about llama-server, what does this mean?
>warning: failed to VirtualLock 8058626048-byte buffer (after previously locking 0 bytes): Invalid access to memory location.
Everything still works. Some guy says it's caused by a "corrupt model" but thing is I tried couple of different ones, they can't all be corrupted?
>https://github.com/ggml-org/llama.cpp/issues/5293