JinxRyu/Ryujinx.HLE/HOS/Services/Audio/AudioRendererManager/IAudioRenderer.cs

416 lines
14 KiB
C#
Raw Normal View History

Add a new JIT compiler for CPU code (#693) * Start of the ARMeilleure project * Refactoring around the old IRAdapter, now renamed to PreAllocator * Optimize the LowestBitSet method * Add CLZ support and fix CLS implementation * Add missing Equals and GetHashCode overrides on some structs, misc small tweaks * Implement the ByteSwap IR instruction, and some refactoring on the assembler * Implement the DivideUI IR instruction and fix 64-bits IDIV * Correct constant operand type on CSINC * Move division instructions implementation to InstEmitDiv * Fix destination type for the ConditionalSelect IR instruction * Implement UMULH and SMULH, with new IR instructions * Fix some issues with shift instructions * Fix constant types for BFM instructions * Fix up new tests using the new V128 struct * Update tests * Move DIV tests to a separate file * Add support for calls, and some instructions that depends on them * Start adding support for SIMD & FP types, along with some of the related ARM instructions * Fix some typos and the divide instruction with FP operands * Fix wrong method call on Clz_V * Implement ARM FP & SIMD move instructions, Saddlv_V, and misc. fixes * Implement SIMD logical instructions and more misc. fixes * Fix PSRAD x86 instruction encoding, TRN, UABD and UABDL implementations * Implement float conversion instruction, merge in LDj3SNuD fixes, and some other misc. fixes * Implement SIMD shift instruction and fix Dup_V * Add SCVTF and UCVTF (vector, fixed-point) variants to the opcode table * Fix check with tolerance on tester * Implement FP & SIMD comparison instructions, and some fixes * Update FCVT (Scalar) encoding on the table to support the Half-float variants * Support passing V128 structs, some cleanup on the register allocator, merge LDj3SNuD fixes * Use old memory access methods, made a start on SIMD memory insts support, some fixes * Fix float constant passed to functions, save and restore non-volatile XMM registers, other fixes * Fix arguments count with struct return values, other fixes * More instructions * Misc. fixes and integrate LDj3SNuD fixes * Update tests * Add a faster linear scan allocator, unwinding support on windows, and other changes * Update Ryujinx.HLE * Update Ryujinx.Graphics * Fix V128 return pointer passing, RCX is clobbered * Update Ryujinx.Tests * Update ITimeZoneService * Stop using GetFunctionPointer as that can't be called from native code, misc. fixes and tweaks * Use generic GetFunctionPointerForDelegate method and other tweaks * Some refactoring on the code generator, assert on invalid operations and use a separate enum for intrinsics * Remove some unused code on the assembler * Fix REX.W prefix regression on float conversion instructions, add some sort of profiler * Add hardware capability detection * Fix regression on Sha1h and revert Fcm** changes * Add SSE2-only paths on vector extract and insert, some refactoring on the pre-allocator * Fix silly mistake introduced on last commit on CpuId * Generate inline stack probes when the stack allocation is too large * Initial support for the System-V ABI * Support multiple destination operands * Fix SSE2 VectorInsert8 path, and other fixes * Change placement of XMM callee save and restore code to match other compilers * Rename Dest to Destination and Inst to Instruction * Fix a regression related to calls and the V128 type * Add an extra space on comments to match code style * Some refactoring * Fix vector insert FP32 SSE2 path * Port over the ARM32 instructions * Avoid memory protection races on JIT Cache * Another fix on VectorInsert FP32 (thanks to LDj3SNuD * Float operands don't need to use the same register when VEX is supported * Add a new register allocator, higher quality code for hot code (tier up), and other tweaks * Some nits, small improvements on the pre allocator * CpuThreadState is gone * Allow changing CPU emulators with a config entry * Add runtime identifiers on the ARMeilleure project * Allow switching between CPUs through a config entry (pt. 2) * Change win10-x64 to win-x64 on projects * Update the Ryujinx project to use ARMeilleure * Ensure that the selected register is valid on the hybrid allocator * Allow exiting on returns to 0 (should fix test regression) * Remove register assignments for most used variables on the hybrid allocator * Do not use fixed registers as spill temp * Add missing namespace and remove unneeded using * Address PR feedback * Fix types, etc * Enable AssumeStrictAbiCompliance by default * Ensure that Spill and Fill don't load or store any more than necessary
2019-08-08 18:56:22 +00:00
using ARMeilleure.Memory;
using Ryujinx.Audio;
using Ryujinx.Audio.Adpcm;
using Ryujinx.Common.Logging;
using Ryujinx.HLE.HOS.Ipc;
using Ryujinx.HLE.HOS.Kernel.Common;
using Ryujinx.HLE.HOS.Kernel.Threading;
using Ryujinx.HLE.Utilities;
using System;
using System.Runtime.InteropServices;
Implement libsoundio as an alternative audio backend (#406) * Audio: Implement libsoundio as an alternative audio backend libsoundio will be preferred over OpenAL if it is available on the machine. If neither are available, it will fallback to a dummy audio renderer that outputs no sound. * Audio: Fix SoundIoRingBuffer documentation * Audio: Unroll and optimize the audio write callback Copying one sample at a time is slow, this unrolls the most common audio channel layouts and manually copies the bytes between source and destination. This is over 2x faster than calling CopyBlockUnaligned every sample. * Audio: Optimize the write callback further This dramatically reduces the audio buffer copy time. When the sample size is one of handled sample sizes the buffer copy operation is almost 10x faster than CopyBlockAligned. This works by copying full samples at a time, rather than the individual bytes that make up the sample. This allows for 2x or 4x faster copy operations depending on sample size. * Audio: Fix typo in Stereo write callback * Audio: Fix Surround (5.1) audio write callback * Audio: Update Documentation * Audio: Use built-in Unsafe.SizeOf<T>() Built-in `SizeOf<T>()` is 10x faster than our `TypeSize<T>` helper. This also helps reduce code surface area. * Audio: Keep fixed buffer style consistent * Audio: Address styling nits * Audio: More style nits * Audio: Add additional documentation * Audio: Move libsoundio bindings internal As per discussion, moving the libsoundio native bindings into Ryujinx.Audio * Audio: Bump Target Framework back up to .NET Core 2.1 * Audio: Remove voice mixing optimizations. Leaves Saturation optimizations in place.
