/dports/lang/pocl/pocl-1.8/tests/regression/ |
H A D | test_flatten_barrier_subs.cpp | 125 #define WG_SIZE 64 in exclusive_scan_cl() macro 126 int GROUP_BLOCK_SIZE_SCAN = (WG_SIZE << 3); in exclusive_scan_cl() 127 int GROUP_BLOCK_SIZE_DISTRIBUTE = (WG_SIZE << 2); in exclusive_scan_cl() 154 cl::NDRange(NUM_GROUPS_BOTTOM_LEVEL_SCAN * WG_SIZE), in exclusive_scan_cl() 155 cl::NDRange(WG_SIZE) in exclusive_scan_cl() 166 cl::NDRange(NUM_GROUPS_MID_LEVEL_SCAN * WG_SIZE), in exclusive_scan_cl() 167 cl::NDRange(WG_SIZE) in exclusive_scan_cl() 177 cl::NDRange(NUM_GROUPS_TOP_LEVEL_SCAN * WG_SIZE), in exclusive_scan_cl() 178 cl::NDRange(WG_SIZE) in exclusive_scan_cl() 188 cl::NDRange(WG_SIZE) in exclusive_scan_cl() [all …]
|
/dports/devel/bullet/bullet3-3.21/src/Bullet3OpenCL/RigidBody/kernels/ |
H A D | batchingKernels.cl | 64 #define WG_SIZE 64 85 #define STACK_SIZE (WG_SIZE*10) 86 //#define STACK_SIZE (WG_SIZE) 89 #define CHECK_SIZE (WG_SIZE) 167 while( ringCap > WG_SIZE ) 170 if( lIdx < ringCap - WG_SIZE ) 200 for(int ii=0; ii<end; ii+=WG_SIZE, srcIdx+=WG_SIZE) 267 for(int i=lIdx; i<STACK_SIZE; i+=WG_SIZE) 276 //for(int i=lIdx; i<CHECK_SIZE; i+=WG_SIZE) 287 for(int i=lIdx; i<ldsStackEnd; i+=WG_SIZE) [all …]
|
/dports/devel/py-bullet3/bullet3-3.21/src/Bullet3OpenCL/RigidBody/kernels/ |
H A D | batchingKernels.cl | 64 #define WG_SIZE 64 85 #define STACK_SIZE (WG_SIZE*10) 86 //#define STACK_SIZE (WG_SIZE) 89 #define CHECK_SIZE (WG_SIZE) 167 while( ringCap > WG_SIZE ) 170 if( lIdx < ringCap - WG_SIZE ) 200 for(int ii=0; ii<end; ii+=WG_SIZE, srcIdx+=WG_SIZE) 267 for(int i=lIdx; i<STACK_SIZE; i+=WG_SIZE) 276 //for(int i=lIdx; i<CHECK_SIZE; i+=WG_SIZE) 287 for(int i=lIdx; i<ldsStackEnd; i+=WG_SIZE) [all …]
|
/dports/devel/godot/godot-3.2.3-stable/thirdparty/bullet/Bullet3OpenCL/RigidBody/kernels/ |
H A D | batchingKernels.cl | 64 #define WG_SIZE 64 85 #define STACK_SIZE (WG_SIZE*10) 86 //#define STACK_SIZE (WG_SIZE) 89 #define CHECK_SIZE (WG_SIZE) 167 while( ringCap > WG_SIZE ) 170 if( lIdx < ringCap - WG_SIZE ) 200 for(int ii=0; ii<end; ii+=WG_SIZE, srcIdx+=WG_SIZE) 267 for(int i=lIdx; i<STACK_SIZE; i+=WG_SIZE) 276 //for(int i=lIdx; i<CHECK_SIZE; i+=WG_SIZE) 287 for(int i=lIdx; i<ldsStackEnd; i+=WG_SIZE) [all …]
|
/dports/devel/godot-tools/godot-3.2.3-stable/thirdparty/bullet/Bullet3OpenCL/RigidBody/kernels/ |
H A D | batchingKernels.cl | 64 #define WG_SIZE 64 85 #define STACK_SIZE (WG_SIZE*10) 86 //#define STACK_SIZE (WG_SIZE) 89 #define CHECK_SIZE (WG_SIZE) 167 while( ringCap > WG_SIZE ) 170 if( lIdx < ringCap - WG_SIZE ) 200 for(int ii=0; ii<end; ii+=WG_SIZE, srcIdx+=WG_SIZE) 267 for(int i=lIdx; i<STACK_SIZE; i+=WG_SIZE) 276 //for(int i=lIdx; i<CHECK_SIZE; i+=WG_SIZE) 287 for(int i=lIdx; i<ldsStackEnd; i+=WG_SIZE) [all …]
|
/dports/devel/bullet/bullet3-3.21/src/Bullet3OpenCL/ParallelPrimitives/kernels/ |
