Skip to content

O
opencv
Commit 0ad03162 authored by Ilya Lavrenov's avatar Ilya Lavrenov
Browse files
Options

refactored and extended arithm operations add/sub/mul/div/absdiff

parent 5ff5fdd7
Show whitespace changes
Inline Side-by-side
Showing with 283 additions and 2832 deletions
modules/ocl/include/opencv2/ocl/ocl.hpp
View file @ 0ad03162
......@@ -409,40 +409,37 @@ namespace cv
CV_EXPORTS void split(const oclMat &src, vector<oclMat> &dst);
////////////////////////////// Arithmetics ///////////////////////////////////
//#if defined DOUBLE_SUPPORT
//typedef double F;
//#else
//typedef float F;
//#endif
// CV_EXPORTS void addWeighted(const oclMat& a,F alpha, const oclMat& b,F beta,F gama, oclMat& c);
CV_EXPORTS void addWeighted(const oclMat &a, double alpha, const oclMat &b, double beta, double gama, oclMat &c);
//! adds one matrix to another (c = a + b)
// supports all types except CV_8SC1,CV_8SC2,CV8SC3 and CV_8SC4
CV_EXPORTS void add(const oclMat &a, const oclMat &b, oclMat &c);
//! adds one matrix to another (c = a + b)
// supports all types except CV_8SC1,CV_8SC2,CV8SC3 and CV_8SC4
CV_EXPORTS void add(const oclMat &a, const oclMat &b, oclMat &c, const oclMat &mask);
CV_EXPORTS void add(const oclMat &a, const oclMat &b, oclMat &c, const oclMat &mask = oclMat());
//! adds scalar to a matrix (c = a + s)
// supports all types except CV_8SC1,CV_8SC2,CV8SC3 and CV_8SC4
CV_EXPORTS void add(const oclMat &a, const Scalar &sc, oclMat &c, const oclMat &mask = oclMat());
//! subtracts one matrix from another (c = a - b)
// supports all types except CV_8SC1,CV_8SC2,CV8SC3 and CV_8SC4
CV_EXPORTS void subtract(const oclMat &a, const oclMat &b, oclMat &c);
//! subtracts one matrix from another (c = a - b)
// supports all types except CV_8SC1,CV_8SC2,CV8SC3 and CV_8SC4
CV_EXPORTS void subtract(const oclMat &a, const oclMat &b, oclMat &c, const oclMat &mask);
CV_EXPORTS void subtract(const oclMat &a, const oclMat &b, oclMat &c, const oclMat &mask = oclMat());
//! subtracts scalar from a matrix (c = a - s)
// supports all types except CV_8SC1,CV_8SC2,CV8SC3 and CV_8SC4
CV_EXPORTS void subtract(const oclMat &a, const Scalar &sc, oclMat &c, const oclMat &mask = oclMat());
//! subtracts scalar from a matrix (c = a - s)
// supports all types except CV_8SC1,CV_8SC2,CV8SC3 and CV_8SC4
CV_EXPORTS void subtract(const Scalar &sc, const oclMat &a, oclMat &c, const oclMat &mask = oclMat());
//! computes element-wise product of the two arrays (c = a * b)
// supports all types except CV_8SC1,CV_8SC2,CV8SC3 and CV_8SC4
CV_EXPORTS void multiply(const oclMat &a, const oclMat &b, oclMat &c, double scale = 1);
//! multiplies matrix to a number (dst = scalar * src)
// supports CV_32FC1 only
CV_EXPORTS void multiply(double scalar, const oclMat &src, oclMat &dst);
//! computes element-wise quotient of the two arrays (c = a / b)
// supports all types except CV_8SC1,CV_8SC2,CV8SC3 and CV_8SC4
CV_EXPORTS void divide(const oclMat &a, const oclMat &b, oclMat &c, double scale = 1);
......
modules/ocl/src/arithm.cpp
View file @ 0ad03162
......@@ -62,11 +62,11 @@ namespace cv
{
namespace ocl
{
////////////////////////////////OpenCL kernel strings/////////////////////
//////////////////////////////// OpenCL kernel strings /////////////////////
extern const char *transpose_kernel;
extern const char *arithm_nonzero;
extern const char *arithm_sum;
extern const char *arithm_2_mat;
extern const char *arithm_sum_3;
extern const char *arithm_minMax;
extern const char *arithm_minMax_mask;
......@@ -74,6 +74,7 @@ namespace cv
extern const char *arithm_minMaxLoc_mask;
extern const char *arithm_LUT;
extern const char *arithm_add;
extern const char *arithm_add_mask;
extern const char *arithm_add_scalar;
extern const char *arithm_add_scalar_mask;
extern const char *arithm_bitwise_binary;
......@@ -83,9 +84,7 @@ namespace cv
extern const char *arithm_bitwise_not;
extern const char *arithm_compare_eq;
extern const char *arithm_compare_ne;
extern const char *arithm_mul;
extern const char *arithm_div;
extern const char *arithm_absdiff;
extern const char *arithm_magnitudeSqr;
extern const char *arithm_transpose;
extern const char *arithm_flip;
extern const char *arithm_flip_rc;
......@@ -97,390 +96,176 @@ namespace cv
extern const char *arithm_addWeighted;
extern const char *arithm_phase;
extern const char *arithm_pow;
extern const char *arithm_magnitudeSqr;
extern const char *arithm_setidentity;
//extern const char * jhp_transpose_kernel;
int64 kernelrealtotal = 0;
int64 kernelalltotal = 0;
int64 reducetotal = 0;
int64 downloadtotal = 0;
int64 alltotal = 0;
}
}
//////////////////////////////////////////////////////////////////////////////
/////////////////////// add subtract multiply divide /////////////////////////
//////////////////////////////////////////////////////////////////////////////
template<typename T>
void arithmetic_run(const oclMat &src1, const oclMat &src2, oclMat &dst,
string kernelName, const char **kernelString, void *_scalar, int op_type = 0)
{
if(!src1.clCxt->supportsFeature(Context::CL_DOUBLE) && src1.type() == CV_64F)
{
CV_Error(CV_GpuNotSupported, "Selected device don't support double\r\n");
return;
}
dst.create(src1.size(), src1.type());
CV_Assert(src1.cols == src2.cols && src2.cols == dst.cols &&
src1.rows == src2.rows && src2.rows == dst.rows);
CV_Assert(src1.type() == src2.type() && src1.type() == dst.type());
CV_Assert(src1.depth() != CV_8S);
Context *clCxt = src1.clCxt;
int channels = dst.oclchannels();
int depth = dst.depth();
int vector_lengths[4][7] = {{4, 0, 4, 4, 1, 1, 1},
{4, 0, 4, 4, 1, 1, 1},
{4, 0, 4, 4, 1, 1, 1},
{4, 0, 4, 4, 1, 1, 1}
};
size_t vector_length = vector_lengths[channels - 1][depth];
int offset_cols = (dst.offset / dst.elemSize1()) & (vector_length - 1);
int cols = divUp(dst.cols * channels + offset_cols, vector_length);
//////////////////////////////////////////////////////////////////////////////
/////////////////////// add subtract multiply divide /////////////////////////
//////////////////////////////////////////////////////////////////////////////
size_t localThreads[3] = { 64, 4, 1 };
size_t globalThreads[3] = { cols, dst.rows, 1 };
enum { ADD = 0, SUB, MUL, DIV, ABS_DIFF };
int dst_step1 = dst.cols * dst.elemSize();
vector<pair<size_t , const void *> > args;
args.push_back( make_pair( sizeof(cl_mem), (void *)&src1.data ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src1.step ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src1.offset ));
args.push_back( make_pair( sizeof(cl_mem), (void *)&src2.data ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src2.step ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src2.offset ));
args.push_back( make_pair( sizeof(cl_mem), (void *)&dst.data ));
args.push_back( make_pair( sizeof(cl_int), (void *)&dst.step ));
args.push_back( make_pair( sizeof(cl_int), (void *)&dst.offset ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src1.rows ));
args.push_back( make_pair( sizeof(cl_int), (void *)&cols ));
args.push_back( make_pair( sizeof(cl_int), (void *)&dst_step1 ));
T scalar;
if(_scalar != NULL)
{
double scalar1 = *((double *)_scalar);
scalar = (T)scalar1;
args.push_back( make_pair( sizeof(T), (void *)&scalar ));
}
switch(op_type)
{
case MAT_ADD:
openCLExecuteKernel(clCxt, kernelString, kernelName, globalThreads, localThreads, args, -1, depth, "-D ARITHM_ADD");
break;
case MAT_SUB:
openCLExecuteKernel(clCxt, kernelString, kernelName, globalThreads, localThreads, args, -1, depth, "-D ARITHM_SUB");
break;
default:
openCLExecuteKernel(clCxt, kernelString, kernelName, globalThreads, localThreads, args, -1, depth);
}
}
static void arithmetic_run(const oclMat &src1, const oclMat &src2, oclMat &dst,
string kernelName, const char **kernelString, int op_type = 0)
{
arithmetic_run<char>(src1, src2, dst, kernelName, kernelString, (void *)NULL, op_type);
}
static void arithmetic_run(const oclMat &src1, const oclMat &src2, oclMat &dst, const oclMat &mask,
string kernelName, const char **kernelString, int op_type = 0)
static void arithmetic_run_generic(const oclMat &src1, const oclMat &src2, const Scalar & scalar, const oclMat & mask,
oclMat &dst, int op_type, bool use_scalar = false)
{
if(!src1.clCxt->supportsFeature(Context::CL_DOUBLE) && src1.type() == CV_64F)
Context *clCxt = src1.clCxt;
bool hasDouble = clCxt->supportsFeature(Context::CL_DOUBLE);
if (!hasDouble && (src1.depth() == CV_64F || src2.depth() == CV_64F || dst.depth() == CV_64F))
{
CV_Error(CV_GpuNotSupported, "Selected device don't support double\r\n");
CV_Error(CV_GpuNotSupported, "Selected device doesn't support double\r\n");
return;
}
dst.create(src1.size(), src1.type());
CV_Assert(src1.cols == src2.cols && src2.cols == dst.cols &&
src1.rows == src2.rows && src2.rows == dst.rows &&
src1.rows == mask.rows && src1.cols == mask.cols);
CV_Assert(src1.type() == src2.type() && src1.type() == dst.type());
CV_Assert(src1.depth() != CV_8S);
CV_Assert(mask.type() == CV_8U);
CV_Assert(src2.empty() || (!src2.empty() && src1.type() == src2.type() && src1.size() == src2.size()));
CV_Assert(mask.empty() || (!mask.empty() && mask.type() == CV_8UC1 && mask.size() == src1.size()));
CV_Assert(op_type >= ADD && op_type <= ABS_DIFF);
Context *clCxt = src1.clCxt;
int channels = dst.oclchannels();
int depth = dst.depth();
dst.create(src1.size(), src1.type());
int vector_lengths[4][7] = {{4, 4, 2, 2, 1, 1, 1},
{2, 2, 1, 1, 1, 1, 1},
{4, 4, 2, 2 , 1, 1, 1},
{1, 1, 1, 1, 1, 1, 1}
};
int oclChannels = src1.oclchannels(), depth = src1.depth();
int src1step1 = src1.step / src1.elemSize(), src1offset1 = src1.offset / src1.elemSize();
int src2step1 = src2.step / src2.elemSize(), src2offset1 = src2.offset / src2.elemSize();
int maskstep1 = mask.step, maskoffset1 = mask.offset / mask.elemSize();
int dststep1 = dst.step / dst.elemSize(), dstoffset1 = dst.offset / dst.elemSize();
oclMat m;
size_t vector_length = vector_lengths[channels - 1][depth];
int offset_cols = ((dst.offset % dst.step) / dst.elemSize()) & (vector_length - 1);
int cols = divUp(dst.cols + offset_cols, vector_length);
size_t localThreads[3] = { 16, 16, 1 };
size_t globalThreads[3] = { dst.cols, dst.rows, 1 };
size_t localThreads[3] = { 64, 4, 1 };
size_t globalThreads[3] = { cols, dst.rows, 1 };
std::string kernelName = op_type == ABS_DIFF ? "arithm_absdiff" : "arithm_binary_op";
const char * const typeMap[] = { "uchar", "char", "ushort", "short", "int", "float", "double" };
const char * const WTypeMap[] = { "short", "short", "int", "int", "int", "float", "double" };
const char operationsMap[] = { '+', '-', '*', '/', '-' };
const char * const channelMap[] = { "", "", "2", "4", "4" };
bool haveScalar = use_scalar || src2.empty();
int WDepth = depth;
if (haveScalar)
WDepth = hasDouble && WDepth == CV_64F ? CV_64F : CV_32F;
if (op_type == DIV)
WDepth = hasDouble ? CV_64F : CV_32F;
else if (op_type == MUL)
WDepth = hasDouble && (depth == CV_32S || depth == CV_64F) ? CV_64F : CV_32F;
std::string buildOptions = format("-D T=%s%s -D WT=%s%s -D convertToT=convert_%s%s%s -D Operation=%c"
" -D convertToWT=convert_%s%s",
typeMap[depth], channelMap[oclChannels],
WTypeMap[WDepth], channelMap[oclChannels],
typeMap[depth], channelMap[oclChannels], (depth >= CV_32F ? "" : (depth == CV_32S ? "_rte" : "_sat_rte")),
operationsMap[op_type], WTypeMap[WDepth], channelMap[oclChannels]);
int dst_step1 = dst.cols * dst.elemSize();
vector<pair<size_t , const void *> > args;
args.push_back( make_pair( sizeof(cl_mem), (void *)&src1.data ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src1.step ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src1.offset ));
args.push_back( make_pair( sizeof(cl_mem), (void *)&src2.data ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src2.step ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src2.offset ));
args.push_back( make_pair( sizeof(cl_mem), (void *)&mask.data ));
args.push_back( make_pair( sizeof(cl_int), (void *)&mask.step ));
args.push_back( make_pair( sizeof(cl_int), (void *)&mask.offset ));
args.push_back( make_pair( sizeof(cl_mem), (void *)&dst.data ));
args.push_back( make_pair( sizeof(cl_int), (void *)&dst.step ));
args.push_back( make_pair( sizeof(cl_int), (void *)&dst.offset ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src1.rows ));
args.push_back( make_pair( sizeof(cl_int), (void *)&cols ));
args.push_back( make_pair( sizeof(cl_int), (void *)&dst_step1 ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src1step1 ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src1offset1 ));