2018-11-15 02:22:50 +00:00
using System.Runtime.Intrinsics;
using System.Runtime.Intrinsics.X86;
namespace Ryujinx.HLE.HOS.Services.Audio.AudioRendererManager
{
class IAudioRenderer : IpcService, IDisposable
{
// This is the amount of samples that are going to be appended
// each time that RequestUpdateAudioRenderer is called. Ideally,
// this value shouldn't be neither too small (to avoid the player
// starving due to running out of samples) or too large (to avoid
// high latency).
private const int MixBufferSamplesCount = 960;
private KEvent _updateEvent;
Add a new JIT compiler for CPU code (#693) * Start of the ARMeilleure project * Refactoring around the old IRAdapter, now renamed to PreAllocator * Optimize the LowestBitSet method * Add CLZ support and fix CLS implementation * Add missing Equals and GetHashCode overrides on some structs, misc small tweaks * Implement the ByteSwap IR instruction, and some refactoring on the assembler * Implement the DivideUI IR instruction and fix 64-bits IDIV * Correct constant operand type on CSINC * Move division instructions implementation to InstEmitDiv * Fix destination type for the ConditionalSelect IR instruction * Implement UMULH and SMULH, with new IR instructions * Fix some issues with shift instructions * Fix constant types for BFM instructions * Fix up new tests using the new V128 struct * Update tests * Move DIV tests to a separate file * Add support for calls, and some instructions that depends on them * Start adding support for SIMD & FP types, along with some of the related ARM instructions * Fix some typos and the divide instruction with FP operands * Fix wrong method call on Clz_V * Implement ARM FP & SIMD move instructions, Saddlv_V, and misc. fixes * Implement SIMD logical instructions and more misc. fixes * Fix PSRAD x86 instruction encoding, TRN, UABD and UABDL implementations * Implement float conversion instruction, merge in LDj3SNuD fixes, and some other misc. fixes * Implement SIMD shift instruction and fix Dup_V * Add SCVTF and UCVTF (vector, fixed-point) variants to the opcode table * Fix check with tolerance on tester * Implement FP & SIMD comparison instructions, and some fixes * Update FCVT (Scalar) encoding on the table to support the Half-float variants * Support passing V128 structs, some cleanup on the register allocator, merge LDj3SNuD fixes * Use old memory access methods, made a start on SIMD memory insts support, some fixes * Fix float constant passed to functions, save and restore non-volatile XMM registers, other fixes * Fix arguments count with struct return values, other fixes * More instructions * Misc. fixes and integrate LDj3SNuD fixes * Update tests * Add a faster linear scan allocator, unwinding support on windows, and other changes * Update Ryujinx.HLE * Update Ryujinx.Graphics * Fix V128 return pointer passing, RCX is clobbered * Update Ryujinx.Tests * Update ITimeZoneService * Stop using GetFunctionPointer as that can't be called from native code, misc. fixes and tweaks * Use generic GetFunctionPointerForDelegate method and other tweaks * Some refactoring on the code generator, assert on invalid operations and use a separate enum for intrinsics * Remove some unused code on the assembler * Fix REX.W prefix regression on float conversion instructions, add some sort of profiler * Add hardware capability detection * Fix regression on Sha1h and revert Fcm** changes * Add SSE2-only paths on vector extract and insert, some refactoring on the pre-allocator * Fix silly mistake introduced on last commit on CpuId * Generate inline stack probes when the stack allocation is too large * Initial support for the System-V ABI * Support multiple destination operands * Fix SSE2 VectorInsert8 path, and other fixes * Change placement of XMM callee save and restore code to match other compilers * Rename Dest to Destination and Inst to Instruction * Fix a regression related to calls and the V128 type * Add an extra space on comments to match code style * Some refactoring * Fix vector insert FP32 SSE2 path * Port over the ARM32 instructions * Avoid memory protection races on JIT Cache * Another fix on VectorInsert FP32 (thanks to LDj3SNuD * Float operands don't need to use the same register when VEX is supported * Add a new register allocator, higher quality code for hot code (tier up), and other tweaks * Some nits, small improvements on the pre allocator * CpuThreadState is gone * Allow changing CPU emulators with a config entry * Add runtime identifiers on the ARMeilleure project * Allow switching between CPUs through a config entry (pt. 2) * Change win10-x64 to win-x64 on projects * Update the Ryujinx project to use ARMeilleure * Ensure that the selected register is valid on the hybrid allocator * Allow exiting on returns to 0 (should fix test regression) * Remove register assignments for most used variables on the hybrid allocator * Do not use fixed registers as spill temp * Add missing namespace and remove unneeded using * Address PR feedback * Fix types, etc * Enable AssumeStrictAbiCompliance by default * Ensure that Spill and Fill don't load or store any more than necessary
