As a followup to his OSS Europe talk on the future of 32-bit support in the kernel, Arnd Bergmann has put together a detailed plan for the eventual removal of high-memory support, which he calls "one of the least popular features of the Linux kernel". The intent is "to gradually phase out highmem over the next 2 years for mainline kernels". This plan is posted as a prompt for a discussion to be held at the Kernel Summit in December, so chances are it will evolve considerably in the next few months.