Package com.jogamp.opencl

Examples of com.jogamp.opencl.CLProgram$CompilerOptions

43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
            int elementCount = 1444477;                                  // Length of arrays to process
            int localWorkSize = min(device.getMaxWorkGroupSize(), 256);  // Local work size dimensions
            int globalWorkSize = roundUp(localWorkSize, elementCount);   // rounded up to the nearest multiple of the localWorkSize

            // load sources, create and build program
            CLProgram program = context.createProgram(HelloJOCL.class.getResourceAsStream("VectorAdd.cl")).build();

            // A, B are input buffers, C is for the result
            CLBuffer<FloatBuffer> clBufferA = context.createFloatBuffer(globalWorkSize, READ_ONLY);
            CLBuffer<FloatBuffer> clBufferB = context.createFloatBuffer(globalWorkSize, READ_ONLY);
            CLBuffer<FloatBuffer> clBufferC = context.createFloatBuffer(globalWorkSize, WRITE_ONLY);

            out.println("used device memory: "
                + (clBufferA.getCLSize()+clBufferB.getCLSize()+clBufferC.getCLSize())/1000000 +"MB");

            // fill input buffers with random numbers
            // (just to have test data; seed is fixed -> results will not change between runs).
            fillBuffer(clBufferA.getBuffer(), 12345);
            fillBuffer(clBufferB.getBuffer(), 67890);

            // get a reference to the kernel function with the name 'VectorAdd'
            // and map the buffers to its input parameters.
            CLKernel kernel = program.createCLKernel("VectorAdd");
            kernel.putArgs(clBufferA, clBufferB, clBufferC).putArg(elementCount);

            // asynchronous write of data to GPU device,
            // followed by blocking read to get the computed results back.
            long time = nanoTime();
View Full Code Here
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
       
        CLContext context = CLContext.create(platform.getMaxFlopsDevice());
       
        try{
            //load and compile program for the chosen device
            CLProgram program = context.createProgram(getStreamFor("Gamma.cl"));
            program.build(CompilerOptions.FAST_RELAXED_MATH);
            assert program.isExecutable();
           
            // load image
            BufferedImage image = readImage("lena.png");
            assert image.getColorModel().getNumComponents() == 3;
           
            float[] pixels = image.getRaster().getPixels(0, 0, image.getWidth(), image.getHeight(), (float[])null);
           
            // copy to direct float buffer
            FloatBuffer fb = Buffers.newDirectFloatBuffer(pixels);
           
            // allocate a OpenCL buffer using the direct fb as working copy
            CLBuffer<FloatBuffer> buffer = context.createBuffer(fb, CLBuffer.Mem.READ_WRITE);
           
            // creade a command queue with benchmarking flag set
            CLCommandQueue queue = context.getDevices()[0].createCommandQueue(Mode.PROFILING_MODE);
           
            int localWorkSize = queue.getDevice().getMaxWorkGroupSize(); // Local work size dimensions
            int globalWorkSize = roundUp(localWorkSize, fb.capacity());  // rounded up to the nearest multiple of the localWorkSize
           
            // create kernel and set function parameters
            CLKernel kernel = program.createCLKernel("gamma");
           
            // original lenna
            show(image, 0, 50, "reference");
           
            // a few gamma corrected versions
View Full Code Here
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
       
        CLContext context = CLContext.create(platform.getMaxFlopsDevice());
       
        try{
            //load and compile program for the chosen device
            CLProgram program = context.createProgram(getStreamFor("Gamma.cl"));
            program.build(CompilerOptions.FAST_RELAXED_MATH);
            assert program.isExecutable();
           
            // load image
            BufferedImage image = readImage("lena.png");
            assert image.getColorModel().getNumComponents() == 3;
           
            float[] pixels = image.getRaster().getPixels(0, 0, image.getWidth(), image.getHeight(), (float[])null);
           
            // copy to direct float buffer
            FloatBuffer fb = Buffers.newDirectFloatBuffer(pixels);
           
            // allocate a OpenCL buffer using the direct fb as working copy
            CLBuffer<FloatBuffer> buffer = context.createBuffer(fb, CLBuffer.Mem.READ_WRITE);
           
            // creade a command queue with benchmarking flag set
            CLCommandQueue queue = context.getDevices()[0].createCommandQueue(Mode.PROFILING_MODE);
           
            int localWorkSize = queue.getDevice().getMaxWorkGroupSize(); // Local work size dimensions
            int globalWorkSize = roundUp(localWorkSize, fb.capacity());  // rounded up to the nearest multiple of the localWorkSize
           
