On 09/10/2015 09:20 AM, Nikolay Shirokovskiy wrote:
From: Michal Privoznik <mprivozn(a)redhat.com>
Direct migration should work if *perform3 is present but *perform
is not. This is situation when driver migration is implemented
after new version of driver function is introduced. We should not
be forced to support old version too as its parameter space is
subspace of newer one.
This patch has been already sent to mailist and has form
that finally suggested by Michal Privoznik. I just include
it in the set as it is not merged yet.
This last paragraph could go below the "---" since you're providing
"extra" information that we don't necessarily want in git history...
---
src/libvirt-domain.c | 13 ++++++++-----
1 files changed, 8 insertions(+), 5 deletions(-)
diff --git a/src/libvirt-domain.c b/src/libvirt-domain.c
index 6f10c74..15de714 100644
--- a/src/libvirt-domain.c
+++ b/src/libvirt-domain.c
@@ -3412,16 +3412,15 @@ virDomainMigrateDirect(virDomainPtr domain,
NULLSTR(xmlin), flags, NULLSTR(dname), NULLSTR(uri),
bandwidth);
- if (!domain->conn->driver->domainMigratePerform) {
- virReportUnsupportedError();
- return -1;
- }
-
/* Perform the migration. The driver isn't supposed to return
* until the migration is complete.
*/
if (VIR_DRV_SUPPORTS_FEATURE(domain->conn->driver, domain->conn,
VIR_DRV_FEATURE_MIGRATION_V3)) {
+ if (!domain->conn->driver->domainMigratePerform3) {
+ virReportUnsupportedError();
+ return -1;
+ }
Ahhh... now I see where patch 10 is going... Direct has no dconnuri...
Hmmm... and I know why Coverity is complaining...
John
VIR_DEBUG("Using migration protocol 3");
/* dconn URI not relevant in direct migration, since no
* target libvirtd is involved */
@@ -3437,6 +3436,10 @@ virDomainMigrateDirect(virDomainPtr domain,
dname,
bandwidth);
} else {
+ if (!domain->conn->driver->domainMigratePerform) {
+ virReportUnsupportedError();
+ return -1;
+ }
VIR_DEBUG("Using migration protocol 2");
if (xmlin) {
virReportError(VIR_ERR_ARGUMENT_UNSUPPORTED, "%s",