| Message ID | 1354740282-20679-5-git-send-email-pbonzini@redhat.com |
|---|---|
| State | New |
| Headers | show |
Paolo Bonzini <pbonzini@redhat.com> writes: > Each device has a reference through the BusChild. This reference > was not accounted for, add it now. > > Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> > --- > hw/qdev.c | 5 +++++ > 1 file changed, 5 insertions(+) > > diff --git a/hw/qdev.c b/hw/qdev.c > index e758131..87dfcb5 100644 > --- a/hw/qdev.c > +++ b/hw/qdev.c > @@ -65,7 +65,10 @@ static void bus_remove_child(BusState *bus, DeviceState *child) > > snprintf(name, sizeof(name), "child[%d]", kid->index); > QTAILQ_REMOVE(&bus->children, kid, sibling); > + > + /* This gives back ownership of kid->child back to us. */ > object_property_del(OBJECT(bus), name, NULL); > + object_unref(OBJECT(kid->child)); > g_free(kid); > return; > } > @@ -83,9 +86,11 @@ static void bus_add_child(BusState *bus, DeviceState *child) > > kid->index = bus->max_index++; > kid->child = child; > + object_ref(OBJECT(kid->child)); > > QTAILQ_INSERT_HEAD(&bus->children, kid, sibling); > > + /* This transfers ownership of kid->child to the property. */ > snprintf(name, sizeof(name), "child[%d]", kid->index); > object_property_add_link(OBJECT(bus), name, > object_get_typename(OBJECT(child)), The link property itself holds a reference no? Or is this not the case because we aren't checking a link when it's added and taking a reference if it's !NULL? Regards, Anthony Liguori > -- > 1.8.0.1
Il 07/01/2013 21:26, Anthony Liguori ha scritto: >> > + object_ref(OBJECT(kid->child)); >> > >> > QTAILQ_INSERT_HEAD(&bus->children, kid, sibling); >> > >> > + /* This transfers ownership of kid->child to the property. */ >> > snprintf(name, sizeof(name), "child[%d]", kid->index); >> > object_property_add_link(OBJECT(bus), name, >> > object_get_typename(OBJECT(child)), > > The link property itself holds a reference no? Or is this not the case > because we aren't checking a link when it's added and taking a reference > if it's !NULL? Yes. I think it's okay this way though, and anyway a separate series that would touch all object_property_add_link callers. Paolo
diff --git a/hw/qdev.c b/hw/qdev.c index e758131..87dfcb5 100644 --- a/hw/qdev.c +++ b/hw/qdev.c @@ -65,7 +65,10 @@ static void bus_remove_child(BusState *bus, DeviceState *child) snprintf(name, sizeof(name), "child[%d]", kid->index); QTAILQ_REMOVE(&bus->children, kid, sibling); + + /* This gives back ownership of kid->child back to us. */ object_property_del(OBJECT(bus), name, NULL); + object_unref(OBJECT(kid->child)); g_free(kid); return; } @@ -83,9 +86,11 @@ static void bus_add_child(BusState *bus, DeviceState *child) kid->index = bus->max_index++; kid->child = child; + object_ref(OBJECT(kid->child)); QTAILQ_INSERT_HEAD(&bus->children, kid, sibling); + /* This transfers ownership of kid->child to the property. */ snprintf(name, sizeof(name), "child[%d]", kid->index); object_property_add_link(OBJECT(bus), name, object_get_typename(OBJECT(child)),
Each device has a reference through the BusChild. This reference was not accounted for, add it now. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> --- hw/qdev.c | 5 +++++ 1 file changed, 5 insertions(+)