OpenVZ Forum


Home » Mailing lists » Devel » [PATCH 0/2] Show per-cpu data in cpuacct stats
[PATCH 2/2] expose fine-grained per-cpu data for cpuacct stats [message #46871 is a reply to message #46869] Wed, 20 June 2012 11:38 Go to previous message
Glauber Costa is currently offline  Glauber Costa
Messages: 916
Registered: October 2011
Senior Member
The cpuacct cgroup already exposes user and system numbers in a per-cgroup
fashion. But they are a summation along the whole group, not a per-cpu figure.
Also, they are coarse-grained version of the stats usually shown at places
like /proc/stat.

I want to have enough cgroup data to emulate the /proc/stat interface. To
achieve that, I am creating a new file "stat_percpu" that displays the
fine-grained per-cpu data. The original data is left alone.

The format of this file is as follows:

1line header
cpux val1 val2 ... valn.

It is the same format used for the cpu part /proc/stat, except for the header,
that may allow us to add fields in the future if they prove themselves needed.

Signed-off-by: Glauber Costa <glommer@parallels.com>
CC: Peter Zijlstra <a.p.zijlstra@chello.nl>
CC: Paul Turner <pjt@google.com>
---
kernel/sched/core.c | 37 +++++++++++++++++++++++++++++++++++++
1 file changed, 37 insertions(+)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 4ce68f6..5c19754 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -8155,6 +8155,39 @@ static int cpuacct_stats_show(struct cgroup *cgrp, struct cftype *cft,
return 0;
}

+static inline void do_fill_seq(struct seq_file *m, struct cpuacct *ca,
+ int cpu, int index)
+{
+ struct kernel_cpustat *kcpustat = per_cpu_ptr(ca->cpustat, cpu);
+ u64 val;
+
+ val = cputime64_to_clock_t(kcpustat->cpustat[index]);
+ seq_put_decimal_ull(m, ' ', val);
+}
+
+static int cpuacct_stats_percpu_show(struct cgroup *cgrp, struct cftype *cft,
+ struct seq_file *m)
+{
+ struct cpuacct *ca = cgroup_ca(cgrp);
+ int cpu;
+
+ seq_printf(m, "user nice system irq softirq guest guest_nice\n");
+
+ for_each_online_cpu(cpu) {
+ seq_printf(m, "cpu%d", cpu);
+ do_fill_seq(m, ca, cpu, CPUTIME_USER);
+ do_fill_seq(m, ca, cpu, CPUTIME_NICE);
+ do_fill_seq(m, ca, cpu, CPUTIME_SYSTEM);
+ do_fill_seq(m, ca, cpu, CPUTIME_IRQ);
+ do_fill_seq(m, ca, cpu, CPUTIME_SOFTIRQ);
+ do_fill_seq(m, ca, cpu, CPUTIME_GUEST);
+ do_fill_seq(m, ca, cpu, CPUTIME_GUEST_NICE);
+ seq_putc(m, '\n');
+ }
+
+ return 0;
+}
+
static struct cftype files[] = {
{
.name = "usage",
@@ -8169,6 +8202,10 @@ static struct cftype files[] = {
.name = "stat",
.read_map = cpuacct_stats_show,
},
+ {
+ .name = "stat_percpu",
+ .read_seq_string = cpuacct_stats_percpu_show,
+ },
{ } /* terminate */
};

--
1.7.10.2
 
Read Message
Read Message
Read Message
Previous Topic: Filter container traffic
Next Topic: [PATCH 0/4] Proposed slab patches as basis for memcg
Goto Forum:
  


Current Time: Mon Nov 04 01:17:46 GMT 2024

Total time taken to generate the page: 0.03476 seconds