H A D | PrefixScanKernels.cl | 25 #define WG_SIZE 128 82 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 87 __local u32 ldsData[WG_SIZE*2]; 95 u32 sum = ScanExclusive(ldsData, WG_SIZE*2, GET_LOCAL_IDX, GET_GROUP_SIZE); 109 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 113 const u32 blockSize = WG_SIZE*2; 127 __attribute__((reqd_work_group_size(WG_SIZE,1,1)))
|
H A D | PrefixScanFloat4Kernels.cl | 25 #define WG_SIZE 128 82 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 86 __local float4 ldsData[WG_SIZE*2]; 94 float4 sum = ScanExclusiveFloat4(ldsData, WG_SIZE*2, GET_LOCAL_IDX, GET_GROUP_SIZE); 109 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 113 const u32 blockSize = WG_SIZE*2; 127 __attribute__((reqd_work_group_size(WG_SIZE,1,1)))
|
H A D | RadixSort32Kernels.cl | 40 #define WG_SIZE 64 41 #define ELEMENTS_PER_WORK_ITEM (256/WG_SIZE) 207 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 210 __local u32 localHistogramMat[NUM_BUCKET*WG_SIZE]; 270 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 273 __local u32 localHistogramMat[NUM_BUCKET*WG_SIZE]; 548 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 622 for(int i=0; i<WG_SIZE/16; i++) 881 for(int i=0; i<WG_SIZE/16; i++) 1004 for (int lIdx=0;lIdx<WG_SIZE;lIdx++) [all …]
|
/dports/devel/godot-tools/godot-3.2.3-stable/thirdparty/bullet/Bullet3OpenCL/ParallelPrimitives/kernels/ |
H A D | PrefixScanFloat4Kernels.cl | 25 #define WG_SIZE 128 82 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 86 __local float4 ldsData[WG_SIZE*2]; 94 float4 sum = ScanExclusiveFloat4(ldsData, WG_SIZE*2, GET_LOCAL_IDX, GET_GROUP_SIZE); 109 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 113 const u32 blockSize = WG_SIZE*2; 127 __attribute__((reqd_work_group_size(WG_SIZE,1,1)))
|
H A D | PrefixScanKernels.cl | 25 #define WG_SIZE 128 82 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 87 __local u32 ldsData[WG_SIZE*2]; 95 u32 sum = ScanExclusive(ldsData, WG_SIZE*2, GET_LOCAL_IDX, GET_GROUP_SIZE); 109 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 113 const u32 blockSize = WG_SIZE*2; 127 __attribute__((reqd_work_group_size(WG_SIZE,1,1)))
|
H A D | RadixSort32Kernels.cl | 40 #define WG_SIZE 64 41 #define ELEMENTS_PER_WORK_ITEM (256/WG_SIZE) 207 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 210 __local u32 localHistogramMat[NUM_BUCKET*WG_SIZE]; 270 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 273 __local u32 localHistogramMat[NUM_BUCKET*WG_SIZE]; 548 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 622 for(int i=0; i<WG_SIZE/16; i++) 881 for(int i=0; i<WG_SIZE/16; i++) 1004 for (int lIdx=0;lIdx<WG_SIZE;lIdx++) [all …]
|
/dports/devel/py-bullet3/bullet3-3.21/src/Bullet3OpenCL/ParallelPrimitives/kernels/ |
