Vai al contenuto

Patches and Overlays

Important: Read this before you start!

This section deals with adding patches to your KNULLI installation manually. This is not a guide for regular updates. Please be absolutely aware that this guide is just a kick-off guide for advanced users who want to learn more about Linux and apply manual changes to their KNULLI installation.

If you are looking for information on how to update your KNULLI device regularly, please head to the Update section instead!

SSH network access to your device is mandatory

Before reading any further, be aware that it is mandatory to be able to access a command line on your KNULLI device via SSH to install patches and create overlays. Consequently, you will not be able to do either unless your device is connected to your Wi-Fi (as explained in the Networking section) and you have prepared SSH access to your device (as explained in the SSH section).

This section covers details about installing patches and persisting changes to the operating system in an overlay. As a regular KNULLI user, you probably do not need to know any of this. This entire section is written for two groups of users:

  • Community members from our Discord who want to try patches before they are included in a new internal or public build.
  • Advanced users who know their way around Linux and want to make changes to their KNULLI setup.

Understanding Linux file systems

If you are a Windows user who has never worked with Linux before, you might want to take a little time to understand how Linux file systems work. Bear in mind that this is a simplified explanation which is only meant to pick you up and give you a brief understanding of how KNULLI works differently from your Windows PC. If you want to learn more about how Linux works, we strongly recommend finding a more extensive source of information.

File systems on Windows machines

On your Windows computer, each drive (or partition) has an assigned drive letter. You might think about them like labeled drawers of a filing cabinet. The main drive that hosts your Windows installation traditionally uses the drive letter C: - this is a relic of a time when drive letters A: and B: were reserved for the two floppy disk drives computers used to have. However, if you connect another drive to your PC, like a thumb drive or a SD card, Windows automatically assigns new drive letters to them, so you can access those drives from your Windows Explorer. Files and folders are addressed by their absolute path which starts with the drive letter. E.g., the folder which holds your pictures is located at

C:\Users\<username>\Pictures

File systems on Linux machines

Linux filesystems work a little differently. On Linux, every drive is treated like a folder. A Linux system always has a so called root that loosely corresponds conceptually to the drive letter C: on a Windows machine. The root is always referred to by the symbol /. By definition, each absolute path always starts with the / symbol.

Other drives are treated like subfolders of the root. With a command called mount, any empty folder of a Linux system can be assigned to any drive of the device. It is even allowed to have more than one folder that points to the same drive. Following the previous metaphor, consider a filing cabinet with only a single drawer.

For example, on your KNULLI device, your games are stored in

/userdata/roms

The KNULLI file system

KNULLI is a Linux, so it uses a Linux file system. The entire KNULLI system is stored in a read-only SquashFS. (You can learn more about SquashFS on Wikipedia.) The only exception is a writable folder called /userdata.

During boot, KNULLI uses the mount command to make the folder /userdata point to a writable drive. By switching between internal and external storage (as explained in the Second SD Card section), you tell KNULLI which drive to mount as /userdata during boot:

  • If you use a single SD card setup (internal storage), KNULLI mounts the SHARE partition of SD 1.
  • If you use a dual SD card setup (external storage), KNULLI mounts the main partition of SD 2.

When accessing your KNULLI device via Windows/SMB network transfer (as explained in the Network Transfer section), you will access a network drive called share that corresponds exactly to your current /userdata folder. The SMB address

\\KNULLI\share

points to the exact same place as the absolute path

/userdata

on your KNULLI device.

The Game Storage section covers in great detail how to use the writable part of your device to add games, BIOS files, bezel decorations, etc. However, when you want to install patches, you want to modify files outside of the writable /userdata folder - that is, where the overlay comes into play.

Many ways lead to Rome

Keep in mind that Linux allows multiple paths to the same files and folders:

/media/SHARE

points to your internal storage while

/media/SHARE_1

points to your external storage (if you have a dual SD card setup).

Consequently, on a single SD card setup, the paths

/media/SHARE/system/batocera.conf

and

/userdata/system/batocera.conf

point to the exact same file.

Following that principle, on a dual SD card setup, the paths

/media/SHARE_1/system/batocera.conf

and

/userdata/system/batocera.conf

point to the exact same file, too.

The overlay

Since everything outside of the /userdata folder of your KNULLI installation is read-only by definition, you will need to create an overlay to make permanent changes to your KNULLI installation. Otherwise, all changes will simply be undone after every reboot of your system.

Your primary SD card (SD1) has a drive called BATOCERA, which is formatted to FAT32 and can be accessed easily from Windows, Linux, and MacOS. On that drive, you will find a folder called boot which contains a file called batocera - that file contains the majority of your KNULLI installation. (This is why you can manually update your KNULLI installation by simply replacing this file, as explained in the Update section.)

The overlay is another file that is just called overlay and resides in the same folder. The overlay is optional - usually, users do not have an overlay in place. The overlay file contains changes you made to the KNULLI installation, which will be applied during boot.

Important: Undo changes

If you have created an overlay that breaks your system, you can always undo all your changes by simply deleting the overlay file from the boot folder of your batocera partition.

