generic-library/opencv
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

linux compiletion error

Browse Source
This commit is contained in:
Andrey Morozov
2010-07-19 10:49:35 +00:00
parent 07825bad1e
commit ace7c7e93c
5 changed files with 213 additions and 47 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
modules/gpu/src/cuda/cuda_shared.hpp
View File

@@ -44,23 +44,26 @@
#define __OPENCV_CUDA_SHARED_HPP__
#include "opencv2/gpu/devmem2d.hpp"
#include "cuda_runtime_api.h"
#include "cuda_runtime_api.h"
namespace cv
{
namespace gpu
{
{
typedef unsigned char uchar;
typedef unsigned short ushort;
typedef unsigned int uint;
typedef unsigned int uint;
extern "C" void error( const char *error_string, const char *file, const int line, const char *func = "");
namespace impl
{
{
static inline int divUp(int a, int b) { return (a % b == 0) ? a/b : a/b + 1; }
extern "C" void stereoBM_GPU(const DevMem2D& left, const DevMem2D& right, DevMem2D& disp, int maxdisp, DevMem2D_<uint>& minSSD_buf);
extern "C" void set_to_without_mask (const DevMem2D& mat, const double * scalar, int depth, int channels);
extern "C" void set_to_with_mask (const DevMem2D& mat, const double * scalar, const DevMem2D& mask, int depth, int channels);
}
}
}
@@ -68,12 +71,12 @@ namespace cv
#if defined(__GNUC__)
#define cudaSafeCall(expr) ___cudaSafeCall(expr, __FILE__, __LINE__, __func__);
#else /* defined(__CUDACC__) || defined(__MSVC__) */
#define cudaSafeCall(expr) ___cudaSafeCall(expr, __FILE__, __LINE__)
#define cudaSafeCall(expr) ___cudaSafeCall(expr, __FILE__, __LINE__)
#endif
static inline void ___cudaSafeCall(cudaError_t err, const char *file, const int line, const char *func = "")
{
if( cudaSuccess != err)
if( cudaSuccess != err)
cv::gpu::error(cudaGetErrorString(err), __FILE__, __LINE__, func);
}

