From: "Doug Smythies" <dsmythies@telus.net>
To: "'Peter Zijlstra'" <a.p.zijlstra@chello.nl>,
"'Jonathan Nieder'" <jrnieder@gmail.com>
Cc: "'Greg Kroah-Hartman'" <gregkh@linuxfoundation.org>,
<linux-kernel@vger.kernel.org>, <stable@vger.kernel.org>,
<torvalds@linux-foundation.org>, <akpm@linux-foundation.org>,
<alan@lxorguk.ukuu.org.uk>, "'Ingo Molnar'" <mingo@kernel.org>,
"Doug Smythies" <dsmythies@telus.net>
Subject: RE: [ 11/37] sched/nohz: Rewrite and fix load-avg computation -- again
Date: Sat, 21 Jul 2012 09:02:21 -0700 [thread overview]
Message-ID: <000001cd675a$3bda1d00$b38e5700$@net> (raw)
In-Reply-To: <1342805139.2583.46.camel@twins>
[-- Attachment #1: Type: text/plain, Size: 2168 bytes --]
> On 2012.07.20 10:26 -0700, Peter Zijlstra wrote:
>> On Fri, 2012-07-20 at 12:13 -0500, Jonathan Nieder wrote:
>> > Peter Zijlstra wrote:
>> >> On Tue, 2012-07-17 at 19:16 -0500, Jonathan Nieder wrote:
>>
>> >> I'm thrilled to see this regression fix for stable@, but are we really
>> >> really sure that it won't cause new regressions?
>> >
>> > Doug Smythies ran a ~68 hour test on it, running various synthetic
loads
>> > of various frequencies against it and comparing the reported load
>> > averages against the expected values and found it to be 'good'.
>> >
>> > This doesn't guarantee we won't find more 'interesting' problems in
>> > there, but it does give me fair confidence in it.
>>
>> Yeah, that sounds good. Very nice to hear.
>>
>> Is the code to generate the synthetic loads and expected results
>> somewhere easy to find (like LTP or tools/testing) to make it easier
>> to keep this code working well in the future?
> /me finds Doug isn't actually on the CC, /me fixes.
Thanks.
> Doug had this web-page with all his testing activities, graphs and code
> etc..
> http://www.smythies.com/~doug/network/load_average/
> Seems to still work.
Those web pages will be there for a long time (years).
> Last time I tried his scripts they weren't very user friendly, and afaik
> he's making the pretty graphs 'manually'. But whatever he's got is there
> I think.
Yes, pretty graphs were manually done.
Yes, scripts lack user friendliness, but everything I used is posted.
> If someone wants to take it and make it pretty and 'usable' for people
> in a hurry I'm sure Doug wouldn't mind.
Someday I might make it more useable myself. Peter's "consume.c" is
very useful also. (I haven't posted it in my web notes yet, but I will.)
The 68 hour test was just one of the tests runs, albeit the main one.
Among the other tests was what I call the "Charles Wang" scenario,
high frequency high loads.
The only operating scenario of potential concern was around higher
loads higher number of processes, where the reported load average
Was a little low and worse than the same conditions without this patch,
although still pretty good (graph attached).
