Unable to run on Manjaro

[AllenDazi]

Hello, I followed the steps in README.md but it can't work. I tried reinstalling it (including rninstalling torch) multiple times. But if I run the lua file, it kept reporting this error:

THCudaCheck FAIL file=/home/dazian/wf2x/torch/extra/cutorch/lib/THC/generic/THCTensorMath.cu line=15 error=8 : invalid device function
/home/dazian/wf2x/torch/install/bin/luajit: ...dazian/wf2x/torch/install/share/lua/5.1/nn/Container.lua:67:
In 1 module of nn.Sequential:
In 1 module of nn.Sequential:
In 1 module of nn.Sequential:
/home/dazian/wf2x/torch/install/share/lua/5.1/nn/THNN.lua:110: cuda runtime error (8) : invalid device function at /home/dazian/wf2x/torch/extra/cutorch/lib/THC/generic/THCTensorMath.cu:15
stack traceback:
[C]: in function 'v'
/home/dazian/wf2x/torch/install/share/lua/5.1/nn/THNN.lua:110: in function 'SpatialConvolutionMM_updateOutput'
.../torch/install/share/lua/5.1/nn/SpatialConvolutionMM.lua:63: in function <.../torch/install/share/lua/5.1/nn/SpatialConvolutionMM.lua:53>
[C]: in function 'xpcall'
...dazian/wf2x/torch/install/share/lua/5.1/nn/Container.lua:63: in function 'rethrowErrors'
...azian/wf2x/torch/install/share/lua/5.1/nn/Sequential.lua:44: in function <...azian/wf2x/torch/install/share/lua/5.1/nn/Sequential.lua:41>
[C]: in function 'xpcall'
...dazian/wf2x/torch/install/share/lua/5.1/nn/Container.lua:63: in function 'rethrowErrors'
...azian/wf2x/torch/install/share/lua/5.1/nn/Sequential.lua:44: in function <...azian/wf2x/torch/install/share/lua/5.1/nn/Sequential.lua:41>
[C]: in function 'xpcall'
...dazian/wf2x/torch/install/share/lua/5.1/nn/Container.lua:63: in function 'rethrowErrors'
...azian/wf2x/torch/install/share/lua/5.1/nn/Sequential.lua:44: in function 'forward'
lib/reconstruct.lua:52: in function 'reconstruct_nn'
lib/reconstruct.lua:200: in function 'scale_rgb'
lib/reconstruct.lua:240: in function 'scale_f'
waifu2x.lua:99: in function 'convert_image'
waifu2x.lua:294: in function 'waifu2x'
waifu2x.lua:299: in main chunk
[C]: in function 'dofile'
...wf2x/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:150: in main chunk
[C]: at 0x004060c0

I tried force_cudnn argument, it doesn't work, and kept reporting errors above.
nvidia_smi report:

| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| 0 GeForce GTX 1660 Off | 00000000:01:00.0 On | N/A |
| 25% 34C P8 6W / 120W | 451MiB / 5941MiB | 4% Default |

[nagadomi]

Is that Torch7 my distro repository( https://github.com/nagadomi/distro ) built with CUDA10? #253 (comment)
NVIDIA Turing is a new architecture and is not supported by the official Torch7 and older CUDA versions.

[AllenDazi]

OK, I solved it and it seems to be working just fine. Thank you for your help.