From: "Paweł Staszewski" <pstaszewski@itcare.pl>
To: linux-acpi@vger.kernel.org
Subject: Strange CPU load - acpi_os_read_port
Date: Thu, 22 Jul 2010 21:39:04 +0200 [thread overview]
Message-ID: <4C489E58.7020604@itcare.pl> (raw)
Hello
I have strange problem with cpu load and acpi_os_read_port
please look at
perf top - output:
PerfTop: 1852 irqs/sec kernel:99.4% exact: 0.0% [1000Hz
cycles], (all, 16 CPUs)
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
samples pcnt function DSO
_______ _____ _______________________________
_________________
2217.00 13.7% acpi_os_read_port
[kernel.kallsyms]
613.00 3.8% igb_poll
[kernel.kallsyms]
546.00 3.4% acpi_idle_enter_bm
[kernel.kallsyms]
482.00 3.0% _raw_spin_lock
[kernel.kallsyms]
469.00 2.9% ipt_do_table
[kernel.kallsyms]
415.00 2.6% find_busiest_group
[kernel.kallsyms]
341.00 2.1% tick_dev_program_event
[kernel.kallsyms]
299.00 1.9% u32_classify
[kernel.kallsyms]
299.00 1.9% __slab_free
[kernel.kallsyms]
294.00 1.8% timer_stats_update_stats
[kernel.kallsyms]
235.00 1.5% leave_mm
[kernel.kallsyms]
231.00 1.4% skb_release_data
[kernel.kallsyms]
227.00 1.4% __hrtimer_start_range_ns
[kernel.kallsyms]
220.00 1.4% tick_program_event
[kernel.kallsyms]
211.00 1.3% irq_entries_start
[kernel.kallsyms]
206.00 1.3% cpuidle_idle_call
[kernel.kallsyms]
181.00 1.1% acpi_hw_validate_io_request
[kernel.kallsyms]
178.00 1.1% apic_timer_interrupt
[kernel.kallsyms]
172.00 1.1% consume_skb
[kernel.kallsyms]
162.00 1.0% getnstimeofday
[kernel.kallsyms]
162.00 1.0% rb_erase
[kernel.kallsyms]
161.00 1.0% read_tsc
[kernel.kallsyms]
154.00 1.0% hrtimer_interrupt
[kernel.kallsyms]
148.00 0.9% _raw_spin_unlock_irqrestore
[kernel.kallsyms]
148.00 0.9% napi_schedule
[kernel.kallsyms]
136.00 0.8% __do_softirq
[kernel.kallsyms]
135.00 0.8% dma_issue_pending_all
[kernel.kallsyms]
130.00 0.8% virt_to_head_page
[kernel.kallsyms]
125.00 0.8% clockevents_program_event
[kernel.kallsyms]
118.00 0.7% ladder_select_state
[kernel.kallsyms]
114.00 0.7% hrtimer_force_reprogram
[kernel.kallsyms]
108.00 0.7% timekeeping_get_ns
[kernel.kallsyms]
107.00 0.7% __dma_has_cap
[kernel.kallsyms]
105.00 0.7% dev_queue_xmit
[kernel.kallsyms]
103.00 0.6% acpi_idle_do_entry
[kernel.kallsyms]
102.00 0.6% kfree
[kernel.kallsyms]
98.00 0.6% sched_clock
[kernel.kallsyms]
95.00 0.6% load_balance
[kernel.kallsyms]
95.00 0.6% pskb_expand_head
[kernel.kallsyms]
88.00 0.5% kmem_cache_free
[kernel.kallsyms]
86.00 0.5% run_timer_softirq
[kernel.kallsyms]
84.00 0.5% __rcu_pending
[kernel.kallsyms]
82.00 0.5% acpi_processor_ffh_cstate_enter
[kernel.kallsyms]
80.00 0.5% notifier_call_chain
[kernel.kallsyms]
80.00 0.5% acpi_hw_read_port
[kernel.kallsyms]
79.00 0.5% nf_iterate
[kernel.kallsyms]
77.00 0.5% dev_gro_receive