switch (op_type)
if (!src2.empty())
{
case MAT_ADD:
openCLExecuteKernel(clCxt, kernelString, kernelName, globalThreads, localThreads, args, channels, depth, "-D ARITHM_ADD");
break;
case MAT_SUB:
openCLExecuteKernel(clCxt, kernelString, kernelName, globalThreads, localThreads, args, channels, depth, "-D ARITHM_SUB");
break;
default:
openCLExecuteKernel(clCxt, kernelString, kernelName, globalThreads, localThreads, args, channels, depth);
}
}
void cv::ocl::add(const oclMat &src1, const oclMat &src2, oclMat &dst)
{
arithmetic_run(src1, src2, dst, "arithm_add", &arithm_add, MAT_ADD);
}
void cv::ocl::add(const oclMat &src1, const oclMat &src2, oclMat &dst, const oclMat &mask)
{
arithmetic_run(src1, src2, dst, mask, "arithm_add_with_mask", &arithm_add, MAT_ADD);
}
void cv::ocl::subtract(const oclMat &src1, const oclMat &src2, oclMat &dst)
{
arithmetic_run(src1, src2, dst, "arithm_add", &arithm_add, MAT_SUB);
}
void cv::ocl::subtract(const oclMat &src1, const oclMat &src2, oclMat &dst, const oclMat &mask)
{
arithmetic_run(src1, src2, dst, mask, "arithm_add_with_mask", &arithm_add, MAT_SUB);
}
typedef void (*MulDivFunc)(const oclMat &src1, const oclMat &src2, oclMat &dst, string kernelName,
const char **kernelString, void *scalar);
void cv::ocl::multiply(const oclMat &src1, const oclMat &src2, oclMat &dst, double scalar)
{
if(src1.clCxt->supportsFeature(Context::CL_DOUBLE) && (src1.depth() == CV_64F))
arithmetic_run<double>(src1, src2, dst, "arithm_mul", &arithm_mul, (void *)(&scalar));
else
arithmetic_run<float>(src1, src2, dst, "arithm_mul", &arithm_mul, (void *)(&scalar));
}
void cv::ocl::divide(const oclMat &src1, const oclMat &src2, oclMat &dst, double scalar)
{
if(src1.clCxt->supportsFeature(Context::CL_DOUBLE))
arithmetic_run<double>(src1, src2, dst, "arithm_div", &arithm_div, (void *)(&scalar));
else
arithmetic_run<float>(src1, src2, dst, "arithm_div", &arithm_div, (void *)(&scalar));
args.push_back( make_pair( sizeof(cl_mem), (void *)&src2.data ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src2step1 ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src2offset1 ));
}
template <typename WT , typename CL_WT>
void arithmetic_scalar_run(const oclMat &src1, const Scalar &src2, oclMat &dst, const oclMat &mask, string kernelName, const char **kernelString, int isMatSubScalar)
{
if(!src1.clCxt->supportsFeature(Context::CL_DOUBLE) && src1.type() == CV_64F)
{
CV_Error(CV_GpuNotSupported, "Selected device don't support double\r\n");
return;
kernelName += "_mat";
}
dst.create(src1.size(), src1.type());
CV_Assert(src1.cols == dst.cols && src1.rows == dst.rows &&
src1.type() == dst.type());
//CV_Assert(src1.depth() != CV_8S);
if(mask.data)
if (haveScalar)
{
CV_Assert(mask.type() == CV_8U && src1.rows == mask.rows && src1.cols == mask.cols);
}
Context *clCxt = src1.clCxt;
int channels = dst.oclchannels();
int depth = dst.depth();
WT s[4] = { saturate_cast<WT>(src2.val[0]), saturate_cast<WT>(src2.val[1]),
saturate_cast<WT>(src2.val[2]), saturate_cast<WT>(src2.val[3])
};
int vector_lengths[4][7] = {{4, 0, 2, 2, 1, 1, 1},
{2, 0, 1, 1, 1, 1, 1},
{4, 0, 2, 2 , 1, 1, 1},
{1, 0, 1, 1, 1, 1, 1}
};
const int WDepthMap[] = { CV_16S, CV_16S, CV_32S, CV_32S, CV_32S, CV_32F, CV_64F };
m.create(1, 1, CV_MAKE_TYPE(WDepthMap[WDepth], oclChannels));
m.setTo(scalar);
size_t vector_length = vector_lengths[channels - 1][depth];
int offset_cols = ((dst.offset % dst.step) / dst.elemSize()) & (vector_length - 1);
int cols = divUp(dst.cols + offset_cols, vector_length);
size_t localThreads[3] = { 64, 4, 1 };
size_t globalThreads[3] = { cols, dst.rows, 1 };
int dst_step1 = dst.cols * dst.elemSize();
vector<pair<size_t , const void *> > args;
args.push_back( make_pair( sizeof(cl_mem) , (void *)&src1.data ));
args.push_back( make_pair( sizeof(cl_int) , (void *)&src1.step ));
args.push_back( make_pair( sizeof(cl_int) , (void *)&src1.offset));
args.push_back( make_pair( sizeof(cl_mem) , (void *)&dst.data ));
args.push_back( make_pair( sizeof(cl_int) , (void *)&dst.step ));
args.push_back( make_pair( sizeof(cl_int) , (void *)&dst.offset));
args.push_back( make_pair( sizeof(cl_mem), (void *)&m.data ));
if(mask.data)
{
args.push_back( make_pair( sizeof(cl_mem) , (void *)&mask.data ));
args.push_back( make_pair( sizeof(cl_int) , (void *)&mask.step ));
args.push_back( make_pair( sizeof(cl_int) , (void *)&mask.offset));
kernelName += "_scalar";
}
args.push_back( make_pair( sizeof(CL_WT) , (void *)&s ));
args.push_back( make_pair( sizeof(cl_int) , (void *)&src1.rows ));
args.push_back( make_pair( sizeof(cl_int) , (void *)&cols ));
args.push_back( make_pair( sizeof(cl_int) , (void *)&dst_step1 ));
if(isMatSubScalar != 0)
openCLExecuteKernel(clCxt, kernelString, kernelName, globalThreads, localThreads, args, channels, depth, "-D ARITHM_SUB");
else
openCLExecuteKernel(clCxt, kernelString, kernelName, globalThreads, localThreads, args, channels, depth, "-D ARITHM_ADD");
}
static void arithmetic_scalar_run(const oclMat &src, oclMat &dst, string kernelName, const char **kernelString, double scalar)
{
if(!src.clCxt->supportsFeature(Context::CL_DOUBLE) && src.type() == CV_64F)
if (!mask.empty())
{
CV_Error(CV_GpuNotSupported, "Selected device don't support double\r\n");
return;
}
dst.create(src.size(), src.type());
CV_Assert(src.cols == dst.cols && src.rows == dst.rows);
CV_Assert(src.type() == dst.type());
CV_Assert(src.depth() != CV_8S);
Context *clCxt = src.clCxt;
int channels = dst.oclchannels();
int depth = dst.depth();
int vector_lengths[4][7] = {{4, 0, 4, 4, 1, 1, 1},
{4, 0, 4, 4, 1, 1, 1},
{4, 0, 4, 4 , 1, 1, 1},
{4, 0, 4, 4, 1, 1, 1}
};
args.push_back( make_pair( sizeof(cl_mem), (void *)&mask.data ));
args.push_back( make_pair( sizeof(cl_int), (void *)&maskstep1 ));
args.push_back( make_pair( sizeof(cl_int), (void *)&maskoffset1 ));
size_t vector_length = vector_lengths[channels - 1][depth];
int offset_cols = (dst.offset / dst.elemSize1()) & (vector_length - 1);
int cols = divUp(dst.cols * channels + offset_cols, vector_length);
kernelName += "_mask";
}
size_t localThreads[3] = { 64, 4, 1 };
size_t globalThreads[3] = { cols, dst.rows, 1 };
if (op_type == DIV)
kernelName += "_div";
int dst_step1 = dst.cols * dst.elemSize();
vector<pair<size_t , const void *> > args;
args.push_back( make_pair( sizeof(cl_mem), (void *)&src.data ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src.step ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src.offset ));
args.push_back( make_pair( sizeof(cl_mem), (void *)&dst.data ));
args.push_back( make_pair( sizeof(cl_int), (void *)&dst.step ));
args.push_back( make_pair( sizeof(cl_int), (void *)&dst.offset ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src.rows ));
args.push_back( make_pair( sizeof(cl_int), (void *)&cols ));
args.push_back( make_pair( sizeof(cl_int), (void *)&dst_step1 ));
args.push_back( make_pair( sizeof(cl_int), (void *)&dststep1 ));
args.push_back( make_pair( sizeof(cl_int), (void *)&dstoffset1 ));
float f_scalar = (float)scalar;
if(src.clCxt->supportsFeature(Context::CL_DOUBLE))
args.push_back( make_pair( sizeof(cl_double), (void *)&scalar ));
else
{
args.push_back( make_pair( sizeof(cl_float), (void *)&f_scalar));
}
args.push_back( make_pair( sizeof(cl_int), (void *)&src1.cols ));
args.push_back( make_pair( sizeof(cl_int), (void *)&src1.rows ));
openCLExecuteKernel(clCxt, kernelString, kernelName, globalThreads, localThreads, args, -1, depth);
openCLExecuteKernel(clCxt, mask.empty() ?
(!src2.empty() ? &arithm_add : &arithm_add_scalar) :
(!src2.empty() ? &arithm_add_mask : &arithm_add_scalar_mask),
kernelName, globalThreads, localThreads,
args, -1, -1, buildOptions.c_str());
}
typedef void (*ArithmeticFuncS)(const oclMat &src1, const Scalar &src2, oclMat &dst, const oclMat &mask, string kernelName, const char **kernelString, int isMatSubScalar);
static void arithmetic_scalar(const oclMat &src1, const Scalar &src2, oclMat &dst, const oclMat &mask, string kernelName, const char **kernelString, int isMatSubScalar)
{
static ArithmeticFuncS tab[8] =
{
arithmetic_scalar_run<int, cl_int4>,
arithmetic_scalar_run<int, cl_int4>,
arithmetic_scalar_run<int, cl_int4>,
arithmetic_scalar_run<int, cl_int4>,
arithmetic_scalar_run<int, cl_int4>,
arithmetic_scalar_run<float, cl_float4>,
arithmetic_scalar_run<double, cl_double4>,
0
};
ArithmeticFuncS func = tab[src1.depth()];
if(func == 0)
cv::ocl::error("Unsupported arithmetic operation", __FILE__, __LINE__);
func(src1, src2, dst, mask, kernelName, kernelString, isMatSubScalar);
}
static void arithmetic_scalar(const oclMat &src1, const Scalar &src2, oclMat &dst, const oclMat &mask, string kernelName, const char **kernelString)
void cv::ocl::add(const oclMat &src1, const oclMat &src2, oclMat &dst, const oclMat &mask)
{
arithmetic_scalar(src1, src2, dst, mask, kernelName, kernelString, 0);
arithmetic_run_generic(src1, src2, Scalar(), mask, dst, ADD);
}
void cv::ocl::add(const oclMat &src1, const Scalar &src2, oclMat &dst, const oclMat &mask)
{
string kernelName = mask.data ? "arithm_s_add_with_mask" : "arithm_s_add";
const char **kernelString = mask.data ? &arithm_add_scalar_mask : &arithm_add_scalar;
arithmetic_run_generic(src1, oclMat(), src2, mask, dst, ADD);
}
arithmetic_scalar( src1, src2, dst, mask, kernelName, kernelString);
void cv::ocl::subtract(const oclMat &src1, const oclMat &src2, oclMat &dst, const oclMat &mask)
{
arithmetic_run_generic(src1, src2, Scalar(), mask, dst, SUB);
}
void cv::ocl::subtract(const oclMat &src1, const Scalar &src2, oclMat &dst, const oclMat &mask)
{
string kernelName = mask.data ? "arithm_s_add_with_mask" : "arithm_s_add";
const char **kernelString = mask.data ? &arithm_add_scalar_mask : &arithm_add_scalar;
arithmetic_scalar( src1, src2, dst, mask, kernelName, kernelString, 1);
arithmetic_run_generic(src1, oclMat(), src2, mask, dst, SUB);
}
void cv::ocl::subtract(const Scalar &src2, const oclMat &src1, oclMat &dst, const oclMat &mask)
void cv::ocl::multiply(const oclMat &src1, const oclMat &src2, oclMat &dst, double scalar)
{
string kernelName = mask.data ? "arithm_s_add_with_mask" : "arithm_s_add";
const char **kernelString = mask.data ? &arithm_add_scalar_mask : &arithm_add_scalar;
arithmetic_scalar( src1, src2, dst, mask, kernelName, kernelString, -1);
const bool use_scalar = !(std::abs(scalar - 1.0) < std::numeric_limits<double>::epsilon());
arithmetic_run_generic(src1, src2, Scalar::all(scalar), oclMat(), dst, MUL, use_scalar);
}
void cv::ocl::multiply(double scalar, const oclMat &src, oclMat &dst)
{
string kernelName = "arithm_muls";
arithmetic_scalar_run( src, dst, kernelName, &arithm_mul, scalar);
arithmetic_run_generic(src, oclMat(), Scalar::all(scalar), oclMat(), dst, MUL);
}
void cv::ocl::divide(double scalar, const oclMat &src, oclMat &dst)
void cv::ocl::divide(const oclMat &src1, const oclMat &src2, oclMat &dst, double scalar)
{
if(!src.clCxt->supportsFeature(Context::CL_DOUBLE))
{
CV_Error(CV_GpuNotSupported, "Selected device don't support double\r\n");
return;
}
const bool use_scalar = !(std::abs(scalar - 1.0) < std::numeric_limits<double>::epsilon());
arithmetic_run_generic(src1, src2, Scalar::all(scalar), oclMat(), dst, DIV, use_scalar);
}
string kernelName = "arithm_s_div";
arithmetic_scalar_run(src, dst, kernelName, &arithm_div, scalar);
void cv::ocl::divide(double scalar, const oclMat &src, oclMat &dst)
{
arithmetic_run_generic(src, oclMat(), Scalar::all(scalar), oclMat(), dst, DIV);
}
//////////////////////////////////////////////////////////////////////////////
///////////////////////////////// Absdiff ///////////////////////////////////
//////////////////////////////////////////////////////////////////////////////
void cv::ocl::absdiff(const oclMat &src1, const oclMat &src2, oclMat &dst)
{
arithmetic_run(src1, src2, dst, "arithm_absdiff", &arithm_absdiff);
arithmetic_run_generic(src1, src2, Scalar(), oclMat(), dst, ABS_DIFF);
}
void cv::ocl::absdiff(const oclMat &src1, const Scalar &src2, oclMat &dst)
{
string kernelName = "arithm_s_absdiff";
oclMat mask;
arithmetic_scalar( src1, src2, dst, mask, kernelName, &arithm_absdiff);
arithmetic_run_generic(src1, oclMat(), src2, oclMat(), dst, ABS_DIFF);
}
//////////////////////////////////////////////////////////////////////////////
///////////////////////////////// compare ///////////////////////////////////
//////////////////////////////////////////////////////////////////////////////
......
