84
Will kernel-level anti-cheat ever work on linux? (lemmy.zip)
submitted 1 year ago by to c/linux@lemmy.ml
82 comments fedilink hide all child comments
 

From both a technical perspective and if the maintainers of these anti-cheat will consider porting or re-writing kernel level anti-cheat to work on linux, is it possible? Do you think that the maintainers of kernel level anti-cheat will be adamant in not doing it, or that the kernel even supports it or will support it. I think that if it ever happens, there will be a influx of people moving to linux, or abandoning their duelboots, and that alot of people will hate that such a thing is available on linux.

you are viewing a single comment's thread
view the rest of the comments
[–] 13 points 1 year ago (11 children)

I'm not a programmer or cheater or anything, but I think the answer is yes and no. Yes it could technically be done and even work as intended as long as the device is locked down to prevent the user from replacing the shipped kernel (which would be a bad thing for users). However, savvy people could (in theory) make custom kernels that lie to the kernel module, causing the module to report there is no cheating when there is. It's my understanding that it's close to the current situation with Windows and virtual machines and anticheat: you can cheat by running your game in a VM and then have that virtual hardware extract secret information or flip bits in the right spots. Most competitive games will refuse to run in a VM for this reason.

  • permalink
  • fedilink
  • source
  • hideshow 11 child comments
    •
  • [–] 3 points 1 year ago (7 children)

    Kernel level anti cheats require secure boot. You can't just "lie" and load an unsigned kernel.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 7 child comments
    •
  • [–] 7 points 1 year ago* (last edited 1 year ago) (5 children)

    You can add your own signing keys to the UEFI and boot an modified bootloader and Kernel that you have signed yourself. So yes, it is possible to "lie"

    For such a locked down system, akin to game consoles or smartphones, would be needed. And even those get jail broken and manipulated, so "total security" on there is not complete but easier to check and ensure. Another way to make sure that the code is not manipulated would be to put all those games into the cloud and have every player only play via streaming. All the code would then run on secured, locked down and verified machines.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 5 child comments
    •
  • [–] 2 points 1 year ago* (last edited 1 year ago)

    Another technique that helps is to limit the amount of information shared with clients to need to know info. This can be computationally intensive server-side and hard to get right .. but it can help in many cases. There are evolving techniques to do this.

    In FPS games, there can also be streaming input validation. eg. Accurate fire requires the right sequence of events and/or is used for cheat detection. At the point where cheats have to emulate human behaviour, with human-like reaction times, the value of cheating drops.

    That's the advanced stuff. Many games don't even check whether people are running around out of bounds, flying through the air etc. Known bugs and map exploits don't get fixed for years.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 1 point 1 year ago (3 children)

    And then your keys will be rejected by the anticheat. Just because you can sign your kernel and load it does not mean a kernel module can't verify who signed it.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 3 child comments
    •
  • [–] 2 points 1 year ago (2 children)

    Yes, but with a modified Kernel you can fake what the anticheat reads when it checks the key, so you just feed it the key it wants to see instead of your own. The anticheat module would need run on a higher level then the Kernel itself to prevent that, for example alongside the CPU (like the Intel Management Engine).

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 1 point 1 year ago

    Linux secure boot was a little weird last I checked. The kernel and modules don't need to be secure boot signed. Most distros can use shim to pass secure boot and then take over the secure boot process.

    There are dkms kernel modules that are user compiled. These are signed using a machine owner key. So the machine owner could for sure compile their own malicious version and still be in a secure boot context.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 0 points 1 year ago (2 children)

    This is where TPMs, measured boot, and remote attestation come in.

    You can run whatever kernel you want, but if it is not an approved kernel, you wouldn't be able to attest to running an approved kernel; allowing whatever DRM scheme the developer put in to active.

    I believe this is how the higher levels of Android's Play Integrity system work.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 3 points 1 year ago (1 child)

    When Microsoft first proposed something like that a couple decades ago, it was widely seen as the nightmarish corporate power grab it was. Even mainstream, non-techy publications were critical.

    I believe this is how the higher levels of Android’s Play Integrity system work.

    It is.

    How the fuck did this become acceptable?

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 1 child comment
    •