It’s been a long time in the works, but a memory management feature intended to give machine learning or other GPU-powered applications a major performance boost is close to making it into one of the next revisions of the kernel.Heterogenous memory management (HMM) allows a device’s driver to mirror the address space for a process under its own memory management. As Red Hat developer Jérôme Glisse explains, this makes it easier for hardware devices like GPUs to directly access the memory of a process without the extra overhead of copying anything.
It also doesn’t violate the memory protection features afforded by modern OSes.To read this article in full or to leave a comment, please click here

Leave a Reply