2019-08-08 18:56:22 +00:00
private IMemoryManager _memory;
private IAalOutput _audioOut;
private AudioRendererParameter _params;
private MemoryPoolContext[] _memoryPools;
private VoiceContext[] _voices;
private int _track;
private PlayState _playState;
public IAudioRenderer(
Horizon system,
Add a new JIT compiler for CPU code (#693) * Start of the ARMeilleure project * Refactoring around the old IRAdapter, now renamed to PreAllocator * Optimize the LowestBitSet method * Add CLZ support and fix CLS implementation * Add missing Equals and GetHashCode overrides on some structs, misc small tweaks * Implement the ByteSwap IR instruction, and some refactoring on the assembler * Implement the DivideUI IR instruction and fix 64-bits IDIV * Correct constant operand type on CSINC * Move division instructions implementation to InstEmitDiv * Fix destination type for the ConditionalSelect IR instruction * Implement UMULH and SMULH, with new IR instructions * Fix some issues with shift instructions * Fix constant types for BFM instructions * Fix up new tests using the new V128 struct * Update tests * Move DIV tests to a separate file * Add support for calls, and some instructions that depends on them * Start adding support for SIMD & FP types, along with some of the related ARM instructions * Fix some typos and the divide instruction with FP operands * Fix wrong method call on Clz_V * Implement ARM FP & SIMD move instructions, Saddlv_V, and misc. fixes * Implement SIMD logical instructions and more misc. fixes * Fix PSRAD x86 instruction encoding, TRN, UABD and UABDL implementations * Implement float conversion instruction, merge in LDj3SNuD fixes, and some other misc. fixes * Implement SIMD shift instruction and fix Dup_V * Add SCVTF and UCVTF (vector, fixed-point) variants to the opcode table * Fix check with tolerance on tester * Implement FP & SIMD comparison instructions, and some fixes * Update FCVT (Scalar) encoding on the table to support the Half-float variants * Support passing V128 structs, some cleanup on the register allocator, merge LDj3SNuD fixes * Use old memory access methods, made a start on SIMD memory insts support, some fixes * Fix float constant passed to functions, save and restore non-volatile XMM registers, other fixes * Fix arguments count with struct return values, other fixes * More instructions * Misc. fixes and integrate LDj3SNuD fixes * Update tests * Add a faster linear scan allocator, unwinding support on windows, and other changes * Update Ryujinx.HLE * Update Ryujinx.Graphics * Fix V128 return pointer passing, RCX is clobbered * Update Ryujinx.Tests * Update ITimeZoneService * Stop using GetFunctionPointer as that can't be called from native code, misc. fixes and tweaks * Use generic GetFunctionPointerForDelegate method and other tweaks * Some refactoring on the code generator, assert on invalid operations and use a separate enum for intrinsics * Remove some unused code on the assembler * Fix REX.W prefix regression on float conversion instructions, add some sort of profiler * Add hardware capability detection * Fix regression on Sha1h and revert Fcm** changes * Add SSE2-only paths on vector extract and insert, some refactoring on the pre-allocator * Fix silly mistake introduced on last commit on CpuId * Generate inline stack probes when the stack allocation is too large * Initial support for the System-V ABI * Support multiple destination operands * Fix SSE2 VectorInsert8 path, and other fixes * Change placement of XMM callee save and restore code to match other compilers * Rename Dest to Destination and Inst to Instruction * Fix a regression related to calls and the V128 type * Add an extra space on comments to match code style * Some refactoring * Fix vector insert FP32 SSE2 path * Port over the ARM32 instructions * Avoid memory protection