The latest binary for linux, cuda 6.5, did not work on my machine (mint linux, driver 390.116, dual 1080Tis) .
I decided to download and extracted the source file ( mfaktc-0.21.tar.gz) and installed the cuda toolkit from the mint software manager.
I had to edit the Makefile in the src directory to change the path to the cuda installation.
Code:
# where is the CUDA Toolkit installed?
CUDA_DIR = /usr/lib/cuda
Since the 1080Ti is compute capability 6.1 and the cuda toolkit I installed is incompatible with compute 1.0x and 2.x, I changed these lines too.
Code:
# generate code for various compute capabilities
#NVCCFLAGS += --generate-code arch=compute_11,code=sm_11 # CC 1.1, 1.2 and 1.3 GPUs will use this code (1.0 is not possible for mfaktc)
#NVCCFLAGS += --generate-code arch=compute_20,code=sm_20 # CC 2.x GPUs will use this code, one code fits all!
NVCCFLAGS += --generate-code arch=compute_30,code=sm_30 # all CC 3.x GPUs _COULD_ use this code
NVCCFLAGS += --generate-code arch=compute_35,code=sm_35 # but CC 3.5 (3.2?) _CAN_ use funnel shift which is useful for mfaktc
NVCCFLAGS += --generate-code arch=compute_50,code=sm_50 # CC 5.x GPUs will use this code
NVCCFLAGS += --generate-code arch=compute_60,code=sm_60 # CC 6.x GPUs will use this code