rudyCUDA_deviceInfo.cu

/**
  * @file rudyCUDA_deviceInfo.cu
	* @author Kurt Robert Rudolph
	* @description This file defines functions which 
	* gather info about a device
	*/
#include "rudyCUDA_deviceInfo.h"

rudyCUDA_state rudyCUDA_deviceInfo_gather( rudyCUDA_deviceInfo ** devices) {
	int i;
	HANDLE_ERROR( cudaMalloc( (rudyCUDA_deviceInfo**) devices, sizeof( rudyCUDA_deviceInfo)));
	HANDLE_ERROR( cudaGetDeviceCount( (int*) &((*devices)->deviceCount)));
	HANDLE_ERROR( cudaMalloc( (cudaDeviceProp**) &(*devices)->devicePropertiesArray, sizeof( cudaDeviceProp*) * (*devices)->deviceCount));
	for( i= 0; i< (*devices)->deviceCount; i++) {
		HANDLE_ERROR( cudaMalloc( (cudaDeviceProp***) &(*devices)->devicePropertiesArray[i], sizeof( cudaDeviceProp)));	
		HANDLE_ERROR( cudaGetDeviceProperties( (*devices)->devicePropertiesArray[i], i));
	}
	return rudyCUDA_state_success;
}

rudyCUDA_state rudyCUDA_deviceInfo_free( rudyCUDA_deviceInfo ** devices) {
	int i;	
	for( i= 0; i< (*devices)->deviceCount; i++) {
		cudaFree( (*devices)->devicePropertiesArray[i]);
		(*devices)->devicePropertiesArray[i] = NULL;
	}
	cudaFree( (*devices)->devicePropertiesArray);
	(*devices)->devicePropertiesArray = NULL;
	cudaFree( (*devices));
	(*devices) = NULL;
	return rudyCUDA_state_success;
}

rudyCUDA_state rudyCUDA_deviceInfo_print( rudyCUDA_deviceInfo ** devices) {
	int i;
	for ( i=0; i< (*devices)->deviceCount; i++) {
		printf( "   --- General Information for device %d ---\n", i);
		printf( "Name:  %s\n", (*devices)->devicePropertiesArray[i]->name);
		printf( "Compute capability:  %d.%d\n", (*devices)->devicePropertiesArray[i]->major, (*devices)->devicePropertiesArray[i]->minor);
		printf( "Clock rate:  %d\n", (*devices)->devicePropertiesArray[i]->clockRate);
		printf( "Device copy overlap:  ");
		if( (*devices)->devicePropertiesArray[i]->deviceOverlap)
			printf( "Enabled\n");
		else
			printf( "Disabled\n");
		printf( "Kernel execution timeout :  ");
		if( (*devices)->devicePropertiesArray[i]->kernelExecTimeoutEnabled)
			printf( "Enabled\n");
		else
			printf( "Disabled\n");
		printf( "   --- Memory Information for device %d ---\n", i);
		printf( "Total global mem:  %ld\n", (*devices)->devicePropertiesArray[i]->totalGlobalMem);
		printf( "Total constant Mem:  %ld\n", (*devices)->devicePropertiesArray[i]->totalConstMem);
		printf( "Max mem pitch:  %ld\n", (*devices)->devicePropertiesArray[i]->memPitch);
		printf( "Texture Alignment:  %ld\n", (*devices)->devicePropertiesArray[i]->textureAlignment);
		printf( "   --- MP Information for device %d ---\n", i);
		printf( "Multiprocessor count:  %d\n", (*devices)->devicePropertiesArray[i]->multiProcessorCount);
		printf( "Shared mem per mp:  %ld\n", (*devices)->devicePropertiesArray[i]->sharedMemPerBlock);
		printf( "Registers per mp:  %d\n", (*devices)->devicePropertiesArray[i]->regsPerBlock);
		printf( "Threads in warp:  %d\n", (*devices)->devicePropertiesArray[i]->warpSize);
		printf( "Max threads per block:  %d\n",	(*devices)->devicePropertiesArray[i]->maxThreadsPerBlock);
		printf( "Max thread dimensions:  (%d, %d, %d)\n",	(*devices)->devicePropertiesArray[i]->maxThreadsDim[0], (*devices)->devicePropertiesArray[i]->maxThreadsDim[1], (*devices)->devicePropertiesArray[i]->maxThreadsDim[2]);
		printf( "Max grid dimensions:  (%d, %d, %d)\n", (*devices)->devicePropertiesArray[i]->maxGridSize[0], (*devices)->devicePropertiesArray[i]->maxGridSize[1], (*devices)->devicePropertiesArray[i]->maxGridSize[2]);
		printf( "\n");
	}
  return rudyCUDA_state_success;
}