[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 Tue, Aug 14, 2001 at 03:25:04PM +0200, Yann Guidon wrote:
> > > hi !
> > >
> > > Juergen Goeritz wrote:
> > > > On Mon, 13 Aug 2001, Yann Guidon wrote:

<snip>
> > While we're at it: something that I miss in most CPUs is `variable'
> > register addressing, that is, accessing a register when its number
> > is not known at compile time -- like get and put in the F-CPU, but for
> > ordinary registers.  I know that this is likely to cause trouble with the
> > scheduler, but what about mirroring the general registers in the SR area
> > and simply use get and put?
> 
> This kind of readressing register is a little bit like memory adressing
> for DSP. Register just contain address. But it's the same thing that
> using memory, it make a much longer pipeline, it's hard to manage for
> switching context,...

Going through SRs will solve the problem about the pipeline.
Of course don't expect to use this kind of tricks in "fast" code.
It is only provided to avoid self-modifying code or 64* code replication.
The number of lost cycles is still unknown and it will be implemented only
if the BIST can be appropriately hacked. There should be no modification
of the instruction set.


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