H A D | PrefixScanFloat4Kernels.cl | 25 #define WG_SIZE 128 82 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 86 __local float4 ldsData[WG_SIZE*2]; 94 float4 sum = ScanExclusiveFloat4(ldsData, WG_SIZE*2, GET_LOCAL_IDX, GET_GROUP_SIZE); 109 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 113 const u32 blockSize = WG_SIZE*2; 127 __attribute__((reqd_work_group_size(WG_SIZE,1,1)))
|
H A D | PrefixScanKernels.cl | 25 #define WG_SIZE 128 82 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 87 __local u32 ldsData[WG_SIZE*2]; 95 u32 sum = ScanExclusive(ldsData, WG_SIZE*2, GET_LOCAL_IDX, GET_GROUP_SIZE); 109 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 113 const u32 blockSize = WG_SIZE*2; 127 __attribute__((reqd_work_group_size(WG_SIZE,1,1)))
|
H A D | RadixSort32Kernels.cl | 40 #define WG_SIZE 64 41 #define ELEMENTS_PER_WORK_ITEM (256/WG_SIZE) 207 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 210 __local u32 localHistogramMat[NUM_BUCKET*WG_SIZE]; 270 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 273 __local u32 localHistogramMat[NUM_BUCKET*WG_SIZE]; 548 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 622 for(int i=0; i<WG_SIZE/16; i++) 881 for(int i=0; i<WG_SIZE/16; i++) 1004 for (int lIdx=0;lIdx<WG_SIZE;lIdx++) [all …]
|
/dports/devel/godot/godot-3.2.3-stable/thirdparty/bullet/Bullet3OpenCL/ParallelPrimitives/kernels/ |
H A D | PrefixScanKernels.cl | 25 #define WG_SIZE 128 82 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 87 __local u32 ldsData[WG_SIZE*2]; 95 u32 sum = ScanExclusive(ldsData, WG_SIZE*2, GET_LOCAL_IDX, GET_GROUP_SIZE); 109 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 113 const u32 blockSize = WG_SIZE*2; 127 __attribute__((reqd_work_group_size(WG_SIZE,1,1)))
|
H A D | PrefixScanFloat4Kernels.cl | 25 #define WG_SIZE 128 82 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 86 __local float4 ldsData[WG_SIZE*2]; 94 float4 sum = ScanExclusiveFloat4(ldsData, WG_SIZE*2, GET_LOCAL_IDX, GET_GROUP_SIZE); 109 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 113 const u32 blockSize = WG_SIZE*2; 127 __attribute__((reqd_work_group_size(WG_SIZE,1,1)))
|
H A D | RadixSort32Kernels.cl | 40 #define WG_SIZE 64 41 #define ELEMENTS_PER_WORK_ITEM (256/WG_SIZE) 207 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 210 __local u32 localHistogramMat[NUM_BUCKET*WG_SIZE]; 270 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 273 __local u32 localHistogramMat[NUM_BUCKET*WG_SIZE]; 548 __attribute__((reqd_work_group_size(WG_SIZE,1,1))) 622 for(int i=0; i<WG_SIZE/16; i++) 881 for(int i=0; i<WG_SIZE/16; i++) 1004 for (int lIdx=0;lIdx<WG_SIZE;lIdx++) [all …]
|
/dports/devel/bullet/bullet3-3.21/src/Bullet3OpenCL/ParallelPrimitives/ |
