Bart/bullet3
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Apple contribution for OSX SSE and iOS NEON optimizations unit tests, thanks to Jordan Hubbard, Ian Ollmann and Hristo Hristov.

Browse Source 
For OSX:
cd build
./premake_osx xcode4
for iOS:
cd build
./ios_build.sh
./ios_run.sh

Also integrated the branches/StackAllocation to make it easier to multi-thread collision detection in the near future. It avoids changing the btCollisionObject while performing collision detection.

As this is a large patch, some stuff might be temporarily broken, I'll keep an eye out on issues.
This commit is contained in:
erwin.coumans
2012-06-07 00:56:30 +00:00
parent 777b92a2ad
commit 73b217fb07
323 changed files with 30730 additions and 13635 deletions
Download Patch File Download Diff File Expand all files Collapse all files

164
Test/Source/Tests/Test_v3dot.cpp Normal file
View File

@@ -0,0 +1,164 @@
//
// Test_v3dot.cpp
// BulletTest
//
// Copyright (c) 2011 Apple Inc.
//
#include "LinearMath/btScalar.h"
#if defined (BT_USE_SSE_IN_API) || defined (BT_USE_NEON)
#include "Test_v3dot.h"
#include "vector.h"
#include "Utils.h"
#include "main.h"
#include <math.h>
#include <string.h>
#include <LinearMath/btVector3.h>
// reference code for testing purposes
static inline
btScalar v3dot_ref(
const btVector3& v1,
const btVector3& v2);
#define LOOPCOUNT 1000
#define NUM_CYCLES 10000
int Test_v3dot(void)
{
btVector3 v1, v2;
float x,y,z,w;
// Init the data
x = RANDF_01;
y = RANDF_01;
z = RANDF_01;
w = BT_NAN; // w channel NaN
v1.setValue(x,y,z);
v1.setW(w);
x = RANDF_01;
y = RANDF_01;
z = RANDF_01;
v2.setValue(x,y,z);
v2.setW(w);
float correctDot0, testDot0;
{
correctDot0 = w;
testDot0 = w; ;
correctDot0 = v3dot_ref(v1, v2);
testDot0 = v1.dot(v2);
if( fabsf(correctDot0 - testDot0) > FLT_EPSILON * 4 )
{
vlog( "Error - v3dot result error! %f != %f \n", correctDot0, testDot0);
return 1;
}
}
#define DATA_SIZE 1024
btVector3 vec3_arr1[DATA_SIZE];
btVector3 vec3_arr2[DATA_SIZE];
btScalar res_arr[DATA_SIZE];
uint64_t scalarTime;
uint64_t vectorTime;
size_t j, k;
for( k = 0; k < DATA_SIZE; k++ )
{
x = RANDF_01;
y = RANDF_01;
z = RANDF_01;
vec3_arr1[k].setValue(x,y,z);
vec3_arr1[k].setW(w);
x = RANDF_01;
y = RANDF_01;
z = RANDF_01;
vec3_arr2[k].setValue(x,y,z);
vec3_arr2[k].setW(w);
res_arr[k] = w;
}
{
uint64_t startTime, bestTime, currentTime;
bestTime = -1LL;
scalarTime = 0;
for (j = 0; j < NUM_CYCLES; j++)
{
startTime = ReadTicks();
for( k = 0; k+4 <= LOOPCOUNT; k+=4 )
{
size_t k32 = (k & (DATA_SIZE-1));
res_arr[k32] = v3dot_ref( vec3_arr1[k32], vec3_arr2[k32]); k32++;
res_arr[k32] = v3dot_ref( vec3_arr1[k32], vec3_arr2[k32]); k32++;
res_arr[k32] = v3dot_ref( vec3_arr1[k32], vec3_arr2[k32]); k32++;
res_arr[k32] = v3dot_ref( vec3_arr1[k32], vec3_arr2[k32]);
}
currentTime = ReadTicks() - startTime;
scalarTime += currentTime;
if( currentTime < bestTime )
bestTime = currentTime;
}
if( 0 == gReportAverageTimes )
scalarTime = bestTime;
else
scalarTime /= NUM_CYCLES;
}
{
uint64_t startTime, bestTime, currentTime;
bestTime = -1LL;
vectorTime = 0;
for (j = 0; j < NUM_CYCLES; j++)
{
startTime = ReadTicks();
for( k = 0; k+4 <= LOOPCOUNT; k+=4 )
{
size_t k32 = k & (DATA_SIZE -1);
res_arr[k32] = vec3_arr1[k32].dot(vec3_arr2[k32]); k32++;
res_arr[k32] = vec3_arr1[k32].dot(vec3_arr2[k32]); k32++;
res_arr[k32] = vec3_arr1[k32].dot(vec3_arr2[k32]); k32++;
res_arr[k32] = vec3_arr1[k32].dot(vec3_arr2[k32]);
}
currentTime = ReadTicks() - startTime;
vectorTime += currentTime;
if( currentTime < bestTime )
bestTime = currentTime;
}
if( 0 == gReportAverageTimes )
vectorTime = bestTime;
else
vectorTime /= NUM_CYCLES;
}
vlog( "Timing:\n" );
vlog( " \t scalar\t vector\n" );
vlog( " \t%10.4f\t%10.4f\n", TicksToCycles( scalarTime ) / LOOPCOUNT, TicksToCycles( vectorTime ) / LOOPCOUNT );
return 0;
}
static btScalar v3dot_ref(const btVector3& v1,
const btVector3& v2)
{
return (v1.m_floats[0] * v2.m_floats[0] +
v1.m_floats[1] * v2.m_floats[1] +
v1.m_floats[2] * v2.m_floats[2]);
}
#endif //BT_USE_SSE