PGI User Forum

[alechand]

Hi Mat,
i installed the cuda 5.0 samples for ubuntu 11.10 64 bits from the website you recommended.

here is one output:

######################################

alechand@pcsantos2:~/NVIDIA_CUDA-5.0_Samples/0_Simple/matrixMul$ make
/usr/local/cuda-5.0/bin/nvcc -m64 -gencode arch=compute_10,code=sm_10 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_30,code=sm_30 -gencode arch=compute_35,code=sm_35 -I/usr/local/cuda-5.0/include -I. -I.. -I../../common/inc -o matrixMul.o -c matrixMul.cu
g++ -m64 -o matrixMul matrixMul.o -L/usr/local/cuda-5.0/lib64 -lcudart
mkdir -p ../../bin/linux/release
cp matrixMul ../../bin/linux/release

#######################################


and here is other:


#####################################

alechand@pcsantos2:~/NVIDIA_CUDA-5.0_Samples/6_Advanced/segmentationTreeThrust$ make
/usr/local/cuda-5.0/bin/nvcc -m64 -gencode arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20 -gencode arch=compute_30,code=sm_30 -gencode arch=compute_35,code=sm_35 -I/usr/local/cuda-5.0/include -I. -I.. -I../../common/inc -o segmentationTree.o -c segmentationTree.cu
g++ -m64 -o segmentationTreeThrust segmentationTree.o -L/usr/local/cuda-5.0/lib64 -lcudart
mkdir -p ../../bin/linux/release
cp segmentationTreeThrust ../../bin/linux/release

###########################################


Considering my driver is installed,
I have problems with my graphic card right?

thanks

[mkcolg]

These are just the build logs. Were you able to run the resulting binaries?

[alechand]

Sorry i forget it... here it is:

#########################3

alechand@pcsantos2:~/NVIDIA_CUDA-5.0_Samples/bin/linux/release$ ./matrixMul
[Matrix Multiply Using CUDA] - Starting...
GPU Device 0: "GeForce GTX 680" with compute capability 3.0

MatrixA(320,320), MatrixB(640,320)
Computing result using CUDA Kernel...
done
Failed to create start event (error code unspecified launch failure)!
alechand@pcsantos2:~/NVIDIA_CUDA-5.0_Samples/bin/linux/release$ ./segmentationTreeThrust
./segmentationTreeThrust Starting...

GPU Device 0: "GeForce GTX 680" with compute capability 3.0

* Building segmentation tree... Algorithm failed (unspecified launch failure)
alechand@pcsantos2:~/NVIDIA_CUDA-5.0_Samples/bin/linux/release$ ./clock
CUDA Clock sample
GPU Device 0: "GeForce GTX 680" with compute capability 3.0

CUDA error at clock.cu:127 code=4(cudaErrorLaunchFailure) "cudaMemcpy(timer, dtimer, sizeof(clock_t) * NUM_BLOCKS * 2, cudaMemcpyDeviceToHost)"
CUDA error at clock.cu:129 code=4(cudaErrorLaunchFailure) "cudaFree(dinput)"
CUDA error at clock.cu:130 code=4(cudaErrorLaunchFailure) "cudaFree(doutput)"
CUDA error at clock.cu:131 code=4(cudaErrorLaunchFailure) "cudaFree(dtimer)"
Total clocks = -1536394840
alechand@pcsantos2:~/NVIDIA_CUDA-5.0_Samples/bin/linux/release$
alechand@pcsantos2:~/NVIDIA_CUDA-5.0_Samples/bin/linux/release$ ./inlinePTX
CUDA inline PTX assembler sample
GPU Device 0: "GeForce GTX 680" with compute capability 3.0

CUDA error at inlinePTX.cu:70 code=4(cudaErrorLaunchFailure) "cudaDeviceSynchronize()"
CUDA error at inlinePTX.cu:75 code=4(cudaErrorLaunchFailure) "cudaMallocHost(&h_d_ptr, N *sizeof(int))"
CUDA error at inlinePTX.cu:76 code=4(cudaErrorLaunchFailure) "cudaMemcpy(h_d_ptr, d_ptr, N *sizeof(int), cudaMemcpyDeviceToHost)"
Segmentation fault
alechand@pcsantos2:~/NVIDIA_CUDA-5.0_Samples/bin/linux/release$

#################################3

I compiled two more programs clock and inlinePTX (above).

I am also using the forum of Nvidia, just to inform you what is happening...
thanks

[mkcolg]