races on JIT Cache * Another fix on VectorInsert FP32 (thanks to LDj3SNuD * Float operands don't need to use the same register when VEX is supported * Add a new register allocator, higher quality code for hot code (tier up), and other tweaks * Some nits, small improvements on the pre allocator * CpuThreadState is gone * Allow changing CPU emulators with a config entry * Add runtime identifiers on the ARMeilleure project * Allow switching between CPUs through a config entry (pt. 2) * Change win10-x64 to win-x64 on projects * Update the Ryujinx project to use ARMeilleure * Ensure that the selected register is valid on the hybrid allocator * Allow exiting on returns to 0 (should fix test regression) * Remove register assignments for most used variables on the hybrid allocator * Do not use fixed registers as spill temp * Add missing namespace and remove unneeded using * Address PR feedback * Fix types, etc * Enable AssumeStrictAbiCompliance by default * Ensure that Spill and Fill don't load or store any more than necessary
2019-08-08 18:56:22 +00:00
IMemoryManager memory,
IAalOutput audioOut,
AudioRendererParameter Params)
{
_updateEvent = new KEvent(system);
_memory = memory;
_audioOut = audioOut;
_params = Params;
_track = audioOut.OpenTrack(
AudioRendererConsts.HostSampleRate,
AudioRendererConsts.HostChannelsCount,
AudioCallback);
_memoryPools = CreateArray<MemoryPoolContext>(Params.EffectCount + Params.VoiceCount * 4);
_voices = CreateArray<VoiceContext>(Params.VoiceCount);
InitializeAudioOut();
_playState = PlayState.Stopped;
}
[Command(0)]
// GetSampleRate() -> u32
public ResultCode GetSampleRate(ServiceCtx context)
{
context.ResponseData.Write(_params.SampleRate);
return ResultCode.Success;
}
[Command(1)]
// GetSampleCount() -> u32
public ResultCode GetSampleCount(ServiceCtx context)
{
context.ResponseData.Write(_params.SampleCount);
return ResultCode.Success;
}
[Command(2)]
// GetMixBufferCount() -> u32
public ResultCode GetMixBufferCount(ServiceCtx context)
{
context.ResponseData.Write(_params.SubMixCount);
return ResultCode.Success;
}
[Command(3)]
// GetState() -> u32
public ResultCode GetState(ServiceCtx context)
{
context.ResponseData.Write((int)_playState);
Logger.PrintStub(LogClass.ServiceAudio, new { State = Enum.GetName(typeof(PlayState), _playState) });
return ResultCode.Success;
}
private void AudioCallback()
{
_updateEvent.ReadableEvent.Signal();
}
private static T[] CreateArray<T>(int size) where T : new()
{
T[] output = new T[size];
for (int index = 0; index < size; index++)
{
output[index] = new T();
}
return output;
}
private void InitializeAudioOut()
{
AppendMixedBuffer(0);
AppendMixedBuffer(1);
AppendMixedBuffer(2);
_audioOut.Start(_track);
}
[Command(4)]
// RequestUpdateAudioRenderer(buffer<nn::audio::detail::AudioRendererUpdateDataHeader, 5>)
// -> (buffer<nn::audio::detail::AudioRendererUpdateDataHeader, 6>, buffer<nn::audio::detail::AudioRendererUpdateDataHeader, 6>)
public ResultCode RequestUpdateAudioRenderer(ServiceCtx context)
{
long outputPosition = context.Request.ReceiveBuff[0].Position;
long outputSize = context.Request.ReceiveBuff[0].Size;
MemoryHelper.FillWithZeros(context.Memory, outputPosition, (int)outputSize);
long inputPosition = context.Request.SendBuff[0].Position;
StructReader reader = new StructReader(context.Memory, inputPosition);
StructWriter writer = new StructWriter(context.Memory, outputPosition);
UpdateDataHeader inputHeader = reader.Read<UpdateDataHeader>();
BehaviorInfo behaviorInfo = new BehaviorInfo();
behaviorInfo.SetUserLibRevision(inputHeader.Revision);
reader.Read<BehaviorIn>(inputHeader.BehaviorSize);
MemoryPoolIn[] memoryPoolsIn = reader.Read<MemoryPoolIn>(inputHeader.MemoryPoolSize);
for (int index = 0; index < memoryPoolsIn.Length; index++)
{
MemoryPoolIn memoryPool = memoryPoolsIn[index];
if (memoryPool.State == MemoryPoolState.RequestAttach)
{
_memoryPools[index].OutStatus.State = MemoryPoolState.Attached;
}
else if (memoryPool.State == MemoryPoolState.RequestDetach)
{
_memoryPools[index].OutStatus.State = MemoryPoolState.Detached;
}
}
reader.Read<VoiceChannelResourceIn>(inputHeader.VoiceResourceSize);
VoiceIn[] voicesIn = reader.Read<VoiceIn>(inputHeader.VoiceSize);
for (int index = 0; index < voicesIn.Length; index++)
{
VoiceIn voice = voicesIn[index];
VoiceContext voiceCtx = _voices[index];
voiceCtx.SetAcquireState(voice.Acquired != 0);
if (voice.Acquired == 0)
{
continue;
}
if (voice.FirstUpdate != 0)
{
voiceCtx.AdpcmCtx = GetAdpcmDecoderContext(
