From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:53945) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1YsSrb-0008KD-J4 for qemu-devel@nongnu.org; Wed, 13 May 2015 05:22:04 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1YsSra-0007TJ-L9 for qemu-devel@nongnu.org; Wed, 13 May 2015 05:22:03 -0400 Date: Wed, 13 May 2015 11:21:52 +0200 From: Kevin Wolf Message-ID: <20150513092152.GD4263@noname.str.redhat.com> References: <1431105726-3682-1-git-send-email-kwolf@redhat.com> <1431105726-3682-20-git-send-email-kwolf@redhat.com> <55527341.6030004@redhat.com> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="E13BgyNx05feLLmH" Content-Disposition: inline In-Reply-To: <55527341.6030004@redhat.com> Subject: Re: [Qemu-devel] [PATCH 19/34] qcow2: Make qcow2_update_options() suitable for transactions List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Eric Blake Cc: mreitz@redhat.com, qemu-devel@nongnu.org, qemu-block@nongnu.org, armbru@redhat.com --E13BgyNx05feLLmH Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Am 12.05.2015 um 23:40 hat Eric Blake geschrieben: > On 05/08/2015 11:21 AM, Kevin Wolf wrote: > > Before we can allow updating options at runtime with bdrv_reopen(), we > > need to split the function into prepare/commit/abort parts. > >=20 > > Signed-off-by: Kevin Wolf > > --- > > block/qcow2.c | 101 ++++++++++++++++++++++++++++++++++++++------------= -------- > > 1 file changed, 67 insertions(+), 34 deletions(-) > >=20 >=20 > In isolation, it looks like a valid conversion, so: >=20 > Reviewed-by: Eric Blake >=20 > However, given that we are having a conversation on another thread about > semantics for prepare vs. commit being the action that actually changes > the in-memory representation, is this the correct split for however we > end up resolving that bug? Reopen transactions are a different thing than QMP transactions, so I hope that they are not affected by that discussion. Generally, my assumption with reopen transactions is that the change only takes effect in commit. Kevin --E13BgyNx05feLLmH Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQIcBAEBAgAGBQJVUxewAAoJEH8JsnLIjy/Wm64QAJhN5+EWTS9zdOoNxQM6w4fP QT+aRHkeLVHaOiHUizcqKO2SyX/MdZTaoq7EhG3ZJRNF4vJHwNOXtEQEQauphQZH 1EBcS9MfU2Qj6qIKKSrtS9jy2V3iuloYjZFQz2eh4Y3m4dXxz46kpgBXf6v9f/SR LndY3wZ3g+wzILhnPXYBd2F2iwKmz97+OW8i6D8r3C2MsLt6vhBhJus1+l4yvweQ CjTsLJ3WVqePPym0VpcF58Q8QoE8xCFamqd+Nvqa9GyQynYPED+RUcIpuT33cfA+ KQkTDLGYAyjVCHaJj/zJqNppNTOLRKVCpkSV1d2Wj4j7p8KJ5RC+EEEso1HZ+DFt Tn3JJTZSYYDWBNrMsIewyR76Avx3tpGct3Iud0JjCQEBFTRjnf+GcmZu9nxvAMDJ r+H2QRPCDQZf4l1KIf38OytX0PJOFFU+H3HFvGX6KhaJNxtIcjQmMN3/AkarymwC i6mE4r57p/gASH78d2bg5YAjVAopaqtEN3aMrNTn2TvtIcVolrnNH5HbdOgU+Qo0 UFiy7XT+m+ZNe6pYVYP6q1qTrOD0JoKqLD0Ga00DC0xqj6hgK2ijA0FTfrFR6+Fs TVnGTmd2PMwSPN5U/a/aaG5Xmu8eUZijt89tT/DJaXOjsvuW0qth0iZn4urSzszH mcGusMOYNwfty/QoRm8W =fq5s -----END PGP SIGNATURE----- --E13BgyNx05feLLmH--