From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:33689) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1YTAb3-0004CJ-Fd for qemu-devel@nongnu.org; Wed, 04 Mar 2015 09:48:26 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1YTAb0-0004V6-8Y for qemu-devel@nongnu.org; Wed, 04 Mar 2015 09:48:25 -0500 Message-ID: <54F71B34.6070206@suse.de> Date: Wed, 04 Mar 2015 15:48:20 +0100 From: Alexander Graf MIME-Version: 1.0 References: <1425432686-29851-1-git-send-email-aik@ozlabs.ru> <54F6FA53.5080106@suse.de> <54F71336.60806@suse.de> In-Reply-To: <54F71336.60806@suse.de> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH] target-ppc: Add versions to server CPU descriptions List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: =?windows-1252?Q?Andreas_F=E4rber?= , Alexey Kardashevskiy , qemu-devel@nongnu.org Cc: qemu-ppc@nongnu.org On 04.03.15 15:14, Andreas F=E4rber wrote: > Alex, >=20 > Am 04.03.2015 um 13:28 schrieb Alexander Graf: >> On 04.03.15 02:31, Alexey Kardashevskiy wrote: >>> 5b79b1c "target-ppc: Create versionless CPU class per family if KVM" = added >>> a dynamic CPU class registration with the name of the CPU family whic= h >>> QEMU is running on. For example, this allowed specifying "-cpu POWER7= " >>> on every version of POWER7 machine, not just the one which POWER7 was >>> an alias of. I.e. before 5b79b1c, "-cpu POWER7" would not work on rea= l >>> POWER7 2.1 and would work on POWER7 2.3 only. The same story for POWE= R8. >>> >>> However that patch broke POWER5+ support as POWER5+ CPU uses the same >>> name as the CPU class so dynamic registering of the POWER5+ class fai= led. >>> >>> This redefines POWER5+ server CPUs by adding a version to them and ad= ding >>> an alias for TCG case. KVM will use dynamically registered CPUs. >>> >>> While we are here, do the same for 970 CPU. >>> >>> Signed-off-by: Alexey Kardashevskiy >> >> Thanks, applied to ppc-next. >=20 > As Alexey predicted, I object. >=20 > The 970 part looks good and could be applied immediately if it were a > separate patch. >=20 > But the POWER5+ part I have my doubts about: Was there really a v0.0??? IIUC most IBM POWER PVRs are regular in that their lower 16 bits are major/minor. The PVR we have as "POWER5+" has those bits as 0, thus v0.0 is the correct translation of that. Whether that CPU ever existed is a different question and arguably out of scope for the patch. > Others start with v1.0 and I have: >=20 > revision : 2.1 (pvr 003b 0201) I wouldn't object to removing the v0.0 version altogether and just make POWER5+ be an alias to v2.1 instead. But that's something for a follow-up patch ;). Alex