From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dario Faggioli Subject: Re: [PATCH v1 2/3] sched_credit2.c : runqueue_per_core code Date: Thu, 12 Mar 2015 17:07:44 +0000 Message-ID: <1426180062.7023.71.camel@citrix.com> References: <20150312145736.GA8478@gmail.com> <5501CB270200007800069467@mail.emea.novell.com> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============4537247544990500238==" Return-path: In-Reply-To: <5501CB270200007800069467@mail.emea.novell.com> Content-Language: en-US List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: "JBeulich@suse.com" Cc: "xen-devel@lists.xen.org" , George Dunlap , "uma.sharma523@gmail.com" List-Id: xen-devel@lists.xenproject.org --===============4537247544990500238== Content-Language: en-US Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-VKOKcylLZPv66///DSky" --=-VKOKcylLZPv66///DSky Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Thu, 2015-03-12 at 16:21 +0000, Jan Beulich wrote: > >>> On 12.03.15 at 15:57, wrote: > > @@ -1940,10 +1946,14 @@ static void init_pcpu(const struct scheduler *o= ps, int cpu) > > =20 > > /* Figure out which runqueue to put it in */ > > /* NB: cpu 0 doesn't get a STARTING callback, so we hard-code it t= o runqueue 0. */ > > - if ( cpu =3D=3D 0 ) > > - rqi =3D 0; > > + if ( opt_credit2_runqueue =3D=3D CREDIT2_OPT_RUNQUEUE_SOCKET ) > > + { > > + rqi =3D (cpu) ? cpu_to_socket(cpu) : boot_cpu_to_socket(); > > + } > > else > > - rqi =3D cpu_to_socket(cpu); > > + { > > + rqi =3D (cpu) ? cpu_to_core(cpu) : boot_cpu_to_core(); > > + } >=20 > Rather than extending the bad assumption of CPU 0 being the boot > CPU (What if it gets offlined and this or another one onlined back > as CPU 0?), can't you find a way to avoid depending on the numeric > value of "cpu"? >=20 BTW, while we're here, can we really offline CPU#0? I haven't played much with CPU on/offlining, so sorry if I'm asking something obvious... But I found this, which looks to be working: http://lxr.free-electrons.com/source/Documentation/ABI/testing/sysfs-device= s-system-xen_cpu Which explicitly says "except cpu0 due to several logic restrictions and assumptions." Note that I'm not asking because I think we shouldn't try to make this more abstract, I'm much rather looking for a way to test the case of the boot cpu being offlined. :-) Thanks and Regards, Dario --=-VKOKcylLZPv66///DSky 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 v2 iEYEABECAAYFAlUBx94ACgkQk4XaBE3IOsR0+wCfZIIXiB2SiRIS988s+ZX6rS+F zpUAn2VOh6SkRkMDA0swVx9GSLFrLtOd =MxVG -----END PGP SIGNATURE----- --=-VKOKcylLZPv66///DSky-- --===============4537247544990500238== 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.xen.org http://lists.xen.org/xen-devel --===============4537247544990500238==--