Deploying on virtual machines?

[fidecastro]

Anybody knows if there are issues running this on Virtual Machines?

I'm testing this on a homelab with Proxmox running on relatively old equipment (dual Xeon E5-2690v4), and I can't get it to work. I'm running a Ubuntu 22.04-server virtual machine and have successfully run llama.cpp in it, but I keep getting "illegal instruction" errors when trying to use this repository.

[zhewang1-intc]

hi, you can use gdb to find out the crash position in this repo, pls refer to this issue #944.

once you get the crash position, e.g. intel-extension-for-transformers/intel_extension_for_transformers/llm/library/jblas/jblas/jit_blas_wrapper.h:152, you can set a breakpoint here via
b intel-extension-for-transformers/intel_extension_for_transformers/llm/library/jblas/jblas/jit_blas_wrapper.h:152.
after program stop at this breakpoint, you can use layout asm to switch to the assembly view and execute stepi until program throw a "illegal instruction" error.
If you get the instruction which lead program crash on vm, pls contact us and we will take a look.