K2 installation is not working (not up to date for the new cuda and pytorch versions)

[maxlvov]

Hi, just looked at the wheels (both CPU and CUDA), it seems that the latest supported cuda verison is 11.8, and pytorch version is 2.0.1, which are not up to date anymore. When i try to install from wheel, I get errors like:

ERROR: k2-1.24.3.dev20230718+cuda11.8.torch2.0.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl is not a supported wheel on this platform. ERROR: k2-1.24.3.dev20230725+cpu.torch2.0.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl is not a supported wheel on this platform.

For compilation from source, I also get errors:

-- Enabled languages: CXX;CUDA CMake Error at /usr/share/cmake-3.25/Modules/CMakeDetermineCompilerId.cmake:739 (message): Compiling the CUDA compiler identification source file "CMakeCUDACompilerId.cu" failed.

Compiler: /usr/bin/nvcc

Build flags:

Id flags: --keep;--keep-dir;tmp -v

The output was:

255

$ SPACE=

$ CUDART=cudart

$ HERE=/usr/lib/nvidia-cuda-toolkit/bin

$ THERE=/usr/lib/nvidia-cuda-toolkit/bin

$ _TARGETSIZE=

$ _TARGETDIR=

$ _TARGETSIZE=64

$ NVVMIR_LIBRARY_DIR=/usr/lib/nvidia-cuda-toolkit/libdevice

$

PATH=/usr/lib/nvidia-cuda-toolkit/bin:/usr/local/cuda-12.1/bin:/data2/asr_frameworks/bin:/home/ubuntu/anaconda/anaconda3/condabin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/kaldi/tools/openfst/bin:/kaldi/src/fstbin:/kaldi/tools/srilm/bin/i686-m64

$ LIBRARIES= -L/usr/lib/x86_64-linux-gnu/stubs

$ rm tmp/a_dlink.reg.c

$ gcc -std=c++14 -D__CUDA_ARCH__=300 -E -x c++

-DCUDA_DOUBLE_MATH_FUNCTIONS -DCUDACC -DNVCC -D"CUDACC_VER_BUILD=85" -D"CUDACC_VER_MINOR=1" -D"CUDACC_VER_MAJOR=9" -include "cuda_runtime.h" -m64 "CMakeCUDACompilerId.cu" > "tmp/CMakeCUDACompilerId.cpp1.ii"

$ cicc --c++14 --gnu_version=70500 --allow_managed -arch compute_30 -m64

-ftz=0 -prec_div=1 -prec_sqrt=1 -fmad=1 --include_file_name "CMakeCUDACompilerId.fatbin.c" -tused -nvvmir-library "/usr/lib/nvidia-cuda-toolkit/libdevice/libdevice.10.bc" --gen_module_id_file --module_id_file_name "tmp/CMakeCUDACompilerId.module_id" --orig_src_file_name "CMakeCUDACompilerId.cu" --gen_c_file_name "tmp/CMakeCUDACompilerId.cudafe1.c" --stub_file_name "tmp/CMakeCUDACompilerId.cudafe1.stub.c" --gen_device_file_name "tmp/CMakeCUDACompilerId.cudafe1.gpu" "tmp/CMakeCUDACompilerId.cpp1.ii" -o "tmp/CMakeCUDACompilerId.ptx"

$ ptxas -arch=sm_30 -m64 "tmp/CMakeCUDACompilerId.ptx" -o

"tmp/CMakeCUDACompilerId.sm_30.cubin"

ptxas fatal : Value 'sm_30' is not defined for option 'gpu-name'

--error 0xff --

Call Stack (most recent call first): /usr/share/cmake-3.25/Modules/CMakeDetermineCompilerId.cmake:6 (CMAKE_DETERMINE_COMPILER_ID_BUILD) /usr/share/cmake-3.25/Modules/CMakeDetermineCompilerId.cmake:48 (__determine_compiler_id_test) /usr/share/cmake-3.25/Modules/CMakeDetermineCUDACompiler.cmake:307 (CMAKE_DETERMINE_COMPILER_ID) CMakeLists.txt:50 (project)

-- Configuring incomplete, errors occurred!

[csukuangfj]

https://k2-fsa.github.io/k2/installation/from_wheels.html

please see the above page

[csukuangfj]

we support up to torch 2.1.2

[maxlvov]

For CPU based package, not CUDA Linux..

[csukuangfj]

You can find the following link in the above page

https://k2-fsa.github.io/k2/cpu.html

Not sure whether you have ever looked at it. As I said, we support up to torch 2.1.2, regardless of CPU or CUDA.