From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dario Faggioli Subject: Re: [PATCH] xen/sched_credit: Use delay to control scheduling frequency Date: Tue, 20 Dec 2011 11:51:24 +0100 Message-ID: <1324378284.6616.14.camel@Abyss> References: <114495d1ff947737518b.1324332781@wsm-ep-n0> <1324372056.23729.8.camel@zakaz.uk.xensource.com> <1324375765.6616.9.camel@Abyss> <1324376802.2143.154.camel@elijah> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============3414353275376268182==" Return-path: In-Reply-To: <1324376802.2143.154.camel@elijah> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: George Dunlap Cc: George Dunlap , "xen-devel@lists.xensource.com" , Ian Campbell , Jan Beulich , "Lv, Hui" List-Id: xen-devel@lists.xenproject.org --===============3414353275376268182== Content-Type: multipart/signed; micalg="pgp-sha1"; protocol="application/pgp-signature"; boundary="=-kWN+hrKMXvO7JFXpsTxo" --=-kWN+hrKMXvO7JFXpsTxo Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Tue, 2011-12-20 at 10:26 +0000, George Dunlap wrote:=20 > > I think that too, but maybe you can print some kind of warning when it > > happens? >=20 > I guess the consensus is that we should put in some effort to make the > interface more polished. >=20 Exactly! :-) > How about this: >=20 > First, add a ratelimit_us element to csched_priv, just like the current > tslice_ms element. >=20 > When the scheduler comes up (in csched_init), it checks to see if > MICROSECS(sched_ratelimit_us) > MILLISECS(sched_credit_tslice_ms); if > so, it prints a warning, and sets prv->ratelimit_us to > 1000*prv->tslice_ms. >=20 For what it counts, I'm more than fine with just printing some "WARNING: rate-limit > time slice is nonsense", as well as doing as you suggest... Or even with doing both! :-P=20 > In the future, when we implement the domctls to change a scheduler's > ratelimit_us, and tslice_ms, we disallow changes which would violate the > "ratelimit < tslice" rule (returning -EINVAL or something like that). > That makes a lot of sense. Regards, Dario --=20 <> (Raistlin Majere) ------------------------------------------------------------------- Dario Faggioli, http://retis.sssup.it/people/faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK) PhD Candidate, ReTiS Lab, Scuola Superiore Sant'Anna, Pisa (Italy) --=-kWN+hrKMXvO7JFXpsTxo Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) iEYEABECAAYFAk7waKwACgkQk4XaBE3IOsRgbwCfb1trPxijD1sN4lo+FV6N1V5m utoAn2sFfriavHtAbPkBUWC6y/ZqA3ir =i2AM -----END PGP SIGNATURE----- --=-kWN+hrKMXvO7JFXpsTxo-- --===============3414353275376268182== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel --===============3414353275376268182==--