AM5728的opencv SVM问题

我重新编译了一下ti的opencv3.1.0，加入了编译选项opencl_svm，然后运行时提示svminit 返回错误CL_INVALID_VALUE, AM5728的dsp不支持SVM吗？

请关注下面的帖子，要到下周才有回复了。
e2e.ti.com/.../774567

不清楚在自己交叉编译opencv加入OPENCL选项的时候是不是一样的

My source code is from git://git.ti.com/opencv/tiopencv.git . I open the compiled option "WITH_OPENCL_SVM ".
NOTICE:
In file included from /media/gtguo/911001D492D4A8AF/ti/tiopencv/modules/core/src/ocl.cpp:1479:0:
/media/gtguo/911001D492D4A8AF/ti/tiopencv/modules/core/include/opencv2/core/opencl/runtime/opencl_svm_hsa_extension.hpp:78:3: error: expected initializer before 'CL_EXT_SUFFIX__VERSION_1_2'
) CL_EXT_SUFFIX__VERSION_1_2;

WHY?

root@embest-am57xx:~# ./OPENCV_CANNY
haveSVM = 1
haveOPENCL = 1
Error on line 2632 (ocl.cpp): CL_DEVICE_SVM_CAPABILITIES via clGetDeviceInfo failed: -30
sh: -c: line 0: syntax error near unexpected token `('
sh: -c: line 0: `clocl -d -I. -D RADIUSX=2 -D LSIZE0=16 -D LSIZE1=16 -D CN=1 -D BORDER_REFLECT_101 -D NO_EXTRA_EXTRAPOLATION -D NO_BOR'
/tmp/openclt58qQy.cl:59:21: error: array is too large (1207959696 elements)
/tmp/openclt58qQy.cl:72:19: error: array is too large (1140850756 elements)
OpenCL program build log: -D WITH_SOBEL -D cn=1 -D TYPE=uchar -D convert_floatN=convert_float -D floatN=float -D GRP_SIZEX=32 -D GRP_S2
/tmp/openclt58qQy.cl:59:21: error: array is too large (1207959696 elements)
/tmp/openclt58qQy.cl:72:19: error: array is too large (1140850756 elements)

Current OpenCL device:
Type = TI_DSP
Name = TI Multicore C66 DSP
Version = OpenCL 1.1 TI => Is it support OPENCL_SVM ?
Compute units = 2

你用的SDK是什么版本？

完整的OPENCV使用教程请参考这个链接：www.ti.com/.../PROCESSOR-SDK-AM57X

I have used PLSDK 5.2 filesystem to test today, it was not OK stilly !

test.cpp:

#include <iostream>
#include <opencv2/opencv.hpp>
#include <opencv2/core/ocl.hpp>
#include <unistd.h>
using namespace std;
using namespace cv;
using namespace cv::ml;

int main()
{
int result = 0;
float response;
UMat inMat;

cout<<"haveOpenCL:";
cout<<cv::ocl::haveOpenCL()<<endl;
cout<<"useOpenCL:";
cout<<cv::ocl::useOpenCL()<<endl;
cout<<"haveSVM:";
cout<<cv::ocl::haveSVM()<<endl;

Ptr<SVM> svm = Algorithm::load<SVM>("model.xml");
char buffer[40];
for(int i=0;i<100;i++){
memset(buffer,0,sizeof(buffer));
sprintf(buffer,"./data/test_image/0/%d.jpg",i+400);
imread(buffer).copyTo(inMat);
UMat p = inMat.reshape(1, 1);
p.convertTo(p, CV_32FC1);
if(i==99){//while 1
i=0;
response = svm->predict(p);
}
if (response == 0)
{
result++;
}
}
cout<<result<<endl;
return 0;
}

root@am57xx-evm:/opt# ./TEST_SVM

haveOpenCL:1

useOpenCL:1

haveSVM:0

top to check cpu used:
TEST_SVM used 94% CPU, it didn't used DSP by opencl.
1144 root 20 0 267156 18644 15720 R 94.1 1.0 0:15.68 TEST_SVM

这里有opencl的使用指南和示例代码：downloads.ti.com/.../intro.html