From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gregory Haskins Subject: Re: [LTP] LTP RT Tests (Cyclic, rt-migrate, etc) Date: Mon, 20 Jul 2009 10:52:32 -0400 Message-ID: <4A6484B0.3050009@novell.com> References: <1224509719.5152.116.camel@subratamodak.linux.ibm.com> <1246890698.4887.24.camel@subratamodak.linux.ibm.com> <1247076833.5405.48.camel@subratamodak.linux.ibm.com> <200907102200.12804.sripathik@in.ibm.com> <1247244446.5040.0.camel@subratamodak.linux.ibm.com> <4A57884C.10401@us.ibm.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="------------enig67FB8764C4CDB6A6E1270BBB" Cc: Darren Hart , subrata@linux.vnet.ibm.com, Sripathi Kodi , Gowrishankar , ltp-list , linux-rt-users , Clark Williams To: Steven Rostedt Return-path: Received: from victor.provo.novell.com ([137.65.250.26]:39872 "EHLO victor.provo.novell.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751379AbZGTOwo (ORCPT ); Mon, 20 Jul 2009 10:52:44 -0400 In-Reply-To: Sender: linux-rt-users-owner@vger.kernel.org List-ID: This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enig67FB8764C4CDB6A6E1270BBB Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Steven Rostedt wrote: > On Fri, 10 Jul 2009, Darren Hart wrote: > > =20 >> Subrata Modak wrote: >> =20 >>> On Fri, 2009-07-10 at 22:00 +0530, Sripathi Kodi wrote: >>> =20 >>>> On Wednesday 08 July 2009 23:43:53 Subrata Modak wrote: >>>> =20 >>>>> Darren/Sri/Gowri, >>>>> >>>>> Where do you want me to put this exactly inside the RT tree ? >>>>> =20 >>>> Hi Subrata, >>>> >>>> Going by how the tests are organized currently, I think this should = go >>>> into it's own directory under testcases/realtime/func. We will need = to >>>> add a makefile to it. Are you looking at us to help you with this? >>>> =20 >>> Correct. Please send me a patch which integrates it into RT tests bui= ld, >>> install & run. >>> =20 >> Just got back from a week vacation and am burning through mail as fast= as I >> can :-) Haven't had a look yet, but does this test use librttest.h? I= suspect >> not. We'll need to adapt it to run within the existing ltp real-time = testing >> framework, which includes things like buffered output as well as mlock= ing >> support. >> >> Lastly, I'm not sure this test does anything effectively different tha= n >> prio-wake, already in the tree. Just to add to Steven's comments below: At the time that rt-migrate was written, LTP and others lacked sufficient resolution in their testing to reliably find the type of problem that rt-migratate can pinpoint quickly. IIRC, "football" was potentially capable of finding these types of scheduler bugs, but it often failed to find it at all, or it took 24h+ of runtime to find it. Steven's test could find it in under a second or two. And, as Steven mentions below, rt-migrate is additionally designed to look at the top N prio tasks (where N =3D cpu-co= unt) That said, I am not familiar with "prio-wake" so I am not sure if its new or if it has direct overlap with Steven's test or not. >> My other concerns with the test are its >> explicit 1ms preemption criteria (as Steven described it anyway). We = are >> trying to move away from criteria being inherent in measurement tests,= and 1 >> ms seems like an awfully long priority inversion to be an acceptable c= riteria >> to many users. >> >> Steven, am I missing something conceptually here? >> =20 > > Hmm, I missed this email, sorry for the late reply. > > What does prio-wake do? > > This test is what I used to develop the rt scheduler in mainline (as we= ll=20 > as in -rt). It wakes up N+1 tasks with lowering real time priorities. = > Where N is the number of CPUs in the system. Then it makes sure that th= e=20 > these tasks spread out across the CPUs. Most tests just test the highes= t=20 > priority task in the system. But those tests usually miss the second=20 > highest prio task in the system. If you have a second highest prio task= in=20 > the system and a CPU is available to run, then it should run on that CP= U.=20 > But what happens is that it can wait to be migrated and can take millis= ecs=20 > to wake up. > > This test makes sure that all the high prio tasks that are in the runni= ng=20 > state are actually running on a CPU if it can. > > Make sense? > > (BTW, current -rt and mainline now fail this test :-? ) > > -- Steve > > -- > To unsubscribe from this list: send the line "unsubscribe linux-rt-user= s" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > =20 --------------enig67FB8764C4CDB6A6E1270BBB Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.11 (Darwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAkpkhLAACgkQlOSOBdgZUxkhqwCeNiFiUo24BVg7d5UGLepcrqVE NykAni5XgpvTkRpq9SBUHNXnCwEwRt+V =dQlA -----END PGP SIGNATURE----- --------------enig67FB8764C4CDB6A6E1270BBB--