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

Re: [f-cpu] Re: Floating-Point



hi,

nicO wrote:
> Michael Riepe a écrit :
> > On Sun, Aug 19, 2001 at 05:33:05PM -0400, nicO wrote:
> > > Hello,
> > > I'm looking for the "usual" definition of the port of a unit :
> > > 3 data in,
> > Or less.
> Yep!
> > > 2 data out,
> > Or more.
> MORE ? I beleive that the fc0 is 3r2w (which give the number of register
> bank port). So ?

some units have "variable latency". IE the multiplier can give results
with different latency, depending on the data chunk size.
The different outputs are then "mixed" with the Xbar (sorry : the big mux).

> > > SIMD flag,
> > Make that `chunk size'; I use std_ulogic_vector(2 downto 0) for it.
> The same of the instruction ?

i don't know : Michael uses a specific strategy.

> > > If there is something to say if the outed data is correct or not ?
> > Currently not; the scheduler should know.  But we can add a `here is
> > the result, take it or I'll throw it away in the next cycle' signal.
> Maybe, it's not useful but it could at least help for debugging.
at least.

> > Something is missing: the lines that select the instruction to execute
> > (if the unit can handle more than one instruction).
> Could have many enable flag, corresponding to each differents
> instructions ?

i still don't catch the precise context.

> nicO
WHYGEE "programmer ou dormir, il faut choisir"...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
*************************************************************
To unsubscribe, send an e-mail to majordomo@seul.org with
unsubscribe f-cpu       in the body. http://f-cpu.seul.org/