From: Rik van Riel <riel@redhat.com>
To: Glauber Costa <glommer@redhat.com>
Cc: kvm@vger.kernel.org, avi@redhat.com, zamsden@redhat.com,
mtosatti@redhat.com
Subject: Re: [RFC 4/7] change kernel accounting to include steal time
Date: Thu, 26 Aug 2010 17:19:23 -0400 [thread overview]
Message-ID: <4C76DA5B.2030701@redhat.com> (raw)
In-Reply-To: <1282772597-4183-5-git-send-email-glommer@redhat.com>
On 08/25/2010 05:43 PM, Glauber Costa wrote:
> This patch proposes a common steal time implementation. When no
> steal time is accounted, we just add a branch to the current
> accounting code, that shouldn't add much overhead.
>
> When we do want to register steal time, we proceed as following:
> - if we would account user or system time in this tick, and there is
> out-of-cpu time registered, we skip it altogether, and account steal
> time only.
> - if we would account user or system time in this tick, and we got the
> cpu for the whole slice, we proceed normaly.
> - if we are idle in this tick, we flush out-of-cpu time to give it the
> chance to update whatever last-measure internal variable it may have.
>
> This approach is simple, but proved to work well for my test scenarios.
> in a UP guest on UP host, with a cpu-hog in both guest and host shows
> ~ 50 % steal time. steal time is also accounted proportionally, if
> nice values are given to the host cpu-hog.
>
> A cpu-hog in the host with no load in the guest, produces 0 % steal time,
> with 100 % idle, as one would expect.
>
> Signed-off-by: Glauber Costa<glommer@redhat.com>
> ---
> include/linux/sched.h | 1 +
> kernel/sched.c | 29 +++++++++++++++++++++++++++++
> 2 files changed, 30 insertions(+), 0 deletions(-)
>
> diff --git a/include/linux/sched.h b/include/linux/sched.h
> index 0478888..e571ddd 100644
> --- a/include/linux/sched.h
> +++ b/include/linux/sched.h
> @@ -312,6 +312,7 @@ long io_schedule_timeout(long timeout);
> extern void cpu_init (void);
> extern void trap_init(void);
> extern void update_process_times(int user);
> +extern cputime_t (*hypervisor_steal_time)(void);
> extern void scheduler_tick(void);
>
> extern void sched_show_task(struct task_struct *p);
> diff --git a/kernel/sched.c b/kernel/sched.c
> index f52a880..9695c92 100644
> --- a/kernel/sched.c
> +++ b/kernel/sched.c
> @@ -3157,6 +3157,16 @@ unsigned long long thread_group_sched_runtime(struct task_struct *p)
> return ns;
> }
>
> +cputime_t (*hypervisor_steal_time)(void) = NULL;
> +
> +static inline cputime_t get_steal_time_from_hypervisor(void)
> +{
> + if (!hypervisor_steal_time)
> + return 0;
> + return hypervisor_steal_time();
> +}
> +
> +
> /*
> * Account user cpu time to a process.
> * @p: the process that the cpu time gets accounted to
> @@ -3169,6 +3179,12 @@ void account_user_time(struct task_struct *p, cputime_t cputime,
> struct cpu_usage_stat *cpustat =&kstat_this_cpu.cpustat;
> cputime64_t tmp;
>
> + tmp = get_steal_time_from_hypervisor();
> + if (tmp) {
> + account_steal_time(tmp);
> + return;
> + }
> +
> /* Add user time to process. */
> p->utime = cputime_add(p->utime, cputime);
> p->utimescaled = cputime_add(p->utimescaled, cputime_scaled);
I see one problem here.
What if get_steal_time_from_hypervisor() returns a smaller
amount of time than "cputime"?
Would it be better to account tmp as stealtime, and the
difference (cputime - tmp) as user/sys/... time?
--
All rights reversed
next prev parent reply other threads:[~2010-08-26 21:19 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-08-25 21:43 [RFC 0/7] KVM steal time implementation Glauber Costa
2010-08-25 21:43 ` [RFC 1/7] Implement getnsboottime kernel API Glauber Costa
2010-08-25 21:43 ` [RFC 2/7] change headers preparing for steal time Glauber Costa
2010-08-25 21:43 ` [RFC 3/7] measure time out of guest Glauber Costa
2010-08-25 21:43 ` [RFC 4/7] change kernel accounting to include steal time Glauber Costa
2010-08-25 21:43 ` [RFC 5/7] kvm steal time implementation Glauber Costa
2010-08-25 21:43 ` [RFC 6/7] touch softlockup watchdog Glauber Costa
2010-08-25 21:43 ` [RFC 7/7] tell guest about steal time feature Glauber Costa
2010-08-26 22:13 ` [RFC 5/7] kvm steal time implementation Rik van Riel
2010-08-26 22:35 ` Glauber Costa
2010-08-26 17:23 ` [RFC 4/7] change kernel accounting to include steal time Marcelo Tosatti
2010-08-26 20:28 ` Glauber Costa
2010-08-26 20:47 ` Marcelo Tosatti
2010-08-26 21:05 ` Rik van Riel
2010-08-26 21:13 ` Glauber Costa
2010-08-26 21:14 ` Anthony Liguori
2010-08-26 21:40 ` Glauber Costa
2010-08-26 23:12 ` Marcelo Tosatti
2010-08-27 0:33 ` Glauber Costa
2010-08-27 15:25 ` Marcelo Tosatti
2010-08-26 21:19 ` Rik van Riel [this message]
2010-08-26 21:39 ` Glauber Costa
2010-08-29 9:59 ` Avi Kivity
2010-08-29 15:13 ` Rik van Riel
2010-08-29 15:25 ` Avi Kivity
2010-08-29 15:42 ` Rik van Riel
2010-08-29 15:47 ` Avi Kivity
2010-08-30 12:42 ` Glauber Costa
2010-08-30 13:15 ` Avi Kivity
2010-08-26 20:54 ` [RFC 3/7] measure time out of guest Zachary Amsden
2010-08-26 21:14 ` Glauber Costa
2010-08-29 9:53 ` Avi Kivity
2010-08-26 20:44 ` [RFC 2/7] change headers preparing for steal time Zachary Amsden
2010-08-26 21:04 ` Rik van Riel
2010-08-26 21:17 ` Glauber Costa
2010-08-26 22:11 ` Rik van Riel
2010-08-29 9:51 ` Avi Kivity
2010-08-30 12:44 ` Glauber Costa
2010-08-30 13:10 ` Avi Kivity
2010-08-26 19:46 ` [RFC 1/7] Implement getnsboottime kernel API Rik van Riel
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4C76DA5B.2030701@redhat.com \
--to=riel@redhat.com \
--cc=avi@redhat.com \
--cc=glommer@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=mtosatti@redhat.com \
--cc=zamsden@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).