From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:41981) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aY19d-00054F-Lc for qemu-devel@nongnu.org; Mon, 22 Feb 2016 19:48:42 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1aY19Z-0005lN-M4 for qemu-devel@nongnu.org; Mon, 22 Feb 2016 19:48:41 -0500 Received: from ozlabs.org ([103.22.144.67]:51494) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aY19Y-0005kb-SV for qemu-devel@nongnu.org; Mon, 22 Feb 2016 19:48:37 -0500 Date: Tue, 23 Feb 2016 10:28:30 +1100 From: David Gibson Message-ID: <20160222232830.GP2808@voom.fritz.box> References: <1456117285-22273-1-git-send-email-bharata@linux.vnet.ibm.com> <56CB2A09.9040102@suse.de> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="xXygN3QAmJYWdGtb" Content-Disposition: inline In-Reply-To: <56CB2A09.9040102@suse.de> Subject: Re: [Qemu-devel] [RFC PATCH v0 0/8] cpu-package hotplug List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Andreas =?iso-8859-1?Q?F=E4rber?= Cc: ehabkost@redhat.com, aik@ozlabs.ru, armbru@redhat.com, Bharata B Rao , agraf@suse.de, qemu-devel@nongnu.org, pbonzini@redhat.com, imammedo@redhat.com --xXygN3QAmJYWdGtb Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Feb 22, 2016 at 04:32:25PM +0100, Andreas F=E4rber wrote: > Hi Bharata, >=20 > Am 22.02.2016 um 06:01 schrieb Bharata B Rao: > > This is an attempt to implement David Gibson's RFC that was posted at > > https://lists.gnu.org/archive/html/qemu-ppc/2016-02/msg00000.html > > I am not sure if I have followed all the aspects of the RFC fully, but = we > > can make changes going forward. >=20 > I am not familiar with David's RFC beyond what was portrayed on the KVM > call - this is not what we discussed on the call and I don't like it. >=20 > Further, your commits are pretty cryptic to me. Please improve your > commit messages. >=20 > For example, you add a cpu_type field and you assign it the value > TYPE_POWERPC_CPU. That's not the user-chosen CPU type then, it's a base > CPU type that cannot be instantiated. Either name it cpu_base_type or > fill it in with proper values in one patch - that patch on its own does > not create value and does not explain your claim: > "Storing CPU typename in MachineState lets us to create CPU threads > for all architectures in uniform manner from arch-neutral code." > I'm pretty sure that CPU threads cannot be created from that type, as it > would run into an assertion. >=20 > Next, you make a functionally correct refactoring of cpu_generic_init(), > but I don't understand why you duplicate that code. cpu_foo_init() still > expects things to be realized, so instead of realizing once in a central > place you do it in nine different places. Had you touched all helper > functions we might be able to move that to three places, once for > softmmu, once or twice for linux-user and once for bsd-user. But I > rather get the feeling that you misunderstand those legacy helper > functions, they're for -cpu handling and not to my knowledge used for > cpu-add at all. You should not be using them and then won't need to > touch them in this way. By using them in your supposedly QOM code you > are hiding an object_new() call inside deep layers of helper functions > instead of using QOM native functions such as object_initialize(), > object_new() and object_property_set*(). >=20 > Is "CPU package" some IBM sPAPR term? It is new to me and does not match > -smp precedence, so I really don't think we should be forcing that term > on all architectures for no good reason. No, it's not an spapr term. *By design* it doesn't match -smp precedence, as noted elsewhere the point is to not lock the unit of hotplug granularity to a fixed level of the -smp heirarchy, because there doesn't seem to be a level there we can pick which works for all platforms. --=20 David Gibson | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_ | _way_ _around_! http://www.ozlabs.org/~dgibson --xXygN3QAmJYWdGtb Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQIcBAEBAgAGBQJWy5meAAoJEGw4ysog2bOSs7YP/jxLYryo9QX7O/aFmI24jTJQ z7aRj2ZHqqpaWr39QL0WSSt+bSuF8pPNUbr1FDWnqqqxG+0bW+3Y022dtYWM9Yb6 r4v+Cf4WPkb1G7lzQxbgKZ8hxkApgy4b9NN2XUVjcIujcw8pMOzLeK+zAAhGmIEh hkQtCeoFAd2tzLS9JzDJWH4INQ/r/HTp4PqibWENBfXj4YljcW9KdyBnsW5O5Q/K OsX5scsOk3pC5KVWhKYX4YxnsumNp+6eX9gLKtiDy5P1sLLhOGvpTnFkRPBRmut1 Oy4/Ye42EX4GY9rCZGTWZHNUesPpMhG/T3JCaJAEIqoQRy+DPxdabP580Ydi0wyk oOSqCUPV6pwvEVwqotXtvHC6x4QHysaPXRCbcALnlfvmmehchPsgnSt7RAy9Kc3p FTO7TZdWbKFgzoMqb11u7SWkD2xu8By1Ly1uk+xsjdfzh/CTj2Z3EQGT6P/8yu+A uvYdr/sVJwKPAsYLS335fTPsOSNsp+1rpe7e1nKOn3ZPWRKnd9+Uid8N7o4tCmDx K/LDcdL6RSL3Ch3HIT+69lKilRz3uvJDhrOhRf1acwWKOzyivknbgsuicXFehtq7 VaLJq3APNag6hutyM5ZLM4/c6Q2ao94LklSDlliK82s3AmdEtr6VH2U5ME1tTBOp yW7rDL4XpbXU9op9CKxT =mQE7 -----END PGP SIGNATURE----- --xXygN3QAmJYWdGtb--