diff options
| author | Joe Ludwig <[email protected]> | 2013-06-26 15:22:04 -0700 |
|---|---|---|
| committer | Joe Ludwig <[email protected]> | 2013-06-26 15:22:04 -0700 |
| commit | 39ed87570bdb2f86969d4be821c94b722dc71179 (patch) | |
| tree | abc53757f75f40c80278e87650ea92808274aa59 /sp/src/mathlib/randsse.cpp | |
| download | source-sdk-2013-39ed87570bdb2f86969d4be821c94b722dc71179.tar.xz source-sdk-2013-39ed87570bdb2f86969d4be821c94b722dc71179.zip | |
First version of the SOurce SDK 2013
Diffstat (limited to 'sp/src/mathlib/randsse.cpp')
| -rw-r--r-- | sp/src/mathlib/randsse.cpp | 109 |
1 files changed, 109 insertions, 0 deletions
diff --git a/sp/src/mathlib/randsse.cpp b/sp/src/mathlib/randsse.cpp new file mode 100644 index 00000000..d6bd666e --- /dev/null +++ b/sp/src/mathlib/randsse.cpp @@ -0,0 +1,109 @@ +//========= Copyright Valve Corporation, All rights reserved. ============//
+//
+// Purpose: generates 4 randum numbers in the range 0..1 quickly, using SIMD
+//
+//=====================================================================================//
+
+#include <math.h>
+#include <float.h> // Needed for FLT_EPSILON
+#include "basetypes.h"
+#include <memory.h>
+#include "tier0/dbg.h"
+#include "mathlib/mathlib.h"
+#include "mathlib/vector.h"
+#include "mathlib/ssemath.h"
+
+// memdbgon must be the last include file in a .cpp file!!!
+#include "tier0/memdbgon.h"
+
+// see knuth volume 3 for insight.
+
+class SIMDRandStreamContext
+{
+ fltx4 m_RandY[55];
+
+ fltx4 *m_pRand_J, *m_pRand_K;
+
+
+public:
+ void Seed( uint32 seed )
+ {
+ m_pRand_J=m_RandY+23; m_pRand_K=m_RandY+54;
+ for(int i=0;i<55;i++)
+ {
+ for(int j=0;j<4;j++)
+ {
+ SubFloat( m_RandY[i], j) = (seed>>16)/65536.0;
+ seed=(seed+1)*3141592621u;
+ }
+ }
+ }
+
+ inline fltx4 RandSIMD( void )
+ {
+ // ret= rand[k]+rand[j]
+ fltx4 retval=AddSIMD( *m_pRand_K, *m_pRand_J );
+
+ // if ( ret>=1.0) ret-=1.0
+ fltx4 overflow_mask=CmpGeSIMD( retval, Four_Ones );
+ retval=SubSIMD( retval, AndSIMD( Four_Ones, overflow_mask ) );
+
+ *m_pRand_K = retval;
+
+ // update pointers w/ wrap-around
+ if ( --m_pRand_J < m_RandY )
+ m_pRand_J=m_RandY+54;
+ if ( --m_pRand_K < m_RandY )
+ m_pRand_K=m_RandY+54;
+
+ return retval;
+ }
+};
+
+#define MAX_SIMULTANEOUS_RANDOM_STREAMS 32
+
+static SIMDRandStreamContext s_SIMDRandContexts[MAX_SIMULTANEOUS_RANDOM_STREAMS];
+
+static volatile int s_nRandContextsInUse[MAX_SIMULTANEOUS_RANDOM_STREAMS];
+
+void SeedRandSIMD(uint32 seed)
+{
+ for( int i = 0; i<MAX_SIMULTANEOUS_RANDOM_STREAMS; i++)
+ s_SIMDRandContexts[i].Seed( seed+i );
+}
+
+fltx4 RandSIMD( int nContextIndex )
+{
+ return s_SIMDRandContexts[nContextIndex].RandSIMD();
+}
+
+int GetSIMDRandContext( void )
+{
+ for(;;)
+ {
+ for(int i=0; i < NELEMS( s_SIMDRandContexts ); i++)
+ {
+ if ( ! s_nRandContextsInUse[i] ) // available?
+ {
+ // try to take it!
+ if ( ThreadInterlockedAssignIf( &( s_nRandContextsInUse[i]), 1, 0 ) )
+ {
+ return i; // done!
+ }
+ }
+ }
+ Assert(0); // why don't we have enough buffers?
+ ThreadSleep();
+ }
+}
+
+void ReleaseSIMDRandContext( int nContext )
+{
+ s_nRandContextsInUse[ nContext ] = 0;
+}
+
+
+fltx4 RandSIMD( void )
+{
+ return s_SIMDRandContexts[0].RandSIMD();
+}
|