[-- Attachment #2: freq_5proc.png --]
[-- Type: image/png, Size: 30971 bytes --]
next prev parent reply other threads:[~2012-07-21 16:03 UTC|newest]
Thread overview: 67+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-07-17 23:24 [ 00/23] 3.0.38-stable review Greg KH
2012-07-18 0:02 ` [ 01/37] Remove easily user-triggerable BUG from generic_setlease Greg Kroah-Hartman
2012-07-18 0:02 ` [ 02/37] media: cx231xx: dont DMA to random addresses Greg Kroah-Hartman
2012-07-18 0:02 ` [ 03/37] hwmon: (it87) Preserve configuration register bits on init Greg Kroah-Hartman
2012-07-18 0:02 ` [ 04/37] block: fix infinite loop in __getblk_slow Greg Kroah-Hartman
2012-07-18 0:02 ` [ 05/37] media: dvb-core: Release semaphore on error path dvb_register_device() Greg Kroah-Hartman
2012-07-18 0:02 ` [ 06/37] mtd: nandsim: dont open code a do_div helper Greg Kroah-Hartman
2012-07-18 0:02 ` [ 07/37] md/raid1: fix use-after-free bug in RAID1 data-check code Greg Kroah-Hartman
2012-07-18 0:02 ` [ 08/37] ARM: SAMSUNG: fix race in s3c_adc_start for ADC Greg Kroah-Hartman
2012-07-18 0:02 ` [ 09/37] ACPICA: Fix possible fault in return package object repair code Greg Kroah-Hartman
2012-07-18 0:02 ` [ 10/37] cpufreq / ACPI: Fix not loading acpi-cpufreq driver regression Greg Kroah-Hartman
2012-07-18 0:02 ` [ 11/37] sched/nohz: Rewrite and fix load-avg computation -- again Greg Kroah-Hartman
2012-07-18 0:16 ` Jonathan Nieder
2012-07-20 17:04 ` Peter Zijlstra
2012-07-20 17:13 ` Jonathan Nieder
2012-07-20 17:25 ` Peter Zijlstra
2012-07-21 16:02 ` Doug Smythies [this message]
2012-07-18 0:02 ` [ 12/37] intel_ips: blacklist HP ProBook laptops Greg Kroah-Hartman
2012-07-18 0:02 ` [ 13/37] fifo: Do not restart open() if it already found a partner Greg Kroah-Hartman
2012-07-18 0:02 ` [ 14/37] rt2x00usb: fix indexes ordering on RX queue kick Greg Kroah-Hartman
2012-07-18 0:02 ` [ 15/37] e1000e: Correct link check logic for 82571 serdes Greg Kroah-Hartman
2012-07-18 0:02 ` [ 16/37] iwlegacy: always monitor for stuck queues Greg Kroah-Hartman
2012-07-18 0:02 ` [ 17/37] iwlegacy: dont mess up the SCD when removing a key Greg Kroah-Hartman
2012-07-18 0:02 ` [ 18/37] rpmsg: fix dependency on initialization order Greg Kroah-Hartman
2012-07-18 0:02 ` [ 19/37] mac80211: destroy assoc_data correctly if assoc fails Greg Kroah-Hartman
2012-07-18 0:02 ` [ 20/37] stmmac: Fix for nfs hang on multiple reboot Greg Kroah-Hartman
2012-07-18 0:02 ` [ 21/37] bonding: debugfs and network namespaces are incompatible Greg Kroah-Hartman
2012-07-18 0:02 ` [ 22/37] bonding: Manage /proc/net/bonding/ entries from the netdev events Greg Kroah-Hartman
2012-07-18 0:03 ` [ 23/37] Input: bcm5974 - Add support for 2012 MacBook Pro Retina Greg Kroah-Hartman
2012-07-18 0:03 ` [ 24/37] Input: xpad - handle all variations of Mad Catz Beat Pad Greg Kroah-Hartman
2012-07-18 0:03 ` [ 25/37] Input: xpad - add signature for Razer Onza Tournament Edition Greg Kroah-Hartman
2012-07-18 0:03 ` [ 26/37] Input: xpad - add Andamiro Pump It Up pad Greg Kroah-Hartman
2012-07-18 0:03 ` [ 27/37] HID: add support for 2012 MacBook Pro Retina Greg Kroah-Hartman
2012-07-18 0:03 ` [ 28/37] clk: Check parent for NULL in clk_change_rate Greg Kroah-Hartman
2012-07-18 0:03 ` [ 29/37] cfg80211: check iface combinations only when iface is running Greg Kroah-Hartman
2012-07-18 0:03 ` [ 30/37] hrtimer: Provide clock_was_set_delayed() Greg Kroah-Hartman