modules/ocl/src/opencl/arithm_add.cl
View file @ 0ad03162
......@@ -52,809 +52,105 @@
#endif
#endif
#ifdef ARITHM_ADD
#define ARITHM_OP(A,B) ((A)+(B))
#elif defined ARITHM_SUB
#define ARITHM_OP(A,B) ((A)-(B))
#endif
//////////////////////////////////////////////////////////////////////////////////////////////////////
/////////////////////////////////////////////ADD////////////////////////////////////////////////////
///////////////////////////////////////////////////////////////////////////////////////////////////////
/**************************************add without mask**************************************/
__kernel void arithm_add_D0 (__global uchar *src1, int src1_step, int src1_offset,
__global uchar *src2, int src2_step, int src2_offset,
__global uchar *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align (dst_offset & 3)
int src1_index = mad24(y, src1_step, x + src1_offset - dst_align);
int src2_index = mad24(y, src2_step, x + src2_offset - dst_align);
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + x & (int)0xfffffffc);
int src1_index_fix = src1_index < 0 ? 0 : src1_index;
int src2_index_fix = src2_index < 0 ? 0 : src2_index;
uchar4 src1_data = vload4(0, src1 + src1_index_fix);
uchar4 src2_data = vload4(0, src2 + src2_index_fix);
if(src1_index < 0)
{
uchar4 tmp;
tmp.xyzw = (src1_index == -2) ? src1_data.zwxy:src1_data.yzwx;
src1_data.xyzw = (src1_index == -1) ? src1_data.wxyz:tmp.xyzw;
}
if(src2_index < 0)
{
uchar4 tmp;
tmp.xyzw = (src2_index == -2) ? src2_data.zwxy:src2_data.yzwx;
src2_data.xyzw = (src2_index == -1) ? src2_data.wxyz:tmp.xyzw;
}
uchar4 dst_data = *((__global uchar4 *)(dst + dst_index));
short4 tmp = ARITHM_OP(convert_short4_sat(src1_data), convert_short4_sat(src2_data));
uchar4 tmp_data = convert_uchar4_sat(tmp);
dst_data.x = ((dst_index + 0 >= dst_start) && (dst_index + 0 < dst_end)) ? tmp_data.x : dst_data.x;
dst_data.y = ((dst_index + 1 >= dst_start) && (dst_index + 1 < dst_end)) ? tmp_data.y : dst_data.y;
dst_data.z = ((dst_index + 2 >= dst_start) && (dst_index + 2 < dst_end)) ? tmp_data.z : dst_data.z;
dst_data.w = ((dst_index + 3 >= dst_start) && (dst_index + 3 < dst_end)) ? tmp_data.w : dst_data.w;
*((__global uchar4 *)(dst + dst_index)) = dst_data;
}
}
__kernel void arithm_add_D2 (__global ushort *src1, int src1_step, int src1_offset,
__global ushort *src2, int src2_step, int src2_offset,
__global ushort *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset / 2) & 3)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int src2_index = mad24(y, src2_step, (x << 1) + src2_offset - (dst_align << 1));
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffff8);
ushort4 src1_data = vload4(0, (__global ushort *)((__global char *)src1 + src1_index));
ushort4 src2_data = vload4(0, (__global ushort *)((__global char *)src2 + src2_index));
ushort4 dst_data = *((__global ushort4 *)((__global char *)dst + dst_index));
int4 tmp = ARITHM_OP(convert_int4_sat(src1_data), convert_int4_sat(src2_data));
ushort4 tmp_data = convert_ushort4_sat(tmp);
dst_data.x = ((dst_index + 0 >= dst_start) && (dst_index + 0 < dst_end)) ? tmp_data.x : dst_data.x;
dst_data.y = ((dst_index + 2 >= dst_start) && (dst_index + 2 < dst_end)) ? tmp_data.y : dst_data.y;
dst_data.z = ((dst_index + 4 >= dst_start) && (dst_index + 4 < dst_end)) ? tmp_data.z : dst_data.z;
dst_data.w = ((dst_index + 6 >= dst_start) && (dst_index + 6 < dst_end)) ? tmp_data.w : dst_data.w;
*((__global ushort4 *)((__global char *)dst + dst_index)) = dst_data;
}
}
__kernel void arithm_add_D3 (__global short *src1, int src1_step, int src1_offset,
__global short *src2, int src2_step, int src2_offset,
__global short *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset / 2) & 3)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int src2_index = mad24(y, src2_step, (x << 1) + src2_offset - (dst_align << 1));
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffff8);
short4 src1_data = vload4(0, (__global short *)((__global char *)src1 + src1_index));
short4 src2_data = vload4(0, (__global short *)((__global char *)src2 + src2_index));
short4 dst_data = *((__global short4 *)((__global char *)dst + dst_index));
int4 tmp = ARITHM_OP(convert_int4_sat(src1_data), convert_int4_sat(src2_data));
short4 tmp_data = convert_short4_sat(tmp);
dst_data.x = ((dst_index + 0 >= dst_start) && (dst_index + 0 < dst_end)) ? tmp_data.x : dst_data.x;
dst_data.y = ((dst_index + 2 >= dst_start) && (dst_index + 2 < dst_end)) ? tmp_data.y : dst_data.y;
dst_data.z = ((dst_index + 4 >= dst_start) && (dst_index + 4 < dst_end)) ? tmp_data.z : dst_data.z;
dst_data.w = ((dst_index + 6 >= dst_start) && (dst_index + 6 < dst_end)) ? tmp_data.w : dst_data.w;
*((__global short4 *)((__global char *)dst + dst_index)) = dst_data;
}
}
__kernel void arithm_add_D4 (__global int *src1, int src1_step, int src1_offset,
__global int *src2, int src2_step, int src2_offset,
__global int *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 2) + src2_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
int data1 = *((__global int *)((__global char *)src1 + src1_index));
int data2 = *((__global int *)((__global char *)src2 + src2_index));
long tmp = ARITHM_OP((long)(data1), (long)(data2));
*((__global int *)((__global char *)dst + dst_index)) = convert_int_sat(tmp);
}
}
__kernel void arithm_add_D5 (__global float *src1, int src1_step, int src1_offset,
__global float *src2, int src2_step, int src2_offset,
__global float *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 2) + src2_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
float data1 = *((__global float *)((__global char *)src1 + src1_index));
float data2 = *((__global float *)((__global char *)src2 + src2_index));
float tmp = ARITHM_OP(data1, data2);
*((__global float *)((__global char *)dst + dst_index)) = tmp;
}
}
#if defined (DOUBLE_SUPPORT)
__kernel void arithm_add_D6 (__global double *src1, int src1_step, int src1_offset,
__global double *src2, int src2_step, int src2_offset,
__global double *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 3) + src2_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
double data1 = *((__global double *)((__global char *)src1 + src1_index));
double data2 = *((__global double *)((__global char *)src2 + src2_index));
*((__global double *)((__global char *)dst + dst_index)) = ARITHM_OP(data1, data2);
}
}
#endif
/**************************************add with mask**************************************/
__kernel void arithm_add_with_mask_C1_D0 (__global uchar *src1, int src1_step, int src1_offset,
__global uchar *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global uchar *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align (dst_offset & 3)
int src1_index = mad24(y, src1_step, x + src1_offset - dst_align);
int src2_index = mad24(y, src2_step, x + src2_offset - dst_align);
int mask_index = mad24(y, mask_step, x + mask_offset - dst_align);
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + x & (int)0xfffffffc);
int src1_index_fix = src1_index < 0 ? 0 : src1_index;
int src2_index_fix = src2_index < 0 ? 0 : src2_index;
int mask_index_fix = mask_index < 0 ? 0 : mask_index;
uchar4 src1_data = vload4(0, src1 + src1_index_fix);
uchar4 src2_data = vload4(0, src2 + src2_index_fix);
uchar4 mask_data = vload4(0, mask + mask_index_fix);
if(src1_index < 0)
{
uchar4 tmp;
tmp.xyzw = (src1_index == -2) ? src1_data.zwxy:src1_data.yzwx;
src1_data.xyzw = (src1_index == -1) ? src1_data.wxyz:tmp.xyzw;
}
if(src2_index < 0)
{
uchar4 tmp;
tmp.xyzw = (src2_index == -2) ? src2_data.zwxy:src2_data.yzwx;
src2_data.xyzw = (src2_index == -1) ? src2_data.wxyz:tmp.xyzw;
}
if(mask_index < 0)
{
uchar4 tmp;
tmp.xyzw = (mask_index == -2) ? mask_data.zwxy:mask_data.yzwx;
mask_data.xyzw = (mask_index == -1) ? mask_data.wxyz:tmp.xyzw;
}
uchar4 data = *((__global uchar4 *)(dst + dst_index));
short4 tmp = ARITHM_OP(convert_short4_sat(src1_data), convert_short4_sat(src2_data));
uchar4 tmp_data = convert_uchar4_sat(tmp);
data.x = ((mask_data.x) && (dst_index + 0 >= dst_start) && (dst_index + 0 < dst_end)) ? tmp_data.x : data.x;
data.y = ((mask_data.y) && (dst_index + 1 >= dst_start) && (dst_index + 1 < dst_end)) ? tmp_data.y : data.y;
data.z = ((mask_data.z) && (dst_index + 2 >= dst_start) && (dst_index + 2 < dst_end)) ? tmp_data.z : data.z;
data.w = ((mask_data.w) && (dst_index + 3 >= dst_start) && (dst_index + 3 < dst_end)) ? tmp_data.w : data.w;
*((__global uchar4 *)(dst + dst_index)) = data;
}
}
__kernel void arithm_add_with_mask_C1_D2 (__global ushort *src1, int src1_step, int src1_offset,
__global ushort *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global ushort *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 1;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset / 2) & 1)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int src2_index = mad24(y, src2_step, (x << 1) + src2_offset - (dst_align << 1));
int mask_index = mad24(y, mask_step, x + mask_offset - dst_align);
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffffc);
ushort2 src1_data = vload2(0, (__global ushort *)((__global char *)src1 + src1_index));
ushort2 src2_data = vload2(0, (__global ushort *)((__global char *)src2 + src2_index));
uchar2 mask_data = vload2(0, mask + mask_index);
ushort2 data = *((__global ushort2 *)((__global uchar *)dst + dst_index));
int2 tmp = ARITHM_OP(convert_int2_sat(src1_data), convert_int2_sat(src2_data));
ushort2 tmp_data = convert_ushort2_sat(tmp);
data.x = ((mask_data.x) && (dst_index + 0 >= dst_start)) ? tmp_data.x : data.x;
data.y = ((mask_data.y) && (dst_index + 2 < dst_end )) ? tmp_data.y : data.y;
*((__global ushort2 *)((__global uchar *)dst + dst_index)) = data;
}
}
__kernel void arithm_add_with_mask_C1_D3 (__global short *src1, int src1_step, int src1_offset,
__global short *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global short *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 1;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset / 2) & 1)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int src2_index = mad24(y, src2_step, (x << 1) + src2_offset - (dst_align << 1));
int mask_index = mad24(y, mask_step, x + mask_offset - dst_align);
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffffc);
short2 src1_data = vload2(0, (__global short *)((__global char *)src1 + src1_index));
short2 src2_data = vload2(0, (__global short *)((__global char *)src2 + src2_index));
uchar2 mask_data = vload2(0, mask + mask_index);
short2 data = *((__global short2 *)((__global uchar *)dst + dst_index));
int2 tmp = ARITHM_OP(convert_int2_sat(src1_data), convert_int2_sat(src2_data));
short2 tmp_data = convert_short2_sat(tmp);
data.x = ((mask_data.x) && (dst_index + 0 >= dst_start)) ? tmp_data.x : data.x;
data.y = ((mask_data.y) && (dst_index + 2 < dst_end )) ? tmp_data.y : data.y;
*((__global short2 *)((__global uchar *)dst + dst_index)) = data;
}
}
__kernel void arithm_add_with_mask_C1_D4 (__global int *src1, int src1_step, int src1_offset,
__global int *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global int *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 2) + src2_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
uchar mask_data = *(mask + mask_index);
int src_data1 = *((__global int *)((__global char *)src1 + src1_index));
int src_data2 = *((__global int *)((__global char *)src2 + src2_index));
int dst_data = *((__global int *)((__global char *)dst + dst_index));
int data = convert_int_sat(ARITHM_OP((long)src_data1, (long)src_data2));
data = mask_data ? data : dst_data;
*((__global int *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_add_with_mask_C1_D5 (__global float *src1, int src1_step, int src1_offset,
__global float *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global float *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 2) + src2_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
uchar mask_data = *(mask + mask_index);
float src_data1 = *((__global float *)((__global char *)src1 + src1_index));
float src_data2 = *((__global float *)((__global char *)src2 + src2_index));
float dst_data = *((__global float *)((__global char *)dst + dst_index));
float data = ARITHM_OP(src_data1, src_data2);
data = mask_data ? data : dst_data;
*((__global float *)((__global char *)dst + dst_index)) = data;
}
}
#if defined (DOUBLE_SUPPORT)
__kernel void arithm_add_with_mask_C1_D6 (__global double *src1, int src1_step, int src1_offset,
__global double *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global double *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 3) + src2_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
uchar mask_data = *(mask + mask_index);
double src_data1 = *((__global double *)((__global char *)src1 + src1_index));
double src_data2 = *((__global double *)((__global char *)src2 + src2_index));
double dst_data = *((__global double *)((__global char *)dst + dst_index));
double data = ARITHM_OP(src_data1, src_data2);
data = mask_data ? data : dst_data;
*((__global double *)((__global char *)dst + dst_index)) = data;
}
}
#endif
__kernel void arithm_add_with_mask_C2_D0 (__global uchar *src1, int src1_step, int src1_offset,
__global uchar *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global uchar *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 1;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset / 2) & 1)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int src2_index = mad24(y, src2_step, (x << 1) + src2_offset - (dst_align << 1));
int mask_index = mad24(y, mask_step, x + mask_offset - dst_align);
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffffc);
uchar4 src1_data = vload4(0, src1 + src1_index);
uchar4 src2_data = vload4(0, src2 + src2_index);
uchar2 mask_data = vload2(0, mask + mask_index);
