From: Fengguang Wu <fengguang.wu@intel.com>
To: lkp@lists.01.org
Subject: Re: [ARM] f9ff907b0af: -100.0% slabinfo.cfq_queue.num_objs
Date: Thu, 29 May 2014 14:00:12 +0800 [thread overview]
Message-ID: <20140529060012.GF7412@localhost> (raw)
In-Reply-To: <5386BDDB.9010708@intel.com>
[-- Attachment #1: Type: text/plain, Size: 7405 bytes --]
Let's ignore this. I don't see meaningful perf/power changes.
On Thu, May 29, 2014 at 12:55:55PM +0800, Jet Chen wrote:
> TO: Rabin Vincent <rabin@rab.in>
> CC: Russell King <rmk+kernel@arm.linux.org.uk>
> CC: LKML <linux-kernel@vger.kernel.org>
> CC: lkp(a)01.org
>
> FYI, we noticed the below changes on
>
> git://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master
> commit f9ff907b0af561dcde4683f7c9f71dc0f41d3be3 ("ARM: 8048/1: fix v7-M setup stack location")
>
> test case: lkp-st02/dd-write/11HDD-RAID5-cfq-ext4-1dd
>
> v3.15-rc6 f9ff907b0af561dcde4683f7c
> --------------- -------------------------
> 273 ~ 3% -100.0% 0 ~ 0% TOTAL slabinfo.cfq_queue.num_objs
> 65541468 ~ 1% -100.0% 0 ~ 0% TOTAL ftrace.balance_dirty_pages.md0.time
> 273 ~ 3% -100.0% 0 ~ 0% TOTAL slabinfo.cfq_queue.active_objs
> 969 ~ 8% -100.0% 0 ~ 0% TOTAL slabinfo.ip6_dst_cache.num_objs
> 948 ~ 8% -100.0% 0 ~ 0% TOTAL slabinfo.ip6_dst_cache.active_objs
> 12399 ~ 1% -100.0% 0 ~ 0% TOTAL ftrace.bdi_dirty_ratelimit.md0.balanced_dirty_ratelimit
> 14234 ~ 2% -100.0% 0 ~ 0% TOTAL ftrace.bdi_dirty_ratelimit.md0.task_ratelimit
> 10649 ~ 1% -100.0% 0 ~ 0% TOTAL ftrace.bdi_dirty_ratelimit.md0.dirty_ratelimit
> 17409 ~ 2% -100.0% 0 ~ 0% TOTAL ftrace.bdi_dirty_ratelimit.md0.dirty_rate
> 13787 ~ 1% -100.0% 0 ~ 0% TOTAL ftrace.bdi_dirty_ratelimit.md0.awrite_bw
> 14464 ~ 1% -100.0% 0 ~ 0% TOTAL ftrace.bdi_dirty_ratelimit.md0.write_bw
> 65322936 ~ 1% -100.0% 0 ~ 0% TOTAL ftrace.bdi_dirty_ratelimit.md0.time
> 31 ~ 3% -100.0% 0 ~ 0% TOTAL ftrace.balance_dirty_pages.md0.dirtied_pause
> 31 ~ 3% -100.0% 0 ~ 0% TOTAL ftrace.balance_dirty_pages.md0.dirtied
> 14350 ~ 1% -100.0% 0 ~ 0% TOTAL ftrace.balance_dirty_pages.md0.task_ratelimit
> 10633 ~ 1% -100.0% 0 ~ 0% TOTAL ftrace.balance_dirty_pages.md0.dirty_ratelimit
> 13720 ~ 1% -100.0% 0 ~ 0% TOTAL ftrace.balance_dirty_pages.md0.bdi_dirty
> 15220 ~ 1% -100.0% 0 ~ 0% TOTAL ftrace.balance_dirty_pages.md0.bdi_setpoint
