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

[lkcl]

On February 16, 2022 2:50:54 AM UTC, Jacob Lifshay <programmerjake at gmail.com> wrote:

>ctr5 <- ctr1 + 4

if you keep coming up with barriers that allow you to avoid exploring the original problem i'm just going to stop responding.

the object of the exercise is to demonstrate that a larger number of RSes are required and larger DMs if the pipelines are longer.

this is fundamental and obvious but for some reason you seek to avoid exploring the issue by creating artificial barriers and to be absolutely frank it is tiresome.

i'd like to see you help collaboratively get to the core of the issue without me having to smash through a constant barrage of strawman "radiobutton" arguments.

once that is done then the limitations can be explored but if you keep putting artificial barriers even to prevent getting to a first premise i'm just going to stop responding and move on to higher priority items.

l.