static void Main(string[] args)
{
int[] r1 = new int[]
{ 8, 2, 3, 4 };
int[] r2 = new int[]
{ 4, 3, 2, 5 };
int[] r3 = new int[4];
int rowSize = r1.Length;
// pick first platform
ComputePlatform platform = ComputePlatform.Platforms[0];
// create context with all gpu devices
ComputeContext context = new ComputeContext(ComputeDeviceTypes.Gpu,
new ComputeContextPropertyList(platform), null, IntPtr.Zero);
// create a command queue with first gpu found
ComputeCommandQueue queue = new ComputeCommandQueue(context,
context.Devices[0], ComputeCommandQueueFlags.None);
// load opencl source and
// create program with opencl source
ComputeProgram program = new ComputeProgram(context, CalculateKernel);
// compile opencl source
program.Build(null, null, null, IntPtr.Zero);
// load chosen kernel from program
ComputeKernel kernel = program.CreateKernel("Calc");
// allocate a memory buffer with the message (the int array)
ComputeBuffer <int> row1Buffer = new ComputeBuffer <int>(context,
ComputeMemoryFlags.ReadOnly | ComputeMemoryFlags.UseHostPointer, r1);
// allocate a memory buffer with the message (the int array)
ComputeBuffer <int> row2Buffer = new ComputeBuffer <int>(context,
ComputeMemoryFlags.ReadOnly | ComputeMemoryFlags.UseHostPointer, r2);
// allocate a memory buffer with the message (the int array)
ComputeBuffer <int> resultBuffer = new ComputeBuffer <int>(context,
ComputeMemoryFlags.ReadOnly | ComputeMemoryFlags.UseHostPointer, new int[4]);
kernel.SetMemoryArgument(0, row1Buffer); // set the integer array
kernel.SetMemoryArgument(1, row2Buffer); // set the integer array
kernel.SetValueArgument(2, rowSize); // set the array size
kernel.SetMemoryArgument(3, resultBuffer); // set the integer array
// execute kernel
queue.ExecuteTask(kernel, null);
// wait for completion
queue.Finish();
GCHandle arrCHandle = GCHandle.Alloc(r3, GCHandleType.Pinned);
queue.Read <int>(resultBuffer, true, 0, r3.Length, arrCHandle.AddrOfPinnedObject(), null);
Console.WriteLine("display result from gpu buffer:");
for (int i = 0; i < r3.Length; i++)
{
Console.WriteLine(r3[i]);
}
arrCHandle.Free();
row1Buffer.Dispose();
row2Buffer.Dispose();
kernel.Dispose();
program.Dispose();
queue.Dispose();
context.Dispose();
Console.WriteLine("Finished");
Console.ReadKey();
}
public static Tuple <List <List <int> >, TimeSpan> MultiplyParallel(List <List <int> > matrixOne, List <List <int> > matrixTwo)
{
if (!isRegularMatrix(matrixOne) || !isRegularMatrix(matrixTwo))
{
throw new ArgumentException("Non regular matrix detected. Rows size mismatch detected.");
}
if (matrixOne[0].Count != matrixTwo.Count)
{
throw new ArgumentException("Matrixes is not compatible. Columns count of first matrix is not equal to rows count of second matrix.");
}
List <List <int> > result = new List <List <int> >();
ComputePlatform platform = GetGPU();
if (platform is null)
{
throw new PlatformNotSupportedException("Platform doesn't have a dedicated GPU. Run is impossible.");
}
ComputeContext context = new ComputeContext(ComputeDeviceTypes.Gpu, new ComputeContextPropertyList(platform), null, IntPtr.Zero);