uchar4 data = *((__global uchar4 *)(dst + dst_index));
short4 tmp = ARITHM_OP(convert_short4_sat(src1_data), convert_short4_sat(src2_data));
uchar4 tmp_data = convert_uchar4_sat(tmp);
data.xy = ((mask_data.x) && (dst_index + 0 >= dst_start)) ? tmp_data.xy : data.xy;
data.zw = ((mask_data.y) && (dst_index + 2 < dst_end )) ? tmp_data.zw : data.zw;
*((__global uchar4 *)(dst + dst_index)) = data;
}
}
__kernel void arithm_add_with_mask_C2_D2 (__global ushort *src1, int src1_step, int src1_offset,
__global ushort *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global ushort *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 2) + src2_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
uchar mask_data = *(mask + mask_index);
ushort2 src_data1 = *((__global ushort2 *)((__global char *)src1 + src1_index));
ushort2 src_data2 = *((__global ushort2 *)((__global char *)src2 + src2_index));
ushort2 dst_data = *((__global ushort2 *)((__global char *)dst + dst_index));
int2 tmp = ARITHM_OP(convert_int2_sat(src_data1), convert_int2_sat(src_data2));
ushort2 data = convert_ushort2_sat(tmp);
data = mask_data ? data : dst_data;
*((__global ushort2 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_add_with_mask_C2_D3 (__global short *src1, int src1_step, int src1_offset,
__global short *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global short *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 2) + src2_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
uchar mask_data = *(mask + mask_index);
short2 src_data1 = *((__global short2 *)((__global char *)src1 + src1_index));
short2 src_data2 = *((__global short2 *)((__global char *)src2 + src2_index));
short2 dst_data = *((__global short2 *)((__global char *)dst + dst_index));
int2 tmp = ARITHM_OP(convert_int2_sat(src_data1), convert_int2_sat(src_data2));
short2 data = convert_short2_sat(tmp);
data = mask_data ? data : dst_data;
*((__global short2 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_add_with_mask_C2_D4 (__global int *src1, int src1_step, int src1_offset,
__global int *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global int *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 3) + src2_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
uchar mask_data = *(mask + mask_index);
int2 src_data1 = *((__global int2 *)((__global char *)src1 + src1_index));
int2 src_data2 = *((__global int2 *)((__global char *)src2 + src2_index));
int2 dst_data = *((__global int2 *)((__global char *)dst + dst_index));
int2 data = convert_int2_sat(ARITHM_OP(convert_long2_sat(src_data1), convert_long2_sat(src_data2)));
data = mask_data ? data : dst_data;
*((__global int2 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_add_with_mask_C2_D5 (__global float *src1, int src1_step, int src1_offset,
__global float *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global float *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 3) + src2_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
uchar mask_data = *(mask + mask_index);
float2 src_data1 = *((__global float2 *)((__global char *)src1 + src1_index));
float2 src_data2 = *((__global float2 *)((__global char *)src2 + src2_index));
float2 dst_data = *((__global float2 *)((__global char *)dst + dst_index));
float2 data = ARITHM_OP(src_data1, src_data2);
data = mask_data ? data : dst_data;
*((__global float2 *)((__global char *)dst + dst_index)) = data;
}
}
///////////////////////////////////////////// ADD ////////////////////////////////////////////////////
//////////////////////////////////////////////////////////////////////////////////////////////////////
#if defined (DOUBLE_SUPPORT)
__kernel void arithm_add_with_mask_C2_D6 (__global double *src1, int src1_step, int src1_offset,
__global double *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global double *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
__kernel void arithm_binary_op_mat(__global T *src1, int src1_step, int src1_offset,
__global T *src2, int src2_step, int src2_offset,
__global T *dst, int dst_step, int dst_offset,
int cols, int rows)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 4) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 4) + src2_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 4) + dst_offset);
int src1_index = mad24(y, src1_step, x + src1_offset);
int src2_index = mad24(y, src2_step, x + src2_offset);
int dst_index = mad24(y, dst_step, x + dst_offset);
uchar mask_data = *(mask + mask_index);
double2 src_data1 = *((__global double2 *)((__global char *)src1 + src1_index));
double2 src_data2 = *((__global double2 *)((__global char *)src2 + src2_index));
double2 dst_data = *((__global double2 *)((__global char *)dst + dst_index));
double2 data = ARITHM_OP(src_data1, src_data2);
data = mask_data ? data : dst_data;
*((__global double2 *)((__global char *)dst + dst_index)) = data;
dst[dst_index] = convertToT(convertToWT(src1[src1_index]) Operation convertToWT(src2[src2_index]));
}
}
#endif
__kernel void arithm_add_with_mask_C4_D0 (__global uchar *src1, int src1_step, int src1_offset,
__global uchar *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global uchar *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
__kernel void arithm_binary_op_mat_div(__global T *src1, int src1_step, int src1_offset,
__global T *src2, int src2_step, int src2_offset,
__global T *dst, int dst_step, int dst_offset,
int cols, int rows)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 2) + src2_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
int src1_index = mad24(y, src1_step, x + src1_offset);
int src2_index = mad24(y, src2_step, x + src2_offset);
int dst_index = mad24(y, dst_step, x + dst_offset);
uchar mask_data = *(mask + mask_index);
uchar4 src_data1 = *((__global uchar4 *)(src1 + src1_index));
uchar4 src_data2 = *((__global uchar4 *)(src2 + src2_index));
uchar4 dst_data = *((__global uchar4 *)(dst + dst_index));
uchar4 data = convert_uchar4_sat(ARITHM_OP(convert_short4_sat(src_data1), convert_short4_sat(src_data2)));
data = mask_data ? data : dst_data;
*((__global uchar4 *)(dst + dst_index)) = data;
T zero = (T)(0);
dst[dst_index] = src2[src2_index] == zero ? zero : convertToT(convertToWT(src1[src1_index]) / convertToWT(src2[src2_index]));
}
}
__kernel void arithm_add_with_mask_C4_D2 (__global ushort *src1, int src1_step, int src1_offset,
__global ushort *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global ushort *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 3) + src2_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
uchar mask_data = *(mask + mask_index);
ushort4 src_data1 = *((__global ushort4 *)((__global char *)src1 + src1_index));
ushort4 src_data2 = *((__global ushort4 *)((__global char *)src2 + src2_index));
ushort4 dst_data = *((__global ushort4 *)((__global char *)dst + dst_index));
ushort4 data = convert_ushort4_sat(ARITHM_OP(convert_int4_sat(src_data1), convert_int4_sat(src_data2)));
data = mask_data ? data : dst_data;
*((__global ushort4 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_add_with_mask_C4_D3 (__global short *src1, int src1_step, int src1_offset,
__global short *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global short *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
__kernel void arithm_absdiff_mat(__global T *src1, int src1_step, int src1_offset,
__global T *src2, int src2_step, int src2_offset,
__global T *dst, int dst_step, int dst_offset,
int cols, int rows)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 3) + src2_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
uchar mask_data = *(mask + mask_index);
short4 src_data1 = *((__global short4 *)((__global char *)src1 + src1_index));
short4 src_data2 = *((__global short4 *)((__global char *)src2 + src2_index));
short4 dst_data = *((__global short4 *)((__global char *)dst + dst_index));
short4 data = convert_short4_sat(ARITHM_OP(convert_int4_sat(src_data1), convert_int4_sat(src_data2)));
data = mask_data ? data : dst_data;
int src1_index = mad24(y, src1_step, x + src1_offset);
int src2_index = mad24(y, src2_step, x + src2_offset);
int dst_index = mad24(y, dst_step, x + dst_offset);
*((__global short4 *)((__global char *)dst + dst_index)) = data;
WT value = convertToWT(src1[src1_index]) - convertToWT(src2[src2_index]);
value = value > (WT)(0) ? value : -value;
dst[dst_index] = convertToT(value);
}
}
__kernel void arithm_add_with_mask_C4_D4 (__global int *src1, int src1_step, int src1_offset,
__global int *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global int *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 4) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 4) + src2_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 4) + dst_offset);
uchar mask_data = *(mask + mask_index);
int4 src_data1 = *((__global int4 *)((__global char *)src1 + src1_index));
int4 src_data2 = *((__global int4 *)((__global char *)src2 + src2_index));
int4 dst_data = *((__global int4 *)((__global char *)dst + dst_index));
int4 data = convert_int4_sat(ARITHM_OP(convert_long4_sat(src_data1), convert_long4_sat(src_data2)));
data = mask_data ? data : dst_data;
*((__global int4 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_add_with_mask_C4_D5 (__global float *src1, int src1_step, int src1_offset,
__global float *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global float *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
// add mat with scale for multiply
__kernel void arithm_binary_op_mat_scalar(__global T *src1, int src1_step, int src1_offset,
__global T *src2, int src2_step, int src2_offset,
__global WT *scalar,
__global T *dst, int dst_step, int dst_offset,
int cols, int rows)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 4) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 4) + src2_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 4) + dst_offset);
int src1_index = mad24(y, src1_step, x + src1_offset);
int src2_index = mad24(y, src2_step, x + src2_offset);
int dst_index = mad24(y, dst_step, x + dst_offset);
uchar mask_data = *(mask + mask_index);
float4 src_data1 = *((__global float4 *)((__global char *)src1 + src1_index));
float4 src_data2 = *((__global float4 *)((__global char *)src2 + src2_index));
float4 dst_data = *((__global float4 *)((__global char *)dst + dst_index));
float4 data = ARITHM_OP(src_data1, src_data2);
data = mask_data ? data : dst_data;
*((__global float4 *)((__global char *)dst + dst_index)) = data;
dst[dst_index] = convertToT(convertToWT(src1[src1_index]) * scalar[0] * convertToWT(src2[src2_index]));
}
}
#if defined (DOUBLE_SUPPORT)
__kernel void arithm_add_with_mask_C4_D6 (__global double *src1, int src1_step, int src1_offset,
__global double *src2, int src2_step, int src2_offset,
__global uchar *mask, int mask_step, int mask_offset,
__global double *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1)
// add mat with scale for divide
__kernel void arithm_binary_op_mat_scalar_div(__global T *src1, int src1_step, int src1_offset,
__global T *src2, int src2_step, int src2_offset,
__global WT *scalar,
__global T *dst, int dst_step, int dst_offset,
int cols, int rows)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 5) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 5) + src2_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 5) + dst_offset);
int src1_index = mad24(y, src1_step, x + src1_offset);
int src2_index = mad24(y, src2_step, x + src2_offset);
int dst_index = mad24(y, dst_step, x + dst_offset);
uchar mask_data = *(mask + mask_index);
double4 src_data1 = *((__global double4 *)((__global char *)src1 + src1_index));
double4 src_data2 = *((__global double4 *)((__global char *)src2 + src2_index));
double4 dst_data = *((__global double4 *)((__global char *)dst + dst_index));
double4 data = ARITHM_OP(src_data1, src_data2);
data = mask_data ? data : dst_data;
*((__global double4 *)((__global char *)dst + dst_index)) = data;
T zero = (T)(0);
dst[dst_index] = src2[src2_index] == zero ? zero :
convertToT(convertToWT(src1[src1_index]) * scalar[0] / convertToWT(src2[src2_index]));
}
}
#endif
modules/ocl/src/opencl/arithm_add_mask.cl 0 → 100644
View file @ 0ad03162
/*M///////////////////////////////////////////////////////////////////////////////////////
//
// IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
//
// By downloading, copying, installing or using the software you agree to this license.
// If you do not agree to this license, do not download, install,
// copy or use the software.
//
//
// License Agreement
// For Open Source Computer Vision Library
//
// Copyright (C) 2010-2012, Institute Of Software Chinese Academy Of Science, all rights reserved.
// Copyright (C) 2010-2012, Advanced Micro Devices, Inc., all rights reserved.
// Third party copyrights are property of their respective owners.
//
// @Authors
// Jia Haipeng, jiahaipeng95@gmail.com
//
// Redistribution and use in source and binary forms, with or without modification,
// are permitted provided that the following conditions are met:
//
// * Redistribution's of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
//
// * Redistribution's in binary form must reproduce the above copyright notice,
// this list of conditions and the following disclaimer in the documentation
// and/or other oclMaterials provided with the distribution.
//
// * The name of the copyright holders may not be used to endorse or promote products
// derived from this software without specific prior written permission.
//
// This software is provided by the copyright holders and contributors as is and
// any express or implied warranties, including, but not limited to, the implied
// warranties of merchantability and fitness for a particular purpose are disclaimed.