We strongly recommend to always remove your current overlay when you update your KNULLI installation, because patches and other modifications stored in the overlay might be in conflict with the new KNULLI version.

Installing patches (or manual modifications)

Patches

Some community members on our Discord server occasionally provide patches you can install manually. These patches are usually new features which are either considered or already confirmed to be included in a future KNULLI release. Be aware that patches might be experimental and sometimes just a prototype for a final solution!

By installing a patch and reporting feedback to the developers, you provide valuable assistance to the development of KNULLI. However, patches are often not yet approved by the community or the lead developers of the KNULLI project, consequently, there are no guarantees that they will work for you.

Before installing a patch, make sure you find the source trustworthy.

Most patch installations (or manual modifications) to your KNULLI system follow a simple pattern:

  • Add, replace, or remove files outside of your /userdata folder
  • Make some of the files executable by applying the command chmod +x to the files.
  • Create or update your overlay by executing the command batocera-save-overlay which makes all the changes persistent.
  • Reboot your device.

Some developers simplify this process for you, by providing a patch installer which does most of these steps for you. However, even if you received a patch installer, you will still have to make the patch installer executable and run it from an SSH command line.

Putting files into the right places

There are several options to put files in the right places outside of the /userdata folder.

  • Some patches come with a patch installer which moves the file for you. In this case, simply store the patch installer and all patch files in your /userdata folder according to the specifications made by the developer.
  • You can use SCP/FTP to access the entire KNULLI file system. You will find a subsection on how to use SCP/FTP in the Network Transfer section. This approach allows to put the files directly to their final destination.
  • You can copy the files to your /userdata folder by the same method you add your games and BIOS files. Afterwards, you can move the files from their location inside /userdata to their final destination
    • either by using the OD-Commander tool you will find in the Ports category of your KNULLI device
    • or by connecting via SSH (as explained in the SSH section) and copying/moving the files via cp/mv on the command line, e.g.,
      cp /userdata/system/some-file.sh /usr/bin/some-file.sh
      
      or
      mv /userdata/system/some-file.sh /usr/bin/some-file.sh
      
      where the first argument is always the path to the source file you want to copy/move and the second argument is always the final destination where you want the file to end up.

Making files executable

After you have moved the files in the right places, you need to make sure that all files which are supposed to be executable actually are executable.

Linux file system permissions

On Linux file systems, three different permissions can be set on each file or folder:

  • The read (r) permission allows to read the contents of the file.
  • The write (w) permission allows to modify or even delete the file and/or its content.
  • The execute (x) permission allows to execute the file, which is a special precaution to avoid users accidentally launching dangerous programs. Each program has to be actively set to actually be executable.

The three permissions can be set on each file for three different "user classes":

  • The owner is a single user who owns the file. (It doesn't necessarily has to be the file's creator.)
  • The group is a single group of users who might have a certain interest in the file.
  • The others are all other users who are not the owner and not members of the group.

As a result, the permissions of each file can be represented by a 3-by-3 matrix. In this example, everyone can read and execute the file, but only the owner can modify it:

Read Write Execute
Owner
Group
Others

This corresponds to a string representation of

rwxr-xr-x

where the first 3 letters represent the owner, the next 3 letters represent the group and the last 3 letters represent the others.

Setting Linux file permissions

Make sure that files need to be executable

Before you make a file executable, make sure the file actually needs to be executable. Do not randomly mark files as executable "just to be safe". If you are unsure, check back with the patch developer and ask if a file should be executable or not.

Again, you have several options to set the file permissions on the files of your KNULLI system and make sure a file is executable:

  • If using WinSCP, you can right-click a file and select Properties.
    • In the Common tab, go to the Permissions section and make sure all the Xes are checked if the file is required to be executable.
  • If connected via SSH (as explained in the SSH section), you can
    • use the ls command to check the permissions of a file, e.g.,
      ls -l /userdata/system/patch-installer.sh
      
      to see if a file called patch-installer.sh is executable. The response will look like
      -rwxr-xr-x 1 root root 458 Aug 3 22:20 /userdata/system/patch-installer.sh
      
      and you can tell from the part rwxr-xr-x if a file is executable for everyone.
    • use the chmod command to change the permissions of a file and make it executable, e.g.,
      chmod +x /userdata/system/patch-installer.sh
      

If provided: Run the patch installer

If your patch came with a patch installer which you have made executable in the previous step, you can now execute the patch installer from the SSH command line. To do so, establish a SSH connection (as explained in the SSH section) and launch the installer by calling the script by its name from the command line (Mind the dot at the beginning of the command!), e.g.,

./userdata/system/patch-installer.sh

Have a close look at the output of this command, maybe copy-paste it to a text file for debugging purposes and reach out to the Discord community if you have problems understanding the output.

If required: Create overlay manually

Some patch installers implicitly create or update the overlay. However, some do not. If your patch did not come with a patch installer, if you want to patch in your own modifications, or if you are not sure whether the patch you applied created an overlay, you better do it yourself. To do so, establish a SSH connection (as explained in the SSH section) and execute the following command:

batocera-save-overlay
Once the overlay was created/updated successfully, you can type
reboot
to reboot your system and check if your changes are still in place afterwards.