From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:34526) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gA7iY-00027l-LJ for qemu-devel@nongnu.org; Wed, 10 Oct 2018 02:11:35 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1gA7iW-0008FU-1j for qemu-devel@nongnu.org; Wed, 10 Oct 2018 02:11:34 -0400 Received: from mx1.redhat.com ([209.132.183.28]:56880) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1gA7iV-0007rZ-Om for qemu-devel@nongnu.org; Wed, 10 Oct 2018 02:11:31 -0400 Received: from smtp.corp.redhat.com (int-mx07.intmail.prod.int.phx2.redhat.com [10.5.11.22]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 3934D30001D3 for ; Wed, 10 Oct 2018 06:11:17 +0000 (UTC) Received: from blackfin.pond.sub.org (ovpn-116-50.ams2.redhat.com [10.36.116.50]) by smtp.corp.redhat.com (Postfix) with ESMTPS id E65411054FBA for ; Wed, 10 Oct 2018 06:11:16 +0000 (UTC) From: Markus Armbruster Date: Wed, 10 Oct 2018 08:11:13 +0200 Message-Id: <20181010061113.6057-6-armbru@redhat.com> In-Reply-To: <20181010061113.6057-1-armbru@redhat.com> References: <20181010061113.6057-1-armbru@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Subject: [Qemu-devel] [PULL 5/5] ivshmem: Fix unplug of device "ivshmem-plain" List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Commit 2aece63c8a "hostmem: detect host backend memory is being used properly" fixed "ivshmem-plain" to reject memory backends that are already in use, and to block their deletion while in use. Two bugs escaped review: * New ivshmem_plain_exit() fails to call ivshmem_exit(). This breaks unplug. Reproducer: migration after unplug still fails with "Migration is disabled when using feature 'peer mode' in device 'ivshmem'". * It failed to update legacy "ivshmem". Harmless, because it creates the memory backend itself, and nothing else should use it. Fix by moving the two host_memory_backend_set_mapped() calls into ivshmem_common_realize() and ivshmem_exit(), guarded by s->hostmem. Fixes: 2aece63c8a9d2c3a8ff41d2febc4cdeff2633331 Signed-off-by: Markus Armbruster Message-Id: <20180926163709.22876-1-armbru@redhat.com> Reviewed-by: Marc-Andr=C3=A9 Lureau Reviewed-by: Paolo Bonzini --- hw/misc/ivshmem.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/hw/misc/ivshmem.c b/hw/misc/ivshmem.c index 6febbabcaa..8cb17b9dd4 100644 --- a/hw/misc/ivshmem.c +++ b/hw/misc/ivshmem.c @@ -911,6 +911,7 @@ static void ivshmem_common_realize(PCIDevice *dev, Er= ror **errp) IVSHMEM_DPRINTF("using hostmem\n"); =20 s->ivshmem_bar2 =3D host_memory_backend_get_memory(s->hostmem); + host_memory_backend_set_mapped(s->hostmem, true); } else { Chardev *chr =3D qemu_chr_fe_get_driver(&s->server_chr); assert(chr); @@ -993,6 +994,10 @@ static void ivshmem_exit(PCIDevice *dev) vmstate_unregister_ram(s->ivshmem_bar2, DEVICE(dev)); } =20 + if (s->hostmem) { + host_memory_backend_set_mapped(s->hostmem, false); + } + if (s->peers) { for (i =3D 0; i < s->nb_peers; i++) { close_peer_eventfds(s, i); @@ -1101,14 +1106,6 @@ static void ivshmem_plain_realize(PCIDevice *dev, = Error **errp) } =20 ivshmem_common_realize(dev, errp); - host_memory_backend_set_mapped(s->hostmem, true); -} - -static void ivshmem_plain_exit(PCIDevice *pci_dev) -{ - IVShmemState *s =3D IVSHMEM_COMMON(pci_dev); - - host_memory_backend_set_mapped(s->hostmem, false); } =20 static void ivshmem_plain_class_init(ObjectClass *klass, void *data) @@ -1117,7 +1114,6 @@ static void ivshmem_plain_class_init(ObjectClass *k= lass, void *data) PCIDeviceClass *k =3D PCI_DEVICE_CLASS(klass); =20 k->realize =3D ivshmem_plain_realize; - k->exit =3D ivshmem_plain_exit; dc->props =3D ivshmem_plain_properties; dc->vmsd =3D &ivshmem_plain_vmsd; } --=20 2.17.1