[Libre-soc-dev] effect of more decode pipe stages on hardware requirements for execution resources for OoO processors

[Jacob Lifshay]

On Tue, Feb 15, 2022, 16:49 Luke Kenneth Casson Leighton <lkcl at lkcl.net>
wrote:

> On Wed, Feb 16, 2022 at 12:41 AM Luke Kenneth Casson Leighton
> <lkcl at lkcl.net> wrote:
>
> > an illustration of the problem can be done by making two hypothetical
> > 8-way-multi-issue processors:
> >
> > 1) the decode stage take 8 stages and the execution only 1.
> > 2) the decode stage is 1 clock and execution only 1.
>
> assume that registers are decoded at stage 1 in both cases,
> such that all 8 instructions may be successfully issued to
> in-flight buffers per clock cycle.
>

i'm assuming that instructions are pushed to the reorder buffer and other
buffers only after decoding finishes, cuz that's what every OoO design i've
seen does.

>
> also assume that both the LD and ST are satisfied from cache
> in 1 clock cycle.
>

already done.

Jacob