[libre-riscv-dev] Introduction

[Jacob Lifshay]

just so you know, we're planning on a separate pipelined alu for
fdiv/fsqrt/frsqrt/idiv/irem that is possibly shared between 2 or 4 cores.

I'm also planning on a unified ALU for i8/i16/i32/i64/f16/f32/f64 where the
alu is split into lanes with separate instructions for each 32-bit half.
So, the multiplier should be capable of 64-bit fmadd, 2x32-bit fmadd,
4x16-bit fmadd, 1x32-bit fmadd + 2x16-bit fmadd (in either order), and all
(8/16/32/64) sizes of integer mul/mulhsu/mulh/mulhu in 2 groups of 32-bits.
We can implement fmul using fmadd with 0 (make sure that we get the right
sign bit for 0 for all rounding modes).

Jacob


On Thu, Jan 31, 2019, 11:30 Aleksandar Kostovic <
alexandar.kostovic at gmail.com wrote:

> >
> > nice.  i did set up some vim macros for highlighting the migen
> > keywords, i have yet to work out how to get them to sit seamlessly on
> > top of the python vim auto-detected bindings (without hacking global
> > system files in /usr/share/vim....)
>
>
> Actually i dont quite like VIM for anything except editing bash scrpits and
> quick edits. For any kind of software needs i use sublime and for HDL stuff
> i use ModelSim's built-in text editor. I can do it all in VIM but i just
> prefer other solutions.
>
> when you're comfortable and have that feeling best described as "a V8
> > supercharged engine revving at 8000 RPM but the clutch pedal is still
> > on the floor", do let me know, as there are some preliminary tests,
> > simulations or self-contained projects that could be done (one being a
> > modular pipelined IEEE754 ALU)
>
>
> Oh believe me i know the feeling. I have it already but its just the lack
> of time due to school recently that was stopping me. Now i have way more
> free time starting friday so there is big window to jump start any project.
> I personally have been reading a bit about floating point and how it works
> already, so i have some concepts in mind. I will let you know when i am
> ready to start that . Just expect it to be soon :)
> _______________________________________________
> libre-riscv-dev mailing list
> libre-riscv-dev at lists.libre-riscv.org
> http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev
>