Hi Michael, Could you share your opinion about this new patch please?
On Thu, Apr 11, 2024 at 10:04 AM Lidong Zhong <lidong.zh...@suse.com> wrote: > We noticed the following nuisance messages during boot process > > [ 7.120610][ T1060] vio vio: uevent: failed to send synthetic uevent > [ 7.122281][ T1060] vio 4000: uevent: failed to send synthetic uevent > [ 7.122304][ T1060] vio 4001: uevent: failed to send synthetic uevent > [ 7.122324][ T1060] vio 4002: uevent: failedto send synthetic uevent > [ 7.122345][ T1060] vio 4004: uevent: failed to send synthetic uevent > > It's caused by either vio_register_device_node() fails to set dev->of_node > or the missing "compatible" property. To keep align with the definition of > modalias in modalias_show(), we remove the return of ENODEV in such cases. > The failure messages can also be suppressed with this change. > > v2: keep align the definition of modalias with modalias_show() > Signed-off-by: Lidong Zhong <lidong.zh...@suse.com> > --- > arch/powerpc/platforms/pseries/vio.c | 8 ++------ > 1 file changed, 2 insertions(+), 6 deletions(-) > > diff --git a/arch/powerpc/platforms/pseries/vio.c > b/arch/powerpc/platforms/pseries/vio.c > index 90ff85c879bf..b2babfdbc40b 100644 > --- a/arch/powerpc/platforms/pseries/vio.c > +++ b/arch/powerpc/platforms/pseries/vio.c > @@ -1592,13 +1592,9 @@ static int vio_hotplug(const struct device *dev, > struct kobj_uevent_env *env) > const char *cp; > > dn = dev->of_node; > - if (!dn) > - return -ENODEV; > - cp = of_get_property(dn, "compatible", NULL); > - if (!cp) > - return -ENODEV; > + if (dn && (cp = of_get_property(dn, "compatible", NULL))) > + add_uevent_var(env, "MODALIAS=vio:T%sS%s", vio_dev->type, > cp); > > - add_uevent_var(env, "MODALIAS=vio:T%sS%s", vio_dev->type, cp); > return 0; > } > > -- > 2.35.3 > > -- Regards, Lidong Zhong
