Re: [PATCH 3/3] nodedev: trigger mdev device definition update on udev add and remove

On 3/15/22 09:56, Boris Fiuczynski wrote: > When nodedev objects are added and removed if possible check if mdev-types is > supported by the object and trigger a mdev device definition update to correct > the associated parent nodedevs. > > Signed-off-by: Boris Fiuczynski <fiuczy(a)linux.ibm.com&gt; > --- > src/node_device/node_device_udev.c | 9 +++++++++ > 1 file changed, 9 insertions(+) > > diff --git a/src/node_device/node_device_udev.c b/src/node_device/node_device_udev.c > index b0a5e6302c..e17373a0b0 100644 > --- a/src/node_device/node_device_udev.c > +++ b/src/node_device/node_device_udev.c > @@ -1440,6 +1440,10 @@ udevRemoveOneDeviceSysPath(const char *path) > } > virNodeDeviceObjEndAPI(&obj); > > + /* cannot check for mdev_types since they have already been removed */ > + if (nodeDeviceUpdateMediatedDevices() < 0) > + VIR_WARN("mdevctl failed to updated mediated devices"); failed to update > + > virObjectEventStateQueue(driver->nodeDeviceEventState, event); > return 0; > } > @@ -1507,6 +1511,7 @@ udevAddOneDevice(struct udev_device *device) > bool persistent = false; > bool autostart = false; > bool is_mdev; > + bool is_mdev_types = false; has_mdev_cap perhaps? > > def = g_new0(virNodeDeviceDef, 1); > > @@ -1562,8 +1567,12 @@ udevAddOneDevice(struct udev_device *device) > event = virNodeDeviceEventUpdateNew(objdef->name); > > virNodeDeviceObjSetActive(obj, true); > + is_mdev_types = virNodeDeviceObjHasCap(obj, VIR_NODE_DEV_CAP_MDEV_TYPES); > virNodeDeviceObjEndAPI(&obj); > > + if (is_mdev_types && nodeDeviceUpdateMediatedDevices() < 0) > + VIR_WARN("mdevctl failed to updated mediated devices"); s/updated/update/ > + > ret = 0; > > cleanup: Otherwise looking good. I'll let Jonathon express his thoughts, but you have my: Reviewed-by: Michal Privoznik <mprivozn(a)redhat.com&gt; Michal