[kernel.kallsyms]
77.00 0.5% rb_insert_color
[kernel.kallsyms]
75.00 0.5% find_next_bit
[kernel.kallsyms]
74.00 0.5% skb_header_pointer
[kernel.kallsyms]
72.00 0.4% run_posix_cpu_timers
[kernel.kallsyms]
72.00 0.4% hfsc_enqueue
[kernel.kallsyms]
69.00 0.4% run_rebalance_domains
[kernel.kallsyms]
68.00 0.4% __napi_complete
[kernel.kallsyms]
67.00 0.4% ktime_get
[kernel.kallsyms]
67.00 0.4% scheduler_tick
[kernel.kallsyms]
66.00 0.4% skb_release_head_state
[kernel.kallsyms]
64.00 0.4% __run_hrtimer
[kernel.kallsyms]
64.00 0.4% acpi_hw_validate_register
[kernel.kallsyms]
62.00 0.4% ip_route_input_common
[kernel.kallsyms]
mpstat -P 1,2 1 10
Linux 2.6.35-rc5 (eP-RTR-IX-01) 07/22/10 _x86_64_
(16 CPU)
23:39:26 CPU %usr %nice %sys %iowait %irq %soft
%steal %guest %idle
23:39:27 1 0.00 0.00 0.00 0.00 0.00 52.00
0.00 0.00 48.00
23:39:27 2 0.00 0.00 0.00 0.00 0.00 17.00
0.00 0.00 83.00
23:39:28 1 0.00 0.00 0.00 0.00 0.00 39.00
0.00 0.00 61.00
23:39:28 2 0.00 0.00 0.00 0.00 0.00 8.00
0.00 0.00 92.00
23:39:29 1 0.00 0.00 0.00 0.00 0.00 47.00
0.00 0.00 53.00
23:39:29 2 0.00 0.00 0.00 0.00 0.00 9.00
0.00 0.00 91.00
23:39:30 1 58.00 0.00 0.00 0.00 0.00 42.00
0.00 0.00 0.00
23:39:30 2 0.00 0.00 0.00 0.00 0.00 18.00
0.00 0.00 82.00
23:39:31 1 9.00 0.00 0.00 0.00 0.00 51.00
0.00 0.00 40.00
23:39:31 2 2.00 0.00 0.00 0.00 0.00 18.00
0.00 0.00 80.00
23:39:32 1 0.00 0.00 0.00 0.00 0.00 39.00
0.00 0.00 61.00
23:39:32 2 2.00 0.00 0.00 0.00 0.00 20.00
0.00 0.00 78.00
23:39:33 1 0.00 0.00 0.00 0.00 0.00 48.00
0.00 0.00 52.00
23:39:33 2 0.00 0.00 0.00 0.00 0.00 13.00
0.00 0.00 87.00
23:39:34 1 0.00 0.00 0.00 0.00 0.00 52.00
0.00 0.00 48.00
23:39:34 2 0.00 0.00 0.00 0.00 0.00 9.00
0.00 0.00 91.00
23:39:35 1 0.00 0.00 0.00 0.00 0.00 52.00
0.00 0.00 48.00
23:39:35 2 0.00 0.00 0.00 0.00 0.00 18.00
0.00 0.00 82.00
23:39:36 1 0.00 0.00 0.00 0.00 0.00 50.00
0.00 0.00 50.00
23:39:36 2 0.00 0.00 0.00 0.00 0.00 9.00
0.00 0.00 91.00
Average: 1 6.70 0.00 0.00 0.00 0.00 47.20
0.00 0.00 46.10
Average: 2 0.40 0.00 0.00 0.00 0.00 13.90
0.00 0.00 85.70
Server is acting as a bgp router - 500Mbit/s RX / 500Mbit/s TX traffic.
But the topper of cpu load is acpi_os_read_port - why ?
Thanks
Pawel
next reply other threads:[~2010-07-22 19:45 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-07-22 19:39 Paweł Staszewski [this message]
2010-07-22 19:44 ` Strange CPU load - acpi_os_read_port Paweł Staszewski
2010-08-15 3:33 ` Len Brown
2010-08-15 10:25 ` Paweł Staszewski
2010-08-15 14:05 ` Len Brown
2010-08-19 9:11 ` Paweł Staszewski
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=4C489E58.7020604@itcare.pl \
--to=pstaszewski@itcare.pl \
--cc=linux-acpi@vger.kernel.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.