voice.AdpcmCoeffsPosition,
voice.AdpcmCoeffsSize);
voiceCtx.SampleFormat = voice.SampleFormat;
voiceCtx.SampleRate = voice.SampleRate;
voiceCtx.ChannelsCount = voice.ChannelsCount;
voiceCtx.SetBufferIndex(voice.BaseWaveBufferIndex);
}
voiceCtx.WaveBuffers[0] = voice.WaveBuffer0;
voiceCtx.WaveBuffers[1] = voice.WaveBuffer1;
voiceCtx.WaveBuffers[2] = voice.WaveBuffer2;
voiceCtx.WaveBuffers[3] = voice.WaveBuffer3;
voiceCtx.Volume = voice.Volume;
voiceCtx.PlayState = voice.PlayState;
}
UpdateAudio();
UpdateDataHeader outputHeader = new UpdateDataHeader();
int updateHeaderSize = Marshal.SizeOf<UpdateDataHeader>();
outputHeader.Revision = AudioRendererConsts.RevMagic;
outputHeader.BehaviorSize = 0xb0;
outputHeader.MemoryPoolSize = (_params.EffectCount + _params.VoiceCount * 4) * 0x10;
outputHeader.VoiceSize = _params.VoiceCount * 0x10;
outputHeader.EffectSize = _params.EffectCount * 0x10;
outputHeader.SinkSize = _params.SinkCount * 0x20;
outputHeader.PerformanceManagerSize = 0x10;
if (behaviorInfo.IsElapsedFrameCountSupported())
{
outputHeader.ElapsedFrameCountInfoSize = 0x10;
}
outputHeader.TotalSize = updateHeaderSize +
outputHeader.BehaviorSize +
outputHeader.MemoryPoolSize +
outputHeader.VoiceSize +
outputHeader.EffectSize +
outputHeader.SinkSize +
outputHeader.PerformanceManagerSize +
outputHeader.ElapsedFrameCountInfoSize;
writer.Write(outputHeader);
foreach (MemoryPoolContext memoryPool in _memoryPools)
{
writer.Write(memoryPool.OutStatus);
}
foreach (VoiceContext voice in _voices)
{
writer.Write(voice.OutStatus);
}
return ResultCode.Success;
}
[Command(5)]
// Start()
public ResultCode StartAudioRenderer(ServiceCtx context)
{
Logger.PrintStub(LogClass.ServiceAudio);
_playState = PlayState.Playing;
return ResultCode.Success;
}
[Command(6)]
// Stop()
public ResultCode StopAudioRenderer(ServiceCtx context)
{
Logger.PrintStub(LogClass.ServiceAudio);
_playState = PlayState.Stopped;
return ResultCode.Success;
}
[Command(7)]
// QuerySystemEvent() -> handle<copy, event>
public ResultCode QuerySystemEvent(ServiceCtx context)
{
if (context.Process.HandleTable.GenerateHandle(_updateEvent.ReadableEvent, out int handle) != KernelResult.Success)
{
throw new InvalidOperationException("Out of handles!");
}
context.Response.HandleDesc = IpcHandleDesc.MakeCopy(handle);
return ResultCode.Success;
}
private AdpcmDecoderContext GetAdpcmDecoderContext(long position, long size)
{
if (size == 0)
{
return null;
}
AdpcmDecoderContext context = new AdpcmDecoderContext
{
Coefficients = new short[size >> 1]
};
for (int offset = 0; offset < size; offset += 2)
{
context.Coefficients[offset >> 1] = _memory.ReadInt16(position + offset);
}
return context;
}
private void UpdateAudio()
{
long[] released = _audioOut.GetReleasedBuffers(_track, 2);
for (int index = 0; index < released.Length; index++)
{
AppendMixedBuffer(released[index]);
}
}
private void AppendMixedBuffer(long tag)
{
int[] mixBuffer = new int[MixBufferSamplesCount * AudioRendererConsts.HostChannelsCount];
foreach (VoiceContext voice in _voices)
{
if (!voice.Playing || voice.CurrentWaveBuffer.Size == 0)
{
continue;
}
int outOffset = 0;
int pendingSamples = MixBufferSamplesCount;
float volume = voice.Volume;
while (pendingSamples > 0)
{
int[] samples = voice.GetBufferData(_memory, pendingSamples, out int returnedSamples);
if (returnedSamples == 0)
{
break;
}
pendingSamples -= returnedSamples;
for (int offset = 0; offset < samples.Length; offset++)
{
mixBuffer[outOffset++] += (int)(samples[offset] * voice.Volume);
}
}
}
_audioOut.AppendBuffer(_track, tag, GetFinalBuffer(mixBuffer));
}
private unsafe static short[] GetFinalBuffer(int[] buffer)
{
short[] output = new short[buffer.Length];
int offset = 0;
Implement libsoundio as an alternative audio backend (#406) * Audio: Implement libsoundio as an alternative audio backend libsoundio will be preferred over OpenAL if it is available on the machine. If neither are available, it will fallback to a dummy audio renderer that outputs no sound. * Audio: Fix SoundIoRingBuffer documentation * Audio: Unroll and optimize the audio write callback Copying one sample at a time is slow, this unrolls the most common audio channel layouts and manually copies the bytes between source and destination. This is over 2x faster than calling CopyBlockUnaligned every sample. * Audio: Optimize the write callback further This dramatically reduces the audio buffer copy time. When the sample size is one of handled sample sizes the buffer copy operation