icyleaf icyleaf

My laptop specs: MSI Bravo 15 B7E

CPU: AMD Ryzen™ 5 7535HS Processor with AMD XDNA™ architecture 6 cores, Max Boost Clock 4.55 GHz
GPU: AMD Radeon™ RX 6550M 4GB GDDR6
RAM: 32GB DDR5-4800

Clone ollama git clone --recursive https://github.com/ollama/ollama.git
go to directory cd ollama

Date of the guide : December 17, 2023

Introduction

In this post, I will provide the solution that worked on my system on how to install Radeon Open Compute (ROCm) on Arch (linux-6.6.7.arch1-1) for RX 6900 XT (Should work on other 6000 series). ROCm is an open-source software platform that allows GPU-accelerated computation. This tool is a prerequist to use GPU Acceleration on TensorFlow or PyTorch. In this guide I will use Paru as my AUR package helper, feel free to use any other (https://wiki.archlinux.org/title/AUR_helpers). I will assume you have a working operating system and know what you do with it (Otherwise Arch will be painfull for you).

How to setup VA-API within Proxmox LXC Unprivileged container

Proxmox configuration

No drivers need to be installed on the proxmox, from now called host.

Find GIDs of video and render group on host:

cat /etc/group | grep video

cat /etc/group | grep render

Clear Systemd Journal Logs in Ubuntu 20.04

Systemd has its own logging system called the journal, and the log files are stored in /var/log/journal.

Check current disk usage of journal files

sudo journalctl --disk-usage

Rotate journal files

Active journal files will be marked as archived, so that they are never written to in future.

	#######
	## These results are still open to the public. See
	## https://blog.ktz.me/the-best-media-server-cpu-in-the-world/
	## for analysis of them.

	# https://github.com/ironicbadger/quicksync_calc

	# zoidberg - dell 7040 sff pc
	CPU TEST FILE BITRATE TIME AVG_FPS AVG_SPEED AVG_WATTS
	i5-6600T h264_1080p_cpu ribblehead_1080p_h264 18952 kb/s 116.352s 29.88 1.04x N/A

	#!/usr/bin/env bash

	# current_shell_rcfile() {
	# if [ -n "$BASH_VERSION" ]; then
	# printf "%s\n" ~/.bashrc
	# elif [ -n "$ZSH_VERSION" ]; then
	# printf "%s\n" ~/.zshrc
	# elif [ -n "$FISH_VERSION" ]; then
	# printf "%s\n" ~/.config/fish/config.fish
	# elif [ -n "$XONSH_VERSION" ]; then

	# Note: This will only work on Navi21 GPUs (6800/6900+).
	# See: https://github.com/RadeonOpenCompute/ROCm/issues/1668#issuecomment-1043994570

	# Install Conda (latest from https://docs.conda.io/en/latest/miniconda.html#linux-installers)
	wget https://repo.anaconda.com/miniconda/Miniconda3-py39_4.12.0-Linux-x86_64.sh
	bash Miniconda3-py39_4.12.0-Linux-x86_64.sh
	# follow the prompts to install it, and run `conda` to make sure it's working.

	# Install git and curl, and clone the stable-diffusion repo
	sudo apt install -y git curl

	job "hello-world" {
	datacenters = ["dc1"]
	namespace = "default"
	type = "service"

	group "redis" {
	# Specify number of replicas of redis needed.
	count = 1

	# Specify networking for the group, port allocs.

	#!/bin/bash

	export LC_CTYPE=en_US.UTF-8
	export LC_ALL=en_US.UTF-8

	# Must run on Proxmox VE 7 server
	# Not sure how to handle a cluster - either run on each node or copy template after creating on one?
	# e.g. $ ssh [email protected] < proxmox-create-cloud-template.sh

	SRC_IMG="https://cloud-images.ubuntu.com/focal/current/focal-server-cloudimg-amd64-disk-kvm.img"