9:51 a.m.
See 3/4
Peter Krempa (4):
syms: Properly export 'virDomainDeviceDefValidate'
qemu: hotplug: Validate definition of 'FS' device after address
allocation
qemu: validate: Fix check for unsupported FS-device bootindex use on
un-assigned addresses
NEWS: Mention fix for broken 'fs' device bootindex support check
NEWS.rst | 6 ++++++
src/libvirt_private.syms | 1 +
src/qemu/qemu_hotplug.c | 4 ++++
src/qemu/qemu_validate.c | 5 ++++-
4 files changed, 15 insertions(+), 1 deletion(-)
--
2.45.1
9:51 a.m.
New subject: [PATCH 1/4] syms: Properly export 'virDomainDeviceDefValidate'
While the function is exported via header, the symbol itself was not.
Signed-off-by: Peter Krempa <pkrempa(a)redhat.com>
---
src/libvirt_private.syms | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms
index f0f7aa8654..653c84a520 100644
--- a/src/libvirt_private.syms
+++ b/src/libvirt_private.syms
@@ -796,6 +796,7 @@ virDomainDefPostParse;
virDomainActualNetDefValidate;
virDomainDefOSValidate;
virDomainDefValidate;
+virDomainDeviceDefValidate;
virDomainDeviceValidateAliasForHotplug;
virDomainDiskDefSourceLUNValidate;
virDomainDiskDefValidateStartupPolicy;
--
2.45.1
9:51 a.m.
New subject: [PATCH 2/4] qemu: hotplug: Validate definition of 'FS' device after address allocation
Some of the checks make sense only after the address is allocated and
thus we need to re-do the validation after the address is assigned.
Signed-off-by: Peter Krempa <pkrempa(a)redhat.com>
---
src/qemu/qemu_hotplug.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c
index 4739beead8..416e89d00d 100644
--- a/src/qemu/qemu_hotplug.c
+++ b/src/qemu/qemu_hotplug.c
@@ -40,6 +40,7 @@
#include "domain_audit.h"
#include "domain_cgroup.h"
#include "domain_interface.h"
+#include "domain_validate.h"
#include "netdev_bandwidth_conf.h"
#include "domain_nwfilter.h"
#include "virlog.h"
@@ -3216,6 +3217,9 @@ qemuDomainAttachFSDevice(virQEMUDriver *driver,
qemuAssignDeviceFSAlias(vm->def, fs);
+ if (virDomainDeviceDefValidate(&dev, vm->def, 0, driver->xmlopt,
priv->qemuCaps) < 0)
+ goto cleanup;
+
chardev = virDomainChrSourceDefNew(priv->driver->xmlopt);
chardev->type = VIR_DOMAIN_CHR_TYPE_UNIX;
chardev->data.nix.path = qemuDomainGetVHostUserFSSocketPath(priv, fs);
--
2.45.1
9:51 a.m.
New subject: [PATCH 3/4] qemu: validate: Fix check for unsupported FS-device bootindex use on un-assigned addresses
When hot-plugging a FS device with un-assigned address with a bootindex
the recently-added validation check would fail as validation on hotplug
is done prior to address assignment.
To fix this problem we can simply relax the check to also pass on _NONE
addresses. Unsupported configurations will still be caught as previous
commit re-checks the definition after address assignment prior to
hotplug.
Resolves: https://issues.redhat.com/browse/RHEL-39271
Fixes: 4690058b6d3dab672bd18ff69c83392245253024
Signed-off-by: Peter Krempa <pkrempa(a)redhat.com>
---
src/qemu/qemu_validate.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/src/qemu/qemu_validate.c b/src/qemu/qemu_validate.c
index ac1940cb31..c08e1538f9 100644
--- a/src/qemu/qemu_validate.c
+++ b/src/qemu/qemu_validate.c
@@ -4519,7 +4519,10 @@ qemuValidateDomainDeviceDefFS(virDomainFSDef *fs,
return -1;
}
- if (fs->info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_PCI) {
+ /* Address type may be _NONE when validating and will be assigned
+ * later during hotplug */
+ if (fs->info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_NONE &&
+ fs->info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_PCI) {
virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s",
_("setting virtiofs boot order is supported only with
PCI bus"));
return -1;
--
2.45.1
9:51 a.m.
New subject: [PATCH 4/4] NEWS: Mention fix for broken 'fs' device bootindex support check
Signed-off-by: Peter Krempa <pkrempa(a)redhat.com>
---
NEWS.rst | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/NEWS.rst b/NEWS.rst
index 14505116b1..edb4a9965d 100644
--- a/NEWS.rst
+++ b/NEWS.rst
@@ -70,6 +70,12 @@ v10.4.0 (unreleased)
The bug was introduced in `v10.1.0 (2024-03-01)`_.
+ * qemu: Fix hotplug of ``virtiofs`` filesystem device with ``<boot order=`` set
+
+ The bug was introduced in `v10.3.0 (2024-05-02)`_ when attempting to reject
+ unsupported configurations. During hotplug the addresses are
+ assigned after validation and thus errorneously reject valid configs.
+
v10.3.0 (2024-05-02)
====================
--
2.45.1
2:44 a.m.
New subject: [PATCH 0/4] qemu: Fix broken check rejecting virtiofs hotplug
with 'bootindex' set
On 5/30/24 16:51, Peter Krempa wrote:
See 3/4
Peter Krempa (4):
syms: Properly export 'virDomainDeviceDefValidate'
qemu: hotplug: Validate definition of 'FS' device after address
allocation
qemu: validate: Fix check for unsupported FS-device bootindex use on
un-assigned addresses
NEWS: Mention fix for broken 'fs' device bootindex support check
NEWS.rst | 6 ++++++
src/libvirt_private.syms | 1 +
src/qemu/qemu_hotplug.c | 4 ++++
src/qemu/qemu_validate.c | 5 ++++-
4 files changed, 15 insertions(+), 1 deletion(-)
Reviewed-by: Michal Privoznik <mprivozn(a)redhat.com>
Michal
175
days inactive
176
days old
5 comments
2 participants
participants (2)
-
Michal Prívozník -
Peter Krempa