On 04/19/2013 07:43 AM, Paul Turner wrote:
We could use a name along the lines of "load_index.c"; it calls to
mind where we actually consume these values and (being specific) is
less overloaded than load_avg.
Hmm... How about using "load_machine.c" "load_cpu.c" and "load_entity.c"?
Thanks,
Char
From: Charles Wang
Positive load weight of rq.cfs can not represent positive load weight
of se->cfs_rq. And when se->cfs_rq's load is 0, the slice calculated
by sched_slice is not that sensible.
Use se->cfs_rq for load checking instead of rq->cfs. And correct the
comments.
Cc
calc_load_exit_idle depends on updated jiffies, so you shouldn't move
this before tick_do_update_jiffies64.
And why should we do nohz_balance_enter_idle in tick_nohz_idle_exit?
It's nohz_balance_exit_idle here.
Regards,
Charles
On 10/30/2012 04:27 AM, Steven Rostedt wrote:
［snipped］
> @@ -57
Yes, it's a new way to do scheduler test. But why use kernel
threads? The info, total time, run time, wait time, preempt number,
all can be collected from tasks' sched info from /proc/pid/sched and
/proc/pid/stat.
I don't understand clearly about "pure scheduler performance" here. In
order to
On 10/06/2012 01:23 AM, Peter Zijlstra wrote:
On Fri, 2012-10-05 at 10:10 -0700, Jonathan Nieder wrote:
Peter Zijlstra wrote:
On Thu, 2012-10-04 at 15:27 -0700, Greg Kroah-Hartman wrote:
I'm puzzled as well. Any ideas if I should do anything here or not?
So I think the current v3.5.5 code
On 10/05/2012 09:39 AM, Jonathan Nieder wrote:
Peter Zijlstra wrote:
I can't find wtf went wrong either, the initial patch 5167e8d5417bf5c
contains both hunks, but in that case the fixup 749c8814f0 doesn't make
sense, not can I find anything in merge commits using:
git log -S calc_load_exit
The HZ you configured is 100, and cs is 350+ per second, so
there will be 3.5cs per tick. This may cause loadavg caculation
not correctly.
This problem was discussed in the following link:
https://lkml.org/lkml/2012/6/12/130
If your kernel alread has Peter's latest fix patch
sched/nohz: Rewrite
Commit-ID: 749c8814f08f12baa4a9c2812a7c6ede7d69507d
Gitweb: http://git.kernel.org/tip/749c8814f08f12baa4a9c2812a7c6ede7d69507d
Author: Charles Wang
AuthorDate: Mon, 20 Aug 2012 16:02:33 +0800
Committer: Ingo Molnar
CommitDate: Tue, 4 Sep 2012 14:30:29 +0200
sched: Add missing call to
No email address provided by Azat Khuzhin, so I don't know
how to let him know this. - -!
On 08/20/2012 04:02 PM, Charles Wang wrote:
From Charles Wang
Azat Khuzhin reported "high loadavg in linux-3.6"
After checking for upstream's code, I found Pete
>From Charles Wang
Azat Khuzhin reported "high loadavg in linux-3.6"
After checking for upstream's code, I found Peter's patch
(Commit id:5167e8d5417bf5c322a703d2927daec727ea40dd) not be
fully applied, missing the call for calc_load_exit_idle.
After that idle exit in samp
10 matches
Mail list logo
