From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yoshiaki Tamura Subject: [PATCH 06/19] virtio: decrement last_avail_idx with inuse before saving. Date: Fri, 14 Jan 2011 02:15:28 +0900 Message-ID: <1294938941-9532-7-git-send-email-tamura.yoshiaki@lab.ntt.co.jp> References: <1294938941-9532-1-git-send-email-tamura.yoshiaki@lab.ntt.co.jp> Cc: avi@redhat.com, anthony@codemonkey.ws, aliguori@us.ibm.com, mtosatti@redhat.com, dlaor@redhat.com, mst@redhat.com, kwolf@redhat.com, ananth@in.ibm.com, psuriset@linux.vnet.ibm.com, vatsa@linux.vnet.ibm.com, stefanha@linux.vnet.ibm.com, ohmura.kei@lab.ntt.co.jp, Yoshiaki Tamura To: kvm@vger.kernel.org, qemu-devel@nongnu.org Return-path: Received: from sh.osrg.net ([192.16.179.4]:41985 "EHLO sh.osrg.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757293Ab1AMRUa (ORCPT ); Thu, 13 Jan 2011 12:20:30 -0500 In-Reply-To: <1294938941-9532-1-git-send-email-tamura.yoshiaki@lab.ntt.co.jp> Sender: kvm-owner@vger.kernel.org List-ID: For regular migration inuse == 0 always as requests are flushed before save. However, event-tap log when enabled introduces an extra queue for requests which is not being flushed, thus the last inuse requests are left in the event-tap queue. Move the last_avail_idx value sent to the remote back to make it repeat the last inuse requests. Signed-off-by: Michael S. Tsirkin Signed-off-by: Yoshiaki Tamura --- hw/virtio.c | 10 +++++++++- 1 files changed, 9 insertions(+), 1 deletions(-) diff --git a/hw/virtio.c b/hw/virtio.c index 07dbf86..408fad5 100644 --- a/hw/virtio.c +++ b/hw/virtio.c @@ -665,12 +665,20 @@ void virtio_save(VirtIODevice *vdev, QEMUFile *f) qemu_put_be32(f, i); for (i = 0; i < VIRTIO_PCI_QUEUE_MAX; i++) { + /* For regular migration inuse == 0 always as + * requests are flushed before save. However, + * event-tap log when enabled introduces an extra + * queue for requests which is not being flushed, + * thus the last inuse requests are left in the event-tap queue. + * Move the last_avail_idx value sent to the remote back + * to make it repeat the last inuse requests. */ + uint16_t last_avail = vdev->vq[i].last_avail_idx - vdev->vq[i].inuse; if (vdev->vq[i].vring.num == 0) break; qemu_put_be32(f, vdev->vq[i].vring.num); qemu_put_be64(f, vdev->vq[i].pa); - qemu_put_be16s(f, &vdev->vq[i].last_avail_idx); + qemu_put_be16s(f, &last_avail); if (vdev->binding->save_queue) vdev->binding->save_queue(vdev->binding_opaque, i, f); } -- 1.7.1.2 From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [140.186.70.92] (port=38327 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1PdQqO-0008H1-Oq for qemu-devel@nongnu.org; Thu, 13 Jan 2011 12:20:21 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1PdQqH-0007GQ-OY for qemu-devel@nongnu.org; Thu, 13 Jan 2011 12:20:16 -0500 Received: from sh.osrg.net ([192.16.179.4]:42688) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1PdQqH-0007GB-8A for qemu-devel@nongnu.org; Thu, 13 Jan 2011 12:20:09 -0500 From: Yoshiaki Tamura Date: Fri, 14 Jan 2011 02:15:28 +0900 Message-Id: <1294938941-9532-7-git-send-email-tamura.yoshiaki@lab.ntt.co.jp> In-Reply-To: <1294938941-9532-1-git-send-email-tamura.yoshiaki@lab.ntt.co.jp> References: <1294938941-9532-1-git-send-email-tamura.yoshiaki@lab.ntt.co.jp> Subject: [Qemu-devel] [PATCH 06/19] virtio: decrement last_avail_idx with inuse before saving. List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: kvm@vger.kernel.org, qemu-devel@nongnu.org Cc: kwolf@redhat.com, aliguori@us.ibm.com, mtosatti@redhat.com, ananth@in.ibm.com, mst@redhat.com, dlaor@redhat.com, vatsa@linux.vnet.ibm.com, Yoshiaki Tamura , ohmura.kei@lab.ntt.co.jp, avi@redhat.com, psuriset@linux.vnet.ibm.com, stefanha@linux.vnet.ibm.com For regular migration inuse == 0 always as requests are flushed before save. However, event-tap log when enabled introduces an extra queue for requests which is not being flushed, thus the last inuse requests are left in the event-tap queue. Move the last_avail_idx value sent to the remote back to make it repeat the last inuse requests. Signed-off-by: Michael S. Tsirkin Signed-off-by: Yoshiaki Tamura --- hw/virtio.c | 10 +++++++++- 1 files changed, 9 insertions(+), 1 deletions(-) diff --git a/hw/virtio.c b/hw/virtio.c index 07dbf86..408fad5 100644 --- a/hw/virtio.c +++ b/hw/virtio.c @@ -665,12 +665,20 @@ void virtio_save(VirtIODevice *vdev, QEMUFile *f) qemu_put_be32(f, i); for (i = 0; i < VIRTIO_PCI_QUEUE_MAX; i++) { + /* For regular migration inuse == 0 always as + * requests are flushed before save. However, + * event-tap log when enabled introduces an extra + * queue for requests which is not being flushed, + * thus the last inuse requests are left in the event-tap queue. + * Move the last_avail_idx value sent to the remote back + * to make it repeat the last inuse requests. */ + uint16_t last_avail = vdev->vq[i].last_avail_idx - vdev->vq[i].inuse; if (vdev->vq[i].vring.num == 0) break; qemu_put_be32(f, vdev->vq[i].vring.num); qemu_put_be64(f, vdev->vq[i].pa); - qemu_put_be16s(f, &vdev->vq[i].last_avail_idx); + qemu_put_be16s(f, &last_avail); if (vdev->binding->save_queue) vdev->binding->save_queue(vdev->binding_opaque, i, f); } -- 1.7.1.2