H A D | b3RadixSort32CL.cpp | 233 b3Assert(WG_SIZE == 64); in execute() 245 int blockSize = ELEMENTS_PER_WORK_ITEM * WG_SIZE; //set at 256 in execute() 284 int num = NUM_WGS * WG_SIZE; in execute() 285 launcher.launch1D(num, WG_SIZE); in execute() 346 launcher.launch1D(nWGs * WG_SIZE, WG_SIZE); in execute() 557 b3Assert(WG_SIZE == 64); in execute() 569 int blockSize = ELEMENTS_PER_WORK_ITEM * WG_SIZE; //set at 256 in execute() 595 int num = NUM_WGS * WG_SIZE; in execute() 596 launcher.launch1D(num, WG_SIZE); in execute() 627 launcher.launch1D(nWGs * WG_SIZE, WG_SIZE); in execute()
|
H A D | b3RadixSort32CL.h | 57 WG_SIZE = 64, enumerator 59 ELEMENTS_PER_WORK_ITEM = (BLOCK_SIZE / WG_SIZE),
|
/dports/devel/godot-tools/godot-3.2.3-stable/thirdparty/bullet/Bullet3OpenCL/ParallelPrimitives/ |
H A D | b3RadixSort32CL.cpp | 233 b3Assert(WG_SIZE == 64); in execute() 245 int blockSize = ELEMENTS_PER_WORK_ITEM * WG_SIZE; //set at 256 in execute() 284 int num = NUM_WGS * WG_SIZE; in execute() 285 launcher.launch1D(num, WG_SIZE); in execute() 346 launcher.launch1D(nWGs * WG_SIZE, WG_SIZE); in execute() 557 b3Assert(WG_SIZE == 64); in execute() 569 int blockSize = ELEMENTS_PER_WORK_ITEM * WG_SIZE; //set at 256 in execute() 595 int num = NUM_WGS * WG_SIZE; in execute() 596 launcher.launch1D(num, WG_SIZE); in execute() 627 launcher.launch1D(nWGs * WG_SIZE, WG_SIZE); in execute()
|
H A D | b3RadixSort32CL.h | 57 WG_SIZE = 64, enumerator 59 ELEMENTS_PER_WORK_ITEM = (BLOCK_SIZE / WG_SIZE),
|
/dports/devel/py-bullet3/bullet3-3.21/src/Bullet3OpenCL/ParallelPrimitives/ |
H A D | b3RadixSort32CL.cpp | 233 b3Assert(WG_SIZE == 64); in execute() 245 int blockSize = ELEMENTS_PER_WORK_ITEM * WG_SIZE; //set at 256 in execute() 284 int num = NUM_WGS * WG_SIZE; in execute() 285 launcher.launch1D(num, WG_SIZE); in execute() 346 launcher.launch1D(nWGs * WG_SIZE, WG_SIZE); in execute() 557 b3Assert(WG_SIZE == 64); in execute() 569 int blockSize = ELEMENTS_PER_WORK_ITEM * WG_SIZE; //set at 256 in execute() 595 int num = NUM_WGS * WG_SIZE; in execute() 596 launcher.launch1D(num, WG_SIZE); in execute() 627 launcher.launch1D(nWGs * WG_SIZE, WG_SIZE); in execute()
|
H A D | b3RadixSort32CL.h | 57 WG_SIZE = 64, enumerator 59 ELEMENTS_PER_WORK_ITEM = (BLOCK_SIZE / WG_SIZE),
|
/dports/devel/godot/godot-3.2.3-stable/thirdparty/bullet/Bullet3OpenCL/ParallelPrimitives/ |
H A D | b3RadixSort32CL.cpp | 233 b3Assert(WG_SIZE == 64); in execute() 245 int blockSize = ELEMENTS_PER_WORK_ITEM * WG_SIZE; //set at 256 in execute() 284 int num = NUM_WGS * WG_SIZE; in execute() 285 launcher.launch1D(num, WG_SIZE); in execute() 346 launcher.launch1D(nWGs * WG_SIZE, WG_SIZE); in execute() 557 b3Assert(WG_SIZE == 64); in execute() 569 int blockSize = ELEMENTS_PER_WORK_ITEM * WG_SIZE; //set at 256 in execute() 595 int num = NUM_WGS * WG_SIZE; in execute() 596 launcher.launch1D(num, WG_SIZE); in execute() 627 launcher.launch1D(nWGs * WG_SIZE, WG_SIZE); in execute()
|
H A D | b3RadixSort32CL.h | 57 WG_SIZE = 64, enumerator 59 ELEMENTS_PER_WORK_ITEM = (BLOCK_SIZE / WG_SIZE),
|