150
modules/gpu/src/cuda/matrix_operations.cu Normal file
View File

@@ -0,0 +1,150 @@
/*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) 2000-2008, Intel Corporation, all rights reserved.
// Copyright (C) 2009, Willow Garage Inc., all rights reserved.
// Third party copyrights are property of their respective owners.
//
// 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 materials 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*/
#include <stddef.h>
#include "cuda_shared.hpp"
#include "cuda_runtime.h"
__constant__ float scalar_d[4];
namespace mat_operators
{
template <typename T, int channels, int count = channels>
struct unroll
{
__device__ static void unroll_set(T * mat, size_t i)
{
mat[i] = static_cast<T>(scalar_d[i % channels]);
unroll<T, channels, count - 1>::unroll_set(mat, i+1);
}
__device__ static void unroll_set_with_mask(T * mat, float mask, size_t i)
{
mat[i] = mask * static_cast<T>(scalar_d[i % channels]);
unroll<T, channels, count - 1>::unroll_set_with_mask(mat, mask, i+1);
}
};
template <typename T, int channels>
struct unroll<T,channels,0>
{
__device__ static void unroll_set(T * , size_t){}
__device__ static void unroll_set_with_mask(T * , float, size_t){}
};
template <typename T, int channels>
__global__ void kernel_set_to_without_mask(T * mat)
{
size_t i = (blockIdx.x * blockDim.x + threadIdx.x) * sizeof(T);
unroll<T, channels>::unroll_set(mat, i);
}
template <typename T, int channels>
__global__ void kernel_set_to_with_mask(T * mat, const float * mask)
{
size_t i = (blockIdx.x * blockDim.x + threadIdx.x) * sizeof(T);
unroll<T, channels>::unroll_set_with_mask(mat, i, mask[i]);
}
}
extern "C" void cv::gpu::impl::set_to_with_mask(const DevMem2D& mat, const double * scalar, const DevMem2D& mask, int depth, int channels)
{
scalar_d[0] = scalar[0];
scalar_d[1] = scalar[1];
scalar_d[2] = scalar[2];
scalar_d[3] = scalar[3];
dim3 numBlocks(mat.rows * mat.step / 256, 1, 1);
dim3 threadsPerBlock(256);
if (channels == 1)
{
if (depth == 1) ::mat_operators::kernel_set_to_with_mask<unsigned char, 1><<<numBlocks,threadsPerBlock>>>(mat.ptr, (float *)mask.ptr);
if (depth == 2) ::mat_operators::kernel_set_to_with_mask<unsigned short, 1><<<numBlocks,threadsPerBlock>>>((unsigned short *)mat.ptr, (float *)mask.ptr);
if (depth == 4) ::mat_operators::kernel_set_to_with_mask<unsigned int, 1><<<numBlocks,threadsPerBlock>>>((unsigned int *)mat.ptr, (float *)mask.ptr);
}
if (channels == 2)
{
if (depth == 1) ::mat_operators::kernel_set_to_with_mask<unsigned char, 2><<<numBlocks,threadsPerBlock>>>(mat.ptr, (float *)mask.ptr);
if (depth == 2) ::mat_operators::kernel_set_to_with_mask<unsigned short, 2><<<numBlocks,threadsPerBlock>>>((unsigned short *)mat.ptr, (float *)mask.ptr);
if (depth == 4) ::mat_operators::kernel_set_to_with_mask<unsigned int, 2><<<numBlocks,threadsPerBlock>>>((unsigned int *)mat.ptr, (float *)mask.ptr);
}
if (channels == 3)
{
if (depth == 1) ::mat_operators::kernel_set_to_with_mask<unsigned char, 3><<<numBlocks,threadsPerBlock>>>(mat.ptr, (float *)mask.ptr);
if (depth == 2) ::mat_operators::kernel_set_to_with_mask<unsigned short, 3><<<numBlocks,threadsPerBlock>>>((unsigned short *)mat.ptr, (float *)mask.ptr);
if (depth == 4) ::mat_operators::kernel_set_to_with_mask<unsigned int, 3><<<numBlocks,threadsPerBlock>>>((unsigned int *)mat.ptr, (float *)mask.ptr);
}
}
extern "C" void cv::gpu::impl::set_to_without_mask(const DevMem2D& mat, const double * scalar, int depth, int channels)
{
scalar_d[0] = scalar[0];
scalar_d[1] = scalar[1];
scalar_d[2] = scalar[2];
scalar_d[3] = scalar[3];
int numBlocks = mat.rows * mat.step / 256;
dim3 threadsPerBlock(256);
if (channels == 1)
{
if (depth == 1) ::mat_operators::kernel_set_to_without_mask<unsigned char, 1><<<numBlocks,threadsPerBlock>>>(mat.ptr);
if (depth == 2) ::mat_operators::kernel_set_to_without_mask<unsigned short, 1><<<numBlocks,threadsPerBlock>>>((unsigned short *)mat.ptr);
if (depth == 4) ::mat_operators::kernel_set_to_without_mask<unsigned int, 1><<<numBlocks,threadsPerBlock>>>((unsigned int *)mat.ptr);
}
if (channels == 2)
{
if (depth == 1) ::mat_operators::kernel_set_to_without_mask<unsigned char, 2><<<numBlocks,threadsPerBlock>>>(mat.ptr);
if (depth == 2) ::mat_operators::kernel_set_to_without_mask<unsigned short, 2><<<numBlocks,threadsPerBlock>>>((unsigned short *)mat.ptr);
if (depth == 4) ::mat_operators::kernel_set_to_without_mask<unsigned int, 2><<<numBlocks,threadsPerBlock>>>((unsigned int *)mat.ptr);
}
if (channels == 3)
{
if (depth == 1) ::mat_operators::kernel_set_to_without_mask<unsigned char, 3><<<numBlocks,threadsPerBlock>>>(mat.ptr);
if (depth == 2) ::mat_operators::kernel_set_to_without_mask<unsigned short, 3><<<numBlocks,threadsPerBlock>>>((unsigned short *)mat.ptr);
if (depth == 4) ::mat_operators::kernel_set_to_without_mask<unsigned int, 3><<<numBlocks,threadsPerBlock>>>((unsigned int *)mat.ptr);
}
}