high performance computing on graphics processing units: hgpu.org

Legacy device drivers implement both device resource management and isolation. This results in a large code base with a wide high-level interface making the driver vulnerable to security attacks. This is particularly problematic for increasingly popular accelerators like GPUs that have large, complex drivers. We solve this problem with library drivers, a new driver architecture. A library driver implements resource management as an untrusted library in the application process address space, and implements isolation as a kernel module that is smaller and has a narrower lower-level interface (i.e., closer to hardware) than a legacy driver. We articulate a set of device and platform hardware properties that are required to retrofit a legacy driver into a library driver. To demonstrate the feasibility and superiority of library drivers, we present Glider, a library driver implementation for two GPUs of popular brands, Radeon and Intel. Glider reduces the TCB size and attack surface by about 35% and 84% respectively for a Radeon HD 6450 GPU and by about 38% and 90% respectively for an Intel Ivy Bridge GPU. Moreover, it incurs no performance cost. Indeed, Glider outperforms a legacy driver for applications requiring intensive interactions with the device driver, such as applications using the OpenGL immediate mode API.