jocl.CLProgram.create() line 74

Here is an excerpt from a slightly modified version of ...
http://dev.loci.wisc.edu/svn/software/branches/maven/projects/opencl-decon/src/demos/SobelFilterExample.java

... // begin source reference

// Create a context from GPU
context = CLContext.create( Type.GPU );
               
// create the program
program = context.createProgram( openCLCodeString );
               
System.out.println( "My source is : " +  program.getSource());
program.build( "" );
               
...// end source reference


That generates the output

...//begin output

Retrieving test image...  
#pragma OPENCL EXTENSION cl_khr_fp64: enable
__kernel void sobel( __global float* input,
        __global float* output,
     int width,
     int height )
{  
    int x = get_global_id(0);
    int y = get_global_id(1);
    float p0, p1, p2, p3, p5, p6, p7, p8 = 0;
    int offset = y * width + x;
   
        if( x < 1 || y < 1 || x > width - 2 || y > height - 2 )
        {
          output[offset] = 0;
        }
        else
        {
            p0 = input[offset - width - 1] ;
            p1 = input[offset - width] ;
            p2 = input[offset - width + 1] ;
            p3 = input[offset - 1] ;
            p5 = input[offset + 1] ;
            p6 = input[offset + width - 1] ;
            p7 = input[offset + width] ;
            p8 = input[offset + width + 1] ;
       
            double sum1 = p0 + 2*p1 + p2 - p6 - 2*p7 - p8;  //GY
            double sum2 = p0 + 2*p3 + p6 - p2 - 2*p5 - p8;  //GX
           
            output[offset] = sqrt(  sum1*sum1 + sum2*sum2 );
        }
}


Starting iteration... 0
My source is : #pragm
The average OpenCL set up time is 0
The average OpenCL IO transfer time is 0
The average OpenCL execution time is 0
The average Java execution time is 0
com.jogamp.opencl.CLException$CLBuildProgramFailureException:
CLDevice [id: 16918016 name: GeForce 8600M GT type: GPU profile: FULL_PROFILE] build log:
<program source>:1:2: error: invalid preprocessing directive
 #pragm
  ^ [error: CL_BUILD_PROGRAM_FAILURE]
        at com.jogamp.opencl.CLException.newException(CLException.java:78)
        at com.jogamp.opencl.CLProgram.build(CLProgram.java:363)
        at com.jogamp.opencl.CLProgram.build(CLProgram.java:245)
        at demos.SobelFilterExample.init(SobelFilterExample.java:69)
        at demos.SobelFilterExample.runTest(SobelFilterExample.java:244)
        at demos.SobelFilterExample.main(SobelFilterExample.java:197)

... // end output

Executed on OSX 10.6.5
Runs fine on Ubuntu...

The static CLProgram create(CLContext context, String src) {} method doesn't seem like it has time to finish copying the string from source before returning since the source available in the output error message is of varied length for independent runs.

I would appreciate any thoughts or insight into this issue.

Regards,

Rick Lentz

Hello Rick,

thanks for the report. According to the spec the method should not return until the source is "loaded into the program". The JNI code looks also good at the first glance (proper reference locking etc).

Next problem is: i can't reproduce it.

Could you try the snippet below once with and once without the with xxx marked line?

        //... create context using high level api as you did before
        //...
        long context = highLevelContext.ID;
        IntBuffer intBuffer = Buffers.newDirectIntBuffer(1);

        CL cl = CLPlatform.getLowLevelCLInterface();
        PointerBuffer lenght = (PointerBuffer)PointerBuffer.allocateDirect(1).put(programSource.length());
        String[] srcArray = new String[] {programSource};
        final long program = cl.clCreateProgramWithSource(context, 1, srcArray, lenght, intBuffer);
        checkError("on clCreateProgramWithSource", intBuffer.get(0));

         // Build the program
         ret = cl.clBuildProgram(program, 0, null, null, null);
         checkError("on clBuildProgram", ret);

 xxx   System.out.println(length +" "+srcArray); // please ignore, just a artificial reference lock


    private final void checkError(String msg, int ret) {
        if(ret != CL.CL_SUCCESS)
            throw CLException.newException(ret, msg);
    }

thanks and best regards,

michael


On 01/14/2011 12:31 AM, Rick Lentz @ imagejdev.org [via jogamp] wrote:

Here is an excerpt from a slightly modified version of ...
http://dev.loci.wisc.edu/svn/software/branches/maven/projects/opencl-decon/src/demos/SobelFilterExample.java

