From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:45963) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1davW8-0003Oo-Ua for qemu-devel@nongnu.org; Thu, 27 Jul 2017 23:00:46 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1davW4-00017i-Ol for qemu-devel@nongnu.org; Thu, 27 Jul 2017 23:00:44 -0400 Date: Fri, 28 Jul 2017 12:59:13 +1000 From: David Gibson Message-ID: <20170728025913.GA3098@umbus.fritz.box> References: <150100547373.27487.3154210751350595400.stgit@bahia> <150100564746.27487.5195312465666688402.stgit@bahia> <20170726052443.GF8978@umbus.fritz.box> <20170727185042.14889377@bahia.lan> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="X1bOJ3K7DJ5YkBrT" Content-Disposition: inline In-Reply-To: <20170727185042.14889377@bahia.lan> Subject: Re: [Qemu-devel] [for-2.11 PATCH 13/26] qdev: store DeviceState's canonical path to use when unparenting List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Greg Kurz Cc: qemu-devel@nongnu.org, "Michael S. Tsirkin" , Michael Roth , qemu-ppc@nongnu.org, Bharata B Rao , Paolo Bonzini , Daniel Henrique Barboza --X1bOJ3K7DJ5YkBrT Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Thu, Jul 27, 2017 at 06:50:42PM +0200, Greg Kurz wrote: > On Wed, 26 Jul 2017 15:24:43 +1000 > David Gibson wrote: >=20 > > On Tue, Jul 25, 2017 at 08:00:47PM +0200, Greg Kurz wrote: > > > From: Michael Roth > > >=20 > > > device_unparent(dev, ...) is called when a device is unparented, > > > either directly, or as a result of a parent device being > > > finalized, and handles some final cleanup for the device. Part > > > of this includes emiting a DEVICE_DELETED QMP event to notify > > > management, which includes the device's path in the composition > > > tree as provided by object_get_canonical_path(). > > >=20 > > > object_get_canonical_path() assumes the device is still connected > > > to the machine/root container, and will assert otherwise, but > > > in some situations this isn't the case: > > >=20 > > > If the parent is finalized as a result of object_unparent(), it > > > will still be attached to the composition tree at the time any > > > children are unparented as a result of that same call to > > > object_unparent(). However, in some cases, object_unparent() > > > will complete without finalizing the parent device, due to > > > lingering references that won't be released till some time later. > > > One such example is if the parent has MemoryRegion children (which > > > take a ref on their parent), who in turn have AddressSpace's (which > > > take a ref on their regions), since those AddressSpaces get cleaned > > > up asynchronously by the RCU thread. > > >=20 > > > In this case qdev:device_unparent() may be called for a child Device > > > that no longer has a path to the root/machine container, causing > > > object_get_canonical_path() to assert. > > >=20 > > > Fix this by storing the canonical path during realize() so the > > > information will still be available for device_unparent() in such > > > cases. =20 > >=20 > > Hm. I'm no expert on the QOM model, but I'm not sure this is the > > right approach. > >=20 > > I would have thought the right time to emit the DEVICE_DELETED message > > would be when the device leaves the main composition tree, even if it > > could be finalized later. > >=20 > > If we made that the case, does this problem go away? > >=20 >=20 > I'm no expert either and I confess I took this patch simply because it was > in Michael's original patchset. :) >=20 > But according to Michael's answer, it seems that the issue has a broader > scope than just PHB hotplug... Ok. I see Michael has posted this and a couple of other things separately. Let's hope that can get resolved upstream, and rebase this series on top of the result. >=20 > > > Cc: Michael S. Tsirkin > > > Cc: Paolo Bonzini > > > Signed-off-by: Michael Roth > > > Signed-off-by: Greg Kurz > > > --- > > > Changes since RFC: > > > - rebased against ppc-for-2.10 > > > --- > > > hw/core/qdev.c | 15 ++++++++++++--- > > > include/hw/qdev-core.h | 1 + > > > 2 files changed, 13 insertions(+), 3 deletions(-) > > >=20 > > > diff --git a/hw/core/qdev.c b/hw/core/qdev.c > > > index 606ab53c42cd..a64b35c16251 100644 > > > --- a/hw/core/qdev.c > > > +++ b/hw/core/qdev.c > > > @@ -928,6 +928,12 @@ static void device_set_realized(Object *obj, boo= l value, Error **errp) > > > goto post_realize_fail; > > > } > > > =20 > > > + /* always re-initialize since we clean up in device_unparent= () instead > > > + * of unrealize() > > > + */ > > > + g_free(dev->canonical_path); > > > + dev->canonical_path =3D object_get_canonical_path(OBJECT(dev= )); > > > + > > > if (qdev_get_vmsd(dev)) { > > > if (vmstate_register_with_alias_id(dev, -1, qdev_get_vms= d(dev), dev, > > > dev->instance_id_alia= s, > > > @@ -984,6 +990,7 @@ child_realize_fail: > > > } > > > =20 > > > post_realize_fail: > > > + g_free(dev->canonical_path); > > > if (dc->unrealize) { > > > dc->unrealize(dev, NULL); > > > } > > > @@ -1102,10 +1109,12 @@ static void device_unparent(Object *obj) > > > =20 > > > /* Only send event if the device had been completely realized */ > > > if (dev->pending_deleted_event) { > > > - gchar *path =3D object_get_canonical_path(OBJECT(dev)); > > > + g_assert(dev->canonical_path); > > > =20 > > > - qapi_event_send_device_deleted(!!dev->id, dev->id, path, &er= ror_abort); > > > - g_free(path); > > > + qapi_event_send_device_deleted(!!dev->id, dev->id, dev->cano= nical_path, > > > + &error_abort); > > > + g_free(dev->canonical_path); > > > + dev->canonical_path =3D NULL; > > > } > > > =20 > > > qemu_opts_del(dev->opts); > > > diff --git a/include/hw/qdev-core.h b/include/hw/qdev-core.h > > > index ae317286a480..9237b6849ff3 100644 > > > --- a/include/hw/qdev-core.h > > > +++ b/include/hw/qdev-core.h > > > @@ -153,6 +153,7 @@ struct DeviceState { > > > /*< public >*/ > > > =20 > > > const char *id; > > > + char *canonical_path; > > > bool realized; > > > bool pending_deleted_event; > > > QemuOpts *opts; > > > =20 > >=20 >=20 --=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 --X1bOJ3K7DJ5YkBrT Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQIzBAEBCAAdFiEEdfRlhq5hpmzETofcbDjKyiDZs5IFAll6qH4ACgkQbDjKyiDZ s5KNuw/9Ezo8Sm6lns7pBGAlqOBjR2WeEielEeKcuHk39JPGbtOXH6/4DWUtOPGf +NwrOiMFFxPrugdBz4ns632u5PM7SRp9+swFsRlm7QXVI0VJ4NA0nVAHLc/JhRbY 5PzWEGFyN98LyWMwe8M7IqE8fUKXq8DmSfiX9qlcXIk2cncFTf8cIAuxG3v64tYI exNfG9XZJzJN77H6IIaijorpO/A/SBJV/y8ixmBWvoT8//29YHK/DPu8PthIaYIJ hMEydKjVkMjfi8InVERHBgj+/VLyB/PhI+1n5s+cuYUIc+8ioUYTvK0j8RZ/vCww 3XfiZMVjzmcW7Cl59ZbAIbSXdaC39SQr+ytcN2M/VtlY1gdj+rkxyTiFWeh+5n6V ksM7jLirI0uQSbV7A5QLLePhPJUYXj3zdyMyN+BK2sO/jh/ZxkePyCm4FfkOqWN1 4VnffhOR5bMjXJT051MomRQh3LWQ9PivTDbhsLhzmQUVUAGks15MFcqXDb5rWBsz fDlw/kGOuriG5CQOELyBkxnAVdsunjGb5DCATJZ2s4drhGzZWMi3IoY+nlbEXT57 oZgYDW1QiuoAf7uBU3hXhR1WrXWxz/X9cwENnUrFQWNobtAnqNeHR9nmeI4KuvSY 8BNXfvhvwfDdxEPnEXdmt2qv7jtPcty+oW9p23RMwn4CDK8XUsE= =1+4o -----END PGP SIGNATURE----- --X1bOJ3K7DJ5YkBrT--