Even if it did, you wouldn't stop DMA cheaters from hacking the game. This is the slippery slope we see on Windows with Ring 0 anticheat being defeated with spoofed PCI hardware.
Even if it did, you wouldn't stop DMA cheaters from hacking the game. This is the slippery slope we see on Windows with Ring 0 anticheat being defeated with spoofed PCI hardware.
I know that Nvidia’s been implementing attestation for the PCI links to AI GPUs recently, and iirc USB has supported it for a while now (though I’m not sure whether USB-IF has formalized it). I would be shocked if there isn’t already a DDR standard for same coming up in 6 or 7. It’ll take a while to for all this reach the consumer market but the alternative is, as you correctly observe, the gradual decline of trustworthiness of not only attested Win11 gaming PCs but also gaming consoles — but emptying the cesspool of the software cheaters in the meantime remains a sensible step to take.
See also:
https://ieeexplore.ieee.org/document/6258448 (2012)
https://memlab.ece.gatech.edu/papers/DSN_2023_2.pdf (2023)
https://pcisig.com/blog/ide-and-tdisp-overview-pcie%C2%AE-te... (2025)
https://docs.trustauthority.intel.com/main/articles/articles... (2026)
Which is why I said, even if Linux enforced TPMs, secure runtimes and signed kernels, you'd still be loading firmware blobs and trusting HID/PCI hardware just the same as Windows. Linux would still be attacked, singled-out and made a pariah regardless, so it's good that Valve and the kernel devs generally ignore these design suggestions.