Skip to content
Projects
Groups
Snippets
Help
Loading...
Sign in / Register
Toggle navigation
O
opencv
Project
Project
Details
Activity
Cycle Analytics
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Issues
0
Issues
0
List
Board
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Charts
Packages
Packages
Wiki
Wiki
Snippets
Snippets
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Charts
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
submodule
opencv
Commits
ace7c7e9
Commit
ace7c7e9
authored
Jul 19, 2010
by
Andrey Morozov
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
linux compiletion error
parent
07825bad
Show whitespace changes
Inline
Side-by-side
Showing
5 changed files
with
178 additions
and
12 deletions
+178
-12
matrix_operations.hpp
modules/gpu/include/opencv2/gpu/matrix_operations.hpp
+8
-8
cuda_shared.hpp
modules/gpu/src/cuda/cuda_shared.hpp
+3
-0
matrix_operations.cu
modules/gpu/src/cuda/matrix_operations.cu
+150
-0
cudastream.cpp
modules/gpu/src/cudastream.cpp
+1
-0
matrix_operations.cpp
modules/gpu/src/matrix_operations.cpp
+16
-4
No files found.
modules/gpu/include/opencv2/gpu/matrix_operations.hpp
View file @
ace7c7e9
...
...
@@ -342,27 +342,27 @@ static inline void swap( GpuMat& a, GpuMat& b ) { a.swap(b); }
//////////////////////////////// MatPL ////////////////////////////////
///////////////////////////////////////////////////////////////////////
MatPL
::
MatPL
()
:
flags
(
0
),
rows
(
0
),
cols
(
0
),
step
(
0
),
data
(
0
),
refcount
(
0
),
datastart
(
0
),
dataend
(
0
)
{}
MatPL
::
MatPL
(
int
_rows
,
int
_cols
,
int
_type
)
:
flags
(
0
),
rows
(
0
),
cols
(
0
),
step
(
0
),
data
(
0
),
refcount
(
0
),
datastart
(
0
),
dataend
(
0
)
inline
MatPL
::
MatPL
()
:
flags
(
0
),
rows
(
0
),
cols
(
0
),
step
(
0
),
data
(
0
),
refcount
(
0
),
datastart
(
0
),
dataend
(
0
)
{}
inline
MatPL
::
MatPL
(
int
_rows
,
int
_cols
,
int
_type
)
:
flags
(
0
),
rows
(
0
),
cols
(
0
),
step
(
0
),
data
(
0
),
refcount
(
0
),
datastart
(
0
),
dataend
(
0
)
{
if
(
_rows
>
0
&&
_cols
>
0
)
create
(
_rows
,
_cols
,
_type
);
}
MatPL
::
MatPL
(
Size
_size
,
int
_type
)
:
flags
(
0
),
rows
(
0
),
cols
(
0
),
step
(
0
),
data
(
0
),
refcount
(
0
),
datastart
(
0
),
dataend
(
0
)
inline
MatPL
::
MatPL
(
Size
_size
,
int
_type
)
:
flags
(
0
),
rows
(
0
),
cols
(
0
),
step
(
0
),
data
(
0
),
refcount
(
0
),
datastart
(
0
),
dataend
(
0
)
{
if
(
_size
.
height
>
0
&&
_size
.
width
>
0
)
create
(
_size
.
height
,
_size
.
width
,
_type
);
}
MatPL
::
MatPL
(
const
MatPL
&
m
)
:
flags
(
m
.
flags
),
rows
(
m
.
rows
),
cols
(
m
.
cols
),
step
(
m
.
step
),
data
(
m
.
data
),
refcount
(
m
.
refcount
),
datastart
(
0
),
dataend
(
0
)
inline
MatPL
::
MatPL
(
const
MatPL
&
m
)
:
flags
(
m
.
flags
),
rows
(
m
.
rows
),
cols
(
m
.
cols
),
step
(
m
.
step
),
data
(
m
.
data
),
refcount
(
m
.