2012-07-18 0:03 ` [ 31/37] timekeeping: Fix leapsecond triggered load spike issue Greg Kroah-Hartman
2012-07-18 0:03 ` [ 32/37] timekeeping: Maintain ktime_t based offsets for hrtimers Greg Kroah-Hartman
2012-07-18 0:03 ` [ 33/37] hrtimers: Move lock held region in hrtimer_interrupt() Greg Kroah-Hartman
2012-07-18 0:03 ` [ 34/37] timekeeping: Provide hrtimer update function Greg Kroah-Hartman
2012-07-18 0:03 ` [ 35/37] hrtimer: Update hrtimer base offsets each hrtimer_interrupt Greg Kroah-Hartman
2012-07-18 0:03 ` [ 37/37] NFC: Export nfc.h to userland Greg Kroah-Hartman
2012-07-18 0:14 ` [ 01/37] Remove easily user-triggerable BUG from generic_setlease Greg KH
2012-07-18 13:36 ` Nick Bowler
2012-07-18 17:58 ` formail doing weird things (was: [ 01/37] Remove easily user-triggerable BUG from generic_setlease) Roland Eggner
2012-07-18 0:11 ` [ 01/23] hwmon: (it87) Preserve configuration register bits on init Greg Kroah-Hartman
2012-07-18 0:11 ` [ 02/23] block: fix infinite loop in __getblk_slow Greg Kroah-Hartman
2012-07-18 0:11 ` [ 03/23] media: dvb-core: Release semaphore on error path dvb_register_device() Greg Kroah-Hartman
2012-07-18 0:11 ` [ 04/23] mtd: nandsim: dont open code a do_div helper Greg Kroah-Hartman
2012-07-18 0:11 ` [ 05/23] ARM: SAMSUNG: fix race in s3c_adc_start for ADC Greg Kroah-Hartman
2012-07-18 0:11 ` [ 06/23] intel_ips: blacklist HP ProBook laptops Greg Kroah-Hartman
2012-07-18 0:12 ` [ 07/23] fifo: Do not restart open() if it already found a partner Greg Kroah-Hartman
2012-07-18 0:12 ` [ 08/23] rt2x00usb: fix indexes ordering on RX queue kick Greg Kroah-Hartman
2012-07-18 0:12 ` [ 09/23] e1000e: Correct link check logic for 82571 serdes Greg Kroah-Hartman
2012-07-18 0:12 ` [ 10/23] Input: xpad - add Andamiro Pump It Up pad Greg Kroah-Hartman
2012-07-18 0:12 ` [ 11/23] tcp: drop SYN+FIN messages Greg Kroah-Hartman
2012-07-18 0:12 ` [ 12/23] cfg80211: check iface combinations only when iface is running Greg Kroah-Hartman
2012-07-18 0:12 ` [ 13/23] ntp: Fix leap-second hrtimer livelock Greg Kroah-Hartman
2012-07-18 0:12 ` [ 14/23] ntp: Correct TAI offset during leap second Greg Kroah-Hartman
2012-07-18 0:12 ` [ 15/23] timekeeping: Fix CLOCK_MONOTONIC inconsistency during leapsecond Greg Kroah-Hartman
2012-07-18 0:12 ` [ 16/23] time: Move common updates to a function Greg Kroah-Hartman
2012-07-18 0:12 ` [ 17/23] hrtimer: Provide clock_was_set_delayed() Greg Kroah-Hartman
2012-07-18 0:12 ` [ 18/23] timekeeping: Fix leapsecond triggered load spike issue Greg Kroah-Hartman
2012-07-18 0:12 ` [ 19/23] timekeeping: Maintain ktime_t based offsets for hrtimers Greg Kroah-Hartman
2012-07-18 0:12 ` [ 20/23] hrtimers: Move lock held region in hrtimer_interrupt() Greg Kroah-Hartman
2012-07-18 0:12 ` [ 21/23] timekeeping: Provide hrtimer update function Greg Kroah-Hartman
2012-07-18 0:12 ` [ 22/23] hrtimer: Update hrtimer base offsets each hrtimer_interrupt Greg Kroah-Hartman
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='000001cd675a$3bda1d00$b38e5700$@net' \
--to=dsmythies@telus.net \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=gregkh@linuxfoundation.org \
--cc=jrnieder@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=stable@vger.kernel.org \
--cc=torvalds@linux-foundation.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 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).