From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:59395) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1etzUF-0006el-NA for qemu-devel@nongnu.org; Thu, 08 Mar 2018 12:37:52 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1etzUA-0007YJ-Pz for qemu-devel@nongnu.org; Thu, 08 Mar 2018 12:37:51 -0500 Date: Thu, 8 Mar 2018 17:37:42 +0000 From: Stefan Hajnoczi Message-ID: <20180308173742.GH2769@stefanha-x1.localdomain> References: <20180306204819.11266-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="MGu/vTNewDGZ7tmp" Content-Disposition: inline In-Reply-To: Subject: Re: [Qemu-devel] [Qemu-block] [PATCH 0/2] block: fix nbd-server-stop crash after blockdev-snapshot-sync List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Eric Blake Cc: Stefan Hajnoczi , qemu-devel@nongnu.org, Kevin Wolf , Max Reitz , qemu-block@nongnu.org, Stefano Panella --MGu/vTNewDGZ7tmp Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Mar 07, 2018 at 05:27:45PM -0600, Eric Blake wrote: > On 03/06/2018 02:48 PM, Stefan Hajnoczi wrote: > > The blockdev-snapshot-sync command uses bdrv_append() to update all par= ents to > > point at the external snapshot node. This breaks BlockBackend's > > blk_add/remove_aio_context_notifier(), which doesn't expect a BDS chang= e. > >=20 > > Patch 1 fixes this by tracking AioContext notifiers in BlockBackend. > >=20 > > See the test case in Patch 2 for a reproducer. > >=20 > > Stefan Hajnoczi (2): > > block: let blk_add/remove_aio_context_notifier() tolerate BDS changes > > iotests: add 208 nbd-server + blockdev-snapshot-sync test case > >=20 > > block/block-backend.c | 63 +++++++++++++++++++++++++++++++++++++= +++++++++ > > block/trace-events | 2 ++ > > tests/qemu-iotests/208 | 55 +++++++++++++++++++++++++++++++++++++= +++ > > tests/qemu-iotests/208.out | 9 +++++++ > > tests/qemu-iotests/group | 1 + > > 5 files changed, 130 insertions(+) > > create mode 100755 tests/qemu-iotests/208 > > create mode 100644 tests/qemu-iotests/208.out >=20 > Whose tree should this series go through? MAINTAINERS didn't flag it as > directly touching any files that normally affect my NBD queue, but given > that the iotest that reproduces the problem uses NBD, I'm fine if you want > it to go through me. Good question. Max and Kevin maintain block/block-backend.c so one of them should be happy with this series before it gets merged. When a patch affects multiple trees, the last sub-maintainer to review it can do the merge. So if they have already posted their R-b when you are finished, then feel free to merge it! And vice versa. Stefan --MGu/vTNewDGZ7tmp Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQEcBAEBAgAGBQJaoXTmAAoJEJykq7OBq3PIaugIAJso7FA42ABKiaZ5UwWO7OD3 swZLZrpOzcgP4xCo6KIxgj8Dp0l1dBq2lZHGUf8PreAuWkgheyF2s49ivs+Vhu5b jh1NoBknwQ31SSEKIOG/se+d3RfZeHmuzJ98vwK10Y2BspnMY2ApEqk64g0ZFKM3 ofveBp3iJcTg4T2vfRF4ka4mYtuuMsECJSM+w7Ea/uFJyDivnO1Gqt+vhpuMv0L7 x0+h2viEvveFP9IWLwtq6eQhV8hIH/NlVL+CY6Zuc1aedJkQ0rm5fqNh79Ui4dwk B3EwmuKZnxtdK0SUg6S0uCeerQGBwjkW7G8t+cQ3QAL6k9CV7kKCZxciezbeuYw= =VTom -----END PGP SIGNATURE----- --MGu/vTNewDGZ7tmp--