Sorry, cudaMallocManaged() is not implemented in the current version.

data61 / cuda-fixnum

Extended-precision modular arithmetic library that targets CUDA.

Other

41 stars 28 forks source link

Sorry, cudaMallocManaged() is not implemented in the current version. #72

Open guiming-shi opened 3 years ago

guiming-shi commented 3 years ago

~/cuda-fixnum# ./bench/bench Function: mul_lo, #elts: 1e3 fixnum digit total data time Kops/s bits bits (MiB) (seconds) 2021-08-01 09:49:54 [ERROR] Sorry, cudaMallocManaged() is not implemented in the current version. If user application keeps reporting errors, please use Ctrl+C to exit Fatal CUDA error at ./src/array/fixnum_array.cu:52 : unified memory allocation (default attach) : cudaErrorNotSupported VirtaiTech Resource. Build-20201209_195723-4f0a3fd

How can i solve it.

Jason9704 commented 1 year ago

it is problem of bitsandbytes version, please try to change optimizer