Misc. bug: AMX is not ready to be used!

[kandakji]

Name and Version

version: 5439 (33983057)
built with cc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu

Operating systems

Linux

Which llama.cpp modules do you know to be affected?

llama-server

Command line

Problem description & steps to reproduce

When running llama-server through the .devops/cpu.Dockerfile to deploy a model on an c7i.2xlarge instance (which supports AMX) on Amazon SageMaker, however, the logs say AMX is not ready to be used!

model used: qwen2.5-3b-instruct-q8_0.gguf

I've tried adding the flags for AMX but still getting the same log entry.

-DGGML_NATIVE=OFF
-DGGML_AVX512=ON
-DGGML_AVX512_BF16=ON
-DGGML_AVX512_VBMI=ON
-DGGML_AVX512_VNNI=ON
-DGGML_AMX_TILE=ON
-DGGML_AMX_INT8=ON
-DGGML_AMX_BF16=ON

First Bad Commit

No response

Relevant log output

AMX is not ready to be used!

[slaren]

This error indicates that AMX is not supported by the operating system, or the request to enable it is being denied for some reason.

[github-actions]

This issue was closed because it has been inactive for 14 days since being marked as stale.