[waifu2x-converter-cpp] nvcc fatal: Value 'sm_30' is not defined for option 'gpu-architecture'

[wtquin]

Cannot install waifu2x-converter-cpp, gives the following error:

==> Starting build()...
-- The C compiler identification is GNU 13.1.1
-- The CXX compiler identification is GNU 13.1.1
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Check for working C compiler: /usr/bin/cc - skipped
-- Detecting C compile features
-- Detecting C compile features - done
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check for working CXX compiler: /usr/bin/c++ - skipped
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Found git, set version to: v5.3.3 (makepkg-b45498b)
-- System is: Linux (Linux)
-- Looking for CL_VERSION_3_0
-- Looking for CL_VERSION_3_0 - found
-- Found OpenCL: /usr/lib/libOpenCL.so (found version "3.0")
-- Found OpenCV: /usr (found version "4.7.0")
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD - Success
-- Found Threads: TRUE
-- Found CUDA: /opt/cuda (found version "12.1")
--
-- Config summary:
--      OpenCV: 4.7.0
--      OpenCL: 3.0
--      CUDA: 12.1
--      Unicode: TRUE
--      Installing models to: /usr/share/waifu2x-converter-cpp
--      Not building test binaries
--      Building for: Unix Makefiles-x86_64
--
-- Configuring done (0.3s)
-- Generating done (0.0s)
-- Build files have been written to: /home/quinten/.cache/yay/waifu2x-converter-cpp/src/waifu2x-converter-cpp
[  4%] Building C object CMakeFiles/conv.dir/conv.c.o
[  8%] Linking C executable conv
[  8%] Built target conv
[ 13%] Generating modelHandler_OpenCL.cl.h
[ 17%] Generating modelHandler_CUDA.ptx30
nvcc fatal   : Value 'sm_30' is not defined for option 'gpu-architecture'
make[2]: *** [CMakeFiles/gensrcs.dir/build.make:84: modelHandler_CUDA.ptx30] Error 1
make[1]: *** [CMakeFiles/Makefile2:141: CMakeFiles/gensrcs.dir/all] Error 2
make: *** [Makefile:146: all] Error 2
==> ERROR: A failure occurred in build().
Aborting...
-> error making: waifu2x-converter-cpp-exit status 4
-> Failed to install the following packages. Manual intervention is required:
waifu2x-converter-cpp - exit status 4

Seems like newer versions of nvcc do not support Kepler arch anymore (as per this issue from PyCuda)

Installed CUDA Version 12.1

$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2023 NVIDIA Corporation
Built on Mon_Apr__3_17:16:06_PDT_2023
Cuda compilation tools, release 12.1, V12.1.105
Build cuda_12.1.r12.1/compiler.32688072_0


$ nvidia-smi
Sun May  7 18:51:50 2023
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 530.41.03              Driver Version: 530.41.03    CUDA Version: 12.1     |
|-----------------------------------------+----------------------+----------------------+
| GPU  Name                  Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf            Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                                         |                      |               MIG M. |
|=========================================+======================+======================|
|   0  NVIDIA GeForce RTX 2070         Off| 00000000:01:00.0  On |                  N/A |
| 12%   55C    P0               37W / 185W|   1189MiB /  8192MiB |     28%      Default |
|                                         |                      |                  N/A |
+-----------------------------------------+----------------------+----------------------+