is almost 10x faster than CopyBlockAligned. This works by copying full samples at a time, rather than the individual bytes that make up the sample. This allows for 2x or 4x faster copy operations depending on sample size. * Audio: Fix typo in Stereo write callback * Audio: Fix Surround (5.1) audio write callback * Audio: Update Documentation * Audio: Use built-in Unsafe.SizeOf<T>() Built-in `SizeOf<T>()` is 10x faster than our `TypeSize<T>` helper. This also helps reduce code surface area. * Audio: Keep fixed buffer style consistent * Audio: Address styling nits * Audio: More style nits * Audio: Add additional documentation * Audio: Move libsoundio bindings internal As per discussion, moving the libsoundio native bindings into Ryujinx.Audio * Audio: Bump Target Framework back up to .NET Core 2.1 * Audio: Remove voice mixing optimizations. Leaves Saturation optimizations in place.
2018-11-15 02:22:50 +00:00
// Perform Saturation using SSE2 if supported
if (Sse2.IsSupported)
{
fixed (int* inptr = buffer)
fixed (short* outptr = output)
Implement libsoundio as an alternative audio backend (#406) * Audio: Implement libsoundio as an alternative audio backend libsoundio will be preferred over OpenAL if it is available on the machine. If neither are available, it will fallback to a dummy audio renderer that outputs no sound. * Audio: Fix SoundIoRingBuffer documentation * Audio: Unroll and optimize the audio write callback Copying one sample at a time is slow, this unrolls the most common audio channel layouts and manually copies the bytes between source and destination. This is over 2x faster than calling CopyBlockUnaligned every sample. * Audio: Optimize the write callback further This dramatically reduces the audio buffer copy time. When the sample size is one of handled sample sizes the buffer copy operation is almost 10x faster than CopyBlockAligned. This works by copying full samples at a time, rather than the individual bytes that make up the sample. This allows for 2x or 4x faster copy operations depending on sample size. * Audio: Fix typo in Stereo write callback * Audio: Fix Surround (5.1) audio write callback * Audio: Update Documentation * Audio: Use built-in Unsafe.SizeOf<T>() Built-in `SizeOf<T>()` is 10x faster than our `TypeSize<T>` helper. This also helps reduce code surface area. * Audio: Keep fixed buffer style consistent * Audio: Address styling nits * Audio: More style nits * Audio: Add additional documentation * Audio: Move libsoundio bindings internal As per discussion, moving the libsoundio native bindings into Ryujinx.Audio * Audio: Bump Target Framework back up to .NET Core 2.1 * Audio: Remove voice mixing optimizations. Leaves Saturation optimizations in place.
2018-11-15 02:22:50 +00:00
{
for (; offset + 32 <= buffer.Length; offset += 32)
Implement libsoundio as an alternative audio backend (#406) * Audio: Implement libsoundio as an alternative audio backend libsoundio will be preferred over OpenAL if it is available on the machine. If neither are available, it will fallback to a dummy audio renderer that outputs no sound. * Audio: Fix SoundIoRingBuffer documentation * Audio: Unroll and optimize the audio write callback Copying one sample at a time is slow, this unrolls the most common audio channel layouts and manually copies the bytes between source and destination. This is over 2x faster than calling CopyBlockUnaligned every sample. * Audio: Optimize the write callback further This dramatically reduces the audio buffer copy time. When the sample size is one of handled sample sizes the buffer copy operation is almost 10x faster than CopyBlockAligned. This works by copying full samples at a time, rather than the individual bytes that make up the sample. This allows for 2x or 4x faster copy operations depending on sample size. * Audio: Fix typo in Stereo write callback * Audio: Fix Surround (5.1) audio write callback * Audio: Update Documentation * Audio: Use built-in Unsafe.SizeOf<T>() Built-in `SizeOf<T>()` is 10x faster than our `TypeSize<T>` helper. This also helps reduce code surface area. * Audio: Keep fixed buffer style consistent * Audio: Address styling nits * Audio: More style nits * Audio: Add additional documentation * Audio: Move libsoundio bindings internal As per discussion, moving the libsoundio native bindings into Ryujinx.Audio * Audio: Bump Target Framework back up to .NET Core 2.1 * Audio: Remove voice mixing optimizations. Leaves Saturation optimizations in place.
