[libre-riscv-dev] [isa-dev] Re: FP transcendentals (trigonometry, root/exp/log) proposal

[Jacob Lifshay]

Atif Zafar asked me to post the following observations:

A couple of observations reading the past set of emails:



   1. Vulkan is the future of 3d graphics. It provides the lowest level
   access to graphics hardware of any API
   2. Much of the functionality of other APIs - WebGL, OpenGL, OpenGL/ES,
   DirectX etc. can be built on top of Vulkan and there are efforts to this
   regard.
   3. High level features such as graphics pipeline commands - i.e. single
   plane 3d clipping of triangles - need to be built in a higher level API


So I agree with you that we should look at SPIR-V and the Vulkan ISA
seriously. Now that ISA is very complex and many of the instructions may
possibly be reconstructed from simpler ones. We need to thus perhaps look
at a "minimized" subset of the Vulkan ISA instructions that truly define
the atomic operations from which the full ISA can be constructed. So the
instruction decode hardware can implement this "higher-level ISA" - perhaps
in microcode - from the "atomic ISA" at runtime while hardware support is
only provided for the "atomic ISA".


>From the SIGGRAPH BOF it was clear there are competing interests. Some
people wanted explicit texture mapping instructions while others wanted HPC
type threaded vector extensions. Although each of these can be accommodated
we need to adjudicate the location in the process pipeline where they
belong - atomic ISA, higher-level ISA or higher-level graphics library.