... // begin source reference

// Create a context from GPU
context = CLContext.create( Type.GPU );
               
// create the program
program = context.createProgram( openCLCodeString );
               
System.out.println( "My source is : " +  program.getSource());
program.build( "" );
               
...// end source reference


That generates the output

...//begin output

Retrieving test image...  
#pragma OPENCL EXTENSION cl_khr_fp64: enable
__kernel void sobel( __global float* input,
        __global float* output,
     int width,
     int height )
{  
    int x = get_global_id(0);
    int y = get_global_id(1);
    float p0, p1, p2, p3, p5, p6, p7, p8 = 0;
    int offset = y * width + x;
   
        if( x < 1 || y < 1 || x > width - 2 || y > height - 2 )
        {
          output[offset] = 0;
        }
        else
        {
            p0 = input[offset - width - 1] ;
            p1 = input[offset - width] ;
            p2 = input[offset - width + 1] ;
            p3 = input[offset - 1] ;
            p5 = input[offset + 1] ;
            p6 = input[offset + width - 1] ;
            p7 = input[offset + width] ;
            p8 = input[offset + width + 1] ;
       
            double sum1 = p0 + 2*p1 + p2 - p6 - 2*p7 - p8;  //GY
            double sum2 = p0 + 2*p3 + p6 - p2 - 2*p5 - p8;  //GX
           
            output[offset] = sqrt(  sum1*sum1 + sum2*sum2 );
        }
}


Starting iteration... 0
My source is : #pragm
The average OpenCL set up time is 0
The average OpenCL IO transfer time is 0
The average OpenCL execution time is 0
The average Java execution time is 0
com.jogamp.opencl.CLException$CLBuildProgramFailureException:
CLDevice [id: 16918016 name: GeForce 8600M GT type: GPU profile: FULL_PROFILE] build log:
<program source>:1:2: error: invalid preprocessing directive
 #pragm
  ^ [error: CL_BUILD_PROGRAM_FAILURE]
        at com.jogamp.opencl.CLException.newException(CLException.java:78)
        at com.jogamp.opencl.CLProgram.build(CLProgram.java:363)
        at com.jogamp.opencl.CLProgram.build(CLProgram.java:245)
        at demos.SobelFilterExample.init(SobelFilterExample.java:69)
        at demos.SobelFilterExample.runTest(SobelFilterExample.java:244)
        at demos.SobelFilterExample.main(SobelFilterExample.java:197)

... // end output

Executed on OSX 10.6.5
Runs fine on Ubuntu...

The static CLProgram create(CLContext context, String src) {} method doesn't seem like it has time to finish copying the string from source before returning since the source available in the output error message is of varied length for independent runs.

I would appreciate any thoughts or insight into this issue.

Regards,

Rick Lentz

---

View message @ http://jogamp.762907.n3.nabble.com/jocl-CLProgram-create-line-74-tp2250272p2250272.html
To start a new topic under jogamp, email [hidden email]
To unsubscribe from jogamp, click here.

-- 
http://michael-bien.com/

Michael,

  Thank you for your excellent work on JOCL and for your response to my question. :-)

Here is the snippet that I tried:

        long context = this.context.ID;
        IntBuffer intBuffer = Buffers.newDirectIntBuffer(1);

        CL cl = CLPlatform.getLowLevelCLInterface();
        PointerBuffer pointBuffer = (PointerBuffer)PointerBuffer.allocateDirect(1).put( openCLCodeString.length() );
        String[] srcArray = new String[] { openCLCodeString };
        final long programLong = cl.clCreateProgramWithSource( context, 1, srcArray, pointBuffer, intBuffer );
        checkError("on clCreateProgramWithSource", intBuffer.get(0) );
       
         // Build the program
         int ret = cl.clBuildProgram( programLong, 0, null, null, null );
         
         // Print out the point buffer and the srcArray variable (as a String)
         String outSourceString = "";
         for(String sourceString : srcArray )
        outSourceString += sourceString;
         System.out.println( "Result value is " + ret + " pointBuffer is " + pointBuffer + " the input OpenCL code is "  + outSourceString ); // please ignore, just a artificial reference lock
       
         
         checkError("on clBuildProgram", ret);

And the results are for the uncommented println() version is:


