ppsspp/Core/Util/AudioFormat.cpp
Henrik Rydgård a68ddd0a8d Merge separate NEON functions into the normal functions.
We no longer support non-NEON ARM.

It's nice also to have the NEON and SSE implementations "close" to each
other, easier to port optimizations back and forth etc.
2022-04-12 23:43:21 +02:00

154 lines
5.3 KiB
C++

// Copyright (c) 2012- PPSSPP Project.
// 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, version 2.0 or later versions.
// 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 2.0 for more details.
// A copy of the GPL 2.0 should have been included with the program.
// If not, see http://www.gnu.org/licenses/
// Official git repository and contact information can be found at
// https://github.com/hrydgard/ppsspp and http://www.ppsspp.org/.
#include "ppsspp_config.h"
#include "Common/Common.h"
#include "Common/CPUDetect.h"
#include "Core/Util/AudioFormat.h"
#ifdef _M_SSE
#include <emmintrin.h>
#endif
#if PPSSPP_ARCH(ARM_NEON)
#if defined(_MSC_VER) && PPSSPP_ARCH(ARM64)
#include <arm64_neon.h>
#else
#include <arm_neon.h>
#endif
#endif // PPSSPP_ARCH(ARM_NEON)
// TODO: This shouldn't be a global.
#if PPSSPP_ARCH(ARM_NEON)
alignas(16) static s16 volumeValues[4] = {};
#endif
void AdjustVolumeBlock(s16 *out, s16 *in, size_t size, int leftVol, int rightVol) {
#ifdef _M_SSE
if (leftVol <= 0x7fff && -leftVol <= 0x8000 && rightVol <= 0x7fff && -rightVol <= 0x8000) {
__m128i volume = _mm_set_epi16(leftVol, rightVol, leftVol, rightVol, leftVol, rightVol, leftVol, rightVol);
while (size >= 16) {
__m128i indata1 = _mm_loadu_si128((__m128i *)in);
__m128i indata2 = _mm_loadu_si128((__m128i *)(in + 8));
_mm_storeu_si128((__m128i *)out, _mm_mulhi_epi16(indata1, volume));
_mm_storeu_si128((__m128i *)(out + 8), _mm_mulhi_epi16(indata2, volume));
in += 16;
out += 16;
size -= 16;
}
}
#if 0
// This path wraps instead of clamps in _mm_slli_epi16. Needs fixing.
else {
// We have to shift inside the loop to avoid the signed 16-bit multiply issue.
int leftShift = 0;
int leftVol16 = leftVol;
while (leftVol16 > 0x7fff || -leftVol16 > 0x8000) {
++leftShift;
leftVol16 >>= 1;
}
int rightShift = 0;
int rightVol16 = rightVol;
while (rightVol16 > 0x7fff || -rightVol16 > 0x8000) {
++rightShift;
rightVol16 >>= 1;
}
__m128i volume = _mm_set_epi16(leftVol16, rightVol16, leftVol16, rightVol16, leftVol16, rightVol16, leftVol16, rightVol16);
while (size >= 16) {
__m128i indata1 = _mm_loadu_si128((__m128i *)in);
__m128i indata2 = _mm_loadu_si128((__m128i *)(in + 8));
_mm_storeu_si128((__m128i *)out, _mm_slli_epi16(_mm_mulhi_epi16(indata1, volume), leftShift));
_mm_storeu_si128((__m128i *)(out + 8), _mm_slli_epi16(_mm_mulhi_epi16(indata2, volume), rightShift));
in += 16;
out += 16;
size -= 16;
}
}
#endif
#elif PPSSPP_ARCH(ARM_NEON)
if (leftVol <= 0xFFFF && -leftVol <= 0x10000 && rightVol <= 0xFFFF && -rightVol <= 0x10000) {
// Note: vqshrn_n_s32 takes a const argument, so we always go with 1 here, 15 there.
volumeValues[0] = leftVol >> 1;
volumeValues[1] = rightVol >> 1;
volumeValues[2] = leftVol >> 1;
volumeValues[3] = rightVol >> 1;
const int16x4_t vol = vld1_s16(volumeValues);
while (size >= 16) {
int16x8_t indata1 = vld1q_s16(in);
int16x8_t indata2 = vld1q_s16(in + 8);
int32x4_t outh1 = vmull_s16(vget_high_s16(indata1), vol);
int32x4_t outh2 = vmull_s16(vget_high_s16(indata2), vol);
int32x4_t outl1 = vmull_s16(vget_low_s16(indata1), vol);
int32x4_t outl2 = vmull_s16(vget_low_s16(indata2), vol);
int16x8_t outdata1 = vcombine_s16(vqshrn_n_s32(outl1, 15), vqshrn_n_s32(outh1, 15));
int16x8_t outdata2 = vcombine_s16(vqshrn_n_s32(outl2, 15), vqshrn_n_s32(outh2, 15));
vst1q_s16(out, outdata1);
vst1q_s16(out + 8, outdata2);
in += 16;
out += 16;
size -= 16;
}
}
#endif
if (leftVol <= 0x7fff && -leftVol <= 0x8000 && rightVol <= 0x7fff && -rightVol <= 0x8000) {
for (size_t i = 0; i < size; i += 2) {
out[i] = ApplySampleVolume(in[i], leftVol);
out[i + 1] = ApplySampleVolume(in[i + 1], rightVol);
}
} else {
for (size_t i = 0; i < size; i += 2) {
out[i] = ApplySampleVolume20Bit(in[i], leftVol);
out[i + 1] = ApplySampleVolume20Bit(in[i + 1], rightVol);
}
}
}
void ConvertS16ToF32(float *out, const s16 *in, size_t size) {
#ifdef _M_SSE
const __m128i zero = _mm_setzero_si128();
const __m128 scale = _mm_set_ps1(1.0f / 32767.0f);
while (size >= 16) {
__m128i indata1 = _mm_loadu_si128((__m128i *)in + 0);
__m128i indata2 = _mm_loadu_si128((__m128i *)in + 1);
// Now we unpack with "sign extension", by unpacking with 0xFFFF if zero is greater.
__m128i insign1 = _mm_cmpgt_epi16(zero, indata1);
__m128i insign2 = _mm_cmpgt_epi16(zero, indata2);
__m128i indata1lo = _mm_unpacklo_epi16(indata1, insign1);
__m128i indata1hi = _mm_unpackhi_epi16(indata1, insign1);
__m128i indata2lo = _mm_unpacklo_epi16(indata2, insign2);
__m128i indata2hi = _mm_unpackhi_epi16(indata2, insign2);
_mm_storeu_ps(out + 0, _mm_mul_ps(_mm_cvtepi32_ps(indata1lo), scale));
_mm_storeu_ps(out + 4, _mm_mul_ps(_mm_cvtepi32_ps(indata1hi), scale));
_mm_storeu_ps(out + 8, _mm_mul_ps(_mm_cvtepi32_ps(indata2lo), scale));
_mm_storeu_ps(out + 12, _mm_mul_ps(_mm_cvtepi32_ps(indata2hi), scale));
in += 16;
out += 16;
size -= 16;
}
#endif
for (size_t i = 0; i < size; i++) {
out[i] = in[i] * (1.0f / 32767.0f);
}
}