Brock Pytlik wrote:
Mark Haywood wrote:
Brock Pytlik wrote:
[snip]
Thanks, that made a huge difference. I'm now spending about 65-70%
of time in C1 (down from 80%), but now I'm seeing my P-states change
instead of being pegged at 2531Mhz. When I'm essentially idling, I'm
@ 800Mhz, and popping up to higher P states periodically.
That seems like a lot of time to be spending in C0. Is your system
supposed to be essentially idle? I see a lot of wakeups due to audiohd?
Yeah, I'm not sure why audiohd is firing that much. I was "idle" not
idle. I had a few apps running, firefox, thunderbird, pidgin if I
remember right, but nothing that was taxing the machine in any
meaningful way. Whatever it was though apparently went away b/c I'm
now at 88% in C1 and audiohd isn't showing up at all in that list. I'm
not sure why it was in the first place.
Hmm. event-mode might work for you now then.
One other thing I'm noticing that seems suspicious is that my
percentage is always at 100% for one P state. It seems somewhat
unlikely (to me at least) that it's changing P states exactly on
every 5 second interval that powertop's using to sample. Is it
possible that the percentage isn't doing quite what's expected?
This doesn't surprise me all that much. In poll-mode the P-State
transitions are much less frequent. You probably won't see them
change often at all unless you either go from being idle to pretty
busy or from busy to idle. In event-mode you should see transitions
many times a second even on a somewhat idle system.
Well, I guess I expected that the percentage being reported was the
percent of time (roughly) spent in that P-state during the last
sampling interval (5 seconds). So, when I taxed the machine and made
it jump from 800Mhz to 2531Mhz, I was surprised that during that
interval, the percentage was still 100%. For example if I started
taxing the machine 2.5 seconds after the last refresh, I would've
expected 50% @ 800Mhz and 50% @ 2531Mhz. Perhaps I'm just
misunderstanding the meaning of the number I'm seeing, but that's what
this sentence from the man page would lead me to believe:
The tool analyzes system activity periodically and displays a
summary of how long the processor is executing at each supported
power state.
No. I think you have it right. I just figured you might have misses the
5 second window where it jumped. Maybe one of the powertop experts will
jump into with their opinion on this.
Mark
Thanks,
Brock
Mark
Brock
_______________________________________________
indiana-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/indiana-discuss