// In no event shall the Intel Corporation or contributors be liable for any direct,
// indirect, incidental, special, exemplary, or consequential damages
// (including, but not limited to, procurement of substitute goods or services;
// loss of use, data, or profits; or business interruption) however caused
// and on any theory of liability, whether in contract, strict liability,
// or tort (including negligence or otherwise) arising in any way out of
// the use of this software, even if advised of the possibility of such damage.
//
//M*/
#if defined (DOUBLE_SUPPORT)
#ifdef cl_khr_fp64
#pragma OPENCL EXTENSION cl_khr_fp64:enable
#elif defined (cl_amd_fp64)
#pragma OPENCL EXTENSION cl_amd_fp64:enable
#endif
#endif
//////////////////////////////////////////////////////////////////////////////////
///////////////////////////////// add with mask //////////////////////////////////
//////////////////////////////////////////////////////////////////////////////////
__kernel void arithm_binary_op_mat_mask(__global T * src1, int src1_step, int src1_offset,
__global T * src2, int src2_step, int src2_offset,
__global uchar * mask, int mask_step, int mask_offset,
__global T * dst, int dst_step, int dst_offset,
int cols, int rows)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int mask_index = mad24(y, mask_step, x + mask_offset);
if (mask[mask_index])
{
int src1_index = mad24(y, src1_step, x + src1_offset);
int src2_index = mad24(y, src2_step, x + src2_offset);
int dst_index = mad24(y, dst_step, dst_offset + x);
dst[dst_index] = convertToT(convertToWT(src1[src1_index]) Operation convertToWT(src2[src2_index]));
}
}
}
modules/ocl/src/opencl/arithm_add_scalar.cl
View file @ 0ad03162
......@@ -51,463 +51,61 @@
#endif
#endif
#ifdef ARITHM_ADD
#define ARITHM_OP(A,B) ((A)+(B))
#elif defined ARITHM_SUB
#define ARITHM_OP(A,B) ((A)-(B))
#endif
/**************************************add with scalar without mask**************************************/
__kernel void arithm_s_add_C1_D0 (__global uchar *src1, int src1_step, int src1_offset,
__global uchar *dst, int dst_step, int dst_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align (dst_offset & 3)
int src1_index = mad24(y, src1_step, x + src1_offset - dst_align);
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + x & (int)0xfffffffc);
int src1_index_fix = src1_index < 0 ? 0 : src1_index;
uchar4 src1_data = vload4(0, src1 + src1_index_fix);
int4 src2_data = (int4)(src2.x, src2.x, src2.x, src2.x);
if(src1_index < 0)
{
uchar4 tmp;
tmp.xyzw = (src1_index == -2) ? src1_data.zwxy:src1_data.yzwx;
src1_data.xyzw = (src1_index == -1) ? src1_data.wxyz:tmp.xyzw;
}
uchar4 data = *((__global uchar4 *)(dst + dst_index));
int4 tmp = ARITHM_OP(convert_int4_sat(src1_data), src2_data);
uchar4 tmp_data = convert_uchar4_sat(tmp);
data.x = ((dst_index + 0 >= dst_start) && (dst_index + 0 < dst_end)) ? tmp_data.x : data.x;
data.y = ((dst_index + 1 >= dst_start) && (dst_index + 1 < dst_end)) ? tmp_data.y : data.y;
data.z = ((dst_index + 2 >= dst_start) && (dst_index + 2 < dst_end)) ? tmp_data.z : data.z;
data.w = ((dst_index + 3 >= dst_start) && (dst_index + 3 < dst_end)) ? tmp_data.w : data.w;
*((__global uchar4 *)(dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_C1_D2 (__global ushort *src1, int src1_step, int src1_offset,
__global ushort *dst, int dst_step, int dst_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 1;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset >> 1) & 1)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffffc);
ushort2 src1_data = vload2(0, (__global ushort *)((__global char *)src1 + src1_index));
int2 src2_data = (int2)(src2.x, src2.x);
ushort2 data = *((__global ushort2 *)((__global uchar *)dst + dst_index));
int2 tmp = ARITHM_OP(convert_int2_sat(src1_data), src2_data);
ushort2 tmp_data = convert_ushort2_sat(tmp);
data.x = (dst_index + 0 >= dst_start) ? tmp_data.x : data.x;
data.y = (dst_index + 2 < dst_end ) ? tmp_data.y : data.y;
*((__global ushort2 *)((__global uchar *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_C1_D3 (__global short *src1, int src1_step, int src1_offset,
__global short *dst, int dst_step, int dst_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 1;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset >> 1) & 1)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffffc);
short2 src1_data = vload2(0, (__global short *)((__global char *)src1 + src1_index));
int2 src2_data = (int2)(src2.x, src2.x);
short2 data = *((__global short2 *)((__global uchar *)dst + dst_index));
int2 tmp = ARITHM_OP(convert_int2_sat(src1_data), src2_data);
short2 tmp_data = convert_short2_sat(tmp);
data.x = (dst_index + 0 >= dst_start) ? tmp_data.x : data.x;
data.y = (dst_index + 2 < dst_end ) ? tmp_data.y : data.y;
*((__global short2 *)((__global uchar *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_C1_D4 (__global int *src1, int src1_step, int src1_offset,
__global int *dst, int dst_step, int dst_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
int src_data1 = *((__global int *)((__global char *)src1 + src1_index));
int src_data2 = src2.x;
int dst_data = *((__global int *)((__global char *)dst + dst_index));
int data = convert_int_sat(ARITHM_OP((long)src_data1, (long)src_data2));
*((__global int *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_C1_D5 (__global float *src1, int src1_step, int src1_offset,
__global float *dst, int dst_step, int dst_offset,
float4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
float src_data1 = *((__global float *)((__global char *)src1 + src1_index));
float src_data2 = src2.x;
float dst_data = *((__global float *)((__global char *)dst + dst_index));
float data = ARITHM_OP(src_data1, src_data2);
*((__global float *)((__global char *)dst + dst_index)) = data;
}
}
#if defined (DOUBLE_SUPPORT)
__kernel void arithm_s_add_C1_D6 (__global double *src1, int src1_step, int src1_offset,
__global double *dst, int dst_step, int dst_offset,
double4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
double src_data1 = *((__global double *)((__global char *)src1 + src1_index));
double src2_data = src2.x;
double dst_data = *((__global double *)((__global char *)dst + dst_index));
double data = ARITHM_OP(src_data1, src2_data);
*((__global double *)((__global char *)dst + dst_index)) = data;
}
}
#endif
__kernel void arithm_s_add_C2_D0 (__global uchar *src1, int src1_step, int src1_offset,
__global uchar *dst, int dst_step, int dst_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 1;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset >> 1) & 1)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffffc);
uchar4 src1_data = vload4(0, src1 + src1_index);
int4 src2_data = (int4)(src2.x, src2.y, src2.x, src2.y);
uchar4 data = *((__global uchar4 *)(dst + dst_index));
int4 tmp = ARITHM_OP(convert_int4_sat(src1_data), src2_data);
uchar4 tmp_data = convert_uchar4_sat(tmp);
data.xy = (dst_index + 0 >= dst_start) ? tmp_data.xy : data.xy;
data.zw = (dst_index + 2 < dst_end ) ? tmp_data.zw : data.zw;
*((__global uchar4 *)(dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_C2_D2 (__global ushort *src1, int src1_step, int src1_offset,
__global ushort *dst, int dst_step, int dst_offset,
int4 src2, int rows, int cols, int dst_step1)
{
///////////////////////////////////////////////////////////////////////////////////
///////////////////////////////// Add with scalar /////////////////////////////////
///////////////////////////////////////////////////////////////////////////////////
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
ushort2 src_data1 = *((__global ushort2 *)((__global char *)src1 + src1_index));
int2 src_data2 = (int2)(src2.x, src2.y);
ushort2 dst_data = *((__global ushort2 *)((__global char *)dst + dst_index));
int2 tmp = ARITHM_OP(convert_int2_sat(src_data1), src_data2);
ushort2 data = convert_ushort2_sat(tmp);
*((__global ushort2 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_C2_D3 (__global short *src1, int src1_step, int src1_offset,
__global short *dst, int dst_step, int dst_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
short2 src_data1 = *((__global short2 *)((__global char *)src1 + src1_index));
int2 src_data2 = (int2)(src2.x, src2.y);
short2 dst_data = *((__global short2 *)((__global char *)dst + dst_index));
int2 tmp = ARITHM_OP(convert_int2_sat(src_data1), src_data2);
short2 data = convert_short2_sat(tmp);
*((__global short2 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_C2_D4 (__global int *src1, int src1_step, int src1_offset,
__global int *dst, int dst_step, int dst_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
int2 src_data1 = *((__global int2 *)((__global char *)src1 + src1_index));
int2 src_data2 = (int2)(src2.x, src2.y);
int2 dst_data = *((__global int2 *)((__global char *)dst + dst_index));
int2 data = convert_int2_sat(ARITHM_OP(convert_long2_sat(src_data1), convert_long2_sat(src_data2)));
*((__global int2 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_C2_D5 (__global float *src1, int src1_step, int src1_offset,
__global float *dst, int dst_step, int dst_offset,
float4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
float2 src_data1 = *((__global float2 *)((__global char *)src1 + src1_index));
float2 src_data2 = (float2)(src2.x, src2.y);
float2 dst_data = *((__global float2 *)((__global char *)dst + dst_index));
float2 data = ARITHM_OP(src_data1, src_data2);
*((__global float2 *)((__global char *)dst + dst_index)) = data;
}
}
#if defined (DOUBLE_SUPPORT)
__kernel void arithm_s_add_C2_D6 (__global double *src1, int src1_step, int src1_offset,
__global double *dst, int dst_step, int dst_offset,
double4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 4) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 4) + dst_offset);
double2 src_data1 = *((__global double2 *)((__global char *)src1 + src1_index));
double2 src_data2 = (double2)(src2.x, src2.y);
double2 dst_data = *((__global double2 *)((__global char *)dst + dst_index));
double2 data = ARITHM_OP(src_data1, src_data2);
*((__global double2 *)((__global char *)dst + dst_index)) = data;
}
}
#endif
__kernel void arithm_s_add_C4_D0 (__global uchar *src1, int src1_step, int src1_offset,
__global uchar *dst, int dst_step, int dst_offset,
int4 src2, int rows, int cols, int dst_step1)
__kernel void arithm_binary_op_scalar (__global T *src1, int src1_step, int src1_offset,
__global WT *scalar,
__global T *dst, int dst_step, int dst_offset,
int cols, int rows)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
uchar4 src_data1 = *((__global uchar4 *)(src1 + src1_index));
int src1_index = mad24(y, src1_step, x + src1_offset);
int dst_index = mad24(y, dst_step, x + dst_offset);
uchar4 data = convert_uchar4_sat(ARITHM_OP(convert_int4_sat(src_data1), src2));
*((__global uchar4 *)(dst + dst_index)) = data;
dst[dst_index] = convertToT(convertToWT(src1[src1_index]) Operation scalar[0]);
}
}
__kernel void arithm_s_add_C4_D2 (__global ushort *src1, int src1_step, int src1_offset,
__global ushort *dst, int dst_step, int dst_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
ushort4 src_data1 = *((__global ushort4 *)((__global char *)src1 + src1_index));
ushort4 data = convert_ushort4_sat(ARITHM_OP(convert_int4_sat(src_data1), src2));
*((__global ushort4 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_C4_D3 (__global short *src1, int src1_step, int src1_offset,
__global short *dst, int dst_step, int dst_offset,
int4 src2, int rows, int cols, int dst_step1)
__kernel void arithm_absdiff_scalar(__global T *src1, int src1_step, int src1_offset,
__global WT *src2,
__global T *dst, int dst_step, int dst_offset,
int cols, int rows)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
short4 src_data1 = *((__global short4 *)((__global char *)src1 + src1_index));
short4 data = convert_short4_sat(ARITHM_OP(convert_int4_sat(src_data1), src2));
int src1_index = mad24(y, src1_step, x + src1_offset);
int dst_index = mad24(y, dst_step, x + dst_offset);
*((__global short4 *)((__global char *)dst + dst_index)) = data;
WT value = convertToWT(src1[src1_index]) - src2[0];
value = value > (WT)(0) ? value : -value;
dst[dst_index] = convertToT(value);
}
}
__kernel void arithm_s_add_C4_D4 (__global int *src1, int src1_step, int src1_offset,
__global int *dst, int dst_step, int dst_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 4) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 4) + dst_offset);
int4 src_data1 = *((__global int4 *)((__global char *)src1 + src1_index));
int4 data = convert_int4_sat(ARITHM_OP(convert_long4_sat(src_data1), convert_long4_sat(src2)));
*((__global int4 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_C4_D5 (__global float *src1, int src1_step, int src1_offset,
__global float *dst, int dst_step, int dst_offset,
float4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 4) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 4) + dst_offset);
float4 src_data1 = *((__global float4 *)((__global char *)src1 + src1_index));
float4 data = ARITHM_OP(src_data1, src2);
*((__global float4 *)((__global char *)dst + dst_index)) = data;
}
}
#if defined (DOUBLE_SUPPORT)
__kernel void arithm_s_add_C4_D6 (__global double *src1, int src1_step, int src1_offset,
__global double *dst, int dst_step, int dst_offset,
double4 src2, int rows, int cols, int dst_step1)
// scalar divide to matrix
__kernel void arithm_binary_op_scalar_div(__global T *src1, int src1_step, int src1_offset,
__global WT *scalar,
__global T *dst, int dst_step, int dst_offset,
int cols, int rows)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 5) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 5) + dst_offset);
int src1_index = mad24(y, src1_step, x + src1_offset);
int dst_index = mad24(y, dst_step, x + dst_offset);
double4 src_data1 = *((__global double4 *)((__global char *)src1 + src1_index));
double4 data = ARITHM_OP(src_data1, src2);
*((__global double4 *)((__global char *)dst + dst_index)) = data;
T zero = (T)(0);
dst[dst_index] = src1[src1_index] == zero ? zero : convertToT(scalar[0] / convertToWT(src1[src1_index]));
}
}
#endif
modules/ocl/src/opencl/arithm_add_scalar_mask.cl
View file @ 0ad03162
......@@ -51,561 +51,28 @@
#endif
#endif
#ifdef ARITHM_ADD
#define ARITHM_OP(A,B) ((A)+(B))
#elif defined ARITHM_SUB
#define ARITHM_OP(A,B) ((A)-(B))
#endif
