AnotherFoxGuy/webrtc-audio-processing
1
0
Fork 0
Code Pull Requests Activity

Bump to WebRTC M120 release

Browse Source 
Some API deprecation -- ExperimentalAgc and ExperimentalNs are gone.
We're continuing to carry iSAC even though it's gone upstream, but maybe
we'll want to drop that soon.
This commit is contained in:
Arun Raghavan
2023-12-12 10:42:58 -05:00
parent 9a202fb8c2
commit c6abf6cd3f
479 changed files with 20900 additions and 11996 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
webrtc/common_audio/resampler/sinc_resampler.cc
View File

@ -80,7 +80,7 @@
// 8) Else, if we're not on the second load, goto (4).
//
// Note: we're glossing over how the sub-sample handling works with
// |virtual_source_idx_|, etc.
// `virtual_source_idx_`, etc.
// MSVC++ requires this to be set before any other includes to get M_PI.
#define _USE_MATH_DEFINES
@ -102,7 +102,7 @@ namespace webrtc {
namespace {
double SincScaleFactor(double io_ratio) {
// |sinc_scale_factor| is basically the normalized cutoff frequency of the
// `sinc_scale_factor` is basically the normalized cutoff frequency of the
// low-pass filter.
double sinc_scale_factor = io_ratio > 1.0 ? 1.0 / io_ratio : 1.0;
@ -126,8 +126,8 @@ void SincResampler::InitializeCPUSpecificFeatures() {
#if defined(WEBRTC_HAS_NEON)
convolve_proc_ = Convolve_NEON;
#elif defined(WEBRTC_ARCH_X86_FAMILY)
// Using AVX2 instead of SSE2 when AVX2 supported.
if (GetCPUInfo(kAVX2))
// Using AVX2 instead of SSE2 when AVX2/FMA3 supported.
if (GetCPUInfo(kAVX2) && GetCPUInfo(kFMA3))
convolve_proc_ = Convolve_AVX2;
else if (GetCPUInfo(kSSE2))
convolve_proc_ = Convolve_SSE;
@ -238,7 +238,7 @@ void SincResampler::SetRatio(double io_sample_rate_ratio) {
io_sample_rate_ratio_ = io_sample_rate_ratio;
// Optimize reinitialization by reusing values which are independent of
// |sinc_scale_factor|. Provides a 3x speedup.
// `sinc_scale_factor`. Provides a 3x speedup.
const double sinc_scale_factor = SincScaleFactor(io_sample_rate_ratio_);
for (size_t offset_idx = 0; offset_idx <= kKernelOffsetCount; ++offset_idx) {
for (size_t i = 0; i < kKernelSize; ++i) {
@ -268,8 +268,8 @@ void SincResampler::Resample(size_t frames, float* destination) {
const double current_io_ratio = io_sample_rate_ratio_;
const float* const kernel_ptr = kernel_storage_.get();
while (remaining_frames) {
// |i| may be negative if the last Resample() call ended on an iteration
// that put |virtual_source_idx_| over the limit.
// `i` may be negative if the last Resample() call ended on an iteration
// that put `virtual_source_idx_` over the limit.
//
// Note: The loop construct here can severely impact performance on ARM
// or when built with clang. See https://codereview.chromium.org/18566009/
@ -278,7 +278,7 @@ void SincResampler::Resample(size_t frames, float* destination) {
i > 0; --i) {
RTC_DCHECK_LT(virtual_source_idx_, block_size_);
// |virtual_source_idx_| lies in between two kernel offsets so figure out
// `virtual_source_idx_` lies in between two kernel offsets so figure out
// what they are.
const int source_idx = static_cast<int>(virtual_source_idx_);
const double subsample_remainder = virtual_source_idx_ - source_idx;
@ -288,16 +288,16 @@ void SincResampler::Resample(size_t frames, float* destination) {
const int offset_idx = static_cast<int>(virtual_offset_idx);
// We'll compute "convolutions" for the two kernels which straddle
// |virtual_source_idx_|.
// `virtual_source_idx_`.
const float* const k1 = kernel_ptr + offset_idx * kKernelSize;
const float* const k2 = k1 + kKernelSize;
// Ensure |k1|, |k2| are 32-byte aligned for SIMD usage. Should always be
// Ensure `k1`, `k2` are 32-byte aligned for SIMD usage. Should always be
// true so long as kKernelSize is a multiple of 32.
RTC_DCHECK_EQ(0, reinterpret_cast<uintptr_t>(k1) % 32);
RTC_DCHECK_EQ(0, reinterpret_cast<uintptr_t>(k2) % 32);
// Initialize input pointer based on quantized |virtual_source_idx_|.
// Initialize input pointer based on quantized `virtual_source_idx_`.
const float* const input_ptr = r1_ + source_idx;
// Figure out how much to weight each kernel's "convolution".