[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [f-cpu] opcodes : missing m4



hi,

jaap stolk wrote:
> --- Yann Guidon <whygee@f-cpu.org> wrote:
> > hi !
>   -- snip -
> > not in the decoder : the decoder decodes, and Xbar
> > transmits.
> ok i put it in the xbar. :-)
great ! things become so simple now :-)

> > > can't stall the xbar, because we need to keep re-
> > > loading the data in case there is a bypass.
> > hmmmmmmmgrmblgrmbl.... did i miss something ?....
> to keep it in hardware terms:( i won't say re-run the xbar..)
> the decoder may discover a (posible) bypass during
> the stall, and change the input signals of the x-bar.
> that will do the bypass.
oh god, these HW/SW collisions...
now i get it.

> > Don't you think it would be better to split eu_rop2
> > to solve all these troubles ?
> > For example, we remove the fanout/xbar stage
> > and we let the Xbar function do the switches when
> > the control signals are ok....
> i think its more of a docoder (or xbar) task.
remember : "the decoder decodes, and the Xbar transmits." :-)

> doing anything in a EU before there is any data
> is a bit confusing.
sure. oh i forgot : "an EU processes data".

> > ok. do i remove the pipeline stage (with tmp_function)
> > from eu_rop2_cycle() ? i believe that there should
> > be no mention of "stalled" in the execution units
> i agree.
ok so i do it.

> the stalled signal is not strictly needed, in the EU.
sure because an EU is a strict, straight pipeline.
control signals come before and after, but not inside
(ideally).

> the decoder is stalled, so it will re-provide the
> ROP2_function. but it will save a lot of computation
> (or bit changes) when we stop the ROP2_function
> signal.
mmmm now you manage this in the Xbar so i can simply
strip the pipeline worries out of ROP2, right ?

WHYGEE
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
*************************************************************
To unsubscribe, send an e-mail to majordomo@seul.org with
unsubscribe f-cpu       in the body. http://f-cpu.seul.org/