/**************************************add with scalar with mask**************************************/
__kernel void arithm_s_add_with_mask_C1_D0 (__global uchar *src1, int src1_step, int src1_offset,
__global uchar *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align (dst_offset & 3)
int src1_index = mad24(y, src1_step, x + src1_offset - dst_align);
int mask_index = mad24(y, mask_step, x + mask_offset - dst_align);
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + x & (int)0xfffffffc);
int src1_index_fix = src1_index < 0 ? 0 : src1_index;
int mask_index_fix = mask_index < 0 ? 0 : mask_index;
uchar4 src1_data = vload4(0, src1 + src1_index_fix);
int4 src2_data = (int4)(src2.x, src2.x, src2.x, src2.x);
uchar4 mask_data = vload4(0, mask + mask_index_fix);
if(src1_index < 0)
{
uchar4 tmp;
tmp.xyzw = (src1_index == -2) ? src1_data.zwxy:src1_data.yzwx;
src1_data.xyzw = (src1_index == -1) ? src1_data.wxyz:tmp.xyzw;
}
if(mask_index < 0)
{
uchar4 tmp;
tmp.xyzw = (mask_index == -2) ? mask_data.zwxy:mask_data.yzwx;
mask_data.xyzw = (mask_index == -1) ? mask_data.wxyz:tmp.xyzw;
}
uchar4 data = *((__global uchar4 *)(dst + dst_index));
int4 tmp = ARITHM_OP(convert_int4_sat(src1_data), src2_data);
uchar4 tmp_data = convert_uchar4_sat(tmp);
data.x = ((mask_data.x) && (dst_index + 0 >= dst_start) && (dst_index + 0 < dst_end)) ? tmp_data.x : data.x;
data.y = ((mask_data.y) && (dst_index + 1 >= dst_start) && (dst_index + 1 < dst_end)) ? tmp_data.y : data.y;
data.z = ((mask_data.z) && (dst_index + 2 >= dst_start) && (dst_index + 2 < dst_end)) ? tmp_data.z : data.z;
data.w = ((mask_data.w) && (dst_index + 3 >= dst_start) && (dst_index + 3 < dst_end)) ? tmp_data.w : data.w;
*((__global uchar4 *)(dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_with_mask_C1_D2 (__global ushort *src1, int src1_step, int src1_offset,
__global ushort *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 1;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset / 2) & 1)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int mask_index = mad24(y, mask_step, x + mask_offset - dst_align);
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffffc);
ushort2 src1_data = vload2(0, (__global ushort *)((__global char *)src1 + src1_index));
int2 src2_data = (int2)(src2.x, src2.x);
uchar2 mask_data = vload2(0, mask + mask_index);
ushort2 data = *((__global ushort2 *)((__global uchar *)dst + dst_index));
int2 tmp = ARITHM_OP(convert_int2_sat(src1_data), src2_data);
ushort2 tmp_data = convert_ushort2_sat(tmp);
data.x = ((mask_data.x) && (dst_index + 0 >= dst_start)) ? tmp_data.x : data.x;
data.y = ((mask_data.y) && (dst_index + 2 < dst_end )) ? tmp_data.y : data.y;
*((__global ushort2 *)((__global uchar *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_with_mask_C1_D3 (__global short *src1, int src1_step, int src1_offset,
__global short *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 1;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset / 2) & 1)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int mask_index = mad24(y, mask_step, x + mask_offset - dst_align);
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffffc);
short2 src1_data = vload2(0, (__global short *)((__global char *)src1 + src1_index));
int2 src2_data = (int2)(src2.x, src2.x);
uchar2 mask_data = vload2(0, mask + mask_index);
short2 data = *((__global short2 *)((__global uchar *)dst + dst_index));
int2 tmp = ARITHM_OP(convert_int2_sat(src1_data), src2_data);
short2 tmp_data = convert_short2_sat(tmp);
data.x = ((mask_data.x) && (dst_index + 0 >= dst_start)) ? tmp_data.x : data.x;
data.y = ((mask_data.y) && (dst_index + 2 < dst_end )) ? tmp_data.y : data.y;
*((__global short2 *)((__global uchar *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_with_mask_C1_D4 (__global int *src1, int src1_step, int src1_offset,
__global int *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
uchar mask_data = *(mask + mask_index);
int src_data1 = *((__global int *)((__global char *)src1 + src1_index));
int src_data2 = src2.x;
int dst_data = *((__global int *)((__global char *)dst + dst_index));
int data = convert_int_sat(ARITHM_OP((long)src_data1, (long)src_data2));
data = mask_data ? data : dst_data;
*((__global int *)((__global char *)dst + dst_index)) = data;
}
}
///////////////////////////////////////////////////////////////////////////////////
//////////////////////////// Add with scalar with mask ////////////////////////////
///////////////////////////////////////////////////////////////////////////////////
__kernel void arithm_s_add_with_mask_C1_D5 (__global float *src1, int src1_step, int src1_offset,
__global float *dst, int dst_step, int dst_offset,
__kernel void arithm_binary_op_scalar_mask(__global T *src1, int src1_step, int src1_offset,
__global WT *scalar,
__global uchar *mask, int mask_step, int mask_offset,
float4 src2, int rows, int cols, int dst_step1)
__global T *dst, int dst_step, int dst_offset,
int cols, int rows)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
uchar mask_data = *(mask + mask_index);
float src_data1 = *((__global float *)((__global char *)src1 + src1_index));
float src_data2 = src2.x;
float dst_data = *((__global float *)((__global char *)dst + dst_index));
float data = ARITHM_OP(src_data1, src_data2);
data = mask_data ? data : dst_data;
*((__global float *)((__global char *)dst + dst_index)) = data;
}
}
#if defined (DOUBLE_SUPPORT)
__kernel void arithm_s_add_with_mask_C1_D6 (__global double *src1, int src1_step, int src1_offset,
__global double *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
double4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
uchar mask_data = *(mask + mask_index);
double src_data1 = *((__global double *)((__global char *)src1 + src1_index));
double src_data2 = src2.x;
double dst_data = *((__global double *)((__global char *)dst + dst_index));
double data = ARITHM_OP(src_data1, src_data2);
data = mask_data ? data : dst_data;
*((__global double *)((__global char *)dst + dst_index)) = data;
}
}
#endif
__kernel void arithm_s_add_with_mask_C2_D0 (__global uchar *src1, int src1_step, int src1_offset,
__global uchar *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 1;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset / 2) & 1)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int mask_index = mad24(y, mask_step, x + mask_offset - dst_align);
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffffc);
uchar4 src1_data = vload4(0, src1 + src1_index);
int4 src2_data = (int4)(src2.x, src2.y, src2.x, src2.y);
uchar2 mask_data = vload2(0, mask + mask_index);
uchar4 data = *((__global uchar4 *)(dst + dst_index));
int4 tmp = ARITHM_OP(convert_int4_sat(src1_data), src2_data);
uchar4 tmp_data = convert_uchar4_sat(tmp);
data.xy = ((mask_data.x) && (dst_index + 0 >= dst_start)) ? tmp_data.xy : data.xy;
data.zw = ((mask_data.y) && (dst_index + 2 < dst_end )) ? tmp_data.zw : data.zw;
*((__global uchar4 *)(dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_with_mask_C2_D2 (__global ushort *src1, int src1_step, int src1_offset,
__global ushort *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
if (mask[mask_index])
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
uchar mask_data = *(mask + mask_index);
ushort2 src_data1 = *((__global ushort2 *)((__global char *)src1 + src1_index));
int2 src_data2 = (int2)(src2.x, src2.y);
ushort2 dst_data = *((__global ushort2 *)((__global char *)dst + dst_index));
int src1_index = mad24(y, src1_step, x + src1_offset);
int dst_index = mad24(y, dst_step, dst_offset + x);
int2 tmp = ARITHM_OP(convert_int2_sat(src_data1), src_data2);
ushort2 data = convert_ushort2_sat(tmp);
data = mask_data ? data : dst_data;
*((__global ushort2 *)((__global char *)dst + dst_index)) = data;
dst[dst_index] = convertToT(convertToWT(src1[src1_index]) Operation scalar[0]);
}
}
__kernel void arithm_s_add_with_mask_C2_D3 (__global short *src1, int src1_step, int src1_offset,
__global short *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
uchar mask_data = *(mask + mask_index);
short2 src_data1 = *((__global short2 *)((__global char *)src1 + src1_index));
int2 src_data2 = (int2)(src2.x, src2.y);
short2 dst_data = *((__global short2 *)((__global char *)dst + dst_index));
int2 tmp = ARITHM_OP(convert_int2_sat(src_data1), src_data2);
short2 data = convert_short2_sat(tmp);
data = mask_data ? data : dst_data;
*((__global short2 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_with_mask_C2_D4 (__global int *src1, int src1_step, int src1_offset,
__global int *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
uchar mask_data = *(mask + mask_index);
int2 src_data1 = *((__global int2 *)((__global char *)src1 + src1_index));
int2 src_data2 = (int2)(src2.x, src2.y);
int2 dst_data = *((__global int2 *)((__global char *)dst + dst_index));
int2 data = convert_int2_sat(ARITHM_OP(convert_long2_sat(src_data1), convert_long2_sat(src_data2)));
data = mask_data ? data : dst_data;
*((__global int2 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_with_mask_C2_D5 (__global float *src1, int src1_step, int src1_offset,
__global float *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
float4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
uchar mask_data = *(mask + mask_index);
float2 src_data1 = *((__global float2 *)((__global char *)src1 + src1_index));
float2 src_data2 = (float2)(src2.x, src2.y);
float2 dst_data = *((__global float2 *)((__global char *)dst + dst_index));
float2 data = ARITHM_OP(src_data1, src_data2);
data = mask_data ? data : dst_data;
*((__global float2 *)((__global char *)dst + dst_index)) = data;
}
}
#if defined (DOUBLE_SUPPORT)
__kernel void arithm_s_add_with_mask_C2_D6 (__global double *src1, int src1_step, int src1_offset,
__global double *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
double4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 4) + src1_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 4) + dst_offset);
uchar mask_data = *(mask + mask_index);
double2 src_data1 = *((__global double2 *)((__global char *)src1 + src1_index));
double2 src_data2 = (double2)(src2.x, src2.y);
double2 dst_data = *((__global double2 *)((__global char *)dst + dst_index));
double2 data = ARITHM_OP(src_data1, src_data2);
data = mask_data ? data : dst_data;
*((__global double2 *)((__global char *)dst + dst_index)) = data;
}
}
#endif
__kernel void arithm_s_add_with_mask_C4_D0 (__global uchar *src1, int src1_step, int src1_offset,
__global uchar *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
uchar mask_data = *(mask + mask_index);
uchar4 src_data1 = *((__global uchar4 *)(src1 + src1_index));
uchar4 dst_data = *((__global uchar4 *)(dst + dst_index));
uchar4 data = convert_uchar4_sat(ARITHM_OP(convert_int4_sat(src_data1), src2));
data = mask_data ? data : dst_data;
*((__global uchar4 *)(dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_with_mask_C4_D2 (__global ushort *src1, int src1_step, int src1_offset,
__global ushort *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
uchar mask_data = *(mask + mask_index);
ushort4 src_data1 = *((__global ushort4 *)((__global char *)src1 + src1_index));
ushort4 dst_data = *((__global ushort4 *)((__global char *)dst + dst_index));
ushort4 data = convert_ushort4_sat(ARITHM_OP(convert_int4_sat(src_data1), src2));
data = mask_data ? data : dst_data;
*((__global ushort4 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_with_mask_C4_D3 (__global short *src1, int src1_step, int src1_offset,
__global short *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
uchar mask_data = *(mask + mask_index);
short4 src_data1 = *((__global short4 *)((__global char *)src1 + src1_index));
short4 dst_data = *((__global short4 *)((__global char *)dst + dst_index));
short4 data = convert_short4_sat(ARITHM_OP(convert_int4_sat(src_data1), src2));
data = mask_data ? data : dst_data;
*((__global short4 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_with_mask_C4_D4 (__global int *src1, int src1_step, int src1_offset,
__global int *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
int4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 4) + src1_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 4) + dst_offset);
uchar mask_data = *(mask + mask_index);
int4 src_data1 = *((__global int4 *)((__global char *)src1 + src1_index));
int4 dst_data = *((__global int4 *)((__global char *)dst + dst_index));
int4 data = convert_int4_sat(ARITHM_OP(convert_long4_sat(src_data1), convert_long4_sat(src2)));
data = mask_data ? data : dst_data;
*((__global int4 *)((__global char *)dst + dst_index)) = data;
}
}
__kernel void arithm_s_add_with_mask_C4_D5 (__global float *src1, int src1_step, int src1_offset,
__global float *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
float4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 4) + src1_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 4) + dst_offset);
uchar mask_data = *(mask + mask_index);
float4 src_data1 = *((__global float4 *)((__global char *)src1 + src1_index));
float4 dst_data = *((__global float4 *)((__global char *)dst + dst_index));
float4 data = ARITHM_OP(src_data1, src2);
data = mask_data ? data : dst_data;
*((__global float4 *)((__global char *)dst + dst_index)) = data;
}
}
#if defined (DOUBLE_SUPPORT)
__kernel void arithm_s_add_with_mask_C4_D6 (__global double *src1, int src1_step, int src1_offset,
__global double *dst, int dst_step, int dst_offset,
__global uchar *mask, int mask_step, int mask_offset,
double4 src2, int rows, int cols, int dst_step1)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 5) + src1_offset);
int mask_index = mad24(y, mask_step, x + mask_offset);
int dst_index = mad24(y, dst_step, (x << 5) + dst_offset);
uchar mask_data = *(mask + mask_index);
double4 src_data1 = *((__global double4 *)((__global char *)src1 + src1_index));
double4 dst_data = *((__global double4 *)((__global char *)dst + dst_index));
double4 data = ARITHM_OP(src_data1, src2);
data = mask_data ? data : dst_data;
*((__global double4 *)((__global char *)dst + dst_index)) = data;
}
}
#endif
modules/ocl/src/opencl/arithm_div.cl deleted 100644 → 0
View file @ 5ff5fdd7
/*M///////////////////////////////////////////////////////////////////////////////////////
//
// IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
//
// By downloading, copying, installing or using the software you agree to this license.
