complexmul_ps(a3Val, z3);
|
|
|
|
z0 = _mm256_complexmul_ps(z0, dz_reg);
|
|
z1 = _mm256_complexmul_ps(z1, dz_reg);
|
|
z2 = _mm256_complexmul_ps(z2, dz_reg);
|
|
z3 = _mm256_complexmul_ps(z3, dz_reg);
|
|
|
|
for (vec_ind = 0; vec_ind < num_a_vectors; ++vec_ind)
|
|
{
|
|
x0Val[vec_ind] = _mm256_loadu_ps(bPtr[vec_ind]); // t0|t1|t2|t3|t4|t5|t6|t7
|
|
x1Val[vec_ind] = _mm256_loadu_ps(bPtr[vec_ind] + 8);
|
|
x0loVal[vec_ind] = _mm256_unpacklo_ps(x0Val[vec_ind], x0Val[vec_ind]); // t0|t0|t1|t1|t4|t4|t5|t5
|
|
x0hiVal[vec_ind] = _mm256_unpackhi_ps(x0Val[vec_ind], x0Val[vec_ind]); // t2|t2|t3|t3|t6|t6|t7|t7
|
|
x1loVal[vec_ind] = _mm256_unpacklo_ps(x1Val[vec_ind], x1Val[vec_ind]);
|
|
x1hiVal[vec_ind] = _mm256_unpackhi_ps(x1Val[vec_ind], x1Val[vec_ind]);
|
|
|
|
// TODO: it may be possible to rearrange swizzling to better pipeline data
|
|
b0Val[vec_ind] = _mm256_permute2f128_ps(x0loVal[vec_ind], x0hiVal[vec_ind], 0x20); // t0|t0|t1|t1|t2|t2|t3|t3
|
|
b1Val[vec_ind] = _mm256_permute2f128_ps(x0loVal[vec_ind], x0hiVal[vec_ind], 0x31); // t4|t4|t5|t5|t6|t6|t7|t7
|
|
b2Val[vec_ind] = _mm256_permute2f128_ps(x1loVal[vec_ind], x1hiVal[vec_ind], 0x20);
|
|
b3Val[vec_ind] = _mm256_permute2f128_ps(x1loVal[vec_ind], x1hiVal[vec_ind], 0x31);
|
|
|
|
c0Val[vec_ind] = _mm256_mul_ps(a0Val, b0Val[vec_ind]);
|
|
c1Val[vec_ind] = _mm256_mul_ps(a1Val, b1Val[vec_ind]);
|
|
c2Val[vec_ind] = _mm256_mul_ps(a2Val, b2Val[vec_ind]);
|
|
c3Val[vec_ind] = _mm256_mul_ps(a3Val, b3Val[vec_ind]);
|
|
|
|
dotProdVal0[vec_ind] = _mm256_add_ps(c0Val[vec_ind], dotProdVal0[vec_ind]);
|
|
dotProdVal1[vec_ind] = _mm256_add_ps(c1Val[vec_ind], dotProdVal1[vec_ind]);
|
|
dotProdVal2[vec_ind] = _mm256_add_ps(c2Val[vec_ind], dotProdVal2[vec_ind]);
|
|
dotProdVal3[vec_ind] = _mm256_add_ps(c3Val[vec_ind], dotProdVal3[vec_ind]);
|
|
|
|
bPtr[vec_ind] += 16;
|
|
}
|
|
|
|
// Force the rotators back onto the unit circle
|
|
if ((number % 64) == 0)
|
|
{
|
|
z0 = _mm256_complexnormalise_ps(z0);
|
|
z1 = _mm256_complexnormalise_ps(z1);
|
|
z2 = _mm256_complexnormalise_ps(z2);
|
|
z3 = _mm256_complexnormalise_ps(z3);
|
|
}
|
|
|
|
aPtr += 32;
|
|
}
|
|
__VOLK_ATTR_ALIGNED(32)
|
|
lv_32fc_t dotProductVector[4];
|
|
|
|
for (vec_ind = 0; vec_ind < num_a_vectors; ++vec_ind)
|
|
{
|
|
dotProdVal0[vec_ind] = _mm256_add_ps(dotProdVal0[vec_ind], dotProdVal1[vec_ind]);
|
|
dotProdVal0[vec_ind] = _mm256_add_ps(dotProdVal0[vec_ind], dotProdVal2[vec_ind]);
|
|
dotProdVal0[vec_ind] = _mm256_add_ps(dotProdVal0[vec_ind], dotProdVal3[vec_ind]);
|
|
|
|
_mm256_store_ps((float*)dotProductVector, dotProdVal0[vec_ind]); // Store the results back into the dot product vector
|
|
|
|
result[vec_ind] = lv_cmake(0.0f, 0.0f);
|
|
for (i = 0; i < 4; ++i)
|
|
{
|
|
result[vec_ind] += dotProductVector[i];
|
|
}
|
|
}
|
|
|
|
z0 = _mm256_complexnormalise_ps(z0);
|
|
_mm256_store_ps((float*)phase_vec, z0);
|
|
_phase = phase_vec[0];
|
|
|
|
number = sixteenthPoints * 16;
|
|
for (; number < num_points; number++)
|
|
{
|
|
wo = in_common[number] * _phase;
|
|
_phase *= phase_inc;
|
|
|
|
for (vec_ind = 0; vec_ind < num_a_vectors; ++vec_ind)
|
|
{
|
|
result[vec_ind] += wo * in_a[vec_ind][number];
|
|
}
|
|
}
|
|
|
|
*phase = _phase;
|
|
#else
|
|
volk_gnsssdr_32fc_32f_rotator_dot_prod_32fc_xn_generic_reload(result, in_common, phase_inc, phase, in_a, num_a_vectors, num_points);
|
|
#endif
|
|
}
|
|
|
|
|
|
#endif /* LV_HAVE_AVX */
|
|
|
|
#endif /* INCLUDED_volk_gnsssdr_32fc_32f_rotator_dot_prod_32fc_xn_H */
|