2018-11-15 02:22:50 +00:00
{
// Unroll the loop a little to ensure the CPU pipeline
// is always full.
Vector128<int> block1A = Sse2.LoadVector128(inptr + offset + 0);
Vector128<int> block1B = Sse2.LoadVector128(inptr + offset + 4);
Implement libsoundio as an alternative audio backend (#406) * Audio: Implement libsoundio as an alternative audio backend libsoundio will be preferred over OpenAL if it is available on the machine. If neither are available, it will fallback to a dummy audio renderer that outputs no sound. * Audio: Fix SoundIoRingBuffer documentation * Audio: Unroll and optimize the audio write callback Copying one sample at a time is slow, this unrolls the most common audio channel layouts and manually copies the bytes between source and destination. This is over 2x faster than calling CopyBlockUnaligned every sample. * Audio: Optimize the write callback further This dramatically reduces the audio buffer copy time. When the sample size is one of handled sample sizes the buffer copy operation is almost 10x faster than CopyBlockAligned. This works by copying full samples at a time, rather than the individual bytes that make up the sample. This allows for 2x or 4x faster copy operations depending on sample size. * Audio: Fix typo in Stereo write callback * Audio: Fix Surround (5.1) audio write callback * Audio: Update Documentation * Audio: Use built-in Unsafe.SizeOf<T>() Built-in `SizeOf<T>()` is 10x faster than our `TypeSize<T>` helper. This also helps reduce code surface area. * Audio: Keep fixed buffer style consistent * Audio: Address styling nits * Audio: More style nits * Audio: Add additional documentation * Audio: Move libsoundio bindings internal As per discussion, moving the libsoundio native bindings into Ryujinx.Audio * Audio: Bump Target Framework back up to .NET Core 2.1 * Audio: Remove voice mixing optimizations. Leaves Saturation optimizations in place.
2018-11-15 02:22:50 +00:00
Vector128<int> block2A = Sse2.LoadVector128(inptr + offset + 8);
Vector128<int> block2B = Sse2.LoadVector128(inptr + offset + 12);
Implement libsoundio as an alternative audio backend (#406) * Audio: Implement libsoundio as an alternative audio backend libsoundio will be preferred over OpenAL if it is available on the machine. If neither are available, it will fallback to a dummy audio renderer that outputs no sound. * Audio: Fix SoundIoRingBuffer documentation * Audio: Unroll and optimize the audio write callback Copying one sample at a time is slow, this unrolls the most common audio channel layouts and manually copies the bytes between source and destination. This is over 2x faster than calling CopyBlockUnaligned every sample. * Audio: Optimize the write callback further This dramatically reduces the audio buffer copy time. When the sample size is one of handled sample sizes the buffer copy operation is almost 10x faster than CopyBlockAligned. This works by copying full samples at a time, rather than the individual bytes that make up the sample. This allows for 2x or 4x faster copy operations depending on sample size. * Audio: Fix typo in Stereo write callback * Audio: Fix Surround (5.1) audio write callback * Audio: Update Documentation * Audio: Use built-in Unsafe.SizeOf<T>() Built-in `SizeOf<T>()` is 10x faster than our `TypeSize<T>` helper. This also helps reduce code surface area. * Audio: Keep fixed buffer style consistent * Audio: Address styling nits * Audio: More style nits * Audio: Add additional documentation * Audio: Move libsoundio bindings internal As per discussion, moving the libsoundio native bindings into Ryujinx.Audio * Audio: Bump Target Framework back up to .NET Core 2.1 * Audio: Remove voice mixing optimizations. Leaves Saturation optimizations in place.
2018-11-15 02:22:50 +00:00
Vector128<int> block3A = Sse2.LoadVector128(inptr + offset + 16);
Vector128<int> block3B = Sse2.LoadVector128(inptr + offset + 20);
Implement libsoundio as an alternative audio backend (#406) * Audio: Implement libsoundio as an alternative audio backend libsoundio will be preferred over OpenAL if it is available on the machine. If neither are available, it will fallback to a dummy audio renderer that outputs no sound. * Audio: Fix SoundIoRingBuffer documentation * Audio: Unroll and optimize the audio write callback Copying one sample at a time is slow, this unrolls the most common audio channel layouts and manually copies the bytes between source and destination. This is over 2x faster than calling CopyBlockUnaligned every sample. * Audio: Optimize the write callback further This dramatically reduces the audio buffer copy time. When the sample size is one of handled sample sizes the buffer copy operation is almost 10x faster than CopyBlockAligned. This works by copying full samples at a time, rather than the individual bytes that make up the sample. This allows for 2x or 4x faster copy operations depending on sample size. * Audio: Fix typo in Stereo write callback * Audio: Fix Surround (5.1) audio write callback * Audio: Update Documentation * Audio: Use built-in Unsafe.SizeOf<T>() Built-in `SizeOf<T>()` is 10x faster than our `TypeSize<T>` helper. This also helps reduce code surface area. * Audio: Keep fixed buffer style consistent * Audio: Address styling nits * Audio: More style nits * Audio: Add additional documentation * Audio: Move libsoundio bindings internal As per discussion, moving the libsoundio native bindings into Ryujinx.Audio * Audio: Bump Target Framework back up to .NET Core 2.1 * Audio: Remove voice mixing optimizations. Leaves Saturation optimizations in place.
