On 08/04/2011 10:22 AM, Jim Fehlig wrote:
Discussed previously:
https://www.redhat.com/archives/libvir-list/2011-August/msg00166.html
The qemu migration speed default is 32MiB/s as defined in migration.c
/* Migration speed throttling */
static int64_t max_throttle = (32<< 20);
There is no reason to throttle migration when targeting a file. For
dump and save operations, set migration speed to unlimited prior to
migration and restore to default value after migration. Default units
is MB for migrate_set_speed monitor command, so (INT64_MAX / (1024 * 1024))
is used for unlimited migration speed.
Tested with both json and text monitors.
---
src/qemu/qemu_migration.c | 12 ++++++++++++
1 files changed, 12 insertions(+), 0 deletions(-)
Nice idea, but incomplete patch.
diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c
index 7aeea69..4542289 100644
--- a/src/qemu/qemu_migration.c
+++ b/src/qemu/qemu_migration.c
@@ -2676,6 +2676,13 @@ qemuMigrationToFile(struct qemud_driver *driver, virDomainObjPtr
vm,
virCommandPtr cmd = NULL;
int pipeFD[2] = { -1, -1 };
+ /* No need for qemu default of 32MiB/s when migrating to a file.
+ Default speed unit is MB, so set to unlimited with INT64_MAX / 1M.
+ Failure to change migration speed is not fatal. */
+ qemuDomainObjEnterMonitor(driver, vm);
+ qemuMonitorSetMigrationSpeed(priv->mon, INT64_MAX / (1024 * 1024));
+ qemuDomainObjExitMonitor(driver, vm);
+
if (qemuCapsGet(priv->qemuCaps, QEMU_CAPS_MIGRATE_QEMU_FD)&&
Missing a check that the domain is still active after you exit the
monitor (in both places). Furthermore, since migrationToFile is only
used inside async jobs, we should be using
qemuDomainObjEnterMonitorAsync, not qemuDomainObjEnterMonitor (fixing
that will add in the check for a still-active vm, but then you have to
be careful to only call qemuDomainObjExitMonitorWithDriver if you
successfully entered the monitor).
--
Eric Blake eblake(a)redhat.com +1-801-349-2682
Libvirt virtualization library
http://libvirt.org