vectorAddDrv "CUDA_ERROR_INVALID_PTX"

Accelerated Computing CUDA CUDA Programming and Performance
1

I am running CentOs 7 and installed
sudo rpm -i cuda-repo-rhel7-10-1-local-10.1.168-418.67-1.0-1.x86_64.rpm
sudo yum clean all
sudo yum install cuda

Inside Terminal windows I Did The Following:
cd /home/jibarra/NVIDIA_CUDA-10.1_Samples/0_Simple/vectorAddDrv

(base) [jibarra@redsky vectorAddDrv]$ make
/usr/local/cuda/bin/nvcc -ccbin g++ -m64 -gencode arch=compute_30,code=compute_30 -o vectorAddDrv vectorAddDrv.o -L/usr/local/cuda/lib64/stubs -lcuda
mkdir -p …/…/bin/x86_64/linux/release
cp vectorAddDrv …/…/bin/x86_64/linux/release

(base) [jibarra@redsky vectorAddDrv]$ ./vectorAddDrv
Vector Addition (Driver API)
MapSMtoCores for SM 2.1 is undefined. Default to use 64 Cores/SM

Using CUDA Device [0]: Quadro 600
findModulePath found file at <./vectorAdd_kernel64.ptx>
initCUDA loading module: <./vectorAdd_kernel64.ptx>
checkCudaErrors() Driver API error = 0218 “CUDA_ERROR_INVALID_PTX” from file <vectorAddDrv.cpp>, line 114.

Please advise, thanks.
Jesse

2

You’re attempting to use CUDA 10.1 on a Fermi GPU (compute capability 2.1)

It is not supported. The last CUDA version that supported that GPU is CUDA 8.0

3

Thank you Robert.