> 13720 ~ 1% -100.0% 0 ~ 0% TOTAL ftrace.balance_dirty_pages.md0.dirty
> 15251 ~ 1% -100.0% 0 ~ 0% TOTAL ftrace.balance_dirty_pages.md0.setpoint
> 17638 ~ 1% -100.0% 0 ~ 0% TOTAL ftrace.balance_dirty_pages.md0.limit
> 36455 ~ 0% +221.6% 117256 ~ 0% TOTAL proc-vmstat.unevictable_pgs_culled
> 986338 ~ 0% -55.5% 438629 ~ 0% TOTAL meminfo.Unevictable
> 246584 ~ 0% -55.5% 109657 ~ 0% TOTAL proc-vmstat.nr_unevictable
> 84787 ~20% -47.1% 44851 ~10% TOTAL meminfo.DirectMap4k
> 11266582 ~ 1% -38.7% 6903348 ~ 0% TOTAL ftrace.writeback_single_inode.md0.index
> 10996 ~ 9% -25.9% 8143 ~14% TOTAL meminfo.AnonHugePages
> 1127 ~ 2% +29.8% 1462 ~ 1% TOTAL slabinfo.ftrace_event_file.num_objs
> 1127 ~ 2% +29.8% 1462 ~ 1% TOTAL slabinfo.ftrace_event_file.active_objs
> 1.47 ~12% -20.7% 1.17 ~ 7% TOTAL perf-profile.cpu-cycles.do_get_write_access.jbd2_journal_get_write_access.__ext4_journal_get_write_access.ext4_reserve_inode_write.ext4_mark_inode_dirty
> 15826442 ~ 0% -15.5% 13381229 ~ 0% TOTAL ftrace.global_dirty_state.written
> 15944180 ~ 0% -15.3% 13502162 ~ 0% TOTAL ftrace.global_dirty_state.dirtied
> 2906 ~ 1% +14.0% 3314 ~ 1% TOTAL slabinfo.shared_policy_node.active_objs
> 2906 ~ 1% +14.0% 3314 ~ 1% TOTAL slabinfo.shared_policy_node.num_objs
> 348891 ~ 0% +14.4% 399046 ~ 0% TOTAL ftrace.global_dirty_state.limit
> 394528 ~ 0% +11.8% 441163 ~ 0% TOTAL meminfo.MemFree
> 98671 ~ 0% +11.8% 110310 ~ 0% TOTAL proc-vmstat.nr_free_pages
> 400850 ~ 0% +11.0% 444838 ~ 0% TOTAL vmstat.memory.free
> 108208 ~ 0% +11.0% 120084 ~ 0% TOTAL proc-vmstat.nr_dirty
> 432868 ~ 0% +11.0% 480357 ~ 0% TOTAL meminfo.Dirty
> 79363 ~ 1% -8.9% 72291 ~ 0% TOTAL proc-vmstat.workingset_nodereclaim
> 65550 ~ 2% -9.1% 59612 ~ 1% TOTAL softirqs.SCHED
> 12.06 ~ 5% +530.1% 75.99 ~ 8% TOTAL iostat.sdd.rrqm/s
> 50.78 ~ 5% +530.2% 320.04 ~ 8% TOTAL iostat.sdd.rkB/s
> 12.29 ~ 5% +520.0% 76.21 ~ 7% TOTAL iostat.sde.rrqm/s
> 51.66 ~ 5% +520.9% 320.74 ~ 7% TOTAL iostat.sde.rkB/s
> 49.28 ~ 6% +542.9% 316.82 ~ 7% TOTAL iostat.sdc.rkB/s
> 53.30 ~ 6% +498.9% 319.26 ~ 7% TOTAL iostat.sdf.rkB/s
> 12.62 ~ 5% +496.4% 75.27 ~ 6% TOTAL iostat.sdg.rrqm/s
> 53.21 ~ 5% +496.2% 317.20 ~ 6% TOTAL iostat.sdg.rkB/s
> 12.37 ~ 4% +509.6% 75.41 ~ 7% TOTAL iostat.sdh.rrqm/s
> 51.89 ~ 4% +511.8% 317.47 ~ 7% TOTAL iostat.sdh.rkB/s
> 12.60 ~ 5% +500.9% 75.73 ~ 8% TOTAL iostat.sdi.rrqm/s
