4:58 a.m.
I went through coverity report and found these.
Michal Privoznik (2):
virDomainMigrateUnmanagedParams: Don't blindly dereference @dconnuri
virt-host-validate-common: Print warning on missing IOMMU
src/libvirt-domain.c | 2 +-
src/qemu/qemu_migration.c | 2 +-
tools/virt-host-validate-common.c | 3 ---
3 files changed, 2 insertions(+), 5 deletions(-)
--
2.4.10
4:58 a.m.
New subject: [libvirt] [PATCH 1/2] virDomainMigrateUnmanagedParams: Don't blindly dereference @dconnuri
This function may be called with @dconnuri == NULL, e.g. from
virDomainMigrateToURI3() if the flags are missing
VIR_MIGRATE_PEER2PEER flag. Moreover, all later functions called
from here do wrap it into NULLSTR() so why not do the same here?
Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com>
---
src/libvirt-domain.c | 2 +-
src/qemu/qemu_migration.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/libvirt-domain.c b/src/libvirt-domain.c
index 7290892..677a9ad 100644
--- a/src/libvirt-domain.c
+++ b/src/libvirt-domain.c
@@ -3428,7 +3428,7 @@ virDomainMigrateUnmanagedParams(virDomainPtr domain,
unsigned int flags)
{
VIR_DOMAIN_DEBUG(domain, "dconnuri=%s, params=%p, nparams=%d, flags=%x",
- dconnuri, params, nparams, flags);
+ NULLSTR(dconnuri), params, nparams, flags);
VIR_TYPED_PARAMS_DEBUG(params, nparams);
if ((flags & VIR_MIGRATE_PEER2PEER) &&
diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c
index bb708a3..44ca91a 100644
--- a/src/qemu/qemu_migration.c
+++ b/src/qemu/qemu_migration.c
@@ -5646,7 +5646,7 @@ qemuMigrationPerform(virQEMUDriverPtr driver,
cookieout, cookieoutlen,
flags, resource);
} else {
- return qemuMigrationPerformJob(driver, conn, vm, xmlin, dconnuri,
+ return qemuMigrationPerformJob(driver, conn, vm, xmlin, NULL,
uri, graphicsuri, listenAddress,
nmigrate_disks, migrate_disks,
cookiein, cookieinlen,
--
2.4.10
4:58 a.m.
New subject: [libvirt] [PATCH 2/2] virt-host-validate-common: Print warning on missing IOMMU
No only coverity warns about this, but it kind of makes sense
too. We have a test whether host supports IOMMU. Some platforms
don't have it, I know. But in that case we should print a message
that it's unknown whether platform has it or not.
Before:
(no output)
After:
QEMU: Checking for device assignment IOMMU support : WARN
(Unknown if this platform has IOMMU support)
Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com>
---
tools/virt-host-validate-common.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/tools/virt-host-validate-common.c b/tools/virt-host-validate-common.c
index dd1c414..8ebf73e 100644
--- a/tools/virt-host-validate-common.c
+++ b/tools/virt-host-validate-common.c
@@ -367,9 +367,6 @@ int virHostValidateIOMMU(const char *hvname,
isIntel = true;
else if (virHostValidateHasCPUFlag("svm"))
isAMD = true;
- else
- /* XXX PPC/ARM/etc support */
- return 0;
virHostMsgCheck(hvname, "%s", _("for device assignment IOMMU
support"));
--
2.4.10
8:11 a.m.
On 01/07/2016 05:58 AM, Michal Privoznik wrote:
I went through coverity report and found these.
Michal Privoznik (2):
virDomainMigrateUnmanagedParams: Don't blindly dereference @dconnuri
virt-host-validate-common: Print warning on missing IOMMU
src/libvirt-domain.c | 2 +-
src/qemu/qemu_migration.c | 2 +-
tools/virt-host-validate-common.c | 3 ---
3 files changed, 2 insertions(+), 5 deletions(-)
ACK to both
John
3240
days inactive
3242
days old
3 comments
2 participants
participants (2)
-
John Ferlan -
Michal Privoznik