Starting iteration... 0
Result value is -11 pointBuffer is PointerBuffer:AbstractBuffer[capacity 1, position 1, elementSize 8, ByteBuffer.capacity 8] the input OpenCL code is #pragma OPENCL EXTENSION cl_khr_fp64: enable
__kernel void sobel( __global float* input,
        __global float* output,
     int width,
     int height )
{  
    int x = get_global_id(0);
    int y = get_global_id(1);
    int offset = y * width + x;
   
    float p0, p1, p2, p3, p5, p6, p7, p8 = 0;
   
   
        if( x < 1 || y < 1 || x > width - 2 || y > height - 2 )
        {
          output[offset] = 0;
        }
        else
        {
            p0 = input[offset - width - 1] ;
            p1 = input[offset - width] ;
            p2 = input[offset - width + 1] ;
            p3 = input[offset - 1] ;
            p5 = input[offset + 1] ;
            p6 = input[offset + width - 1] ;
            p7 = input[offset + width] ;
            p8 = input[offset + width + 1] ;
       
            double sum1 = p0 + 2*p1 + p2 - p6 - 2*p7 - p8;  //GY
            double sum2 = p0 + 2*p3 + p6 - p2 - 2*p5 - p8;  //GX
           
            output[offset] = sqrt(  sum1*sum1 + sum2*sum2 );
        }
}

The average OpenCL set up time is 0
The average OpenCL IO transfer time is 0
The average OpenCL execution time is 0
The average Java execution time is 0
com.jogamp.opencl.CLException$CLBuildProgramFailureException: on clBuildProgram [error: CL_BUILD_PROGRAM_FAILURE]
        at com.jogamp.opencl.CLException.newException(CLException.java:78)
        at demos.SobelFilterExample.checkError(SobelFilterExample.java:62)
        at demos.SobelFilterExample.init(SobelFilterExample.java:96)
        at demos.SobelFilterExample.runTest(SobelFilterExample.java:275)
        at demos.SobelFilterExample.main(SobelFilterExample.java:228)

The result for the comment version is:

Starting iteration... 0
com.jogamp.opencl.CLException$CLBuildProgramFailureException: on clBuildProgram [error: CL_BUILD_PROGRAM_FAILURE]
The average OpenCL set up time is 0
The average OpenCL IO transfer time is 0
The average OpenCL execution time is 0
The average Java execution time is 0
        at com.jogamp.opencl.CLException.newException(CLException.java:78)
        at demos.SobelFilterExample.checkError(SobelFilterExample.java:66)
        at demos.SobelFilterExample.init(SobelFilterExample.java:100)
        at demos.SobelFilterExample.runTest(SobelFilterExample.java:275)
        at demos.SobelFilterExample.main(SobelFilterExample.java:228)

I am stuck on what might be causing the compile error.  Thank you for sharing your insight.
Best regards,

Rick Lentz

Rick,

please run the attached test. I slightly modified the kernel to use floats to make it portable across implementations. I did a few tests on linux and windows with amd, intel and nv CL implementations and it passed all the time.

You mentioned that its working on ubuntu, so its only reproducible on your mac?

I have a few exams next week, I am not sure how much time i will be able to spend for this bug the next few days..

regards,

michael

On 01/14/2011 04:41 AM, Rick Lentz @ imagejdev.org [via jogamp] wrote:

Michael,

  Thank you for your excellent work on JOCL and for your response to my question. :-)

Here is the snippet that I tried:

        long context = this.context.ID;
        IntBuffer intBuffer = Buffers.newDirectIntBuffer(1);

        CL cl = CLPlatform.getLowLevelCLInterface();
        PointerBuffer pointBuffer = (PointerBuffer)PointerBuffer.allocateDirect(1).put( openCLCodeString.length() );
        String[] srcArray = new String[] { openCLCodeString };
        final long programLong = cl.clCreateProgramWithSource( context, 1, srcArray, pointBuffer, intBuffer );
        checkError("on clCreateProgramWithSource", intBuffer.get(0) );
       
         // Build the program
         int ret = cl.clBuildProgram( programLong, 0, null, null, null );
         
         // Print out the point buffer and the srcArray variable (as a String)
         String outSourceString = "";
         for(String sourceString : srcArray )
        outSourceString += sourceString;
         System.out.println( "Result value is " + ret + " pointBuffer is " + pointBuffer + " the input OpenCL code is "  + outSourceString ); // please ignore, just a artificial reference lock
       
         
         checkError("on clBuildProgram", ret);

And the results are for the uncommented println() version is:


