From c12edc6452023b72b3eb25aeddf56b7bf5e9c627 Mon Sep 17 00:00:00 2001 Message-Id: In-Reply-To: References: From: Stefan Hajnoczi Date: Fri, 9 Aug 2013 15:56:48 +0200 Subject: [PATCH 13/13] dataplane: refuse to start if device is already in use RH-Author: Stefan Hajnoczi Message-id: <1376063808-9815-1-git-send-email-stefanha@redhat.com> Patchwork-id: 53136 O-Subject: [PATCH RHEL6.5 qemu-kvm] dataplane: refuse to start if device is already in use Bugzilla: 995530 RH-Acked-by: Kevin Wolf RH-Acked-by: Laszlo Ersek RH-Acked-by: Fam Zheng BZ: 995530 Brew: https://brewweb.devel.redhat.com/taskinfo?taskID=6150474 Upstream-status: b0f2027cde31c645524256763672e09eeb204a9a Dataplane must check whether a block device is in use before launching the dataplane thread. This is necessary since the thread does not synchronize with the main loop and I/O requests could cause corruption. One example is when a drive is added and a block job is started before hotplugging the virtio-blk-pci adapter. In this case we must not use dataplane mode. Cc: qemu-stable@nongnu.org Reviewed-by: Kevin Wolf Signed-off-by: Stefan Hajnoczi (cherry picked from commit b0f2027cde31c645524256763672e09eeb204a9a) Signed-off-by: Stefan Hajnoczi Conflicts: hw/block/dataplane/virtio-blk.c Upstream hw/ was reorganized but RHEL 6.5 qemu-kvm still uses hw/dataplane/virtio-blk.c. --- Laszlo pointed out that this fix could help RHEL6.5, not just RHEL7.0 where I backported the live migration feature. Although dataplane is tech preview in RHEL6.5, it's good to avoid data corruption so let's take this fix. hw/dataplane/virtio-blk.c | 8 ++++++++ 1 file changed, 8 insertions(+) Signed-off-by: Michal Novotny --- hw/dataplane/virtio-blk.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/hw/dataplane/virtio-blk.c b/hw/dataplane/virtio-blk.c index 7d67e11..9485411 100644 --- a/hw/dataplane/virtio-blk.c +++ b/hw/dataplane/virtio-blk.c @@ -391,6 +391,14 @@ bool virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *blk, return false; } + /* If dataplane is (re-)enabled while the guest is running there could be + * block jobs that can conflict. + */ + if (bdrv_in_use(blk->conf.bs)) { + error_report("cannot start dataplane thread while device is in use"); + return false; + } + fd = raw_get_aio_fd(blk->conf.bs); if (fd < 0) { error_report("drive is incompatible with x-data-plane, " -- 1.7.11.7