Re: Re[2]: [avr-gcc-list] Re: GCC-AVR Update (20082003)

[Keith Gudger]

You are right that there is a performance hit.  However, the AT43USB AVRs
run at 12 MHz, and the USB functionality only takes about 5% of the
available cycles.  We have not had any apps where CPU speed was any
consideration - even using floating point calculations.

Keith

On Mon, 25 Aug 2003, Theodore A. Roth wrote:

> 
> 
> On Mon, 25 Aug 2003, Keith Gudger wrote:
> 
> > Here are some recent numbers:
> >
> > ON my application, a USB HID device implemented in the AT43USB355, in IAR
> > it takes 12K, in avr-gcc, it takes 16K.
> >
> > The difference is that IAR has "cross-call optimization".  What this does
> > is look for common code snippets.  If they're longer than 2-3 words (I'm
> > guessing) they substitute a rcall.  It really helps in my code.  It only
> > seems to work inside of each source file, so your source files need to be
> > larger than you might think is optimum.  Part of why it works for me is
> > that a lot of the USB functionality is packed in just a couple of files,
> > giving fodder to this optimizer.
> 
> Just out of curiousity, what kind of performance hit do you take
> because of this? Every rcall/ret is going to cost you 7 clock cycles
> with 4k of code doing this, it seems like it would add up fairly
> quick. Can you disable this optimization is speed critical code
> sections?
> 
> Ted Roth
> 
> _______________________________________________
> avr-gcc-list mailing list
> address@hidden
> http://www.avr1.org/mailman/listinfo/avr-gcc-list
>