On Tue, Mar 06, 2007 at 06:59:14AM +0900, YAMAMOTO Takashi wrote:

> > > I think both of those should be per-LWP, and the sysctl/libkvm interface
> > > should mimic the old behaviour by adding the values for all LWPs in a
> > > process.
> > As I understand, you would tend to "LWPize" this part, hence, there would be
> > more data moving from proc to lwp. One should also keep in mind that top(1)
> > and ps(1) interfaces are defined by POSIX.
> > Can you elaborate with this?
> 
> i don't think top is in posix.
> anyway i guess they can be calculated on-demand as we currently do
> for p_rtime.

Which reminds me.. p_pctcpu is wrong because it's not scaled by the number
of CPUs. So you can have top reporting e.g. a multithreaded app using 400%
CPU. :)

Andrew