Problem with HANDLE_ERROR?

Accelerated Computing CUDA CUDA Programming and Performance
1

I’m trying out examples in the book “CUDA by exmaple”. The first two printf examples compiled and worked, but now i came up with a problem with this code which is as given.

It seems to be a problem with HANDLE_ERROR, I’m not used to Visual Studio (i use code blocks) and trying NVIDIA Parallel Nsight 2.1 with it, it is the only way I found to easily program CUDA.

Here is the debug:

1>------ Build started: Project: Hello World, Configuration: Debug Win32 ------

1>Compiling with CUDA Build Rule...

1>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v4.1\bin\nvcc.exe"  -G0  -gencode=arch=compute_10,code=\"sm_10,compute_10\" -gencode=arch=compute_20,code=\"sm_20,compute_20\"  --machine 32 -ccbin "c:\Program Files (x86)\Microsoft Visual Studio 9.0\VC\bin"    -Xcompiler "/EHsc /W3 /nologo /O2 /Zi   /MT  "  -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v4.1\include" -maxrregcount=0  --compile -o "Debug/kernel.cu.obj" kernel.cu 

1>kernel.cu

1>c:/Users/Jonathan/Documents/Visual Studio 2008/Projects/Hello World/Hello World/kernel.cu(17): error: identifier "HANDLE_ERROR" is undefined

1>1 error detected in the compilation of "C:/Users/Jonathan/AppData/Local/Temp/tmpxft_00001d1c_00000000-8_kernel.compute_20.cpp1.ii".

1>Project : error PRJ0019: A tool returned an error code from "Compiling with CUDA Build Rule..."

1>Build log was saved at "file://c:\Users\Jonathan\Documents\Visual Studio 2008\Projects\Hello World\Hello World\Debug\BuildLog.htm"

1>Hello World - 2 error(s), 0 warning(s)

========== Build: 0 succeeded, 1 failed, 0 up-to-date, 0 skipped ==========

Here is the code:

#include "cuda_runtime.h"

#include "device_launch_parameters.h"

#include <stdio.h>

#include <iostream>

__global__ void add( int a, int b, int *c )

{

	*c = a + b;

}

int main( void )

{

	int c;

	int *dev_c;

	HANDLE_ERROR( cudaMalloc( (void**)&dev_c, sizeof(int) ) );

	add<<<1,1>>>( 2, 7, dev_c );

	HANDLE_ERROR( cudaMemcpy( &c,

							  dev_c,

	                          sizeof(int),

	                          cudaMemcpyDeviceToHost ) );

	printf( "2 + 7 = %d\n", c );

	cudaFree( dev_c );

	return 0;

}

Thanks a bunch!

2

Well I just needed to get rid of that “HANDLE_ERROR” nonsense. Must of been in the books cd of headers…

working version

#include "cuda_runtime.h"

#include "device_launch_parameters.h"

#include <stdio.h>

#include <iostream>

__global__ void add( int a, int b, int *c )

{

	*c = a + b;

}

int main( void )

{

	int c;

	int *dev_c;

	cudaMalloc( (void**)&dev_c, sizeof(int) );

	add<<<1,1>>>( 2, 7, dev_c );

	cudaMemcpy( &c,dev_c,sizeof(int),cudaMemcpyDeviceToHost );

	printf( "2 + 7 = %d\n", c );

	cudaFree( dev_c );

	getchar();

	return 0;

}
3

I’m also having the same problem :(
Have you solved this problem?

G.

4

Here is a link to all the examples in the book. http://developer.nvidia.com/sites/default/files/akamai/cuda/files/cuda_by_example.zip

You don’t have to type anything and the header files are already in there. Just compile with nvcc and run.

5

I have bought this book from google play. is there any CD with book? the book.h is not defined as header. So, I have to type all program manually for learning. any idea?

6

google “cuda by example”

take the first hit (nvidia developer)

read on that page until you find the link to download the source code

7

I am facing the same problem with the same book.

8

Problem Solved! You need to put this patch into your code directly after the header files. Then it will work. I is working for me now.

#define HANDLE_ERROR( err ) ( HandleError( err, FILE, LINE ) )
static void HandleError(cudaError_t err, const char *file, int line)
{if (err != cudaSuccess)
{printf(“%s in %s at line %d\n”, cudaGetErrorString(err),file, line);
exit(EXIT_FAILURE);
}
}