            // create kernel and set function parameters
            CLKernel kernel = program.createCLKernel("gamma");
           
            // original lenna
            show(image, 0, 50, "reference");
           
            // a few gamma corrected versions
View Full Code Here
93
94
95
96
97
98
99
100
101
102
103
104
105
106
        out.println("    creating bitonic sort program");

        CLContext context = queue.getContext();

        CLProgram program = context.createProgram(getClass().getResourceAsStream("BitonicSort.cl"))
                                   .build(define("LOCAL_SIZE_LIMIT", LOCAL_SIZE_LIMIT));

        Map<String, CLKernel> kernelMap = program.createCLKernels();

        out.println("    checking minimum supported workgroup size");
        //Check for work group size
        CLDevice device = queue.getDevice();
        long szBitonicSortLocal  = kernelMap.get(BITONIC_SORT_LOCAL).getWorkGroupSize(device);
View Full Code Here
93
94
95
96
97
98
99
100
101
102
103
104
105
106
        out.println("    creating bitonic sort program");

        CLContext context = queue.getContext();

        CLProgram program = context.createProgram(getClass().getResourceAsStream("BitonicSort.cl"))
                                   .build(define("LOCAL_SIZE_LIMIT", LOCAL_SIZE_LIMIT));

        Map<String, CLKernel> kernelMap = program.createCLKernels();

        out.println("    checking minimum supported workgroup size");
        //Check for work group size
        CLDevice device = queue.getDevice();
        long szBitonicSortLocal  = kernelMap.get(BITONIC_SORT_LOCAL).getWorkGroupSize(device);
View Full Code Here
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
        }
    }

    private void initCL() {

        CLProgram program;
        try {
            program = clContext.createProgram(getClass().getResourceAsStream("JoglInterop.cl"));
            program.build();
            System.out.println(program.getBuildStatus());
            System.out.println(program.isExecutable());
            System.out.println(program.getBuildLog());
        } catch (IOException ex) {
            throw new RuntimeException("can not handle exception", ex);
        }

        commandQueue = clContext.getMaxFlopsDevice().createCommandQueue();

        clBuffer = clContext.createFromGLBuffer(glObjects[VERTICES], BUFFER_SIZE, CLGLBuffer.Mem.WRITE_ONLY);
        System.out.println("clsize: "+clBuffer.getCLSize());

        System.out.println("cl buffer type: " + clBuffer.getGLObjectType());
        System.out.println("shared with gl buffer: " + clBuffer.getGLObjectID());

        kernel = program.createCLKernel("sineWave")
                        .putArg(clBuffer)
                        .putArg(MESH_SIZE)
                        .rewind();

        System.out.println("cl initialised");
View Full Code Here
349
350
351
352
353
354
355
356
357
358
359
360
361
        rGBuffer = cl.createFloatBuffer(width * height * 2, Mem.READ_WRITE);
        gGBuffer = cl.createFloatBuffer(width * height * 2, Mem.READ_WRITE);
        bGBuffer = cl.createFloatBuffer(width * height * 2, Mem.READ_WRITE);
        if (false) {
            try {
                CLProgram p = cl.createProgram(new FileInputStream("/home/notzed/cl/fft-512.cl"));
                p.build();
                fft512 = p.createCLKernel("fft0");
            } catch (IOException ex) {
                Logger.getLogger(BlurTest.class.getName()).log(Level.SEVERE, null, ex);
            }
        } else {
            fft = new CLFFTPlan(cl, new int[]{width, height}, CLFFTPlan.CLFFTDataFormat.InterleavedComplexFormat);
View Full Code Here
TOP

Related Classes of com.jogamp.opencl.CLProgram$CompilerOptions

  • com.jogamp.common.nio.CachedBufferFactory
  • com.jogamp.common.nio.NativeSizeBuffer
  • com.jogamp.common.nio.PointerBuffer
  • com.jogamp.opencl.demos.bitonicsort.BitonicSort
  • com.jogamp.opencl.demos.fft.BlurTest
  • com.jogamp.opencl.demos.gamma.CLSimpleGammaCorrection
  • com.jogamp.opencl.demos.hellojocl.HelloJOCL
  • com.jogamp.opencl.demos.joglinterop.GLCLInteroperabilityDemo
  • com.jogamp.opencl.llb.CLKernelBinding
  • com.jogamp.opencl.llb.CLProgramBinding
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.