Starting iteration... 0
Result value is -11 pointBuffer is PointerBuffer:AbstractBuffer[capacity 1, position 1, elementSize 8, ByteBuffer.capacity 8] the input OpenCL code is #pragma OPENCL EXTENSION cl_khr_fp64: enable
__kernel void sobel( __global float* input,
        __global float* output,
     int width,
     int height )
{  
    int x = get_global_id(0);
    int y = get_global_id(1);
    int offset = y * width + x;
   
    float p0, p1, p2, p3, p5, p6, p7, p8 = 0;
   
   
        if( x < 1 || y < 1 || x > width - 2 || y > height - 2 )
        {
          output[offset] = 0;
        }
        else
        {
            p0 = input[offset - width - 1] ;
            p1 = input[offset - width] ;
            p2 = input[offset - width + 1] ;
            p3 = input[offset - 1] ;
            p5 = input[offset + 1] ;
            p6 = input[offset + width - 1] ;
            p7 = input[offset + width] ;
            p8 = input[offset + width + 1] ;
       
            double sum1 = p0 + 2*p1 + p2 - p6 - 2*p7 - p8;  //GY
            double sum2 = p0 + 2*p3 + p6 - p2 - 2*p5 - p8;  //GX
           
            output[offset] = sqrt(  sum1*sum1 + sum2*sum2 );
        }
}

The average OpenCL set up time is 0
The average OpenCL IO transfer time is 0
The average OpenCL execution time is 0
The average Java execution time is 0
com.jogamp.opencl.CLException$CLBuildProgramFailureException: on clBuildProgram [error: CL_BUILD_PROGRAM_FAILURE]
        at com.jogamp.opencl.CLException.newException(CLException.java:78)
        at demos.SobelFilterExample.checkError(SobelFilterExample.java:62)
        at demos.SobelFilterExample.init(SobelFilterExample.java:96)
        at demos.SobelFilterExample.runTest(SobelFilterExample.java:275)
        at demos.SobelFilterExample.main(SobelFilterExample.java:228)

The result for the comment version is:

Starting iteration... 0
com.jogamp.opencl.CLException$CLBuildProgramFailureException: on clBuildProgram [error: CL_BUILD_PROGRAM_FAILURE]
The average OpenCL set up time is 0
The average OpenCL IO transfer time is 0
The average OpenCL execution time is 0
The average Java execution time is 0
        at com.jogamp.opencl.CLException.newException(CLException.java:78)
        at demos.SobelFilterExample.checkError(SobelFilterExample.java:66)
        at demos.SobelFilterExample.init(SobelFilterExample.java:100)
        at demos.SobelFilterExample.runTest(SobelFilterExample.java:275)
        at demos.SobelFilterExample.main(SobelFilterExample.java:228)

I am stuck on what might be causing the compile error.  Thank you for sharing your insight.
Best regards,

Rick Lentz

---

View message @ http://jogamp.762907.n3.nabble.com/jocl-CLProgram-create-line-74-tp2250272p2250716.html
To start a new topic under jogamp, email [hidden email]
To unsubscribe from jogamp, click here.

-- 
http://michael-bien.com/

sorry, i uploaded the wrong testcase version, please take this one.

On 01/14/2011 04:41 AM, Rick Lentz @ imagejdev.org [via jogamp] wrote:

Michael,

  Thank you for your excellent work on JOCL and for your response to my question. :-)

Here is the snippet that I tried:

        long context = this.context.ID;
        IntBuffer intBuffer = Buffers.newDirectIntBuffer(1);

        CL cl = CLPlatform.getLowLevelCLInterface();
        PointerBuffer pointBuffer = (PointerBuffer)PointerBuffer.allocateDirect(1).put( openCLCodeString.length() );
        String[] srcArray = new String[] { openCLCodeString };
        final long programLong = cl.clCreateProgramWithSource( context, 1, srcArray, pointBuffer, intBuffer );
        checkError("on clCreateProgramWithSource", intBuffer.get(0) );
       
         // Build the program
         int ret = cl.clBuildProgram( programLong, 0, null, null, null );
         
         // Print out the point buffer and the srcArray variable (as a String)
         String outSourceString = "";
         for(String sourceString : srcArray )
        outSourceString += sourceString;
         System.out.println( "Result value is " + ret + " pointBuffer is " + pointBuffer + " the input OpenCL code is "  + outSourceString ); // please ignore, just a artificial reference lock
       
         
         checkError("on clBuildProgram", ret);

And the results are for the uncommented println() version is:


