Loading of a VFIO device's data can take a substantial amount of time as
the device may need to allocate resources, prepare internal data
structures, etc. This can increase migration downtime, especially for
VFIO devices with a lot of resources.

To solve this, VFIO migration uAPI defines "initial bytes" as part of
its precopy data stream. Initial bytes can be used in various ways to
improve VFIO migration performance. For example, it can be used to
transfer device metadata to pre-allocate resources in the destination.
However, for this to work we need to make sure that all initial bytes
are sent and loaded in the destination before the source VM is stopped.

Use migration precopy initial data capability to make sure a VFIO
device's initial bytes are sent and loaded in the destination before the
source stops the VM and attempts to complete the migration.
This can significantly reduce migration downtime.

Signed-off-by: Avihai Horon <avih...@nvidia.com>
---
 include/hw/vfio/vfio-common.h |  3 +++
 hw/vfio/migration.c           | 48 ++++++++++++++++++++++++++++++++++-
 2 files changed, 50 insertions(+), 1 deletion(-)

diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h
index fa42955d4c..dd3b052682 100644
--- a/include/hw/vfio/vfio-common.h
+++ b/include/hw/vfio/vfio-common.h
@@ -69,6 +69,9 @@ typedef struct VFIOMigration {
     uint64_t precopy_init_size;
     uint64_t precopy_dirty_size;
     uint64_t mig_flags;
+    bool initial_data_active;
+    bool initial_data_sent;
+    bool initial_data_loaded;
 } VFIOMigration;
 
 typedef struct VFIOAddressSpace {
diff --git a/hw/vfio/migration.c b/hw/vfio/migration.c
index 980be1f614..23f4f1f8a5 100644
--- a/hw/vfio/migration.c
+++ b/hw/vfio/migration.c
@@ -45,6 +45,7 @@
 #define VFIO_MIG_FLAG_DEV_CONFIG_STATE  (0xffffffffef100002ULL)
 #define VFIO_MIG_FLAG_DEV_SETUP_STATE   (0xffffffffef100003ULL)
 #define VFIO_MIG_FLAG_DEV_DATA_STATE    (0xffffffffef100004ULL)
+#define VFIO_MIG_FLAG_DEV_INIT_DATA_SENT (0xffffffffef100005ULL)
 
 /*
  * This is an arbitrary size based on migration of mlx5 devices, where 
typically
@@ -372,6 +373,8 @@ static void vfio_save_cleanup(void *opaque)
 
     g_free(migration->data_buffer);
     migration->data_buffer = NULL;
+    migration->initial_data_sent = false;
+    migration->initial_data_active = false;
     vfio_migration_cleanup(vbasedev);
     trace_vfio_save_cleanup(vbasedev->name);
 }
@@ -447,10 +450,17 @@ static int vfio_save_iterate(QEMUFile *f, void *opaque)
     if (data_size < 0) {
         return data_size;
     }
-    qemu_put_be64(f, VFIO_MIG_FLAG_END_OF_STATE);
 
     vfio_update_estimated_pending_data(migration, data_size);
 
+    if (migration->initial_data_active && !migration->precopy_init_size &&
+        !migration->initial_data_sent) {
+        qemu_put_be64(f, VFIO_MIG_FLAG_DEV_INIT_DATA_SENT);
+        migration->initial_data_sent = true;
+    } else {
+        qemu_put_be64(f, VFIO_MIG_FLAG_END_OF_STATE);
+    }
+
     trace_vfio_save_iterate(vbasedev->name);
 
     /*
@@ -568,6 +578,12 @@ static int vfio_load_state(QEMUFile *f, void *opaque, int 
version_id)
             }
             break;
         }
+        case VFIO_MIG_FLAG_DEV_INIT_DATA_SENT:
+        {
+            vbasedev->migration->initial_data_loaded = true;
+
+            return 0;
+        }
         default:
             error_report("%s: Unknown tag 0x%"PRIx64, vbasedev->name, data);
             return -EINVAL;
@@ -582,6 +598,33 @@ static int vfio_load_state(QEMUFile *f, void *opaque, int 
version_id)
     return ret;
 }
 
+static bool vfio_initial_data_advise(void *opaque)
+{
+    VFIODevice *vbasedev = opaque;
+    VFIOMigration *migration = vbasedev->migration;
+
+    migration->initial_data_active =
+        migration->mig_flags & VFIO_MIGRATION_PRE_COPY;
+
+    return migration->initial_data_active;
+}
+
+static bool vfio_is_initial_data_active(void *opaque)
+{
+    VFIODevice *vbasedev = opaque;
+    VFIOMigration *migration = vbasedev->migration;
+
+    return migration->initial_data_active;
+}
+
+static bool vfio_initial_data_loaded(void *opaque)
+{
+    VFIODevice *vbasedev = opaque;
+    VFIOMigration *migration = vbasedev->migration;
+
+    return migration->initial_data_loaded;
+}
+
 static const SaveVMHandlers savevm_vfio_handlers = {
     .save_setup = vfio_save_setup,
     .save_cleanup = vfio_save_cleanup,
@@ -594,6 +637,9 @@ static const SaveVMHandlers savevm_vfio_handlers = {
     .load_setup = vfio_load_setup,
     .load_cleanup = vfio_load_cleanup,
     .load_state = vfio_load_state,
+    .initial_data_advise = vfio_initial_data_advise,
+    .is_initial_data_active = vfio_is_initial_data_active,
+    .initial_data_loaded = vfio_initial_data_loaded,
 };
 
 /* ---------------------------------------------------------------------- */
-- 
2.26.3


Reply via email to