# cuDNN nvcc fatal: failed to preprocess host compiler properties

**URL:** <https://forums.developer.nvidia.com/t/cudnn-nvcc-fatal-failed-to-preprocess-host-compiler-properties/144777>\
**Category:** CUDA Setup and Installation\
**Created:** [July 30, 2020, 3:21pm UTC](https://forums.developer.nvidia.com/t/cudnn-nvcc-fatal-failed-to-preprocess-host-compiler-properties/144777 "2020-07-30T15:21:25Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![wchan902](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@wchan902](https://forums.developer.nvidia.com/u/wchan902)\
**Post date:** [July 30, 2020, 3:21pm UTC](https://forums.developer.nvidia.com/t/cudnn-nvcc-fatal-failed-to-preprocess-host-compiler-properties/144777/1 "2020-07-30T15:21:25Z")

</div>

Hello,

I have a PC running a recent AMD CPU and RTX 2080 and Ubuntu 20.04.

I installed the Linux driver v11, CUDA 11 and cuDNN. All no issues. I ran nvcc and nvidia-smi. All good

But when I tried to verify the install of cuDNN by follow the “make clean && make”, I got the following error:

g++ no such file or directory  
nvcc fatal: failed to preprocess host compiler properties  
Freeimage is not set up correctly

Here is the exact error message:

FreeImage is not set up correctly. Please ensure FreeImage is set up correctly. \<\<\<  
[@] /usr/local/cuda/bin/nvcc -ccbin g++ -I/usr/local/cuda/include -I/usr/local/cuda/targets/ppc64le-linux/include -IFreeImage/include -m64 -gencode arch=compute\_35,code=sm\_35 -gencode arch=compute\_50,code=sm\_50 -gencode arch=compute\_53,code=sm\_53 -gencode arch=compute\_60,code=sm\_60 -gencode arch=compute\_61,code=sm\_61 -gencode arch=compute\_62,code=sm\_62 -gencode arch=compute\_70,code=sm\_70 -gencode arch=compute\_72,code=sm\_72 -gencode arch=compute\_75,code=sm\_75 -gencode arch=compute\_80,code=sm\_80 -gencode arch=compute\_80,code=compute\_80 -o fp16\_dev.o -c fp16\_dev.cu  
[@] g++ -I/usr/local/cuda/include -I/usr/local/cuda/targets/ppc64le-linux/include -IFreeImage/include -o fp16\_emu.o -c fp16\_emu.cpp  
[@] g++ -I/usr/local/cuda/include -I/usr/local/cuda/targets/ppc64le-linux/include -IFreeImage/include -o mnistCUDNN.o -c mnistCUDNN.cpp  
[@] /usr/local/cuda/bin/nvcc -ccbin g++ -m64 -gencode arch=compute\_35,code=sm\_35 -gencode arch=compute\_50,code=sm\_50 -gencode arch=compute\_53,code=sm\_53 -gencode arch=compute\_60,code=sm\_60 -gencode arch=compute\_61,code=sm\_61 -gencode arch=compute\_62,code=sm\_62 -gencode arch=compute\_70,code=sm\_70 -gencode arch=compute\_72,code=sm\_72 -gencode arch=compute\_75,code=sm\_75 -gencode arch=compute\_80,code=sm\_80 -gencode arch=compute\_80,code=compute\_80 -o mnistCUDNN fp16\_dev.o fp16\_emu.o mnistCUDNN.o -I/usr/local/cuda/include -I/usr/local/cuda/targets/ppc64le-linux/include -IFreeImage/include -L/usr/local/cuda/lib64 -L/usr/local/cuda/targets/ppc64le-linux/lib -lcublasLt -LFreeImage/lib/linux/x86\_64 -LFreeImage/lib/linux -lcudart -lcublas -lcudnn -lfreeimage -lstdc++ -lm

I noticed that the error message points to /usr/local/cuda/targets/ppc64le-linux/include but there is no ppc64le-linux folder. Instead the folder name is x86\_64-linux.

Is there something wrong with the mnistcudnn.cpp that I downloaded (come with the deb package) from nvidia.

Thanks