Gerard Braad gbraad

Lima on runner

Have been testing a bit more with Lima and other cncf-projects. Got my own prepped cloud images now start with lima instead of the macadam tool I worked on before. Seems to work, but some minor issues along the way; issues filed and let's see. So far, glad to actually move to a more community maintained stack of tooling.

https://github.com/gbraad-dotfiles/upstream/commit/676bc9f03566972e5869c6c2b442c8361440be6b

cd ~/
git clone https://github.com/gbraad-dotfiles/upstream .dotfiles

Nerdctl on runner

ARCH=$(uname -m | sed 's/x86_64/amd64/;s/aarch64/arm64/')
NERDCTL_VERSION=$(curl -fsSL https://api.github.com/repos/containerd/nerdctl/releases/latest |
   grep '"tag_name"' | cut -d'"' -f4 | tr -d 'v')
curl -fsSL "https://github.com/containerd/nerdctl/releases/latest/download/nerdctl-${NERDCTL_VERSION}-linux-${ARCH}.tar.gz" \
   | sudo tar -C /usr/local/bin -xz nerdctl

Test openSUSE cloud image with machine

Needs macadam from: https://github.com/gbraad-dotfiles/macadam/releases/tag/v0.2.1a as this provides UEFI firmware support.

git config -f ~/.config/dotfiles/local/machine.ini \
    'disks.opensuse-cloud' \
    'https://download.opensuse.org/tumbleweed/appliances/openSUSE-Tumbleweed-Minimal-VM.x86_64-Cloud.qcow2'

Import devenv WSL image into container runtime

Note

This will describe in short how to get your old wsl environment running in a container runtime, such as containerd or Podman. While it is possible to use vhdx, I had better results with tarballs. A vhdx needs to be converted, use qemy-nbd etc, as you can not boot this as a VM unless you install the kernel and modules.

Prepare WSL export

  37, 39      42, 44, 46
36, 38, 40, 41, 43, 45, 47, 

  49, ...
48, ...

Here's how you configure it: First, configure your KORG DRUMLOGUE AS FOLLOWS:

Press SHIFT+LIVE[GLOBAL]
Press RS/7 ---- Set Channel to 1-12 (Note, you need Firmware upgrade 1.1) ---- Route: Pick your MIDI Connection (USB or USB+MIDI]

Here’s how to map your LaunchPad Pro MK3 to Drumlogue like in the video, using Novation Components:

Launch Components

Logue BackgroundFX

Store the runtime descriptor passed during initialization so we can access the get_raw_input() API later during rendering.

  // Runtime descriptor for background effect support
  // Provides access to get_raw_input() API which returns audio input unaffected by effect on/off state
  // This enables the effect to run continuously without requiring HOLD (XY Freeze) to be pressed
  static unit_runtime_desc_t s_runtime_desc;

Outline

Why

Not integrated, but Instant Development Environments.
- quick and dirty ?
- quick to set up and emphemeral
https://hub.docker.com/r/gbraad/c9ide

	--------- beginning of system
	02-06 23:31:02.838 11765 11855 I Regroovelizer-Native: === ELF LOADER START ===
	02-06 23:31:02.838 11765 11855 I Regroovelizer-Native: Loading: /data/user/0/com.regroovelizer/cache/selected_unit.nts3unit
	02-06 23:31:02.838 11765 11855 I Regroovelizer-Native: [ELF] Type=3 Machine=40 Entry=0x0
	02-06 23:31:02.838 11765 11855 I Regroovelizer-Native: [ELF] Memory range: 0x0 - 0x3d44 (15684 bytes)
	02-06 23:31:02.838 11765 11855 I Regroovelizer-Native: [ELF] Allocated 15684 bytes at 0xea937000
	02-06 23:31:02.838 11765 11855 I Regroovelizer-Native: [ELF] Loaded segment 1: vaddr=0x0 size=12796 flags=R-X
	02-06 23:31:02.838 11765 11855 I Regroovelizer-Native: [ELF] Loaded segment 2: vaddr=0x3200 size=816 flags=R--
	02-06 23:31:02.838 11765 11855 I Regroovelizer-Native: [ELF] Loaded segment 3: vaddr=0x3530 size=2068 flags=RW-
	02-06 23:31:02.838 11765 11855 I Regroovelizer-Native: Loaded at 0xea937000, size 15684

	// BasePreset - Base class for all Three.js presets
	class ThreeJSBasePreset {
	constructor(scene, camera, renderer, audioContext) {
	this.scene = scene;
	this.camera = camera;
	this.renderer = renderer;
	this.audioContext = audioContext;

	// Audio-reactive data (updated by ThreeJSRenderer)
	this.frequencyData = { bass: 0, mid: 0, high: 0 };