Messages in this thread | | | Subject | Re: High CPU load when machine is idle (related to PROBLEM: Unusually high load average when idle in 2.6.35, 2.6.35.1 and later) | From | Peter Zijlstra <> | Date | Tue, 30 Nov 2010 18:29:31 +0100 |
| |
On Tue, 2010-11-30 at 17:53 +0100, Damien Wyart wrote: > * Peter Zijlstra <peterz@infradead.org> [101130 15:59]: > > OK, so here's a less crufty patch that gets the same result on my > > machine, load drops down to 0.00 after a while. > > Seems OK too here. > > > It seems a bit slower to reach 0.00, but that could be because > > I actually changed the load computation for NO_HZ=n as well, I added > > a rounding factor in calc_load(), we no longer truncate the division. > > Yes, really feels slower, but in the two configurations as you write. > Then this is a matter of convention, do not know which version is more > "correct" or "natural", but this is coherent in the two modes (HZ and > NO_HZ). > > Do you plan to push this one to upstream/stable, or do you plan further > modifications?
No this is about it, I've written a bit more comments, and I'll need to come up with a changelog, I'll also wait for Kyle's testers to come back, but yes, then I'll push it upstream/stable.
| |