It was heartening to hear an observer at the #rust for #qemu bof at #kvmforum say we seemed a lot more chill about its introduction than the kernel guys. The main conclusion seemed to be that we should publish a roadmap to make our desire to migrate is a genuine desire rather than an experiment. Aside from device models there is interest in migrating block, virto and #softfloat subsystems to rust.

Hearsay has it that 64bit #DoublePrecision #FloatingPoint on #GPU is #slow, but #SoftFloat using two 32bit integers is even much #slower (I tested with #OpenCL).