> 52.95 ~ 5% +502.2% 318.87 ~ 8% TOTAL iostat.sdi.rkB/s
> 12.88 ~ 5% +491.9% 76.26 ~ 8% TOTAL iostat.sdj.rrqm/s
> 54.51 ~ 5% +490.0% 321.59 ~ 8% TOTAL iostat.sdj.rkB/s
> 12.69 ~ 6% +501.6% 76.35 ~ 8% TOTAL iostat.sdk.rrqm/s
> 53.90 ~ 6% +497.5% 322.02 ~ 8% TOTAL iostat.sdk.rkB/s
> 12.69 ~ 5% +501.4% 76.33 ~ 8% TOTAL iostat.sdl.rrqm/s
> 53.93 ~ 5% +497.4% 322.15 ~ 8% TOTAL iostat.sdl.rkB/s
> 50.24 ~ 4% +536.4% 319.69 ~ 8% TOTAL iostat.sdb.rkB/s
> 11.94 ~ 4% +535.0% 75.84 ~ 8% TOTAL iostat.sdb.rrqm/s
> 11.70 ~ 6% +542.7% 75.21 ~ 7% TOTAL iostat.sdc.rrqm/s
> 12.68 ~ 7% +498.0% 75.82 ~ 7% TOTAL iostat.sdf.rrqm/s
> 332 ~ 0% +41.9% 471 ~ 0% TOTAL iostat.md0.avgrq-sz
> 3616709 ~ 1% -28.9% 2573038 ~ 1% TOTAL perf-stat.context-switches
> 51043 ~ 1% +10.4% 56342 ~ 3% TOTAL perf-stat.cpu-migrations
> 6.45e+11 ~ 0% -6.7% 6.021e+11 ~ 0% TOTAL perf-stat.dTLB-loads
> 758 ~ 0% -5.9% 713 ~ 0% TOTAL iostat.md0.w/s
>
> Legend:
> ~XX% - stddev percent
> [+-]XX% - change percent
>
>
>
>
> Disclaimer:
> Results have been estimated based on internal Intel analysis and are provided
> for informational purposes only. Any difference in system hardware or software
> design or configuration may affect actual performance.
>
> Thanks,
> Jet
>
> mdadm -q --create /dev/md0 --chunk=256 --level=raid5 --raid-devices=11 --force --assume-clean /dev/sdb1 /dev/sdc1 /dev/sdd1 /dev/sde1 /dev/sdf1 /dev/sdg1 /dev/sdh1 /dev/sdi1 /dev/sdj1 /dev/sdk1 /dev/sdl1
> echo 1 > /sys/kernel/debug/tracing/events/writeback/balance_dirty_pages/enable
> echo 1 > /sys/kernel/debug/tracing/events/writeback/bdi_dirty_ratelimit/enable
> echo 1 > /sys/kernel/debug/tracing/events/writeback/global_dirty_state/enable
> echo 1 > /sys/kernel/debug/tracing/events/writeback/writeback_single_inode/enable
> mkfs -t ext4 -q /dev/md0
> mount -t ext4 /dev/md0 /fs/md0
> dd if=/dev/zero of=/fs/md0/zero-1 status=none &
> sleep 600
> killall -9 dd
>
prev parent reply other threads:[~2014-05-29 6:00 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-29 4:55 [ARM] f9ff907b0af: -100.0% slabinfo.cfq_queue.num_objs Jet Chen
2014-05-29 6:00 ` Fengguang Wu [this message]
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=20140529060012.GF7412@localhost \
--to=fengguang.wu@intel.com \
--cc=lkp@lists.01.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.