llama.cpp

f77261a7 - ggml: bypass code incompatible with CUDA < 11.1 (whisper/2020)

ggml: bypass code incompatible with CUDA < 11.1 (whisper/2020) `cudaHostRegisterReadOnly` parameter was only introduced in CUDA 11.1 See this issue for more details: https://github.com/ggerganov/examples/whisper/whisper.cpp/issues/2007