On Tue, 2005-03-08 at 08:29 +0100, Guillaume Thouvenin wrote:
> TODO:
> - Run the lmbench with the user space application that manages
> group of processus. if fork connector is not used the only
> overhead is a test in the do_fork() routine.
For information here are some results when using the process creation
tests "lat_proc fork". Test was run ten times thus the average is
computed with the ten metrics.
with a kernel 2.6.11-rc4-mm1
max value = 164.0588 msec
min value = 159.8571 msec
average = 161.7012 msec
with a kernel 2.6.11-rc4-mm1 and the cnfork (cn_fork_enable == 0)
max value = 163.3438 msec
min value = 159.8857 msec
average = 160.9447 msec
with a kernel 2.6.11-rc4-mm1 and the cnfork (cn_fork_enable == 1)
max value = 177.6885 msec
min value = 170.9057 msec
average = 173.7675 msec
So we see that when the fork connector is disable the time it takes to
split a process into two copies is the same (and it's normal) and when
the fork connector is enable, it's around 7.5% slower.
Best,
Guillaume
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
