Date: Sun, 29 Apr 2012 16:45:50 +0200 From: Oliver Pinter <oliver.pntr@gmail.com> To: Alexander Motin <mav@freebsd.org> Cc: Davide Italiano <davide@freebsd.org>, Albert Shih <Albert.Shih@obspm.fr>, Luigi Rizzo <rizzo@iet.unipi.it>, freebsd-stable@freebsd.org, Ian Smith <smithi@nimnet.asn.au> Subject: Re: High load event idl. Message-ID: <CAPjTQNFOJUR-%2B%2BcR_0fdL-wQKzxEYy2b9ZTfq7UjqXB2c-FJ9A@mail.gmail.com> In-Reply-To: <4F9D3B1F.2030208@FreeBSD.org> References: <20120427203013.GB60961@pcjas.obspm.fr> <CAPjTQNFsHZQLp8oMwhjkAWLnYZ5mPv9kr9=X5GhqHqExoHM0yw@mail.gmail.com> <20120427213459.GA61125@pcjas.obspm.fr> <4F9B946D.3030607@FreeBSD.org> <CAPjTQNGts290DyjORNfir8_rZ5S_vdog%2BJMEBA9mc2vJhUa=jg@mail.gmail.com> <4F9CCEF2.6050609@FreeBSD.org> <20120429155512.M91148@sola.nimnet.asn.au> <4F9CDE91.1060300@FreeBSD.org> <CAPjTQNF=0Hgq_7BxeK_8o6DRQ%2BUJ_r94Y3PqwF8f_ccDeA_hHQ@mail.gmail.com> <4F9D2F0C.4050501@FreeBSD.org> <CAPjTQNGesMffKZupENy0YWjC8LToutCyHh9uGqd-XKTN24goew@mail.gmail.com> <4F9D3B1F.2030208@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 4/29/12, Alexander Motin <mav@freebsd.org> wrote: > On 04/29/12 15:27, Oliver Pinter wrote: >> http://oliverp.teteny.bme.hu/freebsd/ktr/ > > OK. Now there is no dummynet, but I've found there two more things: > 1. for some reason some acpi_thremal thread seems to consume about > 0.37s of time every 10s. I have no idea what is this. It's not 0.7 load, > but still strange at least. > 2. I suspect another possible synchronization between ehci driver and > loadavg as result of interrupt sharing between HPET timer used for time > events and EHCI USB hardware. Not sure what to do about this. Please > send _verbose_ dmesg to check whether this interrupt sharing is > unavoidable. > >> On 4/29/12, Alexander Motin<mav@freebsd.org> wrote: >>> On 04/29/12 15:04, Oliver Pinter wrote: >>>> Removing dummynet from kernel don't chanage anything, that is releated >>>> to load average. The loadavg hold to 0.70 +/- 0.2. (single user : sh + >>>> top) >>> >>> New ktr dump? >>> >>>> On 4/29/12, Alexander Motin<mav@freebsd.org> wrote: >>>>> On 04/29/12 09:09, Ian Smith wrote: >>>>>> On Sun, 29 Apr 2012 08:17:38 +0300, Alexander Motin wrote: >>>>>> > On 04/29/12 01:53, Oliver Pinter wrote: >>>>>> > > Attached the ktr file. This is on core2duo P9400 cpu ( >>>>>> > > smbios.system.product="HP ProBook 5310m (WD792EA#ABU)" >>>>>> ). >>>>>> The >>>>>> workload >>>>>> > > is only a single user boost: sh + top running, but the >>>>>> load >>>>>> average is >>>>>> > > near 0.5. >>>>>> > >>>>>> > ktr shows no real load there. But it shows that you are >>>>>> using >>>>>> dummynet, that >>>>>> > schedules its runs on every hardclock tick. I believe that >>>>>> load >>>>>> you >>>>>> see is >>>>>> > the result or synchronization between dummynet calls and >>>>>> loadvg >>>>>> sampling, >>>>>> > both of which called from hardclock. I think removing >>>>>> dummynet >>>>>> from >>>>>> equation, >>>>>> > should hide this problem and also reduce you laptops power >>>>>> consumption. >>>>>> > >>>>>> > What's about fixing this, it is loadavg sampling algorithm >>>>>> that >>>>>> should be >>>>>> > changed. Fixing dummynet to not run on every hardclock tick >>>>>> would >>>>>> also be >>>>>> > great. >>>>>> >>>>>> Wading in out of my depth, and copying Luigi in case he misses it .. >>>>>> but >>>>>> even back in the olden days when HZ defaulted to 100, one was advised >>>>>> to >>>>>> use HZ>= 1000 for smooth dummynet traffic shaping dispatch >>>>>> scheduling. >>>>>> >>>>>> I wonder, with the newer clocks and timers, whether there is another >>>>>> clock that could be used for dummynet scheduling, that would not have >>>>>> this effect (even if largely cosmetic?) on load average calculation? >>>>> >>>>> First of all, the easiest solution would be to make dummynet to >>>>> schedule >>>>> callout not automatically, but on first queued packet. I believe that >>>>> in >>>>> case of laptop the queue should be empty most of time and the callout >>>>> calls are completely useless there. Luigi promised to look on this >>>>> once. >>>>> >>>>> What's about better precision/removing synchronization -- there is >>>>> starting GSoC project now (by davide@) to rewrite callout(9) subsystem >>>>> to use better precision allowed by new timer drivers. While now it is >>>>> possible to get raw access to additional timer hardware available on >>>>> some systems, I don't think it is a good idea. http://freebsd.1045724.n5.nabble.com/amd64-160419-acpi-thermal-kernel-thread-high-CPU-usage-td4765266.html but this "high cpu load" is gone, releated to acpi_thermal in 2011 september > > > -- > Alexander Motin >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAPjTQNFOJUR-%2B%2BcR_0fdL-wQKzxEYy2b9ZTfq7UjqXB2c-FJ9A>