On Nov 29, 2007, at 1:14 PM, Eric Saxe wrote:
> Hi Pat,
>
> Pat Bredenberg wrote:
>>
>> On Nov 29, 2007, at 12:57 PM, Eric Saxe wrote:
>>
>>>
>>> I'm thinking the cpu_change_speed and cstate_transition DTrace
>>> probes
>>> (which are already in the PowerTop repo) are generally useful. So
>>> i'm
>>> interested in putting them back to Nevada. Thoughts? Moving forward,
>>> this could eliminate the PowerTop's dependence on the kernel.
>>>
>>> Thoughts?
>> Just my $0.02 from the more user-end side of things, I'd think that'd
>> be beneficial to any replacement tool in the works for mpstat(1)
>> as we
>> move from a CPU-centric way of doing things to a more
>> core/socket-centric. I have nothing specific in mind, of course. :)
>
> Just to clarify, when you say "that'd be beneficial", are you
> referring
> to using DTrace probes vs kstats? ...or did you mean something else?
>
I'm speaking out of ignorance as to what may/may not be coming
around to replace mpstat(1). I think I had just envisioned a tool
that wouldn't need kstats if it could poke the kernel with the Dtrace
probes instead. Of course, mpstat(1) has used kstats rather
effectively for eons now, so maybe I'll just go back to read-only
mode for a while. :)
A follow-up comment to your point I offer is I see no reason for it
to not be generally available from what I've been reading on this
alias up to this point.
Thanks,
Pat
> Thanks,
> -Eric
> _______________________________________________
> tesla-dev mailing list
> tesla-dev at opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/tesla-dev
