HMM has been the multi-year effort to allow device memory to be transparently used by any device process and for mirroring process address space on a device. This has big implications for modern graphics hardware among other devices. NVIDIA has been one of the companies supporting the HMM effort heavily while it should also be of relevance to AMD and other GPU vendors, FPGA hardware, and other possible use-cases.
HMM is ready to be supported by NVIDIA's closed-source kernel driver and NVIDIA developers have also been working on patches for Nouveau so it can be used by this open-source NVIDIA driver too. More background information on Heterogeneous Memory Management is covered in our earlier articles with tracking HMM for a while.
Jerome has been pushing for getting it merged for a while and he's looking to get an answer either way soon. He wrote this week with the HMM v18 patches, "I do not wish to compete for the patchset with the highest revision count and i would like a clear cut position on wether it can be merge or not. If not i would like to know why because i am more than willing to address any issues people might have. I just don't want to keep submitting it over and over until i end up in hell. So please consider applying for 4.12."
Hopefully we'll have an answer soon whether this interesting feature will be accepted for Linux 4.12.