Reverse-engineering the RK3588 NPU: Hacking limits to run vision transformers
Posted by rcarmo 8 hours ago
Comments
Comment by Neywiny 6 hours ago
Comment by kvuj 6 hours ago
Comment by PunchyHamster 5 hours ago
Comment by Neywiny 4 hours ago
Comment by jauntywundrkind 6 hours ago
For what it's worth, it seems like there's a bunch of open source NPU work in progress too. There's a layer "TEFLON" for Gallium3D shared by most of these drivers, that TensorFlow can use. Then hardware drivers for Rockchip (via ROCKET driver), and Vivante (with their Etnaviv drivers). It'd be extra interesting now to see how (or if?) they've dealt with the system constraints (small scratchpad size) here. https://www.phoronix.com/news/Gallium3D-Teflon-Merged https://www.phoronix.com/news/Rockchip-NPU-Linux-Mesa https://www.phoronix.com/news/Two-NPU-Accel-Drivers-2026
Comment by doctorpangloss 5 hours ago