refcount
),
datastart
(
0
),
dataend
(
0
)
{
if
(
refcount
)
CV_XADD
(
refcount
,
1
);
}
MatPL
::
MatPL
(
const
Mat
&
m
)
:
flags
(
0
),
rows
(
0
),
cols
(
0
),
step
(
0
),
data
(
0
),
refcount
(
0
),
datastart
(
0
),
dataend
(
0
)
inline
MatPL
::
MatPL
(
const
Mat
&
m
)
:
flags
(
0
),
rows
(
0
),
cols
(
0
),
step
(
0
),
data
(
0
),
refcount
(
0
),
datastart
(
0
),
dataend
(
0
)
{
if
(
m
.
rows
>
0
&&
m
.
cols
>
0
)
create
(
m
.
size
(),
m
.
type
()
);
...
...
@@ -371,11 +371,11 @@ MatPL::MatPL(const Mat& m) : flags(0), rows(0), cols(0), step(0), data(0), refco
m
.
copyTo
(
tmp
);
}
MatPL
::~
MatPL
()
inline
MatPL
::~
MatPL
()
{
release
();
}
MatPL
&
MatPL
::
operator
=
(
const
MatPL
&
m
)
inline
MatPL
&
MatPL
::
operator
=
(
const
MatPL
&
m
)
{
if
(
this
!=
&
m
)
{
...
...
@@ -392,7 +392,7 @@ MatPL& MatPL::operator = (const MatPL& m)
return
*
this
;
}
MatPL
MatPL
::
clone
()
const
inline
MatPL
MatPL
::
clone
()
const
{
MatPL
m
(
size
(),
type
());
Mat
to
=
m
;
...
...
modules/gpu/src/cuda/cuda_shared.hpp
View file @
ace7c7e9
...
...
@@ -61,6 +61,9 @@ namespace cv
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
);
}
}
}
...
...
modules/gpu/src/cuda/matrix_operations.cu
0 → 100644
View file @
ace7c7e9
/*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);
}
}
modules/gpu/src/cudastream.cpp
View file @
ace7c7e9
...
...
@@ -138,6 +138,7 @@ bool cv::gpu::CudaStream::queryIfComplete()
return
err
==
cudaSuccess
;
cudaSafeCall
(
err
);
return
false
;
}
void
cv
::
gpu
::
CudaStream
::
waitForCompletion
()
{
cudaSafeCall
(
cudaStreamSynchronize
(
impl
->
stream
)
);
}
...
...
modules/gpu/src/matrix_operations.cpp
View file @
ace7c7e9
...
...
@@ -109,15 +109,27 @@ void cv::gpu::GpuMat::convertTo( GpuMat& /*m*/, int /*rtype*/, double /*alpha*/,
CV_Assert
(
!
"Not implemented"
);
}
GpuMat
&
cv
::
gpu
::
GpuMat
::
operator
=
(
const
Scalar
&
/*s*/
)
GpuMat
&
GpuMat
::
operator
=
(
const
Scalar
&
s
)
{
CV_Assert
(
!
"Not implemented"
);
cv
::
gpu
::
impl
::
set_to_without_mask
(
*
this
,
s
.
val
,
this
->
depth
(),
this
->
channels
());
return
*
this
;
}
GpuMat
&
cv
::
gpu
::
GpuMat
::
setTo
(
const
Scalar
&
/*s*/
,
const
GpuMat
&
/*mask*/
)
GpuMat
&
GpuMat
::
setTo
(
const
Scalar
&
s
,
const
GpuMat
&
mask
)
{
CV_Assert
(
!
"Not implemented"
);
CV_Assert
(
mask
.
type
()
==
CV_8U
);
CV_DbgAssert
(
!
this
->
empty
());
if
(
mask
.
empty
())
{
cv
::
gpu
::
impl
::
set_to_without_mask
(
*
this
,
s
.
val
,
this
->
depth
(),
this
->
channels
());
}
else
{
cv
::
gpu
::
impl
::
set_to_with_mask
(
*
this
,
s
.
val
,
mask
,
this
->
depth
(),
this
->
channels
());
}
return
*
this
;
}
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment