Date: Wed, 25 Jan 2023 06:10:14 +1100 From: Ian Smith <smithi@nimnet.asn.au> To: Stavros Filargyropoulos <stav@stav.cc> Cc: freebsd-questions@freebsd.org Subject: Re: Load 0.20 on a freshly installed idle system Message-ID: <9A7F4436-4888-42E2-93B9-7A8BA342877B@nimnet.asn.au> In-Reply-To: <CAGAKwchY_pU5PP0XX2UsUOSf027ZfbdMgd3wXzaKpu=Fi=J2Uw@mail.gmail.com> References: <CAGAKwcjmzz%2B6NMaK2YpjWZPB282RvrKV6zQ%2BSLHsDj3zqXwvGw@mail.gmail.com> <CAFbbPuhX%2BrJFLn8ALVNbCujf1tG3DufVWGSknPkJ%2BKdovFq1xQ@mail.gmail.com> <CAGAKwcgFWukBrB_6SZVAoC4YKyPVhVMGED4aB6Z4E1fwZAO6mg@mail.gmail.com> <CAFbbPuhyJDewvE_a6MQpiGrpUew7EtxZKwt_=RHODNUyCYNduw@mail.gmail.com> <CAGAKwcghHH5t1_dtWPkOCwxPvDra%2BmgNFTQsG65i1_7pzTDkug@mail.gmail.com> <CAFbbPujMDQgo=zvSc5orcT=v3ftrVPXDjZE-D%2BUxt3%2BdOjFmUA@mail.gmail.com> <CAGAKwcid8j3C28wiypO87X9HT9Wq8cfTNHWnOQvcQgAcWrFuLw@mail.gmail.com> <CAGAKwcgZz8gujtd8gPBoC5nKxE00q1Pau6%2BiOUjxBR-UMMNpKw@mail.gmail.com> <1889CB4C-4B11-4178-8446-97DA9E5EC35A@nimnet.asn.au> <CAGAKwchY_pU5PP0XX2UsUOSf027ZfbdMgd3wXzaKpu=Fi=J2Uw@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 23 January 2023 6:31:27 am AEDT, Stavros Filargyropoulos <stav@stav=2Ecc= > wrote: > On Fri, Jan 20, 2023 at 8:44 PM Ian Smith <smithi@nimnet=2Easn=2Eau> > wrote: > > This has been bugging quite a few people for quite a few years: > > > > https://bugs=2Efreebsd=2Eorg/bugzilla/show_bug=2Ecgi?id=3D173541 > Bug is now marked as fixed and closed but no reference to the commit > that fixed it or what the root cause was=2E >=20 > I still see the problem in STABLE where it was supposedly fixed=2E Yes; I updated to 12=2E4-R yesterday (due anyway) assuming 'months ago' sh= ould have been before 12=2E4 left 12-STABLE, but see no change=2E I emailed gbe@ as well, in case he's buried in PR mail=2E > $ uname -a > FreeBSD nas 13=2E1-STABLE FreeBSD 13=2E1-STABLE #0 > stable/13-n253535-4a8af507ebe5: Fri Jan 20 03:21:10 UTC 2023 > root@releng3=2Enyi=2Efreebsd=2Eorg:/usr/obj/usr/src/amd64=2Eamd64/sys/G= ENERIC > amd64 >=20 > last pid: 1545; load averages: 0=2E31, 0=2E11, 0=2E04 > Where do we go from here? I put quite some time exploring this back in 2016, as did Jeremy Chadwick,= so I'd pretty much given up since 0=2E1 or 0=2E2 nowadays is better than 0= =2E6 then, and noone could explain how my magic fix for 9=2E2-STABLE worked= , let alone me=2E I now understand a bit better what Alexander Motin (mav@) was explaining i= n posts #9, #27 and now #46 and #48, and accept the trade-off between idle = accuracy and performance (particularly minimal power consumption on laptops= ) that the current scheme provides=2E That said, it would be great if someone wanted to implement a parallel set= of event callouts specifically for load measurement that added very little= to interrupt rate, and thus power use=2E For example, using periodic rather than one-shot event timers shows low, e= ven 0=2E0 LAs at idle, but by costing 2000 interrupts per second per CPU!, = 8000/s on my i5, adding ~25% idle power consumption=2E cheers, Ian
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?9A7F4436-4888-42E2-93B9-7A8BA342877B>