whoreson
whoreson
Makefile needs to be modified because 10's nvcc doesn't have the --forward-unknown-to-host-compiler option, nor the -arch=native [cuda10.patch](https://github.com/ggerganov/llama.cpp/files/13400052/cuda10.patch)
Nice, that patch does fix the compile issue. However, something else is up: ```cuBLAS error 8 at ggml-cuda.cu:8498 current device: 0 GGML_ASSERT: ggml-cuda.cu:8498: !"cuBLAS error" ``` But it actually dies...
Okay, it's broken since https://github.com/ggerganov/llama.cpp/commit/bcc0eb4591bec5ec02fad3f2bdcb1b265052ea56 Which is the "per-layer KV cache + quantum K cache" update.
Hmm, it does compile with CUDA 10.2 (but not with CUDA 10.1 which I previously used). I didn't even bother compiling a proper gcc, just disabled the version check in...
2bf8d0f7c4cc1235755ad06961ca761e458c5e55 broke it on CUDA 10.2 @slaren @JohannesGaessler
@slaren Very well, if reading feedback from testers is tiresome, I shall cease providing it.
No clue who came up with this "harassment" meme, I sent you an e-mail in december with a question about this (having no github acc then) and received no answer,...
Wow, this is so cool. Easy to convert existing models, quantization.. very nice. https://github.com/bes-dev/stable_diffusion.openvino
Wow, this is still not fixed/merged?..
Simply use httplib 0.19.0 (if you even must), and never upgrade again. The dev is too incompetent to maintain even Windows 7 compatibility (and has declared it "unsupported" instead), why...