Starting iteration... 0
Result value is -11 pointBuffer is PointerBuffer:AbstractBuffer[capacity 1, position 1, elementSize 8, ByteBuffer.capacity 8] the input OpenCL code is #pragma OPENCL EXTENSION cl_khr_fp64: enable
__kernel void sobel( __global float* input,
        __global float* output,
     int width,
     int height )
{  
    int x = get_global_id(0);
    int y = get_global_id(1);
    int offset = y * width + x;
   
    float p0, p1, p2, p3, p5, p6, p7, p8 = 0;
   
   
        if( x < 1 || y < 1 || x > width - 2 || y > height - 2 )
        {
          output[offset] = 0;
        }
        else
        {
            p0 = input[offset - width - 1] ;
            p1 = input[offset - width] ;
            p2 = input[offset - width + 1] ;
            p3 = input[offset - 1] ;
            p5 = input[offset + 1] ;
            p6 = input[offset + width - 1] ;
            p7 = input[offset + width] ;
            p8 = input[offset + width + 1] ;
       
            double sum1 = p0 + 2*p1 + p2 - p6 - 2*p7 - p8;  //GY
            double sum2 = p0 + 2*p3 + p6 - p2 - 2*p5 - p8;  //GX
           
            output[offset] = sqrt(  sum1*sum1 + sum2*sum2 );
        }
}

The average OpenCL set up time is 0
The average OpenCL IO transfer time is 0
The average OpenCL execution time is 0
The average Java execution time is 0
com.jogamp.opencl.CLException$CLBuildProgramFailureException: on clBuildProgram [error: CL_BUILD_PROGRAM_FAILURE]
        at com.jogamp.opencl.CLException.newException(CLException.java:78)
        at demos.SobelFilterExample.checkError(SobelFilterExample.java:62)
        at demos.SobelFilterExample.init(SobelFilterExample.java:96)
        at demos.SobelFilterExample.runTest(SobelFilterExample.java:275)
        at demos.SobelFilterExample.main(SobelFilterExample.java:228)

The result for the comment version is:

Starting iteration... 0
com.jogamp.opencl.CLException$CLBuildProgramFailureException: on clBuildProgram [error: CL_BUILD_PROGRAM_FAILURE]
The average OpenCL set up time is 0
The average OpenCL IO transfer time is 0
The average OpenCL execution time is 0
The average Java execution time is 0
        at com.jogamp.opencl.CLException.newException(CLException.java:78)
        at demos.SobelFilterExample.checkError(SobelFilterExample.java:66)
        at demos.SobelFilterExample.init(SobelFilterExample.java:100)
        at demos.SobelFilterExample.runTest(SobelFilterExample.java:275)
        at demos.SobelFilterExample.main(SobelFilterExample.java:228)

I am stuck on what might be causing the compile error.  Thank you for sharing your insight.
Best regards,

Rick Lentz

---

View message @ http://jogamp.762907.n3.nabble.com/jocl-CLProgram-create-line-74-tp2250272p2250716.html
To start a new topic under jogamp, email [hidden email]
To unsubscribe from jogamp, click here.

-- 
http://michael-bien.com/

Michael,

  I have run the test provided:

OS: Mac OS X
ARCH: x86_64
VM: Java HotSpot(TM) 64-Bit Server VM
lib path: /Users/rick/Documents/workspace-sts-2.5.1.RELEASE/OpenCLIterative3DDeconvolution/lib:/Users/rick/Documents/workspace-sts-2.5.1.RELEASE/OpenCLIterative3DDeconvolution/lib/APPLE:/Users/rick/Documents/workspace-sts-2.5.1.RELEASE/OpenCLIterative3DDeconvolution/lib/APPLE
CLContext [id: 4301395024, platform: Apple, profile: FULL_PROFILE, devices: 2]
OpenCL 1.0 (Aug 22 2010 18:08:16)
please ignore [Ljava.lang.String;@604e280cPointerBuffer:AbstractBuffer[capacity 1, position 1, elementSize 8, ByteBuffer.capacity 8]
src:    __kernel
-> success

Here is the Junit output (failure trace):

