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

Re: [f-cpu] Relative branch



On Wed, 11 Dec 2002 15:48:41 +0100
Michael Riepe <michael@stud.uni-hannover.de> wrote:

> On Wed, Dec 11, 2002 at 11:29:38AM +0100, devik wrote:
> > Just short idea, I was found that in compiled
> > code for fcpu is missing shorter/faster jump.
> > 
> > Why dont we have jump which instead of registers
> > to use as return address save and target adres
> > have instead 12bit immediate which would REPLACE
> > lower 8 bits (for example) of IP.
> 
> Umh. That will require heavy alignments, which may lead to increasing
> aliasing problems. And it's hard to handle inside the compiler and
> linker. You have to know in advance where the page boundaries are -
> that is, every object file will need 4k alignment. The resulting
> executable will contain lots of holes, and while the size of the
> executed code decreases, the file size will actually increase,
> resulting in more (and more frequent) memory loads which slow down
> execution. It's a waste of space and memory bandwidth, IMHO.

Should be more investigate, no ? Because that could be a compiler choice
?

In case of unconditionnal jump, it could be pretty fast !

> 
> -- 
>  Michael "Tired" Riepe <Michael.Riepe@stud.uni-hannover.de>
>  "All I wanna do is have a little fun before I die"
> *************************************************************
> To unsubscribe, send an e-mail to majordomo@seul.org with
> unsubscribe f-cpu       in the body. http://f-cpu.seul.org/
> _____________________________________________________________________
> GRAND JEU SMS : Pour gagner un NOKIA 7650, envoyez le mot IF au 61321
> (prix d'un SMS + 0.35 euro). Un SMS vous dira si vous avez gagn_.
> R_glement : http://www.ifrance.com/_reloc/sign.sms
*************************************************************
To unsubscribe, send an e-mail to majordomo@seul.org with
unsubscribe f-cpu       in the body. http://f-cpu.seul.org/