// If you do not agree to this license, do not download, install,
// copy or use the software.
//
//
// License Agreement
// For Open Source Computer Vision Library
//
// Copyright (C) 2010-2012, Institute Of Software Chinese Academy Of Science, all rights reserved.
// Copyright (C) 2010-2012, Advanced Micro Devices, Inc., all rights reserved.
// Third party copyrights are property of their respective owners.
//
// @Authors
// Jia Haipeng, jiahaipeng95@gmail.com
//
// Redistribution and use in source and binary forms, with or without modification,
// are permitted provided that the following conditions are met:
//
// * Redistribution's of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
//
// * Redistribution's in binary form must reproduce the above copyright notice,
// this list of conditions and the following disclaimer in the documentation
// and/or other oclMaterials provided with the distribution.
//
// * The name of the copyright holders may not be used to endorse or promote products
// derived from this software without specific prior written permission.
//
// This software is provided by the copyright holders and contributors as is and
// any express or implied warranties, including, but not limited to, the implied
// warranties of merchantability and fitness for a particular purpose are disclaimed.
// In no event shall the Intel Corporation or contributors be liable for any direct,
// indirect, incidental, special, exemplary, or consequential damages
// (including, but not limited to, procurement of substitute goods or services;
// loss of use, data, or profits; or business interruption) however caused
// and on any theory of liability, whether in contract, strict liability,
// or tort (including negligence or otherwise) arising in any way out of
// the use of this software, even if advised of the possibility of such damage.
//
//M*/
#if defined (DOUBLE_SUPPORT)
#ifdef cl_khr_fp64
#pragma OPENCL EXTENSION cl_khr_fp64:enable
#elif defined (cl_amd_fp64)
#pragma OPENCL EXTENSION cl_amd_fp64:enable
#endif
typedef double F ;
typedef double4 F4;
#define convert_F4 convert_double4
#define convert_F double
#else
typedef float F;
typedef float4 F4;
#define convert_F4 convert_float4
#define convert_F float
#endif
inline uchar round2_uchar(F v)
{
return convert_uchar_sat(round(v));
}
inline ushort round2_ushort(F v)
{
return convert_ushort_sat(round(v));
}
inline short round2_short(F v)
{
return convert_short_sat(round(v));
}
inline int round2_int(F v)
{
return convert_int_sat(round(v));
}
///////////////////////////////////////////////////////////////////////////////////////
////////////////////////////divide///////////////////////////////////////////////////
//////////////////////////////////////////////////////////////////////////////////////
/**********************************div*********************************************/
__kernel void arithm_div_D0 (__global uchar *src1, int src1_step, int src1_offset,
__global uchar *src2, int src2_step, int src2_offset,
__global uchar *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, F scalar)
{
int2 coor = (int2)(get_global_id(0), get_global_id(1));
if (coor.x < cols && coor.y < rows)
{
coor.x = coor.x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align (dst_offset & 3)
int2 src_index = (int2)(mad24(coor.y, src1_step, coor.x + src1_offset - dst_align),
mad24(coor.y, src2_step, coor.x + src2_offset - dst_align));
int4 dst_args = (int4)(mad24(coor.y, dst_step, dst_offset),
mad24(coor.y, dst_step, dst_offset + dst_step1),
mad24(coor.y, dst_step, dst_offset + coor.x & (int)0xfffffffc),
0);
uchar4 src1_data = vload4(0, src1 + src_index.x);
uchar4 src2_data = vload4(0, src2 + src_index.y);
uchar4 dst_data = *((__global uchar4 *)(dst + dst_args.z));
F4 tmp = convert_F4(src1_data) * scalar;
uchar4 tmp_data;
tmp_data.x = ((tmp.x == 0) || (src2_data.x == 0)) ? 0 : round2_uchar(tmp.x / src2_data.x);
tmp_data.y = ((tmp.y == 0) || (src2_data.y == 0)) ? 0 : round2_uchar(tmp.y / src2_data.y);
tmp_data.z = ((tmp.z == 0) || (src2_data.z == 0)) ? 0 : round2_uchar(tmp.z / src2_data.z);
tmp_data.w = ((tmp.w == 0) || (src2_data.w == 0)) ? 0 : round2_uchar(tmp.w / src2_data.w);
dst_data.x = ((dst_args.z + 0 >= dst_args.x) && (dst_args.z + 0 < dst_args.y)) ? tmp_data.x : dst_data.x;
dst_data.y = ((dst_args.z + 1 >= dst_args.x) && (dst_args.z + 1 < dst_args.y)) ? tmp_data.y : dst_data.y;
dst_data.z = ((dst_args.z + 2 >= dst_args.x) && (dst_args.z + 2 < dst_args.y)) ? tmp_data.z : dst_data.z;
dst_data.w = ((dst_args.z + 3 >= dst_args.x) && (dst_args.z + 3 < dst_args.y)) ? tmp_data.w : dst_data.w;
*((__global uchar4 *)(dst + dst_args.z)) = dst_data;
}
}
__kernel void arithm_div_D2 (__global ushort *src1, int src1_step, int src1_offset,
__global ushort *src2, int src2_step, int src2_offset,
__global ushort *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, F scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset >> 1) & 3)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int src2_index = mad24(y, src2_step, (x << 1) + src2_offset - (dst_align << 1));
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffff8);
ushort4 src1_data = vload4(0, (__global ushort *)((__global char *)src1 + src1_index));
ushort4 src2_data = vload4(0, (__global ushort *)((__global char *)src2 + src2_index));
ushort4 dst_data = *((__global ushort4 *)((__global char *)dst + dst_index));
F4 tmp = convert_F4(src1_data) * scalar;
ushort4 tmp_data;
tmp_data.x = ((tmp.x == 0) || (src2_data.x == 0)) ? 0 : round2_ushort(tmp.x / (F)src2_data.x);
tmp_data.y = ((tmp.y == 0) || (src2_data.y == 0)) ? 0 : round2_ushort(tmp.y / (F)src2_data.y);
tmp_data.z = ((tmp.z == 0) || (src2_data.z == 0)) ? 0 : round2_ushort(tmp.z / (F)src2_data.z);
tmp_data.w = ((tmp.w == 0) || (src2_data.w == 0)) ? 0 : round2_ushort(tmp.w / (F)src2_data.w);
dst_data.x = ((dst_index + 0 >= dst_start) && (dst_index + 0 < dst_end)) ? tmp_data.x : dst_data.x;
dst_data.y = ((dst_index + 2 >= dst_start) && (dst_index + 2 < dst_end)) ? tmp_data.y : dst_data.y;
dst_data.z = ((dst_index + 4 >= dst_start) && (dst_index + 4 < dst_end)) ? tmp_data.z : dst_data.z;
dst_data.w = ((dst_index + 6 >= dst_start) && (dst_index + 6 < dst_end)) ? tmp_data.w : dst_data.w;
*((__global ushort4 *)((__global char *)dst + dst_index)) = dst_data;
}
}
__kernel void arithm_div_D3 (__global short *src1, int src1_step, int src1_offset,
__global short *src2, int src2_step, int src2_offset,
__global short *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, F scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset >> 1) & 3)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int src2_index = mad24(y, src2_step, (x << 1) + src2_offset - (dst_align << 1));
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffff8);
short4 src1_data = vload4(0, (__global short *)((__global char *)src1 + src1_index));
short4 src2_data = vload4(0, (__global short *)((__global char *)src2 + src2_index));
short4 dst_data = *((__global short4 *)((__global char *)dst + dst_index));
F4 tmp = convert_F4(src1_data) * scalar;
short4 tmp_data;
tmp_data.x = ((tmp.x == 0) || (src2_data.x == 0)) ? 0 : round2_short(tmp.x / (F)src2_data.x);
tmp_data.y = ((tmp.y == 0) || (src2_data.y == 0)) ? 0 : round2_short(tmp.y / (F)src2_data.y);
tmp_data.z = ((tmp.z == 0) || (src2_data.z == 0)) ? 0 : round2_short(tmp.z / (F)src2_data.z);
tmp_data.w = ((tmp.w == 0) || (src2_data.w == 0)) ? 0 : round2_short(tmp.w / (F)src2_data.w);
dst_data.x = ((dst_index + 0 >= dst_start) && (dst_index + 0 < dst_end)) ? tmp_data.x : dst_data.x;
dst_data.y = ((dst_index + 2 >= dst_start) && (dst_index + 2 < dst_end)) ? tmp_data.y : dst_data.y;
dst_data.z = ((dst_index + 4 >= dst_start) && (dst_index + 4 < dst_end)) ? tmp_data.z : dst_data.z;
dst_data.w = ((dst_index + 6 >= dst_start) && (dst_index + 6 < dst_end)) ? tmp_data.w : dst_data.w;
*((__global short4 *)((__global char *)dst + dst_index)) = dst_data;
}
}
__kernel void arithm_div_D4 (__global int *src1, int src1_step, int src1_offset,
__global int *src2, int src2_step, int src2_offset,
__global int *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, F scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 2) + src2_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
int data1 = *((__global int *)((__global char *)src1 + src1_index));
int data2 = *((__global int *)((__global char *)src2 + src2_index));
F tmp = (convert_F)(data1) * scalar;
int tmp_data = (tmp == 0 || data2 == 0) ? 0 : round2_int(tmp / (convert_F)(data2));
*((__global int *)((__global char *)dst + dst_index)) =tmp_data;
}
}
__kernel void arithm_div_D5 (__global float *src1, int src1_step, int src1_offset,
__global float *src2, int src2_step, int src2_offset,
__global float *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, F scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 2) + src2_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
float data1 = *((__global float *)((__global char *)src1 + src1_index));
float data2 = *((__global float *)((__global char *)src2 + src2_index));
F tmp = (convert_F)(data1) * scalar;
float tmp_data = (tmp == 0 || data2 == 0) ? 0 : convert_float(tmp / (convert_F)(data2));
*((__global float *)((__global char *)dst + dst_index)) = tmp_data;
}
}
#if defined (DOUBLE_SUPPORT)
__kernel void arithm_div_D6 (__global double *src1, int src1_step, int src1_offset,
__global double *src2, int src2_step, int src2_offset,
__global double *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, double scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 3) + src2_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
double data1 = *((__global double *)((__global char *)src1 + src1_index));
double data2 = *((__global double *)((__global char *)src2 + src2_index));
double tmp = data1 * scalar;
double tmp_data = (tmp == 0 || data2 == 0) ? 0 : (tmp / data2);
*((__global double *)((__global char *)dst + dst_index)) = tmp_data;
}
}
#endif
/************************************div with scalar************************************/
__kernel void arithm_s_div_D0 (__global uchar *src, int src_step, int src_offset,
__global uchar *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, F scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align (dst_offset & 3)
int src_index = mad24(y, src_step, x + src_offset - dst_align);
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + x & (int)0xfffffffc);
uchar4 src_data = vload4(0, src + src_index);
uchar4 dst_data = *((__global uchar4 *)(dst + dst_index));
uchar4 tmp_data;
tmp_data.x = ((scalar == 0) || (src_data.x == 0)) ? 0 : round2_uchar(scalar / (F)src_data.x);
tmp_data.y = ((scalar == 0) || (src_data.y == 0)) ? 0 : round2_uchar(scalar / (F)src_data.y);
tmp_data.z = ((scalar == 0) || (src_data.z == 0)) ? 0 : round2_uchar(scalar / (F)src_data.z);
tmp_data.w = ((scalar == 0) || (src_data.w == 0)) ? 0 : round2_uchar(scalar / (F)src_data.w);
dst_data.x = ((dst_index + 0 >= dst_start) && (dst_index + 0 < dst_end)) ? tmp_data.x : dst_data.x;
dst_data.y = ((dst_index + 1 >= dst_start) && (dst_index + 1 < dst_end)) ? tmp_data.y : dst_data.y;
dst_data.z = ((dst_index + 2 >= dst_start) && (dst_index + 2 < dst_end)) ? tmp_data.z : dst_data.z;
dst_data.w = ((dst_index + 3 >= dst_start) && (dst_index + 3 < dst_end)) ? tmp_data.w : dst_data.w;
*((__global uchar4 *)(dst + dst_index)) = dst_data;
}
}
__kernel void arithm_s_div_D2 (__global ushort *src, int src_step, int src_offset,
__global ushort *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, F scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset >> 1) & 3)
int src_index = mad24(y, src_step, (x << 1) + src_offset - (dst_align << 1));
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffff8);
ushort4 src_data = vload4(0, (__global ushort *)((__global char *)src + src_index));
ushort4 dst_data = *((__global ushort4 *)((__global char *)dst + dst_index));
ushort4 tmp_data;
tmp_data.x = ((scalar == 0) || (src_data.x == 0)) ? 0 : round2_ushort(scalar / (F)src_data.x);
tmp_data.y = ((scalar == 0) || (src_data.y == 0)) ? 0 : round2_ushort(scalar / (F)src_data.y);
tmp_data.z = ((scalar == 0) || (src_data.z == 0)) ? 0 : round2_ushort(scalar / (F)src_data.z);
tmp_data.w = ((scalar == 0) || (src_data.w == 0)) ? 0 : round2_ushort(scalar / (F)src_data.w);
dst_data.x = ((dst_index + 0 >= dst_start) && (dst_index + 0 < dst_end)) ? tmp_data.x : dst_data.x;
dst_data.y = ((dst_index + 2 >= dst_start) && (dst_index + 2 < dst_end)) ? tmp_data.y : dst_data.y;
dst_data.z = ((dst_index + 4 >= dst_start) && (dst_index + 4 < dst_end)) ? tmp_data.z : dst_data.z;
dst_data.w = ((dst_index + 6 >= dst_start) && (dst_index + 6 < dst_end)) ? tmp_data.w : dst_data.w;
*((__global ushort4 *)((__global char *)dst + dst_index)) = dst_data;
}
}
__kernel void arithm_s_div_D3 (__global short *src, int src_step, int src_offset,
__global short *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, F scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset >> 1) & 3)
int src_index = mad24(y, src_step, (x << 1) + src_offset - (dst_align << 1));
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffff8);
short4 src_data = vload4(0, (__global short *)((__global char *)src + src_index));
short4 dst_data = *((__global short4 *)((__global char *)dst + dst_index));
short4 tmp_data;
tmp_data.x = ((scalar == 0) || (src_data.x == 0)) ? 0 : round2_short(scalar / (F)src_data.x);
tmp_data.y = ((scalar == 0) || (src_data.y == 0)) ? 0 : round2_short(scalar / (F)src_data.y);
tmp_data.z = ((scalar == 0) || (src_data.z == 0)) ? 0 : round2_short(scalar / (F)src_data.z);
tmp_data.w = ((scalar == 0) || (src_data.w == 0)) ? 0 : round2_short(scalar / (F)src_data.w);
dst_data.x = ((dst_index + 0 >= dst_start) && (dst_index + 0 < dst_end)) ? tmp_data.x : dst_data.x;
dst_data.y = ((dst_index + 2 >= dst_start) && (dst_index + 2 < dst_end)) ? tmp_data.y : dst_data.y;
dst_data.z = ((dst_index + 4 >= dst_start) && (dst_index + 4 < dst_end)) ? tmp_data.z : dst_data.z;
dst_data.w = ((dst_index + 6 >= dst_start) && (dst_index + 6 < dst_end)) ? tmp_data.w : dst_data.w;
*((__global short4 *)((__global char *)dst + dst_index)) = dst_data;
}
}
__kernel void arithm_s_div_D4 (__global int *src, int src_step, int src_offset,
__global int *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, F scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src_index = mad24(y, src_step, (x << 2) + src_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
int data = *((__global int *)((__global char *)src + src_index));
int tmp_data = (scalar == 0 || data == 0) ? 0 : round2_int(scalar / (convert_F)(data));
*((__global int *)((__global char *)dst + dst_index)) =tmp_data;
}
}
__kernel void arithm_s_div_D5 (__global float *src, int src_step, int src_offset,
__global float *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, F scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src_index = mad24(y, src_step, (x << 2) + src_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
float data = *((__global float *)((__global char *)src + src_index));
float tmp_data = (scalar == 0 || data == 0) ? 0 : convert_float(scalar / (convert_F)(data));
*((__global float *)((__global char *)dst + dst_index)) = tmp_data;
}
}
#if defined (DOUBLE_SUPPORT)
__kernel void arithm_s_div_D6 (__global double *src, int src_step, int src_offset,
__global double *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, double scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src_index = mad24(y, src_step, (x << 3) + src_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
double data = *((__global double *)((__global char *)src + src_index));
double tmp_data = (scalar == 0 || data == 0) ? 0 : (scalar / data);
*((__global double *)((__global char *)dst + dst_index)) = tmp_data;
}
}
#endif
modules/ocl/src/opencl/arithm_mul.cl deleted 100644 → 0
View file @ 5ff5fdd7
/*M///////////////////////////////////////////////////////////////////////////////////////
//
// IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
//
// By downloading, copying, installing or using the software you agree to this license.
