Paulo Marques wrote:
Wouter van Gulik wrote:
[...]
I tested the clzsi2 against all 4.2 billion cases :P It would have
saved me a half a day if avrtest was twice as fast :). I tested the
rest against the test case.
Great. Finally I have a good excuse to tweak the performance of avrtest ;)
Using this excuse, I modified the internal structure of avrtest a little
to gain some performance.
The new version is around 3 times faster than the old one, and does
around ~46 million cycles per second on my P4 at 3GHz (which gives an
average of 65 P4 cycles per one AVR cycle). The exact numbers will
depend on the code being executed, so this is just a rough estimate.
I've just committed the latest version to winavr CVS. You can get it by
doing:
cvs -z3 -d:pserver:[EMAIL PROTECTED]:/cvsroot/winavr
co -P avrtest
or browse the repository:
http://winavr.cvs.sourceforge.net/winavr/avrtest/
Enjoy,
--
Paulo Marques
Software Development Department - Grupo PIE, S.A.
Phone: +351 252 290600, Fax: +351 252 290601
Web: www.grupopie.com
_______________________________________________
AVR-GCC-list mailing list
[email protected]
http://lists.nongnu.org/mailman/listinfo/avr-gcc-list