com.jogamp.opencl.CLException$CLBuildProgramFailureException: on clBuildProgram [error: CL_BUILD_PROGRAM_FAILURE]
        at com.jogamp.opencl.CLException.newException(CLException.java:78)
        at demos.ProgramTest.checkError(ProgramTest.java:109)
        at demos.ProgramTest.buildProgramTest(ProgramTest.java:85)
        at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
        at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
        at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
        at java.lang.reflect.Method.invoke(Method.java:597)
        at org.junit.runners.model.FrameworkMethod$1.runReflectiveCall(FrameworkMethod.java:44)
        at org.junit.internal.runners.model.ReflectiveCallable.run(ReflectiveCallable.java:15)
        at org.junit.runners.model.FrameworkMethod.invokeExplosively(FrameworkMethod.java:41)
        at org.junit.internal.runners.statements.InvokeMethod.evaluate(InvokeMethod.java:20)
        at org.junit.runners.BlockJUnit4ClassRunner.runChild(BlockJUnit4ClassRunner.java:76)
        at org.junit.runners.BlockJUnit4ClassRunner.runChild(BlockJUnit4ClassRunner.java:50)
        at org.junit.runners.ParentRunner$3.run(ParentRunner.java:193)
        at org.junit.runners.ParentRunner$1.schedule(ParentRunner.java:52)
        at org.junit.runners.ParentRunner.runChildren(ParentRunner.java:191)
        at org.junit.runners.ParentRunner.access$000(ParentRunner.java:42)
        at org.junit.runners.ParentRunner$2.evaluate(ParentRunner.java:184)
        at org.junit.internal.runners.statements.RunBefores.evaluate(RunBefores.java:28)
        at org.junit.runners.ParentRunner.run(ParentRunner.java:236)
        at org.eclipse.jdt.internal.junit4.runner.JUnit4TestReference.run(JUnit4TestReference.java:49)
        at org.eclipse.jdt.internal.junit.runner.TestExecution.run(TestExecution.java:38)
        at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.runTests(RemoteTestRunner.java:467)
        at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.runTests(RemoteTestRunner.java:683)
        at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.run(RemoteTestRunner.java:390)
        at org.eclipse.jdt.internal.junit.runner.RemoteTestRunner.main(RemoteTestRunner.java:197)

I tested on another Mac just to ensure that it was not a single case regarding APPLE.  Here is the console output /results (Failed):

OS: Mac OS X
ARCH: x86_64
VM: Java HotSpot(TM) 64-Bit Server VM
lib path: /Users/hinerm/Documents/workspace/OpenCLIterative3DDeconvolution/lib:/Users/hinerm/Documents/workspace/OpenCLIterative3DDeconvolution/lib/APPLE
CLContext [id: 4296115824, platform: Apple, profile: FULL_PROFILE, devices: 2]
OpenCL 1.0 (Aug 22 2010 18:08:16)
please ignore [Ljava.lang.String;@29c56c60PointerBuffer:AbstractBuffer[capacity 1, position 1, elementSize 8, ByteBuffer.capacity 8]
src:    __kernel vo
-> success


**JUnit reports the same error as above***


A third test was run on Ubuntu 9.10 x64.  This test passed:


OS: Linux
ARCH: amd64
VM: Java HotSpot(TM) 64-Bit Server VM
lib path: /home/chrysalisbiosciencesadmin/workspace/OpenCLIterative3DDeconvolution/lib:/home/chrysalisbiosciencesadmin/workspace/OpenCLIterative3DDeconvolution/lib/LINUX64:/home/chrysalisbiosciencesadmin/workspace/OpenCLIterative3DDeconvolution/lib/LINUX64
CLContext [id: 1093557648, platform: NVIDIA CUDA, profile: FULL_PROFILE, devices: 3]
OpenCL 1.0 CUDA 3.2.1
please ignore [Ljava.lang.String;@509ced8ePointerBuffer:AbstractBuffer[capacity 1, position 1, elementSize 8, ByteBuffer.capacity 8]
src:    __kernel void sobel( __global float* input, __global float* output, int width,  int height ) {    
       int x = get_global_id(0);  
       int y = get_global_id(1);  
       int offset = y * width + x;  
                                                               
         float p0, p1, p2, p3, p5, p6, p7, p8 = 0;            
                                                               
                                                               
              if( x < 1 || y < 1 || x > width - 2 || y > height - 2 )  
              {  
                 output[offset] = 0;  
              }  
              else  
              {  
                  p0 = input[offset - width - 1] ;  
                  p1 = input[offset - width] ;  
                  p2 = input[offset - width + 1] ;  
                  p3 = input[offset - 1] ;  
                  p5 = input[offset + 1] ;  
                  p6 = input[offset + width - 1] ;  
                  p7 = input[offset + width] ;  
                  p8 = input[offset + width + 1] ;  
   
                  float sum1 = p0 + 2*p1 + p2 - p6 - 2*p7 - p8;  //GY  
                  float sum2 = p0 + 2*p3 + p6 - p2 - 2*p5 - p8;  //GX  
   
                  output[offset] = sqrt(  sum1*sum1 + sum2*sum2 );  
              }  
      }  
-> success

thank you Rick, i filed a bug report and set its priority to blocker: https://jogamp.org/bugzilla/show_bug.cgi?id=458
(see description)

