You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
54 lines
2.7 KiB
54 lines
2.7 KiB
/////////////////////////////////////////////////////////////////////////////////// |
|
// Copyright (C) 2026 Jon Beniston, M7RCE <jon@beniston.com> // |
|
// Some code by AI // |
|
// // |
|
// This program is free software; you can redistribute it and/or modify // |
|
// it under the terms of the GNU General Public License as published by // |
|
// the Free Software Foundation as version 3 of the License, or // |
|
// (at your option) any later version. // |
|
// // |
|
// This program is distributed in the hope that it will be useful, // |
|
// but WITHOUT ANY WARRANTY; without even the implied warranty of // |
|
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // |
|
// GNU General Public License V3 for more details. // |
|
// // |
|
// You should have received a copy of the GNU General Public License // |
|
// along with this program. If not, see <http://www.gnu.org/licenses/>. // |
|
/////////////////////////////////////////////////////////////////////////////////// |
|
|
|
// LDPC decoder for 32 codewords at a time, with AVX2. Compiled with AVX2 enabled, so only used if the CPU has it |
|
|
|
#include <immintrin.h> |
|
|
|
#include "dvbt2ldpckernel.h" |
|
|
|
namespace dvbt2 { |
|
namespace { |
|
|
|
struct VectorAVX2 |
|
{ |
|
typedef __m256i T; |
|
static const int W = 32; |
|
|
|
static T zero() { return _mm256_setzero_si256(); } |
|
static T set1(char x) { return _mm256_set1_epi8(x); } |
|
static T adds(T a, T b) { return _mm256_adds_epi8(a, b); } |
|
static T subs(T a, T b) { return _mm256_subs_epi8(a, b); } |
|
static T addsu(T a, T b) { return _mm256_adds_epu8(a, b); } |
|
static T abs(T a) { return _mm256_abs_epi8(a); } |
|
static T min(T a, T b) { return _mm256_min_epi8(a, b); } |
|
static T max(T a, T b) { return _mm256_max_epi8(a, b); } |
|
static T cmpgt(T a, T b) { return _mm256_cmpgt_epi8(a, b); } |
|
static T sign(T a, T b) { return _mm256_sign_epi8(a, b); } |
|
static T xor_(T a, T b) { return _mm256_xor_si256(a, b); } |
|
static T and_(T a, T b) { return _mm256_and_si256(a, b); } |
|
static T or_(T a, T b) { return _mm256_or_si256(a, b); } |
|
static void store(uint8_t *p, T a) { _mm256_storeu_si256(reinterpret_cast<T *>(p), a); } |
|
static uint64_t movemask(T a) { return (uint32_t) _mm256_movemask_epi8(a); } |
|
}; |
|
|
|
} // namespace |
|
|
|
const LDPCKernel ldpcKernelAVX2 = {VectorAVX2::W, decodeLDPC<VectorAVX2>}; |
|
|
|
} // namespace dvbt2
|
|
|