On Mon, 2006-08-21 at 12:49 +0300, Aivars Kalvans wrote:
> I did include simple C program that produced such result in previous
> mail. It uses http://en.wikipedia.org/wiki/RDTSC to count CPU ticks.
Ah - I missed that; sorry. I get a huge amount of jitter from these
numbers - but I guess the lowest numbers are static; I imagine
time-slicing screws up RDTSC, indeed - (for me) these numbers look
unreliable enough to be pretty scary ;-)
My numbers for AMD64 gave (lowest of lots) [ but now unrepeatable ]:
branching: 28 cycles
function call: 60 cycles
inline multiply: 30 cycles
But really - I'm fairly convinced the numbers are meaningless and/or
wrong :-)
I provoked the fn. call to be so low by running it once first outside
the timing section :-) without that it was ~1000+ cycles, strangely
removing the fn. call gave the inline multiply a cost of ~1000+ cycles
instead ;-) urk ...
HTH,
Michael.
--
[EMAIL PROTECTED] <><, Pseudo Engineer, itinerant idiot
_______________________________________________
Performance-list mailing list
[email protected]
http://mail.gnome.org/mailman/listinfo/performance-list
