Researchers at the University of Toronto published a technical paper titled “GPUThor: Amplifying Rowhammer Attacks via Non-Uniform Patterns to Exploit ECC-Protected GPUs.”
“We present GPUThor, a high-intensity Rowhammer attack on NVIDIA GPUs leveraging non-uniform hammering. GPUThor reverse engineers GPU memory-access coalescing behavior to enable non-uniform hammering patterns on GPUs, that activate aggressor rows more intensely than decoy rows. Additionally, by identifying refresh instances when in-DRAM mitigations are applied, it constructs longer attack patterns that escape mitigation across refresh intervals, further increasing hammering intensity. ”
Lin, Chris S., Joyce Qu, Aditya Rajeev, and Gururaj Saileshwar. “GPUThor: Amplifying Rowhammer Attacks via Non-Uniform Patterns to Exploit ECC-Protected GPUs.” arXiv preprint arXiv:2609.16546 (2026).
The post High-Intensity Rowhammer Attack On GPUs Leveraging Non-uniform Hammering (U. of Toronto) appeared first on Semiconductor Engineering.
Source link