ComputeCommandQueue queue = new ComputeCommandQueue(context, context.Devices[0], ComputeCommandQueueFlags.None);
ComputeProgram program = new ComputeProgram(context, CalculateKernel);
program.Build(null, null, null, IntPtr.Zero);
ComputeKernel kernel = program.CreateKernel("Multiply");
List <ComputeBuffer <int> > rowsMatrixOne = matrixOne.TransformMatrixToComputerBuffersOfRows(context);
List <ComputeBuffer <int> > columnsMatrixTwo = matrixTwo.TransformMatrixToComputerBuffersOfColumns(context);
List <ComputeBuffer <int> > resultRowsMatrix = TwoDToOneDResult(matrixOne.Count, matrixTwo[0].Count, context);
Stopwatch stopwatch = new Stopwatch();
stopwatch.Start();
for (int i = 0; i < resultRowsMatrix.Count; ++i)
{
for (int j = 0; j < resultRowsMatrix[i].Count; ++j)
{
kernel.SetMemoryArgument(0, rowsMatrixOne[i]);
kernel.SetMemoryArgument(1, columnsMatrixTwo[j]);
kernel.SetMemoryArgument(2, resultRowsMatrix[i]);
kernel.SetValueArgument(3, matrixTwo.Count);
kernel.SetValueArgument(4, j);
queue.ExecuteTask(kernel, null);
}
}
queue.Finish();
stopwatch.Stop();
for (int i = 0; i < resultRowsMatrix.Count; ++i)
{
int[] res = new int[resultRowsMatrix[i].Count];
GCHandle gCHandle = GCHandle.Alloc(res, GCHandleType.Pinned);
queue.Read <int>(resultRowsMatrix[i], true, 0, res.Length, gCHandle.AddrOfPinnedObject(), null);
result.Add(new List <int>(res));
}
return(new Tuple <List <List <int> >, TimeSpan>(result, stopwatch.Elapsed));
}
static void Main(string[] args)
{
int[] r1 = new int[]
{1, 2, 3, 4};
int[] r2 = new int[]
{4, 3, 2, 1};
int rowSize = r1.Length;
// pick first platform
ComputePlatform platform = ComputePlatform.Platforms[0];
// create context with all gpu devices
ComputeContext context = new ComputeContext(ComputeDeviceTypes.Gpu,
new ComputeContextPropertyList(platform), null, IntPtr.Zero);
// create a command queue with first gpu found
ComputeCommandQueue queue = new ComputeCommandQueue(context,
context.Devices[0], ComputeCommandQueueFlags.None);
// load opencl source and
// create program with opencl source
ComputeProgram program = new ComputeProgram(context, CalculateKernel);
// compile opencl source
program.Build(null, null, null, IntPtr.Zero);
// load chosen kernel from program
ComputeKernel kernel = program.CreateKernel("Calc");
// allocate a memory buffer with the message (the int array)
ComputeBuffer<int> row1Buffer = new ComputeBuffer<int>(context,
ComputeMemoryFlags.ReadOnly | ComputeMemoryFlags.UseHostPointer, r1);
// allocate a memory buffer with the message (the int array)
ComputeBuffer<int> row2Buffer = new ComputeBuffer<int>(context,
ComputeMemoryFlags.ReadOnly | ComputeMemoryFlags.UseHostPointer, r2);
kernel.SetMemoryArgument(0, row1Buffer); // set the integer array
kernel.SetMemoryArgument(1, row2Buffer); // set the integer array
kernel.SetValueArgument(2, rowSize); // set the array size
// execute kernel
queue.ExecuteTask(kernel, null);
// wait for completion
queue.Finish();
Console.WriteLine("Finished");
Console.ReadKey();
}
// 4719 ms 4096x4096@512 iter with 1 core?
