On Sun, 12 May 2024, Christoph Fritz wrote: > On Fri, 2024-05-10 at 17:21 +0300, Ilpo Järvinen wrote: > > On Thu, 9 May 2024, Christoph Fritz wrote: > ... > > > diff --git a/drivers/tty/serdev/core.c b/drivers/tty/serdev/core.c > > > index 613cb356b918d..23a1e76cb553b 100644 > > > --- a/drivers/tty/serdev/core.c > > > +++ b/drivers/tty/serdev/core.c > > > @@ -339,6 +339,17 @@ unsigned int serdev_device_set_baudrate(struct serdev_device *serdev, unsigned i > > > } > > > EXPORT_SYMBOL_GPL(serdev_device_set_baudrate); > > > > > > +void serdev_device_set_break_detection(struct serdev_device *serdev, bool enable) > > > +{ > > > + struct serdev_controller *ctrl = serdev->ctrl; > > > + > > > + if (!ctrl || !ctrl->ops->set_break_detection) > > > + return; > > > > Why you need to test for !ctrl? > > In our case we don't, it's an extra check like all the other functions > here: > > https://elixir.bootlin.com/linux/v6.9-rc7/source/drivers/tty/serdev/core.c#L330 Okay. It might be I just don't understand all the structs in serdev well enough to understand when ctrl actually end ups being NULL so I don't object leaving it as is. -- i.