-michael

On 01/14/2011 04:35 PM, Rick Lentz @ imagejdev.org [via jogamp] wrote:

I tested on another Mac just to ensure that it was not a single case regarding APPLE.  Here is the console output /results (Failed):

OS: Mac OS X
ARCH: x86_64
VM: Java HotSpot(TM) 64-Bit Server VM
lib path: /Users/hinerm/Documents/workspace/OpenCLIterative3DDeconvolution/lib:/Users/hinerm/Documents/workspace/OpenCLIterative3DDeconvolution/lib/APPLE
CLContext [id: 4296115824, platform: Apple, profile: FULL_PROFILE, devices: 2]
OpenCL 1.0 (Aug 22 2010 18:08:16)
please ignore [Ljava.lang.String;@29c56c60PointerBuffer:AbstractBuffer[capacity 1, position 1, elementSize 8, ByteBuffer.capacity 8]
src:    __kernel vo
-> success


**JUnit reports the same error as above***


A third test was run on Ubuntu 9.10 x64.  This test passed:


OS: Linux
ARCH: amd64
VM: Java HotSpot(TM) 64-Bit Server VM
lib path: /home/chrysalisbiosciencesadmin/workspace/OpenCLIterative3DDeconvolution/lib:/home/chrysalisbiosciencesadmin/workspace/OpenCLIterative3DDeconvolution/lib/LINUX64:/home/chrysalisbiosciencesadmin/workspace/OpenCLIterative3DDeconvolution/lib/LINUX64
CLContext [id: 1093557648, platform: NVIDIA CUDA, profile: FULL_PROFILE, devices: 3]
OpenCL 1.0 CUDA 3.2.1
please ignore [Ljava.lang.String;@509ced8ePointerBuffer:AbstractBuffer[capacity 1, position 1, elementSize 8, ByteBuffer.capacity 8]
src:    __kernel void sobel( __global float* input, __global float* output, int width,  int height ) {    
       int x = get_global_id(0);  
       int y = get_global_id(1);  
       int offset = y * width + x;  
                                                               
         float p0, p1, p2, p3, p5, p6, p7, p8 = 0;            
                                                               
                                                               
              if( x < 1 || y < 1 || x > width - 2 || y > height - 2 )  
              {  
                 output[offset] = 0;  
              }  
              else  
              {  
                  p0 = input[offset - width - 1] ;  
                  p1 = input[offset - width] ;  
                  p2 = input[offset - width + 1] ;  
                  p3 = input[offset - 1] ;  
                  p5 = input[offset + 1] ;  
                  p6 = input[offset + width - 1] ;  
                  p7 = input[offset + width] ;  
                  p8 = input[offset + width + 1] ;  
   
                  float sum1 = p0 + 2*p1 + p2 - p6 - 2*p7 - p8;  //GY  
                  float sum2 = p0 + 2*p3 + p6 - p2 - 2*p5 - p8;  //GX  
   
                  output[offset] = sqrt(  sum1*sum1 + sum2*sum2 );  
              }  
      }  
-> success

---

View message @ http://jogamp.762907.n3.nabble.com/jocl-CLProgram-create-line-74-tp2250272p2256102.html
To start a new topic under jogamp, email [hidden email]
To unsubscribe from jogamp, click here.

-- 
http://michael-bien.com/

there are a few options how we could fix this issue.

(the main problem is that we can't pass a String down to OpenCL, we have to convert it first into char**. That is fine as long we can delete the char** right after the function returns)

e.g the quickest fix would be to switch from String to (direct) ByteBuffer for the source parameter and document it that the user is responsible to hold a reference on that buffer (and don't modify it) until the program is compiled. This is not very nice from the public api perspective thats why i would like to prevent that somehow.

it could quickly lead to bugs if the user would do things like:
ByteBuffer src = readSource1();
program1 = createProgramFromSource..
src = readSource2();
program2 = createProgramFromSource..

compile program1
compile program2

The high level binding would be less problematic since its statefull... but this would lead to additional copy of the sources in e.g CLProgram as ByteBuffer if we want to keep the CLProgram.create(String src) api. (Note: the driver stores the sources too...)

have to think about that...
any feedback appreciated.

-michael

On 01/15/2011 03:17 AM, Rick Lentz wrote:

Hi Michael,

  The only thing I thought about adding was Mac OS version (10.6.5).  Thank you again for your help.  I am not sure what options are available to address this issue.  If I can help by posting in the Apple Developer forums - or in another way, please let me know.

