CPU: new framework for optimized CPU-specific routines

The new CPU framework replaces the old BasicOps framework. It is more
flexible and the build process isn't such a mess anymore (pre-compiled
assembler files etc.). It will hopefully see some improvements and
extensions soon.

Signed-off-by: Tobias Doerffel <tobias.doerffel@gmail.com>

include/Cpu.h

@@ -1,8 +1,8 @@
 /*
- * basic_ops.h - basic memory operations
+ * Cpu.h - CPU specific accellerated operations
+ *
+ * Copyright (c) 2008-2009 Tobias Doerffel <tobydox/at/users.sourceforge.net>
  *
- * Copyright (c) 2008 Tobias Doerffel <tobydox/at/users.sourceforge.net>
- * 
  * This file is part of Linux MultiMedia Studio - http://lmms.sourceforge.net
  *
  * This program is free software; you can redistribute it and/or
@@ -22,9 +22,8 @@
  *
  */
 
-
-#ifndef _BASIC_OPS_H
-#define _BASIC_OPS_H
+#ifndef _CPU_H
+#define _CPU_H
 
 #include "lmms_basics.h"
 
@@ -32,56 +31,64 @@
 #include <stdbool.h>
 #endif
 
-void initBasicOps( void );
+#ifdef __cplusplus
+namespace CPU
+{
+#endif
 
-void * alignedMalloc( int _bytes );
-void alignedFree( void * _buf );
+void init();
 
-sampleFrameA * alignedAllocFrames( int _frames );
-void alignedFreeFrames( sampleFrameA * _buf );
+void * memAlloc( int _bytes );
+void memFree( void * _buf );
+
+sampleFrameA * allocFrames( int _frames );
+void freeFrames( sampleFrameA * _buf );
 
 
-// all aligned* functions assume data to be 16 byte aligned and size to be
-// multiples of 64
-typedef void (*alignedMemCpyFunc)( void * RP _dst, const void * RP _src,
+// all functions assume data to be 16 byte  and size to be
+// multiples of 64 (except for unaligned*())
+typedef void (*MemCpyFunc)( void * RP _dst, const void * RP _src,
 								int _size );
-typedef void (*alignedMemClearFunc)( void * RP _dst, int _size );
-typedef void (*alignedBufApplyGainFunc)( sampleFrameA * RP _dst,
+typedef void (*MemClearFunc)( void * RP _dst, int _size );
+typedef void (*BufApplyGainFunc)( sampleFrameA * RP _dst,
 						float _gain, int _frames );
-typedef void (*alignedBufMixFunc)( sampleFrameA * RP _dst,
+typedef void (*BufMixFunc)( sampleFrameA * RP _dst,
 						const sampleFrameA * RP _src,
 								int _frames );
-typedef void (*alignedBufMixLRCoeffFunc)( sampleFrameA * RP _dst,
+typedef void (*BufMixLRCoeffFunc)( sampleFrameA * RP _dst,
 						const sampleFrameA * RP _src,
 						float _left, float _right,
 								int _frames );
-typedef void (*unalignedBufMixLRCoeffFunc)( sampleFrame * RP _dst,
+typedef void (*UnalignedBufMixLRCoeffFunc)( sampleFrame * RP _dst,
 						const sampleFrame * RP _src,
 						float _left, float _right,
 								int _frames );
-typedef void (*alignedBufWetDryMixFunc)( sampleFrameA * RP _dst,
+typedef void (*BufWetDryMixFunc)( sampleFrameA * RP _dst,
 					const sampleFrameA * RP _src,
 					float _wet, float _dry, int _frames );
-typedef void (*alignedBufWetDryMixSplittedFunc)( sampleFrameA * RP _dst,
+typedef void (*BufWetDryMixSplittedFunc)( sampleFrameA * RP _dst,
 					const float * RP _left,
 					const float * RP _right,
 					float _wet, float _dry, int _frames );
-typedef int (*alignedConvertToS16Func)( const sampleFrameA * RP _src,
+typedef int (*ConvertToS16Func)( const sampleFrameA * RP _src,
 					intSampleFrameA * RP _dst,
 					const fpp_t _frames,
 					const float _master_gain,
 					const bool _convert_endian );
 
-extern alignedMemCpyFunc alignedMemCpy;
-extern alignedMemClearFunc alignedMemClear;
-extern alignedBufApplyGainFunc alignedBufApplyGain;
-extern alignedBufMixFunc alignedBufMix;
-extern alignedBufMixLRCoeffFunc alignedBufMixLRCoeff;
-extern unalignedBufMixLRCoeffFunc unalignedBufMixLRCoeff;
-extern alignedBufWetDryMixFunc alignedBufWetDryMix;
-extern alignedBufWetDryMixSplittedFunc alignedBufWetDryMixSplitted;
-extern alignedConvertToS16Func alignedConvertToS16;
+extern MemCpyFunc memCpy;
+extern MemClearFunc memClear;
+extern BufApplyGainFunc bufApplyGain;
+extern BufMixFunc bufMix;
+extern BufMixLRCoeffFunc bufMixLRCoeff;
+extern UnalignedBufMixLRCoeffFunc unalignedBufMixLRCoeff;
+extern BufWetDryMixFunc bufWetDryMix;
+extern BufWetDryMixSplittedFunc bufWetDryMixSplitted;
+extern ConvertToS16Func convertToS16;
 
+#ifdef __cplusplus
+}
+#endif
 
 #ifdef LMMS_HOST_X86
 #define X86_OPTIMIZATIONS

include/audio_dummy.h

@@ -1,8 +1,8 @@
 /*
  * audio_dummy.h - dummy-audio-device
  *
- * Copyright (c) 2004-2008 Tobias Doerffel <tobydox/at/users.sourceforge.net>
- * 
+ * Copyright (c) 2004-2009 Tobias Doerffel <tobydox/at/users.sourceforge.net>
+ *
  * This file is part of Linux MultiMedia Studio - http://lmms.sourceforge.net
  *
  * This program is free software; you can redistribute it and/or
@@ -22,12 +22,11 @@
  *
  */
 
-
 #ifndef _AUDIO_DUMMY_H
 #define _AUDIO_DUMMY_H
 
 #include "audio_device.h"
-#include "basic_ops.h"
+#include "Cpu.h"
 #include "micro_timer.h"
 
 
@@ -45,7 +44,7 @@ public:
 		stopProcessing();
 	}
 
-	inline static QString name( void )
+	inline static QString name()
 	{
 		return( QT_TRANSLATE_NOOP( "setupWidget",
 						"Dummy (no sound output)" ) );
@@ -64,11 +63,11 @@ public:
 		{
 		}
 
-		virtual void saveSettings( void )
+		virtual void saveSettings()
 		{
 		}
 
-		virtual void show( void )
+		virtual void show()
 		{
 			parentWidget()->hide();
 			QWidget::show();
@@ -78,12 +77,12 @@ public:
 
 
 private:
-	virtual void startProcessing( void )
+	virtual void startProcessing()
 	{
 		start();
 	}
 
-	virtual void stopProcessing( void )
+	virtual void stopProcessing()
 	{
 		if( isRunning() )
 		{
@@ -92,7 +91,7 @@ private:
 		}
 	}
 
-	virtual void run( void )
+	virtual void run()
 	{
 		microTimer timer;
 		while( true )
@@ -104,7 +103,7 @@ private:
 			{
 				break;
 			}
-			alignedFreeFrames( b );
+			CPU::freeFrames( b );
 
 			const Sint32 microseconds = static_cast<Sint32>(
 					getMixer()->framesPerPeriod() *