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

Re: [f-cpu] RC5, F-CPU and srotl

> > > Depends on your definitions of `really' and `simd' ;)
> > I think that you have understand what I mean. ;-)
> Didn't you see the smiley?
I forgot to put one, so I correct the past ;-)

> > > > 	An other thing about the srotl, it currently double the asm code.
> > > > It mean that, if we have a real srotl operation, we will have the
> > > > same or better performance than the K7 core (who is actually the best
> > > > for this algorithm). So my question is what is the cost of having a
> > > > real srotl instruction ?
> > > A more complex shifter unit, with more delay.
> > It was what I was thinking, but is it so important ?
> If you don't mind that shift operations take more than 1 cycle...

> > So, we need to patch the manual.
> In many places. I made a long list of ambiguities and errors months
> ago (e.g. look for the phrase "Dark and Dusty Corners", September 2001).
Did you post something ? (If yes give me the title of your post and I will
do the update).

I currently scheduled to do :
	- Verify all the instruction set, that they didn't have any error like the 
rotl instruction.
	- Add the new load and store operation that whas asked by christophe
	- Update loadm/storem to be up to date with what we say on 
the mailing-list
	- Add the dnetc-rc5 F-CPU code sample
	- Find a way to add an index (if some one know howto to do that in latex)
	- Put all the different manual (english, french, german) in the same

If someone want to add something, only ask. When I finished to do that I will
use a cvs so that we can work more easily.


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