Best regards,

Rick Lentz

On Fri, Jan 14, 2011 at 11:26 AM, Michael Bien <[hidden email]> wrote:

the mailing list doesn't seem to accept my mails anymore... mail is attached..
feel free to add any info to the bug report if you like.

-michael

-------- Original Message --------

Subject:	Re: jocl.CLProgram.create() line 74
Date:	Fri, 14 Jan 2011 16:55:32 +0100
From:	Michael Bien [hidden email]
To:	Rick Lentz @ imagejdev.org [via jogamp] [hidden email]

thank you Rick, i filed a bug report and set its priority to blocker: https://jogamp.org/bugzilla/show_bug.cgi?id=458
(see description)

-michael

On 01/14/2011 04:35 PM, Rick Lentz @ imagejdev.org [via jogamp] wrote:

I tested on another Mac just to ensure that it was not a single case regarding APPLE.  Here is the console output /results (Failed):

OS: Mac OS X
ARCH: x86_64
VM: Java HotSpot(TM) 64-Bit Server VM
lib path: /Users/hinerm/Documents/workspace/OpenCLIterative3DDeconvolution/lib:/Users/hinerm/Documents/workspace/OpenCLIterative3DDeconvolution/lib/APPLE
CLContext [id: 4296115824, platform: Apple, profile: FULL_PROFILE, devices: 2]
OpenCL 1.0 (Aug 22 2010 18:08:16)
please ignore [Ljava.lang.String;@29c56c60PointerBuffer:AbstractBuffer[capacity 1, position 1, elementSize 8, ByteBuffer.capacity 8]
src:    __kernel vo
-> success


**JUnit reports the same error as above***


A third test was run on Ubuntu 9.10 x64.  This test passed:


OS: Linux
ARCH: amd64
VM: Java HotSpot(TM) 64-Bit Server VM
lib path: /home/chrysalisbiosciencesadmin/workspace/OpenCLIterative3DDeconvolution/lib:/home/chrysalisbiosciencesadmin/workspace/OpenCLIterative3DDeconvolution/lib/LINUX64:/home/chrysalisbiosciencesadmin/workspace/OpenCLIterative3DDeconvolution/lib/LINUX64
CLContext [id: 1093557648, platform: NVIDIA CUDA, profile: FULL_PROFILE, devices: 3]
OpenCL 1.0 CUDA 3.2.1
please ignore [Ljava.lang.String;@509ced8ePointerBuffer:AbstractBuffer[capacity 1, position 1, elementSize 8, ByteBuffer.capacity 8]
src:    __kernel void sobel( __global float* input, __global float* output, int width,  int height ) {    
       int x = get_global_id(0);  
       int y = get_global_id(1);  
       int offset = y * width + x;  
                                                               
         float p0, p1, p2, p3, p5, p6, p7, p8 = 0;            
                                                               
                                                               
              if( x < 1 || y < 1 || x > width - 2 || y > height - 2 )  
              {  
                 output[offset] = 0;  
              }  
              else  
              {  
                  p0 = input[offset - width - 1] ;  
                  p1 = input[offset - width] ;  
                  p2 = input[offset - width + 1] ;  
                  p3 = input[offset - 1] ;  
                  p5 = input[offset + 1] ;  
                  p6 = input[offset + width - 1] ;  
                  p7 = input[offset + width] ;  
                  p8 = input[offset + width + 1] ;  
   
                  float sum1 = p0 + 2*p1 + p2 - p6 - 2*p7 - p8;  //GY  
                  float sum2 = p0 + 2*p3 + p6 - p2 - 2*p5 - p8;  //GX  
   
                  output[offset] = sqrt(  sum1*sum1 + sum2*sum2 );  
              }  
      }  
-> success

---

View message @ http://jogamp.762907.n3.nabble.com/jocl-CLProgram-create-line-74-tp2250272p2256102.html
To start a new topic under jogamp, email [hidden email]
To unsubscribe from jogamp, click here.

-- 
http://michael-bien.com/

-- 
http://michael-bien.com/

bug fixed in build 246:
http://jogamp.org/chuck/view/fwd/job/jocl/label=macosx-10_6-x86_64/246/

commit:
https://github.com/mbien/jocl/commit/6ca42ce154439b8c2f7f76ce4300ae08678f223a

bug:
https://jogamp.org/bugzilla/show_bug.cgi?id=458