// If you do not agree to this license, do not download, install,
// copy or use the software.
//
//
// License Agreement
// For Open Source Computer Vision Library
//
// Copyright (C) 2010-2012, Institute Of Software Chinese Academy Of Science, all rights reserved.
// Copyright (C) 2010-2012, Advanced Micro Devices, Inc., all rights reserved.
// Third party copyrights are property of their respective owners.
//
// @Authors
// Jia Haipeng, jiahaipeng95@gmail.com
//
// Redistribution and use in source and binary forms, with or without modification,
// are permitted provided that the following conditions are met:
//
// * Redistribution's of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
//
// * Redistribution's in binary form must reproduce the above copyright notice,
// this list of conditions and the following disclaimer in the documentation
// and/or other GpuMaterials provided with the distribution.
//
// * The name of the copyright holders may not be used to endorse or promote products
// derived from this software without specific prior written permission.
//
// This software is provided by the copyright holders and contributors as is and
// any express or implied warranties, including, but not limited to, the implied
// warranties of merchantability and fitness for a particular purpose are disclaimed.
// In no event shall the Intel Corporation or contributors be liable for any direct,
// indirect, incidental, special, exemplary, or consequential damages
// (including, but not limited to, procurement of substitute goods or services;
// loss of use, data, or profits; or business interruption) however caused
// and on any theory of liability, whether in contract, strict liability,
// or tort (including negligence or otherwise) arising in any way out of
// the use of this software, even if advised of the possibility of such damage.
//
//M*/
#if defined (DOUBLE_SUPPORT)
#ifdef cl_khr_fp64
#pragma OPENCL EXTENSION cl_khr_fp64:enable
#elif defined (cl_amd_fp64)
#pragma OPENCL EXTENSION cl_amd_fp64:enable
#endif
#endif
int4 round_int4(float4 v)
{
v.s0 = v.s0 + (v.s0 > 0 ? 0.5 : -0.5);
v.s1 = v.s1 + (v.s1 > 0 ? 0.5 : -0.5);
v.s2 = v.s2 + (v.s2 > 0 ? 0.5 : -0.5);
v.s3 = v.s3 + (v.s3 > 0 ? 0.5 : -0.5);
return convert_int4_sat(v);
}
uint4 round_uint4(float4 v)
{
v.s0 = v.s0 + (v.s0 > 0 ? 0.5 : -0.5);
v.s1 = v.s1 + (v.s1 > 0 ? 0.5 : -0.5);
v.s2 = v.s2 + (v.s2 > 0 ? 0.5 : -0.5);
v.s3 = v.s3 + (v.s3 > 0 ? 0.5 : -0.5);
return convert_uint4_sat(v);
}
long round_int(float v)
{
v = v + (v > 0 ? 0.5 : -0.5);
return convert_int_sat(v);
}
//////////////////////////////////////////////////////////////////////////////////////////////////////
/////////////////////////////////////////////multiply//////////////////////////////////////////////////
///////////////////////////////////////////////////////////////////////////////////////////////////////
/**************************************add without mask**************************************/
__kernel void arithm_mul_D0 (__global uchar *src1, int src1_step, int src1_offset,
__global uchar *src2, int src2_step, int src2_offset,
__global uchar *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, float scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align (dst_offset & 3)
int src1_index = mad24(y, src1_step, x + src1_offset - dst_align);
int src2_index = mad24(y, src2_step, x + src2_offset - dst_align);
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + x & (int)0xfffffffc);
uchar4 src1_data ,src2_data;
src1_data.x= src1_index+0 >= 0 ? src1[src1_index+0] : 0;
src1_data.y= src1_index+1 >= 0 ? src1[src1_index+1] : 0;
src1_data.z= src1_index+2 >= 0 ? src1[src1_index+2] : 0;
src1_data.w= src1_index+3 >= 0 ? src1[src1_index+3] : 0;
src2_data.x= src2_index+0 >= 0 ? src2[src2_index+0] : 0;
src2_data.y= src2_index+1 >= 0 ? src2[src2_index+1] : 0;
src2_data.z= src2_index+2 >= 0 ? src2[src2_index+2] : 0;
src2_data.w= src2_index+3 >= 0 ? src2[src2_index+3] : 0;
uchar4 dst_data = *((__global uchar4 *)(dst + dst_index));
int4 tmp = convert_int4_sat(src1_data) * convert_int4_sat(src2_data);
tmp = round_int4(convert_float4(tmp) * scalar);
uchar4 tmp_data = convert_uchar4_sat(tmp);
dst_data.x = ((dst_index + 0 >= dst_start) && (dst_index + 0 < dst_end)) ? tmp_data.x : dst_data.x;
dst_data.y = ((dst_index + 1 >= dst_start) && (dst_index + 1 < dst_end)) ? tmp_data.y : dst_data.y;
dst_data.z = ((dst_index + 2 >= dst_start) && (dst_index + 2 < dst_end)) ? tmp_data.z : dst_data.z;
dst_data.w = ((dst_index + 3 >= dst_start) && (dst_index + 3 < dst_end)) ? tmp_data.w : dst_data.w;
*((__global uchar4 *)(dst + dst_index)) = dst_data;
}
}
__kernel void arithm_mul_D2 (__global ushort *src1, int src1_step, int src1_offset,
__global ushort *src2, int src2_step, int src2_offset,
__global ushort *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, float scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset >> 1) & 3)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int src2_index = mad24(y, src2_step, (x << 1) + src2_offset - (dst_align << 1));
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffff8);
ushort4 src1_data = vload4(0, (__global ushort *)((__global char *)src1 + src1_index));
ushort4 src2_data = vload4(0, (__global ushort *)((__global char *)src2 + src2_index));
ushort4 dst_data = *((__global ushort4 *)((__global char *)dst + dst_index));
uint4 tmp = convert_uint4_sat(src1_data) * convert_uint4_sat(src2_data);
tmp = round_uint4(convert_float4(tmp) * scalar);
ushort4 tmp_data = convert_ushort4_sat(tmp);
dst_data.x = ((dst_index + 0 >= dst_start) && (dst_index + 0 < dst_end)) ? tmp_data.x : dst_data.x;
dst_data.y = ((dst_index + 2 >= dst_start) && (dst_index + 2 < dst_end)) ? tmp_data.y : dst_data.y;
dst_data.z = ((dst_index + 4 >= dst_start) && (dst_index + 4 < dst_end)) ? tmp_data.z : dst_data.z;
dst_data.w = ((dst_index + 6 >= dst_start) && (dst_index + 6 < dst_end)) ? tmp_data.w : dst_data.w;
*((__global ushort4 *)((__global char *)dst + dst_index)) = dst_data;
}
}
__kernel void arithm_mul_D3 (__global short *src1, int src1_step, int src1_offset,
__global short *src2, int src2_step, int src2_offset,
__global short *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, float scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
x = x << 2;
#ifdef dst_align
#undef dst_align
#endif
#define dst_align ((dst_offset >> 1) & 3)
int src1_index = mad24(y, src1_step, (x << 1) + src1_offset - (dst_align << 1));
int src2_index = mad24(y, src2_step, (x << 1) + src2_offset - (dst_align << 1));
int dst_start = mad24(y, dst_step, dst_offset);
int dst_end = mad24(y, dst_step, dst_offset + dst_step1);
int dst_index = mad24(y, dst_step, dst_offset + (x << 1) & (int)0xfffffff8);
short4 src1_data = vload4(0, (__global short *)((__global char *)src1 + src1_index));
short4 src2_data = vload4(0, (__global short *)((__global char *)src2 + src2_index));
short4 dst_data = *((__global short4 *)((__global char *)dst + dst_index));
int4 tmp = convert_int4_sat(src1_data) * convert_int4_sat(src2_data);
tmp = round_int4(convert_float4(tmp) * scalar);
short4 tmp_data = convert_short4_sat(tmp);
dst_data.x = ((dst_index + 0 >= dst_start) && (dst_index + 0 < dst_end)) ? tmp_data.x : dst_data.x;
dst_data.y = ((dst_index + 2 >= dst_start) && (dst_index + 2 < dst_end)) ? tmp_data.y : dst_data.y;
dst_data.z = ((dst_index + 4 >= dst_start) && (dst_index + 4 < dst_end)) ? tmp_data.z : dst_data.z;
dst_data.w = ((dst_index + 6 >= dst_start) && (dst_index + 6 < dst_end)) ? tmp_data.w : dst_data.w;
*((__global short4 *)((__global char *)dst + dst_index)) = dst_data;
}
}
__kernel void arithm_mul_D4 (__global int *src1, int src1_step, int src1_offset,
__global int *src2, int src2_step, int src2_offset,
__global int *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, float scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 2) + src2_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
int data1 = *((__global int *)((__global char *)src1 + src1_index));
int data2 = *((__global int *)((__global char *)src2 + src2_index));
int tmp = data1 * data2;
tmp = round_int((float)tmp * scalar);
*((__global int *)((__global char *)dst + dst_index)) = convert_int_sat(tmp);
}
}
__kernel void arithm_mul_D5 (__global float *src1, int src1_step, int src1_offset,
__global float *src2, int src2_step, int src2_offset,
__global float *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, float scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 2) + src2_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
float data1 = *((__global float *)((__global char *)src1 + src1_index));
float data2 = *((__global float *)((__global char *)src2 + src2_index));
float tmp = data1 * data2;
tmp = tmp * scalar;
*((__global float *)((__global char *)dst + dst_index)) = tmp;
}
}
#if defined (DOUBLE_SUPPORT)
__kernel void arithm_mul_D6 (__global double *src1, int src1_step, int src1_offset,
__global double *src2, int src2_step, int src2_offset,
__global double *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, double scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 3) + src1_offset);
int src2_index = mad24(y, src2_step, (x << 3) + src2_offset);
int dst_index = mad24(y, dst_step, (x << 3) + dst_offset);
double data1 = *((__global double *)((__global char *)src1 + src1_index));
double data2 = *((__global double *)((__global char *)src2 + src2_index));
double tmp = data1 * data2;
tmp = tmp * scalar;
*((__global double *)((__global char *)dst + dst_index)) = tmp;
}
}
#endif
#ifdef DOUBLE_SUPPORT
#define SCALAR_TYPE double
#else
#define SCALAR_TYPE float
#endif
__kernel void arithm_muls_D5 (__global float *src1, int src1_step, int src1_offset,
__global float *dst, int dst_step, int dst_offset,
int rows, int cols, int dst_step1, SCALAR_TYPE scalar)
{
int x = get_global_id(0);
int y = get_global_id(1);
if (x < cols && y < rows)
{
int src1_index = mad24(y, src1_step, (x << 2) + src1_offset);
int dst_index = mad24(y, dst_step, (x << 2) + dst_offset);
float data1 = *((__global float *)((__global char *)src1 + src1_index));
float tmp = data1 * scalar;
*((__global float *)((__global char *)dst + dst_index)) = tmp;
}
}
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment