8000 Compile bug: NVIDIA A800-SXM4-40GB ggml_cuda_init failed · Issue #13059 · ggml-org/llama.cpp · GitHub
[go: up one dir, main page]

Skip to content

Compile bug: NVIDIA A800-SXM4-40GB ggml_cuda_init failed #13059

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
lld1995 opened this issue Apr 22, 2025 · 1 comment
Open

Compile bug: NVIDIA A800-SXM4-40GB ggml_cuda_init failed #13059

lld1995 opened this issue Apr 22, 2025 · 1 comment

Comments

@lld1995
Copy link
lld1995 commented Apr 22, 2025

Git commit

now master

Operating systems

Linux

GGML backends

CUDA

Problem description & steps to reproduce

cmake .. -DGGML_CUDA=ON -DCMAKE_CUDA_ARCHITECTURES="80;86;87;89;90"

just build it and run

First Bad Commit

No response

Compile command

./llama-server --model /data/llama-model/DeepSeek-R1-GGUF-q4/DeepSeek-R1-UD-IQ1_M/DeepSeek-R1-UD-IQ1_M-00001-of-00004.gguf --ctx-size 26000 --n-gpu-layers 63 --host 0.0.0.0 --parallel 2 --cache-reuse 4096 --port 9990 --metrics --jinja -fa

Relevant log output

run error:
ggml_cuda_init: failed to initialize CUDA: system not yet initialized
warning: no usable GPU found, --gpu-layers option will be ignored
warning: one possible reason is that llama.cpp was compiled without GPU support
warning: consult docs/build.md for compilation instructions
@lld1995
Copy link
Author
lld1995 commented Apr 22, 2025

I guess it may driver error...

@github-actions github-actions bot added the stale label May 23, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant
0