static long Method03(float ymin, float xmin, float width, int[] message)
{
// pick first platform
ComputePlatform platform = ComputePlatform.Platforms[0];
// create context with all gpu devices
ComputeContext context = new ComputeContext(ComputeDeviceTypes.Gpu,
new ComputeContextPropertyList(platform), null, IntPtr.Zero);
// create a command queue with first gpu found
ComputeCommandQueue queue = new ComputeCommandQueue(context,
context.Devices[0], ComputeCommandQueueFlags.None);
// load opencl source
StreamReader streamReader = new StreamReader("Mandel.cl");
string clSource = streamReader.ReadToEnd();
streamReader.Close();
// create program with opencl source
ComputeProgram program = new ComputeProgram(context, clSource);
// compile opencl source
program.Build(null, null, null, IntPtr.Zero);
// load chosen kernel from program
ComputeKernel kernel = program.CreateKernel("mandel");
int messageSize = message.Length;
// allocate a memory buffer with the message
ComputeBuffer <int> messageBuffer = new ComputeBuffer <int>(context,
ComputeMemoryFlags.WriteOnly | ComputeMemoryFlags.UseHostPointer, message);
kernel.SetMemoryArgument(0, messageBuffer);
kernel.SetValueArgument(1, N);
kernel.SetValueArgument(2, ymin);
kernel.SetValueArgument(3, xmin);
kernel.SetValueArgument(4, width);
kernel.SetValueArgument(5, MaxIter);
var watch = System.Diagnostics.Stopwatch.StartNew();
// Execute kernel
queue.ExecuteTask(kernel, null);
// Read data back
unsafe
{
fixed(int *retPtr = message)
{
queue.Read(messageBuffer,
false, 0,
messageSize,
new IntPtr(retPtr),
null);
queue.Finish();
}
}
watch.Stop();
return(watch.ElapsedMilliseconds);
}
public void ProccessOld()
{
// pick first platform
ComputePlatform platform = ComputePlatform.Platforms[0];
// create context with all gpu devices
ComputeContext context = new ComputeContext(ComputeDeviceTypes.Gpu, new ComputeContextPropertyList(platform), null, IntPtr.Zero);
// create a command queue with first gpu found
ComputeCommandQueue queue = new ComputeCommandQueue(context, context.Devices[0], ComputeCommandQueueFlags.None);
// load opencl source
StreamReader streamReader = new StreamReader("1.cl");
string clSource = streamReader.ReadToEnd();
streamReader.Close();
// create program with opencl source
ComputeProgram program = new ComputeProgram(context, clSource);
// compile opencl source
program.Build(null, null, null, IntPtr.Zero);
// load chosen kernel from program
ComputeKernel kernel = program.CreateKernel("sum");
// create a ten integer array and its length
float[] a = new float[] { 1.1f, 2.3f };
float[] b = new float[] { 3.6f, 4.9f };
// allocate a memory buffer with the message (the int array)
ComputeBuffer <float> aBuffer = new ComputeBuffer <float>(context, ComputeMemoryFlags.ReadWrite | ComputeMemoryFlags.UseHostPointer, a);
ComputeBuffer <float> bBuffer = new ComputeBuffer <float>(context, ComputeMemoryFlags.ReadOnly | ComputeMemoryFlags.UseHostPointer, b);
kernel.SetMemoryArgument(0, aBuffer);
kernel.SetMemoryArgument(1, bBuffer);
// execute kernel
queue.ExecuteTask(kernel, null);
// wait for completion
queue.Finish();
queue.ReadFromBuffer <float>(aBuffer, ref a, true, null);
for (int i = 0; i < a.Length; i++)
{
Console.WriteLine(a[i]);
}
/*
* // create a ten integer array and its length
* int[] message = new int[] { 1, 2, 3, 4, 5 };
* int messageSize = message.Length;
*
* // allocate a memory buffer with the message (the int array)
* ComputeBuffer<int> messageBuffer = new ComputeBuffer<int>(context,
* ComputeMemoryFlags.ReadOnly | ComputeMemoryFlags.UseHostPointer, message);
*
* kernel.SetMemoryArgument(0, messageBuffer); // set the integer array
* kernel.SetValueArgument(1, messageSize); // set the array size
*
* // execute kernel
* queue.ExecuteTask(kernel, null);
*
* // wait for completion
* queue.Finish();*/
}
