From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:49030)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <thuth@redhat.com>) id 1bmxS9-0005d1-0L
	for qemu-devel@nongnu.org; Thu, 22 Sep 2016 02:25:50 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <thuth@redhat.com>) id 1bmxS4-0002Uc-O6
	for qemu-devel@nongnu.org; Thu, 22 Sep 2016 02:25:47 -0400
Date: Thu, 22 Sep 2016 08:25:38 +0200
From: Thomas Huth <thuth@redhat.com>
Message-ID: <20160922082538.3d1c5895@thh440s>
In-Reply-To: <1474407935.2857.87.camel@kernel.crashing.org>
References: <75632046-c527-8d5f-6ac0-a40c28239579@redhat.com>
	<1474407935.2857.87.camel@kernel.crashing.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Subject: Re: [Qemu-devel] KVM-PR is broken with current QEMU
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel/>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: qemu-ppc@nongnu.org, QEMU Developers <qemu-devel@nongnu.org>, =?UTF-8?B?Q8OpZHJpYw==?= Le Goater <clg@kaod.org>, David Gibson <david@gibson.dropbear.id.au>

On Wed, 21 Sep 2016 07:45:35 +1000
Benjamin Herrenschmidt <benh@kernel.crashing.org> wrote:

> On Tue, 2016-09-20 at 13:44 +0200, Thomas Huth wrote:
> >=20
> > Seems like KVM PR is using the "degraded" ISA variants (without the
> > 1TB
> > segments), but the new POWERPC_MMU_64K flag has not been added to
> > those.
> > Has this been done on purpose, or was this just by accident?
> > I can make KVM PR working again with the following patch:
> >=20
> > diff --git a/target-ppc/cpu-qom.h b/target-ppc/cpu-qom.h
> > index 2864105..36694cb 100644
> > --- a/target-ppc/cpu-qom.h
> > +++ b/target-ppc/cpu-qom.h
> > @@ -81,6 +81,7 @@ enum powerpc_mmu_t {
> > =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 | POWERPC_MMU_AMR | 0x00000003,
> > =C2=A0=C2=A0=C2=A0=C2=A0 /* Architecture 2.06 "degraded" (no 1T segment=
s)=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 */
> > =C2=A0=C2=A0=C2=A0=C2=A0 POWERPC_MMU_2_06a=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 =3D POWERPC_MMU_64 | POWERPC_MMU_AMR
> > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 | POWERPC_MMU_64K
> > =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 | 0x00000003,
> > =C2=A0=C2=A0=C2=A0=C2=A0 /* Architecture 2.07 variant=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0 */
> > =C2=A0=C2=A0=C2=A0=C2=A0 POWERPC_MMU_2_07=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0 =3D POWERPC_MMU_64 | POWERPC_MMU_1TSEG
> > @@ -88,6 +89,7 @@ enum powerpc_mmu_t {
> > =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 | POWERPC_MMU_AMR | 0x00000004,
> > =C2=A0=C2=A0=C2=A0=C2=A0 /* Architecture 2.07 "degraded" (no 1T segment=
s)=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 */
> > =C2=A0=C2=A0=C2=A0=C2=A0 POWERPC_MMU_2_07a=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 =3D POWERPC_MMU_64 | POWERPC_MMU_AMR
> > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 | POWERPC_MMU_64K
> > =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 | 0x00000004,
> > =C2=A0};
> >=20
> > However, not sure whether this is the right fix ... C=C3=A9dric, Ben, a=
ny
> > ideas?
>=20
> Oh I thought I had removed the degraded variants ... Definitely looks
> like an accident. I *think* PR KVM supports 64K pages, no ? If not,
> then we shouldn't enable the flag.. somebody needs to check the kernel.

Yes, it supports 64k pages - but only on POWER8, not on POWER8E or
POWER8NVL yet. I've posted a patch to fix this here:

https://patchwork.ozlabs.org/patch/672841/

 Thomas