Date: Mon, 9 Nov 2015 01:31:22 +0000 From: Rasool Al-Saadi <ralsaadi@swin.edu.au> To: Bruce Evans <brde@optusnet.com.au>, Hans Petter Selasky <hps@selasky.org> Cc: "freebsd-net@freebsd.org" <freebsd-net@freebsd.org> Subject: RE: Timing issue with Dummynet on high kernel timer interrupt Message-ID: <6545444AE21C2749939E637E56594CEA3C0E4D6B@gsp-ex02.ds.swin.edu.au> In-Reply-To: <20151107223816.D1837@besplex.bde.org> References: <6545444AE21C2749939E637E56594CEA3C0DCCC4@gsp-ex02.ds.swin.edu.au> <5638B7B5.3030802@selasky.org> <6545444AE21C2749939E637E56594CEA3C0DE7FF@gsp-ex02.ds.swin.edu.au> <563B2703.5080402@selasky.org> <6545444AE21C2749939E637E56594CEA3C0E0BD9@gsp-ex02.ds.swin.edu.au> <563C6864.2090907@selasky.org> <CA%2BhQ2%2Bhm2z0MkB-8w5xJM7%2Biz13r_ZjwmpZBnb30_D_48gaf-w@mail.gmail.com> <563C786C.1050305@selasky.org> <CA%2BhQ2%2Bj0WiGgzV119M1ZQiXP5Z7fq7deVxDPkOhvTc7hpTETKw@mail.gmail.com> <563CC186.9000807@selasky.org> <6545444AE21C2749939E637E56594CEA3C0E1B79@gsp-ex02.ds.swin.edu.au> <20151107162915.A893@besplex.bde.org> <563DD6B2.8010607@selasky.org> <20151107223816.D1837@besplex.bde.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Saturday, 7 November 2015 10:51 PM, Bruce Evans wrote:
> On Sat, 7 Nov 2015, Hans Petter Selasky wrote:
>=20
> > On 11/07/15 07:19, Bruce Evans wrote:
> >> I don't know if CTLFLAG_RWTUN with SYSCTL_PROC() actually works for
> >> all types, but it is used with CTLFLAG_STRING for kern.corefile.
> >> This is another bogus undocumented tunable. It is better documented
> >> as a sysctl than most since it is old so it is documented in
> >> sysctl(8). It is not documented as a tunable there of course. It is
> >> also not documented as a tunable in core(5).
> >
> > SYSCTL_PROC() works with RWTUN from /boot/loader.conf, as long as the
> > procedure callback handles early calls during boot.
>=20
> Hmm, that would be messy if it wants to do hardware stuff like
> sysctl_kern_eventtimer_periodic() does.
>=20
> Timercounters only have 1 MI tunable
> (sysctl_kern_timecounter_adjprecision). It has a (cold) check to avoid d=
oing
> too much.
>=20
> Bruce
Thanks a lot all for your help!
The results now reasonable when kern.eventtimer.periodic=3D1. sysctl kern.t=
imecounter.alloweddeviation=3D0 does not make deference (at least in my tes=
t) .
Here is RTT vs Time graph when Hz=3D10000, kern.eventtimer.periodic=3D1 and=
kern.timecounter.alloweddeviation=3D{default (5), 0} https://goo.gl/photos=
/ukxTvpBDeoFCFEPH8
The time deference graph https://goo.gl/photos/AAV8oECBpdvowgsY7
=20
So, what the effects of using periodic mode on the system in general? Why t=
he timing issue disappears when eventtimer.periodic=3D0 and C_DIRECT_EXEC =
is removed from callout_reset_sbt() call?
Cheers,
Rasool
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?6545444AE21C2749939E637E56594CEA3C0E4D6B>