lmbench's timing loops are loops, in other words they want to be able to do
something N times to get accurate numbers. The only time they don't run
it N times is if the run takes longer than some threshold, I think it was
either a second or 5 seconds.
If you tell me what you want to measure I can probably code up something
for you. Or tell you how to do it - there is a lot of infrastructure in
lmbench which makes writing new tests trivial and I would hope you would
do that. For example, suppose you had a new system call "sysWhizzy"
and you wanted to benchmark it. An lmbench program which would do that
would look like
#include "bench.h"
main()
{
BENCH(sysWhizzy(args), 0);
micro("Whizzy", get_n());
}
Pretty easy, no?
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/