2018-11-15 02:22:50 +00:00
Vector128<int> block4A = Sse2.LoadVector128(inptr + offset + 24);
Vector128<int> block4B = Sse2.LoadVector128(inptr + offset + 28);
Implement libsoundio as an alternative audio backend (#406) * Audio: Implement libsoundio as an alternative audio backend libsoundio will be preferred over OpenAL if it is available on the machine. If neither are available, it will fallback to a dummy audio renderer that outputs no sound. * Audio: Fix SoundIoRingBuffer documentation * Audio: Unroll and optimize the audio write callback Copying one sample at a time is slow, this unrolls the most common audio channel layouts and manually copies the bytes between source and destination. This is over 2x faster than calling CopyBlockUnaligned every sample. * Audio: Optimize the write callback further This dramatically reduces the audio buffer copy time. When the sample size is one of handled sample sizes the buffer copy operation is almost 10x faster than CopyBlockAligned. This works by copying full samples at a time, rather than the individual bytes that make up the sample. This allows for 2x or 4x faster copy operations depending on sample size. * Audio: Fix typo in Stereo write callback * Audio: Fix Surround (5.1) audio write callback * Audio: Update Documentation * Audio: Use built-in Unsafe.SizeOf<T>() Built-in `SizeOf<T>()` is 10x faster than our `TypeSize<T>` helper. This also helps reduce code surface area. * Audio: Keep fixed buffer style consistent * Audio: Address styling nits * Audio: More style nits * Audio: Add additional documentation * Audio: Move libsoundio bindings internal As per discussion, moving the libsoundio native bindings into Ryujinx.Audio * Audio: Bump Target Framework back up to .NET Core 2.1 * Audio: Remove voice mixing optimizations. Leaves Saturation optimizations in place.
2018-11-15 02:22:50 +00:00
Vector128<short> output1 = Sse2.PackSignedSaturate(block1A, block1B);
Vector128<short> output2 = Sse2.PackSignedSaturate(block2A, block2B);
Vector128<short> output3 = Sse2.PackSignedSaturate(block3A, block3B);
Vector128<short> output4 = Sse2.PackSignedSaturate(block4A, block4B);
Sse2.Store(outptr + offset + 0, output1);
Sse2.Store(outptr + offset + 8, output2);
Sse2.Store(outptr + offset + 16, output3);
Sse2.Store(outptr + offset + 24, output4);
Implement libsoundio as an alternative audio backend (#406) * Audio: Implement libsoundio as an alternative audio backend libsoundio will be preferred over OpenAL if it is available on the machine. If neither are available, it will fallback to a dummy audio renderer that outputs no sound. * Audio: Fix SoundIoRingBuffer documentation * Audio: Unroll and optimize the audio write callback Copying one sample at a time is slow, this unrolls the most common audio channel layouts and manually copies the bytes between source and destination. This is over 2x faster than calling CopyBlockUnaligned every sample. * Audio: Optimize the write callback further This dramatically reduces the audio buffer copy time. When the sample size is one of handled sample sizes the buffer copy operation is almost 10x faster than CopyBlockAligned. This works by copying full samples at a time, rather than the individual bytes that make up the sample. This allows for 2x or 4x faster copy operations depending on sample size. * Audio: Fix typo in Stereo write callback * Audio: Fix Surround (5.1) audio write callback * Audio: Update Documentation * Audio: Use built-in Unsafe.SizeOf<T>() Built-in `SizeOf<T>()` is 10x faster than our `TypeSize<T>` helper. This also helps reduce code surface area. * Audio: Keep fixed buffer style consistent * Audio: Address styling nits * Audio: More style nits * Audio: Add additional documentation * Audio: Move libsoundio bindings internal As per discussion, moving the libsoundio native bindings into Ryujinx.Audio * Audio: Bump Target Framework back up to .NET Core 2.1 * Audio: Remove voice mixing optimizations. Leaves Saturation optimizations in place.
2018-11-15 02:22:50 +00:00
}
}
}
// Process left overs
for (; offset < buffer.Length; offset++)
{
output[offset] = DspUtils.Saturate(buffer[offset]);
}
return output;
}
public void Dispose()
{
Dispose(true);
}
protected virtual void Dispose(bool disposing)
{
if (disposing)
{
_audioOut.CloseTrack(_track);
}
}
}
}