[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[RFC PATCH v2 10/10] virtio_net: register incremental migration handlers
|
From: |
Eugenio Pérez |
|
Subject: |
[RFC PATCH v2 10/10] virtio_net: register incremental migration handlers |
|
Date: |
Tue, 28 Nov 2023 11:43:03 +0100 |
This way VirtIONet can detect when the incoming migration starts.
While registering in the backend (nc->peer) seems more logical, we need
nic dma address space, and we cannot get it from the backend.
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
---
This could be done in vhost_vdpa or VirtIODevice struct, but future
series will add state restore through CVQ so it's easier to start in
VirtIONet directly. If we need to make this more generic, we can move
to VirtIODevice and expose callbacks from VirtIONet class.
Also, the pointer may not be the best id, but there are not a lot of
things initialized in n.
---
hw/net/virtio-net.c | 35 +++++++++++++++++++++++++++++++++++
1 file changed, 35 insertions(+)
diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
index 80c56f0cfc..374d0b4ec8 100644
--- a/hw/net/virtio-net.c
+++ b/hw/net/virtio-net.c
@@ -38,6 +38,7 @@
#include "qapi/qapi-events-migration.h"
#include "hw/virtio/virtio-access.h"
#include "migration/misc.h"
+#include "migration/register.h"
#include "standard-headers/linux/ethtool.h"
#include "sysemu/sysemu.h"
#include "trace.h"
@@ -3810,9 +3811,39 @@ static void virtio_net_device_unrealize(DeviceState *dev)
virtio_cleanup(vdev);
}
+static int virtio_net_load_setup(QEMUFile *f, void *opaque)
+{
+ VirtIONet *n = opaque;
+ NetClientState *nc = qemu_get_subqueue(n->nic, 0);
+
+ if (nc->peer->info->load_setup) {
+ return nc->peer->info->load_setup(nc->peer, n->nic);
+ }
+
+ return 0;
+}
+
+static int virtio_net_load_cleanup(void *opaque)
+{
+ VirtIONet *n = opaque;
+ NetClientState *nc = qemu_get_subqueue(n->nic, 0);
+
+ if (nc->peer->info->load_cleanup) {
+ return nc->peer->info->load_cleanup(nc->peer, n->nic);
+ }
+
+ return 0;
+}
+
+static const SaveVMHandlers savevm_virtio_net_handlers = {
+ .load_setup = virtio_net_load_setup,
+ .load_cleanup = virtio_net_load_cleanup,
+};
+
static void virtio_net_instance_init(Object *obj)
{
VirtIONet *n = VIRTIO_NET(obj);
+ g_autoptr(GString) id = g_string_new(NULL);
/*
* The default config_size is sizeof(struct virtio_net_config).
@@ -3824,6 +3855,10 @@ static void virtio_net_instance_init(Object *obj)
DEVICE(n));
ebpf_rss_init(&n->ebpf_rss);
+
+ g_string_printf(id, "%p", n);
+ register_savevm_live(id->str, VMSTATE_INSTANCE_ID_ANY, 1,
+ &savevm_virtio_net_handlers, n);
}
static int virtio_net_pre_save(void *opaque)
--
2.39.3
- [RFC PATCH v2 00/10] Map memory at destination .load_setup in vDPA-net migration, Eugenio Pérez, 2023/11/28
- [RFC PATCH v2 01/10] vdpa: do not set virtio status bits if unneeded, Eugenio Pérez, 2023/11/28
- [RFC PATCH v2 02/10] vdpa: make batch_begin_once early return, Eugenio Pérez, 2023/11/28
- [RFC PATCH v2 04/10] vdpa: extract out _dma_end_batch from _listener_commit, Eugenio Pérez, 2023/11/28
- [RFC PATCH v2 03/10] vdpa: merge _begin_batch into _batch_begin_once, Eugenio Pérez, 2023/11/28
- [RFC PATCH v2 05/10] vdpa: factor out stop path of vhost_vdpa_dev_start, Eugenio Pérez, 2023/11/28
- [RFC PATCH v2 06/10] vdpa: check for iova tree initialized at net_client_start, Eugenio Pérez, 2023/11/28
- [RFC PATCH v2 08/10] vdpa: add vhost_vdpa_load_setup, Eugenio Pérez, 2023/11/28
- [RFC PATCH v2 07/10] vdpa: set backend capabilities at vhost_vdpa_init, Eugenio Pérez, 2023/11/28
- [RFC PATCH v2 10/10] virtio_net: register incremental migration handlers,
Eugenio Pérez <=
- [RFC PATCH v2 09/10] vdpa: add vhost_vdpa_net_load_setup NetClient callback, Eugenio Pérez, 2023/11/28